• 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.

OcUK RX6000 series review thread

Soldato
Joined
8 Mar 2010
Posts
4,967
Location
Aberdeenshire
I definitely think AMD have some tuning to do with 4K, let's not forget just how new the Infinity Cache is for them to code for! At a guess I'd say they prioritised 1080p & 1440p performance, as that's where 95%+ of gamers still are.


Loving this XT though, it's fantastic at Ultrawide 3440!
Remember most of the 18games average 4k results or whatever are done without SAM. So it probably could match the 3080 or pass it in many games with it enabled...if you have the right system ofc.
 
Permabanned
Joined
31 Aug 2013
Posts
3,364
Location
Scotland
Remember most of the 18games average 4k results or whatever are done without SAM. So it probably could match the 3080 or pass it in many games with it enabled...if you have the right system ofc.

SAM is not limited to AMD though, and may even benefit Nvidia more as they use faster VRAM.
 
Permabanned
Joined
31 Aug 2013
Posts
3,364
Location
Scotland
I am specifically addressing your statement. If you know something about how a GPU with GDDR6x can transfer data to a CPU faster than a GPU with GDDR6 over PCIE 3/4 feel free to share it with the rest of us.

Well it's over PCIE 4 to start with and I said Nvidia may benefit more due to faster VRAM. Now I am assuming that it won't be all sent in one transfer and so not only does it require the initial request and all the handshaking that requires but many buffered batches, each having to interrupt to some extent what is going on with VRAM access on the GPU itself before the storage of data. Now if Nvidia can store that data faster due to having faster VRAM then the transfer is going to take less time. Hence Nvidia may benifit more from SAM than AMD.

Of course this is without any research at all into how SAM works, but simply based on past experience of coding data transfers between systems.

Maybe it's easier to think about the difference in copying a file from one location to another and compare that copy with the destination being a HD and another destination being a SSD. It's the same copy and the same hardware, just the storage is different.
 
Last edited:
Soldato
Joined
12 May 2014
Posts
5,236
Well it's over PCIE 4 to start with and I said Nvidia may benefit more due to faster VRAM. Now I am assuming that it won't be all sent in one transfer and so not only does it require the initial request and all the handshaking that requires but many buffered batches, each having to interrupt to some extent what is going on with VRAM access on the GPU itself before the storage of data. Now if Nvidia can store that data faster due to having faster VRAM then the transfer is going to take less time. Hence Nvidia may benifit more from SAM than AMD.

Of course this is without any research at all into how SAM works, but simply based on past experience of coding data transfers between systems.

Maybe it's easier to think about the difference in copying a file from one location to another and compare that copy with the destination being a HD and another destination being a SSD. It's the same copy and the same hardware, just the storage is different.
So you entire argument hinges on GDDR6x being able to store that data faster during instances were there are heavy data request happening on the VRAM?
 
Back
Top Bottom