Question Speculation: RDNA2 + CDNA Architectures thread

Page 48 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

uzzi38

Platinum Member
Oct 16, 2019
2,703
6,405
146
All die sizes are within 5mm^2. The poster here has been right on some things in the past afaik, and to his credit was the first to saying 505mm^2 for Navi21, which other people have backed up. Even still though, take the following with a pich of salt.

Navi21 - 505mm^2

Navi22 - 340mm^2

Navi23 - 240mm^2

Source is the following post: https://www.ptt.cc/bbs/PC_Shopping/M.1588075782.A.C1E.html
 

uzzi38

Platinum Member
Oct 16, 2019
2,703
6,405
146
Last time AMD had the chance to release a really good price/perf GPU lineup because Nvidia went crazy on prices and low on perf improvements, they went ahead and only bothered with increasing the prices to match Nvidia gpus in order to offer just a bit better performance for the same money, while lacking all the extras like RT and DLSS.
Back then AMD had no chance of winning a price war with 7nm costs. They had a lot to lose if they tried, so they didn't.

That won't be the same for next gen. AMD still retains an advantage in PPA, they'll likely have an advantage on perf/W if the Series X is any sign of what RDNA2 should perform like (which will have knock on effects with cooler design and costs), on the high end especially they'll likely have greater leeway in terms of BOM (GDDR6X requirements in terms of signal integrity are madness) and the icing on the cake? 8nm yields with really large dies are a complete unknown, but given how cut down the 3080 is compared to the 3090 and on top of that, the apparent difference in silicon quality (20% extra CUDA cores and 14 extra GD6X chips but only a 30W boost to TDP while rumours are still pointing to a ~20% gain over the 3080), whereas N7's yields are approaching minimal defect density.

So im a little exceptical, specially because it has been A WEEK, it they had something they would have show it by now, why let RTX3000 gain momentum like this?

Because that really worked out for AMD the last few times they tried, didn't it?

RTG are learning from the mistakes they've committed over the last few years. Being leaky and creating hype prior to a launch are two of them
 

randomhero

Member
Apr 28, 2020
184
251
136
AMD are keeping things tightly under lid.
But there is lots of hints. My personal belief is that they will have GPU competitive with rtx 3090.
If I'm wrong, then I'm wrong, but I cannot shake this feeling.
Die size of xbox apu without cpu part is around 300-310 mm2, 384 bit memory controler and phy included,56cu gpu.That sort of gpu could be around rtx2080ti performance or even 5-10% faster.
Rumoured big navi die size is 505 mm2. That is 60% biger die that probably uses hbm2e.That can carry a big punch.
I will not even go with all those memory compression patents recently released.
I think we are all set for a little surpris.
 

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,249
136
AMD are keeping things tightly under lid.
But there is lots of hints. My personal belief is that they will have GPU competitive with rtx 3090.
If I'm wrong, then I'm wrong, but I cannot shake this feeling.
Die size of xbox apu without cpu part is around 300-310 mm2, 384 bit memory controler and phy included,56cu gpu.That sort of gpu could be around rtx2080ti performance or even 5-10% faster.
Rumoured big navi die size is 505 mm2. That is 60% biger die that probably uses hbm2e.That can carry a big punch.
I will not even go with all those memory compression patents recently released.
I think we are all set for a little surpris.

Raw horsepower isn't enough anymore. The battlefield already shifted to the gimmicky stuff.

Intel and Nvidia will stop at nothing to insure their dominant positions.

Voting with the wallet helps, but in the end theres more sheep then hurders/sheep dogs!

Looking forward to reviews once launched.
 

GodisanAtheist

Diamond Member
Nov 16, 2006
7,063
7,489
136
Because that really worked out for AMD the last few times they tried, didn't it?

RTG are learning from the mistakes they've committed over the last few years. Being leaky and creating hype prior to a launch are two of them

- I don't really understand this sentiment. It wasn't the pre-release marketing or hype that made AMD have a non-competitive product, it was just not having a competitive product in Fiji or Vega.

Hyping something to the moon that cannot even reach your competitor's second tier is a bad idea. Promoting something that is neck and neck with your competitor's top tier is a good idea.

If AMD had something performance competitive with NV that is ready for release in the next couple months, then they would and should show it to bolster their own fans confidence and maybe give pause to the fence sitters who can go either way on a card.

The gap that they are leaving is actually being filled with "50-60% faster than the 2080Ti" levels of hype, which is potentially more dangerous when they do reveal.
 
Reactions: SMU_Pony and Lodix

Glo.

Diamond Member
Apr 25, 2015
5,762
4,667
136
AMD is not releasing and not leaking any info, because Nvidia hasn't shown everything they have in the bag.

Releasing any info right now is like opening yourself for "competition". And AMD wants to win, one way or another.
 
Reactions: kurosaki and Tlh97

Konan

Senior member
Jul 28, 2017
360
291
106
AMD is not releasing and not leaking any info, because Nvidia hasn't shown everything they have in the bag.

Releasing any info right now is like opening yourself for "competition". And AMD wants to win, one way or another.

But they have a Fornite Arena announcement
It is a game of chess. NV went first, AMD will go next and then NV will counter.
 

TESKATLIPOKA

Platinum Member
May 1, 2020
2,429
2,914
136
I think its worth quoting this, with a little summary that can be gathered from rumors, to get a little context.

N23 - 40 CUs/192 bit bus.
N22 - 60 CUs/256 bit bus.
N21 - 80 CUs/384 bit bus.
192bit bus => 6BG Vram
Does anyone believe that AMD will release M23 with only 6GB Vram, while performing 10% faster than 20820 Super?
The other option would be 12GB Vram, but that's also unlikely.

The next problem is 192bit bus.
5700XT has 256bit 14Ghz => 448GB/s
N23 with 192bit 14Ghz => 336GB/s
N23 with 192bit 16Ghz => 384GB/s
Even If N23 had 16Ghz memory 5700 XT would still have 17% higher bandwidth and this N23 is supposedly as fast as 2080 Super or faster which means 39% over 5700 XT(2080 Super is 26% faster than 5700 XT and another 10% over that is 1.26*1.1 = 1.39). That would mean N23 has 62% more effective memory subsystem compared to 5700 XT, which is in my opinion a nonsense.

BTW those TBPs you mentioned once more would mean >80-100% better performance/W ratio compared to Navi, yet AMD mentioned only 50% better performance/W ratio. AMD managed only 50% in the previous generation by using 7nm and a new architecture(RDNA1) and now RDNA2 on a barely better process should have over 80-100%? In my opinion this is another nonsense.
 

GodisanAtheist

Diamond Member
Nov 16, 2006
7,063
7,489
136
AMD is not releasing and not leaking any info, because Nvidia hasn't shown everything they have in the bag.

Releasing any info right now is like opening yourself for "competition". And AMD wants to win, one way or another.

- This is fair, but if AMD doesn't try to steal some of NV's thunder come launch/review day, then I'll openly take it as a bad sign.

NV has crapped on countless AMD launches with total confidence, if AMD cannot do the same I have to assume they do not have faith in their own products.

I am happy to be wrong and I hope that I am wrong, but someone's gotta keep this train on the tracks...
 

Glo.

Diamond Member
Apr 25, 2015
5,762
4,667
136
- This is fair, but if AMD doesn't try to steal some of NV's thunder come launch/review day, then I'll openly take it as a bad sign.

NV has crapped on countless AMD launches with total confidence, if AMD cannot do the same I have to assume they do not have faith in their own products.

I am happy to be wrong and I hope that I am wrong, but someone's gotta keep this train on the tracks...
Well, lets see what will happen.
 

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,249
136
- This is fair, but if AMD doesn't try to steal some of NV's thunder come launch/review day, then I'll openly take it as a bad sign.

NV has crapped on countless AMD launches with total confidence, if AMD cannot do the same I have to assume they do not have faith in their own products.

I am happy to be wrong and I hope that I am wrong, but someone's gotta keep this train on the tracks...

AMD's going to sell all that they produce anyways. I'd rather they don't join the ruthless (can't type that here) and just keep on being AMD.

Nice guys finish last? I'd rather be last then trip the guy in front of me.

Console effect? You guys should thank GOD that AMD rules it.
 

Glo.

Diamond Member
Apr 25, 2015
5,762
4,667
136
AMD's going to sell all that they produce anyways. I'd rather they don't join the ruthless (can't type that here) and just keep on being AMD.

Nice guys finish last? I'd rather be last then trip the guy in front of me.

Console effect? You guys should thank GOD that AMD rules it.
Remember guys, Mining craze incoming, once more. No matter how many wafers AMD and Nvidia will have at their disposal for dGPUs everything may be sold out, for months...
 

randomhero

Member
Apr 28, 2020
184
251
136
Raw horsepower isn't enough anymore. The battlefield already shifted to the gimmicky stuff.

Intel and Nvidia will stop at nothing to insure their dominant positions.

Voting with the wallet helps, but in the end theres more sheep then hurders/sheep dogs!

Looking forward to reviews once launched.
Gimmicky stuff included.
 

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,249
136
Remember guys, Mining craze incoming, once more. No matter how many wafers AMD and Nvidia will have at their disposal for dGPUs everything may be sold out, for months...

Well luckily I'm happy with my RX 5700 as I'm a 1080p part time peasant gamer.

Ampere should be the new king of mining! Big Navi won't be so good at mining....You know those sucky drivers and all that everybody brings up. Ampere will just work!
 

Midwayman

Diamond Member
Jan 28, 2000
5,723
325
126
- This is fair, but if AMD doesn't try to steal some of NV's thunder come launch/review day, then I'll openly take it as a bad sign.

Likely. I'd probably wait until a few days before launch. Maybe the 14th? Give the info a couple days to peak. The only complication here is the consoles. I wonder if they have any agreements with them about announcement timing?
 

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,249
136
Likely. I'd probably wait until a few days before launch. Maybe the 14th? Give the info a couple days to peak. The only complication here is the consoles. I wonder if they have any agreements with them about announcement timing?

Rumor is PS5 announcement on the 9th.

Agreements?
 

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,249
136
Sony already shot that down actually.

That's good news for me. I was actually contemplating buying one. If they did announce and pre-order on the 9th my impulse buy would have most likely kicked in. It'll give me more time to talk myself out of it.

I'd imagine no matter when launched they'll sell out quickly. I feel sorry for those who have to deal with the scumbag scalpers!
 
Reactions: Tlh97 and Konan

jpiniero

Lifer
Oct 1, 2010
14,839
5,456
136
That's good news for me. I was actually contemplating buying one. If they did announce and pre-order on the 9th my impulse buy would have most likely kicked in. It'll give me more time to talk myself out of it.

They are pretty obviously waiting until the last second. How far they can go without upsetting retailers I have no clue. The Switch was 7 weeks from price announcement to launch, btw.
 
Reactions: Tlh97 and Konan

Tup3x

Golden Member
Dec 31, 2016
1,012
1,002
136
- This is fair, but if AMD doesn't try to steal some of NV's thunder come launch/review day, then I'll openly take it as a bad sign.

NV has crapped on countless AMD launches with total confidence, if AMD cannot do the same I have to assume they do not have faith in their own products.

I am happy to be wrong and I hope that I am wrong, but someone's gotta keep this train on the tracks...
At least one thing is sure: NVIDIA will sell truckloads of cards while AMD has nothing to offer.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |