Question Speculation: RDNA2 + CDNA Architectures thread

[uzzi38]

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

 

[dzoni2k2]

I dont understand why try to push the Navi 22 agaist the GTX3070 in a dumb corporate attempt to try to get a big margin out of a card that will not sell very well, place it against the 3060 at the same price and win the mainstream sector to gain market share!

Why would they place N21 against 3060 when N22 is more than enough? N22 is between 2080s and 2080Ti. 3060 won't touch that. Not to mention 3060 has 6GB of VRAM lmao.

 

[GodisanAtheist]

I dont understand why try to push the Navi 22 agaist the GTX3070 in a dumb corporate attempt to try to get a big margin out of a card that will not sell very well, place it against the 3060 at the same price and win the mainstream sector to gain market share!

If AMD has a superior product they should be using it to provide a significant better performance than they Nvidia counterparts to win market, not get huge margins out of products that will sell in low volume, SPECIALLY with Nvidia having supply issues.

- If AMD has a product that performs like a 3070, they're going to price it like a 3070.

Supply is tightly constrained at the moment with RDNA 2 competing for fab space with Zen3, XSX, and PS5. No reason to sell for less now when they'll sell everything they make. Can drop prices later when/if supply outstrips demand.

 

[topmounter]

I'd say AMD needs a direct hit, if margins and supplies allow, they need to undercut Nvidia where they can. If the leaked benchmarks hold, I would say Big Navi is more of a Ryzen2 opportunity than a Ryzen3 opportunity.

 

[HurleyBird]

"Price to gain marketshare" vs "Price to make money" depends on two things imho.

First is how supply constrained they are. Second is how well they anticipate they'll be positioned versus Nvidia in the future.

To explain that second one, well, if AMD is thinking that they'll beat Nvidia by a significant margin with RDNA 3, then it would make sense to price RDNA 2 lower and take as much market and mind share as possible so that when RDNA 3 comes along they're in a better position to capitalize on its success.

On the other hand, If AMD expects that RDNA 3 will lose then it doesn't make sense to price RDNA 2 to gain share since they'll just lose it in the future anyway.

 

[kurosaki]

"Price to gain marketshare" vs "Price to make money" depends on two things imho.

First is how supply constrained they are. Second is how well they anticipate they'll be positioned versus Nvidia in the future.

To explain that second one, well, if AMD is thinking that they'll beat Nvidia by a significant margin with RDNA 3, then it would make sense to price RDNA 2 lower and take as much market and mind share as possible so that when RDNA 3 comes along they're in a better position to capitalize on its success.

On the other hand, If AMD expects that RDNA 3 will lose then it doesn't make sense to price RDNA 2 to gain share since they'll just lose it in the future anyway.

Strange thing as a company, not to think they would beat competition in a year, if they beat them now..
You got to have faith! Duh duh duuh, du du duh!
Why would they don't want to snag 80% market share? It's an especially easy win if they can show the competitor not only incompetent, but also greedy. (It's an especially easy win this round if priced well, good minshare is not underestimated) If the price marketed accordingly, nothing will change. If they show real bang for the buck, people will start to stream. If AMD Don have capacity increasing consumer base, fine. If they want to dominate, they will have to show some goodwill to win people over. A high quality gaming card for 300 bucks?
There lies sweet spot..

 

[Shivansps]

- If AMD has a product that performs like a 3070, they're going to price it like a 3070.

Supply is tightly constrained at the moment with RDNA 2 competing for fab space with Zen3, XSX, and PS5. No reason to sell for less now when they'll sell everything they make. Can drop prices later when/if supply outstrips demand.

It depends on supply? yes, but they already wasted this chance on RDNA1 with the "pricing like Nvidia counterpart".

 

[Karnak]

N7 was still expensive back then. Although we don't know exact numbers, but 16 months later it should be considerably cheaper.

 

[Geranium]

"Price to gain marketshare" vs "Price to make money" depends on two things imho.

First is how supply constrained they are. Second is how well they anticipate they'll be positioned versus Nvidia in the future.

To explain that second one, well, if AMD is thinking that they'll beat Nvidia by a significant margin with RDNA 3, then it would make sense to price RDNA 2 lower and take as much market and mind share as possible so that when RDNA 3 comes along they're in a better position to capitalize on its success.

On the other hand, If AMD expects that RDNA 3 will lose then it doesn't make sense to price RDNA 2 to gain share since they'll just lose it in the future anyway.

AMD will not go for price war. AMD tried that in past, even with flagship like 290X and Fury X, but it didn't work. AMD will price their card according to its performance, just like current RDNA cards.

 

[Geranium]

It depends on supply? yes, but they already wasted this chance on RDNA1 with the "pricing like Nvidia counterpart".

Pricing lower in past didn't raised AMD's market share with cheaper wafer cost. So why now why AMD would price its card lower with higher wafer cost??

 

[maddie]

AMD will not go for price war. AMD tried that in past, even with flagship like 290X and Fury X, but it didn't work. AMD will price their card according to its performance, just like current RDNA cards.

Will Nvidia, if AMD starts to take market share? Sometimes you don't choose your war, it chooses you.

 

[JujuFish]

AMD will not go for price war. AMD tried that in past, even with flagship like 290X and Fury X, but it didn't work. AMD will price their card according to its performance, just like current RDNA cards.

I expect a mild undercut. With the pent up demand for hardware, combined with limited supply, they probably don't need to even do a slight undercut.

 

[DeathReborn]

AMD will not go for price war. AMD tried that in past, even with flagship like 290X and Fury X, but it didn't work. AMD will price their card according to its performance, just like current RDNA cards.

To be fair though at the time Nvidia didn't have issues getting cards to market, if AMD can flood the market at a small undercut they will make a mint & gain a truckload of market share too.

 

[zinfamous]

Pricing lower in past didn't raised AMD's market share with cheaper wafer cost. So why now why AMD would price its card lower with higher wafer cost??

Yeah, Opteron and FX were not cheap, at all. That pricing never hurt their market share, so there isn't much reason to think they would implement that strategy in any part of their sector if they have an actual performance lead.

 

[zinfamous]

To be fair though at the time Nvidia didn't have issues getting cards to market, if AMD can flood the market at a small undercut they will make a mint & gain a truckload of market share too.

They will be facing the same supply issues as Nvidia this year and the next. I don't why some don't understand this.

They will sell everything they print, at whatever price they sell them at. There isn't much reason to significantly undercut nVidia in price

 

[GodisanAtheist]

It depends on supply? yes, but they already wasted this chance on RDNA1 with the "pricing like Nvidia counterpart".

- I figure AMD will still come in a little cheaper than their NV equivalent, probably $50-$70 USD, but they won't price down a whole tier vs NV.

As was mentioned in another post, RDNA was AMD's "Zen" moment and RDNA2 will be AMD's Zen2 moment: better in some things, slower in others, but a general return to competitiveness.

 

[PhoBoChai]

Is there a more neutral, or rather, non-RTX optimized RT game out there currently?

I think not.

Comparing RDNA2 RT performance in current RTX titles or benchmarks, is like comparing any prior AMD GPU on GameWorks titles, no? There is an innate advantage to NV GPUs in these games, and RTX in particular since only NV has had a say in how these features are used and optimized.

We should revisit this talk of RT performance, once a few next-gen console AAA games make it to the PC with RDNA2 optimized RT.

 

[undertaker101]

They will be facing the same supply issues as Nvidia this year and the next. I don't why some don't understand this.

They will sell everything they print, at whatever price they sell them at. There isn't much reason to significantly undercut nVidia in price

Why do you say that? They have much more flexibility with TSMC and their CPUs and motherboards have been mostly available. Yes the x570 was OOS for a few weeks over the summer but nothing even approaching the 3080 debacle. Read the jd.com review leak where AIBs are doctoring reviews to make the supply seem larger than it is..you know Jensen has been lying his teeth off.

 

[AmericanLocomotive]

The impressive part to me is that it's a relatively small die with a 192-bit bus that punches above its weight class in terms of performance, 4GB more VRAM than the 3070 included. AMD looks like they are extremely well positioned this generation. The only question left is RT performance, where it isn't looking all that rosy according to the new rumors.

AMD/ATi were famous for building small, high-clocked dies that punched well above their weight. Even all the way back to the Radeon x800 days in 2004. Their dies were always smaller, with fewer transistors, but clocked far higher. nvidia has had a thing for giant dies for a while. AMD/ATi never really seemed to figure out how to make a giant GPU die work well, and that's where they really started running into trouble.

While RT performance probably won't be great, I still don't think it'll be particularly relevant this product cycle.

 

[AnandThenMan]

I have buyers version of blue balls at this point.

 

[PhoBoChai]

While RT performance probably won't be great, I still don't think it'll be particularly relevant this product cycle.

This product cycle, as in the next 2-3 years?

Pretty sure RT is very relevant, even next year in 2021 with the new wave of AAA console ports we're about to receive.

As for RT not being great, I would reserve judgement until I see some RDNA2 optimized RT in games.

 

[GoodRevrnd]

Is there a more neutral, or rather, non-RTX optimized RT game out there currently?

I think not.

Comparing RDNA2 RT performance in current RTX titles or benchmarks, is like comparing any prior AMD GPU on GameWorks titles, no? There is an innate advantage to NV GPUs in these games, and RTX in particular since only NV has had a say in how these features are used and optimized.

We should revisit this talk of RT performance, once a few next-gen console AAA games make it to the PC with RDNA2 optimized RT.

I think that relevance depends on how long you keep a card. We're probably 2 years out from meaningfully optimized raytraced console games? In the meantime for me I think it comes down to how well will it run RT Cyberpunk and how well will it run everything else non-RT.

 

[Qwertilot]

Why do you say that? They have much more flexibility with TSMC and their CPUs and motherboards have been mostly available. Yes the x570 was OOS for a few weeks over the summer but nothing even approaching the 3080 debacle. Read the jd.com review leak where AIBs are doctoring reviews to make the supply seem larger than it is..you know Jensen has been lying his teeth off.

Look at the demand there was for Ampere's top chips? It's a nearly bottomless pit.

If AMD are broadly competitive (likely) then market share will just come down to who can produce more cards.

Everyone has won and all shall have prizes!

If that same level of increased demand hits Zen3, ps5, xbx etc then there really won't be much flexibility left for AMD to produce extra GPU's vs their original plans.

The console launches likely get top priority as there's vast amounts of future revenue from selling games at stake for MS & Sony.

NV do at least have SS almost to themselves, warts and all.

 

[moinmoin]

This product cycle, as in the next 2-3 years?

Pretty sure RT is very relevant, even next year in 2021 with the new wave of AAA console ports we're about to receive.

As for RT not being great, I would reserve judgement until I see some RDNA2 optimized RT in games.

RT will be very relevant through the adoption and usage in the... *drumroll* ...upcoming console games, which are likely to favor AMD's cards due to the optimizations necessary for using RT at acceptable frequencies (and those are actually the more likely the weaker AMD's RT should turn out). It will be very interesting to see how different Nvidia's and AMD's implementation of RT really are and how the performance impact of those to all the different optimizations will be in the coming years.

 

[misuspita]

That we have heard in the past when AMD GPU was in both consoles, and still Nvidia did, with throwing enough money at developers , changed the way the PC ports ran in Windows. I would like to think now is different, but history showed that Nvidia plays their cards very well, and this time has also precedence (second generation Ray Tracing in Nvidia), so I dont think AMD can get traction in RT this generation. I just hope the developers will bother implementing both ways of doing RT

 

[zinfamous]

I think that relevance depends on how long you keep a card. We're probably 2 years out from meaningfully optimized raytraced console games? In the meantime for me I think it comes down to how well will it run RT Cyberpunk and how well will it run everything else non-RT.

Cyberpunk's official recommended hardware (recommended, not minimum) are so strangely low that I'm not sure we should worry too much about RT performance across brands, haha.