Discussion Zen 5 Builders thread

[AdamK47]

wow,7200 1 T I don't believe it !

@1.3 volts no less.
View attachment 106111

FCLK?

 

[Markfw]

That's for sure running 1:2 divider.

Not what the screen says ! CR 1T

G.SKILL Trident Z5 RGB Series 32GB (2 x 16GB) 288-Pin PC RAM DDR5 7200 (PC5 57600) Desktop Memory Model F5-7200J3445G16GX2-TZ5RK​

@igor_kavinski Links for those testing apps ?

 

[Markfw]

FCLK?

if its not on the memory screen, how would I tell. CPUz says almost nothing.

 

[PJVol]

if its not on the memory screen, how would I tell.

Use ZenTimings, very useful standalone app.

ZenTimings

ZenTimings is a simple and lightweight app for monitoring memory timings on Ryzen platform.

zentimings.com

 

[Markfw]

It also says 1T, but only running 2000 FLCK

 

[PJVol]

It also says 1T, but only running 2000 FLCK

FCLK is fully decoupled from UMC/DDR Phy clock domains, though at the cost of BW due to FIFO losses.
The desired FCLK is a trade-off between these losses and the Data Fabric throughput.

And, as has always been the case with MCM Zen memory subsystem, a question of prioriy: latency or bandwidth.

As for the memory speed, there's a MCLK threshold after which the UMC operates at a 1:2 ratio. I believe 3600 is well above that, though Zen4 owners may correct me.

 

[AdamK47]

I've found 1:2 divider memory clocks to be completely useless until around 7800 to 8200 memory speeds.

You lose performance over 1:1 6000+ speeds.

 

[igor_kavinski]

@igor_kavinski Links for those testing apps ?

PM'ed you the links.

 

[Hotrod2go]

My 9700X with 2167 FCLK needs 0.095 less volts to run stable compared to my 7600X with the same FCLK speed in the same motherboard. The MemTestPro test for over 700% coverage in my system with an X670E board & EXPO 5600MT/S ram profile. I tweaked only the tREFI & both SCL's minor timings for that ram profile, but in the greater scheme of things, that's just a minor tweak to the ram itself. The surprising thing I found is the low SOC volts now needed, even with LLC of 3 for SOC which is the lowest LLC my board can do for that bios option.

So consequently, my 9700x runs significantly cooler than before thanks to the SOC not needing as much volts to be stable at 2167 FCLK & all this is with first iteration of AGESA code for Zen 5. Also much easier to sustain 5500MHZ PBO boost on all 8 cores with SMT enabled (auto settings for PBO) continuously during this test with MTP.
(note the slower last thread in MTP is because of Windows 11 23H2 scheduling issues)

To get perspective on some environmental factors influencing temps in this screenshot:
1. Antec C8 case, closed up.
2. Case fan profile on silent except cpu on standard.
3. Deep Cool AK620 tower cooler + Thermal Grizzley Kyronaut extreme TIM
4. RX 7900 RTX idling below the CPU
5. Windows power profile on high performance.

 

[B-Riz]

Need some feedback from the forum; I want to do fixed clock testing, starting with Zen5, then work down to Zen4, Zen3, Zen2 and Zen1.

I have a 7600X, no pure 6 core Zen3, a 3600, but no 6 core Zen1.

Would we be happy if I turned off half the 5900X (or limit a 5800X to 6 cores?) and limited the R7 1700 to 6 cores?

I am thinking the clock would have to be around 3.6 Ghz, I don't think the 1700 can do much more than that without too much voltage; I may look for a late production OG R5 1600 non AF.

Fastest GPU is a 7900 XT, then a 3090, then 3080 Ti.

Thoughts / suggestions?

This will take time to work through though. (could take a month or more of here and there testing)

 

[Det0x]

Seems like a slow Tuesday, so i guess i can share some binning results for my 3x 9950X retail + 1x 9950X ES, now that i have sold all my Zen5 cpus i wont daily (lost ~200$ per cpu on the used-market)

A few words before we get into the data:

• The goal was to complete 3x MT runs in a row in Cinebench R23
• Static cpu voltage was set at 1.200 vcore --> around 1.156 vcore get while running (since set in windows only hwinfo motherboard vcore is correct reading)
• Static clockspeeds was upped on each individual CCDs in 25mhz intervals before it could not pass 3x runs back to back
• I tried to keep the watercooled same loop GPU idle temp as close to 20 degrees as possible for all runs (most reliable for me)
• Non of the runs were done at stock 5600MT/s speeds, but not maxed out either --> was just what i was running at the time of testing
• Nevermind the R23 MT score themself as i had lots of stuff running, lowering the scores. Point of interest was clockspeeds
• HWinfo got reset before each run, all runs seems to be about 256w powerusage while running Cinebench R23 MT
• Stupid windows11 printscreen would only take screenshot the same millisecond the benchmark ended on all but the ES, but values in hwinfo hadn't had time to update

Retail #1 9950X sample
Max stable FCLK = 2167mhz
Max static R23 ST clockspeed @ set 1.4core = 5825mhz
Max completed MT clockspeeds = 5500mhz CCD0 / 5450mhz CCD1


Retail #2 9950X sample
Max stable FCLK = 2200mhz
Max static R23 ST clockspeed @ set 1.4core = 5850mhz
Max completed MT clockspeeds = 5550mhz CCD0 / 5425mhz CCD1


Retail #3 9950X sample
Max stable FCLK = 2200mhz
Max static R23 ST clockspeed @ set 1.4core = 5850mhz
Max completed MT clockspeeds = 5500mhz CCD0 / 5425mhz CCD1


ES 9950X delidded sample
Max stable FCLK = 2200mhz
Max static R23 ST clockspeed @ set 1.4core = 5900mhz
Max completed MT clockspeeds = 5475mhz CCD0 / 5525mhz CCD1


I have binning data for all samples at 1.100 vcore set also, but it pretty much shows the same as with 1.200 vcore set

 

[igor_kavinski]

(lost ~200$ per cpu on the used-market)

You wouldn't have lost any money if you had stamped their IHS "Det0x tested, tortured and still alive"

 

[Det0x]

You wouldn't have lost any money if you had stamped their IHS "Det0x tested, tortured and still alive"

I dont want to have to deal with all the questions from selling a "binned" sample
Take the loss and move on.. have learned from selling Zen4 golden samples cpus

Some scrub with a 100$ motherboard + boxed CPU cooler and some micron memory is wondering why the cant copy my settings 1-to-1 and be rank1 on hwbot, and pretty much accusing me of "scamming him, nevermind he have no idea about anything 🤣

 

[Hail The Brain Slug]

I dont want to have to deal with all the questions from selling a "binned" sample
Take the loss and move on.. have learned from selling Zen4 golden samples cpus

Some scrub with a 100$ motherboard + boxed CPU cooler and some micron memory is wondering why the cant copy my settings 1-to-1 and be rank1 on hwbot, and pretty much accusing me of "scamming him, nevermind he have no idea about anything 🤣

Damn if I knew you were going to sell the "bad" samples for $450 I might have bought one instantly. Especially #2 or #3

 

[Det0x]

My 9950X likes round numbers today it seems
First 50k Cinebench r23 and now 60k vray
(on watercooling)

Domdtxdissar`s Cinebench - R23 Multi Core with BenchMate score: 50004 cb with a Ryzen 9 9950X

The Ryzen 9 9950X @ 5701.3MHzscores getScoreFormatted in the Cinebench - R23 Multi Core with BenchMate benchmark. Domdtxdissarranks #10 worldwide and #7 in the hardware class. Find out more at HWBOT.

hwbot.org

V-Ray 6 Benchmark

benchmark.chaos.com

 

[Det0x]

Small update

Managed ~49k in Cinebench R23 MT if i limits myself to 250w (kinda seems like the performance/power sweet spot for this 9950X)


And as a bonus, a Corona benchmark while i'm at it

 

[Det0x]

ASUS new Nitropath is definitely doing some work!

For the first time 8000MT/s memoryspeeds and above are viable on 2DPC (4 memory slots) motherboard for daily usage, with full boot to boot stability (which is something that have plagued previous gens)
I have had some time to play around with the X870E Hero in my small air-cooled livingroom media computer (overkill i know 😅), and i have achieved full stablity running 8000MT/s together with 2x16GB Hynix A-die. And best of all, this is using all stock/auto memory training settings, no fiddling with taps and RTT+proc settings required

Should also note that this computer is also using the IGPU on the IO-die for display, which is something that makes it even harder on the memory-timings/IMC, ontop of this being a 4dimmer motherboard. 👍

Hardware used:

• 9950X @ fully stock boosting
• X870E Hero
• Noctua Air cooler
• 2200mhz FCLK
• Teamgroup UD-5 7800MT/s XMP a-die memory with my custom barts heatsinks @ 8000MT/s CL34-45-42-48
• This computer have no casefans installed

Results with A-die: (most Zen5 cpus capps out at 8000MT/s speeds while a few can do 8100/8200 for daily in optimal cooling/settings)


Full screenshot with 13 hours karhu + 3 hours testmem 1usmus + hwinfo open



Have also done some testing with Hynix M-die, but profile is not finalized yet
(lax timings)


All in all this motherboard is working flawlessly for me atm, have had zero hiccups, and best of all like i said above, full boot-to-boot stability with all auto memorytraining settings 😉

There are also some new settings in the bios i can share with you guys in the coming days.. CO safe curve anyone ? Basically two different CO curves at the same time, one for light work and autoswitch to a other one for heavy work

 

[BTRY B 529th FA BN]

Nice! My local MC has 7 of these boards currently, @ $749.99

Wish this board would of had a 3rd expansion slot for a sound card.

Very nice! Congrats!

 

[lightmanek]

Another 9950X on the forum

Just started testing stock performance with EXPO enabled.
Some ancient entries included for fun

 

[igor_kavinski]

Another 9950X on the forum

Just started testing stock performance with EXPO enabled.

Rapydmark High run with text log and Maxxmem2 screenshots please. Thanks!

Also, maybe turn off EXPO and try going for DDR5-6000 CL28? Should be easy now with mature kits on the market.

 

[lightmanek]

Rapydmark High run with text log and Maxxmem2 screenshots please. Thanks!

Also, maybe turn off EXPO and try going for DDR5-6000 CL28? Should be easy now with mature kits on the market.

I work a bit today ... 6h, but will get into tweaking once I finish my 87 tests at out of the box settings

Will do these two for you, no problem, but might be tomorrow before I can.

PS. A shame as I had there my tweaked 7800X3D result and accidentally deleted it.
I remember ST score was 3052, so in this old bench, Zen 5 is significantly faster (7800X3D was running 5.25GHz ST)

 

[lightmanek]

Rapydmark High run with text log and Maxxmem2 screenshots please. Thanks!

Also, maybe turn off EXPO and try going for DDR5-6000 CL28? Should be easy now with mature kits on the market.

Here are the numebrs, still stock + EXPO, no other tweaks!

Spoiler

Tests report generated by RapydMark v1.2a (64 bits) https://www.frikiscape.com/rapydmark/
=====================================
Date: 10-05-2024 (m-d-y)
Time: 20:43:41
CPU: AMD Ryzen 9 9950X 16-Core Processor (16 Cores)
Threads available: 32
GPU: AMD Radeon RX 7900 XTX
Resolution: 2560 x 1440
Operating System: Windows 10 Pro Build 22631 (Win11)
System memory: 31882 MB
Stress level: High
System ID: 1D788E16
=====================================
Results of all tests:

Total time (all tests): 156.692 seconds

-------------------------------------
Test Type: Matrix multiplication
Iterations: 400M | Performance: 158,353,120 it/s (32th) | Time: 2.526 s

-------------------------------------
Test Type: MulDivAddSub (int,double,int64_t)
Iterations: 400M | Performance: 77,775,616 it/s (32th) | Time: 5.143 s

