Take-Two going through layoffs, Private Division and other labels affected

[Periple]

10 minutes ago, RocketRockington said:

What it doesn't mean though is that the devs have moved on to work on finishing new features - this would be the case if only a few developers were needed to fix bugs - because of limited bug count or because only certain devs knew how to do the work.  In this case, KSP2 being the buggy mess that it is, with issues in virtually every feature and system in the game, its unlikely that anything but a few artists or maybe a designer or two are working on anything but fixing the game.

In my experience, throwing more people to mess with code they’re not familiar with makes things worse! Certainly in the short term, and possibly even later! If you have too many coders on the same areas of code, they just trip over each other’s feet. And it’ll throw the entire production schedule out of whack!

Also, specialities. You don’t put a systems engineer on graphics, or the multiplayer specialist on physics!

I’m sure priorities have shifted and they’ve switched some people’s tasks, but if they have put everyone on bugfixing then I think that’s a panic move and will make things worse. It’s certainly not what I’d do under these circumstances!

Edited March 13, 2023 by Periple

[kerbiloid]

8 hours ago, Periple said:

That’s pretty normal really. You work on the whole game in parallel instead of doing things one at a time. You try to get it integrated as soon as possible. Normally the target is to have a vertical slice that implements all of the core systems and has a small set of assets at or near production quality as early as possible.

31 minutes ago, RocketRockington said:

This is completely standard software development practice.

Having semi-failing silicone rockets, test silicone bases?

[RocketRockington]

26 minutes ago, Periple said:

In my experience, throwing more people to mess with code they’re not familiar with makes things worse! Certainly in the short term, and possibly even later! If you have too many coders on the same areas of code, they just trip over each other’s feet. And it’ll throw the entire production schedule out of whack!

Also, specialities. You don’t put a systems engineer on graphics, or the multiplayer specialist on physics!

That's why I mentioned 'virtually every system and feature in the game has bugs'.  You don't have to switch people's specialities in that case - everyone has plenty of work to do in their own features.  And there are also almost always low-complexity code bugs that most engineers can track down and handle, or low-complexity asset most artists can - of course you don't have an artist fix code or vice versa, I'm not saying that.

That's one reason Nate's list of fixes mostly was not showstoppers though, and had a lot of not-very-important issues - likely the few developers who could untangle the really messy issues are swamped, so the other devs are fixing less key issues

9 minutes ago, kerbiloid said:

Having semi-failing silicone rockets, test silicone bases?

No, that part is all KSP2, who knows what else they thought was a good idea, maybe the colonies will be made of jello.  

But doing parallel bug fixing on one branch while QA tests a different one is standard practice - something even Intercept knows how to do.  Probably  

Edited March 13, 2023 by RocketRockington

[kerbiloid]

8 minutes ago, RocketRockington said:

No, that part is all KSP2, who knows what else they thought was a good idea, maybe the colonies will be made of jello.  

But doing parallel bug fixing on one branch while QA tests a different one is standard practice - something even Intercept knows how to do. 

I mean the stability of < 300 part rockets, shown by the youtubers.

[Periple]

1 minute ago, RocketRockington said:

That's why I mentioned 'virtually every system and feature in the game has bugs'.  You don't have to switch people's specialities in that case - everyone has plenty of work to do in their own camp.

But there are going to be people working on systems that aren’t in the EA. Multiplayer, progression, and resource extraction plus other colony mechanics for example. It doesn’t make sense to me to switch them to fix bugs in the EA! So I really don’t think there are just a couple of artists and designers working on other things than fixing bugs!

Also, most devs aren’t even programmers, they’re artists, tech artists, sound engineers, designers, producers etc. Not much they can do to fix bugs. Although I do hope they’ve got artists optimizing assets because some of them really need that!

We’re speculating of course but if they really have had to halt work on everything else to fix bugs then the project really is very badly managed — and I think that move will likely only make things worse. And as I’ve said, I don’t think the project can be THAT badly managed because what we’re seeing looks pretty normal for the studio size and timeline, assuming a full reboot in 2020 and allowing for the time it took to set up the new studio. 

(It would be pretty bad if you assume continuous development since 2017, but I really don’t think this can be the case!)

[RocketRockington]

37 minutes ago, Periple said:

Also, most devs aren’t even programmers, they’re artists, tech artists, sound engineers, designers, producers etc. Not much they can do to fix bugs. Although I do hope they’ve got artists optimizing assets because some of them really need that!

