Question Speculation: RDNA3 + CDNA2 Architectures Thread

[uzzi38]

Man I have been dying to make this one for a while now.

First rumours for RDNA3 are here so new thread time!

Just going to start off with this one for now: kopite7kimi on Twitter: "@VideoCardz Ah, I mean a simple mcm design with 10240 cores is not enough. Because the lift from RDNA2 to RDNA3 is much bigger than from RDNA1 to RDNA2. We should expect many big improvements in GFX11. 🤔" / Twitter

 

[maddie]

As has been explained to you already. The ONLY reason why Intel went for tiles for their mobile MTL and ARL SOCs is because the CPU portion will be manufactured on Intel nodes, and GPU on TSMCs.

And partially, you are correct that Intel is not doing the same thing as AMD. Its the other way around, its AMD who has to compete with Intel's volume and wide availability of their products, which is why they have to build overkill products to sell. If Intel is going to release ARL-P with 384 EUs/3072 ALUs - AMD has to respond.


It isn't silver bullet. But for such small GPU as SP's - its enough to feed the ALUs.

Not if the low external bandwidth introduces a large penalty on misses. There must be balance, witness the 6500XT., same principle, just the bottleneck farther away from the L1

 

[maddie]

It's not a silver bullet solution, but probably the cheapest or easiest one to compensate low BW.
What would you do? Double bus width or use GDDR6 as a system memory?
If It was me, then I would use a single HBM stack.

As has been explained to you already. The ONLY reason why Intel went for tiles for their mobile MTL and ARL SOCs is because the CPU portion will be manufactured on Intel nodes, and GPU on TSMCs.

And partially, you are correct that Intel is not doing the same thing as AMD. Its the other way around, its AMD who has to compete with Intel's volume and wide availability of their products, which is why they have to build overkill products to sell. If Intel is going to release ARL-P with 384 EUs/3072 ALUs - AMD has to respond.


It isn't silver bullet. But for such small GPU as SP's - its enough to feed the ALUs.

Not if the low external bandwidth introduces a large penalty on misses. There must be balance, workable ratios. Witness the 6500XT., same principle, just bottlenecked farther away from the L1. Cache size alone in isolation is simplistic.

 

[TESKATLIPOKA]

Phoenix APU needs 45w to clock 12CU RDNA3 at "up to" 3ghz, but apparently all it takes is a die shrink and twice the CUs can be run at one third the power - while leaving the CPU with enough to live on.

Just as a reference, the Steam deck's 8CU RDNA 2 APU runs at ~1.5ghz..

I am also pretty sceptical, but Steam Deck has a 7nm APU with only 15W TDP, so 8CU clocking at 1.5GHz is actually pretty good. Doesn't mean 24CU 2Ghz 3nm IGP would consume only 15W.

Not if the low external bandwidth introduces a large penalty on misses. There must be balance, workable ratios. Witness the 6500XT., same principle, just bottlenecked farther away from the L1. Cache size alone in isolation is simplistic.

It's not like I don't know about this.
That's why I wanted someone with N23 to downclock Vram to see what happens to performance.
Then we will know exactly what to expect.

Does anybody know someone with this GPU, who would be willing to test It?

 

[maddie]

I am also pretty sceptical, but Steam Deck has a 7nm APU with only 15W TDP, so 8CU clocking at 1.5GHz is actually pretty good. Doesn't mean 24CU 2Ghz 3nm IGP would consume only 15W.

It's not like I don't know about this.
That's why I wanted someone with N23 to downclock Vram to see what happens to performance.
Then we will know exactly what to expect.

Does anybody know someone with this GPU, who would be willing to test It?

That's a good test. Remember the CPU shares memory also.

 

[insertcarehere]

And partially, you are correct that Intel is not doing the same thing as AMD. Its the other way around, its AMD who has to compete with Intel's volume and wide availability of their products, which is why they have to build overkill products to sell. If Intel is going to release ARL-P with 384 EUs/3072 ALUs - AMD has to respond.

Yes, so AMD should overbuild all their APUs with overkill iGPUs that the majority of its customers do not need or want...in case Intel makes an overbuilt tiled product with an overkill iGPU that the majority of its customers do not need or want. Great business strategy.

The last time Intel and AMD made an APU with a ton of hardware dedicated for iGPU was Kaby Lake-G. Notice that neither AMD nor Intel have been anxious to produce spiritual successors to that product, for good reason.

 

[Kronos1996]

If you are sure my 150mm2 design priced at $239 would be too close to Navi 33 then you can share with us how much N33 will be sold for.
My design would certainly have much better performance/price than N24. I can't tell how It would fare against N33, because I don't know Its price.


Mobile N24 is RX 6300M, RX 6450M, RX 6500M, RX 6550M and for mobile workstations Pro W6300M and Pro W6400M

There is such a huge demand for N24 in laptops(consumer + business), that I could find only 3 different laptops with Navi24.
Laptop models with 6500M: HP VICTUS, ThinkPad Z16 G1 and Bravo 15 B5E
Nothing else exist as far as I know.
This doesn't say anything positive about N24's sales.

It would be best If you provided some data to back up what you said.

Data? Every argument made here is based on estimates and numbers pulled out of our collective asses. There’s little public data to base any of this on. But since you asked:


Navi 33

• 248 / $6000 = $24.20
• 8 x $6 = $48
• Silicon + Memory = $72.20

Navi 14

• 384 / $6000 = $15.63
• 8 x $6 = $48
• Silicon + Memory = $63.63

A difference of $9 for something that should be ~50% faster. Similar power class so the PCB/Cooler costs are likely the same. Your ~150mm2 Navi 24 with 6GB could subtract $12 so now the difference is $21. Congratulations you just made the RX 480 for the 4th time at a higher MSRP. The IO doesn’t shrink much so it’d perform pretty similar to the 5500 XT/580/480 again, hence why I used it. You’re product is a very expensive refresh of something that already exists. The only way to get more performance was to make a slightly bigger die which is what they did for Navi 33 while making a barebones Navi 24 for budget laptops and especially business PC’s where AMD lacked an iGPU and needed something to be compettitive against Intel. Also for the business customers who need something with validated GPU drivers.

It serves a purpose, your design does not. It was never meant to be a DIY product in the first place. Navi 33 represents what a next-gen low-end GPU will be given the cost increases and scaling issues with IO. A small die is already mostly IO percentage-wise compared to larger ones. If you include AMD’s 50% margin then it’ll only cost $31-32 more then your design and should pack a meaningful performance improvement around ~50%. That’s a product worth making and even at $300 for a cut down model it would have way higher perf/$.

Business products are sold directly to customers in bulk so of course you won’t find them for sale at Microcenter.

 

[jpiniero]

An N7/N6 wafer is a lot more than 6 grand. It's closer to like 10-11.

Because of the wafer price hikes, N33 is likely more expensive than N23 was when it launched.

 

[Glo.]

Yes, so AMD should overbuild all their APUs with overkill iGPUs that the majority of its customers do not need or want...in case Intel makes an overbuilt tiled product with an overkill iGPU that the majority of its customers do not need or want. Great business strategy.

The last time Intel and AMD made an APU with a ton of hardware dedicated for iGPU was Kaby Lake-G. Notice that neither AMD nor Intel have been anxious to produce spiritual successors to that product, for good reason.

Strix Point will not be overbuilt. It will be in line with what we have had till this point: sub 200 mm2 die sizes.

Genuinely, this argument is completely and utterly ridiculous considering that leaked Driver data suggests that Intel is indeed bringing "overbuilt" iGPUs with MTL-P and ARL-P.

 

[Mopetar]

With the other rumors that NVidia won't have MX parts going forward, iGPUs could certainly step it up a bit more to fill in that gap. Not every consumer needs a beefy GPU, just like not everyone needs an 8-core CPU. Having such parts available does allow the market niches which demand them to be captured.

 

[Heartbreaker]

With the other rumors that NVidia won't have MX parts going forward, iGPUs could certainly step it up a bit more to fill in that gap. Not every consumer needs a beefy GPU, just like not everyone needs an 8-core CPU. Having such parts available does allow the market niches which demand them to be captured.

