Why would nintendo release an underpowered console a year after the release of 360?

[Shifty Geezer]

I think most 40/50+'s will never be hooked bigtime into gaming because by that point in one's life, you realize there's things worth DOING, sharing time with family and grandkids and appreciating life, rather than sitting in front of TV screens hours on end virtually killing virtual things

 

[Qroach]

Raw FLOPS (translates to vertex, 32-bit only) output:
Xbox (twin-vertex shaders), GC (fixed hardware T&L), PS2 (Emotion Engine)
Xbox(10 flops * 2 * 233 mhz) = 4.660 GFLOPS (32-bit, programmable)
GC(w/o lighting) = 3.726 GFLOPS/ (32-bit ops, fixed)
GC(w lighting) = 9.4 GFLOPS (32-bit & 20-bit ops, fixed)
PS2 (VU 1) = 3.08 GFLOPs (32-bit, fully programmable)
PS2 (VU1/VU0/CPU FP) = 6.2 GFLOPS (32-bit, fully programmable)

http://www.segatech.com/gamecube/overview/

But this may be inaccurate, excluding non-programmable XGPU hardware. Lets try total GFLOPS (minus pixel shaders, not including CPU for Xbox and GC):

Ranking (raw, peak, vertex-GFLOPs)
Xbox (21.6 GFLOPS - 2.932 FLOPS (CPU) - 7.456 GFLOPS (pixel shaders, 24-bit)) = 11.2 GFLOPS (32-bit and other, programmable & non-programmable)
Gamecube = 9.4 GFLOPS (32-bit and other, non-programmable)
PS2 = 6.2 GFLOPS (32-bit, fully programmable)

The 21.6 GFLOPS I retrieved from a book (Opening the Xbox), which is Xbox's total system power. For the pixel shaders (3 vector, 1 scalar, * 2 madd * 4 shaders). The rest should be the vertex shaders and related hardware. These comparisons are without any nifty optimizations of course (early z-checks). With CPU (lighting, animation), Gamecube is at 11.3 GFLOPS, making Xbox and GC almost exactly equal in polygon performance, without the XCPU (which isn't contributing to T&L). I'd say GC is better though, because of the aforementioned, probably existing, early z-check, and the fast z-clear (xbox might have fast z-clear as well).

So the ranking (polygon output):
Gamecube (11)
Xbox (11)
PS2 (6)

Xbox and Gamecube are tied, but when Xbox is overloaded with shader effects, like you said, Gamecube wins. With high complexity (requiring lots of z-culling), Gamecube wins. Actual PS2 optimization is limited to VU1, which is at 3.08 GFLOPS. And the Xbox DirectX configuration can slow things down quite a bit, unless push-buffers are used. And I've seen PS2 use good tesselation algorithms. Early Z-checks on the PS2 is hard, but continuous LOD isn't, as well as other software optimization. GC and PS2 will not have the texture resolution on its polygons that Xbox will have.

Ranking (polygons, ingame performance, out of 10):
Gamecube (10)
Xbox (6/7, with pushbuffers)
PS2 (3/4, with VU1 only, assuming good optimization)

Where would N64 be though?
Reality Co-Processor - 4 32-bit ops * 2 (madd) * 62.5 mhz = 500 MFLOPS

Right on the money, as Silicon Graphics told the press that N64's coprocessor could do 500 MIPS.

Ranking (polygons, ingame performance (moderately complex*), out of 10):
Gamecube (10)
Xbox (6/7, with pushbuffers, moderate pixel shader utilization)
PS2 (3/4, with VU1 only, assuming good optimization)
N64 (.5,

Ranking (w/N64, polygons, ingame performance (exception being N64, with no known performance inhibitors; moderately complex), out of 10):

Gamecube (10)
Xbox (6/7, with pushbuffers, moderate pixel shader utilization)
PS2 (3/4, with VU1 only, assuming good optimization - partial VU0 and CPU optimization)
Dreamcast (1.4 SH4 GFLOP capacity - 1.4)
N64 (.5)
PSX/Saturn (<.5)

*complex - many layers of interaction in the 3D scene

I'm not sure about Saturn and PSX...I can't find any FLOP performance for them. If you would like to learn more about these systems, you are bound to find tons of information at any major search engine just using the names of the GPU and CPU of the systems.

This information is completely wrong! Seriously f'ing wrong. my god... Both xbox and PS2 could destroy the gamecube in T&L polygon performance in game. PS2 could probably double what gamecube was capable of and Xbox could possibly triple it.

 

[function]

Strange, you like falling back on those benchmarks when their software doesn't hold up. I take it you haven't seen their top tier franchise Madden'06 comparatively on the GC? Something about those textures Powderkeg.........

"Falling back on"? I'm not on the back foot here.

I've already gone over this in another Nintendo thread (which I'm sure you've read). What you think of EA's games, of how good they are and how much work they've put in, or what quality of assets they've seen fit to use on a particular game has no bearing on their ability to benchmark the hardware.

Not only that, but developers on this board have made reference to the relative performance of systems before. And surprise surprise, they're broadly in line with what EA found. You can find comments about CPU power, fill rate and probably other stuff too. Do a search. It's fun.

I remember talking to you in the past about Revolution, and saying that it wouldn't fit to Nintendo's MO to take on MS and Sony head to head in terms of power. You seemed rather opposed to this line of thinking, but given Miyamoto's comments quoted in this thread and that fact it's certainly true with the cost-effective GC, are you more comfortable with this idea now?

 

[Megadrive1988]

interesting about Xbox1's floating point performance, more down to earth. never seen those numbers before.

 

[Megadrive1988]

I think they could release a console with a R300 at the core and a souped up G3 and we'd see amazing games for it. Hell, you all know that PC graphics chips don't get taken advantage of very well. At non HD resolutions even an "old" 8-pipeline GPU could make some serious imagery, especially on a closed platform..And it would be crazy cheap to make compared to Xbox360 and PS3.

I hope N wipes the floor with Sony and MS If they can hit a unique pricepoint that consumers see as separate from Sony and MS they can make themselves their own market that their competitors can't touch pricewise.

agreed. the R300 blows current console graphics chips (GS, Flipper, NV2A) out of the water, and I expect Hollywood to be significantly better than R300, and somewhat more powerful.

Gamecube released in fall 2001 with Flipper having 4 pixel pipelines, the equivalent of 1 texture unit per pipe, 1 fixed T&L unit @ 162 MHz which at first glance looks roughly on par with an overclocked GeForce256 (GF1 aka NV10) from 2 years earlier, fall 1999 but obviously Flipper blows GeForce1 out of the water, in most areas.

I expect Hollywood to look rather weak if compared to ATI's PC GPUs of 2006 like R580 and R600, but Hollywood will most likely have some very very strong custom features that no other ATI GPU has..that PC GPUs do not have, and that X360 Xenos does not have
(i.e. full NURBS capability as rumored recently). we'll see some totally amazing games on Revolution that are as good as Xbox360 and PS3 in visuals overall. each of the nextgen consoles will have a few advantages over the other two, like this gen, even though the differences in overall graphics will be smaller. (all new consoles have Shader Model 3.0+)


with that said, I would love to see a new upscale 'elite' console emerge, like NEO-GEO, using 2 ATI or 2 Nvidia GPUs, 512-bit bus architecture, and 2 seperate multi core CPUs.
this console would be to X360, Revolution and PS3 what the NEO-GEO was to the TG16, Genesis and SNES in the early 1990s. This elite console woouldn't be a generation beyond X360,Rev,PS3, but of the same generation, only several times more poweful, like NEO-GEO was. this elite system also doubles as a computer. cost? $900-$1000 which is actually less than NEO-GEO's $650 pricetag in 1990 in todays money. software costs the same as PS3,X360 games: $60 which is a helluva lot less than the $200 to $300 for NEO-GEO games. the elite console gets some highend sensory controllers that go beyond whatever Revolution has. the elite console launches in 2006 but lasts until at least 2016, well after the successors to PS3,Rev,X360 arrive.... ok wake me from my dream

 

[Epik]

interesting about Xbox1's floating point performance, more down to earth. never seen those numbers before.

the Xbox was powerful on paper but the bottlenecks in its design robbed it of it theoretical peak performance.Same with Sony.