We’re speculating of course but if they really have had to halt work on everything else to fix bugs then the project really is very badly managed.

(It would be pretty bad if you assume continuous development since 2017, but I really don’t think this can be the case!)

I'm sorry, you seem like a nice and  bright person, less argumentative than most, but in my view anyone who thinks there's a chance that they're not very badly managed is very naive, in my opinion.  

Even if there are artists or what not that aren't working on bugs - they're not on the critical path.  We've been shown assets for ages for the colony game, for instance.  I wouldn't doubt that - if it was just a matter of the art being done - the colony game would be playable already.  The game's primary problem is with engineering and systems work, and I doubt a single engineer is working on anything except bugs and performance right now.  Their sales and reviews are terrible, it's clearly a situation where they're going to be trying to fix that issue before worrying about what's 6 months or a year down the road.  I've been on and around projects that were flaming wrecks and there's no long term thinking happening now - especially given that they're trying to hire 6 more engineers.  Note that if you see those hires get pulled down without new staff actually showing up, that means the project is 100% kaput.

They might not even be wrong - do you think if they don't improve the state of the game in the short term, there even will be a 6 more months for them?  I kind of doubt it.  Tbh from what I hear through the grapevine, I think all of Private division might be on a knife edge.

Edited March 13, 2023 by RocketRockington

[Periple]

4 minutes ago, RocketRockington said:

I'm sorry, you seem like a nice and  bright person, less argumentative than most, but in my view anyone who thinks there's a chance that they're not very badly managed is very naive, in my opinion. 

Then how do you explain that they’ve been able to build a studio from scratch and get to alpha starting from a new codebase and a mostly new team in three years? 

11 minutes ago, RocketRockington said:

The game's primary problem is with engineering and systems work, and I doubt a single engineer is working on anything except bugs and performance right now. 

Do you think it would be a good decision to switch an engineer who’s been working on, say, supply line automation or multiplayer, to fixing bugs in the EA?

7 minutes ago, RocketRockington said:

They might not even be wrong - do you think if they don't improve the state of the game in the short term, there even will be a 6 more months for them?  I kind of doubt it.  Tbh from what I hear through the grapevine, I think all of Private division might be on a knife edge.

I think that they will be in trouble if the project isn’t more or less stable in six months. And yes I think the big risk for the project is T2 pulling the plug on PD, but the state of KSP2 won’t be a big factor with that decision.

But I do think that as long as PD is functioning, they will keep KSP2 going. If the project is clearly off-track in six months, I expect changes in studio leadership or other shake-ups.

In any case we’re still just speculating. We’ll know this time next year which one of us was closer to the mark!

[RocketRockington]

33 minutes ago, Periple said:

Then how do you explain that they’ve been able to build a studio from scratch and get to alpha starting from a new codebase and a mostly new team in three years? 

Well, because I don't think they started from scratch, thats your theory.  I think they at the very least must have had prototypes and design docs.  In a 5 year project, that's typically the first two years.  

And regardless of 'getting to alpha' - their target was to get to (at the very least) a playable beta by now.  They launched a game that's at 51% negative reviews - and if it hadn't had 'Kerbal' in the title it'd probably be 90% negative.   

Also in games, often what seems like the last 10% of the work actually takes another 90% of the time.   A game that is mismanaged doesn't manage its scope or time well, and that's clearly happened here.  

33 minutes ago, Periple said:

Do you think it would be a good decision to switch an engineer who’s been working on, say, supply line automation or multiplayer, to fixing bugs in the EA?

In short, yes.  For multiple reasons.
1. As stated - doesn't matter what they do in a year if they get cancelled in the meanwhile.
2. An engineer who can work on gameplay features (eg supplies) can work on other gameplay features (eg: fuel flow).  Engineers are not so specialized that they work on one system only, ever.  That'd be dumb.  Similarly, if an engineer works on multiplayer, then they should have been working on the core foundation of the game, and should be able to handle many bugs.  If multiplayer isn't ALREADY integrated into the game.  Well, they're up feces creek, and the paddle was RUD'd.   There's no way they're getting it into the game in any reasonable time frame and without going dark for a long time, like NMS.  You don't code a real time physics sim that's meant to work across the network without building it for multiplayer at the start - those two systems HAVE to function together.  If you don't believe me, ask a current programmer - I'm not anymore but that doesn't mean I've forgotten everything I knew.
3. Its likely those engineers dropped what they were doing on future features to work on bugs for the last couple of months at least.  They won't be 'switched'.  They've already been switched.  They must have known what a dumpster fire they were launching, and had all hands on deck improving the release.  If not - their management is even more clueless than my low estimation of them.

33 minutes ago, Periple said:

I think that they will be in trouble if the project isn’t more or less stable in six months. And yes I think the big risk for the project is T2 pulling the plug on PD, but the state of KSP2 won’t be a big factor with that decision.

Yes - so we agree they should be doing everything possible to improve the stability of the build right now.  

And I disagree with you on that last one.  KSP2 is PD's biggest project - both in terms of dollar cost, with the massive overrun and the need to pay for/start a new studio, and the cost of buying the IP and with the future plans they must have for the IP.  They had more than just a sequel in mind.   Outer Worlds was a similar scope project, but Obsidian already had it well underway before they brought it to PD - hence why Obsidian kept the IP, which is now Microsofts, Outer Worlds 2 won't be published by PD.  Also because of that, the royalties go much more in favor of Obsidian.   Moon Studios thing is also similar in scope - but again, Moon Studio has much more of the pie in that case I believe, and also I don't know that one will ever see the light of day.  Everything else PD has done is comparatively small potatoes, and they've had no real successes  of any scale besides Outer Worlds, which again, is now lost to them and they didn't get that big a piece of the pie anyway - and they've screwed up the DLC for it too.

Why do you think the layoffs were focused on Private Division, and not more evenly spread out across Take2?  It's because PD has been a money loser for them, and they're trying to tighten the belt.

Edited March 13, 2023 by RocketRockington

[Fullmetal Analyst]

6 minutes ago, RocketRockington said:

2. An engineer who can work on gameplay features (eg supplies) can work on other gameplay features (eg: fuel flow).  Engineers are not so specialized that they work on one system only, ever.  That'd be dumb.  Similarly, if an engineer works on multiplayer, then they should have been working on the core foundation of the game, and should be able to handle many bugs.  If multiplayer isn't ALREADY integrated into the game.  Well. they're up feces creek, they're no way they're getting it into the game in any reasonable time frame.  You don't code a real time physics sim without building it for multiplayer at the start  - those two systems HAVE to function together.  If you don't believe me, ask another programmer.

i can confirm this xD

[J.Random]

38 minutes ago, Periple said:

build a studio from scratch

[Periple]

3 minutes ago, RocketRockington said:

Well, because I don't think they started from scratch, thats your theory.  I think they at the very least must have had prototypes and design docs.  In a 5 year project, that's typically the first two years.  

It's only been three years since IG started in early 2020. How long do you think it would take to get recruitments done, infrastructure built, and production organized?

– I agree that they must have had design docs.  I don't know how useful any prototypes they may have had would have been though, if they were rebooting! They certainly wouldn't help much with identifying bottlenecks or such, they'd probably mostly be useful for prototyping gameplay... and we already have KSP1 for that!

5 minutes ago, RocketRockington said:

And regardless of 'getting to alpha' - their target was to get to (at the very least) a playable beta by now.  They launched a game that's at 51% negative reviews - and if it hadn't had 'Kerbal' in the title it'd probably be 90% negative.   

No argument there! However that kind of slippage is pretty normal in my experience – not a sign of something being exceptionally badly managed!

6 minutes ago, RocketRockington said:

In short, yes.  For multiple reasons.
1. As stated - doesn't matter what they do in a year if they get cancelled in the meanwhile.

I may be misunderstanding you here, but do you mean that it's just important that they show to PD that they're doing everything they can, even if it doesn't actually help fix the product?

8 minutes ago, RocketRockington said:

2. An engineer who can work on gameplay features (eg supplies) can work on other gameplay features (eg: fuel flow).  Engineers are not so specialized that they work on one system only, ever.  That'd be dumb.  Similarly, if an engineer works on multiplayer, then they should have been working on the core foundation of the game, and should be able to handle many bugs.  If multiplayer isn't ALREADY integrated into the game.  Well. they're up feces creek, they're no way they're getting it into the game in any reasonable time frame.  You don't code a real time physics sim without building it for multiplayer at the start  - those two systems HAVE to function together.  If you don't believe me, ask another programmer.

