20201101, 19:13  #2564  
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest
3·1,637 Posts 
Quote:
The 6900XT has half the memory bandwidth 512GB/sec, 1.44 GFLOPS FP64, vs 1TB/sec & 3.36 GFLOPS for Radeon VII. https://www.techpowerup.com/gpuspec...6900xt.c3481 https://www.techpowerup.com/gpuspecs/radeonvii.c3358 As a point of reference, 5700XT 448GB/s memory bandwidth, FP64 0.6 TFLOPS is around 4050% of Radeon VII speed in gpuowl, and that's with the benefit of power limits imposed on the Radeon VIIs here but the 5700XT running free. https://www.techpowerup.com/gpuspec...5700xt.c3339 Last fiddled with by kriesel on 20201101 at 19:32 

20201101, 20:14  #2565  
Jul 2009
Germany
547 Posts 
Quote:
TeraFLOPS (TFLOPS) = 10^12 FLOPS = 1.000.000.000.000 FLOPS TFLOPS= GFLOPS / 10^3 Last fiddled with by moebius on 20201101 at 20:18 

20201101, 21:23  #2566 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest
1001100101111_{2} Posts 
Still sounds optimistic.
The 6900XT has half the memory bandwidth 512 GB/sec, 1.44 TFLOPS FP64, vs 1TB/sec & 3.36 TFLOPS for Radeon VII. https://www.techpowerup.com/gpuspec...6900xt.c3481 https://www.techpowerup.com/gpuspecs/radeonvii.c3358 As a point of reference, 5700XT 448 GB/s memory bandwidth, FP64 0.6 TFLOPS is around 4050% of Radeon VII speed in gpuowl, and that's with the benefit of power limits imposed on the Radeon VIIs here but the 5700XT running free. https://www.techpowerup.com/gpuspec...5700xt.c3339 (using this convention: https://www.mathsisfun.com/definitio...malpoint.html) Radeon VII 100M P1, 862 us/it; 5700XT 2186; so ~39%. Last fiddled with by kriesel on 20201101 at 21:45 
20201101, 23:08  #2567 
"Viliam Furík"
Jul 2018
Martin, Slovakia
2·193 Posts 

20201101, 23:50  #2568 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest
132F_{16} Posts 
The claim is the cache is 2.17x faster than the GDDR6. 128MB sounds like a lot of cache. But it's not going to hold the whole footprint of a wavefront Gpuowl primality test, much less P1 stage 2. We see prime95 GhzD/day drop off with cache effectiveness at larger fft lengths.
https://www.dailystuff.org/2020/10/2...rtingusd549/ 0.39 x 2.17 is still not the 90% hoped for earlier, and that's with the benefit of unequal power limits It will be interesting to see how it does in real world conditions. Last fiddled with by kriesel on 20201101 at 23:53 
20201102, 00:14  #2569  
Jul 2009
Germany
547 Posts 
Quote:
https://www.techpowerup.com/gpuspec...ctmi100.c3496 Last fiddled with by moebius on 20201102 at 00:17 

20201102, 00:28  #2570  
"Viliam Furík"
Jul 2018
Martin, Slovakia
2·193 Posts 
Quote:
Infinity cache bandwidth can be estimated by subtracting the 256bit bandwidth (~512 GB/s) from 2.17 times the 936 GB/s, resulting in about 1.5 TB/s  1.5 times higher than Radeon VII's. Also, 128 MiB should be plenty for the current wavefront (100M to 110M) as 64 MiB on Ryzen 9 3900X is still enough for it. Based on my testing (results graph attached), 64 MiB is good until the 6400K FFT length. Further calculations predict that 128 MiB should suffice for FFT length up to about 16128K (about 297M exponents), assuming the same behaviour as in 3900X. Last fiddled with by Viliam Furik on 20201102 at 00:28 

20201102, 00:36  #2571 
Jul 2009
Germany
547 Posts 
it's been dealt with....
Last fiddled with by moebius on 20201102 at 00:44 
20201102, 01:08  #2572 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest
11457_{8} Posts 
Radeon VII and 5700XT 100M P1 stage 1 are showing 12 GB gpu ram occupancy in GPUZ with gpuowl V6.11380.
PRP on RX480 and RX550 at ~180M exponent show 11.5 GB occupancy. The article refers to "traditional" gddr6. An RTX3090, which is still too scarce to buy, doesn't seem to me very traditional, nor does the 384bit width. If the cache does pan out the 6800XT looks like a better buy to me than the 6900XT. Last fiddled with by kriesel on 20201102 at 01:48 
20201102, 02:52  #2573  
"Eric"
Jan 2018
USA
11010100_{2} Posts 
Quote:
Actually if those numbers are accurate, it is highly probable that the 6900xt will be more than 2x the performance of 5700xt, since assuming the same FP32 to FP64 ratio with RDNA2 the FP32 performance of 6900xt is 23.04TFLOPs (divide by 16 to get FP64) and with the 5700xt standing only at 9.75TFLOPs, so some 2.36x performance. With the 128MB of cache having huge effective bandwidth for wavefront FFT testing I think there might be a chance. Last fiddled with by xx005fs on 20201102 at 02:52 

20201102, 10:11  #2574  
"Composite as Heck"
Oct 2017
2×3×127 Posts 
Quote:
I still maintain that the types of gamer who bought R7 will want to upgrade to big navi so there may be a glut of them on the secondhand market. If big navi doesn't perform for compute or uses too much power to do so then hoovering up R7's is a good plan B. Of big navi 6800XT is the one to get, 8 extra CU's and a better bin are likely not worth $350 but $70 for 12 extra CU's seems like a nobrainer. 

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
mfakto: an OpenCL program for Mersenne prefactoring  Bdot  GPU Computing  1668  20201222 15:38 
GPUOWL AMD Windows OpenCL issues  xx005fs  GpuOwl  0  20190726 21:37 
Testing an expression for primality  1260  Software  17  20150828 01:35 
Testing Mersenne cofactors for primality?  CRGreathouse  Computer Science & Computational Number Theory  18  20130608 19:12 
Primalitytesting program with multiple types of moduli (PFGWrelated)  Unregistered  Information & Answers  4  20061004 22:38 