Comparison of 5th generation ("32/64-bit") game console hardware

[kool kitty89]

Now I am confused, it is the draw in point that is blended with the background, so there is no 3D behind that because it is all drawing in at that point. In Radiant Emerald no other part of the track ever appears through the transparent floors if that is what you mean.

OK, so definitely a VDP2 effect.

Looking at the VDP2 docs, any of the scroll planes can be either cell mode, or bitmap mode. The limit on bitmaps is what fits in memory. Looking closer at the VDP2 docs, bitmap planes CAN be rotated and scaled, but they have to be in a lower resolution mode to do so. Only scrolling allows bitmaps to be up to 1024x512, while rotation/scaling limits them to 512x512.

OK, and does that apply to both the VDP1 framebuffer and additional VDP2 bitmap planes?

If it can scale in low-res mode, I wonder if that's how the burning force light effects are done. (ie software rendering to a low-res translucent bitmap scaled up 2x -doing it low res on the VDP1 end wouldn't make much sense since you'd still need to draw the same number of pixels either way, but doing that on a separate VDP2 bitmap would make sense -less CPU overhead and less RAM needed)





And another more general note:
For blitter rendering, using a painter's algorithm for rendering (be it 2D or 3D) would be particulary attractive for chunky-pixel renderers rendering few (or especially 1) pixel per write and especially without a z-buffer or similar pixel priority information (for 2D games, much lower res depth could be useful where a 3D z-buffer would be worthless).

It would seem like the Saturn would be particularly favorable for this with VDP1 rendering 1 pixel at a time, and using software Z-Sorting (ie on a per-sprite/primitive or per-model basis) since the Z-sorting could be used to directly organize objects to be drawn with a painter's algorithm. (so you'd just be doing writes to the framebuffer, no read-modify-writes . . . unless you used translucency effects)

On systems rendering phrases of pixels at a time, it would depend more on hardware features to cater as heavily to painter's algorithm type rendering. (a 2-pixel wide bus -like 3DO or PSX- should still get away with only doing writes, but would need logic to support that -either writing a 32-bit word if both destination pixels are opaque, or a single 16-bit write if 1 pixel is transparent, and a 4-pixel wide bus would be more complex still -using single writes in cases of 1/2/4 consecutive opaque pixels, but a read-modify-write in the case of mixed opaque/transparent pixels -again, requiring logic to support that, otherwise requiring a read-modify-write for all rendering)

[Chilly Willy]

OK, and does that apply to both the VDP1 framebuffer and additional VDP2 bitmap planes?

Not to the VDP1 framebuffer, BUT - like I mentioned in another post, VDP1 can be set to transfer the framebuffer in such a way as to do scrolling/scaling/rotation. VDP2 only scrolls its own layers - the VDP1 layer is handled by VDP1 (other than VDP2 compositing it with its own output).

[Team Andromeda]

I am playing the PAL version on my US system. Now that you mention it, I think I have heard rumors that the Japanese version of 2097 runs faster

No there is hardly any improvement in the Japanese version its still runs at 30 fps . I've seen some say there is less pop up in the JP version, but I can't see it myself. So I really do not know how're you getting 20 fps . I'm sure even the Saturn Power demo code disc is running at 30 fps too

Well, I have been being honest this entire time, including the 32X discussion. I am only interested in the facts, and I try to only contradict things when they don't add up. If I say something is wrong,

Me too and these debates are great . But I have never seen the Saturn handle a 3D transparency over a 3D polygon object or character that is anywhere near comparable to a PS one . It was clear from day one on both systems launches of the PS and Saturn (JP) that the Saturn had a issues with such effects while the PS had no such issues - The Shadows in VF were using the mesh like effect while Ellis in PS Toshinden had a lovely translucent clothing (ok I know the shadows weren't transparent but still ) Then a few months in you see the Saturn have no trouble handling brilliant transparent effect on 2D games in like Astal, Darius Gaiden (which pissed on the PS so much its untrue) , Layer Section , but in games like Panzer Dragoon, Gran Chaser you saw the Saturn having issues with transparent effects in 3D, while it was getting pretty standard in almost all PS games

So very in early in it was getting clear to me that the Saturn was a powerhouse for 2D and had great 3D and texture mapping, but the PS had better lighting effects and transparent effects

[kool kitty89]

Not to the VDP1 framebuffer, BUT - like I mentioned in another post, VDP1 can be set to transfer the framebuffer in such a way as to do scrolling/scaling/rotation. VDP2 only scrolls its own layers - the VDP1 layer is handled by VDP1 (other than VDP2 compositing it with its own output).

