Has anyone compared the FFT-performance (1D, 2D, 3D) between GTX 285 and GTX 480? I’ve made some initial tests with a 3D FFT of size 128 x 128 x 128 and got about the same execution time. Isn’t that a bit strange since the GTX 480 has double the amount of processor cores (transfer times not included)? Has the CUFFT library been re-optimized to fit Fermi cards?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| compare cufft in gtx 980 vs gtx 1080 | 1 | 509 | September 20, 2017 | |
| Who fast is the FFT with the new GTX680 | 5 | 2063 | June 10, 2012 | |
| GTX 480 - performance | 8 | 6794 | June 9, 2010 | |
| 2D FFT | 0 | 258 | January 18, 2018 | |
| Accuracy of CUFFT (regarding Fermi) | 1 | 3896 | April 29, 2010 | |
| Best FFT library for Fermi architecture what do you use for best performance? | 4 | 12254 | March 22, 2013 | |
| CUDA Separable Convolution and FFT-Based 2D Convolution | 0 | 417 | July 11, 2020 | |
| GTX280 versus 8800GTX running CUFFT The 8800GTX seems to compute FFTs faster than the GTX280. A fluk | 3 | 4148 | November 9, 2009 | |
| Do 1/2/3D FFTs work correctly on K80 if batch = 1? | 6 | 978 | April 4, 2016 | |
| double precision CUFFT 2.3 is to slow | 2 | 5066 | August 26, 2009 |