Slashdot Mirror
Crossplatform Titles Shortchanging PlayStation 2's Performance?
Thanks to GamerFeed for their new story noting that Sony Europe's research, using their 'performance analyzer', on the latest PS2 games. According to the piece: "The secret (or not so secret) way to unleash the PS2's power is to use its vector units (VUs)... of course, the games that used the VUs fared much better, and the game that scored highest did indeed use the VUs the most." An previous AnandTech analysis of the PS2's hardware explains a little further: "The power of the two VUs exists in the proper use of them as serial counterparts in handling the T&L calculations necessary during 3D rendering, but with the PS2 being... dramatically different from what most developers had seen in the past, getting the most out of the host CPU was quite difficult." The original article, in UK magazine PSM, concludes by pointing out: "A lot of the games that don't really use vector units are ports from other systems."
While there don't seem to be any studies to back this up for the other consoles, I doubt that many cross-platform games found on the X-Box or GameCube have any optimization for those particular platforms.
Really, I doubt the PS2 is the only console who's power isn't being fully used.
Dark Nexus
"Sanity is calming, but madness is more interesting."
Sony is heavily promting development of software that uses the VUs through their Linux development community. I bought the kit a while back, but I don't really have the right skill set to play with them. I'd love to see if, say, a Distributed.Net client could be optimised for them.
What would be perfect is a compiler that could take ordinary C code and turn it into MIPS+VU0+VU1 code automatically, taking the best advantage of all the machine's resources. Unfortunately this pretty much exists in the realm of science fiction.
How we know is more important than what we know.
Forget vector units... I'd like to see multiplatform FPS's make proper use of MICE on PC ports. "Enter the Matrix", "Deus Ex 2" - I'm talking to you.
For great justice.
Funny, most muti-platform games are designed on the PS2 FIRST, not last. There's two I can think of, though: Splinter Cell (which really did look good on all three consoles), and Wreckless: the Yakuza Missions (which is just a shitty game, period).
There're very few that go from GC to PS2 or Xbox to PS2. Normally they're either PS2 to Xbox or GC, or they are developed at the same time for all the consoles that the game will launch on.
So, I wonder how many games equates to 'a lot?'
Perhaps this is a sign to Sony that the next console should be less complicated with its graphics. The Xbox is probably easy enough consider its DirectX basis and the Gamecube architecture uses one graphics chip rather than a convoluted parallized two chip system.
:/ )
While the PS 2 may have a powerful, robust graphics core because of this design, as has been pointed out: Does every company have the time to make use of the specifics, or are they going to do the least amount of work possible when porting to make a game playable and not horrendous looking?
If cross-platform gaming continues to become a huge trend, it would be in Sony's best interest to try and simplify some parts of its graphics to make the best parts of it more accessible to developers that perhaps don't have all the time they need to port the game as well as they should.
In the meantime, though, Sony has a large amount of exclusive content...so they shouldn't be too worried about whether or not they are getting optimization for cross-platform games. People are sure to still buy them but they most likely will definitely pay attention to kick ass exclusive games for being even better.
(PS-If that sounded like complete crap, eh, oh well
A lot of games or programs on any platform including the PC do not take full advantage of the platform. The best way to take advantage of the platform is to code in assembly.
However, this is not really the best use of time for people. Developing in higher programming language in a cross-platform way will end up with better SDKs for the game company.
Here's the priority list I would run my game company.
Focus more on developing good games first. The principal task of a game publisher should be to get his game sold not necessarilly selling the console.
Make it easy for your developers. Choose the platform where it is easiest to program your game in so you can get to market quicker. If your game is good enough, people will buy the console to play your game.
Expand your audience. You can port your games to other platforms at this point. This exercise will promote code abstraction and re-use within your development group.
This is where you start grouping your programmers into the "just-do-it" crowd and "do-it-properly" crowd. They both have their uses.
The "just-do-it" crowd will make you reach your market first. Don't put too much of into this effort. Its going to be too frustrating and boring for most of them, shove them into new development ideas.
The "do-it-properly" crowd will tend to make things slower, but should know how to group and make things better for the code base. They should be promoting code re-use and proper abstractions to help porting to other platforms easier. Don't put too much of them into new development otherwise you'll have conflict with the "just-do-it" crowd.
Learn from your successes. Focus on this one first, as its more positive to your employees. Expand your libraries. Promote code re-use. Think of the sequel.
Learn from your mistakes. Prevent problems in future projects by learning from them and removing the people that don't.
Archie - CIO-for-hire
This makes perfect sense. The developers who use the VU are the ones that take the time. Quick and dirty ports wouldn't use it, but if you take the time to optomize for the VU and such, then chances are you'll take the time to optomize other things. You'll make the sound work well, get rid of the logic glitches, adjust the diffuculty so people don't get stuck because you left out some clues to where the key is hidden, etc. The publishers and developers that take time to make great games spend some of that time one the VUs.
So, indirectly, it makes perfect sense that VUs make a good judge of quality. Now if only other publishers would stop making so much shovelware. The PS2 (and other consoles too, but especially the PS2) has a LOT of great games. But for every great game the PS2 has, it has TONS of shovelware. Unfortunatly, often the shovelware gets great marketing (Finding Nemo, The Matrix: Reloaded, BMX-XXX) and great games (Amplitude, Ico, etc) don't get nearly as much and so they don't do nearly as well, because a large number of games are bought by parents who don't know what games are worth money, or what aren't worth their weight in dust.
Comment forecast: Bits of genius surrounded by a sea of mediocrity.
With few exceptions, cross platform games suck on the PS2.
They generally, in order of appearance look best on the X-Box, slightly worse than the X-Box on the GCN and then like a generation behind on the PS2.
Maybe it's the VU programming versus more direct and easier methods on the GCN and X-Box(I've heard the X-Box is easy as pie to code for[I'd hope so, considering it comes from the same company that came up with the idea of wizards for deriving classes!], and Nintendo has excellent developer support as well, whereas Sony's isn't too hot in comparison) that makes us see this effect.
I somehow doubt that however. I think the PS2 is just the weakest of the 3 systems out(and it is, regardless of whether the problem is dev[and dev support] side, or hardware side).
The key to the enjoyment of pop music is to replace any instance of "love" with "C.H.U.D."
The more eclectic you make your system, the more you limit who will spend the time maximizing its performance. If you remember the Dreamcast, they decided to go with a PowerVR type tile based architecture. It caused problems with ports and some of them required quite some rewriting before they would work properly. End result? Limited third party support, and part of why the console did so poorly.
The PS2 is the 800 lb gorilla so of course it's not going to miss out on any ports. That being said, when Sony chose to get cute with their architecture, they also chose to limit the amount of people that would maximize its power. Was it worth the (hypothetical) 5-10% speed increase in highly specialized apps if non-optimized code runs 20% slower? I'd say no, but for that answer you'd have to ask Sony.
For more info, look at the other docs on SCEA's R&D site and SCEE Technology Group's site.
Don't the studios have a vested interest in lenghtening the life span of a console? Add potential features strategicly so as to have some "wow" left over. The Madden series of games gets better looking every year on the same hardware. sort of like forward expectation control and management.
crazy.
The / in
The difficulty in programming the PS2 VUs has prevented the port of Strongbad's Vector Graphics Based Game. Strongbadzone
Maybe one day technology will catch up. Strongbad E-mail "Video Games"
D6 63 0D 70 89 81 BB 8E 7B 7C 5F 5D 54 EA AB 73
The biggest problem with the ps2 as far as perceived performance and final render quality for ported games is the comparatively limited ram and all those processors contending for the iop bus. Art and sound assets from an xbox game have to be severely cut down to fit in a ps2. Sure, VU assisted rendering would make some things look nicer, but you're still dealing with limited space for textures and sounds compared to xbox.
As far as what systems are used for most cross-platform game dev, what I've seen with this generation of consoles is most iterative design and art work being done with xbox dev kits, and then testing and tweaking finished levels for the ps2 and gamecube. For daily use, the xbox dev tools are simply faster and better integrated.
Instead of Sony bemoaning that developers aren't using the full power of the machine, shouldn't they instead concentrate on how they can make it easier for the developer to unlock that power?
Blame Sony and the PlayStation2. It was Sony, after all, who chose to go with some crazy newfangled hardware-graphics "solution" that everyone knew was going to be hard to program for. I had doubts at the system's announcement and they proved to be founded. What Microsoft did with the XBOX seems to make a lot more sense - make a system out of standard hardware that everyone already knows how to program for. That the PS2 has been around for so long and its full power still remains untapped is testament to its poor hardware design.
Cyde Weys Musings - Scrutinizing the inscrutable
I get asked all the time which machine is most powerful. It is easy to answer that it is the Xbox because it has a 733Mhz processor and NVidia GPU but in reality that is missing the point. When Sony designed the PS2 they were a couple of years earlier than the Xbox so clock speeds were lower, the 300Mhz MIPS RISC processor was very fast for the time, much faster than any available Intel chip, they were still being used in SGI workstations for example and those are beefy pieces of kit. The problem for Sony was to increase the grunt while being limited by clock speed and the best solution is to introduce parallel processing. The VUs are no different to SSE and SSE2 that Intel introduced in the PIII and P4 line but you don't hear people throwing stones at Intel for doing that.