The gentleman up above can't seem to accept the truth.

 

[skilzygw]

MegaDrive 1998 wrote
"with that said, I would love to see a new upscale 'elite' console emerge, like NEO-GEO, using 2 ATI or 2 Nvidia GPUs, 512-bit bus architecture, and 2 seperate multi core CPUs.
this console would be to X360, Revolution and PS3 what the NEO-GEO was to the TG16, Genesis and SNES in the early 1990s. This elite console woouldn't be a generation beyond X360,Rev,PS3, but of the same generation, only several times more poweful, like NEO-GEO was. this elite system also doubles as a computer. cost? $900-$1000 which is actually less than NEO-GEO's $650 pricetag in 1990 in todays money. software costs the same as PS3,X360 games: $60 which is a helluva lot less than the $200 to $300 for NEO-GEO games. the elite console gets some highend sensory controllers that go beyond whatever Revolution has. the elite console launches in 2006 but lasts until at least 2016, well after the successors to PS3,Rev,X360 arrive.... ok wake me from my dream "



You already have that except it's called a PC. Just look at the dream machine from maximumpc magazine. Of course that system is more than the $1000 you said.

 

[skilzygw]

MegaDrive 1998 wrote
"with that said, I would love to see a new upscale 'elite' console emerge, like NEO-GEO, using 2 ATI or 2 Nvidia GPUs, 512-bit bus architecture, and 2 seperate multi core CPUs.
this console would be to X360, Revolution and PS3 what the NEO-GEO was to the TG16, Genesis and SNES in the early 1990s. This elite console woouldn't be a generation beyond X360,Rev,PS3, but of the same generation, only several times more poweful, like NEO-GEO was. this elite system also doubles as a computer. cost? $900-$1000 which is actually less than NEO-GEO's $650 pricetag in 1990 in todays money. software costs the same as PS3,X360 games: $60 which is a helluva lot less than the $200 to $300 for NEO-GEO games. the elite console gets some highend sensory controllers that go beyond whatever Revolution has. the elite console launches in 2006 but lasts until at least 2016, well after the successors to PS3,Rev,X360 arrive.... ok wake me from my dream "

You already have that except it's called a PC. Just look at the dream machine from maximumpc magazine. Of course that system is more than the $1000 you said.

 

[function]

with that said, I would love to see a new upscale 'elite' console emerge, like NEO-GEO

Would be nice, but I image that the cost of creating games that took advantage of it (unless they were short, arcadey style things) couldn't be justified for it's small user base (which would inevitably be relatively small).

I image that Sega's Lindbergh system is quite scaleable and could meet this need, but you'd be quite limited in terms of the software you'd see. And as software drives hardware sales ...

 

[Ooh-videogames]

This information is completely wrong! Seriously f'ing wrong. my god... Both xbox and PS2 could destroy the gamecube in T&L polygon performance in game. PS2 could probably double what gamecube was capable of and Xbox could possibly triple it.

But PS2 doesn't have a built in T&L. Also why did Capcom have to reduce polygon counts for RE4. With a year to port the title compared to most ports, which come in under six months. You would assume that they would have been able to match polygon for polygon and light for light.

 

[mckmas8808]

But PS2 doesn't have a built in T&L. Also why did Capcom have to reduce polygon counts for RE4. With a year to port the title compared to most ports, which come in under six months. You would assume that they would have been able to match polygon for polygon and light for light.

It might have something to do with the bodies that stay on the ground after you kill them. Or maybe it was the 16:9 mode the PS2 has that changed everything.

 

[DuckThor Evil]

Hmm.. It seems that many Nintendo fan is in serious denial here, well whatever makes your world seem brighter.

 

[Ooh-videogames]

It might have something to do with the bodies that stay on the ground after you kill them. Or maybe it was the 16:9 mode the PS2 has that changed everything.

Well you must have not played the Cube verison, because in the earliest part of the game which is the village. The villagers didn't dissolve. The only reason the PS2 version has 16:9 is because many gamers complained about lack of it in the Cube version. Since Kobayashi himself said that the borders in the Cube version were implemented for gameplay reasons and had nothing to do with the hardware.

 

[Ooh-videogames]

Hmm.. It seems that many Nintendo fan is in serious denial here, well whatever makes your world seem brighter.

I don't know, maybe its the better textures or higher polygon characters in most exclusive titles compared to those created exclusively on the PS2.

Or maybe you commenting on the Revolution, if so then ignore the above comment. I'll take a wait and see approach to whatever the specs are for the Revolution.

 

[Powderkeg]

No that would be you .

Not hardly.

That is not true at all .

Yes it is.

Nintendo is launching around june from latest rumors going around the internet . That means that 65nm might be avalible. If not 90nm will have had almost another year to mature over when ms started production . Which means better yields and cheaper parts .

If Nintendo can do 65nm, then MS can do 65nm. Costs are the same for both.
If Nintendo has to use 90nm, then MS also has to use 90nm. One again, the cost is the same for both.

Yeilds will be similar, assuming similar complexity of the processors which would be required to achieve the same power.

End result. Nintendo has to pay the same amount that MS does.

Nintendo can also design a better system with a cheaper lay out saving more .

Using a more expensive optical drive, more expensive RAM, and parts that cost the same as MS is paying? For some strange reason, I doubt that.

They are also targeting a diffrent res which is lower than the xbox 360s . This will allow them to do 3x the work per pixel with the same hardware power that the xbox 360 has .

Yeah, too bad they won't have the same power as MS if they are going to make a cheaper system.

Your not serious are you ?

First to have this debate you need to price out how much the tray loading drive costs ms in orders of 1 million .

Then you need to price out how much the slot loading drive costs nintendo per million .

I think what you will find is that costs may only be 1 or 2$ more when u get to the millions of units .

1 drive or 1 million drives is irrelevent. You are still looking at slot drives costing 4-5 times more than tray.

As for the chip liscensing . Nintendo may have gotten a better deal. Esp from ati. They have had a long relation ship with art-x now part of ati and it seems likely they will be going with a modified desktop part which should be cheaper to license as it is not custom and ati doesn't have to make back its investment solely on the license fee .

It may also be the same with ibm .

Funny, I see the opposite.

MS can use their PC support and DirectX 10/WGF2.0 specifications as leverage with ATI, and for ATI the PC market is MUCH more important than Nintendo's relatively tiny userbase.

Who does ATI give a better deal to:

Nintendo, who might sell 20 million GPU's total if they are lucky, or MS, who is not only expected to sell more GPU's directly, but can also tailor DX10/WGF2.0 for ATI's hardware, giving ATI a performance boost helping them to sell hundreds of millions of PC GPU's?

Tough choice, eh?

Once again nintendo will be using much more mature lines for production. They might even have acess to 65nm for some parts .

This is laughable.

How do you figure Nintendo's brand new production line, possibly using a brand new fabrication process will be more mature than MS's year old and fully developed production?

The TRUTH is, MS will have access to the same manufacturing processes as Nintendo.

There are a slew of factors that one must add together before claiming anything . If nintendo wants to offer a system on the power lvl of the xbox 360 8 or 9 months after the xbox 360 for less money than it can .

