Day changed to 2026-02-02
[02:41] Wolfgang Middlesit: Raphael Nethersmell, that’s worth about 18.6k shells more than what you’re offering.
[02:43] Wolfgang Middlesit: Raphael Nethersmell, that’s worth about 2.5k shells more than what you’re offering.
[02:46] Wolfgang Middlesit: Raphael Nethersmell, that’s worth about 61 shells more than what you’re offering.
[03:05] Raphael Nethersmell: pro tip I shared with Vivian Sporepress and might as well share w/ the log : if you’re building multiple claims of the same resource, you can load up the 10th slot in your inventory with however many enums you want*, as you ziggurate each claim, it’ll decrement the count by one. this saves you from having to click & cluck enums over to the 10th slot each run, which apart from saving you time & effort, saves server traffic.
[03:05] Raphael Nethersmell: *if there’s an upper limit, i’ve not found it yet.
[08:13] Diana Coman: well noted, Raphael Nethersmell and it works for any action that needs something in that slot, whether use or explore on whatever it is. The limit is likely only the maxint upper limit of what can go in a single stack.
[08:14] Diana Coman: meanwhile it turns out that my char ends up overbulk faster than overweight and with …grass of all things, huh.
[08:14] Diana Coman: apparently the trouble with keeping keys is that they are way bulkier than the corresponding lbn
[10:38] Diana Coman: talking on ziggurating claims, there is now available the bot zig job that goes through the keys in one’s inventory, picks one with a claim that is currently visible, with quality above given threshold and for which there is also the corresponding enum + bundle/ingredients, proceeds to the claim’s location, equips enum, moves ingredients to claim (using % as given), requests use on it, takes results from claim as it finds them.
[10:38] Ulrich Logfetch: Quoting 2026-02-02 (Notices and Announcements) Monroe Quarterslit: ** Euloran news: newly deployed client adds bot zig and bot bundle jobs, as well as a visible warning when one’s own char in game is overweight or overbulk (as the client knows it at the time). While the new zig job is known to handle successfuly all the intricacies of ziggurating claims, perfection does not exist and how effective and/or efficient it may prove to be for one player or another depends as well on how well the player understands what they are doing and thus how well they put it to use. So now that there is at least one automation available for each of the more intricate in-game activities, the question is: what will PCs do with it all?**
[10:42] Diana Coman: obviously, given the number and intricacies of steps in there, it’s quite the job and there are plenty of things that can go wrong – on occasion they do go wrong, too! Generally from the relatively short-time (couple of days) testing I got to put it through, it recovers by itself just fine, only it can take a while/end up repeatedly moving stuff back and forth – so if you do mean to use it, it’s probably best to see how it works in your specific setup and then figure out what parameter values and even what sequence of jobs works best for what you want from it (e.g. since ziggurating needs a lot of info on a claim to be able to work with it, it might not be able to start on a freshly explored new claim straight away)
[10:45] Diana Coman: making better (best?) use of things as they are though is quite the nature of the game and where the difference can be made by any player without needing to start or know programming, so let me know if you give it a try.
[10:46] Diana Coman: for the record, the bundle job does exactly what you’d expect, namely aims to make bundles only, in whatever container is set as target when you start the job. Both bundle and zig actually read the blueprint/method so they’ll keep working even if/when the counts change on it.
[12:53] Diana Coman: a note to mind perhaps the location of claims when zig+explore, esp when relatively close to sector border (coming to you from my rather unintended trip to the faraway Barmy Crossing of Madness (some 16 sectors south of PPP according to the handy clientside map, way past that Fertile Knoll of Friskiness)
[15:14] Raphael Nethersmell: Diana Coman, sounds great re zigg’d botting, I’ve updated my client and will try it out later tonight.
[15:24] Diana Coman: cheers.
[16:10] Vivian Sporepress: Diana Coman: spiffy economic reporting from Monroe Quarterslit. just that seeing the deltas now makes me wonder about the opening positions. poor environment… at least it’s big, or so it appears so far
[16:14] Vivian Sporepress: hearing of the NPC’s cash decreasing and so forth makes me wonder, do they have some limited initial pile of it? or can they just go arbitrarily negative if people keep selling stuff to the merchants?
[16:15] Diana Coman: Vivian Sporepress, good to know, thanks. The deltas are literally the result as noticed at that monthly measurement. Why exactly wondering about the opening positions as such? Kind of iffy from a privacy pov to make these public.
[16:17] Diana Coman: well, in the widest sense, keeping all euloran economy meaningful and ticking is part of my job as world-maintainer so I do keep an eye on all these things. In the immediate, you noticed perhaps that an NPC out of cash will say so and refuse the trade since he can’t match it, there is no magicking anything out of thin air.
[16:18] Diana Coman: in general terms, value may get perhaps recycled when/if/as needed but not just created out of nothing.
[16:19] Vivian Sporepress: I suppose one can look at how long the game’s been going and extrapolate to get some rough idea of the totals based on the unreported months. yeah, I don’t know if it’s wisest to publish the totals, it just came to mind to give more context to the changes, or more meaning to the monetary amounts, like knowing the total btc supply limit
[16:20] Diana Coman: heh, more knowledge is always wanted but rarely available just for that reason, certainly
[16:20] Diana Coman: I’m pondering whether a total of amount traded might be added to these reports at some point but it’s not yet top of the list anyhow
[16:21] Vivian Sporepress: I saw some having trouble making change at least with Wolfgang Middlesit, but no such with Rainer Tracksnarl
[16:22] Diana Coman: for the record, one case in which I can see a reason to make the total btc supply limit known and public would be perhaps if there is a public pooling of such btc i.e. literally public subscription
[16:23] Diana Coman: every NPC has his/her limits, even if some may be larger than others!
[16:25] Vivian Sporepress: what do you mean about a public pooling/subscription? I mentioned btc as an example, knowing it’s not directly involved here
[16:26] Vivian Sporepress: and makes sense about limits; iirc Rainer Tracksnarl had at most 9999 of any given item
[16:26] Vivian Sporepress: (or maybe 10k even)
[16:30] Vivian Sporepress: anyway, over here in the Cruel Mound of Pluck I was attempting a full grid explore of the sector, my first such with the public bot. I even packed a lunch so I didn’t have to return home for that; but now it looks like the bot’s overall rep count simply ran out, and I don’t know how to resume the same grid rather than starting a new spiral from here
[16:31] Diana Coman: by public pooling/subscription I meant the case where the in-game value would have a corresponding pot of btc made public and publicly tradeable. Came to mind since it’s just about the only scenario I can currently see where there would be good reason and even required to make the overall sums clearly known and accountable publicly.
[16:33] Vivian Sporepress: on that worthless putrid leather supply, anyone can perhaps find similar now through my previous published map (the moony precipice being in the immediate outer ring from PPP)
[16:33] Ulrich Logfetch: Quoting 2026-01-29 (Eulora 2) Diana Coman: Raphael Nethersmell, that change from one offs to patch of wpl sounds interesting, any hypothesis as to why/how come? I’ve found some more ords but they keep requiring rare resources that nobody found yet and then on top, quite a lot of them as my char reads that enum, too. So not much to do about it atm other than hope either someone finds those rare resources or Rainer Tracksnarl gets a shipment of ord bundles perhaps…
[16:33] Diana Coman: Vivian Sporepress, ah, that is indeed not implemented as such since it would require setting a whole set of new parameters to start a grid from some arbitrary point in it rather than the beginning. Fwiw I simply pick another point in the middle of some empty area and start a new grid there, I didn’t find it yet annoying enough to warrant the trouble of more parameters
[16:36] Vivian Sporepress: I could continue to fill the space at same interval by calculating new spiral centers.
[16:36] Diana Coman: exactly
[16:37] Vivian Sporepress: first I might need to get some visualization pipeline for the new knowledge format, to get something from the raw data.
[16:39] Diana Coman: fwiw I tend to give it quite large step sizes too at least in new places, since it has the double advantage of filling reasonably well the grid and keeping the total number of claims reasonable (given that owned items including claims will not vanish while the owner is nearby, the drunkmove or a gridmove with small step can easily end up with a ton of claims and thus getting bogged down by all that data)
[16:40] Vivian Sporepress: are there some existing tools that consume the res_sect_kw.txt format or are you further awking it or something?
[16:41] Vivian Sporepress: ah, yes you noted sh+awk, http://ossasepia.com/2026/01/11/euloran-exploration-2-0/#footnote_2_7056
[16:42] Vivian Sporepress: bogged down with data or even just keys, as you noted recently.
[16:42] Diana Coman: I have a single sh script that goes through all the data, produces the inputs for graphviz (one per sector) and then runs graphviz to get the .png files too
[16:42] Vivian Sporepress: gnuplot?
[16:43] Diana Coman: gnuplot indeed
[16:44] Diana Coman: at least out of this most recent unplanned trip I ended up with a bunch of data points in all those sectors on the way back to PPP and so the script got a better testing too, it produced the 16 pictures just fine, lolz
[16:46] Diana Coman: correction there: the sh script runs an awk script because there is one such and apparently I even forgot it was enough in there to warrant a separate file
[16:48] Diana Coman: basically the awk does the conversion from kw data to the format wanted by gnuplot, while sh handles the multi-sector, multi-file part
[16:50] Vivian Sporepress: right
[16:53] Vivian Sporepress: I started out in awk but ended up in python, plus a minimal .sh to run the gnuplot for the resulting data file for a given sector; but then, that was before the support of the client knowledge maps and required correlating a lot of data from distinct data change log messages.
[16:59] Diana Coman: this reminds me that I found the original awk textbook and it’s quite striking how it reads precisely like what python supposedly is meant to be for only it rarely is – a simple and very effective tool for scripting, prototyping and data conversion especially. Possibly the environment of each is in the end more the deciding factor as to how they (awk and python) ended up being what they are.
[17:02] Vivian Sporepress: perhaps python’s from a decade where “data conversion” meant accurately parsing email messages and csvs and jsons and whatnot, hence much more “batteries included”; but also clearly going for large-scale with the modula/ada style module system
[17:04] Diana Coman: for the record, the textbook I am talking about is “The AWK Programming Language” by Alfred V. Aho, Brian W. Kernighan and Peter J. Weinberger (my edition is 1988)
[17:05] Diana Coman: and to quote from its very preface, a description which I found to be absolutely and entirely true in practice even today: “The same brevity of expression and convenience of operations make awk valuable for prototyping larger programs. One starts with a few lines, then refines the program until it does the desired job, experimenting with designs by trying alternatives quickly. Since programs are short, it’s easy to get started, and easy to start over when experience suggests a different direction. And it’s straightforward to translate an awk program into another language once the design is right.”
[17:07] Diana Coman: Vivian Sporepress, python got sucked into handling “everything” without any filter or restriction or even definition as to what that everything might be – and yes, exactly as the everything kind of exploded into all-possible-kitchen-sinks (and some impossible ones, too, for good measure).
[17:33] Raphael Nethersmell: basically, python had the unicode derpage baked in even before python3.
[17:37] Diana Coman: pretty much, yeah.
Day changed to 2026-02-03
[14:59] Felicia Solecrimp: omg the npcs are talking !
[14:59] Felicia Solecrimp: they can think !
[15:08] Diana Coman: it’s…possible, though not sure how would it go exactly, “I talk, therefore I think?”
[16:05] Diana Coman: Vivian Sporepress, your linked resource map reminded me that there is one bit of info missing from the current kw resource data, namely the lbn q1 i.e. ‘nothing here’ (since the data collection looks for claims only)
Day changed to 2026-02-04
[15:58] Rainer Tracksnarl: Diana Coman, that’s worth about 8.6k shells more than what you’re offering.
[16:02] Rainer Tracksnarl: Ordinary bundles now available, freshly shipped in though in limited and uncertain supply – get them while they last!
[22:55] Rainer Tracksnarl: Raphael Nethersmell, that’s worth about 34.5k shells more than what you’re offering.
[22:58] Rainer Tracksnarl: Raphael Nethersmell, that’s worth about 8.6k shells more than what you’re offering.
[23:22] Rainer Tracksnarl: Raphael Nethersmell, that’s worth about 226 shells more than what you’re offering.
[23:45] Raphael Nethersmell: I’ve yet to find any ordinaries, but looks like the fomo is gonna cause me to invest. it’ll be hard to know of which resource I’ll find an ordinary, so gotta buy one of each just in case ? hm.
Day changed to 2026-02-05
[08:23] Diana Coman: possibly it depends on how one goes about it – if you go from what you are likely to find, that depends on where you go exploring so if you really want to find just x, the approach would be to explore where there is only x; if you go from what you might need/where the demand for resources might be, you’d look at their use and then basics kind of go into everything anyhow so yeah, possibly one-or-more-of-each is the conclusion anyway.
[16:47] Raphael Nethersmell: makse sense.
[16:50] Raphael Nethersmell: Diana Coman, another refinement on the economic reporting that occured to me this morning was adding mean & median per capita data for PCs and NPCs. I’m curious how I stack up compared to the average NPC. and reporting the mean & median will show the skewness of the populations.
[16:58] Diana Coman: Raphael Nethersmell, that is an interesting idea. How do you see though the different NPCs i.e. the service providers vs non-service providers? And in the longer-term the potential mob/wild npcs aka those attacking on sight rather than exploring?
[16:59] Diana Coman: a note on that increase for the NPCs from this latest economics report: part of it is a recycling of previous environmental damage into the new ord bundles.
[16:59] Ulrich Logfetch: Quoting 2026-02-05 (Notices and Announcements) Ulrich Logfetch: ** Euloran economic activity during one month to day 00 of month 05, year 6 (era 1): PCs’ held value in items increased by 8`720 and their cash decreased by 2`793; NPCs’ held value in items increased by 1`394`023 and their cash increased by 2`793; environment value decreased by 418`476. All amounts are shown in Euloran Shell (10`000 Shells to 1 Copper). **
[17:05] Diana Coman: possibly of interest as a measure, the decrease in environment value is generally what PCs and NPCs together carved out of it one way or another at some point. On the other hand, what said PCs and NPCs managed to hold/make out of it would be the reported increase, so a sort of efficiency measure there, even if quite a rough one.
[18:33] Felicia Solecrimp: Wolfgang Middlesit where are you?
[18:33] Wolfgang Middlesit: Felicia Solecrimp, I am currently at -22.86 99.87 -18.81 in Preposterously Prosperous Place
[18:37] Felicia Solecrimp: my first awesome bought i have been working hard and i got my first own bought spicy mousaka
[18:37] Felicia Solecrimp: and I got 800 crumbly rock in one claim
[18:48] Diana Coman: oh hey, that 800 cr in one claim souns like a mini-pop, congrats Felicia Solecrimp!
[18:48] Diana Coman: enjoy the moussaka, it should get you a few euloran months respite iirc
[19:36] Raphael Nethersmell: Diana Coman, interesting re efficiency. wrt distinguishing npcs, the difference matter in terms of pcs interactions with them, but bottom line is the all need to eat. I expect some wil be more active than others. the first question when looking at the reported data that prompted the suggestion was, I have a good idea how many pcs there are, but well, how many NPCs are there actually ? i’m pretty sure there are some I’ve never glimpsed, especially the relatively little time i’ve spent in ppp.
[19:36] Raphael Nethersmell: Felicia Solecrimp, buen provecho !
[19:57] Raphael Nethersmell: like, approximately how much are V pcs getting ass kicked by npcs.
Day changed to 2026-02-06
[09:27] Diana Coman: heh, V pcs getting ass kicked, aptly put. Raphael Nethersmell, there are currently 11 NPCs in total, including all service providers. Perhaps the number as such may be included in the reports, though I’m not quite sure about the mean and median, I’ll give it some more thought.
[09:31] Diana Coman: The measure you are describing sounds like a sort of gdp per capita and that would be interesting, certainly, if one could get it right (or at least right-enough). And when this right-enough is not so clear, I’m more inclined to let people decide and calculate their own metrics as they see fit rather than stamp one as such through the report. It’s not that clear just how much (and even what, exactly) the average says. The median is likely to be at least more meaningful but it’s also way more expensive to calculate and in time it will get only worse. And calculating the median comes with potential privacy concerns if done centrally i.e. rather too close to individual monitoring.
[09:35] Diana Coman: meanwhile in zig-bot testing, it clearly needs some more robust means to detect work in progress but with those in place, it does reasonably fine. Though it also pushes, of course, for a more discerning use of comms clientside, so I suspect I’ll have to see to that sooner rather than later. Sigh.
[09:36] Diana Coman: and now I got some 150 cdg out of 2 claims, so my char is overbulk and stuck, hmpft
[09:38] Diana Coman: I guess I’ll have to add table-dragging to the bot’s capabilities, too
[15:34] Rainer Tracksnarl: Diana Coman, that’s worth about 1.1k shells more than what you’re offering.
[18:08] Raphael Nethersmell: Diana Coman, makes sense. my interest was my mainly regarding how I’m doing relatively. i think reporting population of pcs and npcs would be sufficient for now, as you said, let people make their own calculations.
Day changed to 2026-02-07
[03:47] Raphael Nethersmell: Diana Coman, meant to say, the bot ziggurating is so pleasant once once I get the hang of it. one thing it has changed in my gameplay is it have caused me to mix my lbn in bulk, which I hadn’t been doing while manually ziggin’. I had attempts like : 6 lbn q11 + 4 lbn q1 + cr(e) q2 -> 21 cr q4 or 5 lbn q11 + 3 lbn q1 + cr(e) q2 -> 25 cr q3 ; now i’ll do 100 lbn q11 + 40 lbn q1 -> 140 lbn q8. and then just zig with stacks of q8, adjusting the amount as needed via euconfing.ini
[03:51] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4198 — I recommend finding a pactch of bbb, which are very low weight and bulk and notbag.jpg value too !
[03:51] Ulrich Logfetch: Quoting 2026-02-06 (Eulora 2) Diana Coman: and now I got some 150 cdg out of 2 claims, so my char is overbulk and stuck, hmpft
[03:54] Raphael Nethersmell: pro tip : if you mistakenly enter a line meant for chat into your client console, just go to your terminal and issue : `tail data/logs_console.txt` and voila, you’ll have it available to copy and paste in the actually chat buffer !
[04:02] Raphael Nethersmell: one way I’ve been “dicerolling” for spots to explore is, if I’m going accross a sector w/ a relocate. sometimes it doesn’t get me to the final destination in one go, speed limit is enforced. issue a single or botted explore wherever it stops me.
[04:30] Raphael Nethersmell: watching just tail -f data/logs_bot_diary.txt , no gui, while running a spree of ziggin’ or exploring gives a window into how text-only client might feel.
[10:48] Diana Coman: Raphael Nethersmell, very glad to hear the bot zig is used and useful, thanks! The bbb recommendation makes total sense indeed, except so far I don’t think I found yet a patch of that (though I did find one of wwb somewhere far away and these are very small and light too, only less valuable). Overnight I let the bot zig+explore+train in the pss/nt patch in ppp and it did ok really, got 1 level further in zig, some 50 pss, some NT and quite a bunch of further ideas as to what the zig still needs to be more robust about.
[10:52] Diana Coman: the text-only client feel of that bot_diary watch is quite interesting really. Conversely, I was noticing the other day just how much the bot makes the GUI delay visible – if one looks at the bot’s actions, it’s always ahead of the graphics, basically the graphics keep playing a sort of catch-up game with the actions.
[11:00] Diana Coman: which points perhaps to a more fundamental shift in there: while the traditional/common/widespread/natural approach is to have the client/gui show “the world as the server/game/authority says you are meant to see it”, what I’m really after is turning this approach on its head, as I’d much rather have, if possible at all, the client/gui show what one does, what their impact is and ultimately the world as it emerges from the interaction with it, perhaps as “discovered”, with the server’s objects as conventions of potential illustration, not more.
[11:03] Diana Coman: ofc I hadn’t quite known just how deep my unhappiness with the usual gui went, it took this much just to start getting a glimpse as to what the deeper trouble is and to start even being able to put it into words, as it tends to happen.
[11:08] Diana Coman: Raphael Nethersmell, on the pick of exploration points, plotting all that resource data tends to be quite helpful as it makes it very easy to start systematically getting to know where things are and one can of course start with very coarse granularity (like Vivian Sporepress did for that mini-map he made with his own bot) and then add detail when + where + as much as wanted
[23:58] Felicia Solecrimp: Diana Coman Can I ask you your advice on whether a hosting service should optimize into using ai features and how can it be better to provide human interaction than machine bots and if whether a decetrilize hosting is actually a good idea ?there are free hosting services for students and developers but aside of ecommerce and streaming websites which coud be the most succesful way in order to offer for a higher price or what could be our differentitator ?
[23:59] Felicia Solecrimp: because i also notice that all hosting services monthly are really low but they increase of price when they show more features .
Day changed to 2026-02-08
[00:02] Felicia Solecrimp: Dedicated Hosting Dedicated hosting means renting an entire physical server just for your site(s). Like owning a house instead of renting an apartment. I think this could be the closest to offer 100 for a hosting server that generates revenue option that i found also
[00:02] Felicia Solecrimp: with this information im looking forward for your opinion
[00:06] Felicia Solecrimp: but i still believe in the importance of human contact for client service because is something that is lacking somehow with all of the automatization now days too
[00:21] Felicia Solecrimp: known host is 143 and inmotion is 110
[00:23] Felicia Solecrimp: ionos is also very popular
[00:24] Felicia Solecrimp: but is very cheap and it has personalized attention
[00:25] Felicia Solecrimp: inmotion has 5 channels in order to talk to people for service
[00:27] Felicia Solecrimp: plenty methods of payment is something else i notice
[00:28] Felicia Solecrimp: a free domain if you register for one year
[02:15] Vivian Sporepress: Felicia Solecrimp: those sorts of data points you should gather into an overall, structured document so people can read at their convenience if they’re even interested – or at least just send them to me privately, though it’s mostly for your own orientation since I’m already somewhat familiar with that market
[02:16] Vivian Sporepress: “decentralized hosting” is still a new buzzword to me, any idea what it might possibly mean?
[02:22] Vivian Sporepress: I think you have a good point that the human factor is a strong differentiator, at least potentially, since I certainly won’t be implementing chatbots for “support” or whatever such nonsense. moreover, with my service the customer gets direct access to the engineering & management behind it, not some powerless call center. the question is if that will be enough and if we can communicate it effectively to the target customer
[02:27] Vivian Sporepress: and you mention payment methods – how many will take direct btc payments (and not through some flaky gateway at best)?
[02:34] Vivian Sporepress: another point worth internalizing is that because of strong identities, we’re practically immune to the biggest threat at other service providers – social engineering: http://trilema.com/2014/o-hai-let-me-verify-your-identity/
[09:05] Diana Coman: Felicia Solecrimp, why don’t you do this properly and exactly as Vivian Sporepress suggest, put together everything you’ve got so far in a structured, coherent piece of writing so that both you and anyone else can read it, add to it, comment on it and even have it as reference at a later point? There is a point to each of these steps and no gain in skipping them. Moreover, then you can simply give a link and have any conversation happen or at least bridged there rather than pushing it here where it doesn’t really quite belong in full anyhow.
[11:20] Diana Coman: on bbb patches, looking back through the logs pre-bot, it turns out that all bbb I found was in PPP, only entirely overlapping with other resources it seems
[11:25] Raphael Nethersmell: Diana Coman, any interest in a trade of bbb patch cords for wwb patch cords ?
[14:40] Diana Coman: Raphael Nethersmell, sure, why not. The one I know is quite far from PPP though, just to make it clear.
[19:26] Vivian Sporepress: newbie question – how does the bot zig pick what quality inputs to use?
[19:57] Diana Coman: Vivian Sporepress, by default it goes with 50% meaning in the middle of the interval for each ingredient. This can be set via parameter in the .ini file (it’s only one parameter though so same for all ingredients).
[19:58] Diana Coman: as for quality, it doesn’t care as such, it’s up to the player to have in inventory what they want used rather than what they don’t want used. And if they want something specific, the way to go about it is to make and have bundles – the bot will use bundles first whenever available for the zig at hand.
[20:00] Diana Coman: should add that there is quite on purpose a loophole to “hide” things from the bot (e.g. if you don’t want them used): the bot looks only at stuff directly in inventory, not those inside container inside inventory, so if you want to keep things away from it, just put them in a table and pick that table up.
[21:31] Vivian Sporepress: that’s a nice trick.
[21:32] Vivian Sporepress: the bot’s ziggurating struggles appear quite reminiscent of my own… still waiting for stuff to update, overshooting because it seemed a move didn’t go through, aborting impossible work due to incomplete info, finally everything’s lined up and still nothing happens…
[21:36] Vivian Sporepress: some of my claims were just now expiring, though seemingly not the one I’m trying to zig; only guess atm is that it’s “trying” to expire but can’t because I’m standing right here, but that still somehow prevents me from building it
[21:37] Vivian Sporepress: ahh no, it’s because my enum readings changed since last checked, so much that even 50% is now over 100%
Day changed to 2026-02-09
[08:23] Diana Coman: yes, the whole of e2 is quite calling for nimbleness rather than blind data crunching hence the troubles poor bot has. Even so, there are quite a few parts where the bot zig could attempt to improve robusteness for sure. If you want to avoid entirely the potential block of enums changing contents while the bot is trying to do something, the bundles are again the answer (since it’s always and by definition ONE bundle, that part doesn’t change)
[11:56] Diana Coman: given that the bot zig job is clearly of interest, I’ve deployed now the cutting edge client version as well. In particular, it should deal much better with any changes in the enum while it’s trying to zig, since now it just re-reads it every time so that it can adjust if needed. There are further changes and checks to hopefully become more quickly aware of items of interest and generally to be better able to deal with the unexpected.
[11:56] Ulrich Logfetch: Quoting 2026-02-09 (Notices and Announcements) Monroe Quarterslit: ** Euloran news: newly deployed client responds to players’ feedback on the bot zig job adding further checks to react more nimbly to changes in enum readings and to improve robustness of bot jobs in general. Furthermore, this version looks for and records LBN as a resource in its own right as and when seen in the world (hence available in the resource knowledge maps afterwards). **
[11:58] Diana Coman: there is one added parameter that can be set to request the zig job to keep track of its starting pos and return to it, though by default this is set to false aka no return (the behaviour as it has been so far). While it can be a waste of time and quite annoying to keep going back and forth if/when one is ziggurating an existing bunch of claims, it can be equally annoying to have a zig job mess up one’s exploration grid or similar pos-related concerns, hence this new option via zig_ret
[12:01] Diana Coman: I’ll add explicitly that these changes are very new and there was very little time to put them thoroughly to test in practice, so if you do update to this latest version, do let me know how it works for you
[16:15] Vivian Sporepress: I’ll try it
[16:19] Diana Coman: cheers
[16:26] Vivian Sporepress: bot none
[16:26] Diana Coman: heh, the multi-input troubles
[16:26] Vivian Sporepress: lol whoops. meant to say: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4199 – what’s that? I can imagine stacking heavy items into a craft table but then not sure how I’d move that without being again over limit
[16:26] Ulrich Logfetch: Quoting 2026-02-06 (Eulora 2) Diana Coman: I guess I’ll have to add table-dragging to the bot’s capabilities, too
[16:29] Diana Coman: it’s a trick from back in e1 and yes, e2 allows it as well: while you can’t move carrying the heavy table, you can drop it, move a bit, pick it up again, repeat. Hence table-dragging and yeah, a pain by hand, though relatively easy and certainly painless by bot (and if you are ever far away when some claim pops with a ton of something, you’ll likely prefer to drag the thing instead of potentially losing it if you leave it there, not to mention it may even take more than one trips to carry it at your char’s capacity)
[16:31] Diana Coman: good old foxybot even used a table at all times, stacking into it whatever it got from the claims and carrying it around so it never got overweight. Plus it allowed one to carry with them more lbn/bundles to build with, there is that side of it, too.
[16:34] Vivian Sporepress: so we have wagons, just no wheels, neat
[16:39] Diana Coman: we have heels not wheels!
[16:40] Diana Coman: though e2 pushes this one further since there is the upper limit of what your char can *lift*. So if the poor bot were to try and fully solve this, it would end up juggling any number of tables, sheesh.
[16:48] Vivian Sporepress: a first bot zig on latest client again overshot on loading lbn and had to move some back; a second somehow found work already ongoing and finished in 3 reps; a third tried to load a bundle instead of lbn and aborted on failing to find such. Overall it still works eventually though
[16:49] Diana Coman: hm, it would normally try to load a bundle only if it found one present
[16:50] Diana Coman: by a second one do you mean you stopped the restarted the bot or just the 2nd time it got to the zig job in a sequence?
[16:50] Vivian Sporepress: second automated run, no manual restart
[16:52] Vivian Sporepress: so far at least it hasn’t hit what seemed the biggest stumbling block yesterday, “aborting impossible work: Missing/incomplete info (stack) for obj …” where that obj was some lbn that no longer existed
[16:53] Diana Coman: anyways, noted to have another look at the checks for the bundle and what it can do to make sure there is one before going for it. Otherwise though there will always be so much that can go wrong/differently/unexpectedly that the real goal is really to still work eventually, not as much some smooth sequence which is very nice while it lasts but then too fragile for what this environment is
[16:53] Diana Coman: yes, that is one thing that I specifically fixed ie if/when it finds some info missing, it will report it as such but request the object so it finds out that it’s gone and otherwise keeps trying rather than bailing out
[16:58] Vivian Sporepress: nice
[16:59] Vivian Sporepress: perhaps it thought I had a bundle because it grabbed the claim marker contents and hadn’t yet noticed the bundle there turning into the item
[17:23] Vivian Sporepress: Diana Coman: looks like euconfig.ini could use the pct entry under BOT-BUNDLE. perhaps also a refresh of ServerName and Comment
[21:27] Diana Coman: Vivian Sporepress, indeed it can, I made the changes now to euconfig.ini, thanks.
[21:28] Vivian Sporepress: thanks to the high volume of time observations now handily provided by the bot diary, I’m finally inferring that Euloran time consists of… months 0-11, days 0-23 and hours 0-19. can’t say yet if the ranges are context-dependent too, how many total days in the year and such
[21:31] Felicia Solecrimp: I found a crafting table nearby rainer
[21:31] Felicia Solecrimp: I told vivian I did not pick it up because i thought it was somebody else table
[21:37] Vivian Sporepress: and I told her that was an ethical mistake, since the game will either enforce ownership or else zap it altogether if nobody grabs it
[21:39] Felicia Solecrimp: By the way sorry for spouting all together the content of the hosting and not making it available in a more organize contetx I think I need to think before acting and how will this repercute on the game otherwise
[21:39] Diana Coman: yeah, feel free to try and pick stuff up, Felicia Solecrimp
[21:39] Felicia Solecrimp: act least on the chat
[21:42] Diana Coman: awareness of context helps at all times, so if it helps towards that, no harm done.
[21:52] Vivian Sporepress: it looks like indeed all months have an even 24 days; I’m just missing the full range from m7 but it’ll be coming back around soon.
[21:59] Vivian Sporepress: and it looks like a quite uniform 20 hours per day.
[22:13] Vivian Sporepress: since I’ve been on this revisit of chronometry & chrononomy. otherwise, now on the hunt for the abandoned eggs… I stumbled on some right here in PPP, in fact two close by each other, just wasn’t quick enough to notice & build them, and it’s not a reliable patch.
Day changed to 2026-02-10
[16:31] Diana Coman: the first PC mention of an euloran almost-complete calendar!
[16:31] Diana Coman: and yeah, I even got that Purple Snail still in PPP but never managed to repeat the catch
[16:32] Diana Coman: the AE are quite heavy though iirc
[17:28] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4241 — the main inefficiency I’ve noticed with zigging is loading multiple stacks of lbn into a single enum. e.g. if it’s set to 8, it’ll load 2 stacks of 8 and then remove one before it tries to build.
[17:28] Ulrich Logfetch: Quoting 2026-02-09 (Eulora 2) Diana Coman: yes, the whole of e2 is quite calling for nimbleness rather than blind data crunching hence the troubles poor bot has. Even so, there are quite a few parts where the bot zig could attempt to improve robusteness for sure. If you want to avoid entirely the potential block of enums changing contents while the bot is trying to do something, the bundles are again the answer (since it’s always and by definition ONE bundle, that part doesn’t change)
[17:54] Diana Coman: Raphael Nethersmell, that would be only because of the data lag as it were – it doesn’t *try* to load 2 stacks, only one, but the trouble is that when it looks again it seems as if the attempt didn’t take, hence it tries again. Possibly increasing the delay/check time might help in that it’s more likely to be up to date by the time it looks again but otherwise there isn’t much the bot can do currently to avoid this sort of trouble.
[17:55] Diana Coman: maybe it could assume that it worked the first time and issue an action request at least when there was only one move left, though dubious
[17:57] Diana Coman: this might perhaps get addressed closer to the root only when the client gets an actual strategy for data refresh/request, but it’s certainly not trivial
[17:59] Diana Coman: meanwhile in possibly more worrisome misfires, it seems that there are cases when some of the outputs don’t make it in time either and so they remain in the claim/end up on the ground afterwards. Perhaps it’s high time to add a pickall (giving also how people kept trying to use takeall for that, lolz)
[19:20] Raphael Nethersmell: Diana Coman, ah, makes sense. I’ll trying increasing the poll_beat on zig to 30 for now and see how it fares.
[21:47] Vivian Sporepress: is it possible the zigbot simply isn’t refreshing the claim marker after requesting the loading of ingredients? not sure then why it would still work sometimes, but I found when I do a reqobj on the marker after its move, the lbn tends to pop right up.
[21:47] Vivian Sporepress: and yes, I noticed items falling on the ground after claim expiry too.
[21:50] Vivian Sporepress: also I note that the explore bot is still prone to targetting non-explorable entities based on whatever was last clicked, to be reset by examining the ground.
[21:52] Vivian Sporepress: (not meaning to be a nag – I just thought I’d read at some point that this had improved)
[22:06] Diana Coman: Vivian Sporepress, no, it’s not possible that it doesn’t refresh, but it IS possible that the refresh ends up “too soon” as it were OR incomplete, hence a second one you send manually sorts it all out
[22:09] Diana Coman: the explore bot normally targets whatever is set as target, and that is by default what you look at, so you do need to mind that before starting the bot. During a run it shouldn’t *change* the target though (so it’s safe to examine whatever). I could just force-set it to none/ground/sector but I’d rather avoid that if possible, simply because the explore is one main type of action, potentially applying to anything. If it really gets too annoying, I suppose I could add a parameter to toggle on/off “no target” in general even, might at least address the trouble while still retaining some flexibility for the longer term.
[22:11] Diana Coman: (and yes, you remember correctly that it improved there – in that it keeps the target rather than ending up with it changed mid-run as it used to happen, which was worse)
[22:14] Felicia Solecrimp: Rainer Tracksnarl where are you?
[22:14] Rainer Tracksnarl: Felicia Solecrimp, I am currently at -18.80 97.41 35.56 in Preposterously Prosperous Place
[22:14] Felicia Solecrimp: Rainer Tracksnarl Where are you ?
[22:18] Diana Coman: should add that the trouble with “send a 2nd request” is that in practice it doesn’t guarantee anything by itself, for the same money could send 3 each time or any other number. The only thing that could perhaps help is literally a higher level layer that keeps track of what is *wanted* and generates requests according to whatever rules it has to get that, so quite the additional logic there and likely non-trivial to reliably get smooth results and be essentially invisible.
Day changed to 2026-02-11
[16:57] Diana Coman: following up on the target for explore, I’ve added now a parameter in the .ini file to have explore always set its target at 0 and thus be entirely immune to any target sets by the GUI whether before or during a bot run. Note that the default for this null_target parameter is actually 0 aka still using the GUI-set target so that the behaviour doesn’t change unless you explicitly set this parameter in you ini file. As I gather that the issue has been quite a nuisance, I’ve redeployed the client so that this change is available.
[16:57] Ulrich Logfetch: Quoting 2026-02-10 (Eulora 2) Diana Coman: the explore bot normally targets whatever is set as target, and that is by default what you look at, so you do need to mind that before starting the bot. During a run it shouldn’t *change* the target though (so it’s safe to examine whatever). I could just force-set it to none/ground/sector but I’d rather avoid that if possible, simply because the explore is one main type of action, potentially applying to anything. If it really gets too annoying, I suppose I could add a parameter to toggle on/off “no target” in general even, might at least address the trouble while still retaining some flexibility for the longer term.
[19:29] Raphael Nethersmell: Felicia Solecrimp, you know you’re exploring right in poor Rainer Tracksnarl’s lap ? mind giving some space ?
[19:30] Raphael Nethersmell: exchange Rainer Tracksnarl
[19:37] Raphael Nethersmell: I could see if he was sitting in a patch of wm or eps or pb, but cr isn’t so hard to find, you know ?
[19:46] Raphael Nethersmell: anyone have lnb q1 they’d be interested in selling me ?
[19:57] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4232 — understood, voila.
[19:57] Ulrich Logfetch: Quoting 2026-02-08 (Eulora 2) Diana Coman: Raphael Nethersmell, sure, why not. The one I know is quite far from PPP though, just to make it clear.
[20:30] Diana Coman: Raphael Nethersmell, nice, here’s mine
[21:35] Raphael Nethersmell: got it, thanks.
Day changed to 2026-02-12
[03:22] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4300 – thanks, will try.
[03:22] Ulrich Logfetch: Quoting 2026-02-11 (Eulora 2) Diana Coman: following up on the target for explore, I’ve added now a parameter in the .ini file to have explore always set its target at 0 and thus be entirely immune to any target sets by the GUI whether before or during a bot run. Note that the default for this null_target parameter is actually 0 aka still using the GUI-set target so that the behaviour doesn’t change unless you explicitly set this parameter in you ini file. As I gather that the issue has been quite a nuisance, I’ve redeployed the client so that this change is available.
[03:24] Vivian Sporepress: http://ossasepia.com/2026/02/02/Notices-and-Announcements-Logs-for-Feb-2026/#197 – sounds like the sustainability & environmental protections programme is having some effect, and profitable one at that
[03:24] Ulrich Logfetch: Quoting 2026-02-11 (Notices and Announcements) Ulrich Logfetch: ** Euloran economic activity during one month to day 00 of month 06, year 6 (era 1): PCs’ held value in items increased by 27`672 and their cash increased by 29`481; NPCs’ held value in items increased by 46`689 and their cash decreased by 29`481; environment value decreased by 449`403. All amounts are shown in Euloran Shell (10`000 Shells to 1 Copper). **
[03:28] Vivian Sporepress: some fresh personal ziggurating records: 1 cr q75, 1 cr q86, 1 cr q93, and 1 lonely nt q98
[03:44] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4305 – I could, but can’t compete price wise with the merchant’s npc-powered supplies
[03:44] Ulrich Logfetch: Quoting 2026-02-11 (Eulora 2) Raphael Nethersmell: anyone have lnb q1 they’d be interested in selling me ?
[04:11] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4202 – there’s a little algebra to find the ratios to do these mixes without rounding loss, short of trekking over to EOJ…
[04:11] Ulrich Logfetch: Quoting 2026-02-07 (Eulora 2) Raphael Nethersmell: Diana Coman, meant to say, the bot ziggurating is so pleasant once once I get the hang of it. one thing it has changed in my gameplay is it have caused me to mix my lbn in bulk, which I hadn’t been doing while manually ziggin’. I had attempts like : 6 lbn q11 + 4 lbn q1 + cr(e) q2 -> 21 cr q4 or 5 lbn q11 + 3 lbn q1 + cr(e) q2 -> 25 cr q3 ; now i’ll do 100 lbn q11 + 40 lbn q1 -> 140 lbn q8. and then just zig with stacks of q8, adjusting the amount as needed via euconfing.ini
[04:14] Vivian Sporepress: say you mix stacks N of qual X and M of qual Y; you get a stack N+M of qual floor[(NX+MY)/(N+M)]
[04:22] Vivian Sporepress: call that output quality Z, assume no rounding and multiply by the denominator: NZ+MZ=NX+MY. by rearranging and dividing, you get N/M = (Y-Z)/(Z-X). plug in your desired Z and available X and Y and there you go
[04:32] Vivian Sporepress: for your example, Z=8, X=1, Y=11 so N/M = 3/7. you want 140 so mix 42 lbn q1 with 98 lbn q11
[08:21] Diana Coman: heh, Vivian Sporepress got it on the mixing, applied algebra ftw! In e2, people had even a shorthand for mixing, simply picking stacks of same count and same *parity* wrt quality (since that ensures that the division has remainder 0 hence no loss), potentially repeating if/as needed. E.g. mix 7 at q5 with 7 at q99 OR 7 at q10 with 7 at q200
[08:21] Diana Coman: meant to say in e1*, ofc
[08:23] Diana Coman: Raphael Nethersmell, fwiw I’ve let the bot overnight to map that bb spot from you and it seems to cover fine a whole 100×100 square (starting to get into lbn on the edges)
[08:23] Diana Coman: on the q1 lbn, I guess it’s currently more difficult for people to get that rather than the q11 since the bot is not aware of lbn as output of an explore request.
[08:25] Diana Coman: and its value is so low as to make it hardly worth much effort really, lbn is 11 shells at q100 so at q1 you need 10 of them just to make 1 single shell, it’s more of a shell-cent than anything, lolz
[08:30] Diana Coman: Vivian Sporepress, wrt merchant prices, they are meant as a reliable baseline only so I wouldn’t even expect that PCs want to trade among them at NPC prices. And as time passes, there is no guarantee that Rainer Tracksnarl keeps the same stock anyhow.
[09:23] Diana Coman: interesting data points on the zig, too, any ideas as to what might be explaining them or how to get such high q more reliably? While high quality means higher value anyhow, there might even be a premium for it since it’s rather rare currently, dunno.
[09:24] Diana Coman: meanwhile in my own gathering/ziggurating experience, my enum q went again down but worse really, not just to q3 from the “high” of q4 but again back to q2, apparently easier to slide down with it than to get anywhere higher.
[13:53] Raphael Nethersmell: in unexpected but welcome outputs, I got an sms q1 out of a ziggurating attempt.
[13:57] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4315 — when I looked at Rainer Tracksnarl’s inventory yesterday I didn’t see any supply.
[13:57] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4305 – I could, but can’t compete price wise with the merchant’s npc-powered supplies
[14:06] Diana Coman: ha, that would be first reported instance of an activity yielding a skill-item, congrats Raphael Nethersmell!
[14:06] Ulrich Logfetch: Quoting 2026-02-04 (Notices and Announcements) Monroe Quarterslit: ** Euloran news: a well-placed source has it that some activities may now yield on occasion skill-giving items, too! Who will be the first to get any such item and what even are the new skills thus available? **
[14:07] Diana Coman: and it occurs to me that Vivian Sporepress is saying there that the 20% markup at Rainer Tracksnarl is …too small?!
[14:11] Raphael Nethersmell: I’d definitely beat Rainer Tracksnarl’s price.
[14:12] Raphael Nethersmell: Diana Coman, it’s actually the second one I’ve got, but I was away during the first one and wasn’t exactly sure how it emerged, this one I witnessed live.
[14:13] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4324 — yeah, I know they’re quite cheap, I just seem to get better zig value with lbn q < 11 so far.
[14:13] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Diana Coman: Raphael Nethersmell, fwiw I’ve let the bot overnight to map that bb spot from you and it seems to cover fine a whole 100×100 square (starting to get into lbn on the edges)
[14:14] Raphael Nethersmell: that is, total return per “click”, i.e. I get lower q outputs, but more of them.
[14:19] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4324 — most bodacious ! i’ve only been drunking moving on the wwb so far, but have zigg’d over 1.1k with bulk and weight capacity to spare.
[14:23] Diana Coman: sounds like you found quite the sweet spot of ziggurating there, getting 2 skill items already and 1.1k wwb, apparently there is plenty to be done at a profit, too
[14:23] Diana Coman: in terms of return per click, you’d need to calculate the qabv really since count is not all there is to it (although it does sound like you are getting more than you put in, in real terms)
[14:27] Diana Coman: as to the lbn q1 supply as limiting factor, it sounds quite very euloran in style really – as soon as one finds some sort of “guaranteed money-making”, they find out that they soon run out of something or the other
[14:28] Raphael Nethersmell: yeah, looks like it’s been worth flipping through all the maps to arrive here. ~40% of that 1.1k is q1 from 1 small claim. I burned pretty much an entire ch to get 2 smalls.
[14:29] Diana Coman: that sounds quite familiar from e1 too, namely you barely get anything for whatever number of attempts and then you get a ton from a lucky hit (it took though all the other ones to get there though)
[14:30] Diana Coman: kind of curious if you’d get more smalls on a lower value resource (since you say the zig gives you either more at lower q or fewer at higher q, then possibly there would be something similar with the claims – if one manages to figure out how would that apply to the claims in the first place, exactly)
[14:32] Raphael Nethersmell: what are the arguments the l parameter accepts for the train command ? I greped lesson on the src dir and only yielded hits in client/euaction.cpp ; in that file I don’t see any mention of what arguments it takes. so far i’ve tried gather, gathering and explore but the console log reports “lesson 0”
[14:33] Raphael Nethersmell: Diana Coman, aha, I did get more smalls on exploring a cdg patch with ch.
[14:34] Diana Coman: Raphael Nethersmell, the l aka lesson is the training session itself, hence the ID of the training session object or 0 if/when there is no such thing (ie when asking for a new one)
[14:35] Raphael Nethersmell: ok
[14:35] Diana Coman: all the interaction commands work like this, referring to the session object itself (and 0 is in general the “null” value)
[14:36] Diana Coman: quite interesting then, maybe if one gets more and/or skill items out of ziggurating small claims, it really is a better use of ch currently on lower q resources than on higher q ones
[14:36] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4346 — “Gravitation is not responsible for your great tumble – after all it tried to prevent your climb every step of the way.”
[14:36] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Diana Coman: that sounds quite familiar from e1 too, namely you barely get anything for whatever number of attempts and then you get a ton from a lucky hit (it took though all the other ones to get there though)
[14:37] Diana Coman: hehe, quite so
[14:39] Raphael Nethersmell: Diana Coman, on your new blog setup was there some unicode conversion of text, e.g. on my gentoo I see “great tumble<80><93> after all” in the log instead of –
[14:40] Diana Coman: Raphael Nethersmell, ah, I get now what you were after with that l parameter but it doesn’t work like that, heh. For the record, on any protocol level command, it’s the protocol spec where you would find more information about what it takes and how it works. As for *what* to train, it simply depends on what the trainer *can* train you in, so a matter of what they know. If you noticed in the training window, it says what the trainer can train and up to what level, as well.
[14:41] Raphael Nethersmell: I also notice some curly quotes in the broswer’s rendering of the log, but those don’t appear in my txt file.
[14:41] Diana Coman: uhm, no explicit unicode conversion but I’ll have a look – fwiw here on centos 6 I see the – fine
[14:43] Raphael Nethersmell: I didn’t mean to imply accustion of the crime of explicit unicode conversion, ftr. lulz.
[14:43] Diana Coman: lolz, no, it’s fine, I’ll add it to the list to double check there, thanks for letting me know
[14:44] Raphael Nethersmell: noted re referencing the spec, thanks.
[15:52] Diana Coman: adding to those zig data points, I got 36 cdg at q227 even, from a bare-handed small, though sadly no skill-item so far.
[16:11] Raphael Nethersmell: how many cft did you use and what quality ?
[17:10] Diana Coman: heh, the right question asked: 31 cft, q61 (that’s what I had in storage)
[17:11] Diana Coman: pondering here if the bot shouldn’t take advantage of all the infrastructure to do a before/after list of things in one’s inventory and log that for each action, perhaps
[17:11] Diana Coman: log the changes, I mean, not the full list each time
[18:27] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4331 – oic. and I find I’ve used up most of mine; I have in stock 5 q1 and 85 q2, asking 10 shell apiece for any of those. or I can just point you to where to dig for them
[18:27] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4315 — when I looked at Rainer Tracksnarl’s inventory yesterday I didn’t see any supply.
[18:28] Vivian Sporepress: I guess the exploring NPCs haven’t been in the lbn patches for a while, and we previously benefitted from an initial glut from before they knew how to trade.
[18:34] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4328 – haven’t found the secret; so far it’s been just doing more, by bot or otherwise and letting the details vary, though lately I’ve been going with straight lbn q11.
[18:34] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Diana Coman: interesting data points on the zig, too, any ideas as to what might be explaining them or how to get such high q more reliably? While high quality means higher value anyhow, there might even be a premium for it since it’s rather rare currently, dunno.
[18:37] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4359 – myeah, looks like the verschlimmbessert wordpress frontend is autocorrecting quotes, hyphens into en-dashes, double-hyphens to em-dashes, three dots to unicode ellipsis and so fifth. for extra lulz, my fabulous ubuntu fails to paste text from browser to terminal *at all* when it contains such magic byte sequences
[18:37] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Raphael Nethersmell: I also notice some curly quotes in the broswer’s rendering of the log, but those don’t appear in my txt file.
[18:40] Vivian Sporepress: ah, that’s xfce4-terminal to blame, classic xterm pastes obediently with some kind of garbage-in garbage-out.
[18:41] Vivian Sporepress: xfce’s ‘mousepad’ text editor has been similarly broken, refusing to save a file at all if there’s some encoding failure
[18:41] Vivian Sporepress: quite python3 style
[18:47] Felicia Solecrimp: Rainer Tracksnarl where are you?
[18:47] Rainer Tracksnarl: Felicia Solecrimp, I am currently at -18.80 97.41 35.56 in Preposterously Prosperous Place
[19:43] Raphael Nethersmell: Felicia Solecrimp, you know he hasn’t moved in like… weeks ? better to note the spot and return to it and only if he’s not there, then ask in the log.
[20:48] Raphael Nethersmell: just when I think I’m about to get overbulked, a bunch of keys turn to lbn and lighten me up !
[02:41] Wolfgang Middlesit: Raphael Nethersmell, that’s worth about 18.6k shells more than what you’re offering.
[02:43] Wolfgang Middlesit: Raphael Nethersmell, that’s worth about 2.5k shells more than what you’re offering.
[02:46] Wolfgang Middlesit: Raphael Nethersmell, that’s worth about 61 shells more than what you’re offering.
[03:05] Raphael Nethersmell: pro tip I shared with Vivian Sporepress and might as well share w/ the log : if you’re building multiple claims of the same resource, you can load up the 10th slot in your inventory with however many enums you want*, as you ziggurate each claim, it’ll decrement the count by one. this saves you from having to click & cluck enums over to the 10th slot each run, which apart from saving you time & effort, saves server traffic.
[03:05] Raphael Nethersmell: *if there’s an upper limit, i’ve not found it yet.
[08:13] Diana Coman: well noted, Raphael Nethersmell and it works for any action that needs something in that slot, whether use or explore on whatever it is. The limit is likely only the maxint upper limit of what can go in a single stack.
[08:14] Diana Coman: meanwhile it turns out that my char ends up overbulk faster than overweight and with …grass of all things, huh.
[08:14] Diana Coman: apparently the trouble with keeping keys is that they are way bulkier than the corresponding lbn
[10:38] Diana Coman: talking on ziggurating claims, there is now available the bot zig job that goes through the keys in one’s inventory, picks one with a claim that is currently visible, with quality above given threshold and for which there is also the corresponding enum + bundle/ingredients, proceeds to the claim’s location, equips enum, moves ingredients to claim (using % as given), requests use on it, takes results from claim as it finds them.
[10:38] Ulrich Logfetch: Quoting 2026-02-02 (Notices and Announcements) Monroe Quarterslit: ** Euloran news: newly deployed client adds bot zig and bot bundle jobs, as well as a visible warning when one’s own char in game is overweight or overbulk (as the client knows it at the time). While the new zig job is known to handle successfuly all the intricacies of ziggurating claims, perfection does not exist and how effective and/or efficient it may prove to be for one player or another depends as well on how well the player understands what they are doing and thus how well they put it to use. So now that there is at least one automation available for each of the more intricate in-game activities, the question is: what will PCs do with it all?**
[10:42] Diana Coman: obviously, given the number and intricacies of steps in there, it’s quite the job and there are plenty of things that can go wrong – on occasion they do go wrong, too! Generally from the relatively short-time (couple of days) testing I got to put it through, it recovers by itself just fine, only it can take a while/end up repeatedly moving stuff back and forth – so if you do mean to use it, it’s probably best to see how it works in your specific setup and then figure out what parameter values and even what sequence of jobs works best for what you want from it (e.g. since ziggurating needs a lot of info on a claim to be able to work with it, it might not be able to start on a freshly explored new claim straight away)
[10:45] Diana Coman: making better (best?) use of things as they are though is quite the nature of the game and where the difference can be made by any player without needing to start or know programming, so let me know if you give it a try.
[10:46] Diana Coman: for the record, the bundle job does exactly what you’d expect, namely aims to make bundles only, in whatever container is set as target when you start the job. Both bundle and zig actually read the blueprint/method so they’ll keep working even if/when the counts change on it.
[12:53] Diana Coman: a note to mind perhaps the location of claims when zig+explore, esp when relatively close to sector border (coming to you from my rather unintended trip to the faraway Barmy Crossing of Madness (some 16 sectors south of PPP according to the handy clientside map, way past that Fertile Knoll of Friskiness)
[15:14] Raphael Nethersmell: Diana Coman, sounds great re zigg’d botting, I’ve updated my client and will try it out later tonight.
[15:24] Diana Coman: cheers.
[16:10] Vivian Sporepress: Diana Coman: spiffy economic reporting from Monroe Quarterslit. just that seeing the deltas now makes me wonder about the opening positions. poor environment… at least it’s big, or so it appears so far
[16:14] Vivian Sporepress: hearing of the NPC’s cash decreasing and so forth makes me wonder, do they have some limited initial pile of it? or can they just go arbitrarily negative if people keep selling stuff to the merchants?
[16:15] Diana Coman: Vivian Sporepress, good to know, thanks. The deltas are literally the result as noticed at that monthly measurement. Why exactly wondering about the opening positions as such? Kind of iffy from a privacy pov to make these public.
[16:17] Diana Coman: well, in the widest sense, keeping all euloran economy meaningful and ticking is part of my job as world-maintainer so I do keep an eye on all these things. In the immediate, you noticed perhaps that an NPC out of cash will say so and refuse the trade since he can’t match it, there is no magicking anything out of thin air.
[16:18] Diana Coman: in general terms, value may get perhaps recycled when/if/as needed but not just created out of nothing.
[16:19] Vivian Sporepress: I suppose one can look at how long the game’s been going and extrapolate to get some rough idea of the totals based on the unreported months. yeah, I don’t know if it’s wisest to publish the totals, it just came to mind to give more context to the changes, or more meaning to the monetary amounts, like knowing the total btc supply limit
[16:20] Diana Coman: heh, more knowledge is always wanted but rarely available just for that reason, certainly
[16:20] Diana Coman: I’m pondering whether a total of amount traded might be added to these reports at some point but it’s not yet top of the list anyhow
[16:21] Vivian Sporepress: I saw some having trouble making change at least with Wolfgang Middlesit, but no such with Rainer Tracksnarl
[16:22] Diana Coman: for the record, one case in which I can see a reason to make the total btc supply limit known and public would be perhaps if there is a public pooling of such btc i.e. literally public subscription
[16:23] Diana Coman: every NPC has his/her limits, even if some may be larger than others!
[16:25] Vivian Sporepress: what do you mean about a public pooling/subscription? I mentioned btc as an example, knowing it’s not directly involved here
[16:26] Vivian Sporepress: and makes sense about limits; iirc Rainer Tracksnarl had at most 9999 of any given item
[16:26] Vivian Sporepress: (or maybe 10k even)
[16:30] Vivian Sporepress: anyway, over here in the Cruel Mound of Pluck I was attempting a full grid explore of the sector, my first such with the public bot. I even packed a lunch so I didn’t have to return home for that; but now it looks like the bot’s overall rep count simply ran out, and I don’t know how to resume the same grid rather than starting a new spiral from here
[16:31] Diana Coman: by public pooling/subscription I meant the case where the in-game value would have a corresponding pot of btc made public and publicly tradeable. Came to mind since it’s just about the only scenario I can currently see where there would be good reason and even required to make the overall sums clearly known and accountable publicly.
[16:33] Vivian Sporepress: on that worthless putrid leather supply, anyone can perhaps find similar now through my previous published map (the moony precipice being in the immediate outer ring from PPP)
[16:33] Ulrich Logfetch: Quoting 2026-01-29 (Eulora 2) Diana Coman: Raphael Nethersmell, that change from one offs to patch of wpl sounds interesting, any hypothesis as to why/how come? I’ve found some more ords but they keep requiring rare resources that nobody found yet and then on top, quite a lot of them as my char reads that enum, too. So not much to do about it atm other than hope either someone finds those rare resources or Rainer Tracksnarl gets a shipment of ord bundles perhaps…
[16:33] Diana Coman: Vivian Sporepress, ah, that is indeed not implemented as such since it would require setting a whole set of new parameters to start a grid from some arbitrary point in it rather than the beginning. Fwiw I simply pick another point in the middle of some empty area and start a new grid there, I didn’t find it yet annoying enough to warrant the trouble of more parameters
[16:36] Vivian Sporepress: I could continue to fill the space at same interval by calculating new spiral centers.
[16:36] Diana Coman: exactly
[16:37] Vivian Sporepress: first I might need to get some visualization pipeline for the new knowledge format, to get something from the raw data.
[16:39] Diana Coman: fwiw I tend to give it quite large step sizes too at least in new places, since it has the double advantage of filling reasonably well the grid and keeping the total number of claims reasonable (given that owned items including claims will not vanish while the owner is nearby, the drunkmove or a gridmove with small step can easily end up with a ton of claims and thus getting bogged down by all that data)
[16:40] Vivian Sporepress: are there some existing tools that consume the res_sect_kw.txt format or are you further awking it or something?
[16:41] Vivian Sporepress: ah, yes you noted sh+awk, http://ossasepia.com/2026/01/11/euloran-exploration-2-0/#footnote_2_7056
[16:42] Vivian Sporepress: bogged down with data or even just keys, as you noted recently.
[16:42] Diana Coman: I have a single sh script that goes through all the data, produces the inputs for graphviz (one per sector) and then runs graphviz to get the .png files too
[16:42] Vivian Sporepress: gnuplot?
[16:43] Diana Coman: gnuplot indeed
[16:44] Diana Coman: at least out of this most recent unplanned trip I ended up with a bunch of data points in all those sectors on the way back to PPP and so the script got a better testing too, it produced the 16 pictures just fine, lolz
[16:46] Diana Coman: correction there: the sh script runs an awk script because there is one such and apparently I even forgot it was enough in there to warrant a separate file
[16:48] Diana Coman: basically the awk does the conversion from kw data to the format wanted by gnuplot, while sh handles the multi-sector, multi-file part
[16:50] Vivian Sporepress: right
[16:53] Vivian Sporepress: I started out in awk but ended up in python, plus a minimal .sh to run the gnuplot for the resulting data file for a given sector; but then, that was before the support of the client knowledge maps and required correlating a lot of data from distinct data change log messages.
[16:59] Diana Coman: this reminds me that I found the original awk textbook and it’s quite striking how it reads precisely like what python supposedly is meant to be for only it rarely is – a simple and very effective tool for scripting, prototyping and data conversion especially. Possibly the environment of each is in the end more the deciding factor as to how they (awk and python) ended up being what they are.
[17:02] Vivian Sporepress: perhaps python’s from a decade where “data conversion” meant accurately parsing email messages and csvs and jsons and whatnot, hence much more “batteries included”; but also clearly going for large-scale with the modula/ada style module system
[17:04] Diana Coman: for the record, the textbook I am talking about is “The AWK Programming Language” by Alfred V. Aho, Brian W. Kernighan and Peter J. Weinberger (my edition is 1988)
[17:05] Diana Coman: and to quote from its very preface, a description which I found to be absolutely and entirely true in practice even today: “The same brevity of expression and convenience of operations make awk valuable for prototyping larger programs. One starts with a few lines, then refines the program until it does the desired job, experimenting with designs by trying alternatives quickly. Since programs are short, it’s easy to get started, and easy to start over when experience suggests a different direction. And it’s straightforward to translate an awk program into another language once the design is right.”
[17:07] Diana Coman: Vivian Sporepress, python got sucked into handling “everything” without any filter or restriction or even definition as to what that everything might be – and yes, exactly as the everything kind of exploded into all-possible-kitchen-sinks (and some impossible ones, too, for good measure).
[17:33] Raphael Nethersmell: basically, python had the unicode derpage baked in even before python3.
[17:37] Diana Coman: pretty much, yeah.
Day changed to 2026-02-03
[14:59] Felicia Solecrimp: omg the npcs are talking !
[14:59] Felicia Solecrimp: they can think !
[15:08] Diana Coman: it’s…possible, though not sure how would it go exactly, “I talk, therefore I think?”
[16:05] Diana Coman: Vivian Sporepress, your linked resource map reminded me that there is one bit of info missing from the current kw resource data, namely the lbn q1 i.e. ‘nothing here’ (since the data collection looks for claims only)
Day changed to 2026-02-04
[15:58] Rainer Tracksnarl: Diana Coman, that’s worth about 8.6k shells more than what you’re offering.
[16:02] Rainer Tracksnarl: Ordinary bundles now available, freshly shipped in though in limited and uncertain supply – get them while they last!
[22:55] Rainer Tracksnarl: Raphael Nethersmell, that’s worth about 34.5k shells more than what you’re offering.
[22:58] Rainer Tracksnarl: Raphael Nethersmell, that’s worth about 8.6k shells more than what you’re offering.
[23:22] Rainer Tracksnarl: Raphael Nethersmell, that’s worth about 226 shells more than what you’re offering.
[23:45] Raphael Nethersmell: I’ve yet to find any ordinaries, but looks like the fomo is gonna cause me to invest. it’ll be hard to know of which resource I’ll find an ordinary, so gotta buy one of each just in case ? hm.
Day changed to 2026-02-05
[08:23] Diana Coman: possibly it depends on how one goes about it – if you go from what you are likely to find, that depends on where you go exploring so if you really want to find just x, the approach would be to explore where there is only x; if you go from what you might need/where the demand for resources might be, you’d look at their use and then basics kind of go into everything anyhow so yeah, possibly one-or-more-of-each is the conclusion anyway.
[16:47] Raphael Nethersmell: makse sense.
[16:50] Raphael Nethersmell: Diana Coman, another refinement on the economic reporting that occured to me this morning was adding mean & median per capita data for PCs and NPCs. I’m curious how I stack up compared to the average NPC. and reporting the mean & median will show the skewness of the populations.
[16:58] Diana Coman: Raphael Nethersmell, that is an interesting idea. How do you see though the different NPCs i.e. the service providers vs non-service providers? And in the longer-term the potential mob/wild npcs aka those attacking on sight rather than exploring?
[16:59] Diana Coman: a note on that increase for the NPCs from this latest economics report: part of it is a recycling of previous environmental damage into the new ord bundles.
[16:59] Ulrich Logfetch: Quoting 2026-02-05 (Notices and Announcements) Ulrich Logfetch: ** Euloran economic activity during one month to day 00 of month 05, year 6 (era 1): PCs’ held value in items increased by 8`720 and their cash decreased by 2`793; NPCs’ held value in items increased by 1`394`023 and their cash increased by 2`793; environment value decreased by 418`476. All amounts are shown in Euloran Shell (10`000 Shells to 1 Copper). **
[17:05] Diana Coman: possibly of interest as a measure, the decrease in environment value is generally what PCs and NPCs together carved out of it one way or another at some point. On the other hand, what said PCs and NPCs managed to hold/make out of it would be the reported increase, so a sort of efficiency measure there, even if quite a rough one.
[18:33] Felicia Solecrimp: Wolfgang Middlesit where are you?
[18:33] Wolfgang Middlesit: Felicia Solecrimp, I am currently at -22.86 99.87 -18.81 in Preposterously Prosperous Place
[18:37] Felicia Solecrimp: my first awesome bought i have been working hard and i got my first own bought spicy mousaka
[18:37] Felicia Solecrimp: and I got 800 crumbly rock in one claim
[18:48] Diana Coman: oh hey, that 800 cr in one claim souns like a mini-pop, congrats Felicia Solecrimp!
[18:48] Diana Coman: enjoy the moussaka, it should get you a few euloran months respite iirc
[19:36] Raphael Nethersmell: Diana Coman, interesting re efficiency. wrt distinguishing npcs, the difference matter in terms of pcs interactions with them, but bottom line is the all need to eat. I expect some wil be more active than others. the first question when looking at the reported data that prompted the suggestion was, I have a good idea how many pcs there are, but well, how many NPCs are there actually ? i’m pretty sure there are some I’ve never glimpsed, especially the relatively little time i’ve spent in ppp.
[19:36] Raphael Nethersmell: Felicia Solecrimp, buen provecho !
[19:57] Raphael Nethersmell: like, approximately how much are V pcs getting ass kicked by npcs.
Day changed to 2026-02-06
[09:27] Diana Coman: heh, V pcs getting ass kicked, aptly put. Raphael Nethersmell, there are currently 11 NPCs in total, including all service providers. Perhaps the number as such may be included in the reports, though I’m not quite sure about the mean and median, I’ll give it some more thought.
[09:31] Diana Coman: The measure you are describing sounds like a sort of gdp per capita and that would be interesting, certainly, if one could get it right (or at least right-enough). And when this right-enough is not so clear, I’m more inclined to let people decide and calculate their own metrics as they see fit rather than stamp one as such through the report. It’s not that clear just how much (and even what, exactly) the average says. The median is likely to be at least more meaningful but it’s also way more expensive to calculate and in time it will get only worse. And calculating the median comes with potential privacy concerns if done centrally i.e. rather too close to individual monitoring.
[09:35] Diana Coman: meanwhile in zig-bot testing, it clearly needs some more robust means to detect work in progress but with those in place, it does reasonably fine. Though it also pushes, of course, for a more discerning use of comms clientside, so I suspect I’ll have to see to that sooner rather than later. Sigh.
[09:36] Diana Coman: and now I got some 150 cdg out of 2 claims, so my char is overbulk and stuck, hmpft
[09:38] Diana Coman: I guess I’ll have to add table-dragging to the bot’s capabilities, too
[15:34] Rainer Tracksnarl: Diana Coman, that’s worth about 1.1k shells more than what you’re offering.
[18:08] Raphael Nethersmell: Diana Coman, makes sense. my interest was my mainly regarding how I’m doing relatively. i think reporting population of pcs and npcs would be sufficient for now, as you said, let people make their own calculations.
Day changed to 2026-02-07
[03:47] Raphael Nethersmell: Diana Coman, meant to say, the bot ziggurating is so pleasant once once I get the hang of it. one thing it has changed in my gameplay is it have caused me to mix my lbn in bulk, which I hadn’t been doing while manually ziggin’. I had attempts like : 6 lbn q11 + 4 lbn q1 + cr(e) q2 -> 21 cr q4 or 5 lbn q11 + 3 lbn q1 + cr(e) q2 -> 25 cr q3 ; now i’ll do 100 lbn q11 + 40 lbn q1 -> 140 lbn q8. and then just zig with stacks of q8, adjusting the amount as needed via euconfing.ini
[03:51] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4198 — I recommend finding a pactch of bbb, which are very low weight and bulk and notbag.jpg value too !
[03:51] Ulrich Logfetch: Quoting 2026-02-06 (Eulora 2) Diana Coman: and now I got some 150 cdg out of 2 claims, so my char is overbulk and stuck, hmpft
[03:54] Raphael Nethersmell: pro tip : if you mistakenly enter a line meant for chat into your client console, just go to your terminal and issue : `tail data/logs_console.txt` and voila, you’ll have it available to copy and paste in the actually chat buffer !
[04:02] Raphael Nethersmell: one way I’ve been “dicerolling” for spots to explore is, if I’m going accross a sector w/ a relocate. sometimes it doesn’t get me to the final destination in one go, speed limit is enforced. issue a single or botted explore wherever it stops me.
[04:30] Raphael Nethersmell: watching just tail -f data/logs_bot_diary.txt , no gui, while running a spree of ziggin’ or exploring gives a window into how text-only client might feel.
[10:48] Diana Coman: Raphael Nethersmell, very glad to hear the bot zig is used and useful, thanks! The bbb recommendation makes total sense indeed, except so far I don’t think I found yet a patch of that (though I did find one of wwb somewhere far away and these are very small and light too, only less valuable). Overnight I let the bot zig+explore+train in the pss/nt patch in ppp and it did ok really, got 1 level further in zig, some 50 pss, some NT and quite a bunch of further ideas as to what the zig still needs to be more robust about.
[10:52] Diana Coman: the text-only client feel of that bot_diary watch is quite interesting really. Conversely, I was noticing the other day just how much the bot makes the GUI delay visible – if one looks at the bot’s actions, it’s always ahead of the graphics, basically the graphics keep playing a sort of catch-up game with the actions.
[11:00] Diana Coman: which points perhaps to a more fundamental shift in there: while the traditional/common/widespread/natural approach is to have the client/gui show “the world as the server/game/authority says you are meant to see it”, what I’m really after is turning this approach on its head, as I’d much rather have, if possible at all, the client/gui show what one does, what their impact is and ultimately the world as it emerges from the interaction with it, perhaps as “discovered”, with the server’s objects as conventions of potential illustration, not more.
[11:03] Diana Coman: ofc I hadn’t quite known just how deep my unhappiness with the usual gui went, it took this much just to start getting a glimpse as to what the deeper trouble is and to start even being able to put it into words, as it tends to happen.
[11:08] Diana Coman: Raphael Nethersmell, on the pick of exploration points, plotting all that resource data tends to be quite helpful as it makes it very easy to start systematically getting to know where things are and one can of course start with very coarse granularity (like Vivian Sporepress did for that mini-map he made with his own bot) and then add detail when + where + as much as wanted
[23:58] Felicia Solecrimp: Diana Coman Can I ask you your advice on whether a hosting service should optimize into using ai features and how can it be better to provide human interaction than machine bots and if whether a decetrilize hosting is actually a good idea ?there are free hosting services for students and developers but aside of ecommerce and streaming websites which coud be the most succesful way in order to offer for a higher price or what could be our differentitator ?
[23:59] Felicia Solecrimp: because i also notice that all hosting services monthly are really low but they increase of price when they show more features .
Day changed to 2026-02-08
[00:02] Felicia Solecrimp: Dedicated Hosting Dedicated hosting means renting an entire physical server just for your site(s). Like owning a house instead of renting an apartment. I think this could be the closest to offer 100 for a hosting server that generates revenue option that i found also
[00:02] Felicia Solecrimp: with this information im looking forward for your opinion
[00:06] Felicia Solecrimp: but i still believe in the importance of human contact for client service because is something that is lacking somehow with all of the automatization now days too
[00:21] Felicia Solecrimp: known host is 143 and inmotion is 110
[00:23] Felicia Solecrimp: ionos is also very popular
[00:24] Felicia Solecrimp: but is very cheap and it has personalized attention
[00:25] Felicia Solecrimp: inmotion has 5 channels in order to talk to people for service
[00:27] Felicia Solecrimp: plenty methods of payment is something else i notice
[00:28] Felicia Solecrimp: a free domain if you register for one year
[02:15] Vivian Sporepress: Felicia Solecrimp: those sorts of data points you should gather into an overall, structured document so people can read at their convenience if they’re even interested – or at least just send them to me privately, though it’s mostly for your own orientation since I’m already somewhat familiar with that market
[02:16] Vivian Sporepress: “decentralized hosting” is still a new buzzword to me, any idea what it might possibly mean?
[02:22] Vivian Sporepress: I think you have a good point that the human factor is a strong differentiator, at least potentially, since I certainly won’t be implementing chatbots for “support” or whatever such nonsense. moreover, with my service the customer gets direct access to the engineering & management behind it, not some powerless call center. the question is if that will be enough and if we can communicate it effectively to the target customer
[02:27] Vivian Sporepress: and you mention payment methods – how many will take direct btc payments (and not through some flaky gateway at best)?
[02:34] Vivian Sporepress: another point worth internalizing is that because of strong identities, we’re practically immune to the biggest threat at other service providers – social engineering: http://trilema.com/2014/o-hai-let-me-verify-your-identity/
[09:05] Diana Coman: Felicia Solecrimp, why don’t you do this properly and exactly as Vivian Sporepress suggest, put together everything you’ve got so far in a structured, coherent piece of writing so that both you and anyone else can read it, add to it, comment on it and even have it as reference at a later point? There is a point to each of these steps and no gain in skipping them. Moreover, then you can simply give a link and have any conversation happen or at least bridged there rather than pushing it here where it doesn’t really quite belong in full anyhow.
[11:20] Diana Coman: on bbb patches, looking back through the logs pre-bot, it turns out that all bbb I found was in PPP, only entirely overlapping with other resources it seems
[11:25] Raphael Nethersmell: Diana Coman, any interest in a trade of bbb patch cords for wwb patch cords ?
[14:40] Diana Coman: Raphael Nethersmell, sure, why not. The one I know is quite far from PPP though, just to make it clear.
[19:26] Vivian Sporepress: newbie question – how does the bot zig pick what quality inputs to use?
[19:57] Diana Coman: Vivian Sporepress, by default it goes with 50% meaning in the middle of the interval for each ingredient. This can be set via parameter in the .ini file (it’s only one parameter though so same for all ingredients).
[19:58] Diana Coman: as for quality, it doesn’t care as such, it’s up to the player to have in inventory what they want used rather than what they don’t want used. And if they want something specific, the way to go about it is to make and have bundles – the bot will use bundles first whenever available for the zig at hand.
[20:00] Diana Coman: should add that there is quite on purpose a loophole to “hide” things from the bot (e.g. if you don’t want them used): the bot looks only at stuff directly in inventory, not those inside container inside inventory, so if you want to keep things away from it, just put them in a table and pick that table up.
[21:31] Vivian Sporepress: that’s a nice trick.
[21:32] Vivian Sporepress: the bot’s ziggurating struggles appear quite reminiscent of my own… still waiting for stuff to update, overshooting because it seemed a move didn’t go through, aborting impossible work due to incomplete info, finally everything’s lined up and still nothing happens…
[21:36] Vivian Sporepress: some of my claims were just now expiring, though seemingly not the one I’m trying to zig; only guess atm is that it’s “trying” to expire but can’t because I’m standing right here, but that still somehow prevents me from building it
[21:37] Vivian Sporepress: ahh no, it’s because my enum readings changed since last checked, so much that even 50% is now over 100%
Day changed to 2026-02-09
[08:23] Diana Coman: yes, the whole of e2 is quite calling for nimbleness rather than blind data crunching hence the troubles poor bot has. Even so, there are quite a few parts where the bot zig could attempt to improve robusteness for sure. If you want to avoid entirely the potential block of enums changing contents while the bot is trying to do something, the bundles are again the answer (since it’s always and by definition ONE bundle, that part doesn’t change)
[11:56] Diana Coman: given that the bot zig job is clearly of interest, I’ve deployed now the cutting edge client version as well. In particular, it should deal much better with any changes in the enum while it’s trying to zig, since now it just re-reads it every time so that it can adjust if needed. There are further changes and checks to hopefully become more quickly aware of items of interest and generally to be better able to deal with the unexpected.
[11:56] Ulrich Logfetch: Quoting 2026-02-09 (Notices and Announcements) Monroe Quarterslit: ** Euloran news: newly deployed client responds to players’ feedback on the bot zig job adding further checks to react more nimbly to changes in enum readings and to improve robustness of bot jobs in general. Furthermore, this version looks for and records LBN as a resource in its own right as and when seen in the world (hence available in the resource knowledge maps afterwards). **
[11:58] Diana Coman: there is one added parameter that can be set to request the zig job to keep track of its starting pos and return to it, though by default this is set to false aka no return (the behaviour as it has been so far). While it can be a waste of time and quite annoying to keep going back and forth if/when one is ziggurating an existing bunch of claims, it can be equally annoying to have a zig job mess up one’s exploration grid or similar pos-related concerns, hence this new option via zig_ret
[12:01] Diana Coman: I’ll add explicitly that these changes are very new and there was very little time to put them thoroughly to test in practice, so if you do update to this latest version, do let me know how it works for you
[16:15] Vivian Sporepress: I’ll try it
[16:19] Diana Coman: cheers
[16:26] Vivian Sporepress: bot none
[16:26] Diana Coman: heh, the multi-input troubles
[16:26] Vivian Sporepress: lol whoops. meant to say: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4199 – what’s that? I can imagine stacking heavy items into a craft table but then not sure how I’d move that without being again over limit
[16:26] Ulrich Logfetch: Quoting 2026-02-06 (Eulora 2) Diana Coman: I guess I’ll have to add table-dragging to the bot’s capabilities, too
[16:29] Diana Coman: it’s a trick from back in e1 and yes, e2 allows it as well: while you can’t move carrying the heavy table, you can drop it, move a bit, pick it up again, repeat. Hence table-dragging and yeah, a pain by hand, though relatively easy and certainly painless by bot (and if you are ever far away when some claim pops with a ton of something, you’ll likely prefer to drag the thing instead of potentially losing it if you leave it there, not to mention it may even take more than one trips to carry it at your char’s capacity)
[16:31] Diana Coman: good old foxybot even used a table at all times, stacking into it whatever it got from the claims and carrying it around so it never got overweight. Plus it allowed one to carry with them more lbn/bundles to build with, there is that side of it, too.
[16:34] Vivian Sporepress: so we have wagons, just no wheels, neat
[16:39] Diana Coman: we have heels not wheels!
[16:40] Diana Coman: though e2 pushes this one further since there is the upper limit of what your char can *lift*. So if the poor bot were to try and fully solve this, it would end up juggling any number of tables, sheesh.
[16:48] Vivian Sporepress: a first bot zig on latest client again overshot on loading lbn and had to move some back; a second somehow found work already ongoing and finished in 3 reps; a third tried to load a bundle instead of lbn and aborted on failing to find such. Overall it still works eventually though
[16:49] Diana Coman: hm, it would normally try to load a bundle only if it found one present
[16:50] Diana Coman: by a second one do you mean you stopped the restarted the bot or just the 2nd time it got to the zig job in a sequence?
[16:50] Vivian Sporepress: second automated run, no manual restart
[16:52] Vivian Sporepress: so far at least it hasn’t hit what seemed the biggest stumbling block yesterday, “aborting impossible work: Missing/incomplete info (stack) for obj …” where that obj was some lbn that no longer existed
[16:53] Diana Coman: anyways, noted to have another look at the checks for the bundle and what it can do to make sure there is one before going for it. Otherwise though there will always be so much that can go wrong/differently/unexpectedly that the real goal is really to still work eventually, not as much some smooth sequence which is very nice while it lasts but then too fragile for what this environment is
[16:53] Diana Coman: yes, that is one thing that I specifically fixed ie if/when it finds some info missing, it will report it as such but request the object so it finds out that it’s gone and otherwise keeps trying rather than bailing out
[16:58] Vivian Sporepress: nice
[16:59] Vivian Sporepress: perhaps it thought I had a bundle because it grabbed the claim marker contents and hadn’t yet noticed the bundle there turning into the item
[17:23] Vivian Sporepress: Diana Coman: looks like euconfig.ini could use the pct entry under BOT-BUNDLE. perhaps also a refresh of ServerName and Comment
[21:27] Diana Coman: Vivian Sporepress, indeed it can, I made the changes now to euconfig.ini, thanks.
[21:28] Vivian Sporepress: thanks to the high volume of time observations now handily provided by the bot diary, I’m finally inferring that Euloran time consists of… months 0-11, days 0-23 and hours 0-19. can’t say yet if the ranges are context-dependent too, how many total days in the year and such
[21:31] Felicia Solecrimp: I found a crafting table nearby rainer
[21:31] Felicia Solecrimp: I told vivian I did not pick it up because i thought it was somebody else table
[21:37] Vivian Sporepress: and I told her that was an ethical mistake, since the game will either enforce ownership or else zap it altogether if nobody grabs it
[21:39] Felicia Solecrimp: By the way sorry for spouting all together the content of the hosting and not making it available in a more organize contetx I think I need to think before acting and how will this repercute on the game otherwise
[21:39] Diana Coman: yeah, feel free to try and pick stuff up, Felicia Solecrimp
[21:39] Felicia Solecrimp: act least on the chat
[21:42] Diana Coman: awareness of context helps at all times, so if it helps towards that, no harm done.
[21:52] Vivian Sporepress: it looks like indeed all months have an even 24 days; I’m just missing the full range from m7 but it’ll be coming back around soon.
[21:59] Vivian Sporepress: and it looks like a quite uniform 20 hours per day.
[22:13] Vivian Sporepress: since I’ve been on this revisit of chronometry & chrononomy. otherwise, now on the hunt for the abandoned eggs… I stumbled on some right here in PPP, in fact two close by each other, just wasn’t quick enough to notice & build them, and it’s not a reliable patch.
Day changed to 2026-02-10
[16:31] Diana Coman: the first PC mention of an euloran almost-complete calendar!
[16:31] Diana Coman: and yeah, I even got that Purple Snail still in PPP but never managed to repeat the catch
[16:32] Diana Coman: the AE are quite heavy though iirc
[17:28] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4241 — the main inefficiency I’ve noticed with zigging is loading multiple stacks of lbn into a single enum. e.g. if it’s set to 8, it’ll load 2 stacks of 8 and then remove one before it tries to build.
[17:28] Ulrich Logfetch: Quoting 2026-02-09 (Eulora 2) Diana Coman: yes, the whole of e2 is quite calling for nimbleness rather than blind data crunching hence the troubles poor bot has. Even so, there are quite a few parts where the bot zig could attempt to improve robusteness for sure. If you want to avoid entirely the potential block of enums changing contents while the bot is trying to do something, the bundles are again the answer (since it’s always and by definition ONE bundle, that part doesn’t change)
[17:54] Diana Coman: Raphael Nethersmell, that would be only because of the data lag as it were – it doesn’t *try* to load 2 stacks, only one, but the trouble is that when it looks again it seems as if the attempt didn’t take, hence it tries again. Possibly increasing the delay/check time might help in that it’s more likely to be up to date by the time it looks again but otherwise there isn’t much the bot can do currently to avoid this sort of trouble.
[17:55] Diana Coman: maybe it could assume that it worked the first time and issue an action request at least when there was only one move left, though dubious
[17:57] Diana Coman: this might perhaps get addressed closer to the root only when the client gets an actual strategy for data refresh/request, but it’s certainly not trivial
[17:59] Diana Coman: meanwhile in possibly more worrisome misfires, it seems that there are cases when some of the outputs don’t make it in time either and so they remain in the claim/end up on the ground afterwards. Perhaps it’s high time to add a pickall (giving also how people kept trying to use takeall for that, lolz)
[19:20] Raphael Nethersmell: Diana Coman, ah, makes sense. I’ll trying increasing the poll_beat on zig to 30 for now and see how it fares.
[21:47] Vivian Sporepress: is it possible the zigbot simply isn’t refreshing the claim marker after requesting the loading of ingredients? not sure then why it would still work sometimes, but I found when I do a reqobj on the marker after its move, the lbn tends to pop right up.
[21:47] Vivian Sporepress: and yes, I noticed items falling on the ground after claim expiry too.
[21:50] Vivian Sporepress: also I note that the explore bot is still prone to targetting non-explorable entities based on whatever was last clicked, to be reset by examining the ground.
[21:52] Vivian Sporepress: (not meaning to be a nag – I just thought I’d read at some point that this had improved)
[22:06] Diana Coman: Vivian Sporepress, no, it’s not possible that it doesn’t refresh, but it IS possible that the refresh ends up “too soon” as it were OR incomplete, hence a second one you send manually sorts it all out
[22:09] Diana Coman: the explore bot normally targets whatever is set as target, and that is by default what you look at, so you do need to mind that before starting the bot. During a run it shouldn’t *change* the target though (so it’s safe to examine whatever). I could just force-set it to none/ground/sector but I’d rather avoid that if possible, simply because the explore is one main type of action, potentially applying to anything. If it really gets too annoying, I suppose I could add a parameter to toggle on/off “no target” in general even, might at least address the trouble while still retaining some flexibility for the longer term.
[22:11] Diana Coman: (and yes, you remember correctly that it improved there – in that it keeps the target rather than ending up with it changed mid-run as it used to happen, which was worse)
[22:14] Felicia Solecrimp: Rainer Tracksnarl where are you?
[22:14] Rainer Tracksnarl: Felicia Solecrimp, I am currently at -18.80 97.41 35.56 in Preposterously Prosperous Place
[22:14] Felicia Solecrimp: Rainer Tracksnarl Where are you ?
[22:18] Diana Coman: should add that the trouble with “send a 2nd request” is that in practice it doesn’t guarantee anything by itself, for the same money could send 3 each time or any other number. The only thing that could perhaps help is literally a higher level layer that keeps track of what is *wanted* and generates requests according to whatever rules it has to get that, so quite the additional logic there and likely non-trivial to reliably get smooth results and be essentially invisible.
Day changed to 2026-02-11
[16:57] Diana Coman: following up on the target for explore, I’ve added now a parameter in the .ini file to have explore always set its target at 0 and thus be entirely immune to any target sets by the GUI whether before or during a bot run. Note that the default for this null_target parameter is actually 0 aka still using the GUI-set target so that the behaviour doesn’t change unless you explicitly set this parameter in you ini file. As I gather that the issue has been quite a nuisance, I’ve redeployed the client so that this change is available.
[16:57] Ulrich Logfetch: Quoting 2026-02-10 (Eulora 2) Diana Coman: the explore bot normally targets whatever is set as target, and that is by default what you look at, so you do need to mind that before starting the bot. During a run it shouldn’t *change* the target though (so it’s safe to examine whatever). I could just force-set it to none/ground/sector but I’d rather avoid that if possible, simply because the explore is one main type of action, potentially applying to anything. If it really gets too annoying, I suppose I could add a parameter to toggle on/off “no target” in general even, might at least address the trouble while still retaining some flexibility for the longer term.
[19:29] Raphael Nethersmell: Felicia Solecrimp, you know you’re exploring right in poor Rainer Tracksnarl’s lap ? mind giving some space ?
[19:30] Raphael Nethersmell: exchange Rainer Tracksnarl
[19:37] Raphael Nethersmell: I could see if he was sitting in a patch of wm or eps or pb, but cr isn’t so hard to find, you know ?
[19:46] Raphael Nethersmell: anyone have lnb q1 they’d be interested in selling me ?
[19:57] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4232 — understood, voila.
[19:57] Ulrich Logfetch: Quoting 2026-02-08 (Eulora 2) Diana Coman: Raphael Nethersmell, sure, why not. The one I know is quite far from PPP though, just to make it clear.
[20:30] Diana Coman: Raphael Nethersmell, nice, here’s mine
[21:35] Raphael Nethersmell: got it, thanks.
Day changed to 2026-02-12
[03:22] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4300 – thanks, will try.
[03:22] Ulrich Logfetch: Quoting 2026-02-11 (Eulora 2) Diana Coman: following up on the target for explore, I’ve added now a parameter in the .ini file to have explore always set its target at 0 and thus be entirely immune to any target sets by the GUI whether before or during a bot run. Note that the default for this null_target parameter is actually 0 aka still using the GUI-set target so that the behaviour doesn’t change unless you explicitly set this parameter in you ini file. As I gather that the issue has been quite a nuisance, I’ve redeployed the client so that this change is available.
[03:24] Vivian Sporepress: http://ossasepia.com/2026/02/02/Notices-and-Announcements-Logs-for-Feb-2026/#197 – sounds like the sustainability & environmental protections programme is having some effect, and profitable one at that
[03:24] Ulrich Logfetch: Quoting 2026-02-11 (Notices and Announcements) Ulrich Logfetch: ** Euloran economic activity during one month to day 00 of month 06, year 6 (era 1): PCs’ held value in items increased by 27`672 and their cash increased by 29`481; NPCs’ held value in items increased by 46`689 and their cash decreased by 29`481; environment value decreased by 449`403. All amounts are shown in Euloran Shell (10`000 Shells to 1 Copper). **
[03:28] Vivian Sporepress: some fresh personal ziggurating records: 1 cr q75, 1 cr q86, 1 cr q93, and 1 lonely nt q98
[03:44] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4305 – I could, but can’t compete price wise with the merchant’s npc-powered supplies
[03:44] Ulrich Logfetch: Quoting 2026-02-11 (Eulora 2) Raphael Nethersmell: anyone have lnb q1 they’d be interested in selling me ?
[04:11] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4202 – there’s a little algebra to find the ratios to do these mixes without rounding loss, short of trekking over to EOJ…
[04:11] Ulrich Logfetch: Quoting 2026-02-07 (Eulora 2) Raphael Nethersmell: Diana Coman, meant to say, the bot ziggurating is so pleasant once once I get the hang of it. one thing it has changed in my gameplay is it have caused me to mix my lbn in bulk, which I hadn’t been doing while manually ziggin’. I had attempts like : 6 lbn q11 + 4 lbn q1 + cr(e) q2 -> 21 cr q4 or 5 lbn q11 + 3 lbn q1 + cr(e) q2 -> 25 cr q3 ; now i’ll do 100 lbn q11 + 40 lbn q1 -> 140 lbn q8. and then just zig with stacks of q8, adjusting the amount as needed via euconfing.ini
[04:14] Vivian Sporepress: say you mix stacks N of qual X and M of qual Y; you get a stack N+M of qual floor[(NX+MY)/(N+M)]
[04:22] Vivian Sporepress: call that output quality Z, assume no rounding and multiply by the denominator: NZ+MZ=NX+MY. by rearranging and dividing, you get N/M = (Y-Z)/(Z-X). plug in your desired Z and available X and Y and there you go
[04:32] Vivian Sporepress: for your example, Z=8, X=1, Y=11 so N/M = 3/7. you want 140 so mix 42 lbn q1 with 98 lbn q11
[08:21] Diana Coman: heh, Vivian Sporepress got it on the mixing, applied algebra ftw! In e2, people had even a shorthand for mixing, simply picking stacks of same count and same *parity* wrt quality (since that ensures that the division has remainder 0 hence no loss), potentially repeating if/as needed. E.g. mix 7 at q5 with 7 at q99 OR 7 at q10 with 7 at q200
[08:21] Diana Coman: meant to say in e1*, ofc
[08:23] Diana Coman: Raphael Nethersmell, fwiw I’ve let the bot overnight to map that bb spot from you and it seems to cover fine a whole 100×100 square (starting to get into lbn on the edges)
[08:23] Diana Coman: on the q1 lbn, I guess it’s currently more difficult for people to get that rather than the q11 since the bot is not aware of lbn as output of an explore request.
[08:25] Diana Coman: and its value is so low as to make it hardly worth much effort really, lbn is 11 shells at q100 so at q1 you need 10 of them just to make 1 single shell, it’s more of a shell-cent than anything, lolz
[08:30] Diana Coman: Vivian Sporepress, wrt merchant prices, they are meant as a reliable baseline only so I wouldn’t even expect that PCs want to trade among them at NPC prices. And as time passes, there is no guarantee that Rainer Tracksnarl keeps the same stock anyhow.
[09:23] Diana Coman: interesting data points on the zig, too, any ideas as to what might be explaining them or how to get such high q more reliably? While high quality means higher value anyhow, there might even be a premium for it since it’s rather rare currently, dunno.
[09:24] Diana Coman: meanwhile in my own gathering/ziggurating experience, my enum q went again down but worse really, not just to q3 from the “high” of q4 but again back to q2, apparently easier to slide down with it than to get anywhere higher.
[13:53] Raphael Nethersmell: in unexpected but welcome outputs, I got an sms q1 out of a ziggurating attempt.
[13:57] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4315 — when I looked at Rainer Tracksnarl’s inventory yesterday I didn’t see any supply.
[13:57] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4305 – I could, but can’t compete price wise with the merchant’s npc-powered supplies
[14:06] Diana Coman: ha, that would be first reported instance of an activity yielding a skill-item, congrats Raphael Nethersmell!
[14:06] Ulrich Logfetch: Quoting 2026-02-04 (Notices and Announcements) Monroe Quarterslit: ** Euloran news: a well-placed source has it that some activities may now yield on occasion skill-giving items, too! Who will be the first to get any such item and what even are the new skills thus available? **
[14:07] Diana Coman: and it occurs to me that Vivian Sporepress is saying there that the 20% markup at Rainer Tracksnarl is …too small?!
[14:11] Raphael Nethersmell: I’d definitely beat Rainer Tracksnarl’s price.
[14:12] Raphael Nethersmell: Diana Coman, it’s actually the second one I’ve got, but I was away during the first one and wasn’t exactly sure how it emerged, this one I witnessed live.
[14:13] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4324 — yeah, I know they’re quite cheap, I just seem to get better zig value with lbn q < 11 so far.
[14:13] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Diana Coman: Raphael Nethersmell, fwiw I’ve let the bot overnight to map that bb spot from you and it seems to cover fine a whole 100×100 square (starting to get into lbn on the edges)
[14:14] Raphael Nethersmell: that is, total return per “click”, i.e. I get lower q outputs, but more of them.
[14:19] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4324 — most bodacious ! i’ve only been drunking moving on the wwb so far, but have zigg’d over 1.1k with bulk and weight capacity to spare.
[14:23] Diana Coman: sounds like you found quite the sweet spot of ziggurating there, getting 2 skill items already and 1.1k wwb, apparently there is plenty to be done at a profit, too
[14:23] Diana Coman: in terms of return per click, you’d need to calculate the qabv really since count is not all there is to it (although it does sound like you are getting more than you put in, in real terms)
[14:27] Diana Coman: as to the lbn q1 supply as limiting factor, it sounds quite very euloran in style really – as soon as one finds some sort of “guaranteed money-making”, they find out that they soon run out of something or the other
[14:28] Raphael Nethersmell: yeah, looks like it’s been worth flipping through all the maps to arrive here. ~40% of that 1.1k is q1 from 1 small claim. I burned pretty much an entire ch to get 2 smalls.
[14:29] Diana Coman: that sounds quite familiar from e1 too, namely you barely get anything for whatever number of attempts and then you get a ton from a lucky hit (it took though all the other ones to get there though)
[14:30] Diana Coman: kind of curious if you’d get more smalls on a lower value resource (since you say the zig gives you either more at lower q or fewer at higher q, then possibly there would be something similar with the claims – if one manages to figure out how would that apply to the claims in the first place, exactly)
[14:32] Raphael Nethersmell: what are the arguments the l parameter accepts for the train command ? I greped lesson on the src dir and only yielded hits in client/euaction.cpp ; in that file I don’t see any mention of what arguments it takes. so far i’ve tried gather, gathering and explore but the console log reports “lesson 0”
[14:33] Raphael Nethersmell: Diana Coman, aha, I did get more smalls on exploring a cdg patch with ch.
[14:34] Diana Coman: Raphael Nethersmell, the l aka lesson is the training session itself, hence the ID of the training session object or 0 if/when there is no such thing (ie when asking for a new one)
[14:35] Raphael Nethersmell: ok
[14:35] Diana Coman: all the interaction commands work like this, referring to the session object itself (and 0 is in general the “null” value)
[14:36] Diana Coman: quite interesting then, maybe if one gets more and/or skill items out of ziggurating small claims, it really is a better use of ch currently on lower q resources than on higher q ones
[14:36] Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4346 — “Gravitation is not responsible for your great tumble – after all it tried to prevent your climb every step of the way.”
[14:36] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Diana Coman: that sounds quite familiar from e1 too, namely you barely get anything for whatever number of attempts and then you get a ton from a lucky hit (it took though all the other ones to get there though)
[14:37] Diana Coman: hehe, quite so
[14:39] Raphael Nethersmell: Diana Coman, on your new blog setup was there some unicode conversion of text, e.g. on my gentoo I see “great tumble
[14:40] Diana Coman: Raphael Nethersmell, ah, I get now what you were after with that l parameter but it doesn’t work like that, heh. For the record, on any protocol level command, it’s the protocol spec where you would find more information about what it takes and how it works. As for *what* to train, it simply depends on what the trainer *can* train you in, so a matter of what they know. If you noticed in the training window, it says what the trainer can train and up to what level, as well.
[14:41] Raphael Nethersmell: I also notice some curly quotes in the broswer’s rendering of the log, but those don’t appear in my txt file.
[14:41] Diana Coman: uhm, no explicit unicode conversion but I’ll have a look – fwiw here on centos 6 I see the – fine
[14:43] Raphael Nethersmell: I didn’t mean to imply accustion of the crime of explicit unicode conversion, ftr. lulz.
[14:43] Diana Coman: lolz, no, it’s fine, I’ll add it to the list to double check there, thanks for letting me know
[14:44] Raphael Nethersmell: noted re referencing the spec, thanks.
[15:52] Diana Coman: adding to those zig data points, I got 36 cdg at q227 even, from a bare-handed small, though sadly no skill-item so far.
[16:11] Raphael Nethersmell: how many cft did you use and what quality ?
[17:10] Diana Coman: heh, the right question asked: 31 cft, q61 (that’s what I had in storage)
[17:11] Diana Coman: pondering here if the bot shouldn’t take advantage of all the infrastructure to do a before/after list of things in one’s inventory and log that for each action, perhaps
[17:11] Diana Coman: log the changes, I mean, not the full list each time
[18:27] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4331 – oic. and I find I’ve used up most of mine; I have in stock 5 q1 and 85 q2, asking 10 shell apiece for any of those. or I can just point you to where to dig for them
[18:27] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Raphael Nethersmell: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4315 — when I looked at Rainer Tracksnarl’s inventory yesterday I didn’t see any supply.
[18:28] Vivian Sporepress: I guess the exploring NPCs haven’t been in the lbn patches for a while, and we previously benefitted from an initial glut from before they knew how to trade.
[18:34] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4328 – haven’t found the secret; so far it’s been just doing more, by bot or otherwise and letting the details vary, though lately I’ve been going with straight lbn q11.
[18:34] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Diana Coman: interesting data points on the zig, too, any ideas as to what might be explaining them or how to get such high q more reliably? While high quality means higher value anyhow, there might even be a premium for it since it’s rather rare currently, dunno.
[18:37] Vivian Sporepress: http://ossasepia.com/2026/02/02/Eulora-2-Logs-for-Feb-2026/#4359 – myeah, looks like the verschlimmbessert wordpress frontend is autocorrecting quotes, hyphens into en-dashes, double-hyphens to em-dashes, three dots to unicode ellipsis and so fifth. for extra lulz, my fabulous ubuntu fails to paste text from browser to terminal *at all* when it contains such magic byte sequences
[18:37] Ulrich Logfetch: Quoting 2026-02-12 (Eulora 2) Raphael Nethersmell: I also notice some curly quotes in the broswer’s rendering of the log, but those don’t appear in my txt file.
[18:40] Vivian Sporepress: ah, that’s xfce4-terminal to blame, classic xterm pastes obediently with some kind of garbage-in garbage-out.
[18:41] Vivian Sporepress: xfce’s ‘mousepad’ text editor has been similarly broken, refusing to save a file at all if there’s some encoding failure
[18:41] Vivian Sporepress: quite python3 style
[18:47] Felicia Solecrimp: Rainer Tracksnarl where are you?
[18:47] Rainer Tracksnarl: Felicia Solecrimp, I am currently at -18.80 97.41 35.56 in Preposterously Prosperous Place
[19:43] Raphael Nethersmell: Felicia Solecrimp, you know he hasn’t moved in like… weeks ? better to note the spot and return to it and only if he’s not there, then ask in the log.
[20:48] Raphael Nethersmell: just when I think I’m about to get overbulked, a bunch of keys turn to lbn and lighten me up !
Comments feed: RSS 2.0