New AMD Listings in Korean RRA Certification Point to Impending Graphics Cards Release - Big Navi?

[AsRock]

AMD should hurry up with Big navi, so I can buy a cheaper RTX 2080 Super that doesn't have tons of driver issues
Come on AMD, I am counting on you

So what ya saying is AMD need to hurry up so you buy from a competitor ?, WOW.

 

[TheoneandonlyMrK]

What kind of change do you expect? A smaller than 505 mm^2 chip is meaningless because there are already 251 mm^2 and 331 mm^2 N7 GPUs.

505 is a small chip compared to the reticle size limit of well over 800 mm^2.

The rumour isn't in any way started by SemiAccurate.
There are other reports from Taiwan and China about the shipped Navi 21 board there.

Which in no way included any specifics such as shader count OR die size.

Meaningless, I am not so presumptuous this time round is all, these could be true facts but as I said I wouldn't bet on them or argue any point with them at my side as fact ,but again it's a rumour, I wouldn't argue too hard against it just wish to temper the hype wagon since hype doesn't help.

Anyway ,Financial viability of a 7nm euv 500mm2 is dubious for a consumer AMD GPU ,but I suppose the price can rise.

 

[medi01]

there's no way AMD is going to catch up yet.

Who knows what it even is.

But there is no serious catching up left, after Navi release.
It beats Turing transistor for transistor.
There might (or might not, as cards are pushed further than needed) be a small gap on power consumption front.
That's it.

As for lack of TI price bracket cards:
1) 99% (literally) of the market could not care less
2) They'll get there pretty soon too, but there is no "catching up" (technology wise) aspect to it. There is no doubt they can make bigger chips, but an underdog (still) has only so many engineers and so much money for R&D.

that "smallest" chip in action

I'd expect from tech savvy users to realize where the catch is in this picture.
2080 can afford to go lower clocks, to give NV more headroom for new products.
XT, on the opposite, is pushed well beyond its comfort zone.

Having a 225W RX5700XT is already a good-enough sign that consumer cards are near the upper TDP cap, so the most we could actually hope for is a 48CU Navi card, maybe on 7nm+ process.

Same die lower clocks is 180w. Nothing stops AMD from making even bigger chip and downclocking them.
Check clock speed on 2080 and 2080Ti.

happened with 1080Ti and 5700 when Nvidia launched Turing.

That alternative history of yours.
Nvidia launched Turing long before 5700 and Navi arrived.

driver issues

driver issues

Messages: 62

Totally not paid, totally not shill, totally not spreading FUD.

epidemic Navi driver issue

And his victims.
Major reviewers experiencing no such problems, despite testing dozens of cards in dozens of configurations, and some using 5700XT for 4+ hours a day.

 

[cucker tarlson]

I'd expect from tech savvy users to realize where the catch is in this picture.
2080 can afford to go lower clocks, to give NV more headroom for new products.
XT, on the opposite, is pushed well beyond its comfort zone.

does strix run lower clocks than fe ?

 

[medi01]

does strix run lower clocks than fe ?

Does strix push clocks hard? Nope.
Why? NV has no need.

 

[cucker tarlson]

Does strix push clocks hard? Nope.
Why? NV has no need.

even 220w for reference is damn high for 250mm2

Doubling down on ReTardX is going to cost space.

do you actually know what you are talking about ?

 

[silentbogo]

Same die lower clocks is 180w.

If you are talking about RX5700 vs XT, then it's a bit misinformed. Not only does it have ~10% lower clocks, but also 10% less CUs. Yes, die is the same, but some parts of that die are fused off.
That's why those strange voices in my head say 48CU is probably the most we'll get from Navi refresh... realistically that's the only way to stay within adequate 250-275W limit for consumer cards.
Also, don't forget that on Turing (and Pascal, and Navi) boost clock is vague, temperature-dependent, and load-dependent. Just pointing at stock 1500-1700MHz means very little, when the card boosts upwards of 1900MHz if cooled properly and not head-banging into TDP cap (w/o overclocking).

 

[efikkan]

Do you, guys, realize that you are arguing over a non-existent product?
Rumors of "Big Navi" have been debunked some time ago, and "Arcturus" is a made-up name that someone came up with almost 2 years ago and everyone still uses it (even though it's not a thing).
Having a 225W RX5700XT is already a good-enough sign that consumer cards are near the upper TDP cap, so the most we could actually hope for is a 48CU Navi card, maybe on 7nm+ process.

You're right about "big Navi" being MIA so far. To my knowledge, the only fact I know about it is that Lisa Su confirmed it is coming, without any specifics. But this could mean anything from a slightly larger chip than Navi 10 to a monster chip. But regardless, it will still be Navi, so set your expectations accordingly.

Most people have already forgotten that there were rumors of "big Polaris" and "big Vega" too…
And AMD "Arcturus" does exist, but it's a Vega-derived data-center product.

Also, the majority of consumer GPU market segments are covered already, and as with Polaris, there's no need to rush for that 0.01% that'll give up their 2080Ti in favor of marginally faster AMD flagship.

Still, Nvidia greatly outsells AMD in the mid-range.
And as you can see in the Steam hardware survey, there are more people with RTX 2080s / 2080 Tis than RX 5700s…

TSMC 12FF process node transistor density: 33.8 MTr/mm^2
TSMC 7nm+ EUV process node transistor density: 115.8 MTr/mn^2

You are confusing maximum density of a node and average density of a chip.

Ampere is not yet available in the wild so we don't have any real die size measurements or even estimates to make. Thus, ideal density scaling is the only metric which we can use. Certainly you can admit that TSMC's 7nm+ is denser than 7nm, providing further density scaling which must be factored into any attempt at calculating an estimated transistor density for real silicon. So now we're over 2x scaling from process alone, approaching 3x. You don't think Nvidia can come up with the last ~33% from uarch and clock speed with a new generation design?

Different parts of chip designs will scale differently. E.g. cache will shrink very well, while hotter parts like FPUs wouldn't.

Don't get me wrong, the jump to 7nm EUV will be a huge upgrade for Nvidia, but we're not talking about a 3-4x density improvement, more like ~30% (+/- 10%).

Non-existant product that passed certification. Yeah seems really non-existant to me for AMD submit something that does not exist.

When did a "big Navi" specifically pass a certification?

The only reason why we have a 225W RDNA right now is that it's the first generation. I have no doubt AMD will improve power efficiency with RDNA 2.0. Problably lower clocks and fitting a 80 CU card inside 300W limit is not a big deal. Plus IPC increase from architectural improvements.

I advice you to learn what IPC actually means, it's not applicable here.

If anything, we should expect the greatest improvements in the first iteration of an "architecture", followed by smaller and smaller improvements.

 

[Lindatje]

RDNA1 = RDNA + GCN.
RDNA 2 = RDNA 2, a 100% new architecture.

If people had paid attention to the presentation of AMD for the RX 5700 / 5700XT (Navi), they could have heard it from Lisa Su themselves. That is what AMD (Lisa) said.

 

[Platinum certified Husky]

So what ya saying is AMD need to hurry up so you buy from a competitor ?, WOW.

AMD release RTX 2080 equivalent or better GPU -> Nvidia lowers the price of RTX 2080.

If AMD release RTX 2080 equivalent or better GPU with actual usable driver, then I will switch to AMD.

It's a win-win situation for consumers.

I don't know why it is so hard for people to understand.

Totally not paid, totally not shill, totally not spreading FUD.

So much salt

Totally no driver issues, totally no random black screen, totally no random core clock drop,totally no high idle power consumption with high refresh rate monitor, totally not being a AMD shill

I mean just do a quick google search and compare the amount of driver issues between AMD and Nvidia.

 

[AsRock]

AMD release RTX 2080 equivalent or better GPU -> Nvidia lowers the price of RTX 2080.

If AMD release RTX 2080 equivalent or better GPU with actual usable driver, then I will switch to AMD.

It's a win-win situation for consumers.

I don't know why it is so hard for people to understand.

Much better , i figured as much but you never know right ?. For example i been thinking of building a AMD system but being mainly a gamer Intel should be what i get, how ever as intels prices are only lower due to AMD and the point that it would be nice to have a AMD setup again ( last one was 939).

Don't know if i am going even bother yet, well i do know theirs no chance i am willing to spend $700+ on just a video card ha.

 

[ARF]

So much salt

Totally no driver issues, totally no random black screen, totally no random core clock drop,totally no high idle power consumption with high refresh rate monitor, totally not being a AMD shill

I mean just do a quick google search and compare the amount of driver issues between AMD and Nvidia.

random black screen - might be users' responsibility - something to do with their particular setup, and their errors in the configuration;
random core clock drop - what is this? Does it influence negatively the frame rate? Maybe it's a designed feature?;
high idle power consumption with high refresh rate monitor - the same as above - maybe it's a designed feature that users have to simply accept?!;

being a AMD shill - totally lack of understanding and blaming the others for own mistakes.

 

[Platinum certified Husky]

random black screen - might be users' responsibility - something to do with their particular setup, and their errors in the configuration;
random core clock drop - what is this? Does it influence negatively the frame rate? Maybe it's a designed feature?;
high idle power consumption with high refresh rate monitor - the same as above - maybe it's a designed feature that users have to simply accept?!;

Ah, the infamous quote used by all AMD fanbois : "It is users' fault that GPU is unstable from the factory, it must be their particular setup, and their errors in the configuration"

If random stuttering (result of core clock dropping) and high idle power consumption are so called "designed feature", I think AMD definitely nailed it

just like I said, do a google search and you will find the answer.

being a AMD shill - totally lack of understanding and blaming the others for own mistakes.

What? Calling others shill randomly is a bad thing now? I wonder who started this

However, I do agree that it is users' mistakes. They should have listened to others' warning

 

[AsRock]

Some times it just isn't the user.

 

[TheoneandonlyMrK]

Some times it just isn't the user.

you lot are turning yet another thread into a flame war , have a word with yourselves eh.

ITS NOT PROVEN WHAT IS CAUSING AMD DRIVER ISSUES IN ANY WAY AT ALL, THATS NOT THAT RELEVANT HERE ANYWAY AND AS EVER THE GREAT MAJORITY USE AMD FINE.

AND LOW AND BEHOLD MANY HAVE ISSUES WITH NVIDIA DRIVERS

complex systems have complex issues, simple people simplify any issue to simply finger pointing.

 

[ratirt]

I still don't get what's the all fuss about. Some have problems with drivers some don't. Isn't this normal for both NV and AMD vendors? What I really do admire is, how sometimes people blow things out of proportions to prove some ridiculous point which is not even true.

Anyway. Celebrate you guys. New graphics is coming and we get to see all the benchmarks for comparison. New stuff new opportunities.

 

[kings]

I still don't get what's the all fuss about. Some have problems with drivers some don't. Isn't this normal for both NV and AMD vendors? What I really do admire is, how sometimes people blow things out of proportions to prove some ridiculous point which is not even true.

When are the retailers saying that the return rate of Navi is more than 5x of Nvidia, they are not just "normal" problems, which as you said, can happen to any manufacturer. And if it were just normal random problems, not even AMD had come public to recognize it and report that they are trying to find a solution.

I don't know why some AMD fans are so defensive about this. Even for those who have the cards it´s bad to ignore the problem just to "defend your brand". If people go into denial, AMD sees no reason to try to fix the problems and improve the experience for their costumers.

 

[Lindatje]

When are the retailers saying that the return rate of Navi is more than 5x of Nvidia, they are not just "normal" problems, which as you said, can happen to any manufacturer

It was not the retailers, it was a poll from hardware unboxed, how many Nvidia fan’s has used this poll...?

I don't know why some AMD fans are so defensive about this. Even for those who have the cards it´s bad to ignore the problem just to "defend your brand".

Problems?
I know many people who have an AMD GPU and no one has a problem, reviewers also have no problems with AMD cards and neither do their family members, just look at hardware unboxed for example.
It all started with a dubious redit message. Now all problems suddenly become AMD's fault, not from the user because of, for example, a wrong OC or Windows settings because the user has changed it. It can't be due to Microsoft's updates, etc. No, it is all due to AMD fault because a few people on redit say so..... They (the redit users) cannot even be checked if it is all really true wat they say.

Because this whole witch hunt is underway, AMD must respond as a large company.

According to Nvidia fans it takes a long time because it might not be an AMD "problem" ....

Nvidia fans enjoy this, they love it. Although I continue to find it strange why people are a big fan of a company, the only thing the companies want is your money nothing more nothing less.

 

[kings]

It was not the retailers, it was a poll from hardware unboxed, how many Nvidia fan’s has used this poll...?

Yes, it was the retailers:

We’ve also spoken with a few retailers to see if they’re seeing issues with 5700 XT boards and we’ve received some interesting reports. Basically the return rate for 5700 XT's is more than 5x that of competing Nvidia products

Gamers are ditching Radeon graphics cards over driver issues

As I planned and was beginning to run our next big graphics card benchmark test, I felt I had to shift gears to discuss AMD's driver woes...

www.techspot.com

 

[AsRock]

you lot are turning yet another thread into a flame war , have a word with yourselves eh.

ITS NOT PROVEN WHAT IS CAUSING AMD DRIVER ISSUES IN ANY WAY AT ALL, THATS NOT THAT RELEVANT HERE ANYWAY AND AS EVER THE GREAT MAJORITY USE AMD FINE.

AND LOW AND BEHOLD MANY HAVE ISSUES WITH NVIDIA DRIVERS

complex systems have complex issues, simple people simplify any issue to simply finger pointing.

Point taken, but come on flame war, WOW.

Anyone who has delt with PC's for a number of years it could be 101+ reasons. Like shit i can get a black screen with the 290X and the 390X i have now which i believe is due to the PCI-e slot and not the card.

 

[medi01]

That's why those strange voices in my head say 48CU is probably the most we'll get from Navi refresh... realistically that's the only way to stay within adequate 250-275W limit for consumer cards.

You were told in the previous post to check what happened to clocks when NV added more CUs.
Yet you repeat this again.

 

[silentbogo]

You were told in the previous post to check what happened to clocks when NV added more CUs.
Yet you repeat this again.

Apparently you haven't read my post about boost clocks either, so I guess it's pointless to have a discussion here.

 

[medi01]

Apparently you haven't read my post a

Yeah, "boost clock is vague", if one "slaps outstanding cooling, it's higher" I guess.
Very bright comment, thank you so much for this, I'll go punish myself for not realizing its greatness in full the first time.

Bigger NV chips clearly drop clocks, which leads to watt/CU to go down (and to wrong impressions about CU scaling, but that beside the point)

 

[Platinum certified Husky]

So much salt

Totally no driver issues, totally no random black screen, totally no random core clock drop,totally no high idle power consumption with high refresh rate monitor, totally not being a AMD shill

I mean just do a quick google search and compare the amount of driver issues between AMD and Nvidia.

There, fixed your cherry picking screenshot for you

 

[medi01]

@UltraThicc
Fuses are burnt on this acc, I'm sorry about that.
Feel free to create new one though.
@cucker tarlson will appreciate.