Originally posted by: chizow
I'm only going off VRAM amounts when strictly comparing between the same architecture, obviously once you expand the scope to encompass other features the results will vary.
Let me backup and say that I think we're in an agreement here.
However, I think you and I are applying the same concept differently.
Again, if the benchmark you linked is "VRAM limited" and only the amount of VRAM will increase performance, why do you see the GTX 260 out performing the 9800 GTX+ 1GB?
Answer: Because factors outside VRAM amounts influence performance. (i.e., shaders, TMUs, ROPs, clocks, and bandwidth all contribute)
The reason why the GTX 260 can outperform the 9800 GTX+ 1GB in your benchmark is because it has more "GPU power" and more bandwidth.
Therefore, I don't see it as being enough to discredit the power of the 4870 512MB solely on its VRAM amount, especially when its GPU power and bandwidth - like the GTX 260 - is greater.
Originally posted by: chizow
Its clear bandwidth, shaders, rops, tmus, fillrate and anything else aren't enough to compensate in some instances, which is why I brought up the 1GB 9800GTX+ to 512MB 4870 comparison to begin with.
Again, by that account the GTX 260 should have lost to the 9800 GTX+ 1GB, but that didn't happen in the bench you linked.
Originally posted by: chizow
Originally posted by: josh6079
I'm not saying it won't, but it would be surprising since even at high resolutions with AA in VRAM-demanding games differences in performance are based on factors outside of VRAM amounts like memory bandwidth and GPU architecture.
How would it be surprising? We've seen it makes enough difference in the past with like architectures when comparing 8800GTX to 9800GTX and 4870 512 to the 1GB version.
Correct, but unlike those examples, the 4870 is a
different architecture than the 9800 GTX+, which was the comparison you were making.
The differences outside of VRAM between the 4870 512MB and the 9800 GTX+ 1GB factor into the overall performance of the product, even in VRAM-intensive situations like you and I have linked.
Originally posted by: chizow
Originally posted by: josh6079
"My" 4870 to 4850 comparison? I never claimed it to be a 4870 to 4850 comparison. I claimed it to be a 4870 512MB to 9800 GTX+ 512MB comparison at a VRAM-stressing situation.
Oh right, I was just a step ahead of you.
A step ahead of me? Put your epee away dude and be less offensive. You began comparing a different card that wasn't in the conversation just because it has the identical but lower clocked GPU core. Considering the topic at had was about
memory and only between
the 4870 and 9800 GTX+, I didn't think it was "a step ahead". Saying it was is simply egotistical.
Originally posted by: chizow
Again, you're comparing between different architectures in that case where the 4800s don't require additional VRAM as much as the G92.
You're discrediting the Anandtech benchmark on account of the G92
needing VRAM more than the 4800s?
:roll:
This is completely absurd beings how there's a
63% difference between the 4870 512MB and the 4870 1GB.
Originally posted by: chizow
But even in the case of your 4870 to 9800GTX+, I wouldn't say bandwidth is the greatest limiting factor again using the 4850 and 1GB 9800GTX+ as examples.
What are you going on about here? I never brought up the 4850, you did. Furthermore, I never suggested the 4850 would even beat a 9800 GTX+ 1GB, especially in the context we're discussing (i.e., VRAM-intensive) I also never claimed bandwidth was
"the greatest limiting factor".
Quit trailing off on tangents.
Originally posted by: chizow
Originally posted by: josh6079
Originally posted by: chizow
...it'd be inaccurate to say that difference is result soley because of VRAM...
When comparing the two products
you're wanting to (i.e., the 4850 and the 4870 512), yes it would be inaccurate to say the difference is soley because of VRAM.
Yep, meant to read bandwidth, which was the inaccurate assumption you made.
Rofl. Right, the blame is on me. Calling my "assumption" inaccurate because you didn't say the right word.
:roll:
Originally posted by: chizow
Which just confirms VRAM is still the greatest limiting factor at that particular resolution and setting.
The bench I showed suggests that VRAM is
a limiting factor. I don't know if it is *drumroll*
"the greatest limiting factor" because the 4870 doesn't come in a 1.5 or 2GB model. If it did we could without a doubt see what, if any, improvement would be had over the 4870 1GB at those settings since all of the other factors would then be the same.