Apparently, Valve keeps getting builds that were never run through Steam at all, which is how games end up with broken launch options or missing depots.

all 19 comments

sorted by: hot top controversial new old
[–] 107 points 4 days ago* (5 children)

I have somehow seen youtube channels of indie devs crashing out over how these are onerous requirements that they will never financially recover from and this means Valve hates indie devs.

While also complaining that their games/updates are taking too long to get actually screened and published.

I have since concluded that those indie devs are hacks, because none of this seems to be unreasonable in anyway.

As 9point6 says... yeah, most of this can pretty easily be handled by adding another step or to to your build automation process, or just, gasp! Do it manually!

Is... is it really that hard to know what your own games dependencies... actually are?

Genuienly ridiculous that Valve has to specifically mandate that uploaded games, you know, need to actually run, don't crash on launch... the bar is apparently in hell.

smdh

  • source
  • hideshow 5 child comments
  • [–] 30 points 4 days ago (3 children)

    We have plenty of CI that launches and tests our shit. Before any prod deploy, I always take 5-15 min to click around before running the deploy job.

    Like... how often are you deploying if you can't take the 5 min to do that? If you're a dev you probably already have the environment up and running.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 4 points 4 days ago (2 children)

    IMO it shouldn't be merged if it's not deployable. So same thing you said but test before merging, after merging it should start the deployment process automatically. The idea here is that ease of development/code changes would drive up the quality.

    If something gets missed, add it to your acceptance tests, it's not that common for things to have to have manual deploys. Dave Farley has a lot of content about CI on YT that I recommend watching (even pre-AI becoming popular)

  • source
  • parent
  • hideshow 2 child comments
  • [–] 5 points 4 days ago* (last edited 4 days ago) (1 child)

    it shouldn’t be merged if it’s not deployable.

    100 % agree. The purpose of post-merge testing is to have an extra layer of security to prevent bugs from leaking out into prod, not that you should ever consciously merge something that is not deployable. In my experience, testing on a feature before merge tends to focus on what you expect could be broken/bugged. The post-merge/pre-deploy testing exists to catch cross-feature bugs and in general find the stuff you never expected to break in the first place. Basically: Test your feature thoroughly before merging, then test the whole app without focusing on any particular feature before deploying.

    after merging it should start the deployment process automatically (...) it’s not that common for things to have to have manual deploys

    There's two sides to this: We have auto-deploy on merge to a staging environment, and a manual deploy to prod. That lets us keep a high rate of development, while we continuously test / use the staging environment internally. The manual prod deploy doesn't cost us any extra time, since it essentially consists of clicking the "deploy to prod" button. The advantage is that we typically get a couple days of delay from something hitting staging before it goes to prod, which has let us catch a couple embarrassing bugs before they hit customers. When that happens, we can improve our testing regime without the bug having ever hit prod in the first place.

    So basically, before I click the "deploy to prod" button, I spend maybe 15 minutes clicking around in the staging environment to double-check that all is good, then I double-check that we don't have any known blocking bugs (we just use a dedicated issue-tag for "I found a deploy-blocker on staging, don't deploy to prod before it's fixed.")

  • source
  • parent
  • hideshow 1 child comment
  • [–] 38 points 4 days ago

    "It works on my machine"itis has hit the indie crowd!!

  • source
  • [–] 33 points 5 days ago (3 children)

    I was about to say this just saved some test engineers' jobs

    But then realised it's an easily written post-deploy test automation stage stuck onto the end of the release pipeline

  • source
  • hideshow 3 child comments
  • [–] 21 points 4 days ago (2 children)

    That'll still fix the problem it's intending to solve - games that don't launch

  • source
  • parent
  • hideshow 2 child comments
  • [–] 5 points 4 days ago (1 child)

    As long as the developer uses a different PC with a different user account name. Otherwise it can still suffer from "it works on my PC" syndrome because they used resources that are local to their device and not included in the game

  • source
  • parent
  • hideshow 1 child comment
  • [–] 6 points 4 days ago

    I've not heard of release pipelines running directly on a dev's own PC before but sure, if they're running it on their own PC it's not the best test. It's still better than not running it at all which sounds like has been happening

  • source
  • parent
  • load more comments (-1 replies)
    [–] 13 points 4 days ago

    The onions voting simulator wasn't that bad, come on.

  • source
  • [–] 8 points 4 days ago (1 child)

    What’s a depot in this context?

  • source
  • hideshow 1 child comment
  • [–] 13 points 4 days ago

    According to SteamWorks Documentation: A depot is a logical grouping of files which are all delivered to a customer as a single group.

    In the simple case, that's probably an EXE and some DLLs plus maybe some graphics/music/etc. files (if they're not packed into the EXE) for a particular version of the game.

  • source
  • parent
  • [–] 5 points 4 days ago
    [–] 6 points 4 days ago

    They should charge a deposit for reviews. If it's clearly slop, then the deposit is forfeited. A review is cheap enough that even small-time studios should be able to afford this.

  • source
  • [–] 2 points 4 days ago

    I read the headline and was like they want devs to play Half-life?

  • source
  • [–] 3 points 4 days ago

    I would trust my pipelines that much

  • source
  • 78 comments in c/pcgaming@lemmy.ca view that post
    [–] 199 points 5 days ago* (last edited 2 days ago) (15 children)

    Meanwhile we were so afraid to hit submit we tested for months even if we were just launching a playtest build. Sweating before every early build submission and our steam review.

    Even if it was actually a rubber stamp im glad we took so much care and am glad steam is setting up more tests like this.

    Edit: i would like to add our reviewers were super kind. Our unreal build had some obscure issue and they fixed it for us, pushed it for us, and showed us clearly what it was so we understood. But. Dayem. Thats good service and support ngl.

  • source
  • hideshow 15 child comments
  • [–] 47 points 5 days ago (9 children)

    Glad there's some devs who do care. Do You mean any specific title here?

    This new law is probably aimed against ai slops or/and super bugged self crashing crap... bethesda :V

  • source
  • parent
  • hideshow 9 child comments
  • [–] 26 points 4 days ago (8 children)

    I am a co founder of Hermertia Wars. Bunch of indie devs making our dream come true by working on an Action RTS that takes characters from our 13+ year lore playing minecraft community, and makes them heros you control while recruiting villagers, soldiers, and directing economy. Like AOE/warcraft 3 meets smash bros. Early game. We have a long project in mind. Would love new playtesters or meet folks interested in indie dev :). We are also a workers cooperative which is the best part.

  • source
  • parent
  • hideshow 8 child comments
  • [–] 6 points 4 days ago

    You got me intrigued.
    Indie, the game type, you being in a workers cooperative, aiming for a long project and thorough development sound so good.
    I'm going to join the playtest!

    What's the preferred way to provide feedback aside from Discord, which I don't use.

  • source
  • parent
  • [–] 6 points 4 days ago (1 child)

    For last few days and probably few weeks more, I'm deep diving another game that I just acquired (Soulframe). Hermertia Wars got wishlisted and installed and defo gonna check, may even do some infodump about at some point.

    Much thanks for doing games with passion.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 3 points 4 days ago (2 children)

    Looks interesting, just joined the playtest!

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 3 days ago (1 child)

    Wow amazing! Its tricky to play against AI at this point as its an arena game so hit me up if you wanna try with people. We have a tournament coming to help with that too!

    We are still easily a year away from serious marketing but with all subsystems built we are having so much fun with features and content now.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 4 days ago (1 child)

    Hell yeah, when it's further along be sure to talk about it on here more

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 2 days ago

    Cool will do :) we are a good year away from more openly marketing. We've done a few game dev conferences and learned so much but our game is ambitious as hell. Took us 2 years just to build sub systems. Finally done that and now we are on features and gameplay. Getting exhilarating. Our Monday night mini tournaments are so fun. But long way to go.

  • source
  • parent
  • [–] 4 points 4 days ago (1 child)

    But looking back, do you think that after you were done with testing, you could have said: "These tests we've run show that our game runs according to our standards" with confidence?

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 2 days ago

    I suppose! Another part of it is we are indie devs. Years of programming experience but not on games. So we are learning a lot as we go! And hope to make new friends along the way looking for a platform to do the same.

  • source
  • parent
  • [–] 119 points 5 days ago (5 children)

    Reminds me of the new Simon the Sorcerer game. The Linux build used a script to launch the actual game. The only thing the script did was to call the game's binary. Trouble was that they misspelled the binary. Just launching it once would have shown them the error.

  • source
  • hideshow 5 child comments
  • [+] -9 points 4 days ago (5 children)
  • [–] 39 points 4 days ago* (last edited 4 days ago) (3 children)

    Now imagine the developer who wrote that terrible code starting to use AI to excrete exponentially more terrible code...

  • source
  • parent
  • hideshow 3 child comments
  • load more comments (-1 replies)
    [–] 76 points 5 days ago (11 children)

    Ngl, kind of surprised that something like this wasnt the case already.

  • source
  • hideshow 11 child comments
  • [–] 44 points 5 days ago (10 children)

    Five minutes seem unreasonably paltry.

  • source
  • parent
  • hideshow 10 child comments
  • [–] 36 points 5 days ago (6 children)

    For real.

    If it were me, I'd probably be swinging the pendulum too far in the other direction, but I'd prefer that the developers clear the whole game to make sure the whole thing works. Submit enough gameplay data to prove that every achievement is unlockable.

  • source
  • parent
  • hideshow 6 child comments
  • [–] 10 points 4 days ago (2 children)

    Every achievement is too much. Some achievements are pure time sinks, and some are designed to be hard to be completed.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 13 points 4 days ago (1 child)

    I said what I said. Every. Achievement.

    (Half joking, but also half serious. It sucks when people accomplish something crazy difficult only to learn that the game is bugged and the achievement fails to unlock or the criteria can't be met)

  • source
  • parent
  • hideshow 1 child comment
  • That was my initial thought, however it may seem unreasonable as some studios develop for multiple storefronts and/or platforms. Should they play the entirety of their game on every platform or storefront, probably, but testing is like security, they don't really care about it because it's not a money maker.

    I miss the days of old when a game would launch mostly bug free (with a few minor exceptions).

  • source
  • parent
  • [–] 1 point 4 days ago* (last edited 3 days ago)

    I think it’s too harsh. The rule should be “devs should be aware of their slop game.” That’ll get ‘em. /s

    Edit: Downvoted for a comment making fun of Steam going too easy on sloppers flooding the indie scene…

    For the “sarcasm doesn’t need a /s” crowd; sometimes even the /s isn’t enough.

  • source
  • parent
  • [–] 1 point 4 days ago

    I'd hope it's on every update (from the screenshot in article seems like last 7 days also works), so they can't expect the devs to do long testing on small bugfixes.

    But also devs will probably do most of the testing outside of steam.

  • source
  • parent
  • [–] 64 points 4 days ago* (1 child)

    Four months or so Blizzard dropped a new season for Diablo 4 (along with a new game update), while being "Steam Deck Verified" and the new version of the game didn't launch at all on Linux anymore.

    The Proton team had to make an experimental Proton build which took around a day before it was fixed iirc.

    So really, even AAA devs can't seem to test launching their games for all scenarios.

  • source
  • hideshow 1 child comment
  • [–] 4 points 4 days ago

    A game can be "Steam Deck Verified" by Valve without the publisher making a request. In fact request by the publisher is the last option after automatic heuristics and Valve identifying a game as improtant.

  • source
  • parent
  • [–] 56 points 5 days ago (2 children)

    We are now living in time where the platform holder asks the developers to at least play their game for 5 minutes. The quality bar is that low.

  • source
  • hideshow 2 child comments
  • [–] 56 points 4 days ago (1 child)

    Tim McSweeney next week:

    Steam's terrible choice to demand that developers play their own game is incredibly anti-consumer and is cyber-bullying me specifically. How can anyone compete in such an unfair ecosystem?

  • source
  • parent
  • hideshow 1 child comment
  • [–] 37 points 4 days ago (1 child)

    Same reason Mario maker required completion before sharing the level you made.

  • source
  • hideshow 1 child comment
  • [–] 34 points 4 days ago (3 children)

    Everyone assumes it's just to stop the slop, but I had the same issue with gta4. Could not get past its launcher.

  • source
  • hideshow 3 child comments
  • [–] 22 points 4 days ago (2 children)

    I fucking despise Rockstar's launcher and their fucking idiotic overlay that is bound to the Home button against my fucking will.

    "Oh you know that button on your keyboard that you paid for? Yea that's like, our property while the game runs, fuck you."

  • source
  • parent
  • hideshow 2 child comments
  • [–] 7 points 4 days ago (1 child)

    Really the only option is to just pirate it at this point

    Actually that shouls have been the only option to begin with come to think of it

  • source
  • parent
  • hideshow 1 child comment
  • [–] 4 points 4 days ago

    Shame on me for buying Rdr2 on sale, should've just pirated it and ignored multiplayer.

    Thought multiplayer would be similar to the main game. My fucking ass. First random mission I tried had me hauling zombie corpses while getting shot by crazy priests controlling said zombies. Fuck that.

  • source
  • parent
  • [–] 61 points 4 days ago (7 children)

    Allegedly, Miyazaki is required to beat every game of From Software (Dark Souls, Sekiro, Elden Eing, etc.) before it can be released.

    Screenshots of an interview with a From Software staff claiming Miyazaki beating Sekiro is part of the release approval standard

    (There are occasionally glitches that allow you to bypass bosses. For this reason, they are called "Miyazaki's passways".)

  • source
  • hideshow 7 child comments
  • [–] 27 points 5 days ago* (4 children)

    5 minutes is paltry. Devs should be required to submit an uncut recording of the first 30 minutes of gameplay from launching via Steam before they can release a game or push an update.

    I don't think that's an unreasonable demand, cause if a game can't launch or survive 30 minutes of normal gameplay, it shouldn't be on the store for people to waste their time and/or money on.

  • source
  • hideshow 4 child comments
  • [–] 28 points 4 days ago (1 child)

    5 minutes is paltry.

    Yes, but I feel like this is intended more to curb AI-generated games that are submitted to Steam without any testing. This is just adding a layer of "a human (or something close-enough to a human) launched the game and didn't find it immediately unsuitable for public release", which I guess wasn't there before.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 16 points 5 days ago (3 children)

    heroes3 remake on Steam, few months ago :D

  • source
  • hideshow 3 child comments
  • [–] 12 points 4 days ago (1 child)

    What the actual fuck

    The last time I hit submit I had three thousand hours clocked on the game

  • source
  • hideshow 1 child comment
  • [–] 11 points 4 days ago (1 child)

    Do it like submitting a mario maker level. If you haven't beat it yourself, no upload for you.

  • source
  • hideshow 1 child comment
  • [–] 20 points 4 days ago (2 children)

    Honestly surprised this wasn't already a rule. The number of times I've bought something on day one and the first thing I see is a launcher error or a missing DLL is kind of absurd for a store that takes 30%.

    The Steam Deck Verified stuff makes it worse, because people trust that badge. If a patch breaks the game on Deck and the badge stays up for weeks, that's on Valve too, not just the dev. Five minutes of a human launching the actual depot would catch most of the dumb ones: wrong launch option, wrong exe, Linux script pointing at nothing.

  • source
  • hideshow 2 child comments
  • [–] 11 points 4 days ago

    I agree with you but also Steam has the easiest refund process in the entire industry. I've had old EA games become fully unplayable because of them switching origin to EAapp and had no way to even ask for assistance let alone get a refund. With Steam you just have to get past the bot and you'll be fine once you can explain the situation to an actual employee.

  • source
  • parent
  • [–] 8 points 4 days ago

    Considering in the past, even before AI, there have been games put in the store without an executable file, this rule makes sense and should've existed long ago.

  • source
  • [–] 21 points 5 days ago* (1 child)

    Can it verify that the player did more than just sit at the main menu? actually started the game?

    Also, hopefully this will cut down on long unskippable opening cutscenes.

    Valve should require video of a full game launch through Steam, through the main menu and then starting a new game, with at least five minutes of actual gameplay to verify that it's not a crashy mess.

    Also also, it should be required for every major update, not just the initial product listing, along with a list of test platform specs.

  • source
  • hideshow 1 child comment
  • [–] 3 points 5 days ago

    Ah yeah? make them 100% the game first

  • source
  • [–] 5 points 4 days ago

    5 MINUTES!? How about 5 months!? I mean WTF!?

  • source
  • [–] 2 points 4 days ago

    Lol.

    Can't say how many times I've screamed "DID THE DEVS EVEN PLAY THIS" while smashing my keyboard.

    Honestly, it seems like a lot of devs just get caught up in the coding aspect and forget about gameplay.

  • source
  • [–] 1 point 4 days ago

    Games with launchers becoming trendy in 1, 2, 3...

    That's a nice way to leverage Steam's own stats against dumbasses.

  • source
  • [–] 1 point 2 days ago

    They should have to play, actively, to at least 2x the refund window and to record all of this and submit it along with.

  • source
  • [–] 1 point 4 days ago

    Those are rookie numbers. I have thousands is hours in my steam game.

  • source