How was Ps2 thought to work ???

[function]

shame that for some people PS2 always comes out losing, either against DC or Xbox.... go figure...

London-boy, in what possible way has the PS2 come out as "losing" in any part of this discussion?

Or has the entire point about using untapped features on the DC gone completely over your head and somehow been interpreted as "DC is better than PS2"?

 

[London Geezer]

shame that for some people PS2 always comes out losing, either against DC or Xbox.... go figure...

London-boy, in what possible way has the PS2 come out as "losing" in any part of this discussion?

Or has the entire point about using untapped features on the DC gone completely over your head and somehow been interpreted as "DC is better than PS2"?

huh?
i wasnt talking about this thread in particular mate...

 

[function]

My bad, I thought it was in regards to this thread. This and the ATI/Xbox threads are the only ones I've been reading in the console forum ATM.

 

[London Geezer]

simon,

You know what. The PS2 is able to do bumpmapping better than the Xbox, it's OK if no game has demonostrated that, because with time the true secrets of the PS2 capability will show you the way. I heard it's able to do real-time raytracing too, just that no one has discover that secret function yet.

Now I know you are a troll. Go cast your nets elsewhere.

he was being sarcastic.... u know, "hidden features of DC" : "hidden features of PS2" = PS2 : Xbox
got that? pretty neat heh
its the same argument ps2 fanboys use when trying to defend the "hidden features" or the "untapped powers" when comparing it to the xbox........ it was supposed to be funny...

 

[Simon F]

he was being sarcastic

Yes that was obvious. I just found it tiresome.

 

[London Geezer]

he was being sarcastic

Yes that was obvious. I just found it tiresome.

no reason to call him a fonboi...

 

[cthellis42]

And, of course, it's not so much an "issue" but more of a "lamentable pondering"...

Indeed. Sometimes I even lamentably ponder the fate of the Jaguar. :?

<cries> My poor Jaguar!

The official death of DC announcement was made by Sega of Japan in January 2001, only 26 months after the machine was launched. This had enormous implications for software development on the machine.

Hm, shit it was. -_- Go revisionist history, go! Hehe... (Frankly I just couldn't remember, and didn't think it was THAT much before the other guys were launching.) But still, my, er... 2 year comments stand! Hehe... Go me!

PS2 had only actually been out a little over 6 months at this point, and it's PR masters had done an awesome job.

Nothing much to say here other than... PH34R THE P33-4R! (Sorry, sorry...)

I'm getting really into this lamentable pondering, as you can see.

Jaguar...!

 

[YPO]

"Many of the people who have posted here more than once already know that. They also know that I don't suffer fools gladly. "

I know you are, but I find it funny that you used that against Marconnelly. Are you an employee of Sony that designed the PS2? If not then how do you know if the DC is better at what? After all that's the argument you used against Marconnelly right? And no, I'm not a programmer or expert in graphics of any sort. I use my own eyes. A lot of first batch PS2 use multi-texture effects. TTT, RRV, Madden....even that crappy racing game featuring those buggy cars had it. There's way less use of multi-texture effects in DC games. Pretty much everyone agrees with that.

"Certainly the rendering effects improved over time, which is only to be expected from a device which possibly had about 4~5x (*) as much silicon as the DC. "

And what exactly are you saying here? That the PS2 is better at those? IF so you are agreeing with Marconnelly, so what's the argument?

Oh yes I'm a troll, only because I'm using your logic in a sarcastic manner.

edit: Thanks for the info London-boy, now we just have to wait for the secret capibility to show up.

 

[Simon F]

"Many of the people who have posted here more than once already know that. They also know that I don't suffer fools gladly. "

I know you are, but I find it funny that you used that against Marconnelly. Are you an employee of Sony that designed the PS2? If not then how do you know if the DC is better at what?

I don't believe that I said that DC was superior at multi-texturing - only questioned the assumption that DC was inferior. If I'm not greatly mistaken, Fafalada, a well known developer of PS2 games and eloquent frequenter of these forums, has listed the blending modes available on the PS2 which, in my humble opinion, seem rather limited. These blend modes are what are generally used to perform multi-texturing. The DC has a very much richer set of functions but, AFAICS, sadly few were used 'in anger' in the games to date. You can see more extensive use of such modes in current PC (and likely XBOX) games.

And no, I'm not a programmer or expert in graphics of any sort. I use my own eyes. A lot of first batch PS2 use multi-texture effects. TTT, RRV, Madden....even that crappy racing game featuring those buggy cars had it. There's way less use of multi-texture effects in DC games. Pretty much everyone agrees with that.

Do you question the existence of atoms? One reason for the frequent use of multi-texturing on PS2 is that it is rather lacking in the texture memory department. One way of increasing the apparent texture information is to use multi-texturing avoiding correlation between the applied textures.

"Certainly the rendering effects improved over time, which is only to be expected from a device which possibly had about 4~5x (*) as much silicon as the DC. "

And what exactly are you saying here? That the PS2 is better at those? IF so you are agreeing with Marconnelly, so what's the argument?

I was implying that by using brute force with simple operations you can achieve some nice effects (once you discover how to do them) and PS2 has quite a bit of brute force.

I said I was going to be blunt with Marconnelly, and perhaps I was unnecessarily offensive, but he was effectively asserting that I didn't know my subject.

Oh yes I'm a troll, only because I'm using your logic in a sarcastic manner.

If you look back at your post, you will see that you entered the discussion with an impertinent question and ended it with outrageous sarcasm. I don't think that was particularly called for.

 

[YPO]

"I don't believe that I said that DC was superior at multi-texturing - only questioned the assumption that DC was inferior"

So what exactly are you saying? That the 2 machines are equal in that aspect? That's the only other choice left. Like I said, I'm no techie so I can't argue with you on the technical front. Maybe DC has more hardwired function or whatnot and the PS2 has brute force. In the end of the day, the games that are out showed PS2's better at it. Not only in multi-texture effects but shadows, particles, etc. The only thing matter is what you see on screen, doesn't matter how it is achived.

"One reason for the frequent use of multi-texturing on PS2 is that it is rather lacking in the texture memory department. One way of increasing the apparent texture information is to use multi-texturing avoiding correlation between the applied textures."

That's just weak. Multi-texturing enhances the look of game. A lot of games use it because the developers wanted to use it, not to get around some limitation. There might be few cases, but I doubt that's the main reason for multitexture usage on PS2. Sorry but multi-textured objects in genereal looks better those with single layer textures.

Sometimes sarcasm is the only way for realization

 

[Crazyace]

I wish there were more blend modes implemented on the GS, but it's still a lot better than the PS1...

The DC chip has excellent blend modes, and an extremely elegant design that gives it exceptional opaque draw capabilities,
as well as the cool per pixel sorting ( very slow, but amazing that it is there at all ) It isn't perfect though..

The GS is optimised for high throughput semitransparent rendering...

EE+GS did a good job with VF4evo , a SH4+ELAN+2xPVRDC with a lot more memory ( not too bad for AM2's first try )

Simon, I think that the reason a lot of early PS2 games had DC poly budgets came down to the fact that they were rushed DC ports or rewrites..

( Plus a lot of the transistor budget in the GS is the embedded ram which allows major cost reduction later in the system life )

I'm looking forward to seeing a PVR DX9/DX10 part, as the deferred rendering paradigm should shine when people start writing >100 line pixel shaders...

 

[Lazy8s]

cthellis42:

I was just saying that if the machine had some HUGE capabilities missed/untapped in it when the games stopped a-developin' I'd wonder just what people were doing in the years previous to not notice/take advantage of them.

Sometimes capabilities go untapped simply because the development community isn't ready for them. Just as technological capacity improves over time, game design is a craft that also evolves with time.

Developers are only now starting to take advantage of polybump-type schemes in their Xbox engines with frequency even though it's the same hardware with the same bump mapping spec that's been around since 2001. Advancement in technique like this could've benefited DC greatly, as developers circa 1999 just weren't very accustomed to the advanced, hardwired features PowerVR2DC efficiently supported (dot product bump mapping being one)... a negative aspect to being the pace setter in technology. Developer familiarity with, and effective utilization of, modern 3D techniques is better now than it was several years ago, even when working on the same hardware.

wazoo:

LeMans team claim more than 3/4M for its racing game

Speaking with Melbourne House at the old Dreamcast Technical Pages message board, some of the specs they actually divulged about Le Mans on DC were a polygon rate of over five million a second, usage of anisotropic filtering, and a liberal application of multitexturing.

 

[cthellis42]

I was just saying that if the machine had some HUGE capabilities missed/untapped in it when the games stopped a-developin' I'd wonder just what people were doing in the years previous to not notice/take advantage of them.

Sometimes capabilities go untapped simply because the development community isn't ready for them. Just as technological capacity improves over time, game design is a craft that also evolves with time.

