Digital Foundry Article Technical Discussion [2019]

Status
Not open for further replies.
This is a different argument.
Agreed, we are making different points.

Out of This World/Another World on the SNES had long load screens. I think SNES Earthworm Jim had load times about the same as or longer than the Sega CD version. On the Genesis cartridge version, these could be shortened by mashing start. Quake 2 on N64 had load screens, and Perfect Dark had very long black screens between maps.
Street Fighter Alpha 2 on SNES has loading between every match, after the characters appear on screen, even. Quake 1 also had loading screens. I'm sure there are other examples.
 
"Did you see the movie The Matrix? Same interface. Same concept. Starting from next year, you can jack into The Matrix!" -- Ken Kutaragi, regarding the PS2.

I see no lie here...

13868_front.jpg
 
"Did you see the movie The Matrix? Same interface. Same concept. Starting from next year, you can jack into The Matrix!" -- Ken Kutaragi, regarding the PS2.
That wasnt for the PS2.
That was his comment during the GSCube presentation, which also had a tech demo of the Matrix. It was the vision he had about the future of the technology.

"First up, Square demoed at SIGGRAPH yesterday. Attendees saw a quick reel of a female character waking up on a lounger [couch] in a space-craft. The hair has been brute-forced [a CG technique] because so many strands were involved. Everything in the scene had been multi-passed to get the skin and eyes just right. Attendees reckoned Kuturagi-san absolutely loved this."Read that again. Now get a load of this. "Eon showed a concept preview of The Matrix. It was from the scene at the beginning of the movie, the hotel ambush. What we got was a city-top view of an animated figure running and jumping across rooftops, but you couldn't distinguish from the actual film. It was in real-time. It had an interactive camera and they were moving it around."Several other demos were shown, one of which was a flight simulator by Silicon Studio, another was a scene from the movie Antz. 150 fighting characters were seen beating it out, each with 8,000 polys eachWhat we have just outlined is the future of videogaming. Highly placed sources within the industry have been talking for some time about the convergence of movies and games, where players interact with film quality graphics indistinguishable from real life. As reported yesterday, the GScube n a CG development environment powered by the guts of 16 PlayStation 2s n is designed to take is wares to broadband networks.Welcome to the age of e-cinema.How all of this will become possible is unclear. But rest assured that you'll be playing movies in some way, shape or form in the next five years. Just ask Ken Kutaragi. "You can communicate to a new cyber-city," he said in an interview with Newsweek earlier this year. "This will be the ideal home server. Did you see the movie 'The Matrix'? Same interface. Same concept. Starting from next year, you can jack into 'The Matrix'." No-one knew just how right he was�
 
It is quite amazing how our perception changes. Back then those graphics were reported as being indistinguishable from the movies. Now to us they look like PS2 games.
The only section that truly looks impressive and we arent sure if it was indeed real time on a GSCube is the Final Fantasy Spirits Within demo in the above video. That looks way too good for its time to be real time.
 
MGS2 for the first time in 2001, or the E3 half life 2 tech presentation, Doom 3 was quite unreal too. Now it all looks kinda bad :p
 
16 × Emotion Engine CPUs clocked at 294.912 MHz
2 GB of DRDRAM Rambus main memory (16 × 128 MB)
(128 MB was a common memory allocation on devkits vs. the 32 MB on shipping units)
Memory Bus Bandwidth 50.3 GB/s (3.1 GB/s × 16)
Floating Point Performance 97.5 GFLOPS (6.1 GFLOPS × 16)
16 × "Graphics Synthesizer "I-32" Graphics Processors clocked at 147.456 MHz
512 MB of eDRAM Video Memory (16 × 32 MB)
(The "I-32" Graphics Synthesizer was a custom variant that contained 32 MB of eDRAM instead of the typical 4 MB)
eDRAM Bandwidth 755 GB/s (47.2 GB/s × 16)
Pixel Fill Rate 37.7 GB/s (2.36 GB/s × 16)
Maximum Polygon Drawing Rate 1.2 Gpolygons/s (73.7 Mpolygons/s × 16)
 
16 × Emotion Engine CPUs clocked at 294.912 MHz
2 GB of DRDRAM Rambus main memory (16 × 128 MB)
(128 MB was a common memory allocation on devkits vs. the 32 MB on shipping units)
Memory Bus Bandwidth 50.3 GB/s (3.1 GB/s × 16)
Floating Point Performance 97.5 GFLOPS (6.1 GFLOPS × 16)
16 × "Graphics Synthesizer "I-32" Graphics Processors clocked at 147.456 MHz
512 MB of eDRAM Video Memory (16 × 32 MB)
(The "I-32" Graphics Synthesizer was a custom variant that contained 32 MB of eDRAM instead of the typical 4 MB)
eDRAM Bandwidth 755 GB/s (47.2 GB/s × 16)
Pixel Fill Rate 37.7 GB/s (2.36 GB/s × 16)
Maximum Polygon Drawing Rate 1.2 Gpolygons/s (73.7 Mpolygons/s × 16)

Could have been a PS3 :p
 
Would it even have been as expensive?

Good question... on first sight it would be much more expensive but Sony wouldn't have the Cell development costs, which ran into the billions? Also, Nvidia wasn't cheap either i'm sure, so wasn't the XDR ram.
But nah, looking at those specs, seems fantasy almost :p
 
True, but I wonder how viable a pared down version might have been?

I just googled the GSCube, and the first result was the GSBox, a prototype of the GSCube, utilising 4 PS2 blades and a custom chip to merge the 4 outputs.

I've not finished reading through the thread yet, but I can't help but wonder why Sony didn't lean in to this approach for the PS3?
  • It would probably have been better for developers, given that they wouldn't have to learn an entirely new architecture. They'd have had to learn how to synchronise multiple streams, but would that have been more difficult than the Cell?
  • It would have guaranteed backwards compatibility.
  • As you said, it wouldn't have required the billions of R&D the Cell cost. Instead, it would've utilised the R&D already spent on the GSBox/GSCube.
  • Four PS2's, each capable of a 640x480 output would've made for a pretty capable 720p console. Gran Turismo 4 was able to output in 1080i, so 1080p for media output would've presumably been doable.
  • Edited in bullet point: Sony wouldn't have had to turn to Nvidia, who had already proven themselves an unhelpful partner to Microsoft.
The paltry amount of memory would've been a problem. 16MB VRAM and 128MB main memory just wouldn't cut it. How expensive would 32MB of EDRAM and 64MB of main memory have been per "PS2 blade"? A total of 128MB of ultra fast VRAM and 256MB main memory would've been rather impressive.

I need to go and have a look at the PS3's specs to properly compare it to this hypothetical PS2Cube.
 
Last edited:
Hmm dunno, but even the 4x ps2 blades spec seem rather impressive, perhaps more so then the PS3's... Seeing how devs really harnessed that PS2, all their first party studios where kinda fond of the PS2, even though it was so hard to program for, it also enabled them to do things the other platforms of the time couldn't.
The Cell/PS3 wasn't that easy to learn either.

It would have guaranteed backwards compatibility.

The inclusion of a whole PS2 in the early 60gb PS3's was rather expensive one might think.

Four PS2's, each capable of a 640x480 output would've made for a pretty capable 720p console. Gran Turismo 4 was able to output in 1080i, so 1080p for media output would've presumably been doable.

Since most games where close to 720p that generation, it wouldn't be too far off with the PS2cube idea :p

Sony would not have to use the GScube either, the idea of the GScube, but a more simple structure, like a heavily improved PS2, more raw power. The PS2 was very flexible, a much more powerfull PS2 wouldn't have that much problems keeping up with the 360/ps3 games we got.
Interesting thought at the least, somewhere it must have been on the table at Sony, but they didn't go for it, i mean the GScube and all...

This potentional PS2cube wouldn't need as much memory as the PS3's 2x 256mb either i think. The PS2 had 32mb against the xbox's 64 but they werent a whole gen apart either, they were clearly playing in the same generation of power.
Not that the PS3 was bad anyway, uncharted for the time was something else :)
 

This is the guy that made the PS2 Duck demo!
Kutaragi wanted him to work on the graphics that dont use polygons. And he managed to do it! And the rubber duck was the result!
Disappointingly enough this technology was never used in actual games, but it was probably a gate to the PS2's untapped powers.
It makes me wonder what we would have got if his work was used further.
The water effects of the tech demo are beyond anything that was done in real time graphics and it probably challenges present implementations.
 
What did the demo use instead of polygons?
To be honest I have no idea, I cant imagine anything without polygons. He describes his efforts to meet Kutaragi's demands for polygon-less visuals and he says something about b-splined curved surfaces patches that tessellated into triangles whatever that is in the tweets. Maybe he means the visuals were constructed with curves that generated triangles rather than being constructed with polygons from the get go. I d like to hear an explanation.
They manage to cram in a form of tessellation, normal and refractive water too,
 
... he says something about b-splined curved surfaces patches that tessellated into triangles whatever that is in the tweets.
That sounds about right. No visible polygons in the rough YT videos I've seen, and it's a tech Sony both talked about (hardware can handle 16m poly/s Bézier patches) and even implemented in hardware on PSP.
 

This is the guy that made the PS2 Duck demo!
Kutaragi wanted him to work on the graphics that dont use polygons. And he managed to do it! And the rubber duck was the result!
Disappointingly enough this technology was never used in actual games, but it was probably a gate to the PS2's untapped powers.
It makes me wonder what we would have got if his work was used further.
The water effects of the tech demo are beyond anything that was done in real time graphics and it probably challenges present implementations.
I didn't know someone had made a demo using bézier surfaces (if that's what he used) of PS2 !

I still remember vividly ;-) the video-game magazines hyping like crazy the nextgen bézier surface feature of PS2 ! But it was all marketing PR as we never saw that in any games. We were also drowned with tons of impressive specs, that were never actually used in 95% of games (very few of them did, like MGS2, yeah that shit was nextgen, but it was a rarity on PS2).

On the other hand, Dreamcast had already nextgen graphics (like 1, 2 years before PS2 games) but few knew about those games and how great those looked (Sonic, Soul Calibur, Shenmue, Toy Commander ...).

Poor Dreamcast (:love:). Killed by deceitful marketing from Sony (that all journalists were happily regurgitating directly from Sony's mouth), bad marketing from Sega and obviously EA :devilish:.
 
Status
Not open for further replies.
Back
Top