By using this site, you agree to our Privacy Policy and our Terms of Use. Close
Pemalite on 29 September 2021
Edit Thread
Bofferbrauer2 said:
Pemalite said:

Once you start taking the improved bandwidth into account, the GPU will definitely be able to breathe, especially in fillrate heavy scenarios.
The Switch is bandwidth limited.

I am probably being conservative by stating "50%" when there is more than a 100% gain in real-world bandwidth.

The point of Pascal though is that, nVidia re-architected the GPU to operate at higher clockrates with minimal impact on power usage... We saw it on the PC with the move from Maxwell to Pascal.

You're really not. I got those 20-40% not from estimates, but from Benchmarks between the TX and the TX2, and that was the bracket of performance increase (in FPS) between the two with similar power consumption.

Gotta' disagree. Because... Bandwidth. 25.6GB/s isn't good for 1080P gaming... There would be a linear increase in performance with a doubling of bandwidth.

I did some testing on this back in the day and posted it on this very forum about how bandwidth scales with resolution. - Around 150-200GB/s is the ideal ballpark for 1080P... Otherwise you are fillrate limited.

So you are going from 25.6GB/s to 59.7GB/s... Which is a 133% increase in bandwidth... Maxwell to Pascal V4 Delta Colour compression gives Pascal a 20% best-case scenario bandwidth improvement on top of that.
https://www.anandtech.com/show/10325/the-nvidia-geforce-gtx-1080-and-1070-founders-edition-review/8

So that potentially makes the difference of 25.6GB/s to 71.64GB/s or 179% advantage to Tegra Pascal.

If you can point me to a point in history where a GPU of the same relative architecture but with a 150%-200% bandwidth advantage getting less than a 50% performance improvement at decent resolutions... I'll eat my hat.

Because I can point to many cases where nVidia/AMD have taken a GPU, but hamstrung bandwidth by including DDR4/DDR3 over GDDR5 and performance took a massive dive.
Like the Radeon 7750, Geforce 1030 etc'.

JEMC said:

Yeah, the memory bus will probably limit the potential of that chip. But we'll see.

And one of the reasons Silicon Lottery is closing is because most CPUs already run close to their limit nowadays, making their "products" less appealing.

The Infinity Cache does muddy the waters significantly though.

Bofferbrauer2 said:

To me, it looks more like Navi 33 will be for a 7600XT or even just 7500XT, not a 7700XT.

  • For one, it seems like AMD is using more shaders at a lower clock speed with the 7000 series. As such, only having 40 CU again would certainly be slower than the 6700XT unless the ipc increases a very lot.
  • Second, it's the third chip in line. The 6700XT was the second-biggest chip, as both the 6800XT and 6800 used a cut-down 6900XT die. I don't think AMD will stray from this basic setup with the next gen

As a result from the "more shader at lower clock", the TDPs also look silly. I very much doubt that Navi 33 will use more than 150W, and more probable closer to 100W-120W

Yeah. Navi 33 looks like the successor to the 6600XT... Which is a good GPU for 1080P gaming.






--::{PC Gaming Master Race}::--