-------------------------------------
Test Type: Trigonometric functions
Iterations: 400M | Performance: 110,619,472 it/s (32th) | Time: 3.616 s

-------------------------------------
Test Type: Floating point operations per second
Iterations: 16K | Performance: 4,797 it/s 1127.219 GFLOPS (32th) | Time: 3.335 s

-------------------------------------
Test Type: Find prime numbers SQR method
Iterations: 200M | Performance: 40,290,088 it/s (32th) | Time: 4.964 s

-------------------------------------
Test Type: Fibonachi sequence (92)
Iterations: 4000M | Performance: 1,064,962,688 it/s (32th) | Time: 3.756 s

-------------------------------------
Test Type: Pi approximation Taylor series method
Iterations: 800 | Performance: 237 it/s (32th) | Time: 3.372 s

-------------------------------------
Test Type: Greatest Common Divisor (GCD) Euclid's algorithm
Iterations: 1600M | Performance: 370,456,160 it/s (32th) | Time: 4.319 s

-------------------------------------
Test Type: Fill Memory 256 MB
Iterations: 60K | Performance: 7,289 it/s (32th) | Time: 8.231 s

-------------------------------------
Test Type: Copy Memory 256 MB
Iterations: 60K | Performance: 11,838 it/s (32th) | Time: 5.068 s

-------------------------------------
Test Type: Memory cache performance (max 4 threads)
Iterations: 40 | Performance: 10 it/s (4th) | Time: 15.549 s
Last result: SizeMem= 8 kB Iterations= 65536 Time= 1 ms
SizeMem= 16 kB Iterations= 32768 Time= 1 ms
SizeMem= 32 kB Iterations= 16384 Time= 1 ms
SizeMem= 64 kB Iterations= 8192 Time= 2 ms
SizeMem= 128 kB Iterations= 4096 Time= 2 ms
SizeMem= 256 kB Iterations= 2048 Time= 2 ms
SizeMem= 512 kB Iterations= 1024 Time= 2 ms
SizeMem= 1 MB Iterations= 512 Time= 3 ms
SizeMem= 2 MB Iterations= 256 Time= 8 ms
SizeMem= 4 MB Iterations= 128 Time= 53 ms
SizeMem= 8 MB Iterations= 64 Time= 49 ms
SizeMem= 16 MB Iterations= 32 Time= 51 ms
SizeMem= 32 MB Iterations= 16 Time= 55 ms
SizeMem= 64 MB Iterations= 8 Time= 62 ms
SizeMem= 128 MB Iterations= 4 Time= 86 ms


-------------------------------------
Test Type: Files Encrypt AES256 (10 MB)
Iterations: 20K | Performance: 2,191 it/s (32th) | Time: 9.128 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt AES256 (1 MB)
Iterations: 8K | Performance: 2,566 it/s (32th) | Time: 3.117 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt Blowfish448 (1 MB)
Iterations: 8K | Performance: 3,565 it/s (32th) | Time: 2.244 s

-------------------------------------
Test Type: Memory Compress/Decompress 1 MB (Gzip)
Iterations: 8K | Performance: 3,187 it/s (32th) | Time: 2.510 s

-------------------------------------
Test Type: Files Compress 10 MB (Gzip)
Iterations: 4K | Performance: 1,159 it/s (32th) | Time: 3.449 s

-------------------------------------
Test Type: RSA Keys Generator (2048 bits)
Iterations: 2K | Performance: 539 it/s (32th) | Time: 3.676 s

-------------------------------------
Test Type: Hash SHA1/256/512 MD4/5 CRC32 RIPEMD160 (512 bytes)
Iterations: 1M | Performance: 18,909 it/s (32th) | Time: 52.884 s

-------------------------------------
Test Type: Random Numbers Generator (2048 bits)
Iterations: 200M | Performance: 35,790,980 it/s (32th) | Time: 5.588 s

-------------------------------------
Test Type: Bitcoin Address Generator
Iterations: 200K | Performance: 22,532 it/s (32th) | Time: 8.876 s

-------------------------------------
Test Type: Distances between points on Earth (Haversine)
Iterations: 1000M | Performance: 187,230,848 it/s (32th) | Time: 5.341 s

-------------------------------------
Test Type: Cathedral 3D Test
Iterations: 1 | Performance: FPS average: 278 | min: 133 | max: 1170
(Max threads) | Time: 88.572 s

For comparision, my previous 16 core Ryzen 9 5950X (Zen 4 era was soldiered on 7600 followed by 7800X3D) - mostly doubling of performance in 2 gens

Spoiler

Tests report generated by RapydMark v1.2a (64 bits) https://www.frikiscape.com/rapydmark/
=====================================
Date: 01-15-2022 (m-d-y)
Time: 12:44:41
CPU: AMD Ryzen 9 5950X 16-Core Processor (16 Cores)
Threads available: 32
GPU: AMD Radeon RX 6800 XT
Resolution: 2560 x 1440
Operating System: Windows 10 Pro Build 19044
System memory: 32688 MB
Stress level: High
System ID: AFF24779
=====================================
Results of all tests:

Total time (all tests): 287.552 seconds

-------------------------------------
Test Type: Matrix multiplication
Iterations: 400M | Performance: 78,492,936 it/s (32th) | Time: 5.096 s

-------------------------------------
Test Type: MulDivAddSub (int,double,int64_t)
Iterations: 400M | Performance: 31,665,612 it/s (32th) | Time: 12.632 s

-------------------------------------
Test Type: Trigonometric functions
Iterations: 400M | Performance: 46,948,356 it/s (32th) | Time: 8.520 s

-------------------------------------
Test Type: Floating point operations per second
Iterations: 16K | Performance: 2,680 it/s 554.197 GFLOPS (32th) | Time: 5.968 s

-------------------------------------
Test Type: Find prime numbers SQR method
Iterations: 200M | Performance: 17,793,594 it/s (32th) | Time: 11.240 s

-------------------------------------
Test Type: Fibonachi sequence (92)
Iterations: 4000M | Performance: 376,222,720 it/s (32th) | Time: 10.632 s

-------------------------------------
Test Type: Pi approximation Taylor series method
Iterations: 800 | Performance: 117 it/s (32th) | Time: 6.816 s

-------------------------------------
Test Type: Greatest Common Divisor (GCD) Euclid's algorithm
Iterations: 1600M | Performance: 165,700,080 it/s (32th) | Time: 9.656 s

-------------------------------------
Test Type: Fill Memory 256 MB
Iterations: 60K | Performance: 4,728 it/s (32th) | Time: 12.688 s

-------------------------------------
Test Type: Copy Memory 256 MB
Iterations: 60K | Performance: 5,634 it/s (32th) | Time: 10.648 s

-------------------------------------
Test Type: Memory cache performance (max 4 threads)
Iterations: 40 | Performance: 6 it/s (4th) | Time: 25.168 s
Last result: SizeMem= 8 kB Iterations= 65536 Time= 9 ms
SizeMem= 16 kB Iterations= 32768 Time= 7 ms
SizeMem= 32 kB Iterations= 16384 Time= 7 ms
SizeMem= 64 kB Iterations= 8192 Time= 7 ms
SizeMem= 128 kB Iterations= 4096 Time= 7 ms
SizeMem= 256 kB Iterations= 2048 Time= 8 ms
SizeMem= 512 kB Iterations= 1024 Time= 35 ms
SizeMem= 1 MB Iterations= 512 Time= 62 ms
SizeMem= 2 MB Iterations= 256 Time= 60 ms
SizeMem= 4 MB Iterations= 128 Time= 57 ms
SizeMem= 8 MB Iterations= 64 Time= 57 ms
SizeMem= 16 MB Iterations= 32 Time= 55 ms
SizeMem= 32 MB Iterations= 16 Time= 64 ms
SizeMem= 64 MB Iterations= 8 Time= 77 ms
SizeMem= 128 MB Iterations= 4 Time= 99 ms


-------------------------------------
Test Type: Files Encrypt AES256 (10 MB)
Iterations: 20K | Performance: 1,134 it/s (32th) | Time: 17.632 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt AES256 (1 MB)
Iterations: 8K | Performance: 1,331 it/s (32th) | Time: 6.008 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt Blowfish448 (1 MB)
Iterations: 8K | Performance: 2,083 it/s (32th) | Time: 3.840 s

-------------------------------------
Test Type: Memory Compress/Decompress 1 MB (Gzip)
Iterations: 8K | Performance: 1,703 it/s (32th) | Time: 4.696 s

-------------------------------------
Test Type: Files Compress 10 MB (Gzip)
Iterations: 4K | Performance: 637 it/s (32th) | Time: 6.272 s

-------------------------------------
Test Type: RSA Keys Generator (2048 bits)
Iterations: 2K | Performance: 234 it/s (32th) | Time: 8.448 s

-------------------------------------
Test Type: Hash SHA1/256/512 MD4/5 CRC32 RIPEMD160 (512 bytes)
Iterations: 1M | Performance: 12,937 it/s (32th) | Time: 77.296 s

-------------------------------------
Test Type: Random Numbers Generator (2048 bits)
Iterations: 200M | Performance: 15,862,944 it/s (32th) | Time: 12.608 s

-------------------------------------
Test Type: Bitcoin Address Generator
Iterations: 200K | Performance: 10,373 it/s (32th) | Time: 19.280 s

-------------------------------------
Test Type: Distances between points on Earth (Haversine)
Iterations: 1000M | Performance: 80,593,168 it/s (32th) | Time: 12.408 s

-------------------------------------
Test Type: Cathedral 3D Test
Iterations: 1 | Performance: 0 | Time: 0

 

[lightmanek]

UPDATE with slight memory tweak + PPT 230W CO -15 and FCLK 103 Sync.

Spoiler

Tests report generated by RapydMark v1.2a (64 bits) https://www.frikiscape.com/rapydmark/
=====================================
Date: 10-05-2024 (m-d-y)
Time: 21:48:23
CPU: AMD Ryzen 9 9950X 16-Core Processor (16 Cores)
Threads available: 32
GPU: AMD Radeon RX 7900 XTX
Resolution: 2560 x 1440
Operating System: Windows 10 Pro Build 22631 (Win11)
System memory: 31882 MB
Stress level: High
System ID: 1D788E16
=====================================
Results of all tests:

Total time (all tests): 151.454 seconds

-------------------------------------
Test Type: Matrix multiplication
Iterations: 400M | Performance: 226,628,896 it/s (32th) | Time: 1.765 s

-------------------------------------
Test Type: MulDivAddSub (int,double,int64_t)
Iterations: 400M | Performance: 80,857,088 it/s (32th) | Time: 4.947 s

-------------------------------------
Test Type: Trigonometric functions
Iterations: 400M | Performance: 111,141,984 it/s (32th) | Time: 3.599 s

-------------------------------------
Test Type: Floating point operations per second
Iterations: 16K | Performance: 4,844 it/s 1143.054 GFLOPS (32th) | Time: 3.303 s

-------------------------------------
Test Type: Find prime numbers SQR method
Iterations: 200M | Performance: 40,908,160 it/s (32th) | Time: 4.889 s

-------------------------------------
Test Type: Fibonachi sequence (92)
Iterations: 4000M | Performance: 1,082,544,000 it/s (32th) | Time: 3.695 s

-------------------------------------
Test Type: Pi approximation Taylor series method
Iterations: 800 | Performance: 237 it/s (32th) | Time: 3.367 s

-------------------------------------
Test Type: Greatest Common Divisor (GCD) Euclid's algorithm
Iterations: 1600M | Performance: 379,957,248 it/s (32th) | Time: 4.211 s

-------------------------------------
Test Type: Fill Memory 256 MB
Iterations: 60K | Performance: 7,774 it/s (32th) | Time: 7.718 s

-------------------------------------
Test Type: Copy Memory 256 MB
Iterations: 60K | Performance: 12,804 it/s (32th) | Time: 4.686 s

-------------------------------------
Test Type: Memory cache performance (max 4 threads)
Iterations: 40 | Performance: 10 it/s (4th) | Time: 15.197 s
Last result: SizeMem= 8 kB Iterations= 65536 Time= 1 ms
SizeMem= 16 kB Iterations= 32768 Time= 1 ms
SizeMem= 32 kB Iterations= 16384 Time= 1 ms
SizeMem= 64 kB Iterations= 8192 Time= 2 ms
SizeMem= 128 kB Iterations= 4096 Time= 2 ms
SizeMem= 256 kB Iterations= 2048 Time= 2 ms
SizeMem= 512 kB Iterations= 1024 Time= 2 ms
SizeMem= 1 MB Iterations= 512 Time= 3 ms
SizeMem= 2 MB Iterations= 256 Time= 7 ms
SizeMem= 4 MB Iterations= 128 Time= 43 ms
SizeMem= 8 MB Iterations= 64 Time= 43 ms
SizeMem= 16 MB Iterations= 32 Time= 49 ms
SizeMem= 32 MB Iterations= 16 Time= 54 ms
SizeMem= 64 MB Iterations= 8 Time= 70 ms
SizeMem= 128 MB Iterations= 4 Time= 85 ms


-------------------------------------
Test Type: Files Encrypt AES256 (10 MB)
Iterations: 20K | Performance: 2,633 it/s (32th) | Time: 7.594 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt AES256 (1 MB)
Iterations: 8K | Performance: 2,659 it/s (32th) | Time: 3.008 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt Blowfish448 (1 MB)
Iterations: 8K | Performance: 3,605 it/s (32th) | Time: 2.219 s

-------------------------------------
Test Type: Memory Compress/Decompress 1 MB (Gzip)
Iterations: 8K | Performance: 3,277 it/s (32th) | Time: 2.441 s

-------------------------------------
Test Type: Files Compress 10 MB (Gzip)
Iterations: 4K | Performance: 1,240 it/s (32th) | Time: 3.225 s

-------------------------------------
Test Type: RSA Keys Generator (2048 bits)
Iterations: 2K | Performance: 508 it/s (32th) | Time: 3.902 s

-------------------------------------
Test Type: Hash SHA1/256/512 MD4/5 CRC32 RIPEMD160 (512 bytes)
Iterations: 1M | Performance: 19,232 it/s (32th) | Time: 51.995 s

-------------------------------------
Test Type: Random Numbers Generator (2048 bits)
Iterations: 200M | Performance: 36,390,104 it/s (32th) | Time: 5.496 s

-------------------------------------
Test Type: Bitcoin Address Generator
Iterations: 200K | Performance: 22,899 it/s (32th) | Time: 8.734 s

-------------------------------------
Test Type: Distances between points on Earth (Haversine)
Iterations: 1000M | Performance: 183,049,616 it/s (32th) | Time: 5.463 s

-------------------------------------
Test Type: Cathedral 3D Test
Iterations: 1 | Performance: FPS average: 285 | min: 136 | max: 1267
(Max threads) | Time: 88.505 s

This is safe for my memory kit (M-Die), but as for CPU, I need more time to find it's limits. So far everything is easy, I just re-applied my 7800X3D settings and it works like a treat. Memory controller is definitely better on this sample compared to 7800X3D. So far very happy with my upgrade!

 

[igor_kavinski]

Here are the numebrs, still stock + EXPO, no other tweaks!

The Maxxmem2 score is good. But in Rapydmark, it seems the test "Hash SHA1/256/512 MD4/5 CRC32 RIPEMD160 (512 bytes)" is giving your 9950X quite a bit of trouble. Are you sure you are on the latest AGESA?

 

[igor_kavinski]

UPDATE with slight memory tweak + PPT 230W CO -15 and FCLK 103 Sync.

Good improvement!

However, something is still a bit off with the Hash test in Rapydmark. For reference, a stock 7950X with DDR5-6400 is able to finish it in about 30 secs while my Zen 2 128 thread Epyc takes about the same time as your 9950X. I think your 9950X should be able to complete this test in 28 secs or less if some mysterious setting is set right.