IRC logs for #openttd on OFTC at 2026-04-10
β΄ go to previous day
00:38:16 <reldred> talltyler: I don't think we'll be seeing much of xarick around anyway...
01:00:06 <talltyler> I want #15341 but it makes so many unnecessary changes, I'll probably have to rewrite it myself... π
01:00:29 <talltyler> Why is it so hard to keep PRs simple?
01:20:42 <reldred> copilot effect I'm thinking
01:28:32 <talltyler> Yeah, admitting to using copilot instantly makes me distrust all future PRs π₯²
01:29:47 <reldred> but no, i'm not expecting them back. They copped a 7day time out on discord for making homophobic comments, then wondered why their reception on IRC wasn't seen favorably, then mouthed off at the mods and ragequit.
01:30:41 <reldred> now we get abigail dropping cbombs instead
01:31:26 <reldred> but yes probably quite a few pr's that could be closed
01:31:27 <talltyler> The Aussies are taking over!
01:31:31 <belajalilija> Itβs part of her culture
01:31:48 <reldred> look its my culture too alright but I at least try to behave myself these days lmao
01:32:18 <belajalilija> Whereβs the fun in that
01:33:42 <reldred> goddamnit, I switched my kvm over to this machine to cue up music and now I'm getting distracted again
01:34:20 <belajalilija> Bet you listen to cuntry music
01:35:18 <reldred> no, goth, darkwave/etc, drum and bass, boomer vintage electronic music, occasionally black metal.
01:35:33 <belajalilija> Mostly good stuff then
01:37:55 <squirejames> belajalilija: Take me home, cuntry roads
01:39:29 <belajalilija> Everyone should listen to this
01:39:58 <reldred> no Iβm listening to proper british music right now
01:40:23 <squirejames> reldred: Bass boosted Rule Britannia? π
01:40:54 <belajalilija> squirejames: Β£1 fish
01:42:04 <belajalilija> We peaked as a culture here
01:42:28 <belajalilija> Truly our gangnam style
02:02:06 <__abigail> reldred: It's all about increasing soft power so we can have our own Manifest Destiny centred on bringing dim sims and chicken salt everywhere
02:02:10 *** Wormnest has quit IRC (Quit: Leaving)
02:23:11 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:23:12 *** WormnestAndroid has joined #openttd
02:23:33 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:23:35 *** WormnestAndroid has joined #openttd
02:23:42 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:23:43 *** WormnestAndroid has joined #openttd
02:26:16 *** WormnestAndroid has quit IRC (Remote host closed the connection)
02:26:19 *** WormnestAndroid has joined #openttd
02:39:56 *** ChanServ sets mode: +v tokai
02:47:01 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
03:11:34 *** Philemon has joined #openttd
03:14:56 *** Phileman has quit IRC (Ping timeout: 480 seconds)
03:15:03 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
03:15:05 *** WormnestAndroid has joined #openttd
06:25:33 *** cu-kai has quit IRC (Quit: Leaving for a bit.)
07:17:57 <reldred> Been a short week for me what with Easter
07:18:04 <reldred> Swear I slept half of it
07:26:15 <rito12> reldred, when have you waken up?
07:48:58 *** Alkel_U3 has quit IRC (Quit: maintenance)
07:49:49 *** Alkel_U3 has joined #openttd
07:59:53 <LordAro> talltyler: they did the "unrelated changes" thing long before copilot was a thing
08:00:45 <andythenorth> if anything, LLMs have built vectors around enough "don't make unrelated changes" articles
08:01:11 <andythenorth> I only use GPT, but it's considerably less prone to scope creep than people
08:08:10 <peter1138> Steam commenters have got bored.
08:09:05 <reldred> yeah I've so-far been utterly unimpressed with both copilot and gpt, and thoroughly unimpressed with coworkers who insist on using them.
08:09:13 <LordAro> from where will we derive joy now then?
08:09:45 <reldred> but most of my job is based on critical thinking, the business already outsources almost all the work that doesn't require it.
08:10:24 <andythenorth> I never have to write css again
08:10:34 <andythenorth> or utility javascript for gui stuff
08:10:48 <reldred> look, fair, I don't have to write any of those things at all
08:10:58 <andythenorth> and GPT knows far more about python standard lib, or common packages than I do
08:11:49 <andythenorth> if it's used like an interactive version of this, with live examples, it's quite good
08:11:56 <reldred> i'm spending most of my time staring at rack layouts of two DC's piecing together various recollections of various conversations from a dozen different teams about several hundred devices and their various interdepencies
08:12:01 <andythenorth> I grew to really hate that book
08:12:07 <LordAro> Anthropic Mythos seems like a system that an eye or two should be kept on
08:12:07 <Borg> the problem is.. when the jobs requiring analitical and critical thinking are not enjoyed by people doing them.. trouble start to arise
08:12:15 <andythenorth> none of the examples worked reliably across all browsers in 2000
08:12:29 <reldred> actually i should be working on some powerbi stuff but I'm bored of that and focusing more on the analytical stuff
08:12:36 <reldred> finding places to save money
08:12:43 <Borg> you cant compare those jobs to simple building jobs.. when you can mostly work nearly mindlessly
08:12:43 <andythenorth> "omg, we have a terrifying new tool, too powerful to release, please buy our stocks"
08:13:01 <andythenorth> "we're one year away from commercial fusion"
08:13:09 <andythenorth> "we have learnt to synthesise gold from lead"
08:13:09 <reldred> coz we dig up coal and sell it, and with the coal resource royalties in this state we end up with no monies
08:13:11 <LordAro> andythenorth: oh for sure
08:13:29 <Borg> I personaly never ever used any LLM yet.. because why should I? I enjoy solving interesting challenges
08:13:37 <andythenorth> "a great undertaking, for which the purpose is yet to be determined"
08:13:43 <reldred> i mean we end up with monies, I get paid fantastic amounts, but we're only keeping the lights on.
08:13:45 <LordAro> but the results are definitely real, even if the costs are still pretty high, they're not high for actual threat actors etc
08:14:11 <andythenorth> oh the actual South Sea Bubble quote is "a company for carrying on an undertaking of great advantage, but nobody to know what it is"
08:14:30 <reldred> my goal is for the next FY to find and get rid of enough stuff to lower our operating costs p/m than what my contract is worth.
08:14:38 <Borg> peter1138: any luck w/ my patch? ;)
08:14:56 <andythenorth> Borg 'getting good results from an LLM' is an interesting challenge
08:15:00 <reldred> i mean we've basically already done that this FY but I didn't come up with the plan that achieved it
08:15:44 <reldred> so in my mind it doesn't count
08:16:33 <reldred> i mean i don't have to, its not a kpi and its not like they can really get rid of me lol we're a 2 person team and that's the smallest we can be
08:16:38 <reldred> but I like the challenge
08:17:36 <Borg> andythenorth: really? I think its boring...
08:17:55 <Borg> you just need to ask.. and ask.. and ask... nothing doing thinking yourself
08:18:25 <Borg> but I suspect.. LLM is enjoyed by people with better language skills than engineering/math
08:18:48 <Borg> but.. whatever works for you...
08:19:13 <LordAro> that sort of attitude drops you into the "i am so very smart because i know science" crowd very quickly
08:19:32 <LordAro> which immediately discounts all opinion privileges ;)
08:19:42 <Borg> LordAro: im not very smart.. but I enjoy science and math
08:20:06 <reldred> I guess in a business setting like what andy's described 'i dont want to write css and this means i don't have to', where there's no 'technical challenge', it's just tedious shite that stops you from solving the real problems... then yeah sure.
08:20:11 <Borg> and im defiinitly better w/ those compared to humanistic skills.. or whatever that thing is called
08:21:04 <Borg> LordAro: but you are right.. such opinions are quicly attacked by crowds of quiche eaters [;
08:21:04 <reldred> I don't have to do any of those things though. Sure I could use it to write my emails to account managers and suppliers but... I'd personally prefer to use my own words.
08:21:54 <reldred> I will probably outsource writing flowery corporate BS for some of our dumb KPI's that force us to write articles hilight us doing our job we're being paid for
08:22:06 <reldred> I just see no need for it
08:22:08 <Borg> reldred: haha ;D thats good usage for those indeed
08:22:29 <Borg> or.. I heard people love to use ChatGPT etc to do corporate bullshit trainings too
08:22:43 <Borg> at least.. this is good use for those
08:22:46 <reldred> But part of me is like, I need to know how to write that dumb flowery corporate BS because I *also need to know how to speak that dumb flowery corporate BS*
08:23:42 <reldred> coz while it's rare I go to corporate HQ (despite being literally right across the river from my house), when I do I'm liable to rub shoulders with those people and I need to know how to navigate those waters and advocate for what we do
08:24:25 <Borg> andythenorth: and its all right.. we can have interests in everything..
08:24:30 <andythenorth> it was pretty close to the "spec driven development" that's currently very high fashion
08:24:34 <Borg> we would run out of fscking time..
08:24:52 <Borg> so many typos today :P I didnt slept well
08:24:54 <reldred> but, I can afford to be a bit blase about AI (I mean, I still hate it), because my job cannot be replaced by it.
08:25:05 <andythenorth> very few jobs can be replaced by AI
08:25:17 <andythenorth> that's just a cover for keeping FAANG share prices high
08:25:22 <Borg> andythenorth: hard to say [;
08:25:36 <Borg> well I dont like the whole idea of LLMs..
08:25:48 <Borg> I bet they have they uses.. for filtering out huge datasets
08:26:00 <Borg> and similar pattern recognitions..
08:26:20 <Borg> but to outsource thinking to AI? hmm...
08:26:49 <reldred> I think they're violently underwhelming and give unreasonable amounts of confidence to people who are violently incompetent - managers or engineers.
08:27:07 <LordAro> i suspect it's less that there are (many) new types of jobs that can be replaced, but more that it's lowered the cost of replacing jobs that could already be automated
08:27:18 <LordAro> i.e. it's changed the values in the xkcd table
08:27:24 <andythenorth> yes, they substitute the simulation of critical thinking for people who lack critical thinking
08:27:28 <andythenorth> 'studies are emerging of this'
08:28:03 <reldred> its only lowered the cost because the people making this stuff aren't charging enough to make a profitable business. I suspect that reality is going to come crumbling down in the next year or two
08:28:22 <reldred> if openai or anthropic want to go publicly traded then they're actually going to have to file proper financials
08:28:32 <andythenorth> $10 / hour compute cost allegedly on my $20 / month GPT subscription
08:28:34 <reldred> and people will realise its a turd wrapped in gold foil
08:29:03 <Borg> andythenorth: it might be other tactic
08:29:13 <reldred> I mean, openai commited to $1.2T of spend for this year and only raised $120B
08:29:35 <Borg> like trap... everyone will start to use AI.. to the point that even basic stuff will need it.. and then you rise price 20x times or more
08:29:52 <Borg> 10 years ago I already saw people.. when google was out. they couldnt do work
08:30:06 <Borg> why taking own notes... google have everything
08:30:12 <Borg> w/ AI.. it brings you to whole new level
08:30:16 <reldred> oh yeah, most orgs if there's a M365 outage basically come to a standstill.
08:31:05 <reldred> and usually at least they still had their phones but then hey lets migrate voice to teams calling
08:31:28 <reldred> and then a global m365 outage happens and your deskphones stop authing lmao
08:31:44 <reldred> and teams goes spinning wheel of death
08:31:47 <Borg> all my stuff is decentralized.. as far as I can..
08:31:54 <LordAro> centralisation is great for businesses
08:31:58 <LordAro> terrible for everyone else
08:32:08 <Borg> centralisation is great for capital owners
08:32:11 <Borg> its bad for everyone else
08:49:48 <andythenorth> the systemic risks are currently enormous but eh
08:50:14 <andythenorth> one bad day in AWS US-East could be temporary civilisational collapse
08:50:19 <reldred> nothing will happen til people start losing their retirement funds over it
08:50:29 <andythenorth> but eh' $0.00002 per compute hour
08:50:34 <andythenorth> and I don't have to own any Dell boxes
08:50:53 <reldred> yeah don't buy any powerstore's atm those things are flaky as fuck
08:50:56 <andythenorth> [even without hallucinating robots, the systemic risks are enormous]
08:51:28 <reldred> we bought four recently, two for each DC, and they've all had like a half a dozen service calls from dell and unisys each and that's in like their first six months of operation
08:51:49 <reldred> and now we're getting ready to roll out a heap more storage refresh work
08:52:04 <reldred> new datadomains and isilon replacements
08:52:29 <reldred> not sure if the isilon replacement is just a newer model isilon or some other product, idk, i don't care much about storage
08:52:47 <reldred> all our r650/r660's have been fine though
08:53:13 <reldred> and we've got a rack full of r760's with dual gpu's in them
08:53:21 <reldred> like a half a million dollar rack right there
08:53:28 <reldred> not used for ai fortunately
08:53:50 <dwfreed> did they fix the replication problems on datadomains, or are they still slow as fuck to replicate a whole array to another one?
08:54:10 <reldred> no idea, I just rack the suckers
08:54:23 <reldred> maybe configure some ip's on them when they go in
08:54:52 <reldred> networking is my wheelhouse, but I don't have to get into the weeds with it with this new job of mine
08:54:53 <__abigail> How many hours do we reckon these have hahaha
08:55:17 <dwfreed> at one company, we had to replicate the whole array of one datadomain to another temporarily so the first one could be reconfigured (which required wiping the storage), and then replicate it back
08:55:23 <reldred> __abigail: bout a few
08:56:01 <dwfreed> the replication was so slow that our regular backups (which weren't that much data!) were putting in more data than the replication could put out to the temp one
08:56:13 <__abigail> I like that they have 11 drives instead of 12
08:56:41 <reldred> yeah i know nothing about hp equipment
08:56:50 <reldred> and even less about storage
08:57:06 <dwfreed> We ended up having to turn off our backups for a few days so the replication could finish
08:58:06 <reldred> i must say I do enjoy my proxmox+ceph hyperconverged setup I'm running
08:58:08 <__abigail> Unfortunately I don't drive so I can't get one lol
08:58:24 <reldred> __abigail: you'd need something with a fibrechannel card in it anyway to use that.
08:59:21 <reldred> and, and I've learned this the hard way, you do not want to bring enterprise equipment home to run it, it's not like a puppy, you will get sick of it VERY quickly.
08:59:35 <reldred> there's a reason I use a trio of little 1L mini PC's
09:00:26 <reldred> its one of the most tragic things about my current job, I have so many opportunities to bring home cool shit but I know from experience that its a Bad Idea
09:00:53 <dwfreed> I have an intel 1U sitting next to me
09:00:53 <__abigail> I found a local enterprise store that has a set of rack rails for the chassis so I'm picking it up on Wednesday
09:00:58 <reldred> shits unbearably loud outside of its natural habitat and it will absolute destroy your power bills.
09:01:16 <dwfreed> It doesn't bother me for the most part, and only pulls 150 W most of the time
09:01:28 <dwfreed> but also I live alone
09:01:41 <reldred> yeah we used to keep ours on top of the fridge
09:02:24 <__abigail> My parents have been very happy with my Jellyfin server so they will be fine to cover the extra power bills. Saves them a fuckload of subscriptions or whatever
09:02:34 <reldred> had a 2u server for a while a few years back, then at my last place I had a housemate who had a big 4u quad socket xeon beast with an ungodly amount of ram but he only ran that in short spurts as a home render farm since he was studyting vfx
09:02:44 <dwfreed> mine is in my office right next to my work desk
09:03:01 <dwfreed> but I can put on headphones and play music during work and not notice the fan anymore
09:03:19 <reldred> Yeah I don't like wearing headphones for long periods
09:03:29 <reldred> over the ear or earbuds
09:03:50 <dwfreed> I wear over ear headphones all day and often forget I'm wearing them
09:04:00 <__abigail> I've got a 4U chassis so there's decent-ish space for cooling
09:04:04 <dwfreed> When I move to Europe, I will probably move this server to a colo somewhere nearby
09:04:36 <reldred> I was distraught, I pulled like 30-40TB in 32 and 64GB sticks of DDR4 ECC, and I couldn't use any of it in my home lab.
09:04:49 <reldred> like a literal brick of ddr4 in my hands
09:04:57 <reldred> more like a cinder block actually
09:05:46 <reldred> but my bloody home lab gear uses SODIMM's not DIMM's
09:06:03 <__abigail> Luckily you can donate all of them to me instead
09:06:14 <__abigail> In around 2018 my high school's IT guys showed me the new 8x32GB set of DDR4 they just got in
09:06:57 <reldred> eh, would have been messy me taking them anyway. It's all getting shipped off to perth to get divvied up and reused
09:08:49 <__abigail> I need storage rather than RAM atm
09:08:49 <__abigail> I have 5TB of brand new 10 year-old HPE drives but they're stuck at my gf's place
09:08:49 <__abigail> I will have to get them all the way to my place on the train (after their first train ride already happened)
09:09:56 <reldred> I need so many things for my lab
09:10:52 <reldred> need to swap my m920q's for m90q's so I can get 10/11th gen intel, and dual nvme, need to get 64gb of ddr4 in the three of them, six 4TB nvme's, then there's the switching
09:11:38 <reldred> currently each of my nodes has a 256gb sata ssd for booting proxmox, a single 512gb nvme for ceph, and an old i5 8500t which doesn't support hyper threading
09:11:55 <__abigail> I have a single 256gb SSD right now lol
09:12:06 <reldred> but, it is running, and it is HA, and i do have jellyfin containerised with iGPU acceleration for transcoding
09:12:13 <dwfreed> I need some new CPUs for my lab as well; I can't run RHEL 10 >.>
09:12:28 <reldred> that jellyfin only has a single video on it though, I mostly use my friends lmao
09:12:35 <__abigail> I don't want to think about the power consumption of the GPU that my server came with but I managed to deal with NVIDIA drivers for the passthrough
09:13:06 <reldred> a lot of stuff with hardware passthrough is easier if you run containers instead of vm's
09:13:18 <reldred> but containers can't do live HA whereas vm's can
09:13:23 <reldred> so critical services are vm's
09:13:29 <reldred> stuff that needs passthrough are containers
09:13:41 <dwfreed> personally, I refuse to run containers on a proxmox host
09:14:01 <reldred> I only use them for what containers are better for
09:14:06 <reldred> eg, hardware passthrough
09:14:26 <__abigail> I have most things in a container, but my web server is on a VM
09:14:41 <reldred> or, critical services where hypervisor based HA would conflict with its application level redundancy
09:15:20 <reldred> back when I used to build virtualised pbx's we'd deploy them as a pair of VM's and then disable DRS on them. Or if we were running DRS we'd only deploy a single VM.
09:15:32 <reldred> Just depended on what the environment looked like
09:17:08 <peter1138> Oh, weather looks a bit pants tomorrow.
09:17:11 <dwfreed> reldred: meanwhile, researchers just published a NVIDIA bug where the GPU can be used to get root on the host
09:17:32 <reldred> dwfreed: fortunately, cybersec is not my problem π
09:17:43 <belajalilija> dwfreed: That sounds very Australian
09:17:47 <reldred> the team who look after that big ass VDI cluster can deal with that.
09:18:13 <reldred> god i don't even have to look after vdi clusters anymore
09:19:47 <reldred> loved my old job (not), had to look after the DC, had to look after the vdi cluster, had to look after the hypervisors, had to manage the customers coz they all hated the new account manager we hired because he had a gun to his head to try and squeeze them for cash, had to look after all the carrier connections, the voice solutions did get moved over to a team with a few people though so that got
09:19:47 <reldred> taken partially off my back.
