Provocative comment by Id member about PS2 (and Gamecube)!

[London Geezer]

I would like to, once again, thank london-boy & chap for derailing yet another topic. Great job, guys.

huh? excuse me, but i was in this thread long after it was derailed by other people... come on this thread has gone from Doom3 on ps2 to EVERY-F**KING-THING ELSE!! how did I derail the topic? by talking about progressive scan when SOMEONE ELSE asked about it? or when i popped in to talk about what DICE in the Xbox is, when SOMEONE ELSE asked what it was?
come on, stop patronising people...

anyway, PLEASE BRING THE OLD CHAP BACK!!! XBOXCHAP was much better than PS2CHAP...

 

[Phil]

derailing it that far that it is no longer relevant to the forum its in - I thought that was clear and didn't need further explaining. Despite the topic already being off-topic (discussion above with PC-Engine), it's at least relevant to the forum and to the topic to some degree. Since people obviously can't ignore these things, perhaps opening a thread for the usual suspects may be a good idea.

 

[London Geezer]

yeah whatever Master...
can u even remember what the topic was in the first place?
if "asking whats wrong with Chap" is derailing the topic so much, then i'm sorry, but i wasnt the only one concerned with his mental health.
we had 1299 discussions which ended and new ones started... whats with the patronising attitude? it's not like u're posting about the topic anyway....
this thread is DEAD, get over it, people were having 82 discussions at the same time because of this.
see, this is what happens when u have 24 pages worth of discussions... maybe u should get used to it.
if we had to open new threads for every discussion had on this one, then the board would be full of similar topics about PS2 and Xbox and their technicalities, one about Doom3, one about what the hell is going on with Chap, and 9182 about stupid little things. which by the way, i did not start.
so next time, think about the fact that not everyone here is a little kid.

 

[chaphack]

ANYWAY, have you peeps seen the new DOA:O trailer?

http://ruliwebfile.intizen.com/mpeg3/tgs2003/ruliweb/ruliweb_doaonline01.mpg

http://ruliwebfile.intizen.com/mpeg3/tgs2003/ruliweb/ruliweb_doaonline02.mpg

looks better/solid than DOA3 to me. gotta love this rain scene.

 

[cthellis42]

Linkee's no workee.

 

[wazoo]

Have you seen Monster of Capcom, the trailer at IGN is amazing

kind of PSO meets Jurassic park meets MGS3.

I dunno abput an official link

 

[Squeak]

1. So you can look at a game like J&D/R&C and magically know it's pushing more polys than any Xbox game to date? Xbox CAN push more polys than PS2 whether you're talking about RAW or in-game. Tests done by EA already give a good idea as to what Xbox is capable of compared to PS2. PS2 can do 66 RAW Mpolys/s while Xbox can do 116 RAW Mpolys/s. Why would Xbox almost double the PS2's RAW poly number then suddenly push less than the PS2 when it's in-game??? I don't see the logic in that.

Take a look at these screens of Jak II. Do you still think that PS2 is inferior in the geometry department?

3. Why do high profile PS2 games not use mipmapping? Maybe because it's a trade off to get better performance somewhere else?

It is difficult to see where a performance increase, so big it would be worth compromising IQ so much, would come from. In fact in some ways mip mapping should increase performance.

Mipmapping should be used as much as possible to clean up the shimmering.

I completely agree.

Trilinear and Anisotropic filtering is superior to Bilinear period.

I didn’t say they were not. I said that few xbox games actually use them

Just because PS2 can emulate it in software doesn't mean it's fast enough to be used in-game. My i486DX could do raytracing too AND??? Xbox has the potential and often does use Trilinear in games instead of Bilinear. Anisotropric is useful for road signs etc. in racing games.

The clamp function on GS is hardwired. It is used to “stretch†textures, it’s just a general function that can be used in a variety of ways.

4. There are Xbox games that have high resolution textures that PS2 can only dream of end of story. Every console has games that display blurry textures. They're most common on PS2 than Xbox.

In most cases the xbox high resolution textures is “just†a low res base texture, with a detail texture on top of it.

6. Nobody said 720p was free. NOTHING is FREE!!! Point is there are 720p games on Xbox while there is none on PS2 even after being on the market for what 4 years now?

Okay, so PS2 can’t fit a whole 720p frame buffer system inside its 4Mb VRAM, so what? It saves huge amounts of bandwidth by having all the buffers on die, I think that’s an acceptable trade off.
Besides 480p looks almost as good, and doesn’t have the performance hit of 720p. The important thing is that it is progressive, that is much more noticeable than a slight increase in resolution.
By the time PS2 is faced out, it will still only be a very small percentage of the world’s population who have access to HDTVs.

7. Point is it came with EVERY Xbox sold. What it is used for depends on the developers.

And?

9. Some games don't NEED a divided screen. Played any tennis or volleyball games lately???

Oh, your mean lard arss sports? I thought we where talking about real videogames?
Seriously, I forgot about those, I love Mario tennis on the N64.

 

[Grall]

Superiority is based on benchmarks and or in-game numbers. What's the highest in-game numbers for ANY game on PS2? What's the highest for Xbox? I rest my case.

Why should anyone beyond the fanboi crowd care? Explain that clearly and in a rational and consise manner in no more than five words, please.

Oh, and you can't use the words "because I love Bill Gates", or any variation thereof.

Again:
Yadda yadda. Who gives a flying you-know-what?

Why do you feel this weird need to have everybody acknowledge this? Just play the games and get over it, okay.


*G*

 

[Crazyace]

Comparing apples and oranges...

Hi Chap,

You are absolutely right when you go on about Xbox numbers being higher.. but appearences are deceptive..

The EA shader paper, which gives some pretty reasonable BM figures..
shows raw transform of the bunny model as 32million verts transformed for PS2 against 94 million for Xbox - a big win for the microsoft machine - but the number of polys generated is 25million against 64 million - which is a lower ratio.
For the skinned models the differences are different - 11million verts against 30 million, but 9 million polys against 14 million... The effective raw improvement is only 50%, which is not going to be totally obvious in a game situation...

I tend to think that the biggest advantage the Xbox has is the memory - effectively twice the ram for graphics assets once code has been taken into consideration.... but I reckon that if the publisher wanted to produce Doom 3 on PS2 , someone somewhere would produce a pretty close version... lights&all, it may not be the same tech, but visuals would be close ( Look at SH2 PS2 and Xbox - vertex shaded torch against pixel shaded torch - the difference is there if you look for it, but it's not something that ruins the game... )

The same thing happened last gen - N64 owners slated PS1 for no filtering and warping textures, and the PS1 programmers generated work arounds that ended up minimising the visual differences as seen by the general public..

 

[Josiah]

I think D3 is far too dependant on pixel shaders etc. that PS2 doesn't have, it wouldn't be a good port. It would end up looking more like Quake3. Hell they could port D3 to Gameboy if they wanted too, but it just wouldn't be the same game.

Quake 2 was a little different scenario, it could run on the PC in software mode to begin with, and didn't really do anything PS1 couldn't do. The main problem was the lack of RAM on PS1, which is just a matter of optimizing to squeeze it in.

 

[Fox5]

Nick- Considering burnout 2 probably pushed more polys than Burnout 1 and was multiplatform, it would also mean all the consoles can handle those numbers.
Same with Timesplitters 2, of which the PS2 had the lowest framerate.
(and I wouldn't consider Burnout 2 or TS2 even near the best looking games on any of the platforms)
BTW, what is this Performance Analyser that will yield major improvements?

 

[wazoo]

Question : with panajev on Ga, we had a short discussion about how you should count pol/sec. According to him, the number of pol is the product of the used pol by the number of textures applied on them (because you have to resend the geometry, sorry Panajev if I misunderstood you). That seems a bit weird, but maybe the only way to understand those crazy 18M pol/sec of GPC (this number does not appear on screen), maybe dividing it by an average of 3 textures, you got a fairly good 6M pol/sec.

I doubt devs count like this, but who knows ??

 

[Crazyace]

It would be an interesting technical exercise

On the one hand you could try to duplicate the dot3 bm look with multiple passes - or you could keep the most important stencil shadows, and apply lighting on a vertex level with a much denser mesh.. with some variant of a portal system to dynamically switch lighting models (rather than geometry ) around shadow areas..