More like the other way around. NVidia is killing parts that are worse than new APU like the 680m.

What OEM is going to buy a discrete GPU when when the iGPU/APU already present is of similar performance.

 

[jpiniero]

With the other rumors that NVidia won't have MX parts going forward, iGPUs could certainly step it up a bit more to fill in that gap. Not every consumer needs a beefy GPU, just like not everyone needs an 8-core CPU. Having such parts available does allow the market niches which demand them to be captured.

Problem is that wafer prices are why nVidia is killing MX. AMD's APUs have the same problem.

 

[Mopetar]

Problem is that wafer prices are why nVidia is killing MX. AMD's APUs have the same problem.

APU avoids the cost of needing an extra board, VRAM, and cooling solution. It also removes any AIB profit margin from the consideration as well.

The economics aren't as bad as you might think. Sure costs are higher, but that means GPUs are more expensive as well. Having a beefier iGPU that can substitute for a discrete card for people who want to game occasionally adds a lot of value.

As long as they're not designing anything that will be horribly bottlenecked by the available memory bandwidth, it has a purpose.

 

[Glo.]

APU avoids the cost of needing an extra board, VRAM, and cooling solution. It also removes any AIB profit margin from the consideration as well.

The economics aren't as bad as you might think. Sure costs are higher, but that means GPUs are more expensive as well. Having a beefier iGPU that can substitute for a discrete card for people who want to game occasionally adds a lot of value.

As long as they're not designing anything that will be horribly bottlenecked by the available memory bandwidth, it has a purpose.

Costs are actually lower for designing and manufacturing monolithic APU then a separate CPU and dGPU, adding everything up(boards, VRAM, general complexity of a platform). Also much more efficient.

 

[Mopetar]

Costs are actually lower for designing and manufacturing monolithic APU then a separate CPU and dGPU, adding everything up(boards, VRAM, general complexity of a platform). Also much more efficient.

That's my point.

 

[jpiniero]

Costs are actually lower for designing and manufacturing monolithic APU then a separate CPU and dGPU, adding everything up(boards, VRAM, general complexity of a platform). Also much more efficient.

Not in the era where Moore's Law is dead. That's why AMD doesn't get much in the way of IGP only gaming focused OEM sales because Cezanne+3050 is cheaper.

 

[Heartbreaker]

APU avoids the cost of needing an extra board, VRAM, and cooling solution. It also removes any AIB profit margin from the consideration as well.

Not seeing anyone suggest APUs are a bad idea.

Just that they aren't going to make Big Console size APUs.

AMDs current balance looks about right for the available BW, and you really can't expect more than that.

 

[zinfamous]

That's why I also included Civ IV. We're really nitpicking at this point, when the leakers were so far off that in reality we didn't get any increase in perf/$ this generation at all.

not really. It's a discussion about GPU performance, so talking looking at 1080p scores and setting any kind of percentage comparison, for this era, is meaningless. Just...anything north of 150fps might as well be 5%-90%-200% different, it really just barely matters. For competitive FPS or whatevers, let's set that bar at 200FPS at and call it a day? It really doesn't matter.

If we're talking CPUs, then yes, it's worthwhile, but if you just want to compare GPU performance by frames, with flagship parts over the last 3 generations now, the only meaningful measure really is 4k FPS. It doesn't matter if you or I or anyone else plays at 4k, but it's the only frame rate you should like at if you want to make real quick comparison between GPUs in just raw performance. You can always interpolate back to your resolution if you want to, but lower resolutions aren't as meaningful for flagship parts and well, looking at architectural improvements between generations (I mean, I mostly play at 3440x1440 or...uh 7680x1440 ...and I like to see those numbers, but I know it isn't the first thing that I need to see. I can always think about the efficiency that I might see. )

 

[insertcarehere]

Strix Point will not be overbuilt. It will be in line with what we have had till this point: sub 200 mm2 die sizes.

Yes, on N3. So as a ballpark for cost x2 that vs 5nm which makes it equivalent to sub 400mm^2 on N5. Yes, 200mm^2 on N3 for Strix Point would be in all likelihood very overkill, I expect it to be a sub 150mm^2 part.

Genuinely, this argument is completely and utterly ridiculous considering that leaked Driver data suggests that Intel is indeed bringing "overbuilt" iGPUs with MTL-P and ARL-P.

I have no idea what ARL-P is supposed to be, but last I heard MTL-P can have a 128EU part as a iGPU option, which is definitely not overbuilt compared to what ADL-P/RPL-P have already.

And you're missing the point that by doing tiling Intel isn't making all their MTL/ARL parts with the maximum level iGPU, there are already 64/128EU tiles for MTL and doubtless something similar for ARL. A monolithic Strix Point will not have such an option to modify the amount of on-die iGPU for end-consumer requirements.

 

[zinfamous]

More like the other way around. NVidia is killing parts that are worse than new APU like the 680m.

What OEM is going to buy a discrete GPU when when the iGPU/APU already present is of similar performance.

yeah it seems more and more that the need for smaller, ultralight and medium laptops is more and more served by standard APUs these days.

The ROG and ALIENWARE and whatever uber gaming brick laptops for the ~$3k+ and strange heat and power choices market...is really the only rational niche audience for those type of mobile IGPs, and best served by flagship parts anyway like the x700-x900 series cards from the manufacturers. ...because you're already paying 3 grand for laptop, right?

 

[MrTeal]

Not seeing anyone suggest APUs are a bad idea.

Just that they aren't going to make Big Console size APUs.

AMDs current balance looks about right for the available BW, and you really can't expect more than that.

Console like APUs would actually be great IMO, you just can't get it in the typical dual channel mainstream ATX system. Maybe not for the typical AT GPU forum user, but if a Tier 1 OEM produced an ITX sized prebuilt with specs similar to a PS5 (maybe a little more RAM) along with maybe an extra PCIe, m.2 and 2.5" SATA port that would be a winner.
Hell, even being able to just buy a PS5 even without any expansion for $7-800 and a copy of Windows would be a pretty great value for 90% of people who want a PC for mid tier gaming and work/school.

 

[insertcarehere]

Console like APUs would actually be great IMO, you just can't get it in the typical dual channel mainstream ATX system. Maybe not for the typical AT GPU forum user, but if a Tier 1 OEM produced an ITX sized prebuilt with specs similar to a PS5 (maybe a little more RAM) along with maybe an extra PCIe, m.2 and 2.5" SATA port that would be a winner.
Hell, even being able to just buy a PS5 even without any expansion for $7-800 and a copy of Windows would be a pretty great value for 90% of people who want a PC for mid tier gaming and work/school.

Intel and AMD have tried this concept before already. Kaby Lake-G was an APU with 1060 max-q level performance and launched in early 2018, with both NUCs and laptops, it's still technically the fastest IGP today. Neither party (especially AMD) pursued this concept any further, most likely for good reason.

Its all well and good saying that a big gaming-focused APU is great for consumers when ultimately, not enough of said consumers purchase said big gaming-focused APU at pricing which would makes sense for AMD/Intel to develop such a product.

 

[biostud]

Here is the latest from the rumor mill/clickbait land:

Dismal AMD RDNA 3 refresh rumor suggests all RDNA 3+ RX 7000 SKUs have been canned - NotebookCheck.net News

 

[moinmoin]

Rumors are always fun whenever they first suggest something new will happen and later on suggest that that has been cancelled again.

All I want is a post mortem on how AMD could be so far off with its projections for RDNA3.

 

[leoneazzurro]

While a refresh may or may not come, the rumor seems quite bullshit as it was said many times that RDNA3+ was not for discrete graphic cards but it was only for APUs and thus the supposedly "fixed" cards would be still RDNA3.

 

[biostud]

While a refresh may or may not come, the rumor seems quite bullshit as it was said many times that RDNA3+ was not for discrete graphic cards but it was only for APUs and thus the supposedly "fixed" cards would be still RDNA3.

I agree, I just wonder why we know so little about N32...