Esp if it isn't looking to do all the things the xbox 360 is doing

So far you haven't given a compelling arguement. At most, you are basing everything on the hypocritical notion that somehow Nintendo is going to have 65nm parts when no one else can make them (If someone else could, MS can use them to manufacture their parts too) and that Nitnendo's brand new production line will be more mature than MS's year old production line that is already operating at full capacity.

The truth is that both Nitnendo and MS will be using the same companies to manufacture their chips, so what one has access to, the other will also, and the price for both will be very similar.

You most definitely haven't explained how they could save $50-$100 per unit in production costs. $5-$10 maybe, but no where near $100 per unit in savings.

And even if you saved $100 on production costs, Nintendo would still lose just as much money as MS if they priced their system $100 lower.

 

[Shifty Geezer]

So far you haven't given a compelling arguement. At most, you are basing everything on the hypocritical notion that somehow Nintendo is going to have 65nm parts when no one else can make them (If someone else could, MS can use them to manufacture their parts too) and that Nitnendo's brand new production line will be more mature than MS's year old production line that is already operating at full capacity.

I think the process argument is more a consideration of design time decisions. XB360 was designed around a 90 nm process. It's chips are specc'd to that, with 200 million transistors ina package that is viable to manufacture at 90nm. The result is components that produce XB360's level of performance. Now a chip designed for 65nm could factor in shrinks and higher clock-speeds and 'stuff' that a process shrink for the 90nm part won't attain. eg. If the maximum viable size for a 90nm part is say 200 million trannies, the maximum size for 65nm might be say 260 million trannies. A chip designed for 90nm can get smaller with the process shrink, and have better yields than at 90nm, but won't attain the same maximum performance as a part designed for 65nm. Waiting for a process shrink potentially gives Nintendo more 'legroom' to balance out cost, performance and temperature.

That said I don't think the bonuses of a process shrink can produce faster AND smaller AND cooler AND cheaper. The fab's will be new, 'immature', presumably need to be fairly profitable to cover development costs and such. And I don't believe Revolution is being designed for a 65nm process either. That'd be one hell of a gamble on Nintendo's part to trust that the fabrication technology will attain a level sufficient to support their designs.

When PS2 received a process shrink remodel Sony squeezed it into a case similar to Revolution's in proportions, and that was with a drop from 250nm thorough 180, 130, and eventually 90nm with component consolidation. Assuming the same shrinkability per process advance, a couple of process shrinks at least would be needed to get XB360 down to Revolution sizes. I can't see a system starting out at 65nm managing to get that much size reduction and yet retain the same performance. If so, why did Sony wait 3 process shrinks before remodelling PS2?

 

[Powderkeg]

I think the process argument is more a consideration of design time decisions. XB360 was designed around a 90 nm process. It's chips are specc'd to that, with 200 million transistors ina package that is viable to manufacture at 90nm. The result is components that produce XB360's level of performance. Now a chip designed for 65nm could factor in shrinks and higher clock-speeds and 'stuff' that a process shrink for the 90nm part won't attain. eg. If the maximum viable size for a 90nm part is say 200 million trannies, the maximum size for 65nm might be say 260 million trannies. A chip designed for 90nm can get smaller with the process shrink, and have better yields than at 90nm, but won't attain the same maximum performance as a part designed for 65nm. Waiting for a process shrink potentially gives Nintendo more 'legroom' to balance out cost, performance and temperature.

That said I don't think the bonuses of a process shrink can produce faster AND smaller AND cooler AND cheaper. The fab's will be new, 'immature', presumably need to be fairly profitable to cover development costs and such. And I don't believe Revolution is being designed for a 65nm process either. That'd be one hell of a gamble on Nintendo's part to trust that the fabrication technology will attain a level sufficient to support their designs.

I can agree with that assessment completely.

When PS2 received a process shrink remodel Sony squeezed it into a case similar to Revolution's in proportions, and that was with a drop from 250nm thorough 180, 130, and eventually 90nm with component consolidation. Assuming the same shrinkability per process advance, a couple of process shrinks at least would be needed to get XB360 down to Revolution sizes. I can't see a system starting out at 65nm managing to get that much size reduction and yet retain the same performance. If so, why did Sony wait 3 process shrinks before remodelling PS2?