 

[randycat99]

That seems a bit weird, but maybe the only way to understand those crazy 18M pol/sec of GPC (this number does not appear on screen), maybe dividing it by an average of 3 textures, you got a fairly good 6M pol/sec.

I doubt devs count like this, but who knows ??

Does the look of GPC strike you as a game that textures that extensively (in the number of 3 layers)? I wouldn't guess it to be so, but I don't know much about the game other than it is a rather good looking Formula racer sort of game.

 

[marconelly!]

Well, GPC game *does* have a *crapload* of very detailed modelled cars on screen at the same time running at 60FPS.

 

[randycat99]

If you are talking about geometry when you say "detail modeled", then I couldn't agree more. I think we are both agreeing that suggests a lot of polys going onto the screen, right?

 

[Megadrive1988]

I think D3 is far too dependant on pixel shaders etc. that PS2 doesn't have, it wouldn't be a good port. It would end up looking more like Quake3. Hell they could port D3 to Gameboy if they wanted too, but it just wouldn't be the same game.

exactly. you can make any game for any machine, but this means its there in name only.

examples:
Virtua Racing and Virtua Fighter 2 on Genesis.
Daytona USA on Saturn
AfterBurner on SMS and NES
Double Dragon on 2600
Street Fighter 2 on Gameboy

Doom 3 and Doom 3 engine based games would be better served on PS3.

 

[cthellis42]

Nick- Considering burnout 2 probably pushed more polys than Burnout 1 and was multiplatform, it would also mean all the consoles can handle those numbers.

I don't think he was challinging any of the other consoles at all, he was just posting a known number. (Which was the topic at hand for all consoles after all. ...at least at that particular point. )

(and I wouldn't consider Burnout 2 or TS2 even near the best looking games on any of the platforms)

And this is, of course, the point coming from all sorts about number of polys or whether a console uses frick and another uses frack, and how many heebie-jeebies one can crank from the hoopla.

In the end, it's the game that matters. And we've all seen good-looking and technically-adept games suck before.

BTW, what is this Performance Analyser that will yield major improvements?

From an article about GDC 2003: "Just like on the PlayStation, SCEA is now sharing its PS2 Performance Analyzer, which measures and analyses a game's performance and helps creators to optimize a game's capabilities on the platform."

I'm not sure if it was only available after that point, or select devs had it before, but it basically gives in-depth analysis as to how one's code is running on the machine: where one had headroom, where bottlenecks are, what percentage of resources are being used...

Hard to find released figures for it from games for whatever reason, though. One of the only ones I know was talking about GT3 (I guess this was a premiere title to show off figures with) only using between 1/4-1/3 of the PS2's resources on the PA.

Does it mean devs have any hope of hitting anywhere near 100%? Of course not. (I think a tech demo has come close, though. Wasn't there one posted on here a while back? I have the file around somewhere, but don't have a link.) But an in-depth analysis toolset is never a bad thing. <grins>

There was a bit more murmuring going on about it beginning of the year (because of GDC I guess), and one of the titles that got some press because of "heavy use of the PA" was Primal. Certainly not a bad game to look at. ^_^

Hence why I'd love to be able to see more games and figures and know which developers may be using it for what... 's just cool. ^_^ Still won't tack anything down to "absolutes" of course, but more information is always interesting.

 

[PC-Engine]

According to Adam Billyard Chief Technology Officer at Criterion, Burnout 1 was pushing 15 million polygons/sec on the PS2 back in October 2001.

According to internal SCEE benchmarking done at that time using the 1st iteration of the Performance Analyser, it was the game pushing the most polygons/sec on the hardware at that time.

This caused great embarressment to a number of developers who had shunned the Renderware middleware as not being up to the job.

Here are some other interesting quotes from the recent Edge magazine Equip special on the PS2.

Steve Ellis (Free Radical Design co-founder)
"In Timesplitters 2, we managed to double the performance we achieved with Timesplitters 1. I believe that with some more difficult optimisations and careful planning, at least another 50% increase is possible."