09:20:24 <reldred> did all the network engineering for the two main dc tenancies we had and fifty plus customer sites
09:20:42 <reldred> :cryingbloodbutstillcool:
09:23:17 <dwfreed> btw, my thing against containers on a proxmox host is about security; if someone breaks out of the container, they now likely have root on the proxmox host, and thus access to all the VMs on it too
09:23:44 <reldred> vm escapes also exist
09:23:55 <reldred> this isn't a production environment
09:23:57 <dwfreed> they are *significantly* less common than container escapes
09:24:07 <reldred> this is my home setup
09:24:34 <dwfreed> so the only way I'd do it is inside a VM, or on a dedicated container host
09:25:13 <__abigail> My parents are the customers
09:25:13 <__abigail> The customers' TV does NOT like HLS so instead they have to start a recording then watch it live
09:25:35 <dwfreed> but also I hate how much things these days abuse containers instead of proper packaging, so I tend to avoid them anyway
09:26:30 <reldred> yeah at work I ran a few docker swarm environments but they were running on a few VM's setup with their own redundant external ingress proxies, setup in a totally seperate vlan, the ingress proxies had a pair of public IP's out of our /23 and strict ingress rules to seperate from the other customers. everything was all zoned off.
09:27:36 <reldred> modern proxmox now I think defaults to unprivileged containers and is a lot more hardened than it used to be but again - not running a customer+internet facing setup here.
09:31:18 <dwfreed> I use my server for work lab environments, so I have to be more mindful of things
09:31:41 <dwfreed> in addition to my personal stuff
09:31:50 <reldred> yeah work stuff never touchs personal stuff
09:32:25 <reldred> hell, I keep my work laptop segregated in its own vlan with no traffic allowed to traverse outside of it.
09:32:54 <reldred> it gets the internet, and 8.8.8.8 as a dns and it can fuck right off outside of that.
09:33:08 <reldred> no internal name resolution or vlan traversal
09:33:59 <reldred> i'm primarily a network engineer, so the point of the lab is to demonstrate on a very very small scale the sort of things I used to build for customers without me having to give up privileged design docs for customer systems
09:34:48 <dwfreed> my work is far more lax about some things (work laptop runs linux, I have full root, no nanny software or anything)
09:35:05 <reldred> so the proxmox setup is a rough analogue for a vsphere+vsan setup I used to deploy customer solutions to at my old work
09:36:21 <LordAro> being a sysadmin for developers definitely males things more complicated
09:36:32 <dwfreed> And I had the available resources on the server to be able to also use it for work, so might as well; this way they don't have to pay for me to have something somewhere, that I don't have as much control over, and would have much higher latency
09:36:44 <LordAro> have to give them admin, and many of them know *just* enough to be able to break everything
09:36:48 <reldred> and the network design is a rough analogue for things I used to design for customers too; concepts like trunking carrier handoffs over a wider L2 network, virtualised routing, etc. and while it doesn't currently do it, it'll be expanding out to demonstrate mlag etc. as I built a couple l2 clos topologies for customers.
09:36:49 <dwfreed> LordAro: yes, "I can't ship your workstation in production" is like a weekly explanation
09:36:49 <LordAro> then complain that nothing works
09:37:17 <LordAro> though many of them are males
09:38:23 <dwfreed> on my next hardware refresh I might get a server for the work lab (so I can update CPU to be able to run RHEL 10 in that lab), but it'll probably still be connected to my home server
09:40:35 <__abigail> LordAro: You're welcome π₯°
09:40:49 <dwfreed> maybe I'll get a fat twin from unix surplus so I have 3 systems
09:55:53 <reldred> obligatory weekly 'yolo merge pr15391 wen'
09:56:31 <reldred> i should compile my own local build of it
10:04:32 *** u0_a351 has joined #openttd
10:09:28 *** rito12 has quit IRC (Ping timeout: 480 seconds)
10:17:33 <reldred> i have once again starting compiling openttd in a single goddamn thread
10:26:32 <peter1138> At least that leaves CPU time for other things.