I don't think the Revolution will ever get smaller than it already is, if for no other reason than the fact that a 90nm chip is so small to begin with that any further shrinkage wouldn't make any noticable difference. You're talking about starting with a chip that's smaller than a fingernail. Even if you made it half that size, have you really saved enough space to make any changes?

 

[swaaye]

agreed. the R300 blows current console graphics chips (GS, Flipper, NV2A) out of the water, and I expect Hollywood to be significantly better than R300, and somewhat more powerful.

Yeah I really am starting to think that it would be INCREDIBLY wise for N to target a new low price. MS and Sony are pushing pricing boundaries with their systems and it's opening up an opportunity that hasn't existed before. If N could hit a ~$200 price they would undercut their competitors and probably be able to still offer games that look the same to most people. $200 less will sell the system. Parents will jump on that over a $400 Xbox360 or PS3. And many parents still remember Nintendo over Xbox or PS. N has a lot of pull with many people. Many parents now grew up with Nintendo systems.

And when I said "R300" I was really trying to convey a level of performance, not a specific chip. I think a chip of R3x0's level could produce more than adequate performance for a 640x480 console. Without a doubt. And considering it would certainly be customized for a console it could be absolutely enough. MS and Sony think everyone needs 1280x720. That is simply not realistic IMO. Kids aren't going to pick up $800+ TVs to go with their $400+ console packages (system, access, games)! And very few people already have HDTVs. I personally only know of 1 of my friends who has one. And he's not even a big console gamer lol. None of my console gaming friends have HDTVs.

Gamecube, N64, and SNES all have engineering in them that screams cost optimization. Sega and Sony didn't understand this, and MS and Sony still don't. Gamecube was VERY cheap to build. N64 was FAR more simple than Saturn and PS1. SNES had a crippled CPU and carts with DSPs to make up for it. I think it's guaranteed that N will make another "cheap", low complexity console. It has ALWAYS worked for them in the past.

It is exciting to think of what could potentially happen if N hit that $200 non-HDTV market! I think trying to be a 3rd high end player would be suicide for them honestly.

 

[Qroach]

But PS2 doesn't have a built in T&L. Also why did Capcom have to reduce polygon counts for RE4. With a year to port the title compared to most ports, which come in under six months. You would assume that they would have been able to match polygon for polygon and light for light.

PS2 hardware doesn't perform transformation and lighting on polygons? what kind of magic do you think is under the hood??

I didnt say PS2 had a T&L unit, I said it can transform and light more polygons faster than the gamecube. really what's under the hood hardware wise doesn't matter in this argument about which could achieve higher real world numbers. Xbox could double what the game cube was capable of PS2 achieved much higher polygon performance then the gamecube. honestly this is a waste of time arguing about over and over again. This argument was finished years ago, there no point taking one develoeprs comments and trying to support your opinion. just look at the data EA released and it says it all. Just talk to the developers on this forum and "please" listen to what they have said.

 

[Ooh-videogames]

PS2 hardware doesn't perform transformation and lighting on polygons? what kind of magic do you think is under the hood??

I didnt say PS2 had a T&L unit, I said it can transform and light more polygons faster than the gamecube. really what's under the hood hardware wise doesn't matter in this argument about which could achieve higher real world numbers. Xbox could double what the game cube was capable of PS2 achieved much higher polygon performance then the gamecube. honestly this is a waste of time arguing about over and over again. This argument was finished years ago, there no point taking one develoeprs comments and trying to support your opinion. just look at the data EA released and it says it all. Just talk to the developers on this forum and "please" listen to what they have said.

I have no doubts about the Xbox capability, I have yet see the proof of what you say about the PS2. Especially considering the PS2 takes a hit on polygon count when you start throwing some lights into mix.