Andrew Perella (Head of Programming at Eutechnyx)
"We hope to push game performance significantly by continuing to do specific PS2 R&D, particularly with respect to making mores use of procedural geometry ability of the Vector Units."

Mike Diskett (Founder member of Mucky Foot)
"When we can actually have the Performance Analyser all the time instead of the occasional week, its going to double the perfromance we can get out of the PS2."

Mike Healey (Senior Programmer at Rebellion)
"I think the Performance Analyser will yield major improvements in performance - we can look forward to much more visually stunning games in the very near future."

Jason Rubin (Naughty Dog co-founder)
Q:How far do you think you're pushing the hardware with Jak II?
"We've got a lot more out of it, but at the same time you're always doing everything you can do at that time. If we do another game we'll get more out of it again. There's a lot more in that system."

Andy Beveridge (SN Systems Director)
"Understanding VU0, is a key part to improving PS2 game performance. Incredibly, most of the first wave of games hardly used VU0 at all! The main bottleneck of many games is the CPU, not the GS, offloading tasks from the CPU MIPS core onto its co-processing Vector units will have a massive impact on overall performance."

You can draw your own conclusions, but it seems clear that a lot of developers that are on their 2nd or 3rd titles for the console believe that there is a lot more they can do.

Doubling performance can mean anything same with the 50% performance increase. I'm sure Xbox developers are getting similar performance increases. Where are the numbers?

 

[PC-Engine]

Why should anyone beyond the fanboi crowd care? Explain that clearly and in a rational and consise manner in no more than five words, please.

Oh, and you can't use the words "because I love Bill Gates", or any variation thereof.

Again:
Yadda yadda. Who gives a flying you-know-what?

Why do you feel this weird need to have everybody acknowledge this? Just play the games and get over it, okay.

Why do you have this weird need to constantly whine like a little girl? Nobody is forcing you to read this. I'd rather you not read further as your whining is extremely annoying...

Comparing bechmarks and in-game numbers will give you a very flawed result, as you are making yourself dependant on the developers ability to get performance out of the system. How do you know that the Xbox developer hasn't tapped more of the hardware opposed to a PS2 developer? I gave you indications why your comparasment is flawed - you have not made any attempt to prove that wrong so far (and benchmarks of the like don't cut it).

Do you even read??? If you want to disregard the benchmarks then go ahead you still haven't answered the question. What is the highest polygon performance number for ANY PS2 game from ANY developer??? Are you saying there are no PS2 developers out there?? and that they're just imaginary??? Can you even grasp this simple concept. It's a very simple comparison. REAL games the best vs the best. What do you have to say about ERP's statement??? You can't say anything because you're not a developer end of story.

I'll remind you that we are talking strictly of hardware superiority, so developers competence/ability/motive to not include progressive out as some sort of evidence is moot. The hardware is capable, period. Don't you see how pointless your argument is?

Nobody said PS2 isn't capable of progressive scan. Point is 99% of Xbox games support progressive scan while 1% of PS2 games support it and PS2 has been on the market far longer than Xbox. That feature alone puts Xbox above PS2 period. Same with DICE and DTS...understand???? If it makes you feel better I'll retract the 720p comment since only a couple of Xbox games support it.

Actually, the bottleneck as far as I know hits in first with the available memory/bandwidth. Still, in some other posts in mainly PS3 threads, Pana I am sure mentioned being a transformation peak on EE at just above 100 tris/sec.

Heh so the gobs of eDRAM bandwidth is of no use at all and is actually a bottleneck???

Why won't someone who really knows chime in?? Any volunteers from the PS2 supporters?

I still don't quite see the point of your entire post though - Why are you comparing specs in the first place? What's the point? What do you hope to achieve?

I'm sorry if you don't see the point, however this is a forum a place for discussion isn't that good enough for you??? I'm making a point using concrete numbers instead of some subjective argument over "well I think PS2 games looks better...well I think Xbox games looks better". The point is Xbox hardware is superior to PS2 hardware in most ways. Whether you accept it or not is up to you...I couldn't care less.

Taking your words, the best evidence are the games, in which I see areas where PS2 is way more impressive, though much less in others.

It's obvious you want to turn it into a subjective matter because the concrete numbers are too hard for you to accept.