@KGB29, good points there, also competition for engines drives the price down which is better for the budgets on games too.

Each studio prefers to work with their own creation their own work.

One engine wont work for GT GOW Uncharted, each game has its own needs.

It would be flat out retarded to use only one engine, if that one fails to live up to expectations all of Sonys studios would suffer due to it. All studios dev and upgrade their engine through out each gen. It works best for them.

 but these game studios are using these game engines for their games only. i wander if any of them is used by third party game developers. so mostly these game engines are not going to be sold. so competition  to drive the price down is not a very valid point.

I prefer that, actually.  I know its expensive, but as an end consumer I'm of the opinion, based on what I've observed, that games running on an engine completely tuned to the concept tend to be superior to more generic games based on middleware engine.

It's an old argument though, and goes right back to when the first 'engine' sourcing started to take place.  I think there's an old but interesting interview with Warren Spector where he reviews this regarding putting Deus Ex on Unreal engine.

There are so many average FPS on U3 engine, for example, or games that built on the engine but end up buggy as a result - probably because at the end of the day the developer using a sourced engine is unlikely to really understand its guts as well as the engine's developers.

I know Ted Price from Insomniac has quoted on this a few times too (he's in the build your own engine camp).  Epic are in the source it camp, but given they make a lot of money from sourcing the engine they develop themselves I'm not sure they can be seen as totally unbiased.

However, I would hope that the core engines you list will see more use/expansion which I'm sure they will - either via further entries in the series or sharing of assets.  Insomniac for example have released 3 full games and an expansion using the same, evolving in-house engine.

well all of sonys studios share technology so while their engines are different they use a lot of the same technology
this allows for a lot of epic and epic looking games that also have a visual diversity you could never achieve with something like the unreal engine 3.

I agree, but even in near future this can't happen anymore really. Unless you have something ready already and you have platform that you have used before. Every gen tends to make things differently and in every gen price of engine goes up. Eventually there have to be similar platforms(backwards) or nobody would make engines for those.

Sounds like really bad documentation problem... or some people just can't do it. Do you mean buggy as bugs or something like bad performance? Bugs happen all the time in the software and you have to have good metrics system to get rid of them in larger projects. So in other words that would be developers failing. And performance is always crappier on engine that is not in-house engine. I am quite sure that it isn't because developers don't understand how to use it. Its more like developers hands are tied to use it only in one way and that way might give bad performance in some cases.

(Btw, I am in that camp too. ^^)

@OT

I think Sony is betting way too much on PS3 and on its software. Even if PS3 would be 10 years on the market, it won't beat next gen consoles that go for graphics & stuff that has been the main selling points of PS3. If MS brings such new console in next few years, it would be like mortal blow for PS3. I am quite sure that it could handle 1080p with much better graphics. Very few *'hc player' would be able to resist it.

(*IMO console games are way more casual than real hard core games.)

These internal Sony devs are very well known for sharing tech and information, I am sure the costs for those engines were minimal thanks to sharing the minds. Also most of those devs you mentioned are due for another game, rest assured most of those devs will not end this gen with just one game.

You will probably find although these compnanies (Sucker Punch, Insomniac, Naughty Dog, Media Molecule, Guerilla and Polyphony) transfer information and techniques between each other to help other companies get the best out of their engines.

On top of this, all of these games are technically superior than what could be done on a generic engine.

This is Sucker Punch's first PS3 game and of course have spent the time on a new engine, but I reckon they should be able to get another 3-4 games out of this engine easily with a little tweak here and there.

Insomniac have actually made 4 games so far and are working on their 5/6 at the moment, all of which have used variations of their 3d engine.

Naughty Dog took some time to get their engine from scratch, but have now tweaked and enhanced for Uncharted 2 and have one of the best in the industry and by looks alone, should be able to stand up in years to come.

Media Molecule's is a first for them, but no doubt will be used again and again.

Guerilla spent years on everything in their engine, but WILL be used for new Killzone games and other genres.

Polyphony need to develop specific engines for themselves, to make sure they get the best out of the system for a racer at high resolutions, no doubt we will see GT5 and possibly GT6 on this engine.

I get the feeling that with all the money Sony sank into the PS3 and development of the Cell i think the PS4 (which we all know there will be one) will either expand the properties of the Cell to make it more powerful or we will see a quad-Cell type processor!

If this is the case, then the engines that Sony and the Dev teams are making at the moment would work on this infrastructure and therefore the time and money spent on them would be worthwhile!

Just my opinion of course

What I meant by buggy is that often, when developers take something like U3, and want to add code for extended features specific to their game (say I want it to underpin an RPG, for example) the end result always seems to be a game that shows the following traits vs pure FPS titles developed on U3, or titles developed by Epic

a) general performance of the engine is not on par with what it should be, due to the changes not sitting well with the underlying code of the engine

b) a more unstable product - i.e. a somewhat higher percentage of quirks and sometimes outright bugs

If you think about it Epic know every line of code in U3 in detail, but a developer who uses it tends to either bolt on code to the core or use the middleware tools without really knowing the underlying code inside out.  Mostly this is okay but it is an environment much more likely to produce issues than someone developing on an engine they know inside out.

At the end of the day I think middleware is here to stay, and it can produce great results also - after all Half Life was built on ID tech - but I think it does work better where the developer is producing something 'in genre' rather than extending the base too much.

If you really want to push out something new I think you have to built an engine unless there is a really good fit.  For example look at Mirrors Edge, with its level orientated structure dictated by the middleware engine vs the realy open air, run anywhere feel of the engine under Assassin's Creed.

I do like U3, and I am excited to see what ID tech 5 looks like, etc. but I think there is always a balance to be struck between knowing when to built from scratch to get the best results and when using middleware will create good enough results and cut costs without harming the game.