@ sieanr

 Yes really. The RSX is powerful, but so is the Xenos. In terms of graphics generation, they are about equal. Yet the RSX has 24 standard pixel pipelines, while the Xenos has 48 pipelines with a unified structure, which every objective test (as in not trying to make Sony, Microsoft, or even Nintendo look better than the other) rates as higher capability.

 Think this doesn't help resolution? The PS2 has the greatest resolution of the last gen, even if the actual graphics are lesser. This is due to the greater pixel/texel fill rate, and having 16 pixel pipelines compared to 4 that the Xbox and GC had (yes, that is in their specs).

 That doesn't mean the PS3 will have bad resolution. It just means that the 360 can do much better than you claim.

 As for the RAM, the XDR has a faster clock speed, but clock speed is not the only meacure of RAM speed. Unlike a processor, RAM has to pause in between cycles. This is latency.* XDR is part of the Rambus DRAM class, which has a slower latency than standard RAM. This is because FMVs require clock speed over latency, hence why the blu-ray playback is soo good. Yet graphics need latency over clock speed. Hence the GDDR3 VRAM on the PS3 is more suited for graphics, while the XDR is used as emergency VRAM.**

 The 360 used all GDDR3 for its RAM. This doesn't mean the PS3 can't fit as much graphics in its RAM, and with proper use of the SPEs, can fit more in there. Yet it does mean that your claim of a gap is false, not matter how much you type "IMO". This is not up to opinion. The facts are there. Look them up.

 *And in case anyone brings up the connection between the XDR and the EIB on the Cell, the EIB only prevents further latency between the parts connected to it (since otherwise all those parts working at once in the cell would clog each other). It can't reduce the latency the parts already have.

 **Yet system RAM will work fine on either type of RAM, so things like physics and AI work fine on the XDR.

@ LordTheNightKnight



Twice the shader power.

You can't just compare the number of pipelines between the two GPUs. The RSX only has 24 pixel shaders and 8 vertex shaders, compared to the 360's 48 unified shaders, but an individual unified shader is not as good as a dedicated shader for any specific task. The RSX's vertex shaders can handle 5 ops per cycle and the pixel shaders can handle 2 ops per sec, whereas all of the Xenos' shaders are only capable of 2 ops per second.

This leads to a a maximum of 136 ops per cycle for the RSX ((24 x 5) + (8 x 2) = 136) compared to a maximum of 96 ops per cycle for the Xenos (48 x 2 = 96). The fact that the RSX is clocked at 550mhz compared to just 500 for the Xenos also gives it an advantage.

Much like the Cell versus the Xenon, and dedicated versus unified RAM, it's power versus efficiency.  The ps3 has more max theoretical power, with the Cell, XDR dedicated CPU RAM, more powerful Vertex shaders, and the ability of the Cell to assist the RSX, but the question is, will developers be able to take advantage of this?

Heavenly Sword, MGS4, Killzone 2, and FFXIII all say yes.

But why do multiplats look like crap on the ps3?

Because they were designed around the 360's more flexible hardware, and to port them to the ps3 would require either a complete reworking of the code or else just cutting things from the game. Developers are money-conscience and chose the latter to save time.

MikeB said: @ LordTheNightKnight Twice the shader power. which every objective test (as in not trying to make Sony, Microsoft, or even Nintendo look better than the other) rates as higher capability No, the Xenos is just more flexible. However the SPUs are also flexible in what ways they can help the overall more powerful RSX. the XDR has a faster clock speed XDR is higher clocked but also offers less latency than DDR2 / DDR3.

 I didn't write that the Xenos was more powerful. I wrote that it was more capable. Last I checked, flexibility was a capability. So you countered your own point. Plus I did state that using the SPEs would allow greater graphics, so you tried to counter me with one of the very points I made. This makes me wonder if you even read my post properly.

 The XDR comment just shows your ignorance. If it had faster latency than DDR3, then it would have been used as the RAM in the RSX, instead of GDDR3, as the combination of size, clock speed, and latency would make it ideal for VRAM. It doesn't have faster latency, and that's why it's just used for system RAM. It works fine for that, but if you can't get your facts straight, your "IMO" looks more like deliberate lying.

Proofreading! You have that backwards, the pixel shaders handle 5 ops per cycle, the vertex shaders 2 ops per cycle (you did it in the right order when you did the math, though). And stop confusing "per cycle" with "per second". If these chips could only do a few ops per second, we'd be seeing pretty awful graphics. :) I'm not trying to give you a hard time, but these things can be confusing enough even with good proofreading.  :)

What's also important is to look at what kind of "ops" you're talking about when you say 5 pixel shader ops per cycle on the RSX vs. 2 on Xenos. As I understand from the spec sheets (which may or may not be detailed enough for this kind of analysis), the RSX can do two vector ops per cycle and two scalar ops per cycle, plus fog, whereas Xenos does one vector and one scalar op per cycle. Well, the fog step happens at the very end of the shader -- the output fragment gets blended with the fog color. This doesn't happen every cycle, it just happens "free" at the end of the process. The same thing happens on Xenos, they just don't count it as a 'shader op' on the spec sheet, since it really isn't one. So what I see is 48 pipelines * (1 vector + 1 scalar) vs. 24 pipelines * (2 vector + 2 scalar) + 8 pipelines * (1 vector + 1 scalar).  Which gives:

RSX: 56 vector + 56 scalar shader ops per cycle @ 550 MHz
Xenos: 48 vector + 48 scalar shader ops per cycle @ 500 MHz

I can see stating that the RSX is the more powerful chip, but let's not overstate it.

Nice job reading my post, but it seems ignoring things you dont like is something you often do.

But I'll give you the benefit of the doubt, incase you missed it.

Can you figure it out now?

Facetious; lacking serious intent; concerned with something nonessential, amusing, or frivolous

MikeB said: @ sieanr Still, the Darkness used the PS3 as its lead platform and that was only 540p. Hmm.... Like it or not, here's the clarification for your problem: "It depends on the type of engine you are doing. In The Darkness they have pretty similar performance, but that is very intentional from our side. We need the two platforms to perform similarly, and therefore we can’t design features that would take advantage of the difference of the two platforms. To my knowledge the PS3 has untapped potential in its seven SPUs" You're welcome.

Funny you failed to mention UT3 being only 720p and only 30fps on the PS3 after you postulated that the 360 would perform best at those resolutions and framerates.

Care to discuss that?

And do you still think Halo 3 could run at 1080p and 60fps on the PS3? If so, I emplore you to post that on Beyond 3D as I'm sure that idea will go over well.

 I did compare them.  What I was saying was that you can't compare them based on number of shaders alone.  You have to get into the tyope of shaders, ops per cycle, etc.

Also, yes, I am aware that not all of the shaders may be used at the same time in a non-unified architecture.  That's why I said it's efficiency vs brute power and drew parallels between the shaders of the Xenos/RSX and the dedicated/unified RAM of the ps3/360.

Basically, you missed exactly what I was trying to say.  I apologize for my miscommunication.