this post was submitted on 23 Jan 2024
131 points (83.2% liked)

Game Development

3444 readers
1 users here now

Welcome to the game development community! This is a place to talk about and post anything related to the field of game development.

Community Wiki

founded 1 year ago
MODERATORS
Ategon@programming.dev
LiquidityC@programming.dev
Frank@programming.dev
131
EDIT: Fake screenshot about some facts from the Palworld development, very loosely based on a really interesting blog post from the dev that's linked in the post body. (programming.dev)
submitted 9 months ago* (last edited 9 months ago) by Mikina@programming.dev to c/gamedev@programming.dev
36 comments fedilink hide all child comments
 

UPDATE: So, apparently it's mostly fake, taken from this article [translation] (where they even mention some kind of VCS).

However, even though it's not as absurd, it's a great read and a pretty wholesome story, so I recommend reading the article instead. And I'm even more convinced that this studio really does not deserve any of the hate they are getting.

Here is my summary of some of the interesting points from the article:

PocketPair started as a three man studio, passionate about game development, that couldn't find an investor for their previous games even though they've had really fleshed out prototypes, to the point where they just said "Game business sucks, we'll make it and release it on our own terms", and started working on games without any investor.

They couldn't hire professionals due to budget constraints. The guy responsible for the animations was a random 20-yo guy they found on Twitter, where he was posting his gun reload animations he self-learned to do and was doing for fun, while working as a store clerk few cities over.

They had no prior game development experience, and the first senior engineer, and first member of the team who actually was a professional game developer, was someone who ranomly contacted them due to liking Craftopia. But he didn't have experience with Unity, only Unreal, so they just said mid-development "Ok, we'll just throw away all we have so far, and we'll switch to Unreal - if you're willing to be a lead engineer, and will teach us Unreal from scratch as we go."

They had no budget. They literally said "Figuring out budget is too much additional work, and we want to focus on our game. Our budget plan is "as long as our account isn't zero, and if it reaches zero, we can always just borrow more money, so we don't need a budget".

For major part of the development, they had no idea you can rig models and share animations between them, and were doing everything manually for each of the model, until someone new came to the team and said "Hey, you know there's an easier way??"

It's a miracle this game even exists as it is, and the developer team sound like someone really passionate about what they are doing, even against all the odds.

This game is definitely not some kind of cheap cash-grab, trying to milk money by copying someone else's IP, and they really don't deserve all the hate they are receiving for it.

all 37 comments
sorted by: hot top controversial new old
[–] mozz@mbin.grits.dev 63 points 9 months ago* (last edited 9 months ago) (1 children)

I think the lesson is not that bucket-o-flash-drives is a better way; I think the lesson is that you can make a not-ideal process work completely fine if you just keep focused on the main point. People made successful software way before version control existed. It just makes it easier but that's all it does.

[–] half_built_pyramids@lemmy.world 15 points 9 months ago (2 children)

Romero talked about how they would just pass floppy disk to each other. It's just bigger disks now.

[–] mozz@mbin.grits.dev 20 points 9 months ago* (last edited 9 months ago) (2 children)

Linux was written up to about version 2.2 or 2.4 or thereabouts with no version control, just diff and patch and email. They invented git because at a certain point they wanted automated tools to make easier and more automated their way of working (which none of the suitable VCSs of the time were capable of), but it wasn't like they couldn't do the job until the tools existed.

[–] EmergMemeHologram@startrek.website 10 points 9 months ago (1 children)

Emailing patches is shockingly similar to git tho

[–] 4am@lemm.ee 9 points 9 months ago

Exactly. They did what we all wish we could do, they took a process that worked for them and automated it.

Thankfully they shared it with the world, and it works for most of us, too.

[–] theterrasque@infosec.pub 1 points 9 months ago (1 children)

https://git-scm.com/book/en/v2/Getting-Started-A-Short-History-of-Git

They invented git because bitkeeper withdrew support

[–] mozz@mbin.grits.dev 4 points 9 months ago* (last edited 9 months ago)

Yes. The little word "suitable" was doing a lot of work in my explanation, it's true. BK was invented by a kernel developer for pretty much exactly the reasons I explained; I just glossed over the part of the route that went diff+patch -> BK -> git as a detour. I was actually a lurker on the linux-kernel mailing list while all this was going on and saw the fireworks in real time.

It's not exactly accurate to say BK withdrew support. Larry McVoy was such a pain in the ass about wanting to control how people used his product that the kernel developers felt the need to write their whole own solution as opposed to continue putting up with him. (Specifically, his pulling Andrew Tridgell's license for dubious reasons was the straw that broke the camel's back.) RMS actually wrote a sarcastic open letter thanking McVoy for providing a good lesson in why it's a bad idea to let your important stuff be dependent on proprietary software.

But the point remains; BK was invented by a member of the kernel community specifically because none of the existing solutions were usable for them, after more than 10 years of one of the biggest and most-distributed software projects in the world having no source control whatsoever.

[–] SurvivalMariner@lemm.ee 7 points 9 months ago* (last edited 9 months ago)

Project zomboid had to start again after flat got burgled and laptop gone. Offsite backups are key for theft and fire. Version control is the easiest and cheapest way.

Someone always knows someone that drinks, smokes and eats crap and lives until mid 90s. Doesn't mean it's good health advice.

Beware anecdotal evidence.

[–] Mikina@programming.dev 11 points 9 months ago (2 children)

I love this so much :D That reads like something I'd expect from ZA/UM, but it also thankfully alleviates most of the major issues I had with the game, which I've already talked about here on Lemmy. I really liked the game, but there was a lot of red flags point to it being just a quick corporate cash grab, where they decided to basically re-skin heir previous game based on with as low effort as possible, to quickly sell it and cash in on the Pokemon thing. It just smelled with corporate greed, and that they did not really cared about the game too much.

But assuming this screenshot is true, I'd say that it's clear that it wasn't development driven and pushed by corporate greed, but really just a few of guys trying their best.

[–] bassomitron@lemmy.world 5 points 9 months ago* (last edited 9 months ago) (1 children)

I guess I'm a little confused, because wasn't their previous game Craftopia? I'm fairly certain that game sold relatively well for an indie game, at least 25,000+ copies if you base it off the All Time Peak Players on Steam Charts. For a small team of just a few people, that's a decent chunk of money (I think it sold at $30 around launch, so roughly $750k for 25k copies sold before Steam takes its cut). Craftopia came out in 2020, so they're saying they've learned virtually nothing in 4 years, not counting the dev time Craftopia had?

I enjoy Palworld, I think it's a fun game that has a lot of potential, but I'm not sure I'm fully buying into some of these responses.

As a side note, a lot of Craftopia people complain they abandoned that game, but looking on Steam it shows several recent updates across the last year, with one even coming out just yesterday and a huge one in November 2023 and a new roadmap posted in December 2023. So, I'm not really sure where those players are coming from regarding that.

Regardless, I'm looking forward to what Palworld grows into, as I really do think they have something rather special here. It's got a lot of rough edges and a couple core design problems, but those can eventually be addressed with some hard work. Hopefully they use the massive cash influx they've achieved with their recent success and hire some competent, seasoned developers to come in and get their shit in order. I'm not holding my breath too much, though (remember when we thought Valheim devs would spend their game success lottery money to massively boost that game's content and polish?).

[–] Mikina@programming.dev 3 points 9 months ago

It turns out that most info from the screenshot is false, there's a better article that's written by the actual developer linked in the updated post.

He did talk about them not having a budget plan, which was a fairly long part of the article, but can be summed up like this:.

Figuring out budget is too much additional work, and we want to focus on our game. Our budget plan is “as long as our account isn’t zero, and if it reaches zero, we can always just borrow more money, so we don’t need a budget

He also further down mentioned actual numbers of how much went into the development:

Judging from Craftopia's sales, it's [the budget] probably around 1 billion yen... Because all those sales are gone.

[–] half_built_pyramids@lemmy.world 3 points 9 months ago (2 children)

Next big game: Digimon, but with guns

[–] ogeist@lemmy.world 6 points 9 months ago

They already have guns, looking at you Gundramon

[–] Kolanaki@yiffit.net 2 points 9 months ago* (last edited 9 months ago) (1 children)

They didn't already do that? 🤔 Pretty sure I remember a digimon that was like a little cowboy with revolvers.

[–] smokin_shinobi@lemmy.world 2 points 9 months ago

They turn into mechagodzilla eventually.

[–] GammaGames@beehaw.org 10 points 9 months ago

5M copies sold, anyone who learns git is a schmuck!

[–] alilbee@lemmy.world 7 points 9 months ago (2 children)

I'm a DevOps person by trade, and I have been playing a lot of Palworld. This is my worst nightmare and I have no idea how any team bigger than one person could have done anything without basic source control. Guess it just goes to show that nobody cares about the details as long as you ship.

[–] 9point6@lemmy.world 9 points 9 months ago (2 children)

My realisation long ago is that the games industry just doesn't work like the rest of the software engineering industry. The most cowboy engineer you've ever worked with is a voice of reason in the game dev industry.

[–] alilbee@lemmy.world 2 points 9 months ago

Yeah, I have a colleague or two who have worked in that space. You could not pay me enough to work with their tools, conditions, and practices. Guess I'm in the wrong sub for that opinion, but I'm just a wanderer stopping by.

[–] SurvivalMariner@lemm.ee 2 points 9 months ago

Games industry is mostly binary files. Especially in Unreal. Perforce is popular from what I've heard from those in the field.

[–] Mikina@programming.dev 6 points 9 months ago* (last edited 9 months ago) (2 children)

It turned out it's not true, they did use VCS. However, they mention a pretty horrifying story about VCS nonetheless.

They were a team without prior or professional gamedev experience, and they were using git. The first senior engineer, and first member of the team who actually was a professional game developer, was someone who ranomly contacted them due to liking Craftopia. But he didn't have experience with Unity, only Unreal, so they just said mid-development "Ok, we'll just throw away all we have so far, and we'll switch to Unreal - if you're willing to be a lead engineer, and will teach us Unreal from scratch as we go."

And then, they also mention this:

Surprisingly, [the new engineer] had no experience using the version control system git.

According to him, Perforce seems to be a better match for Unreal Engine.

But Perforce is too expensive. This is not the amount that a company like us would pay.

If you can't use Perforce, you should at least use svn instead of git.

Fully trusting his words, I also migrated my version control system from git to svn.

[–] alilbee@lemmy.world 5 points 9 months ago

God, I need a drink or two after reading that. Just chaos.

[–] Lesrid@lemm.ee 1 points 9 months ago (1 children)

I haven't touched SVN since my mod installing spree in Garry's Mod.

[–] Mikina@programming.dev 4 points 9 months ago

On the other hand, now that I think about it, SVN may actually be better for Unity projects than git is, at least in some areas.

One major issue with Unity and VCS are the scene and asset files. Trying to mere scene changes when multiple people have worked on the same scene is hell, to the point where it's usually better to just choose one changset and manually re-do the other. I know there is a unity merge tool for that, but since you have no idea what exactly it did, it's been pretty hit or miss. SVN could solve that issue, since you can just lock files.

However, that still doesn't outweights the benefits of virtually every other feature of VCSes.

It's such a shame that Unity are greedy bastards that tend to buy out and heavily paywall amazing projects. I've worked with Plastic on one project, and it's amazing. I've really enjoyed the workflow, and the way the merging works is awesome. But then, Unity came and now it's unaffordable for anyone but larger teams.

Same with Parsec. Parsec has been an amazing alternative for Steam Remote, that had open source SDK and libraries to integrate directly into games. It was a perfect alternative for smaller teams that can't make proper multiplayer. And once Unity bought them, they've removed access to SDK only for companies that directly ask for it - which we (being a small student project done on our free time, that really could use MP since it's two player only local coop game) have done, mentioning that we're really just students and hobbyist.

They response? They basically said "Sure, we can give you access to the SDK, no problem. The first step is to pay us 1 000 000$ for it.". How can anyone be so out of touch?

Maan, I hate Unity.

[–] NostraDavid@programming.dev 4 points 9 months ago

I don't have a creative vision. I just want to make a game that people like

Unironically a good take.

[–] nix@merv.news 3 points 9 months ago (2 children)

What does VCS stand for in this case?

[–] NuXCOM_90Percent@lemmy.zip 9 points 9 months ago (2 children)

Version Control System

At this point: That should be a synonym for "git" unless you REALLY have a good reason not to.

[–] Amaltheamannen@lemmy.ml 5 points 9 months ago (3 children)

You underestimate how many companies with legacy code still run Subversion. Help.

[–] SurvivalMariner@lemm.ee 5 points 9 months ago

Even legacy codebases get migrated easy. SVN etc. belongs in a museum. Best red flag for dead end dev job.

[–] NuXCOM_90Percent@lemmy.zip 3 points 9 months ago

I got stuck organizing the migration at multiple companies. Believe me, I know how many people still use SVN (and CVS...)

And, in some cases, that is your "good reason not to" because of the disruption to development and needing to retrain devs. But it is also a migration that is well worth doing, if only because of how good Gitlab is.

[–] pixxelkick@lemmy.world 2 points 9 months ago (1 children)

You dont typically use git for VC on game files, git sucks a lot for binary files as it tries to diff them. You can use it for your codebase though.

You usually use SVN for game files like models, sounds, etc etc.

[–] BiggestBulb@kbin.run 2 points 9 months ago

Version Control System - something like Git, Subversion, Unity Teams' VCS, etc.

[–] AlexisFR@jlai.lu -1 points 9 months ago* (last edited 9 months ago) (1 children)

That's pretty telling of the state of the industry nowadays.

How did this game even got discovered?

[–] SurvivalMariner@lemm.ee 5 points 9 months ago

It's an indie. Indies just piece stuff together based on the experience of their devs.