If you haven't worked on a codebase before, you can't just jump in and start fixing things. It takes a while to ramp up, and helping you get ramped up takes time from the people who are familiar with the codebase – at the very least, it'll take them more time to review PRs because there will be more naive mistakes there. Moreover, putting more people on the case will only help if there are underserved areas they can jump into. If you already have somebody working on fuel flow, it doesn't help to throw somebody else to work on it – they'll just trip over each other's feet!

Also I'm pretty sure that MP is already integrated into the game and has been from the start, because they know it has to be, and MP engineers have been on board since 2020. They're going to have keep doing what they're doing precisely so they don't end up making decisions that break MP!

What I would have done in this situation is just a few strategic reassignments. I'd inventory the worst bugs that need to be fixed, the systems where they are, and then take the two or three best engineers I have for each of these areas and put them to work on them, both hands-on in the code and especially to support others already working on them with code reviews and such. If it so happened that my best systems engineer was working on supply route automation and I needed her on resource flow, then yeah I would reassign her and push supply route automation further down the roadmap. I would also check if we needed more people in specific roles and push hard to get them filled over the next quarter or so, within the economic constraints I have to deal with.

I don't think IG's engineering team is so thin that they could all be usefully assigned to bugfixing the EA. That would be a panic move and make things worse.

10 minutes ago, RocketRockington said:

3. Its likely those engineers dropped what they were doing on future features to work on bugs for the last couple of months at least.  They won't be 'switched'.  They've already been switched.  They must have known what a dumpster fire they were launching, and had all hands on deck improving the release.  If not - their management is even more clueless than my low estimation of them.

I don't think the game would look like it does if it had had a couple of months of bugfixing. To me it looks like a fairly early alpha: there wouldn't have been an integrated build for them to bugfix!

Also, if you think they got to alpha already several months ago – say, in 2 1/2 years of development – how does that square with your "terribly bad management" theory, especially accounting for what it takes to set up a studio, organize production, fill the key roles, set up the infrastructure etc?

15 minutes ago, RocketRockington said:

Yes - so we agree they should be doing everything possible to improve the stability of the build right now.  

Yes!

15 minutes ago, RocketRockington said:

Why do you think the layoffs were focused on Private Division, and not more evenly spread out across Take2?  It's because PD has been a money loser for them, and they're trying to tighten the belt.

They weren't. PD suffered a lot less than most other departments at T2! I don't know why the headlines made it look like they were particularly hit.

– Finally: I think we're actually in broad agreement about most things! It's just that you interpret the known facts in a really pessimistic/negative light, while I'm much more optimistic. 

However I'm still not able to reconcile two of your core assumptions with the known facts: that KSP2 is a re-skin of KSP1, built on the same codebase, and that KSP2's production has been horribly mismanaged. (Or did you change your mind about the KSP1 codebase? You seem a little less certain about it in the post I'm quoting than before.) 

I don't think the former can be the case because of various observable technical differences (savegame format, loading times, the way background vessels are handled etc) and because I can't believe that PD would be so utterly clueless that after ST failed to make KSP2 that way, they would go "Hey, let's continue to do EXACTLY THE SAME THING but in an internal studio, I'm sure it'll work out this time!"

And I don't think the latter can be the case, because they were able to found a studio, organize production, fill key roles, set up the infrastructure, and get to alpha in three years. That's not easy and you can't do it at all with terribly bad management! I know I'd be daunted if somebody asked me to do that! I mean, I think I could do it, maybe, but I think there's also a pretty big chance that I'd fail – all it would take is one bad hire into a key role, and hiring is always a bit of a crapshoot no matter how well you do your homework!

In sum, from where I'm standing KSP2 development looks pretty normal – not exceptionally good but not exceptionally awful either, and certainly not THE GAME IS DOOMED THE SKY IS FALLING awful. There are two big mistakes I can see: the call to give it to Uber/ST in the first place, and the failure to manage expectations for the EA. Both unfortunate, neither all that unusual. 

[RocketRockington]

4 hours ago, Periple said:

– I agree that they must have had design docs.  I don't know how useful any prototypes they may have had would have been though, if they were rebooting! They certainly wouldn't help much with identifying bottlenecks or such, they'd probably mostly be useful for prototyping gameplay... and we already have KSP1 for that!

