hardware computational power specs how to undersatnd the specs of the different lines of hardware

Dear there.

I have some difficulty to compare the computational capabilities of different lines of video card.
for example, for qudro fx 1700 and tesla c1060

for qudro
and tesla

does the # of Thread Processors with tesla has the same meaning as Config core in Qudro?
how about core clock? does Core clock in qudro has the same meaning as core clock in tesla?

how many thread processor does FX 1700 has?

can I just say the computational power of FX 1700 is: 32 * 460? ( number of cores * core clock)
and the computational power of tesla c1060 is: 240 * 600?

tks

KZ

This table tends to be useful for comparing cards.

You want to compare shader clocks, not core clocks. The table linked above also gives memory bandwidth which in many cases is even more important than computational resources.

I came across CUDA as I want to use it to speed up some image elaboration algorithms but I’m facing the following doubt: Which are the main differences between the different families of video cards from just computational power perspective?

I know Tesla are the best suited for these tasks but they are far more expensive than GeForce and Quadro.
Among the latter two which are the main characteristics I should check to compare the computational power?

Based on what I heard, they are using same/familiar cores but different drivers. ( computational power perspective)

GeForce line is designed for personal pcs for regular gaming/rendering stuff.

Qudro line is designed better for some cad/image processing stuff. and usually for professional purpose with better tech support.

Tesla is designed mainly for computational purpose( they even don’t have a dvi connector for your monitor ), of course, they have good tech support and mostly installed on workstations like qudro.

I also learned that qudro and tesla are well tested with little defect compare to geforce line.

---------- did not get those confirmed with nvidia yet----------

see, my geforce 8600gt has the similar(very close even better) computational power as qudro fx 1700 with my test.

geforce 8600gt has 4 thread processors(multi processors) and total 4*8 = 32 cores, the clock rate is 1.19 ghz with 256 mb memory

qudro fx 1700 has 32 cores too, and the colck rate is 0.92 ghz with 512 MB memory.