Monday, February 14th 2022

Intel Sapphire Rapids Xeon with DDR5 Memory Spotted in AIDA64 and Cinebench R15

Intel's next-generation Xeon processors code-named Sapphire Rapids are on track to hit the market this year. These new processors are supposed to bring a wide array of new and improved features and a chance for Intel to show off its 10 nm SuperFin manufacturing process in the server market. Thanks to the Twitter user YuuKi_AnS, we have some of the first tests run in AIDA64 and Cinebench R15 benchmark suites. Yuuki managed to get ahold of DDR5-enabled Sapphire Rapids Xeon with 48 cores and 96 threads, equipped with a base frequency of 2.3 GHz and boost speeds of 3.3 GHz. The processor tested was an engineering sample with a Q-SPEC designation of "QYFQ" and made for Intel Socket E (LGA-4677). This CPU sample was locked at 270 Watt TDP.

Below, you can see the performance results of this processor, tested in the AIDA64 cache and memory benchmark and Cinebench R15 bench test. There is a comparison between AMD's Milan-X and Xeon Platinum 8380, so the numbers are more in check of what you can expect from the final product.
AIDA64:
Cinebench R15:
Sources: Yuuki_Ans, via VideoCardz
Add your own comment

6 Comments on Intel Sapphire Rapids Xeon with DDR5 Memory Spotted in AIDA64 and Cinebench R15

#1
Crackong
2x 7773X = 128 cores and 10160 MT points only ?
And 118 points single threaded? That is first gen EPYC level.

What have they done to that poor 7773X to make it that low ?
Posted on Reply
#2
usiname
Crackong2x 7773X = 128 cores and 10160 MT points only ?
And 118 points single threaded? That is first gen EPYC level.

What have they done to that poor 7773X to make it that low ?
Intel's special treatment
Posted on Reply
#3
Crackong
usinameIntel's special treatment
Some digging,
From Guru3D's 5950x review, 5950x had 274 points in R15 ST.
So, 274 points at 4.9GHz
So each GHz give 55.92 points
Now 118 / 55.92 = 2.11
So, if assume R15 ST scale linear, that 7773x was locked at 2.1 GHz in R15 ST test.
From the leaks, 7773x should be a 2.2 - 3.5GHz CPU.

Now , for the Ice Lake Xeon 8380
The only Ice Lake counter part was mobile, so take any Ice Lake mobile
Let's take 1065g7
From notebookcheck, 1065g7 had 190 points (max) in R15 ST
So, 190 points at 3.9GHz
Each GHz gives 48.72 points
Now 155/48.72 = 3.18
So, the Ice Lake 8380 was running at 3.2GHz which is, only 0.2GHz below its maximum boost (3.4GHz)

As for the Sapphire Rapids
The closest counterpart was the Alder Lake running the same Golden Cove cores as P cores
But there are surprisingly small amount of 12900k reviews running R15.
From some random videos / article the max was ~285 points in R15 ST.
So, 285 points for 5.2Ghz
So each GHz give 54.81 points
Now 149 / 54.81 = 2.72
So the Sapphire Rapids should be running 2.7 GHz
From the picture it is a 1.5GHz to 3.3GHz CPU

So in conclusion,
This "leak" was comparing 2 Intel CPUs in boost state vs an underclocked AMD CPU.
Noice :)

Edit: Just noticed that super weird RAM setting (2866 CL23) used for the 7773x , what's wrong with that ?
Posted on Reply
#4
usiname
CrackongSome digging,
From Guru3D's 5950x review, 5950x had 274 points in R15 ST.
So, 274 points at 4.9GHz
So each GHz give 55.92 points
Now 118 / 55.92 = 2.11
So, if assume R15 ST scale linear, that 7773x was locked at 2.1 GHz in R15 ST test.
From the leaks, 7773x should be a 2.2 - 3.5GHz CPU.
No, you forget the SMT - 4.6Ghz 5900x score 3900 in R15
www.msi.com/blog/amd-r9-5900x-overclocking-guide
3900/12 = 325 per core with SMT at 4.6ghz
325/4.6 = 70.65 per core with SMT at 1ghz
70.65*128 = 9043 for 128 cores with SMT at 1ghz
9043+12.5% = 10173
so the Epycs were clocked at 1.125 Ghz
Posted on Reply
#5
Crackong
usinameNo, your calculation is wrong, also you forget the SMT - 4.6Ghz 5900x score 3900 in R15
www.msi.com/blog/amd-r9-5900x-overclocking-guide
3900/12 = 325 per core with SMT at 4.6ghz
325/4.6 = 70.65 per core with SMT at 1ghz
70.65*128 = 9043
9043+12.5% = 10173
so the Epycs were clocked at 1.125 Ghz
I was checking the Single thread results only, since the ST result problem are very noticeable.

And,
R15 is pretty much broken with so much threads available
Posted on Reply
#6
Tomorrow
Also SR is using DDR5. Most likely 4800 minimum where as Milan-X is using DDR4. That alone makes a big difference in AIDA64 bandwidth numbers.
Posted on Reply
May 21st, 2024 16:31 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts