PSP NEWS!

Panajev2001a said:
N64 had a more then respectable combiner ( by admission of the own Art-X guys who had worked on the N64's RCP )

N64 had a *color* combiner for fog and alpha blending etc, but to my knowledge it did NOT have any *register* combiners of any kind. No software shows any use of such features anyway. It's very rare to see any multitexturing at all in N64 games since it's got pretty rotten fillrate. :)


*G*
 
Grall said:
Panajev2001a said:
N64 had a more then respectable combiner ( by admission of the own Art-X guys who had worked on the N64's RCP )

N64 had a *color* combiner for fog and alpha blending etc, but to my knowledge it did NOT have any *register* combiners of any kind. No software shows any use of such features anyway. It's very rare to see any multitexturing at all in N64 games since it's got pretty rotten fillrate. :)


*G*

Go ask the Art-X guys, they went on record about Flipper's TEV combiner unit being derived from the N64's GPU ( which is basically the RCP ).
 
Go ask Sega about the SCU's DSP... ;) maybe it was not fast enough and the fill-rate limited its use...

If I could find the quote I would feel much better because it is hard to remember what exactly did they take ( and enhance ) from the N64's GPU... I remember it was the combiner unit, I whish I remembered more...

Sorry Grall...
 
Surely it would have found use SOMEWHERE in a game to enhance some special effect or such, think Link holding up treasure from a chest etc. However, it isn't used ANYWHERE EVER, which makes me believe the feature just isn't there. Remember, RCP is ~4m transistors IN TOTAL, that includes 8k+ of SRAM, vector co-processor and a MIPS R4k CPU core plus all the other stuff, rasterizer, memory controller, sound hardware etc.

The ArtX guys might have extended the functionality of the RCPs color combiner and turned it into Flipper's TEV, but I don't think there's anything like the TEV in the RCP. That was a very primitive chip after all really. It didn't even do proper bilinear for chrissakes, it used three samples per pixel and did very poor interpolation between texels across a polygon surface. Its subpixel precision was great though, zoom in on those screens of scrolling text in Goldeneye/Perfect Dark for example. :)

*G*
 
Tag honey, I know DC supports bumpmapping and that it isn't used for anything EVER, but that doesn't mean N64 has a register combiner in its GPU. Using the "never seen ever" style of argumentation has caused a lot of noise in another thread recently.

I prefer to see evidence things EXIST as compared to the opposite. ;)


*G*
 
Or perhaps it just ISN'T THERE?

The fact I never seen it listed in any kind of specs sheets about the N64 speaks for that as well.

If it was there, I think at least one game would have used it SOMEWHERE, at least on the title screen or something.

*G*
 
Grall said:
Or perhaps it just ISN'T THERE?

The fact I never seen it listed in any kind of specs sheets about the N64 speaks for that as well.

If it was there, I think at least one game would have used it SOMEWHERE, at least on the title screen or something.

*G*

while I see your point, it could just be a oversight during design.

to take a less extreme example BM isn;t used in PS2 titles thogh technically it could be done. the trade off are apparently unbalenced so omitted.
 
jvd said:
There hasn't been a move because of battery life. Everything before mbx would kill the battrey while being used. That is why you haven't seen 3d in phones. Since mbx is both 3d and 2d and uses what .05 watts . You will see it take off . Esp if the price is very cheap. At the begining only a few high end phones will use it. Alot of pda's will use it . But as time goes on mbx and other chips as capable will take over that market. One day pdas will be fully replaced by phones just as capable.

But my question is this: If 3D tech on handhelds will be the next "killer app" that has only been held back by power consumption issues, then why hasn't 3D hardware become standard on desktops where power concerns are largely irrelevant. We have had efficient consumer-level 3D harware on desktops for 7+ years and there has been no movement towards mass adoption yet. Hence my extreme pessimism.
 
akira888 said:
But my question is this: If 3D tech on handhelds will be the next "killer app" that has only been held back by power consumption issues, then why hasn't 3D hardware become standard on desktops where power concerns are largely irrelevant. We have had efficient consumer-level 3D harware on desktops for 7+ years and there has been no movement towards mass adoption yet. Hence my extreme pessimism.

I'm not sure what you mean by this. Today, 3D hardware is pretty pervasive with Intel's enormous integrated marketshare making up a huge part with nVidia trailing and ATI, et al following.

3D has reached a point where everyone has some lvl of accelerated potential (DX7 as a proboble mean), your argument held water in 2000 but it's no longer as applicable. In the near future it will become entirely irrelevent (circa 2005) when Microsoft releases Longhorn, which will require DX7 as a minimum and DX9 preferably for basic GUI functionality.

If anything, I question why there hasn't been a serious challenege to Nintendo's POS products in the past, perhaps arising from the PPC market in the form of a specilized product/standard spearheaded by an industry leader like Microsoft.
 
notAFanB said:
to take a less extreme example BM isn;t used in PS2 titles thogh technically it could be done. the trade off are apparently unbalenced so omitted.

Can I just ask a simple question? Even on xbox that seems to be the best at bumpmapping, you still have to calculate the bend normal at each texel, right? Then wouldn’t it be possible to put one of the PS2s vector units on the task of calculating all those normals for each texel, before sending the texture to the GS buffer? According to Sonys numbers the EE should be able to 150 million 3d points a second, so calculating shading normals for a 128x128 texture shouldn't be a great deal.
 
Vince said:
I'm not sure what you mean by this. Today, 3D hardware is pretty pervasive with Intel's enormous integrated marketshare making up a huge part with nVidia trailing and ATI, et al following.

3D has reached a point where everyone has some lvl of accelerated potential (DX7 as a proboble mean), your argument held water in 2000 but it's no longer as applicable. In the near future it will become entirely irrelevent (circa 2005) when Microsoft releases Longhorn, which will require DX7 as a minimum and DX9 preferably for basic GUI functionality.

If anything, I question why there hasn't been a serious challenege to Nintendo's POS products in the past, perhaps arising from the PPC market in the form of a specilized product/standard spearheaded by an industry leader like Microsoft.

I don't really consider those integrated components to be 3D chipsets. Many of them have Direct X6 or DX7 functionality in the same sense that the GFFX-5200 has DX9 functionality - unusable functionality. Longhorn, from the videos I've seen, looks as if it will require an extremely low level of performance to work correctly. But still from what I know 3D has not entered most people's computer using experience to any appreciable extent, and that will not change until Longhorn, only 9 years after graphic accelerators hit the market. One exception - 3D screensavers. ;)

As for why Ninendo never was challenged by MS before, I can only guess they were waiting to do so in context of their X-box strategy.
 
MS is not going to enter hadheld gaming market. They said they don't have any interest in that.

Pocket PC is entirely oriented to business market, because of their high production price, and because selling them under their production price would make no sense, as Microsoft doesn't make much money on any software for it, nor is any software for PPC selling all that much (*especially* the gaming software)

Basically, PDAs, no matter how cheap, are not selling that well with general population. Not even close to cell phones, for example.
 
Back
Top