Huh, so VDP1 handles its own framebuffer scanning too? (all that's missing is a video DAC?)

So, hypothetically, if VDP1 was used by itelf, all it would need was an external RAMDAC (with 256 entry CLUT for 8bpp mode -so basically an off the shelf VGA/SVGA 18 or 24-bit RAMDAC) and it would be a fully functional, standlone graphics chip?

[Crazyace]

VDP2 reads the framebuffer from VDP1 and mixes it.

[sheath]

No there is hardly any improvement in the Japanese version its still runs at 30 fps . I've seen some say there is less pop up in the JP version, but I can't see it myself. So I really do not know how're you getting 20 fps . I'm sure even the Saturn Power demo code disc is running at 30 fps too

All I can tell you is that it is, download any of the gameplay videos for Wipeout 2097 and use Windows Media Player Classic or Virtual Dub to frame advance. In these games I would frame advance until the race clock on the HUD changed to the next second and then count unique frames until the next second. You will see nearly consistent two new frames, one still, two new, one still.

Me too and these debates are great . But I have never seen the Saturn handle a 3D transparency over a 3D polygon object or character that is anywhere near comparable to a PS one . It was clear from day one on both systems launches of the PS and Saturn (JP) that the Saturn had a issues with such effects while the PS had no such issues - The Shadows in VF were using the mesh like effect while Ellis in PS Toshinden had a lovely translucent clothing (ok I know the shadows weren't transparent but still ) Then a few months in you see the Saturn have no trouble handling brilliant transparent effect on 2D games in like Astal, Darius Gaiden (which pissed on the PS so much its untrue) , Layer Section , but in games like Panzer Dragoon, Gran Chaser you saw the Saturn having issues with transparent effects in 3D, while it was getting pretty standard in almost all PS games

So very in early in it was getting clear to me that the Saturn was a powerhouse for 2D and had great 3D and texture mapping, but the PS had better lighting effects and transparent effects

Early games were reputedly not even using the slave SH2 and many multiplatform games use the VDP2 minimally if at all. Even Konami's port of Symphony of the Night uses dithering for the effects rather than optimizing it for what the Saturn could actually do better than the Playstation in 2D. Simply comparing Die Hard Trilogy to Virtua Cop 1+2 on the Saturn shows how little of the Saturn hardware could handle a similar PS1 port.

I really wish there were more later generation multiplatform titles on the Saturn, that would have made a much better comparison.

[Team Andromeda]

All I can tell you is that it is, download any of the gameplay videos for Wipeout 2097 and use Windows Media Player Classic or Virtual Dub to frame advance. In these games I would frame advance until the race clock on the HUD changed to the next second and then count unique frames until the next second. You will see nearly consistent two new frames, one still, two new, one still.

Please that is fatal . Now we listen to you VF on the 32X is running faster than the Saturn. I don't know what set up you have or when you're getting these vids from , but I can tell you Wipeout 2097 is running closer to 30 fps than it is 20 fps and that VF is 100% running at 30 fps

Early games were reputedly not even using the slave SH2 and many multiplatform games use the VDP2 minimally if at all

The likes of VF remix and Panzer Dragon was making use of the VDP II not a hint of multi platform and Darius Gadien looks bloody awesome even though its a port

Even Konami's port of Symphony of the Night uses dithering for the effects rather than optimizing it for what the Saturn could actually do better than the Playstation in 2D

The game wasn't just 2D it used polygons too (much like Swagaman) and like Core Swagman team said you're run in to trouble when trying to do PS like effects even in a 2D game on the Saturn.

[sheath]

Please that is fatal . Now we listen to you VF on the 32X is running faster than the Saturn. I don't know what set up you have or when you're getting these vids from , but I can tell you Wipeout 2097 is running closer to 30 fps than it is 20 fps and that VF is 100% running at 30 fps

Framerate is one of those things that is absolutely measurable. There is no debate, Wipeout 2097 is 20FPS. Maybe your Japanese Saturn makes a difference, but that would be extremely odd since it is NTSC also.

Also, I did not say that VF 32X has a faster framerate than Saturn VF, I just said it has none of the graphical glitches of Saturn VF. I think it was Trekkies who was saying that Saturn VF was 20FPS or some such. I haven't videoed either version to be able to count the frames on them yet.

The likes of VF remix and Panzer Dragon was making use of the VDP II not a hint of multi platform and Darius Gadien looks bloody awesome even though its a port

Yeah, those are Sega titles, I should have been more clear that I was talking about multi-platform titles.

The game wasn't just 2D it used polygons too (much like Swagaman) and like Core Swagman team said you're run in to trouble when trying to do PS like effects even in a 2D game on the Saturn.

Right, they just ported the polygonal code over to the Saturn, sloppily, and didn't even optimize for load times so the load scenes don't even load until you walk through the door. It was a sloppy port.

[Black_Tiger]

The game wasn't just 2D it used polygons too (much like Swagaman) and like Core Swagman team said you're run in to trouble when trying to do PS like effects even in a 2D game on the Saturn.

You run into even more trouble when trying to do Saturn like effects on Playstation, even in a 3D game.

[TrekkiesUnite118]

Here TA:



Both are recorded from real PAL consoles running authentic PAL versions of the games. You can see the PS1 version is significantly smoother than the Saturn version. The PS1 version is running at 30fps, Saturn is obviously running lower than that. If you look at the one for the first Wipeout, the frame rates look identical. And if we all agree that the original Saturn Wipeout runs at 20fps, then I think we can agree from this evidence that Wipeout 2097 runs at 20 fps as well.

[evilevoix]

It looks much slowr on teh Saturn

[sheath]

I have never liked that video, it makes the Saturn version look worse than it really does. This keeps coming up, I might have to go ahead and redo my comparison video.

[Crazyace]

You will see nearly consistent two new frames, one still, two new, one still.

2 new , 1 still would be 40fps by my reckoning
( 1 new, 2 still would be 20fps if that's what you mean? )

Assuming you are looking at a 60fps capture - maybe if the source video was 30fps the 2new/1still would imply 20fps.

[sheath]

2 new , 1 still would be 40fps by my reckoning
( 1 new, 2 still would be 20fps if that's what you mean? )

Assuming you are looking at a 60fps capture - maybe if the source video was 30fps the 2new/1still would imply 20fps.

It's 30FPS video, 29.97 actually.

[Chilly Willy]

Huh, so VDP1 handles its own framebuffer scanning too? (all that's missing is a video DAC?)

So, hypothetically, if VDP1 was used by itelf, all it would need was an external RAMDAC (with 256 entry CLUT for 8bpp mode -so basically an off the shelf VGA/SVGA 18 or 24-bit RAMDAC) and it would be a fully functional, standlone graphics chip?

No. It's one of those "weird" things about the Saturn. VDP2 asks VDP1 for the next pixel in the frame buffer as it composites the levels to generate the final video out. VDP1 decides WHERE that next pixel comes from. It can actually be the next pixel, or it could be offset some amount horizontally and vertically. By sending the same pixel repeatedly, you get expansion; by skipping pixels, you get shrinking; by changing the vertical offset at the same time as the horizontal, you get rotation.

You would THINK that since they have the hardware in VDP2 to scale/rotate/scroll bitmaps that they would use that hardware on the frame buffer as well, but they don't. It's almost duplicated in VDP1's frame buffer interface... yet another reason to believe the tale that VDP1 was either an afterthought, or radically beefed up in response to the announced PSX specs. If it had been MEANT to be the way it was from the beginning, you would think it would have been just another bitmap, just in a different memory bank than the vram. The way the video works on the Saturn SCREAMS "last minute hack" to anyone familiar with electronics.