AMD Zen 4c Not an E-core, 35% Smaller than Zen 4, but with Identical IPC

[Minus Infinity]

As someone that does more productivity than gaming, it is IMO a huge mistake for AMD to be sticking to it's lower core counts than Intel on desktop especially now they have a great compact core in the 4c. I would have much preferred a 7850X with 8 Zen 4 and 4 Zen 4c cores any day over a v-cache 7800X3D and it would take the fight to the 13700K. 4 Zen 4c cores would be easily competitive with 8 Gracemont cores. And they would cause less scheduler issues as they'll look like regular cores. That Zen 5 is also sticking to same pattern as Zen 4 is even worse. IMO the performance gap between Zen 5 and Arrow Lake will be even larger for MT.

 

[Tomorrow]

As someone that does more productivity than gaming, it is IMO a huge mistake for AMD to be sticking to it's lower core counts than Intel on desktop especially now they have a great compact core in the 4c. I would have much preferred a 7850X with 8 Zen 4 and 4 Zen 4c cores any day over a v-cache 7800X3D and it would take the fight to the 13700K. 4 Zen 4c cores would be easily competitive with 8 Gracemont cores. And they would cause less scheduler issues as they'll look like regular cores. That Zen 5 is also sticking to same pattern as Zen 4 is even worse. IMO the performance gap between Zen 5 and Arrow Lake will be even larger for MT.

Why does it have to be either/or?

7850X could exist for those who need MT and 7800X3D exists for those who want the best gaming performance.
I would hate 7800X3D not existing.

 

[Scrizz]

Maybe 4c stands for Zen 4 ‘compact’. Siena might introduce Zen 4e as an E-core like architecture. AFAIK, Intel’s E-core removes basic functional blocks to achieve a smaller footprint. Zen 4c does not remove any basic or even specialized (bfloat, VINNI, AVX512) functional blocks.

I'm wondering when we'll get a 4c tactical

 

[InVasMani]

Why does it have to be either/or?

7850X could exist for those who need MT and 7800X3D exists for those who want the best gaming performance.
I would hate 7800X3D not existing.

I agree there is reason to celebrate both points of convergence and divergence from the directions AMD/INTEL take with their CPU design architectures it just means a lot more options for everyone.

 

[Squared]

Zen4c would be a much better efficiency core if AMD decides to beat intel at its own game.

SemiAnalysis also said that Zen4 has hard lines between various parts of the core, separating areas different teams worked on. Zen4c eliminates many of these lines so that the different parts can be pushed closer together. But I suspect this also means that Zen4c started development late in the Zen4 development process, and is the primary reason why the first Zen4 chiplet came out last year but the first Zen4c chiplet isn't quite out yet.

On the other hand, Intel's Redwood Cove P cores and Crestmont E cores will be released on the same day when Meteor Lake comes to market. If AMD wants to build a hybrid processor like this, it's probably either going to be six months later than a non-hyrbrid, or AMD will need a different kind of small core that can be developed concurrently with Zen.

 

[Minus Infinity]

Why does it have to be either/or?

7850X could exist for those who need MT and 7800X3D exists for those who want the best gaming performance.
I would hate 7800X3D not existing.

I didn't mean to imply either/or. I just said "I" would have preferred a productivity optimised version. Hence the different name. I have zeroo interest in the v-cache models other than they are more efficient. Now if they could give us a 3rd gen with basically no MT penalty due to lower clocks and v-cache across both ccd's that would be different.

 

[doc7000]

As someone that does more productivity than gaming, it is IMO a huge mistake for AMD to be sticking to it's lower core counts than Intel on desktop especially now they have a great compact core in the 4c. I would have much preferred a 7850X with 8 Zen 4 and 4 Zen 4c cores any day over a v-cache 7800X3D and it would take the fight to the 13700K. 4 Zen 4c cores would be easily competitive with 8 Gracemont cores. And they would cause less scheduler issues as they'll look like regular cores. That Zen 5 is also sticking to same pattern as Zen 4 is even worse. IMO the performance gap between Zen 5 and Arrow Lake will be even larger for MT.

If they were to make a CPU using both Zen 4 and Zen 4c then it would have an 8 core zen 4 chiplet and a 16 core Zen4c chiplet on it for 24 cores and 48 threads, we don't know how this will compare against the current 16 core 7950X as we don't know the real world performance loss for non data center workloads of the Zen 4c chiplets. If the smaller Zen 4c delivers half the performance of a Zen 4 core on productivity workloads then it would be pointless.

IMHO Ryzen 9 should have all been 16 core cpus, with Ryzen 7 being 12 core cpus, Ryzen 5 being 8 core cpus, and Ryzen 3 being 6 core cpus. Sell the cheapest 12 core for around $450 with the cheapest 8 core at $300 and the cheapest 6 core at $200 and lets say $600 for the cheapest 16 core cpu. This would have done a much better job keeping core counts and overall performance much closer when comparing Intel and AMD offerings this time around. Though considering that right now you can get a 7600 for $223, a 7700 for $326, a 7900 for $420, and a 7950X for $576 dollars in the US the pricing is almost there anyways.

 

[R0H1T]

Ryzen 5 being 8 core cpus, and Ryzen 3 being 6 core cpus

There will always be a demand for cheap quad or dual cores, even if just for using it to watch the latest Netflix/Disney trash. That demand isn't going away soon & as long as we have rejects/defective chips being sold we'll continue to see them being listed for 50~100 USD. So that seems unreasonable as far as I'm concerned, 2-3 years down the line when Intel/AMD just eliminate dual cores from the lineup I guess that could make sense but not right now.

 

[zlobby]

Maybe 4c stands for Zen 4 ‘compact’.

My bet is on 'carrera'.

 

[dyonoctis]

There will always be a demand for cheap quad or dual cores, even if just for using it to watch the latest Netflix/Disney trash. That demand isn't going away soon & as long as we have rejects/defective chips being sold we'll continue to see them being listed for 50~100 USD. So that seems unreasonable as far as I'm concerned, 2-3 years down the line when Intel/AMD just eliminate dual cores from the lineup I guess that could make sense but not right now.

AMD already eliminated dual cores from their line-up, and there haven't been a quad core Ryzen since zen 2, the cheapest zen 3 is an hexa core at 94€. The remaining Ryzen 3 can be found for the same price as an hexacore of the same generation for some reason.

 

[AnotherReader]

Little 'c' = little cache

I am pretty surprised at how much space they managed to save by reducing cache though.

Zen 4c actually has more cache than Zen 4 across the entire chiplet. Remember there are 16 Zen4c cores in one chiplet. Despite the reduced L3 per core, this means each chiplet has 32 MB of L3 cache and 16 MB of L2 cache as opposed to regular Zen 4 with 32 MB of L3 and 8 MB of L2.

 

[R0H1T]

AMD already eliminated dual cores from their line-up, and there haven't been a quad core Ryzen since zen 2, the cheapest zen 3 is an hexa core at 94€. The remaining Ryzen 3 can be found for the same price as an hexacore of the same generation for some reason.
View attachment 300945

I was talking about all of their existing lineups, including notebooks, so that's not exactly true ~

TechPowerUp

CPU database with latest specifications for processors launched in recent years. Includes clocks, photos, and compatibility info.

www.techpowerup.com

It will eventually happen as they move almost completely from 7/6nm parts but probably still a couple of years away.

 

[btk2k2]

Guys, the article is about Zen4 c core. Why so much spam?


What is this about?

AMD will probably use the Zen4c CCX layout in PHX2 going forward so they can make the APU smaller and cheaper to produce. It should also improve the low power performance since the v/f curve will be different.

SemiAnalysis also said that Zen4 has hard lines between various parts of the core, separating areas different teams worked on. Zen4c eliminates many of these lines so that the different parts can be pushed closer together. But I suspect this also means that Zen4c started development late in the Zen4 development process, and is the primary reason why the first Zen4 chiplet came out last year but the first Zen4c chiplet isn't quite out yet.

On the other hand, Intel's Redwood Cove P cores and Crestmont E cores will be released on the same day when Meteor Lake comes to market. If AMD wants to build a hybrid processor like this, it's probably either going to be six months later than a non-hyrbrid, or AMD will need a different kind of small core that can be developed concurrently with Zen.

They don't need to run concurrently. AMD can release a full big core part like the 7950X and once the 'c' variant is online can release a new SKU with 1 8c standard CCD and 1 16c dense CCD.

Just like the lag between the 5800X and the 5800X3D was longer than the lag between 7700X and 7800X3D I expect for Zen 5 it will initially be a 16c 2 CCD design and a few months later once the Zen 5c design is finished they can launch a 24c SKU.

In theory they could release a 24c Zen 4 SKU as well, just depends on if there is a business case for such a thing. Personally I think there might be if it gets people to jump onto AM5 before Arrow Lake launches which should improve uptake of Zen 5 onwards.

 

[Od1sseas]

Nobody cares about this. This article is not about your CPU. Learn something about Zen4 c cores.

I'm replying to someone that called me an Intel fanboy. If you didn't actually care you wouldn't have replied to my comment as well.

 

[evolucion8]

I'm replying to someone that called me an Intel fanboy. If you didn't actually care you wouldn't have replied to my comment as well.

Cheering up for a brand won't make it magically better or faster overall.

 

[Tek-Check]

I'm replying to someone that called me an Intel fanboy. If you didn't actually care you wouldn't have replied to my comment as well.

Cut irrelevant discussions and focus on what matters. Don't get rpovoked and let others know what to focus on. Simple.

 

[R-T-B]

@R-T-B just going off of observable Ryzen behaviour in the past 3 generations and its internal temp monitoring, L3 doesn't usually get hot or draw a lot of power. Whether in 16MB, 32MB or Vcache form.

Depends on if your working it. Run small ffts on prime 95 and watch the temps rise.

EDIT: wait, that might just be l1/l2 though. Now I am questioning my logic.

 

[tabascosauz]

Depends on if your working it. Run small ffts on prime 95 and watch the temps rise.

EDIT: wait, that might just be l1/l2 though. Now I am questioning my logic.

L3 (at least the base layer not Vcache, since the temp behaviour does not change for Vermeer-X) shows up as its own temp sensor. It basically never gets hot compared to the rest of the core; whether that's because L3 doesn't get hot or the sensor is not what it seems, who knows.

If L3 had a significant impact on per-core power, I also think we would be able to tell a difference by now between half and full L3 and between full and X3D. Core power is roughly similar at the end of the day.

 

[Altirix]

Intel can pack 4 E-Cores in the same size as 1 P-Core. What about AMD? How many Zen4c cores for one Zen 4 core?

Raptor Cove Core with 2MB L2$ is 7.04mm^2

Gracemont is 1.70mm^2 without L2$, a block of 4 with 4MB L2$ cache is 8.78mm^2 (note the image incorrectly states 2MB for L2$)

Zen 4 with 1MB L2$ is 3.84mm^2

Zen 4C with 1MB L2$ is 2.48mm^2, we can estimate a block of 4 with 4MB L2$ to be about 9.9mm^2 unless ive butchered the math

so it really doesn't matter how many zen4c cores can pack into the area of a zen4 core because intel is getting thrashed in area efficiency.

kinda find it funny this was likely commented in bad faith as you had the numbers to work out they could fit like 1.5 Z4C cores in a Z4 core. which must be an own because Intel could fit 4 cripple cores in the same area as their performance core. turns out Intels P core has awful area efficiency and honestly its the same with their E core, when you consider Z4C, is the same RTL design as Z4 and is about the same size as a Gracemont cluster.

 

[doc7000]

There will always be a demand for cheap quad or dual cores, even if just for using it to watch the latest Netflix/Disney trash. That demand isn't going away soon & as long as we have rejects/defective chips being sold we'll continue to see them being listed for 50~100 USD. So that seems unreasonable as far as I'm concerned, 2-3 years down the line when Intel/AMD just eliminate dual cores from the lineup I guess that could make sense but not right now.

AMD hasn't made a Ryzen 4 core since the 3100 and 3300X CPUs, there may still be demand for 4 core cpus and AMD does sell CPUs under different branding then Ryzen. Same thing with Intel not all of their cpus are sold with core i branding so they can launch lower core count cpus with different branding while keeping the Ryzen where it currently is no less then 6 cores. Compared to CPUs that went on sale back in 2020 and you probably can't even buy brand new.

 

[Squared]

They don't need to run concurrently. AMD can release a full big core part like the 7950X and once the 'c' variant is online can release a new SKU with 1 8c standard CCD and 1 16c dense CCD.

Just like the lag between the 5800X and the 5800X3D was longer than the lag between 7700X and 7800X3D I expect for Zen 5 it will initially be a 16c 2 CCD design and a few months later once the Zen 5c design is finished they can launch a 24c SKU.

In theory they could release a 24c Zen 4 SKU as well, just depends on if there is a business case for such a thing. Personally I think there might be if it gets people to jump onto AM5 before Arrow Lake launches which should improve uptake of Zen 5 onwards.

Sure Zen4c doesn't need to be developed concurrently; but I think as long as AMD's small core is a compacted derivative of the main core, it'll always be a bit later to market and it'll always be limited to fewer precessor models that AMD could afford to wait to release. This will limit how it can compete against Intel's E cores.

 

[InVasMani]

Can't they combine this on one CCX and also use another CCX with X3D and get a nice mix of low latency single thread driven performance with higher multi-thread driven performance at a bit slower latency!? Seems reasonable enough to me.

 

[Mawkzin]

AMD hasn't made a Ryzen 4 core since the 3100 and 3300X CPUs, there may still be demand for 4 core cpus and AMD does sell CPUs under different branding then Ryzen. Same thing with Intel not all of their cpus are sold with core i branding so they can launch lower core count cpus with different branding while keeping the Ryzen where it currently is no less then 6 cores. Compared to CPUs that went on sale back in 2020 and you probably can't even buy brand new.

AMD is selling ryzen 3 5300G 4c/8t, they stop selling the X version because they could make more profit of these 4cores in the datacenter market.

 

[londiste]

Raptor Cove Core with 2MB L2$ is 7.04mm^2
Gracemont is 1.70mm^2 without L2$, a block of 4 with 4MB L2$ cache is 8.78mm^2 (note the image incorrectly states 2MB for L2$)

Zen 4 with 1MB L2$ is 3.84mm^2
Zen 4C with 1MB L2$ is 2.48mm^2, we can estimate a block of 4 with 4MB L2$ to be about 9.9mm^2 unless ive butchered the math

so it really doesn't matter how many zen4c cores can pack into the area of a zen4 core because intel is getting thrashed in area efficiency.

Intel 7 vs TSMC N5 though. So not exactly a direct comparison.

 

[doc7000]

AMD is selling ryzen 3 5300G 4c/8t, they stop selling the X version because they could make more profit of these 4cores in the datacenter market.

They sell that as a cheap APU, they are not selling stand alone 4 core CPUs and I question how many 5300G chips they are even selling. If there was more demand for 4 core CPUs then AMD as well as Intel would offer more of them.

Can't they combine this on one CCX and also use another CCX with X3D and get a nice mix of low latency single thread driven performance with higher multi-thread driven performance at a bit slower latency!? Seems reasonable enough to me.

A much better route for AMD would be to just move to a third chiplet for their CPUs, this would give them up to 24 cores and 48 threads.