10:28:45 <reldred> I cancelled the build like 30% of the way through and restarted it with -j8 and it cracked it like a freakin egg
10:29:00 <reldred> this old 5800x3d is still such a workhorse
10:39:58 <peter1138> Urgh, SQL query timing out :/
11:02:08 <Borg> hah.. I never cared about such network separation between work and corpo laptop
11:02:32 <Borg> actually.. I always did most of a work on desktops :) just RDPed to corpo laptop for Outlook and teams
11:02:50 <reldred> heh, I never did, but now I work for a bonafide megacorp so I keep that shit JAILED lmao
11:03:05 <Borg> reldred: all my personal stuff runs on FW
11:03:18 <_jgr_> Badly configured developer laptops is a popular attack vector into organisations these days
11:03:28 <Borg> besides router runs having FW as well
11:03:44 <Borg> _jgr_: yeah.... thx to all that npm install shit and other stuff
11:03:56 <reldred> a routers firewall does nothing for two machines on the same broadcast domain
11:04:09 <Borg> reldred: thats why said.. desktop additionally runs its own FW too
11:04:17 <peter1138> That's why we all set up our home networks with VLANs and segmentation...
11:04:54 <Borg> my network is flat... VLANs are only on internet side.. because I was dualhomed for long time
11:05:30 <Borg> I went VPN route.. everything is plugged into VPN.. :)
11:05:36 <Borg> and thats where I run stuff
11:06:11 *** u0_a351 has left #openttd (rirc v0.1.7)
11:07:16 <peter1138> talltyler, very xarick updating all at once ;-)
11:09:16 <_jgr_> The "work: needs rebase" label does imply that that is what you should be doing...
11:09:24 <peter1138> It's okay, I even included a smiley this time to make sure everyone knows it's a joke.
11:09:47 <reldred> but its still fun to poke fun
11:10:00 <peter1138> I know, but not everyone else can.
11:10:57 <reldred> yes it is kinda interesting how we retained the ability to do that here in straya but a lot of americans for instance struggle with it.
11:11:15 <reldred> guess we didn't stray as far culturally
11:11:36 <reldred> or you know, throw all the tea in the harbour
11:11:56 <reldred> start a cheeky lil war
11:12:06 <reldred> beg the french for help
11:20:10 <talltyler> Well, somebody removed all the whitespace from settings_type.h π
11:20:28 * Borg is going to test his 'entry mode' RR :)
11:20:40 <talltyler> The worldβs easiest rebase, except for maybe when saveload entires get incremented
11:23:09 <talltyler> Right, thatβs all my rebasing done π
11:29:55 <peter1138> I'm still anti-#9832
11:42:14 <peter1138> LordAro, oof 280W FTP allegedly, probably not bad considering the lack of cycling.
11:59:40 <LordAro> oh hang on, you're ridiculously tall as well, aren't you?
12:04:59 <peter1138> (On average, I'm taller than most people)
12:44:12 *** SigHunter has quit IRC (Remote host closed the connection)
12:46:22 *** SigHunter has joined #openttd
12:57:42 <peter1138> Dunno but Github is being terrible righ tnow.
13:24:31 *** toktik is now known as Guest6984
13:27:55 <_glx_> haha updated but not rebased so preview won't show
13:29:55 *** Guest6984 has quit IRC (Ping timeout: 480 seconds)
13:40:42 <LordAro> anyone fancy renaming those blocks to actually be vaguely understandable? :p
14:05:25 <talltyler> I also rebased so preview should work now π
14:13:11 <talltyler> Rebased so preview works, also fixed the new flip setting's value strings
14:15:33 *** x1vi_zi has quit IRC (Quit: User went offline on Discord a while ago)
14:32:00 <peter1138> Hmm, what happened to my patch that measured throughput/speed at waypoints...
14:42:59 *** MinchinWeb[m] has quit IRC (Ping timeout: 480 seconds)
14:44:00 *** MinchinWeb[m] has joined #openttd
15:45:56 *** Wormnest has joined #openttd
15:48:55 *** gelignite has joined #openttd
16:11:57 <rito12> Are we still on trees?
16:13:37 <peter1138> Currently I'm on trying-to-stay-awake.
16:14:43 <andythenorth> I'm on can-I-nap?
16:18:37 * Philemon sings "NO NAP TILL BROOKLYN!"
16:37:28 *** u0_a351 has joined #openttd
16:38:35 *** rito12 has quit IRC (Read error: Connection reset by peer)
16:38:42 *** u0_a351 has quit IRC (Remote host closed the connection)
16:38:43 *** u0_a351 has joined #openttd
16:55:53 <_glx_> ok I just noticed another issue with deployments, when it succeeds all previous deployments are now marked inactive
16:56:09 <_glx_> was not the case with the old way
17:02:49 *** u0_a351 has left #openttd (rirc v0.1.7)
17:09:16 *** MinchinWeb[m] has quit IRC (Read error: Connection reset by peer)
17:09:33 *** MinchinWeb[m] has joined #openttd
17:31:10 <peter1138> The lines for Rating use the y-axis on the right.
17:31:15 <peter1138> But it's entirely unclear.
17:32:24 <talltyler> OpenTTD enters the era of Big Data
17:59:33 <_glx_> I hope it's the last one π
18:07:38 <peter1138> Date: Mon Jun 2 07:35:48 2025 +0100
18:07:43 <peter1138> Hmm, that was a while ago.
18:08:07 *** k_yro has quit IRC (Remote host closed the connection)
18:11:18 *** toktik has quit IRC (Remote host closed the connection)
18:16:34 *** milenayumekooy has joined #openttd
18:18:53 *** Flygon_ has joined #openttd
18:22:31 *** Flygon has quit IRC (Ping timeout: 480 seconds)
18:28:21 *** milenayumeko has joined #openttd
18:29:27 <olionkey> discord should've deleted that one
18:29:33 <olionkey> it deleted all of the other ones then
18:30:40 <peter1138> > Pouring a concrete foundation starts at around Β£650 for 10 x 10 foot, up to Β£3,000 or so for a large 24 square foot construction.
18:30:58 <peter1138> I'm pretty sure that 10 x 10 foot is more than 24 square foot...
18:36:49 *** ufo-piloot has quit IRC (Quit: you click on fancy icons. i execute code !)
18:37:01 *** ufo-piloot has joined #openttd
18:38:21 <peter1138> Hmm, anyone have a savegame with very busy stations?
18:42:41 *** ufo-piloot has quit IRC (Remote host closed the connection)
18:43:05 *** ufo-piloot has joined #openttd
18:43:53 <peter1138> Ah, the ProZone one will do.
18:56:19 *** MinchinWeb[m] has quit IRC (Ping timeout: 480 seconds)
18:56:22 *** MinchinWeb[m] has joined #openttd
20:18:47 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
20:23:12 *** Borg has quit IRC (Quit: leaving)
20:31:51 *** Flygon_ has quit IRC (Read error: Connection reset by peer)
21:05:48 *** WormnestAndroid has quit IRC (Remote host closed the connection)
21:06:08 *** WormnestAndroid has joined #openttd
21:16:56 <peter1138> Hmm, what's the correct way to saveload a conditional SLEG_STRUCT?
21:38:37 <_glx_> like ` SLEG_CONDSTRUCT("old_ai", SlCompanyOldAI, SL_MIN_VERSION, SLV_107),`?
21:40:26 <_glx_> or ` SLEG_CONDSTRUCTLIST("liveries", SlCompanyLiveries, SLV_34, SL_MAX_VERSION),` for the other conditional type π
21:41:42 <_glx_> oups second one is a list, but same idea
21:54:09 <peter1138> Not version conditional :)
21:54:36 <peter1138> I went with using STRUCTLIST instead of STRUCT, and setting the list size to 0 or 1 depending on whether it exists or not.
21:54:41 <peter1138> Probably not ideal.
21:55:11 <peter1138> Other places that do it conditionally have another saveload variablee that is tested (e.g. stations vs waypoints)
22:15:43 <_glx_> `SlStationFlow` does some conditional saving too
22:27:35 <peter1138> Yeah, that is actually a list though :)
22:37:56 <_glx_> lol, doxygen always complains π
23:10:34 *** MinchinWeb[m] has quit IRC (Ping timeout: 480 seconds)
23:11:39 *** MinchinWeb[m] has joined #openttd
continue to next day β΅
