lock these

[gosh]

first number fumbling from sony now nvidia? thier getting pretty desperate

 

[nAo]

first number fumbling from sony now nvidia? thier getting pretty desperate

I would remind to you this is a technical forum, enlighten us with your analysis. Tell us why those numbers are not correct

 

[gosh]

first number fumbling from sony now nvidia? thier getting pretty desperate

I would remind to you this is a technical forum, enlighten us with your analysis. Tell us why those numbers are not correct

because you cannot judge performance and make "Charts" between seperated and unified architure and the lack of eDram functionality mention in the chart

According to ATI, the R520 has about the same performance as the R500 (Xbox 360) while looking at these numbers a noobie can assume its only 50% of the performance

 

[nAo]

because you cannot judge performance and make "Charts" between seperated and unified architure and the lack of eDram functionality mention in the chart

They're not comparing fillrate or AA performance, eDram is completely out of context here. They're comparing programmable shading ops, I'm still waiting for you to tell use cause those numbers are not correct.

According to ATI, the R520 has about the same performance as the R500 (Xbox 360) while looking at these numbers a noobie can assume its only 50% of the performance

Whe don't know a thing about R520. There's nothing official out there.

 

[gosh]

because you cannot judge performance and make "Charts" between seperated and unified architure and the lack of eDram functionality mention in the chart

They're not comparing fillrate or AA performance, eDram is completely out of context here. They're comparing programmable shading ops, I'm still waiting for you to tell use cause those numbers are not correct.

the comparitive numbering is wrong. please tell me ur intelligent enough to understand this is PR

 

[Shifty Geezer]

@ gosh : This chart compares like-with-like peak metrics, not overall performance factoring in other aspects of the rendering system where the chips are to be employed.

 

[nAo]

the comparitive numbering is wrong. please tell me ur intelligent enough to understand this is PR

Tell us where and why they're wrong and tell us the correct numbers 8)
If I'm smart enough or not this is not your problem, you're making statements you have the burden of the proof

 

[gosh]

@ gosh : This chart compares like-with-like peak metrics, not overall performance factoring in other aspects of the rendering system where the chips are to be employed.

precisely how can you make a chart like with like between a unified vs seperated, a console specific vs pc derivitive.

 

[gosh]

so looking at this chart alone we can assume RSX is way more powerful than xenos?

Responses like this

 

[nAo]

Gosh, you're still not telling us why those numbers are not correct.
The're not comparing REAL WORLD performance, they're comparing THEORETICAL PEAK performance. what's wrong with that?
Unified shading doesn't mean anything in this case, in my opinion.
You disagree? Well, just tell us why you disagree, if you can't back up your statements maybe you shouldn't write here

 

[gosh]

no nAo im still waiting for the response as to how a unified vs seperated archicture can be explained in charts like by like comparison

 

[Shifty Geezer]

@ gosh : This chart compares like-with-like peak metrics, not overall performance factoring in other aspects of the rendering system where the chips are to be employed.

precisely how can you make a chart like with like between a unified vs seperated, a console specific vs pc derivitive.

Well what you do is you measure the clock rates and write those down. You see how many instructions each ALU can issue per clock and write that down. You do the same with other metrics that can be easily measured and write them down, side-by-side for at a glance comparison.

For PERFORMANCE evaluation, which works better, you need benchmarks. But this isn't a performance comparison chart; it's a technical chart.

BEar in mind though, as I said earlier, we should not trust these specs about ATi from nVIDIA, just as we shouldn't trust the comparisons about consoles from console manufacturers or their affiliated parties.

 

[gosh]

Gosh, you're still not telling us why those numbers are not correct.
The're not comparing REAL WORLD performance, they're comparing THEORETICAL PEAK performance. what's wrong with that?
Unified shading doesn't mean anything in this case, in my opinion.
You disagree? Well, just tell use why you disagree, if you can't back up your statements maybe you shouldn't write here

i disagree because the architure of the R500 + the philosophy of using Unified is different from R520, G70 and RSX, its like comparing G5s and Pentium 4s. I am only saying the numbers are wrong because you cant compare those numbers with different architectures, Steve Jobs used to do Benchmarks and Real time performance tests and not charts

 

[gosh]

Yeah, those Xenos numbers are wrong

Then why do they match up exactly to the MS ones?

G5s running at 2.0 Ghz are faster and Pentium 4s running at 3.0 Ghz. difference in architeture is why im against charts like this and for real world charts not BS PR charts

 

[gosh]

Gosh repeating the same stuff 10 times doesn't work here, you need to elaborate your statements, repeating a void mantra many times doesn't makle it any truer.

because youre not understanding. Comparing apple vs oranges archictures cannot apply to charts when you dont have real time performance charts

 

[jvd]

Gosh , Do not troll the forum . I've request that people do not responed to you but if they do in the future i will handle your posts the same way I handled these