RapydMark CPU benchmark

[igor_kavinski]

RapydMark for Windows (frikiscape.com)

Synthetic highly parallel benchmark.




My company's dual socket Ivy Bridge-E beat Comet Lake in total time, only due to more cores. Should be fun to see how 5950X and i9-12900K fare in this benchmark. My guess is that 5950X will win.

Update: 5950X did win! But i9-12900K put up a good fight despite fewer cores.

Performance comparison chart so far thanks to Makaveli:



Public view-only link for the sheet: RapydMark64 - Google Drive

CCD comparison by Det0x:

1 CCD -> 8 cores / 8 threads = 368.309 seconds
1 CCD -> 8 cores / 16 threads = 230.797 seconds
2 CCD's -> 16 cores / 32 threads = 135.978 seconds

Interesting HT/no-HT comparison:

User​	CPU​	Cores​	Threads​	Time​
JoeRambo​	i9-12900K​	8​	16​	212.656​
JoeRambo​	5800X​	8​	16​	263.424​
Det0x​	5950X​	8​	16​	230.797​
Det0x​	5950X​	8​	8​	368.309​
JoeRambo​	i9-12900K​	8​	8​	328.88​

 

[clemsyn]

Need to edit the result of my overclocked OC to 5400. I had something running on the backround that might have affected the result. Fresh from boot, here is the result of my memory OC'd to 5400

 

[Ed1]

Need to edit the result of my overclocked OC to 5400. I had something running on the backround that might have affected the result. Fresh from boot, here is the result of my memory OC'd to 5400

View attachment 53585

I think this test likes fresh boot my score went up too (time lower 254.442)

comparing the memory test from your DDR5 to my DDR4 we can see big changes in the first few memory tests.

File memory - DDR5 =25.805, DDR4 = 38.418
Copy memory- DDR5 =16.752, DDR4 =22.326
memory cache performance - DDR5 =12.600, DDR4 =14.061
Memory Encrypt/Decrypt AES256- DDR5 =9.148, DDR4 =9.687
Memory Encrypt/Decrypt Blowfish448- DDR5 =6.186, DDR4 =6.249
Memory Compress/Decompress 1 MB - DDR5 =7.726, DDR4 =7.703

 

[Zucker2k]

I also did 8 no HT test, even went to CL13, so we can do as Apples to Apples now, so pretty much 5Ghz vs 4.975Ghz for AMD
View attachment 53583

View attachment 53584


So 329s vs 368s with no HT.

Also the mysteries with Matrix Multiplication and Memory Cache performance remain, should not be THAT different.

DDR5 should be even faster.

 

[Leeea]

5900x
pbo2 = yes
overclocked = "no"
speed = 4.4 - 4.8 Ghz ( it spends most of the tests at 4.4 )
dram = 1780 x 2 Mhz = 3600 16 18 18 38 58 1T
user = um?

Spoiler: report file 195.968

Tests report generated by RapydMark v1.2a (64 bits) https://www.frikiscape.com/rapydmark/
=====================================
Date: 11-29-2021 (m-d-y)
Time: 16:12:33
CPU: AMD Ryzen 9 5900X 12-Core Processor (12 Cores)
Threads available: 24
GPU: AMD Radeon RX 6900 XT
Resolution: 2560 x 1440
Operating System: Windows 10 Pro Build 19042
System memory: 32667 MB
Stress level: High
System ID: 9A6B9B34
=====================================
Results of all tests:

Total time (all tests): 195.968 seconds

-------------------------------------
Test Type: Matrix multiplication
Iterations: 400M | Performance: 142,045,456 it/s (24th) | Time: 2.816 s

-------------------------------------
Test Type: MulDivAddSub (int,double,int64_t)
Iterations: 400M | Performance: 44,326,240 it/s (24th) | Time: 9.024 s

-------------------------------------
Test Type: Trigonometric functions
Iterations: 400M | Performance: 60,096,152 it/s (24th) | Time: 6.656 s

-------------------------------------
Test Type: Floating point operations per second
Iterations: 16K | Performance: 3,619 it/s 701.004 GFLOPS (24th) | Time: 4.416 s

-------------------------------------
Test Type: Find prime numbers SQR method
Iterations: 200M | Performance: 23,496,240 it/s (24th) | Time: 8.512 s

-------------------------------------
Test Type: Fibonachi sequence (92)
Iterations: 4000M | Performance: 512,295,072 it/s (24th) | Time: 7.808 s

-------------------------------------
Test Type: Pi approximation Taylor series method
Iterations: 800 | Performance: 158 it/s (24th) | Time: 4.992 s

-------------------------------------
Test Type: Greatest Common Divisor (GCD) Euclid's algorithm
Iterations: 1600M | Performance: 203,252,032 it/s (24th) | Time: 7.872 s

-------------------------------------
Test Type: Fill Memory 256 MB
Iterations: 60K | Performance: 3,972 it/s (24th) | Time: 15.104 s

-------------------------------------
Test Type: Copy Memory 256 MB
Iterations: 60K | Performance: 4,464 it/s (24th) | Time: 13.440 s

-------------------------------------
Test Type: Memory cache performance (max 4 threads)
Iterations: 40 | Performance: 6 it/s (4th) | Time: 24.128 s
Last result: SizeMem= 8 kB Iterations= 65536 Time= 5 ms
SizeMem= 16 kB Iterations= 32768 Time= 4 ms
SizeMem= 32 kB Iterations= 16384 Time= 4 ms
SizeMem= 64 kB Iterations= 8192 Time= 5 ms
SizeMem= 128 kB Iterations= 4096 Time= 5 ms
SizeMem= 256 kB Iterations= 2048 Time= 11 ms
SizeMem= 512 kB Iterations= 1024 Time= 31 ms
SizeMem= 1 MB Iterations= 512 Time= 63 ms
SizeMem= 2 MB Iterations= 256 Time= 58 ms
SizeMem= 4 MB Iterations= 128 Time= 59 ms
SizeMem= 8 MB Iterations= 64 Time= 59 ms
SizeMem= 16 MB Iterations= 32 Time= 62 ms
SizeMem= 32 MB Iterations= 16 Time= 64 ms
SizeMem= 64 MB Iterations= 8 Time= 73 ms
SizeMem= 128 MB Iterations= 4 Time= 91 ms


-------------------------------------
Test Type: Files Encrypt AES256 (10 MB)
Iterations: 20K | Performance: 1,509 it/s (24th) | Time: 13.248 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt AES256 (1 MB)
Iterations: 8K | Performance: 1,372 it/s (24th) | Time: 5.824 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt Blowfish448 (1 MB)
Iterations: 8K | Performance: 2,270 it/s (24th) | Time: 3.520 s

-------------------------------------
Test Type: Memory Compress/Decompress 1 MB (Gzip)
Iterations: 8K | Performance: 1,951 it/s (24th) | Time: 4.096 s

-------------------------------------
Test Type: Files Compress 10 MB (Gzip)
Iterations: 4K | Performance: 808 it/s (24th) | Time: 4.928 s

-------------------------------------
Test Type: RSA Keys Generator (2048 bits)
Iterations: 2K | Performance: 327 it/s (24th) | Time: 6.080 s

-------------------------------------
Test Type: Hash SHA1/256/512 MD4/5 CRC32 RIPEMD160 (512 bytes)
Iterations: 1M | Performance: 43,402 it/s (24th) | Time: 23.040 s

-------------------------------------
Test Type: Random Numbers Generator (2048 bits)
Iterations: 200M | Performance: 20,559,212 it/s (24th) | Time: 9.728 s

-------------------------------------
Test Type: Bitcoin Address Generator
Iterations: 200K | Performance: 18,711 it/s (24th) | Time: 10.688 s

-------------------------------------
Test Type: Distances between points on Earth (Haversine)
Iterations: 1000M | Performance: 99,522,288 it/s (24th) | Time: 10.048 s

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

Spoiler: report file 196.096

Tests report generated by RapydMark v1.2a (64 bits) https://www.frikiscape.com/rapydmark/
=====================================
Date: 11-29-2021 (m-d-y)
Time: 16:17:13
CPU: AMD Ryzen 9 5900X 12-Core Processor (12 Cores)
Threads available: 24
GPU: AMD Radeon RX 6900 XT
Resolution: 2560 x 1440
Operating System: Windows 10 Pro Build 19042
System memory: 32667 MB
Stress level: High
System ID: 9A6B9B34
=====================================
Results of all tests:

Total time (all tests): 196.096 seconds

-------------------------------------
Test Type: Matrix multiplication
Iterations: 400M | Performance: 223,214,272 it/s (24th) | Time: 1.792 s

-------------------------------------
Test Type: MulDivAddSub (int,double,int64_t)
Iterations: 400M | Performance: 41,946,308 it/s (24th) | Time: 9.536 s

-------------------------------------
Test Type: Trigonometric functions
Iterations: 400M | Performance: 61,274,512 it/s (24th) | Time: 6.528 s

-------------------------------------
Test Type: Floating point operations per second
Iterations: 16K | Performance: 3,619 it/s 699.855 GFLOPS (24th) | Time: 4.416 s

-------------------------------------
Test Type: Find prime numbers SQR method
Iterations: 200M | Performance: 23,854,962 it/s (24th) | Time: 8.384 s

-------------------------------------
Test Type: Fibonachi sequence (92)
Iterations: 4000M | Performance: 516,528,928 it/s (24th) | Time: 7.744 s

-------------------------------------
Test Type: Pi approximation Taylor series method
Iterations: 800 | Performance: 156 it/s (24th) | Time: 5.056 s

-------------------------------------
Test Type: Greatest Common Divisor (GCD) Euclid's algorithm
Iterations: 1600M | Performance: 200,000,000 it/s (24th) | Time: 8.000 s

-------------------------------------
Test Type: Fill Memory 256 MB
Iterations: 60K | Performance: 3,922 it/s (24th) | Time: 15.296 s

-------------------------------------
Test Type: Copy Memory 256 MB
Iterations: 60K | Performance: 4,464 it/s (24th) | Time: 13.440 s

-------------------------------------
Test Type: Memory cache performance (max 4 threads)
Iterations: 40 | Performance: 6 it/s (4th) | Time: 24.320 s
Last result: SizeMem= 8 kB Iterations= 65536 Time= 5 ms
SizeMem= 16 kB Iterations= 32768 Time= 5 ms
SizeMem= 32 kB Iterations= 16384 Time= 5 ms
SizeMem= 64 kB Iterations= 8192 Time= 5 ms
SizeMem= 128 kB Iterations= 4096 Time= 4 ms
SizeMem= 256 kB Iterations= 2048 Time= 7 ms
SizeMem= 512 kB Iterations= 1024 Time= 35 ms
SizeMem= 1 MB Iterations= 512 Time= 65 ms
SizeMem= 2 MB Iterations= 256 Time= 57 ms
SizeMem= 4 MB Iterations= 128 Time= 52 ms
SizeMem= 8 MB Iterations= 64 Time= 51 ms
SizeMem= 16 MB Iterations= 32 Time= 52 ms
SizeMem= 32 MB Iterations= 16 Time= 56 ms
SizeMem= 64 MB Iterations= 8 Time= 66 ms
SizeMem= 128 MB Iterations= 4 Time= 92 ms


-------------------------------------
Test Type: Files Encrypt AES256 (10 MB)
Iterations: 20K | Performance: 1,432 it/s (24th) | Time: 13.952 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt AES256 (1 MB)
Iterations: 8K | Performance: 1,452 it/s (24th) | Time: 5.504 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt Blowfish448 (1 MB)
Iterations: 8K | Performance: 2,356 it/s (24th) | Time: 3.392 s

-------------------------------------
Test Type: Memory Compress/Decompress 1 MB (Gzip)
Iterations: 8K | Performance: 1,921 it/s (24th) | Time: 4.160 s

-------------------------------------
Test Type: Files Compress 10 MB (Gzip)
Iterations: 4K | Performance: 750 it/s (24th) | Time: 5.312 s

-------------------------------------
Test Type: RSA Keys Generator (2048 bits)
Iterations: 2K | Performance: 327 it/s (24th) | Time: 6.080 s

-------------------------------------
Test Type: Hash SHA1/256/512 MD4/5 CRC32 RIPEMD160 (512 bytes)
Iterations: 1M | Performance: 43,402 it/s (24th) | Time: 23.040 s

-------------------------------------
Test Type: Random Numbers Generator (2048 bits)
Iterations: 200M | Performance: 20,973,154 it/s (24th) | Time: 9.536 s

-------------------------------------
Test Type: Bitcoin Address Generator
Iterations: 200K | Performance: 19,054 it/s (24th) | Time: 10.496 s

-------------------------------------
Test Type: Distances between points on Earth (Haversine)
Iterations: 1000M | Performance: 98,892,400 it/s (24th) | Time: 10.112 s

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

Spoiler: report file 193.655 (after reboot of computer)

Tests report generated by RapydMark v1.2a (64 bits) https://www.frikiscape.com/rapydmark/
=====================================
Date: 11-29-2021 (m-d-y)
Time: 16:49:42
CPU: AMD Ryzen 9 5900X 12-Core Processor (12 Cores)
Threads available: 24
GPU: AMD Radeon RX 6900 XT
Resolution: 2560 x 1440
Operating System: Windows 10 Pro Build 19042
System memory: 32667 MB
Stress level: High
System ID: 9A6B9B34
=====================================
Results of all tests:

Total time (all tests): 193.655 seconds

-------------------------------------
Test Type: Matrix multiplication
Iterations: 400M | Performance: 150,545,728 it/s (24th) | Time: 2.657 s

-------------------------------------
Test Type: MulDivAddSub (int,double,int64_t)
Iterations: 400M | Performance: 44,717,720 it/s (24th) | Time: 8.945 s

-------------------------------------
Test Type: Trigonometric functions
Iterations: 400M | Performance: 59,453,032 it/s (24th) | Time: 6.728 s

-------------------------------------
Test Type: Floating point operations per second
Iterations: 16K | Performance: 3,666 it/s 703.094 GFLOPS (24th) | Time: 4.360 s

-------------------------------------
Test Type: Find prime numbers SQR method
Iterations: 200M | Performance: 24,810,818 it/s (24th) | Time: 8.061 s

-------------------------------------
Test Type: Fibonachi sequence (92)
Iterations: 4000M | Performance: 515,065,696 it/s (24th) | Time: 7.766 s

-------------------------------------
Test Type: Pi approximation Taylor series method
Iterations: 800 | Performance: 156 it/s (24th) | Time: 5.055 s

-------------------------------------
Test Type: Greatest Common Divisor (GCD) Euclid's algorithm
Iterations: 1600M | Performance: 191,754,560 it/s (24th) | Time: 8.344 s

-------------------------------------
Test Type: Fill Memory 256 MB
Iterations: 60K | Performance: 4,029 it/s (24th) | Time: 14.891 s

-------------------------------------
Test Type: Copy Memory 256 MB
Iterations: 60K | Performance: 4,435 it/s (24th) | Time: 13.527 s

-------------------------------------
Test Type: Memory cache performance (max 4 threads)
Iterations: 40 | Performance: 6 it/s (4th) | Time: 25.120 s
Last result: SizeMem= 8 kB Iterations= 65536 Time= 4 ms
SizeMem= 16 kB Iterations= 32768 Time= 4 ms
SizeMem= 32 kB Iterations= 16384 Time= 4 ms
SizeMem= 64 kB Iterations= 8192 Time= 4 ms
SizeMem= 128 kB Iterations= 4096 Time= 4 ms
SizeMem= 256 kB Iterations= 2048 Time= 7 ms
SizeMem= 512 kB Iterations= 1024 Time= 33 ms
SizeMem= 1 MB Iterations= 512 Time= 69 ms
SizeMem= 2 MB Iterations= 256 Time= 61 ms
SizeMem= 4 MB Iterations= 128 Time= 56 ms
SizeMem= 8 MB Iterations= 64 Time= 54 ms
SizeMem= 16 MB Iterations= 32 Time= 61 ms
SizeMem= 32 MB Iterations= 16 Time= 68 ms
SizeMem= 64 MB Iterations= 8 Time= 83 ms
SizeMem= 128 MB Iterations= 4 Time= 107 ms


-------------------------------------
Test Type: Files Encrypt AES256 (10 MB)
Iterations: 20K | Performance: 1,699 it/s (24th) | Time: 11.766 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt AES256 (1 MB)
Iterations: 8K | Performance: 1,409 it/s (24th) | Time: 5.671 s

-------------------------------------
Test Type: Memory Encrypt/Decrypt Blowfish448 (1 MB)
Iterations: 8K | Performance: 2,345 it/s (24th) | Time: 3.407 s

-------------------------------------
Test Type: Memory Compress/Decompress 1 MB (Gzip)
Iterations: 8K | Performance: 1,959 it/s (24th) | Time: 4.079 s

-------------------------------------
Test Type: Files Compress 10 MB (Gzip)
Iterations: 4K | Performance: 842 it/s (24th) | Time: 4.727 s

-------------------------------------
Test Type: RSA Keys Generator (2048 bits)
Iterations: 2K | Performance: 351 it/s (24th) | Time: 5.665 s

-------------------------------------
Test Type: Hash SHA1/256/512 MD4/5 CRC32 RIPEMD160 (512 bytes)
Iterations: 1M | Performance: 44,264 it/s (24th) | Time: 22.591 s

-------------------------------------
Test Type: Random Numbers Generator (2048 bits)
Iterations: 200M | Performance: 20,682,524 it/s (24th) | Time: 9.670 s

-------------------------------------
Test Type: Bitcoin Address Generator
Iterations: 200K | Performance: 18,764 it/s (24th) | Time: 10.658 s

-------------------------------------
Test Type: Distances between points on Earth (Haversine)
Iterations: 1000M | Performance: 100,331,096 it/s (24th) | Time: 9.967 s

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

 

[CuriousMike]

5700G - stock, high, no cathedral

 

[AdamK47]

 

[nicalandia]

That i3 shouldn't be there. It was done on Medium. Or you could add High or Medium in parentheses.

Hey I just did the "HIGH" setting and it took me 54 minutes to complete it at 100% CPU(No Cathedral 3D). I want my Haswell i3 back on that Chart.




54 minutes to complete

 

[igor_kavinski]

I want my Haswell i3 back on that Chart.

Here ya go

 

[Makaveli]

Here ya go

View attachment 53633

lol that i3 breaks the scaling on the graph. The numbers from 5800H and below will have to be pulled out of the bars to match the above results so it all matches visually. And the intel Xeon should be a blue bar not red.

 

[igor_kavinski]

lol that i3 breaks the scaling on the graph. The numbers from 5800H and below will have to be pulled out of the bars to match the above results so it all matches visually. And the intel Xeon should be a blue bar not red.

Thanks. Chart updated.

 

[nicalandia]

Thanks. Chart updated.

Remember the ThreadRipper Chart that needed like two screens?

https://www.techpowerup.com/img/nThFVmoKBHVtMivq.jpg

My i3 Just beat that with your Original Chart.

 

[igor_kavinski]

Matrix Multiplication is 2.4s vs 9.4s

I find it hard to believe ADL is 4x worse at matrix multiplication,

The impact of cache locality on performance in C through matrix multiplication | by Gunavaran Brihadiswaran | Level Up Coding (gitconnected.com)

Very eye-opening article that shows matrix multiplication performance depends on cache locality.

Also, this may contain a clue: Zen 3: Load/Store and a Massive L3 Cache - AMD Zen 3 Ryzen Deep Dive Review: 5950X, 5900X, 5800X and 5600X Tested (anandtech.com)

The L3 continues to maintain shadow tags of the cores’ L2 contents – so if a cache line is requested by one core and resides on another core in the new core complex, the L3 will know from which core to fetch that line back from.

Something about Ryzen's cache architecture is helping it avoid cache misses when doing matrix multiplication.

 

[nicalandia]

The impact of cache locality on performance in C through matrix multiplication | by Gunavaran Brihadiswaran | Level Up Coding (gitconnected.com)

Very eye-opening article that shows matrix multiplication performance depends on cache locality.

Also, this may contain a clue: Zen 3: Load/Store and a Massive L3 Cache - AMD Zen 3 Ryzen Deep Dive Review: 5950X, 5900X, 5800X and 5600X Tested (anandtech.com)


Something about Ryzen's cache architecture is helping it avoid cache misses when doing matrix multiplication.

Would 3D Cache further improve this?

 

[igor_kavinski]

Would 3D Cache further improve this?

Don't know. I suppose 3D cache would be slower than L3 cache?

 

[nicalandia]

Don't know. I suppose 3D cache would be slower than L3 cache?

it's the same speed

AMD Demonstrates Stacked 3D V-Cache Technology: 192 MB at 2 TB/sec

www.anandtech.com

 

[JoeRambo]

The impact of cache locality on performance in C through matrix multiplication | by Gunavaran Brihadiswaran | Level Up Coding (gitconnected.com)

Very eye-opening article that shows matrix multiplication performance depends on cache locality.

Intel has 30MB of L3 + 1.5MB of L2, AMD 8C has 32MB of L3 and 512KB of L2.

Times without HT 8C vs 8C, at equal clocks of almost 5ghz for both CPUs:
14.3s vs 2.4s

Makes very little sense for such small difference in cache size to result in difference that large when no other test is showing anything at all. At some point one starts to question the validity of benchmark.
Maybe it is exacly 32MB of size and fits ZEN3 caches? Maybe access pattern defeats Intel's prefetcher in some strange way? Maybe AMD's branch prediction is always right.

Hard to tell without source code or Intel's VTune that is working on Alder Lake.

What is interesting, going from CL14 to CL13 has shaven 1s from runtime in that test, that is kinda shouting about ton of L3 cache misses.

No worries about this test tho, Intel has equally nonsensical win in Memory cache performance, so they cancel each other out

 

[nicalandia]

Why is my i3 so slow?? 2C/4T at 3.6 boost. it's 4 time slower than the slowest i7 which is also a Haswell.

 

[igor_kavinski]

Why is my i3 so slow?? 2C/4T at 3.6 boost. it's 4 time slower than the slowest i7 which is also a Haswell.

Maybe thermal throttling? Is the airflow good?

Also, only 3MB cache vs 8MB. No turbo boost.

 

[moinmoin]

Intel has 30MB of L3 + 1.5MB of L2, AMD 8C has 32MB of L3 and 512KB of L2.

Times without HT 8C vs 8C, at equal clocks of almost 5ghz for both CPUs:
14.3s vs 2.4s

Makes very little sense for such small difference in cache size to result in difference that large when no other test is showing anything at all. At some point one starts to question the validity of benchmark.
Maybe it is exacly 32MB of size and fits ZEN3 caches? Maybe access pattern defeats Intel's prefetcher in some strange way? Maybe AMD's branch prediction is always right.

Hard to tell without source code or Intel's VTune that is working on Alder Lake.

What is interesting, going from CL14 to CL13 has shaven 1s from runtime in that test, that is kinda shouting about ton of L3 cache misses.

No worries about this test tho, Intel has equally nonsensical win in Memory cache performance, so they cancel each other out

Latency may play a bigger role than size there. Would also explain the improvement from going from CL14 to CL13.

 

[lightmanek]

My HP Envy 360 15'' Ryzen 4700U with no tweaks apart from W10 High Performance mode.

No Cathedral and fully bloated Windows 10 Pro.

 

[Saylick]

it's the same speed

Patrick Kennedy from ServeTheHome says that the V$ die has a slightly higher latency than the base die, but it's miniscule compared to the latency going to DRAM, so it effectively can be ignored:

The Microsoft team also told us that some of the gains here are enormous. I asked specifically about the additional latency for the stacked cache segments. The response I got was very simple. The base case cache is basically the same. There is a very small latency to go to the stacked cache. On the other hand, that small latency to go through the interconnect up to the stacked die is so much smaller than going off the CCD, through the IO hub, then out to memory and back, that it is effectively irrelevant from a performance penalty perspective.

Interestingly, Microsoft state the following:

It is also important to note that because of the unique packaging AMD has utilized to achieve much larger L3 caches (i.e. vertical die stacking) that the L3 latency distribution will be wider than it was with Milan processors. This does not mean L3 memory latency is worse, per se. Best case L3 latencies should be the same as compared to Milan’s traditional planar approach to L3 packaging. However, worst case L3 latencies will be modestly slower.

 

[igor_kavinski]

My HP Envy 360 15'' Ryzen 4700U with no tweaks apart from W10 High Performance mode.

No Cathedral and fully bloated Windows 10 Pro.

Hmm...I can see the screenshot now but couldn't before. Weird.

Chart updated in original post.

 

[CuriousMike]

My laptop, Microsoft Surface Laptop 3
Intel i5-1035G7 - plugged in - no cathedral

1077.52

 

[AdamK47]

Decided to run this again now that I have DDR5 6000.

 

[NGruia]

I5 12600K
Asus Tuf Z690 WiFi D4
2x8 GB G.Skill 3200 MHz
GTX 750 2 MB (in final build RTX 3090)
Seagate FireCuda 530 2 TB
With Cathedral and wo Cathedral.