I am benchmarking a few GPUs in my machines. I would be happy, if we all could list the GFLOP readout of our graphic cards, as they appear in the logfile.

Here I go with the first ones:
------------------------------
GeForce 210 67 GFLOPs
Intel HD4400 with i3-CPU (2.9GHz) 105 GFLOPS
NVIDIA GT610 156 GFLOPs
NVIDIA GTX970 4087 GFLOPs

---

Kind Regards,
Horst

There is some info also:

https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units

"Processing Power" there gives at least a hint about the relative differences between cards.

Thx for the list. I tried in the past to compare these figures with the data I have. Sometimes they fit, sometimes not.

------------------------------
NVIDIA Quadro FX 380 LP 66 GFLOPs

---

Kind Regards,
Horst

Is your GFLOPS rating for single precision operations or double precision?

I think they are single precision.

---

Kind Regards,
Horst

New Graphics:
------------------------------
GeForce GTX 750 Ti 1472 GFLOPs
Intel HD 4600 with i7-CPU (3.4GHz) 56 GFLOPs

---

Kind Regards,
Horst

(sapphire r9 270x 2gb)(aib)CAL: ATI GPU 0: AMD Radeon HD 7850/7870 series (Pitcairn)6848 GFLOPS peak)
(hd7660 igpu)(a10-5800k)CAL: ATI GPU 1: AMD Radeon HD 7500/7600/8500/8600 series (Devastator) (CAL version 1.4.1848, 512MB 2054 GFLOPS peak)

New Graphics:
------------------------------
AMD/ATI R7 240 (1GB DDR3) 312 GFLOPs

This is a silent card without fan. Probably the fastest 0dB GPU

---

Kind Regards,
Horst

AMD R7 260x OpenCL: AMD/ATI GPU 0: Bonaire, 2048MB available, 1120 GFLOPS peak.

(sapphire r9 270x 2gb)(aib)CAL: ATI GPU 0: AMD Radeon HD 7850/7870 series (Pitcairn)6848 GFLOPS peak)

6848 Gflops seems to be strange for a 270x.
A R9 Fury makes 7168 GFlops (wiki)

New Graphics:
------------------------------
AMD R9 380X 4096MB DDR5 2534 GFLOPS

---

Kind Regards,
Horst

XFX HD7990

OpenCL: AMD/ATI GPU 0: Tahiti (driver version 1702.3 (VM), device version OpenCL 1.2 AMD-APP (1702.3), 2987MB, 2987MB available, 2688 GFLOPS peak)
OpenCL: AMD/ATI GPU 1: Tahiti (driver version 1702.3 (VM), device version OpenCL 1.2 AMD-APP (1702.3), 3035MB, 3035MB available, 2688 GFLOPS peak)

please help me figure out why boinc reports the card this way and will repost more accurate results...been at this long time...never had this kinda "power" before(pre 1999)

GeForce GTX 960 (driver version 359.12, CUDA version 7.5, compute capability 5.2, 2048MB, 1634MB available, 2566 GFLOPS peak)
GeForce GTX 960 (driver version 359.12, CUDA version 7.5, compute capability 5.2, 2048MB, 1634MB available, 2644 GFLOPS peak)

GeForce GTX 760 (driver version 359.12, CUDA version 7.5, compute capability 3.0, 2048MB, 1663MB available, 2650 GFLOPS peak)
GeForce GTX 760 (driver version 359.12, CUDA version 7.5, compute capability 3.0, 2048MB, 1663MB available, 2439 GFLOPS peak)

GeForce GTX 670 (driver version 359.12, CUDA version 7.5, compute capability 3.0, 2048MB, 1659MB available, 2634 GFLOPS peak)
GeForce GTX 670 (driver version 359.12, CUDA version 7.5, compute capability 3.0, 2048MB, 1659MB available, 2845 GFLOPS peak)

GeForce GTX 780 (driver version 347.88, CUDA version 7.0, compute capability 3.5, 3072MB, 2838MB available, 4636 GFLOPS peak)

** Ps. Underclocking the cards or running them at stock speed doesn't affect the results. I wonder where BOINC gets those values.

28-12-15 16:08:41 | | CUDA: NVIDIA GPU 0: GeForce GTX TITAN X (driver version 361.43, CUDA version 8.0, compute capability 5.2, 4096MB, 4025MB available, 8479 GFLOPS peak)
28-12-15 16:08:41 | | OpenCL: NVIDIA GPU 0: GeForce GTX TITAN X (driver version 361.43, device version OpenCL 1.2 CUDA, 12288MB, 4025MB available, 8479 GFLOPS peak)
28-12-15 16:08:41 | | OpenCL: AMD/ATI GPU 0: Hawaii (driver version 1912.5 (VM), device version OpenCL 2.0 AMD-APP (1912.5), 4096MB, 4096MB available, 3200 GFLOPS peak)

28-12-15 16:08:41 | | OpenCL: AMD/ATI GPU 0: Hawaii (driver version 1912.5 (VM), device version OpenCL 2.0 AMD-APP (1912.5), 4096MB, 4096MB available, 3200 GFLOPS peak)

@xixou hi, what model AMD GPU ?

28-12-15 16:08:41 | | OpenCL: AMD/ATI GPU 0: Hawaii (driver version 1912.5 (VM), device version OpenCL 2.0 AMD-APP (1912.5), 4096MB, 4096MB available, 3200 GFLOPS peak)

@xixou hi, what model AMD GPU ?

R9 290

Nvidia GTX 650 (driver 352.41) Peak 813 GFLOPS
Compute capability 3.0; CUDA 6.0; 1024 MB available
Max. performance mode at 1058 Mhz clock.

Is Fury X supported ?

Is Fury X supported ?

Partially.

GCN 1.2 GPUs seem to have problem and it seems to be a driver related issue when running more than one task at a time.

see this thread for example.