• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

;)

Nvidia have a real chance here to deal a good blow on their first batch of cards released. If they let them go for a decent undercutting price, it would stop the half of 7 series cards in it's tracks, especially since a fair few people are aggravated by their ridiculous pricing.

i agree, but i dont think its nvidias style to undercut the competition, thats amd's tactic :p
 
Anyone know how well that compute benchmark represents overall compute performance? some specific scenarios can synthetically give the AMD architecture - even GCN - a massive advantage which isn't represented in realworld compute performance.

That said GK104 won't be a compute monster at best it won't be significantly faster than the GTX580 for compute and was never intended otherwise.
 
Nvidia have a real chance here to deal a good blow on their first batch of cards released. If they let them go for a decent undercutting price, it would stop the half of 7 series cards in it's tracks, especially since a fair few people are aggravated by their ridiculous pricing.

Agree, but I don't think they are smart enough !
I would sell at £299..everybody would buy instantly, which would push prices up a bit. But if they sell at £400 they will not sell many.

It is after all a small cheap card and they must have huge profit margins to work with, I doubt many are fooled by the part number on the box !

Globally it still tough times and people want "Bargain" deals to motivate them to spend ;)
 
Anyone know how well that compute benchmark represents overall compute performance? some specific scenarios can synthetically give the AMD architecture - even GCN - a massive advantage which isn't represented in realworld compute performance.

That said GK104 won't be a compute monster at best it won't be significantly faster than the GTX580 for compute and was never intended otherwise.

I agree. And there's an obvious spin in this article because GTX 680 compute performance ought to to be down on compute performance, as expected. But even so 1 benchmark is not a very useful indication. Even the HD 5870 completely destroy the GTX 580 on the rare compute bench where the problem is can VLIW5-optimized.

Another thing to note is that the bench shows the GTX 680 running at 705Mhz clock.
 
I agree. And there's an obvious spin in this article because GTX 680 compute performance ought to to be down on compute performance, as expected. But even so 1 benchmark is not a very useful indication. Even the HD 5870 completely destroy the GTX 580 on the rare compute bench where the problem is can VLIW5-optimized.

Another thing to note is that the bench shows the GTX 680 running at 705Mhz clock.

Seen a lot of "leaks" from these little "shops" and "factories" where they don't seem to have the drivers setup properly for some reason - not sure if its user error or that the drivers in the box don't work very well, so I'd take any leaked results even from a reliable source with a bit of caution atm.
 
Anyone know how well that compute benchmark represents overall compute performance? some specific scenarios can synthetically give the AMD architecture - even GCN - a massive advantage which isn't represented in realworld compute performance.

That said GK104 won't be a compute monster at best it won't be significantly faster than the GTX580 for compute and was never intended otherwise.

From the hardware canucks admin:

"After going through the synthetic GPGPU programs over the last week, I've come to the conclusion that they aren't worth squat.

If they don't support a particular architecture (case and point: Kepler) or use OpenCL / Directcompute extensions that aren't yet supported at the driver level (ie: Sandra, ComputeMark, etc), then there will be absolutely horrible performance.

Not only that but certain architectures are specifically designed for certain tasks. For example, Tahiti is geared towards Double Precision tasks in a big way but will that superiority actually translate into faster compute performance in actual applications? Only if the applications in question are actually optimized for AMD's implementation of OpenCL."
 
moving-boxes-kitchen-300x300.jpg
 
Back
Top Bottom