If they didn't do engineering prototypes to identify key issues - like 'oh hey our planet rendered can only run smoothly on a 4090' (a card that didn't exist 3 years ago) then again, that's a failing their project management.  Prototyping isn't JUST gameplay prototyping, especially when you have a systems driven game.  They *claimed* to have a working build, with at least Kerbin in the promotional footage for 2019.  It should have been much more than prototypes, since they *claimed* they'd ship in 2020. 

And they carried their project management over to Intercept.  [snip]

4 hours ago, Periple said:

Also, if you think they got to alpha already several months ago – say, in 2 1/2 years of development – how does that square with your "terribly bad management" theory, especially accounting for what it takes to set up a studio, organize production, fill the key roles, set up the infrastructure etc?

Because, first of all, they 'got to alpha' when they were supposed to get to beta.  Failing that is bad management.  Second - their 'alpha' isn't even a feature complete alpha.   Alpha typically means 'game is fully testable, although all content is not in'.  That means all systems are in. They've only gotten to an alpha of part of their game systems - and not even the most difficult one, multiplayer.

Finally - it's the same project management.  Same Creative Director.  Same executive producer until Nate Robison left.  Same Studio Manager.   Same Art Director. The same ones who screwed up things at star theory.  You're the one who's giving them a total pass for entirely messaging up their whole project once.  I'm not.  Games typically don't get a 'do over'.  And the fact that they messed up their do-over this badly?  Sorry man, but this is the height of bad management.  It's beyond bad management tbh.

4 hours ago, Periple said:

They weren't. PD suffered a lot less than most other departments at T2! I don't know why the headlines made it look like they were particularly hit.

Stats on this?  PD is a small division of Take2 and they seem to have lost several staff.   Unless most of Take2 lost >10% of their staff, I'm not buying this.

4 hours ago, Periple said:

However I'm still not able to reconcile two of your core assumptions with the known facts: that KSP2 is a re-skin of KSP1, built on the same codebase, and that KSP2's production has been horribly mismanaged. (Or did you change your mind about the KSP1 codebase? You seem a little less certain about it in the post I'm quoting than before.) 

I never said it was a re-skin of KSP1.   You said that Star Theoy's KSP2 was a reskin of KSP1.  And I don't see how that at all relates to whether they're badly managed or not.   

4 hours ago, Periple said:

And I don't think the latter can be the case, because they were able to found a studio, organize production, fill key roles, set up the infrastructure, and get to alpha in three years. That's not easy and you can't do it at all with terribly bad management! I know I'd be daunted if somebody asked me to do that! I mean, I think I could do it, maybe, but I think there's also a pretty big chance that I'd fail – all it would take is one bad hire into a key role, and hiring is always a bit of a crapshoot no matter how well you do your homework!

Companies can make a mobile game in 6 months if they wanted to.  There absolutely is no 'getting to alpha should always take 5 years, anyone who does it faster is a genius' which is what you seem to be implying.

When I'm working on a project, I look at the resources at hand, the core goals of the project (gameplay pillars, quality bar, target audience, etc), and plan out the best way to reach that goal with the resources at hand, or tell the people setting those goals that they're unrealistic.  When someone FAILS to do that - that's bad management.  EIther failing to get where you're meant to in the allotted time - even if you have to reduce your target - or making it super clear to upper management that its impossible to do what they're asking.   What clearly happened in Intercept's case is neither of those things - these people, who ALREADY had a chance to try to build this exact game, somehow told management they would be able to do it by end of 2021, KNOWING all the challenges they'd face - again, most games don't get the 'do over' and the fact that they did and flubbed it is a massive point against them.

Then, given a massive extension, despite all that - they failed again.  Then, given another massive extension - they failed again and released this garbage which, as I'll remind you, is a barely-alpha version of a cut-down feature set of their target game.  A game they CLAIMED to be able to release by 2021.   

[snip]   If you work at a company, and you say 'I'll be done by next week', and people count on you to get it done by next week - even if that 'next week' goal is very ambitious - and you screw it up, over and over - well, it's terrible time management.  And if what you end up delivering after multiple flubs is a big fail - even if the time frame was STILL ambitious - it's a big failure of time management, especially since you had the chance to figure out what you'd been doing wrong all the other times.

Edited March 13, 2023 by Snark
Redacted by moderator

[Snark]

Some content has been removed and/or redacted. Please don't make things personal, and try to stay on topic.

The topic of this thread is about Take Two and Private Division. The topic is not how you feel about other people posting in the thread.

Thank you for your understanding.