Something being announced on 24th iirc. Seen it a few pages back. Someone had a link to it.
-
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.
Radeon RX 480 "Polaris" Launched at $199
- Thread starter Kaapstad
- Start date
Something being announced on 24th iirc. Seen it a few pages back. Someone had a link to it.
This is what I see, besides clock speeds they don't have anything over Titan X and Ti.
the problem with the vr stuff doesnt the 480 (like the 10xx's) have superior vr magic hardwired in to it so by rights even if its slower than some cards in non vr it could very well be fast in vr ?
i dont know how the vr test works but id expect that said vr trickery would help that score up ?
i dont know how the vr test works but id expect that said vr trickery would help that score up ?
Are your comprehension skills really that low? No that is not what i am saying.. Try again.
This is what I see, besides clock speeds they don't have anything over Titan X and Ti.
With the 1080 it is all about clockspeeds, there is nothing wrong with high clockspeeds but that is where the performance comes from.
The transistor count on that chart for the 980 Ti are not all active as the chip has some disabled compared to the TitanX.
On the 24'th just under half the nation will be busy crying into their morning Croissants.
Unlike the 10xxs which have specific technologies to give them a boost in vr, no such thing has been announced for polaris as far as I'm aware.
Should actually be around 7.5 or more then @ 1266Mhz.
As a comparison reminder.
970 @ Ref Stock 6.8
@ 1500/1950 8.5
The question really becomes if the 480 was intentionally down clocked to offer "vr at low power." If not, then I have to wonder why anyone who is already planning on the ~600 dollar headset purchase wouldn't be willing to save just a little more for a 390x/980.
Lets see what happens when you run a 980 Ti at normal 1080 clockspeeds.
http://www.3dmark.com/hall-of-fame-2/fire+strike+3dmark+score+performance+preset/version+1.1/1+gpu
What is also interesting is there are no 1080s in the top 20.
I think the 980 Ti wins easy on efficiency.
http://www.3dmark.com/hall-of-fame-2/fire+strike+3dmark+score+performance+preset/version+1.1/1+gpu
What is also interesting is there are no 1080s in the top 20.
I think the 980 Ti wins easy on efficiency.
Lets hope it is downclocked because if its beaten by a stock 970 whats the point of this card full stop?
The 970 is inexpensive and low power.
Yeah all the bouncing around with expectations in here is hard to keep up with things.
From one side we have its a 970 performance to the other in it not being close to a 980ti.
I personally will be delighted with something over a 980 as clearly quite a big gap between it and the 980ti to nicely slot into.
i thought they said something at computex about it having vr wizardry in it, have to rewatch that now when i get chance.
I think we're saying the same thing
We are.
It does not matter how you get the performance as long as it is there.
Associate
- Joined
- 15 Aug 2012
- Posts
- 352
We are.
It does not matter how you get the performance as long as it is there.
Does make the situation seem even more artificial though (as in let's keep stuff off the card for when we'll need to get a Titan out).
Pascal Titan will be a monster but this time big Vega should give it some real competition.
(seriously wrong on the 980 a-sync part btw!
)You could say OPS is the measure of how fast a graphics card can process a given workload. Each task is broken down into operations that have to be completed, so if the operations were the same between graphics cards then the performance would be easily derived from the OPS. However things haven't run like that in almost forever - you have lots of shortcuts you can take, so that a complex set of tasks can be carried out with fewer operations or you re-use the result of an operation so as not to have to carry it out again. Most of the DX API features, architectural tricks, tesselation engines, culling etc. are about doing the same tasks with fewer graphics card (shader) operations.
Architectural improvements, and probably what was being got at before with the IPC comment, are about what tricks the GPU does to reduce the number of operations they need to carry out for a given task, or they are about removing other bottlenecks that mean the cards OPS throughput isn't being fully utilised, memory bandwidth for instance.
So you can improve graphics cards by both increasing the OPS (give more shaders, run them faster) and increasing the shortcuts/removing bottlenecks to make better use of those OPS. As I think you were saying, it's unlikely newer gen GCN is going to have worse architecture, so the minimum performance we can expect is one where we scale the (easily calculated) TFLOPs. Sadly that doesn't tell us about other bottlenecks - it sometimes happens that you make a process less efficient in order to ramp up clock speeds so the scaling doesn't work too well. But I doubt we need to worry about that.
Last edited: