Can anyone explain better how the primitive shader is supposed to work (and interact with current APIs) ?
Nice post, however I have still a couple of questions..
At the begin sebbi says that it is not clear how things work for Nvidia, however then he...
I guess the main complain is that the big market quote of Nvidia will slow down its adoption and utilization.
I was even hoping we could have had...
Could you link it, please?
I guess other reasons come before OpenCL, such as price and reliability. Given my luck, I (still) own two of the mbp which had gpu failure, one...
Yep, now it works, thanks
Link is broken
I guess if it had happened a couple of years before, he probabily would have added also the Nsa to his list.
The guy on overclock.net points out...
Take in account also planned obolescence
I guess there are a lot of people who buy only nvidia, no matter what
I love how people jump fast on conclusions to liquidate something they...
I don't know, I was thinking I am gonna run a couple of test by myself using my 680
But how could you explain something like that otherwise?...
Since a couple of days, I am reading a lot of threads regarding Nvidia deliberately downgrading kepler performance in favour of maxwell...
Then someone (maybe you?) should correct it, after all that's the spirit of wikipedia :)
Then maybe it is time to force devs to agree on one standard
In the main 970 ram thread on the Nvidia forum I found this...
Sorry Ryan, but I hardly doubt some memory occupance can descrease the bw that much and just at the very end in that way
Correct me if I am wrong
Yep, it is that, does it make sense what he says?
Moreover, is also the 970m being affected in any way?
It seems like the diagram is still a lie
The 'crossbar' diagram from Anandtech and other places speaking on behalf of Nvidia is STILL a...
I never understood why bw and memory capacity are linked each others.. somebody can explain to me?
Some other amd slides
What about the other Maxwell cards? Is there already some speculations about specifications and times?
Sorry but how an integrated core may alleviate bandwidth constraints?
I'd too, but this is what I read
I read a guy mentioning that his 290x was also about 5° cooler under Mantle
Some of you have heard something related to heat as well?
Seriously, I hate these shits.
Any speculation on CC 5?
My biggest disappointing with Dynamic Parallelism was that spawning threads couldn't...
I guess they refer about dropping the so-called hot clocks technique, where the shaders were running twice the frequency
I think too, 5 SMXs on the Ti and 4 ones on the normal 750
The full system I guess
It looks amazing, I want some real benchs
Cards with a shitty cooler is intended to lower the price as much as possible..
This is something that people willing to install a custom cooler...
I heard about problems cooling VRMs, but it'sfine since it is not done for the 290
Btw, it's a pity they didn't mention any temperature..
If you know that something new is going out and hopefully making prices fall, you mostly would like to wait before buying anything
Yep, you are right..
In far cry 3 there is a +10% with a 100% fan...
I see a lot of discussions without real numbers. With those everything can be much clearer, that is telling us if Mantle is worth or not.
Effectively this is true :smile:
What a pity, I really want to see some real benchs
Sorry, but which is the other thread you mean? This one?
Ach so :grin:
I skipped at the spot, this time I saw it all, very nice :wink:
I agree regarding the caches, I dont regarding registers
You can use TMU only under special...
Possible? The specs say 1,3-1,5 TFLOPS DP
OpenCL and CUDA are very very similar and they perform similar in almost all scenarios.
I just hope for Nvidia that this is gonna be true..
Disappointed for the gpgpu performances
Wait, in summary:
Cuda compute capability 3.x devices have 4 warp schedulers. This means that each SM can execute 4 warps at clock. Each of...
Same SMX same bandwidth I guess
I am also interested in this, can anyone confirm it?
That is a 6990 and the other a 680 (I suppose)
What we should get there? That a single-vga is shorter than a dual?