What is the difference between kepler and fermi

Email Me. Notify Me. Update avatar. Browse or drag an image. File must be at least xpx and less than xpx. GeForce Forums. Join Now or Login. All Topics. Feature Requests. Sort by. Topics details. GeForce Graphics Cards. By Recency Recency Votes Hot. JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding. Previous Next Sort by votes. CaptainSolo Honorable. Jul 12, 3 0 10, 0. How much of a difference does GPU architecture make?

Apr 29, 28, 75, 2, Usually yes. Fermi is better at double precision work 64it FP? If you need to do this your best bet on the cheap is still the GTX If all you care about is gaming then yes, Kepler cards will be faster then their Fermi counterparts. Dec 14, 2, 0 21, CaptainSolo :. But seriously - newer is always better. Well, at least in the case of Nvidia because they handle their hardware with an excellent driver team.

May 11, 12, 1 48, 1, It makes a lot of difference, look at series release benchmarks and you can plainly see that. So how does GK change this?

At the same time, with twice as many functional units NVIDIA also doubled the other execution resources, with 2 warp schedulers becoming 4 warp schedulers, and the register file being doubled from 32K entries to 64K entries. Altogether the SMX now has 15 functional units that the warp schedulers can call on. While that covers the operation of the SMX in a nutshell, there are a couple of other things relating to the SMX that need to be touched upon. Instead the capabilities of the Polymorph Engine itself have been doubled, making each Polymorph Engine 2.

As with the many other things being doubled in GK, the number of GPCs has been doubled from 2 to 4, thereby doubling the number of Raster Engines present. Last, but certainly not least is the complete picture. The ROPs themselves are virtually identical to those found on GF; theoretical performance is the same, and at a lower level the only notable changes are an incremental increase in compression efficiency, and improved polygon merging.

On a technical level, getting to 6GHz is hard, really hard.