AMD Radeon RDNA2 Navi (RX 6500, 6600, 6700, 6800, 6900 XT)

I think i found a programming failure in MacOs 10 for Navi 21 :D It is correct in MacOs 11. So we can say Navi 21 has indeed 8 rasterizer :D

ElluhnkXEAAkyU8

https://www.reddit.com/r/Amd/comments/j7bpzs
Click to expand...
 
Last edited:
Digidi said:
Can sombody explain me why all people guessing that Navi 21 has only 4 Rasterizers? I cant see why it not should be 8 Rasterizers:

Navi21_Navi10_2de1730741a364ff9.png
Click to expand...
I would look to the num_sc_per_sh as a possible indicator of the number of rasterizers, assuming sc stands for scan converter. Packers would be related to primitive order pixel shading, and having more might allow for more independent sections to be tracked in parallel.
 
3dilettante said:
I would look to the num_sc_per_sh as a possible indicator of the number of rasterizers, assuming sc stands for scan converter. Packers would be related to primitive order pixel shading, and having more might allow for more independent sections to be tracked in parallel.
Click to expand...
I did read my post above your one :D But thank you for the information.

Can sombody tell me how i put in a redit link without showing the reddit windows in the post? I need only the link as source. Thank you.
 
Last edited:
Digidi said:
I did read my post above your one :D But thank you for the information.

Can sombody tell me how i put in a redit link without showing the reddit windows in the post? I need only the link as source. Thank you.
Click to expand...

You could wrap the reddit link in quotes
 
One thing I have to say. If independent reviewer benchmarks bear out AMD's claims WRT to RDNA 1, then RDNA 2 is a phenomenally impressive accomplishment for AMD regardless of how it compares to NV.

NV's accomplishment with Ampere compared to Turing is pretty impressive, but this goes beyond that WRT being a IHV centric generation to generation improvement if RDNA 2 ends up being well over twice as fast as RDNA 1.

For AMD (formerly ATI). as a generation to generation uplift? I don't know if it rivals R300 versus R200, but it's pretty damned impressive none-the-less.

Regards,
SB
 
Silent_Buddha said:
For AMD (formerly ATI). as a generation to generation uplift? I don't know if it rivals R300 versus R200, but it's pretty damned impressive none-the-less.
Click to expand...
The RV670 -> RV770 jump also comes to mind.
But on both of these cases there were considerable jumps in performance while using the same node, which is pretty impressive.

RDNA3 in late 2021 using 5nm seems very promising too, at this point.
 
Silent_Buddha said:
One thing I have to say. If independent reviewer benchmarks bear out AMD's claims WRT to RDNA 1, then RDNA 2 is a phenomenally impressive accomplishment for AMD regardless of how it compares to NV.

NV's accomplishment with Ampere compared to Turing is pretty impressive, but this goes beyond that WRT being a IHV centric generation to generation improvement if RDNA 2 ends up being well over twice as fast as RDNA 1.

For AMD (formerly ATI). as a generation to generation uplift? I don't know if it rivals R300 versus R200, but it's pretty damned impressive none-the-less.

Regards,
SB
Click to expand...


I agree. RDNA2 is looking really impressive. If they follow the trends they've set with Ryzen, RDNA3 is something I'm really looking forward to. Lisa Su is earning that compensation package she's getting in terms of where the company has positioned under her leadership.
 
ToTTenTranz said:
The RV670 -> RV770 jump also comes to mind.
But on both of these cases there were considerable jumps in performance while using the same node, which is pretty impressive.

RDNA3 in late 2021 using 5nm seems very promising too, at this point.
Click to expand...

RDNA2 is certainly looking like AMD's Maxwell and while RV770 did significantly increase performance, AFAIK the power efficiency did not increase as much.

There's no certainty that RDNA3 is on 5nm, AMD's conspicuous use of "advanced node" instead of an explicit mention of 5nm as they do with Zen 4 leads me to believe it could be on 6nm. And the roadmap showing RDNA 1,2 and 3 extends till the end of 2022 so it might well be a 2022 product as well.
 
Erinyes said:
RDNA2 is certainly looking like AMD's Maxwell and while RV770 did significantly increase performance, AFAIK the power efficiency did not increase as much.

There's no certainty that RDNA3 is on 5nm, AMD's conspicuous use of "advanced node" instead of an explicit mention of 5nm as they do with Zen 4 leads me to believe it could be on 6nm. And the roadmap showing RDNA 1,2 and 3 extends till the end of 2022 so it might well be a 2022 product as well.
Click to expand...

Didn't David Wang promise yearly cadence? Also wth is N31 entry in MacOS? It's obviously not something RDNA3 based.
 
What about rumors about the existence of other versions of Navi21 with HBM2? The memory bandwidth level for tandem from HBM2 and Infinity Cache is very interesting. Maybe this option will appear in Apple Mac Pro systems.
 
SimBy said:
Temper your expectations. Forza 4 and BF5 run great on AMD hardware and Ampere scales slightly worse at lower res.
Click to expand...

I am. There's no way I believe that in some games the 6800 will give me higher FPS than a 3090 until I see it in reviews. But if it's true...then damn AMD really deliver here.

Is there an announcement on when the NDA will finish?
 
xEx said:
I am. There's no way I believe that in some games the 6800 will give me higher FPS than a 3090 until I see it in reviews. But if it's true...then damn AMD really deliver here.

Is there an announcement on when the NDA will finish?
Click to expand...

It's possible if the game is fillrate limited. Radeon 6800 has a fillrate advantage over the 3090, I think.
 
You must log in or register to reply here.

Similar threads

Kaotik
AMD Radeon Software Driver Announcement thread
2
Replies
29
Views
13K
DegustatoR
D
Back
Top