The problem Sony has is convincing programmers to look beyond the capabilities of the basic MIPS processor and getting them to use the VUs but just as it is difficult to really make SSE kick arse it is difficult to get VUs working well. I used to program massively parallel computers (look up MasPar) for a living and they were hard, one of the reasons the company eventually failed in fact. However, the techniques used to program that beast are the same as used to write code on the VUs and SSE. I have seen applications that increased in speed by a factor of 20 (not 20%, 20x!) through the use of SSE on Pentium chips so when Sony gets annoyed that people are not using the VUs and so making the PS2 look like it isn't very powerful, well, you can see their point.
As for the relative power of the PS2 versus the other hardware, have a look at Gran Tourismo 4, or Killzone and reconsider your position if you blindly believe that power is all about clock speed and DirectX.
"I have the attention span of a strobe lit goldfish, please get to the point quickly!"
This is typical of machines that use a custom chip approach to hardware. Anyone who remembers the appalling port of Street Fighter 2 to the Amiga 500 compared with the graphically superb (but poor playing) Elfmania knows what I mean. I'm sure there are other examples of really bad ports in gaming history...
It it true that porting an application quickly is always going to result in a less impressive application than if you re-wrote from the ground up, writing ASM for all system-critical bits.
On the other hand, if you have a well-designed system then porting should still produce a quality application. This isn't the case for ports to the PS2 (as it wasn't in the case of the Saturn). This just means the PS2 is a bugger to program. That sounds like Sony's fault to me.
All the PS2->X-box conversions I've seen, while not looking as good as Xbox apps could do have mostly looked better than the original PS2 versions..
Combination - fun iPhone puzzling
Usually, a 3rd party is responsible for porting games. Unless it is a big development company or part of a bigger publisher, it just isn't feasible or worth it for one company to port the game to all platforms. Another company which only does porting is usually responsible for the port.
GTA:3 and Vice City are games that I think really suffer from being first made for PS2 and then ported...at least, PS2's limited memory issues is the explanation I heard for the games' tendency to dissappear random traffic cars and pedestrian while your back is turned. I can't help thinking those 2 games would be much better if they had started on Xbox, or GC (as unlikely as that is)
SO YOU'RE GOING TO DIE: The Comic for Dealing with Death
Not really. The manufacturers may have a vested interest in lengthening the life span of a console, but the third-party developers have a vested interest in pushing as much "wow" into their games as possible. Unless they're lazy, distracted, or disinterested.
It's not that EA Sports had the extra "wow" to give on previous games. I seriously doubt that any video game project manager, on a Madden football title or any other, will look at test-play footage and say...
The Madden games look better every year on the same hardware because they produce the games annually, and each year they learn some more about the process from the previous year's efforts.
That said, can developers get lazy? Yes, and on any platform too. Who wants to reinvent the rendering engine when the one in the previous game (or this textbook) worked so well? Sooner or later it happens to everyone. The PS2 has a sharper learning curve. Getting lazy when developing on it does more damage and produces lackluster results. The Xbox was designed with a smoother curve, such that it required less effort to produce good product, which makes it a good sell to the average developer.
Sony's thinking, way back then, was that because no developer wants to produce an average game, they'll take the time to master the technology in the PS2, so they built it with as much potential as they could cram in. Given the topic, that thinking, noble as it might have been, seems to be hurting them now. People who develop something that looks good on the Xbox take a look at the PS2's devkit, say "What the hell is all this?", and just shove some code in a folder and compile it. "Aw hell," they'll say, "this is Sony. They can make anything look good. And if it doesn't, we'll blame them for making their tools so hard to use."
The only way to really compare the two platforms would be to build identical games from scratch using the same basic specs and ALL the tools and tricks that each platform makes available. So far as I know, nobody's done that. (Corrections, anybody? Please?) Were they to do that, my money'd be on the PS2 to pump out more wow, even though it came out months before the Xbox. Steeper learning curve, but higher potential.
You cannot truly appreciate Dilbert until you read it in the original Klingon.
Really, who fucking cares about clock speeds and vector units unless you are actually programming the game? Every Xbox owner likes to pimp their system because it has the best this or that, but what difference does it make if none of the games are fun?
The Xbox has some quality games with good graphics, but the PS2 has so many cool games that you don't see for other systems. I really don't care if the PS2 used a tin can with string to send data across the bus. It was only last month that I finally retired my Dreamcast. now *that* system had some fun games.
WHat exactly is the point of creating games that do more than 30fps? Remember this is a console that PLUGS INTO A TV! There is NO POINT
in writing a game that does 60fps if no one is going to see 50% of the frames. Far better to use the processing power to do the AI or phyics
calculations.
I'm not going to speculate as to the numbers of games written for which console first I can say that it really doesn't mater. As has been pointed out in other responses to this article, there are no Compilers that effectively use the PS2s VUs, so even if the game is original executed on a PS2 it is being most likely designed by these lazy developers that don't want to put in the effort necessary to utilize the VUs. Even out side of the VUs there are other differences in design that developers aren't used too, such as continual utilization of the internal bandwidth(an article I read just over a year ago stated that most games are utilizing less than 10% of the internal bandwidth of the PS2 at most times).
I'm not going to argue over with machine is better (it would be similar to arguing analog vs. digital), I just think people need to realize that the PS2 uses newer technology than the other systems and that takes time for developers to get used to. Luckily for us there are companies out there, such as Naughty Dog, that are willing to try and learn this new system if for no other reason than to be a step ahead when the next generation of consoles come out, and archaic technologies like DirectX are not able to keep up with system demands.
On Slashdot a poster inquired
.html
Why ASM code is so mired.
The story of Mel
Was the answer his question desired.
I'm sorry.
Funnily enough , I was having a conversation with a mate of mine who is a games developer.
...
X-Box games also suffer from dodgy PC ports syndrome. Most X-Box games are simply derived from the Windows version of the source code and simply tweaked a little to run on the X-Box which leads to an obvious performance hit since they are not always developed specifically for the hardware. Obviously it makes a lot of sense for Game studios to build the PC version and then port it to the X-Box, it saves money, development time, and its easy to do since The X-Box really is nothing more than a butchered PC. And lets not forget how old PC Architecture really goes back...
Playstation 2 despite its lower clock rate is a much under-rated machine (mainly by Microsoft zealots). But remember it Is 128bit risc chip with a modern architecture and has some pretty sophisticated parallel Vector processing units to boot. It can shift data around very very quickly. The main let down on the PS2 is its memory limitations.
I dont understand why of recent times there have been an influx of PS2 games with inferior performance. I have seen a number of games which should be no-more processor intensive than other PS2 games. These games tend to be the ones that appear on all three platforms.
It is surely the fault of the programmers for being lazy and letting standards drop in favor of making a quick buck. I am a big believer in pushing boundaries, Real programmers know how to get every last ounce out of the hardware they code for, and then some!! thats where innovation comes from.
Sony have the advantage that they are not using off-the-shelf components. By the time PS3 hits the streets I imagine the hardware will be way more advanced than any off-the-shelf parts Microsoft is planning to stick in the X-Box 2.
I think that it is right for Sony to have performed these analyses and it shows that they are commited to ensuring a high standard on their platform. I hope that the naughty game studios take notice and pay more attention to their products in future.
It also just goes to show that "Cross Platform Development" , although it might look good on your CV is not always the "right" way to do things.
nick
Electronic Music Made Using Linux http://soundcloud.com/polyp
Is easier to spot a port on the xbox than in any other system, games designed for the xbox usually pull effects like glow, soft shadowing, and fov (they are relatively easy to pull by the gfx card). Just Try to compare the looks of otogi, PGR2 or rainbow six with those of ROTK, THUG or jedi knight2 (and those games are so cool what a shame!) crap! even freaking spongebob squarepants looks cool with those effects! Fortunately some companies go the extra mile and pull the same look on all platforms (prince of persia and need for speed underground) but they are still just a few.
Well at least most ports have antialias on (it takes a line of code to turn on/off but ok).
Go ahead MOD my day!
More opinions here
An efficient way to deal with this problem for developers would be to use middleware, per example Renderware not only this helps to make the developing phase a little shorter, each version of the game is optimized for each console. Per example MKDA plays and looks exactly the same in each version. Prince of persia was also made in middleware software and the results are astonishing.
p.s.
Max Payne 2 for xbox takes the cake as one of the worst ports ever, not only they made missions shorter and the auto targeting cheap, it doesnt pull ANY effect of the PC version. it doesnt even feature reflections on mirrors. Oh yeah and it hangs on the same level lots of times.
Go ahead MOD my day!
More opinions here