Can too slow of development kill KSP?

[MajorSpittle]

I love this game but I never play it because there is no Multiplayer (I'm a social person) and the lack of a "story line" or "purpose".

I see the developer blogs and a cheer on the developers (last update was GREAT) but I find it disconcerting the number of people working on the game. Nobody is making any Mining or Colonization parts. Nobody is working on alternate Space centers or Kerbal races. Nobody is working on Kerbal A.I. or Cities and such to dress up the game and lay the foundation for a more immersive environment.

Please don't take this as me taking shots at the developers because I don't mean it that way. I just see so much potential in the game and so much desire from the community to have the game expanded and developed faster.

I hope it is not money holding up the game. I have bought a copy for me and my brother. I would have NO Problem spending that much again for a Mining addon, or a "Colonization Part Pack", or a "Return of the Jebi" StarWars pack......

I don't know much about developing a game, but how can people develop "Kerthane" mods, "Dangerous Re-entry", "Planet Packs", and "Sound Packs" just in their spare time so quickly and the developers not be able to develop for sale mod packs as quickly to help propel production of the game.

I would love to hear from the developer as to where they picture this game being in 10 months? What it will look like and what the development path is? Will I be able to buy add-ons (yes I want to give you money) and will I be able to play online with family and friends?

I am very much looking forward to currency, just as I was looking forward to the asteriods and larger rocket parts. But I played maybe 3 hours after the last update because I had nobody to play the game with. I wish nothing but the greatest success for the developers and hope the community doesn't end up dying off from boredom. I know that sounds real bad to say, but I find myself checking on the game's progress less and less.

[Dedjal]

They are using Unity. Ksp is held together by the force of ducktape. The more they add to the game, the more laggier and prone to crashes it will become.

Unity is a great and 'simple' engine to use, but it is just not enough power to realise the dreams of the devs.

[MajorSpittle]

So the game will never be much more then it is now? Time to release this puppy then and start work on KSP II with a real engine.

[regex]

I don't know much about developing a game, but how can people develop "Kerthane" mods, "Dangerous Re-entry", "Planet Packs", and "Sound Packs" just in their spare time so quickly and the developers not be able to develop for sale mod packs as quickly to help propel production of the game.

Well, for one, modders don't generally have a team of QA people they answer to and can put out a fairly incomplete product and refine it as needed. I'll guaranty you that Kethane took a lot longer than a few months to get to the state it is in now. Deadly Reentry is the product of several people's work as well, over quite some time. You should take a good long look at modder development time, not to put out an initial product, but to put out a polished, relatively bug free and balanced piece of game, to get a solid grasp of how much time it takes the KSP devs to get something out there. Maybe then you'll understand why their small team of devs is actually working pretty hard.

[Vanamonde]

This discussion of the game's development has been moved to the development discussion sub-forum.

Squad is a small group making their first game. I don't think development is particularly slow, in proportion to the number of people working on it. But I must, once again, take issue with thecomparison to the work of mod-makers, which comes up again and again. This complaint fails to take into consideration that mod-makers are making creating alterations or additions to already working code. And who created, maintains, and updates the code that runs the game? Squad does. Of course it's going to be faster and easier to modify or add to a functioning system than it was to create the functioning system in the first place, so there's realy no basis for comparing the two.

To know more about where the game's development is going, follow the Devnotes section of the forum, and read over the planned features list.

[Dedjal]

The game will be more than it is today.

But since the limitations of unity is now felt by the devs. Development will be slower due to having to fix bugs, reprogram and streamlining the codes in the hopes of getting a little more extra out of unity.

And the more they add, the slower the development will become

[softweir]

Unity is a great and 'simple' engine to use, but it is just not enough power to realise the dreams of the devs.

That's putting it too strongly. There is a great deal that can be done with Unity There are some things Unity prevents, but it is progressing and the limitations are being pushed back with every year.

So the game will never be much more then it is now? Time to release this puppy then and start work on KSP II with a real engine.

No way Squad will do that! The moment they announce suspension of KSP 1, sales will fall through the basement. With significantly reduced cash flow they would be very hard-pressed to create a new game, with new code, from an engine they haven't worked with before. No, what will happen is that Squad will complete KSP 1, release it, debug it, and maybe start adding extra content. Then, and only then, they might consider KSP 2 with a new engine. Or maybe not - that is entirely up to them.

I know this thread started in the General Discussion forum so you won't have read the Suggestions forum stickies, but it's worth pointing out that Resources, mining and colonisation are all areas that are listed as "Do not suggest". This isn't because they Won't Happen (they may) but because they are suggested so often that there is very little point in discussing them yet again.

For what it is worth, Squad implemented Resources - and found that the implementation was "not fun to play", and decided to get on with other things and think about it again.

[Dedjal]

For what it is worth, Squad implemented Resources - and found that the implementation was "not fun to play", and decided to get on with other things and think about it again.

And then they implemented science because that, apperently, was fun!

[CaptRobau]

The game will be more than it is today.

But since the limitations of unity is now felt by the devs. Development will be slower due to having to fix bugs, reprogram and streamlining the codes in the hopes of getting a little more extra out of unity.

And the more they add, the slower the development will become

Bullcrap. KSP has made great strides in stability and performance, while still adding new stuff like larger parts and asteroids. It was a few months ago, but the devs said they still hadn't had to rewrite anything Unity-wise to completely fit their needs. Maybe that's changed in the meantime, but at least at that time Unity out of the box was still allowing them to do what they wanted. That means the engine works great for them.

[Sirrobert]

And then they implemented science because that, apperently, was fun!

No, they implented science because it is the core of the career mode.

As for the speed off updates, you have to remember that even big studio titles take several years to finish. KSP really isn't going all that slow. It just feels that way because we are there for the entire ride, rather than just the time between announcement and release that normal titles go through (a few months, during which you don't think about it that much)

[LethalDose]

Slow development of a game won't kill it, but it's certainly not healthy for the development either.

However, slow game development + trying to capitalize on early release + poor communication between the devs and fan base is really really bad for the game. I think a game could do alright with two of those three, but KSP is, IMO, dealing with all 3.

The game development is going slow, probably for several reasons. Based on information on the forums, SQUAD's site and interviews, I think the following facts are relevant:

• Squad isn't primarly a game studio
  
• The team working on KSP seems to be relatively small
  
• KSP got the greenlight to keep HarvesteR from quitting SQUAD several years ago
  
• The game is, admittedly, complex
  

A note on why Harv was going to quit, apparently SQUAD was kept taking on more projects, but wasn't hiring people to keep up with the increased work. (The source article is here: http://www.polygon.com/features/2014/1/27/5338438/kerbal-space-program). So it's well within the realm of possibility that the developers aren't getting the support they need to keep up an optimal development pace

I don't think the devs have good communication with the fan base. I've pointed out the problems I've seen in my previous posts. A summary would be:

• Devs rarely deign to make official announcements about the state of the game (HarvesteR's recent 0.24 update is noteworthy because it happens so rarely)
  
• The devs (or the mods via the devs) primarily seem to address issues only as a last resort.
  
• IMO, the devs do a poor job of addressing issues when they rarely DO come around to it. (See the resources issue, where HarV's statement was repeating the KerbalKon statement nearly verbatim, instead of actually explaining anything).
  
• Most announcement really seem to have a lot of "spin" to try to put KSP and SQUAD in the best possible light, instead of being direct.
  
• There's a lot of very vague communication
  
• There is no officially released definition of "scope complete". One that would be labeled "subject to change" would be more than we have now (the "Planned features" listed above is completely unofficial and cobbled from many sources, some probably obsolete)
  

Finally, early release. I don't have a problem with it. I think it's a great idea, because small developers can get the funds for development when they need them! But you have to make the release work for the good of the game. Extremely infrequent dev communication that is vague when it happens is not good for the game.

If the developer is clear about the game's features and constantly keeping the community updated with the game's current and planned future status, there's no reason for anyone to speculate about the state of the game, and players can make informed decisions.

If the developer is unclear about the game's features and does a poor job he community informed about the game's current and future status, speculation starts and as Harvester pointed out last week, speculation is not good.

So, I'm glad KSP had an early release because I like the game, but early releases are a two-edged sword. I think KSP is getting cut on the back-end because SQUAD didn't realize what they were getting into. I think KSP at release is going to be a good game, but I think it could get to release with much less choppy waters if SQUAD had been more open with the fan base.

Again, it's my opinion and an explanation of where it came from.

[Jean Deaux]

So the game will never be much more then it is now? Time to release this puppy then and start work on KSP II with a real engine.

*Engage deafening applause*

I don't fire up KSP much anymore for the same reason. The game is capable, and quite frankly, deserving of so much more.

[phoenix_ca]

Yeah. Right. Squad will write a new engine for "KSP 2". And pigs fly and fairies are real and I'm actually a two-thousand-ton bagel with little hands to type with on a keyboard.

[Dunbaratu]

This complaint fails to take into consideration that mod-makers are making creating alterations or additions to already working code. And who created, maintains, and updates the code that runs the game? Squad does. Of course it's going to be faster and easier to modify or add to a functioning system than it was to create the functioning system in the first place, so there's realy no basis for comparing the two.

The other thing people don't take into account is that a feature in the stock game requires that it works with every other feature in the stock game. This is as opposed to mods, which don't have to work with other mods because the mod's author can just say "well if you don't feel like running my mod then don't. If you like the other mod better then use it instead." That sort of approach doesn't work in the stock game. Also, mod writers don't have to find a good balance that makes everyone happy. If a mod changes the gameplay in such a way that only a minority of the players will like it, that's fine in a mod. It's not fine in the stock game to do that.

My only worry is that they might have reached market saturation already during early release. By the time they are ready to sell the final product at full price instead of the early access discount, there might not be anyone left who'd be interested in buying it who hasn't bought it already.

Edited May 27, 2014 by Steven Mading

[Xaiier]

So much facepalm.

I will never comprehend how people can have so little understanding of how game development on this scale works. It's not even worth trying to refute points.

[LethalDose]

So much facepalm.

I will never comprehend how people can have so little understanding of how game development on this scale works. It's not even worth trying to refute points.

Well, someone who knows better than anyone else here could enlighten us...

[Specialist290]

So much facepalm.

I will never comprehend how people can have so little understanding of how game development on this scale works. It's not even worth trying to refute points.

I'm going to second LethalDose's sentiment above. You wouldn't mind giving us the benefit of the doubt and presenting your side of the argument as well, would you?

[nobbers12345]

I kind of agree, perhaps not kill it, but definitely won't help the game when it finally releases.

With the implementation of steam sales, I think the amount of players will begin to plateau because anyone who will ever want it has already bought it.

However, the donations people will likely give could possible support the game as it drags along.

[Nobody]

I love this game but I never play it because there is no Multiplayer (I'm a social person) and the lack of a "story line" or "purpose".

I see the developer blogs and a cheer on the developers (last update was GREAT) but I find it disconcerting the number of people working on the game. Nobody is making any Mining or Colonization parts. Nobody is working on alternate Space centers or Kerbal races. Nobody is working on Kerbal A.I. or Cities and such to dress up the game and lay the foundation for a more immersive environment.

I think you found the reason why those things are taking so long. They assigned them to someone with almost no programing experience and then forgot to tell me about it.

Yes development going to slow can hurt the game but so can going to fast. A rushed job will be buggy and unstable.

Edited May 27, 2014 by Nobody

[Northstar1989]

I kind of agree, perhaps not kill it, but definitely won't help the game when it finally releases.

With the implementation of steam sales, I think the amount of players will begin to plateau because anyone who will ever want it has already bought it.

However, the donations people will likely give could possible support the game as it drags along.

It's not as if they couldn't sell expansions to the game if they *really needed to*...

The game is nowhere near market saturation. So many people still haven't even heard of the game. The player-base continues to grow vibrantly, and there are always more kids coming of age into mature gamers capable of playing something like KSP...

The best thing Squad could do would be to increase the number of developers. They have a small team now, and it seems like many of them are working on supplemental things like Google+ spreadsheets (to take a recent example) and running the forums rather than the actual code.

I'm not criticizing- those functions need to be filled. But they could easily double the number of developers working on code by hiring just a few programmers, since the vast majority of the devs currently aren't actually working on coding...

Regards,

Northstar

[allmhuran]

Well, someone who knows better than anyone else here could enlighten us...

Perhaps I can help a little. I've been a professional programmer since 1998. I don't work on games these days, but I did do some game work back in 1999. In any case, the relevant points here aren't specific to game development, but rather just large development projects in general. I've been working on enterprise level projects for most of my career.

So with that said, let me try to say something about this point (and give a bit more detail regarding Vanamonde's earlier comment):

I don't know much about developing a game, but how can people develop "Kerthane" mods, "Dangerous Re-entry", "Planet Packs", and "Sound Packs" just in their spare time so quickly and the developers not be able to develop for sale mod packs as quickly to help propel production of the game.

Large programs are composed of a number of "layers" of code, and the layers are composed of independent "objects" that can communicate with each other. The layers are constructed such that the higher layers depend on the lower layers, but the lower layers don't depend on the higher layers (at least, this is how they should be constructed).

When modders are doing their work, they are working with some of the very high layers. Squad (and Unity) have exposed some of the layers to the community so that mods can be created. Building on top of the higher layers doesn't have any impact on the lower layers, so you don't have to go through and test the whole KSP codebase when making mods. You work against a static and (over the short term) stable layer that has been provided for you.

Squad, in the meantime, works on the whole stack.

Ideally the lower layers don't have to be modified too often, but large design decisions can require that lower layer code gets reworked. When this happens a lot of additional work (largely testing) is done to make sure that you haven't broken any of the higher layers. If you find something has broken, it can often mean a *lot* of work to fix.

In addition to this, Squad's internal plans for future features and updates might include changes anywhere in the stack. An example of this might be switching to 64 bit unity. So they have some idea about what may change in future, and what kind of impact that might have. If some of those changes might be lower level changes, then they have a decision to make: Build the new high level stuff now, then test it and potentially rebuild it again later when you update the lower layers. Or just wait until the lower layers are updated, and then build the higher layers.

The advantage of the former approach is that new features get released quickly. The disadvantage is that it means more work in the long run.

The advantage of the latter approach is that it is an efficient method that results in fewer bugs and less testing. The disadvantage is that the higher level stuff gets updated more slowly.

Since Squad is a relatively small team, the latter approach is a more sensible one to take. You could say "just hire more programmers", but it's not quite that simple. Each time a new programmer is added they start off with no knowledge of what has gone before. They must be brought up to speed, learning the quirks and idiosyncrasies of the existing code base. And since the existing code base is large, this takes a long time. In fact, it's almost impossible to get a new developer on a project that already has a large code base up to speed with everything. The best you can do is hope that the code is modular enough that they only need to know about the areas they are working against directly.

Modders don't have this detailed, multi layer plan. They just have the higher layers that have been exposed for them to use, and build against whatever is available "now". They can't decide whether to hold off and wait for lower level changes to come through, because they're not exposed to those layers in the first place, and don't know in detail what those plans might be. So a modder just goes ahead and builds whatever they think works best using whatever is currently exposed. If things change in the future the mod might break - and indeed, this happens all the time. Many, many KSP mods have "died" due to core game updates. We as a community might not really care - someone else will come along and build a new mod that works with the new version of KSP, and we're happy. But the double, or triple, or quadruple development effort that this process creates isn't something that any respectable development team would want as part of their process, because it ultimately reduces available resources.

Now someone might say "I know mods that have continued to work through several versions of KSP". And that would be correct. Some mods don't need much updating, or only need minor updating, in order to remain compatible. You might think this defeats the argument against approach 1. Not much rework was required after all, so why not just go ahead and add the high level features now?

Ah, but the very reason why not much work was required, the very reason why mods can be built quickly and updated quickly, is that Squad themselves did a very large amount of work to ensure that the high layer stayed consistent through lower layer changes. In other words, Squad has spent a very very large amount of time so that the modders don't have to. Squad's other option would be to say "hey, we made some changes, we were able to make them quickly but pretty much all mods are now broken. Too bad modders, go build your mods again". But if they took this approach the number and quality of mods for KSP would be massively reduced, because people wouldn't bother putting in the effort if their stuff broke all the time.

Edited May 27, 2014 by allmhuran

[darkmuminek]

So much facepalm.

I will never comprehend how people can have so little understanding of how game development on this scale works. It's not even worth trying to refute points.

I just love it when someone criticizes KSP and the fanboy police arrives in seconds with no arguments at all. Frankly speaking KSP is a bad game. Not in terms of idea or gameplay, but because of slow development and horrible communication. Right now the main line of communication between developers and player base are the devnotes which, for me, don't work at all. Most of the time they consist of statements like: "We are doing something right now, we won't tell you what or when it will happen" or "We were at someplace and we're tired but now we are preparing to go someplace else and we will be tired too". The best example is this forum, where mostly the only occasion to get some reaction from the team is "thread closed" when someone dares to post something from the dreaded "what not to suggest" list.

As for the development, we know how it works. We know it takes time and resources. We know and understand it's an early access game but I think SQUAD lost it some time ago. The game's been in development for three or four years and yet most functionalities come from mods. Reentry effects and damage? There's a mod for that. Better atmospheric model? Try FAR. Basic flight and orbital data? Yeah, go and install Kerbal Engineer or MechJeb. Kerbal Alarm Clock, Crew Manifest, KAS, Infernal Robotics, Kethane... these should be stock. Instead we get the, in my opinion, completely unnecessary ARM Pack. It's great we've got asteroids now. Too bad all you can do is dock with them just the same as you can dock with another ship. Too bad you don't focus on what's important. The science system is completely broken. There's no fun in preforming the experiments, there's no sense of achievement. Just spamming the "get science" button. Another example? 64-bit. Unity announced a few months ago that the new version of their engine will have proper 64-bit support for Windows. Did anyone from SQUAD bother to tell us they're looking into it? Don't think so...

This complaint fails to take into consideration that mod-makers are making creating alterations or additions to already working code. And who created, maintains, and updates the code that runs the game? Squad does. Of course it's going to be faster and easier to modify or add to a functioning system than it was to create the functioning system in the first place, so there's realy no basis for comparing the two.

Yes, but I don't understand how one person (or even a group of them), working for free, in their spare time can do a better job than a studio of professional developers. Of course it took a lot of time and work to make the best mods but eventually they got finished, they are compatible with other mods and once again - all of this done by someone who has a family, a job, school and does it for free.

Every time there's talk about development of KSP I compare it to Space Engineers. It's also a complex, sandbox, space game made by a small studio. Yet they manage to release weekly updates. Granted, they have their own engine and the updates are smaller than the KSP ones but still they deliver what the players want, they are in constant contact with their fans, they listen to them and give them complete information about the status of the game and it's planned features. Why can't SQUAD do the same?

As for the OP's question. Yes, I think it can kill a game. To be honest I don't believe we'll ever see a finished KSP. I give it two, maybe three years after which it'll get abandoned. This saddens me because it's a great game with great potential but the developer forgot what's it all about.

[allmhuran]

Most of the time they consist of statements like: "We are doing something right now, we won't tell you what or when it will happen"

<...>

and give them complete information about the status of the game and it's planned features. Why can't SQUAD do the same?

While I just made a large and detailed post "defending" Squad's position on development, I must agree with these specific sentiments about communication.

Squad has taken the position that "letting people know about things that might happen, or current under-construction ideas" is a bad move, because it "sets people up for disappointment", or something along these lines. Squad points to the "resources" de-listing as evidence of this, and seems to assume that the matter is therefore proven.

But that's not the right approach to take.

What Squad has historically done is announce features close to release, but providing very few details about how things might work, and then provided a second update at time of release to say "it's done". But this denies the community the opportunity to provide useful feedback. The community was not upset by the postponement/cancellation of resources merely because they'd heard it was coming. It was upset because no other ideas for "things we might be able to do beyond going places and landing there" were raised, and Squad explicitly said they weren't going to reveal any draft plans for what that might be.

Transparency doesn't mean letting people know something has been released. It means letting people know what your ideas are while they're still being worked on. It doesn't matter if those ideas are changing. Recently we've even seen people in this forum write things like "there's no point talking about contracts until contracts are released". But what if the community can come up with good contract related ideas? Right now people are left to make suggestions in a vacuum - we don't know if the suggestion is even remotely compatible with what Squad might end up doing. There's no starting point that the community can work from when building their suggestions.

This isn't to say that Squad should do whatever the community wants and ignore their own plans. It's simply the acknowledgement that when you have thousands of dedicated players, some of them are going to come up with ideas and post them, it might turn out that they can come up with better ideas than half a dozen people. And if those half dozen people like the ideas, they can incorporate them, or incorporate the aspects that they like, or maybe they don't want to use the idea as stated but it gives them another idea of their own. Squad still gets to put into the game exactly what Squad wants to, their pool of available ideas just becomes larger. But without any information from Squad on, even vaguely, what kind of things they're thinking about, the pool is full of random flotsam and jetsam. And if there's no information provided until something is released, it's too late for the community to bother providing feedback.

Edited May 27, 2014 by allmhuran

[darkmuminek]

I agree with you. Listening to suggestions and sharing your thoughts is very important in not only game development but nearly all aspects of life. It doesn't mean that everyone should start posting their ideas in the suggestions forum nor does it mean Squad should implement everything people write there but having a look from time to time and taking some things into consideration is generally a good idea. Instead the only replies people get there are:

1. It's on the "what not to suggest" list, go to hell.
   
2. There's a mod for that.
   
3. Some general good/bad words from other users that give nothing of value.
   

A normal, or better said - standard, game development process has two parties: the developer who does all the work and the publisher who gives the money and tells the developer whad he'd like in the game. In case of indie games there's no publisher and that's why they're called "indie". However KSP is also an Early Access game meaning that people (us) gave their money to the developer trusting they can deliver a complete product in the near future. This makes us a form of publishers meaning we have the right to:

• 
  
• complain
  
• have expectations
  
• have suggestions
  
• ask questions
  
• get the answers to these questions
  
• get reliable information about the status of the game
  

Once again, Squad. This is your game. You decide how it looks like or what features will it have. However it is because of us, our trust and money that you can make it happen. A thing you (and many other indie devs) seem to forget too often.

[allmhuran]

I think that's going a bit far, but I don't want to open up the "obligations/entitlements" discussion.