• 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.

AMD Navi 23 ‘NVIDIA Killer’ GPU Rumored to Support Hardware Ray Tracing, Coming Next Year

Status
Not open for further replies.
Still no information (or even false rumours) about the ROP counts for RDNA 2 graphics cards. This makes me think that all the rumours so far have been guess work and aren't in fact, based on 'inside sources' or leaks from AIBs. If they were genuine, why is this key piece of information always omitted?

Also, why the complete lack of consistency in the leaked specs, from different websites.
 
It will not be 6GB, it will be 12GB.
I'd be much more surprised if AMD skimp on VRAM, but it's not beyond the realms of possibility.

However I don't think they'll want to launch mid to mid-high PC cards that are inferior in any way to the APUs in the consoles.
 
You guys are assuming that the 6700xt is aimed at the same market as the 5700xt which was pretty much AMDs top card. If the 6700xt is now a £250 card then 6gb ain't so bad.
 
Last edited:
You guys are assuming that the 6700xt is aimed at the same market as the 5700xt which was pretty much AMDs top card. If the 6700xt is now a £250 card then 8gb ain't so bad.

The Radeon VII HBM2 16 GB VRAM is AMD's top card. RX 5700 XT came close to it because of the mess that the RTG currently is.
Everything is borked in that division.
 
Last edited:
Still no information (or even false rumours) about the ROP counts for RDNA 2 graphics


It's anywhere between 64 and 128 at increments of 16...a credible macOS kernel leak suggests that the biggest part has 16 render backends...Xbox series x has 80 Rops so it would be reasonable to assume that Navi 21 has atleast 96
 
Xbox series x has 80 Rios so it would be reasonable to assume that Navi 21 has atleast 96

The ROP count (listed on techpowerup) for the series X is just an estimate, Microsoft haven't confirmed the ROP count. Could easily be 64.

I think 128 is quite unlikely for RDNA 2, only Tesla GPUs have had that many in the past. 112 is the ROP count of the RTX 3090.
 
Last edited:
You're brave, placing faith in a few megabytes of cache to do the job of a wider vram bus. Especially if 'may not be required' is the most concrete evidence you have....
Bravery? Go do one child, I'm not in the mood for your petulant baiting.

I'm simply not making any assumptions. Come review day, if Infinity Cache proves to be a mistake then the "concrete evidence" you so sorely require will be given to you. If AMD's design choice proves to be a good one then excellent. I have reservations about such a radical departure from traditional design but let's see how it pans out. SO where's your "concrete evidence" that it's going to be a bad thing? No? Jog on then.
 
Its a nice, civil debate we're having, isn't it?

I think what i will add is that volume drives cost. ignoring the 'whether it will be faster than a 3080 or not' debate aside, it is an irrefutable fact that 120 million AMD GPU+16GB VRAM configurations will be sold in the next 4 years and that when it comes to RDNA2, cost and volume, AMD have a big advantage that Nvidia do not, this has to have a downward pressure on 6900XT pricing, over the next 6 months.

That volume advantage will not extend to RDNA3 but that's for another thread.
 
Bravery? Go do one child, I'm not in the mood for your petulant baiting.

uuh, what? :rolleyes:

I think you need to step away from the keyboard lad, before you bust a blood vessel. Where'd that aggression come from? Somebody **** in your cornflakes this morning?

'petulant' reaction aside, we have nothing but rumours. rumours built on rumours, no less. I don't need evidence that a rumour isnt true because that's not how it works, is it? but you know that of course.

Come review day, if Infinity Cache proves to be a mistake then the "concrete evidence" you so sorely require will be given to you. If AMD's design choice proves to be a good one then excellent. I have reservations about such a radical departure from traditional design but let's see how it pans out. SO where's your "concrete evidence" that it's going to be a bad thing? No? Jog on then.

sorely require? jog on? WTF is wrong with you?
 
Last edited:
Status
Not open for further replies.
Back
Top Bottom