MW1 on the 360 had technically better quality and better framerate than the PS3 version. MW2 closed the gap some but also had a better framerate on the 360. "nearly identical" is extremely nebulous and you are using it as a launching point for a line of inquiry that doesn't really establish the assumed proposition, namely that Bo suffers moreso than previous versions. The PS3 version of BO have a bigger quality and framerate gap than MW2 on both consoles (not sure about MW1 as the PS3 version had some big performance spikes and IQ issues) but to this I think the answer was already given. Infact the reason BO's performance resembles closer to the MW1 issues versus the closer parity in MW2 seems quite simple, in which case your "but for PS3-only owners, it's hard to just keep your mouth shut" has little to do with the TECHNICAL development issues and more to do with voicing off about the state of software quality. And the answer to the question is in this thread. The question is, technically, was it a good development approach (given the code base, dev cycle, etc, who is to disagree?) and if the PS3 version could be better, in what technical ways, given those constraints, and finally why did the PS3 hardware pose troubles to begin with (again, already covered).
The Game Technology discussion thread *Read first post before posting*
- Thread starter Shifty Geezer
- Start date
MW1 on the 360 had technically better quality and better framerate than the PS3 version. MW2 closed the gap some but also had a better framerate on the 360. "nearly identical" is extremely nebulous and you are using it as a launching point for a line of inquiry that doesn't really establish the assumed proposition, namely that Bo suffers moreso than previous versions. The PS3 version of BO have a bigger quality and framerate gap than MW2 on both consoles (not sure about MW1 as the PS3 version had some big performance spikes and IQ issues) but to this I think the answer was already given. Infact the reason BO's performance resembles closer to the MW1 issues versus the closer parity in MW2 seems quite simple, in which case your "but for PS3-only owners, it's hard to just keep your mouth shut" has little to do with the TECHNICAL development issues and more to do with voicing off about the state of software quality. And the answer to the question is in this thread. The question is, technically, was it a good development approach (given the code base, dev cycle, etc, who is to disagree?) and if the PS3 version could be better, in what technical ways, given those constraints, and finally why did the PS3 hardware pose troubles to begin with (again, already covered).
DonaldDuck
Newcomer
The question is that legacy code is by no means a good approach in order to maximize performance, overall performance, because the frame is the result of the whole amount of computations that are implied in the engine, and if legacy code, by whatever reason, doesn´t fit with hardware in one platform, that base deficit in performance will arise each time you produce a game.
CoD is exactly the type of game that, because of several reasons (tight scheduling, long history, correct performance in the platform with higher sales), has legacy code probably in spades and I suspect that the real reason of the problems with PS3 version is there, not precisely in new and subtle care added in the last instances of the game.
That is, almost every periodic franchise (or not periodic, but "old" in the context of the generation) has some parts of the code that are written with XBOX in mind and, of course, you rewrite, but more obvious is that you don´t re-think the big picture, just because isn´t economically feasible, smart or appropiate. People still buy PS3 Black Ops and in millions, so from a company perspective the 2-3 months of forum crossfire related with the poor (relative to XBOX) performance doesn´t justify the risky decission of making PS3 lead platform...
... and rewrite all the systems almost from scratch.
The reason because the rewriting is so necessary is other question, and it´s pretty clear that the "different" (some will say unusual, others unnaceptable, archaic, dated, strange, flexible... opinions) way in which PS3 unleashes its performance makes big, sometimes definitive problems to programmers, more pronounced if you have legacy code.
We should assume that "old" (single threaded, pc-friendly) code doesn´t love PS3, their SPU´s, assymetric CPU architecture, and I´m almost sure that producing a "top" game in PS3 needs not only algorithm tricks, but changes and conceptions at the architectural level. The big, terrible question that arises when you start from nothing (ok, impressive knowledge of the hardware, but no one line written): how I construct the whole engine? Big design decissions that, if PS3 is not the platform used, drives you to the land of trouble. Performance trouble, not financial, of course. You still have Microsoft.
And this idea has some logic into it. Assuming PS3 is theoretically more powerful (people from MS have said that in this thread) but extremely more complex to program, with more constraints, and requires more deep and tiny adjustments in code, if on the other hand XBOX is smooth, has better tools, and is more open-minded, then you have what we have today: any kind of love from XBOX to PS3 is condemned to relative disaster, but forgetting that something called XBOX 360 exists produce really phenomenal games, and even and minimal, but measurable, difference in performance in favour of the expensive (at different levels) Sony´s hardware when you port your code.
CoD is exactly the type of game that, because of several reasons (tight scheduling, long history, correct performance in the platform with higher sales), has legacy code probably in spades and I suspect that the real reason of the problems with PS3 version is there, not precisely in new and subtle care added in the last instances of the game.
That is, almost every periodic franchise (or not periodic, but "old" in the context of the generation) has some parts of the code that are written with XBOX in mind and, of course, you rewrite, but more obvious is that you don´t re-think the big picture, just because isn´t economically feasible, smart or appropiate. People still buy PS3 Black Ops and in millions, so from a company perspective the 2-3 months of forum crossfire related with the poor (relative to XBOX) performance doesn´t justify the risky decission of making PS3 lead platform...
... and rewrite all the systems almost from scratch.
The reason because the rewriting is so necessary is other question, and it´s pretty clear that the "different" (some will say unusual, others unnaceptable, archaic, dated, strange, flexible... opinions) way in which PS3 unleashes its performance makes big, sometimes definitive problems to programmers, more pronounced if you have legacy code.
We should assume that "old" (single threaded, pc-friendly) code doesn´t love PS3, their SPU´s, assymetric CPU architecture, and I´m almost sure that producing a "top" game in PS3 needs not only algorithm tricks, but changes and conceptions at the architectural level. The big, terrible question that arises when you start from nothing (ok, impressive knowledge of the hardware, but no one line written): how I construct the whole engine? Big design decissions that, if PS3 is not the platform used, drives you to the land of trouble. Performance trouble, not financial, of course. You still have Microsoft.
And this idea has some logic into it. Assuming PS3 is theoretically more powerful (people from MS have said that in this thread) but extremely more complex to program, with more constraints, and requires more deep and tiny adjustments in code, if on the other hand XBOX is smooth, has better tools, and is more open-minded, then you have what we have today: any kind of love from XBOX to PS3 is condemned to relative disaster, but forgetting that something called XBOX 360 exists produce really phenomenal games, and even and minimal, but measurable, difference in performance in favour of the expensive (at different levels) Sony´s hardware when you port your code.
(((interference)))
Veteran
Well if they're paying PS3 coders to get 360 code working on PS3 how much harder can it be to instead only port features that are easy to get up and running on PS3 and for anything more exotic write PS3 specific implementation of the feature or even substitute it for something entirely different (like MLAA instead of MSAA).
It might not be possible for COD which is on a strict 2 year development schedule but for other multiplat titles with more flexible schedules it should be workable.
The problem with that is development time and cost. The reason they attempt parity on both systems is because the engine is the smallest part of the game. Most of the game is level design, art assets, AI behaviours. You want to be able to use these same assets across your platforms, or it'll cost you a lot more in both development and testing time. If you change the engine behaviour for each platform, you can't guarantee that the assets/AI etc will look or perform the same.
djskribbles
Legend
You're right... I forgot that MW1 had a bigger gap between the PS3 and 360 version (compared to MW2), but it wasn't nearly as much as with BO.
Look, all I was saying was that I think constructive criticism is okay, and I was simply trying to explain why PS3 owners were voicing their opinions and asking the questions they were. People seemed surprised that PS3 owners were asking questions of how they could have improved the PS3 version of BO when really, it shouldn't be surprising considering the circumstances. For myself, I haven't said a single word about the matter because I know it's not my place.
Billy Idol
Legend
thanks joker and bkilian for the insight - this explains a lot!
I always thought that it is quite obvious that lead platform gets best results - no matter what platform it is...and that you get parity if you respect a common base tec...this seems to be true in both directions!?
Jokers post explains why typically X leads, and typically gets better version - and I think that this makes a lot of sense.
But it seems to me that some of you guys forgot that there were occasions were PS3 was indeed the lead, resulting in slight better versions (e.g. FF13, Castlevania. and it seems to me that in case of FF13, the EDRAM restriction of the X, was reason for lower resolution and that the devs were not willing to change code parts to support tiling due to time and money and existing code base).
I fully agree with (((interference))) that I would love to see games specialized to each platform using each strenght in multiplatform games. But it seems to me quite illusional: if it is too expensive to even lead on the PS3...it seems to be way too expensive to develop an extra version for the ol Station.
It would really interest me, if you gaming dev guys already know the parts of the code which don't translate so well to PS3 in BO engine...and more interestingly: do you have found a solution/ideas to those bottlenecks in mean time (just theoretical: given enough time and money to change those code parts)?
I always thought that it is quite obvious that lead platform gets best results - no matter what platform it is...and that you get parity if you respect a common base tec...this seems to be true in both directions!?
Jokers post explains why typically X leads, and typically gets better version - and I think that this makes a lot of sense.
But it seems to me that some of you guys forgot that there were occasions were PS3 was indeed the lead, resulting in slight better versions (e.g. FF13, Castlevania. and it seems to me that in case of FF13, the EDRAM restriction of the X, was reason for lower resolution and that the devs were not willing to change code parts to support tiling due to time and money and existing code base).
I fully agree with (((interference))) that I would love to see games specialized to each platform using each strenght in multiplatform games. But it seems to me quite illusional: if it is too expensive to even lead on the PS3...it seems to be way too expensive to develop an extra version for the ol Station.
It would really interest me, if you gaming dev guys already know the parts of the code which don't translate so well to PS3 in BO engine...and more interestingly: do you have found a solution/ideas to those bottlenecks in mean time (just theoretical: given enough time and money to change those code parts)?
(((interference)))
Veteran
Yes but I didn't mean such drastic changes so as to necessitate separate content being created for each system. Stuff like MLAA vs MSAA, more light sources, higher fidelity water simulation, physics enabled particles etc. shouldn't affect things like level design, art or AI to any substantial degree
.
Despite not having a dedicated install, I think the PS3 version will automatically create a cache on the HDD while running just like on a 360 with a HDD.
What I always want to know is how multiplat titles (for example Red Dead Redemption) ran on 360s without HDDs and how that compared to the PS3 version in terms of load times, texture LODing and framerate. As it would be something like magic to get some of these games running of a DVD with less than 512 MB of RAM and still keep up (or perhaps even better) with the PS3 which has Bluray and a HDD.
I have yet to see any kind of analysis done with HDDless 360s.
Actually, I think some multiplat titles already have some graphical features specific to each console - though for the life of me I can't think of one other than Saboteur and Tekken 6 but they were really both better on PS3
Last edited by a moderator:
Break it down even smaller. Which of these would give the best results:
1) A single programmer is given a task, he must make it work on one console.
2) A single programmer is given a task, he must make it work on two consoles.
3) Four programmers are given a task, they must make it work on two consoles.
Assuming programmers of equal talent, and assuming the task in every case is identical, which of the above is more likely to give the best result in the shortest time frame? #1 may equate to ND, #2 may equate to a small multi platform studio, and #3 may equate to a large multi platform studio like Ubisoft. #1 case is looking at sales in the 1 to 5 million range so they are happy to dedicate a coder to the problem since it's just one console anyways. #2 would love to dedicate more coders to the problem, but since they are looking at 500k to 800k in sales, they will just stick with one guy to figure it out on both machines. #3 is looking at solving the problem for three different multi million copy selling games, so they throw as many people as needed at the problem, and hire more if they need to.
That's where I say it's unfair to immediately assume 'multi plat' means they cut back, it depends on many things. The sales projections for Cod BO must have been huge, so I find it hard to believe that they skimped on coders and instead hired someone from the Home Depot parking lot to get the graphics working on the ps3 version.
Yup that's totally true. True story...a few years ago when the nonsense of "point to game X and be pissed that game Y doesn't implement tech Z" started, I joked internally where we were working that we should make a mini game called Cube Master. It would do nothing by draw a cube on screen, but it would be the greatest cube you ever saw! Hundreds of thousands of lights, megabytes of textures, 100+ layers of fur shading, just go crazy and make it the most epic looking cube ever. Then anytime another game came out and bragged about a graphics feature, we would mock it. 100,000 lights you say? Cube Master has 500,000! The point being to emphasize how silly the whole situation became.
I don't think Barbarian can ever get into any detail that would satisfy anyone really, it's just too risky. I'm semi retired so heck, I don't care about posting
Although even then I have to make sure to not posting about something I was told in confidence naturally.
Last edited by a moderator:
DonaldDuck
Newcomer
The technical question I still have, a very illustrative one I think, is: taking into consideration that PS3 and XBOX have a different hardware architecture, how different must be the software code in order to max out the power of each console?
Because the answer to that question could clarify the differences relating multiplatform games. You know, at the architectural level, is it possible that the "same" structure for the engine could unleash optimum performance in both platforms?
And what I think is that the differences between hardware for each console are big enough to force you, as a project manager, to start the development of an engine suited for each HW (specially with PS3, because it´s far more intrincated and complex to program) if you want to success in that aspect.
Some people insist (joker, with its huge experience, is among them) that the biggest studios and the biggest franchises have nailed the specs and adjusted the code to a point of "maximum" excellence, that´s it, a level of quality that is over the rest of exclusive studios. But my question is still there: How different is the architecture of PS3 and XBOX code for RDR? The "core" code of both engines, isn´t the one for XBOX programmed at the beginning of the lifecycle for the generation?
It´s absolutely clear that that engine has been fine-tunned, obssesively is you want, but the big decissions have been made years before, and now only the algorithms and the medium to low level structure of the modules of those engines are changed with each iteration of the engine (games).
That´s my belief and the main reason for me to support that opinion is budgeting. And money is sacred. It´s the key and the bar for everything related to projects. A project manager will not authorise a deep change at the technical level that compromises the time frame of the project or the cost implied. If you have an engine good enough you go with it. AC2 (there´s an interview in DF, by the way, really interesting) had a draconian timeframe. You (me, just in case) have to decide if the proposal of the main software engineer about changing the scope of the engine to obtain better performance with PS3 is meaningful. And you´ll say "NO" and will answer: "Do the best with micromanagement". And go.
Obviously, the muscle of the project, the software coders and engineers, can do miracles in its realms of modules. They will respect the place and interfaces the big engine has imposed and inside the frontiers the code each time will be better. Time will be reduced, lines will be shortened and optimised... And each iteration of a game will be better coded, for sure. But the question is still there:
Do you believe that the same engine approach (taken years ago, at the beggining of the generation) is appropiate to maximize performance with both platforms, being them so different in hardware architecture? I don´t believe so, and my logic tell me that the more intrincated, more complex, archaic and with poorer development tools hardware will have the worst part.
In this precise case, that platform is PS3, and it shows.
Obviously, if the premise is wrong (that´s it, "PS3 is different ENOUGH in hardware architecture to need an specific approach, at the big level") then my opinion will be invalid. But I don´t think it´s the case. Several developers (the interviews in DF are marvelous) stated that the way PS3 need in order to perform is really very specific, needing not only fine tunning, but a holistic approach and, even more relevant to the question, different and more rigid that the one XBOX needs.
Again: is the premise valid, or no?
I have the impression that the engine behind RDR in PS3 and XBOX is very different, extremely customized, but at the low-medium level, maybe not enough to max out the performance in the hardware more "abrupt". Just because the engine for XBOX arrived first, hasn´t the intrincate type of adjustments the engines for PS3 seems to have,and the most important reason, isn´t necessary in order to sell. Money, again is the key. And being honest RDR in PS3 is brilliant... just not "perfect" related with XBOX standards. But a project manager sees how the game sells zillions, and the engineering cost is reduced... --> Epic win.
Because the answer to that question could clarify the differences relating multiplatform games. You know, at the architectural level, is it possible that the "same" structure for the engine could unleash optimum performance in both platforms?
And what I think is that the differences between hardware for each console are big enough to force you, as a project manager, to start the development of an engine suited for each HW (specially with PS3, because it´s far more intrincated and complex to program) if you want to success in that aspect.
Some people insist (joker, with its huge experience, is among them) that the biggest studios and the biggest franchises have nailed the specs and adjusted the code to a point of "maximum" excellence, that´s it, a level of quality that is over the rest of exclusive studios. But my question is still there: How different is the architecture of PS3 and XBOX code for RDR? The "core" code of both engines, isn´t the one for XBOX programmed at the beginning of the lifecycle for the generation?
It´s absolutely clear that that engine has been fine-tunned, obssesively is you want, but the big decissions have been made years before, and now only the algorithms and the medium to low level structure of the modules of those engines are changed with each iteration of the engine (games).
That´s my belief and the main reason for me to support that opinion is budgeting. And money is sacred. It´s the key and the bar for everything related to projects. A project manager will not authorise a deep change at the technical level that compromises the time frame of the project or the cost implied. If you have an engine good enough you go with it. AC2 (there´s an interview in DF, by the way, really interesting) had a draconian timeframe. You (me, just in case) have to decide if the proposal of the main software engineer about changing the scope of the engine to obtain better performance with PS3 is meaningful. And you´ll say "NO" and will answer: "Do the best with micromanagement". And go.
Obviously, the muscle of the project, the software coders and engineers, can do miracles in its realms of modules. They will respect the place and interfaces the big engine has imposed and inside the frontiers the code each time will be better. Time will be reduced, lines will be shortened and optimised... And each iteration of a game will be better coded, for sure. But the question is still there:
Do you believe that the same engine approach (taken years ago, at the beggining of the generation) is appropiate to maximize performance with both platforms, being them so different in hardware architecture? I don´t believe so, and my logic tell me that the more intrincated, more complex, archaic and with poorer development tools hardware will have the worst part.
In this precise case, that platform is PS3, and it shows.
Obviously, if the premise is wrong (that´s it, "PS3 is different ENOUGH in hardware architecture to need an specific approach, at the big level") then my opinion will be invalid. But I don´t think it´s the case. Several developers (the interviews in DF are marvelous) stated that the way PS3 need in order to perform is really very specific, needing not only fine tunning, but a holistic approach and, even more relevant to the question, different and more rigid that the one XBOX needs.
Again: is the premise valid, or no?
I have the impression that the engine behind RDR in PS3 and XBOX is very different, extremely customized, but at the low-medium level, maybe not enough to max out the performance in the hardware more "abrupt". Just because the engine for XBOX arrived first, hasn´t the intrincate type of adjustments the engines for PS3 seems to have,and the most important reason, isn´t necessary in order to sell. Money, again is the key. And being honest RDR in PS3 is brilliant... just not "perfect" related with XBOX standards. But a project manager sees how the game sells zillions, and the engineering cost is reduced... --> Epic win.
DonaldDuck
Newcomer
Apart from this, other take is inherited code, something that in CoD precisely seems very relevant.
Old code is cheap (in fact, has no cost, you have it). It´s probably old (the first CoD was a PC installment some 7-8 years ago, one core, etc). Using it is a temptation. You build a multiplatform engine using the better tools availables for the most part (XBOX in this case), recycle as much code as you can (PC CPU routines, from my point of view, are more easily adapted to the parallel, symetric CPU architecture of XBOX than to the assymetric of Cell), etc.
At the end you have an engine build around limitations of old inherited code, using the more comfortable development tools, and tested with the more flexible hardware, for the most part; even more, that´s was undoubtly the case at the beginning of the gen, when Sony´s tools were a nightmare.
Then things are better, but, again, you inherit code from the previous installments of your engine, and that engine was really in love with the more straight architecture. You, as a PS3 programmer, are at odds, because now there´s a new timeframe (that thinks that "a big enough engine for PS3 exists") and your duty is corrupted by limits imposed to you from above. The wheel is tourning, you have to proceed, and improve of course... as much as you sensibly can.
And that´s the problem: how far you can go.
I´m sure that the memory budget is a terrible factor against PS3 development. Terrible, because it´s almost irresoluble, because the memory is occupied ¡by other programmers that are thinking exactly the same as you! The big brain that fuck you from above will not take the risk of rebuilding the block assignments ir order to mantain programmable code really as tiny as possible, among other considerations. If in God of War III the executable code occupied only 6MB may be the case that all the engenieering process was thought around the idea of preserving memory. But if you use old code, if you come from an XBOX way of live with obviously more memory available, then you´re screwed. No more no less. And you have to eat all the trash talk about lazyness with PS3 programmers.
Because I don´t know if PS3 is more or less powerful than XBOX (that´s really a complex question, if it´s ever a question at all). But what I suspect is that PS3 is by no means benefited from some circumstances (year later, more complex, poor tools, etc) that make very probably that the code from publishers (even the most respected) will not be really indulgent with it.
That´s it, the problem is not programmers. The problem is reality and bussiness. As always.
Old code is cheap (in fact, has no cost, you have it). It´s probably old (the first CoD was a PC installment some 7-8 years ago, one core, etc). Using it is a temptation. You build a multiplatform engine using the better tools availables for the most part (XBOX in this case), recycle as much code as you can (PC CPU routines, from my point of view, are more easily adapted to the parallel, symetric CPU architecture of XBOX than to the assymetric of Cell), etc.
At the end you have an engine build around limitations of old inherited code, using the more comfortable development tools, and tested with the more flexible hardware, for the most part; even more, that´s was undoubtly the case at the beginning of the gen, when Sony´s tools were a nightmare.
Then things are better, but, again, you inherit code from the previous installments of your engine, and that engine was really in love with the more straight architecture. You, as a PS3 programmer, are at odds, because now there´s a new timeframe (that thinks that "a big enough engine for PS3 exists") and your duty is corrupted by limits imposed to you from above. The wheel is tourning, you have to proceed, and improve of course... as much as you sensibly can.
And that´s the problem: how far you can go.
I´m sure that the memory budget is a terrible factor against PS3 development. Terrible, because it´s almost irresoluble, because the memory is occupied ¡by other programmers that are thinking exactly the same as you! The big brain that fuck you from above will not take the risk of rebuilding the block assignments ir order to mantain programmable code really as tiny as possible, among other considerations. If in God of War III the executable code occupied only 6MB may be the case that all the engenieering process was thought around the idea of preserving memory. But if you use old code, if you come from an XBOX way of live with obviously more memory available, then you´re screwed. No more no less. And you have to eat all the trash talk about lazyness with PS3 programmers.
Because I don´t know if PS3 is more or less powerful than XBOX (that´s really a complex question, if it´s ever a question at all). But what I suspect is that PS3 is by no means benefited from some circumstances (year later, more complex, poor tools, etc) that make very probably that the code from publishers (even the most respected) will not be really indulgent with it.
That´s it, the problem is not programmers. The problem is reality and bussiness. As always.
There is absolutely no way/point to compare different games from both systems,be it multiplatform or exclusives not only because they are probably different genres,and budgets/time ratios are different but because different engines handle things quite differently.
For example,while i was looking at UDK build notes couple days ago i saw that in UE3,SSAO takes ~2.5 ms on 360.Than I looked at latest Crytek tech papers and they posted their optimized SSAO which is better in performance and quality by quite margin in comparison with old one.It takes mere 1ms on 360.Thats just one little thing where the difference is quite considerable and there is hundreds of things like that that engines have to handle.They all have differently written code and techniques so there is absolutely no point in comparing them.
For example,while i was looking at UDK build notes couple days ago i saw that in UE3,SSAO takes ~2.5 ms on 360.Than I looked at latest Crytek tech papers and they posted their optimized SSAO which is better in performance and quality by quite margin in comparison with old one.It takes mere 1ms on 360.Thats just one little thing where the difference is quite considerable and there is hundreds of things like that that engines have to handle.They all have differently written code and techniques so there is absolutely no point in comparing them.
Last edited by a moderator:
The upcoming port of Mass Effect 2 for the PS3 is going to be an interesting case study. It is built on UE3, so a multiplatform engine; and we've already seen that the X360 and the PC versions are basically the same in every way. It's also certain that the Xbox version has been heavily optimized, both in terms of the engine (running much better compared to the first game, with less glitches) and in terms of the artwork (characters look more detailed despite using the same texture budgets, levels are nicer too).
On the other hand the PS3 version might be able to use an updated build of UE3, and/or any further optimizations that Bioware may have implemented for ME3. We've also seen close parity in other UE3 based titles, although none of them were really pushing the 360 as well as ME2 seemed to.
So, I'm interested to see the results...
On the other hand the PS3 version might be able to use an updated build of UE3, and/or any further optimizations that Bioware may have implemented for ME3. We've also seen close parity in other UE3 based titles, although none of them were really pushing the 360 as well as ME2 seemed to.
So, I'm interested to see the results...
Prophecy2k
Veteran
I was under the impression that the engine used in ME games was based on UE2.5 and has since diverged considerably from the most current UE codebase (i hear that they re-wrote the renderer almost entirely, amongst other things). I'm not sure about how much ME2 on PS3 will tell us about UE3 performance on PS3 or vice-versa, as i don't think the engine used for ME2 on Xbox360, PC or PS3 strongly resembles the current UE3 codebase in any meaningful way.
Now i may be wrong, but i've heard quite a bit to support this view...
360 UE3 always seems to benefit quite a lot from Gears or War development. I haven't seen much improve recently on the PS3 front. But I guess we'll find out soon. I wouldn't be surprised if it has a mandatory intall that copies the content equivalent to the second disc on the 360 to the PS3's HDD.
I don't think this only has to do with GeoW, Sweeney stated that for UE3 they decided not move finer grained task management. So they uses coarse grained task management which is not optimal on the 360 but even less on the PS3.360 UE3 always seems to benefit quite a lot from Gears or War development. I haven't seen much improve recently on the PS3 front. But I guess we'll find out soon. I wouldn't be surprised if it has a mandatory intall that copies the content equivalent to the second disc on the 360 to the PS3's HDD.
EDIT
Read the post more as a question than an affirmation.
Last edited by a moderator:
No.
An engine is more than a renderer. Mass Effect is likewise more than Gears of War.
I think you confuse the ME and Bioshock games - Irrational talked a lot about how they rewrote significant parts in both UE2 and now for Bioshock Infinite in UE3, whereas Bioware didn't really make that many modifications, or at least I haven't head or read about it anywhere. The first Mass Effect game is definitely UE3 based though.
What made the games look more unique was that they used different settings (for example a strong film noise filter to deviate from the generic UE3 look). ME1 even had the same delayed texture load issue that we've seen in Gears1, only with much more frequent appearance due to the more non-linear levels. And the other, more important factor is the art itself, which is what really made the game stand out.
Similar threads