I know that and stated that. At the time, I was riding under a different assumption (forgetting when offhand Sega officially pulled the plug), but even still you can look at the difference in quality between launch titles and games that made it under the 2 year mark and see some tremendous furtherance. You can even look at some of the games by the same developers and compare them to later titles (Square's "Bouncer" to "FFX" for instance) and mark the advancements they made on their own. Then you can perhaps mark the continued advances (say, SH2 to SH3, or FFX to FFX-2) and gain a reasonable gauge.

The Dreamcast had ~18 months as the clear frontrunner and had many great-looking titles, not to mention being an easier platform to develop for than the PS2. I'm just skeptical of anyone who things there was magic coming... Slow advancements and some new tricks--same as always.

Developers are only now starting to take advantage of polybump-type schemes in their Xbox engines with frequency even though it's the same hardware with the same bump mapping spec that's been around since 2001.

I was given to understand this comes mainly from most developers really sucking at BM since 2001...

Advancement in technique like this could've benefited DC greatly, as developers circa 1999 just weren't very accustomed to the advanced, hardwired features PowerVR2DC efficiently supported (dot product bump mapping being one)... a negative aspect to being the pace setter in technology. Developer familiarity with, and effective utilization of, modern 3D techniques is better now than it was several years ago, even when working on the same hardware.

Certainly. Some developers were already pushing past "theoretical limits" and providing 5mpps with lots of nifty gee-gaws. <grins> But as most point out, many times providing new "stuff" will come at a cost elsewhere, and so the net effect would be...? Slow, gradual steps, I imagine. Same as always.

I meanly figure if the DC had more "free" capabilities, even if they were really HARD to program for, ONE of the developers would have been pushing insanely for it, just to show off to the others. And that, of course, would leave a title of note sitting around for us to look at and take into effect while continuing to ponder lamentables... 8)

 

[wazoo]

wazoo:

LeMans team claim more than 3/4M for its racing game

Speaking with Melbourne House at the old Dreamcast Technical Pages message board, some of the specs they actually divulged about Le Mans on DC were a polygon rate of over five million a second, usage of anisotropic filtering, and a liberal application of multitexturing
.

Yes, I was refering to this thread. Anyway, if you put so much faith into that, then you have to trust their claim about 20M pol/sec for GPC on the ps2, which put the ps2 well above the DC

 

[Crazyace]

and they stated 18M poly's for their PS2 racing game...

Currently 'polybump' technology relies on renormalisation maps as well as dot3 blending - DC can probally implement it ( Even N64 could have a good stab ), but you may end up lacking CPU resources...

I tend to prefer actual high poly character over faked bm effects, DOOM has very angular sillohettes that spoil the illusion. DOA3 characters are much more impressive ( though they still suffer from 'plastic' material attributes, which may well be intentional )

 

[YPO]

"some of the specs they actually divulged about Le Mans on DC were a polygon rate of over five million a second,"

Strangely their spec for the PS2 version stated 2 million pps. I think you've left out the keyword "engine" here Deadmeat. Just like the GT3 engine was capable of 10 million pps when the actual game didn't push that number.

 

[cthellis42]

<blinks> How did Deadmeat get involved in this...?

...and should we start running now?

 

[ShinHoshi]

Currently 'polybump' technology relies on renormalisation maps as well as dot3 blending - DC can probally implement it ( Even N64 could have a good stab ), but you may end up lacking CPU resources...

Perhaps I am going a little out of topic...well, perhaps no, sure. But what was the N64 problem with pushing polygons ??? Was the isue related to the fact of only having one processor for doing everything (graphics, sound, IA...) ?

 

[Fox5]

DC could have had a major increase, perhaps if a new way of programming for it was discovered, but with such an easy to program for system, I doubt it. Conker's Bad Fur Day and Perfect Dark(and evne paper mario) had huge improvements over even games released at the same time as them, and besides the blurriness and low framerate, I'd say they were on par with dreamcast games. Conker's Bad Fur Day had large enviroments and draw distances(also seen in super mario 64, but most n64 games did not have that), good character models, great textures, actual fluidly moving textures, great particle effects, an ok frame rate for n64, good sound, and many other impressive things not seen in most n64 games. Ok, maybe not. And then Perfect Dark had great particle and lighting effects, complex enviroments, good textures, and good physics. However, the n64 did not start with very efficient programming methods(the ucodes or whatever), so rare had to make their own. Wouldn't be surprised if they disabled some of the n64's features, like bilnear filtering(though it doesn't look like it), to do what they did.

 

[zurich]

cthellis,

IMO, The Bouncer cutscenes were far superior to FF X's (hair being the most obvious.. *shudders at Yuna/Lu Lu's hair*)