Jump to content

Threadripper/Asus performance confusion using DOCP/XMP

I did Google this but didn't find something so specific. Maybe someone has an idea of what's going on.

Check system specs in my signature.

 

Story:

 

With DOCP (AMD's version of XMP) turned on in the BIOS, the RAM runs at 3600 but I've noticed that my CPU wouldn't hit 100% while running cinebench r23, not even getting hot enough for the fans to go much faster. That reflected in the score: I was getting around 41000 while I found online that this Threadripper CPU should get around 43000. (I also tested with all the performance enhancements I could find set to maximum in the BIOS, still couldn't hit 100% usage).

 

So I set the BIOS to the defaults (DOCP off). The RAM went back to stock 2333, but the CPU now hits 100% on cinebench and the score went UP to the expected >43000 with technically slower RAM! A video export test from Davinci took 32 seconds.

 

I understand the performance gain at lower frequency can be a timing issue with the XMP profile, so for the next test I kept the BIOS on defaults with DOCP off again, but this time I manually entered all the timings, frequency (3600) and voltage for the RAM according to the XMP profile which I found using Thaiphoon software. With this setup, cinebench is hitting a lower score again (about 41900) but the video export speed actually went quite a bit faster, around 24 seconds.

 

I may be missing something here, but how can I get faster results on Davinci when the settings give me a lower score on cinebench?

 

Threadripper 3970x | ASUS ROG Strix TRX40-E Gaming | Corsair HX1200i | 64 GB DDR4 Ripjaws-V 3600Mhz CL18-22-22-42 | Gigabyte RTX 3090 Xtreme Waterforce | Obsidian Series 750D Airflow | 250GB WD Blue SSD SN500 NVMe M.2 | 2x Samsung 860 Evo 1 TB Aorus 8 TB NVMe PCIe SSD | Nocuta NH-U14S TR4-SP3 | 3x Corsair ML120 Pro RGB | 1x Corsair ML140 Pro RGB | 1x Noctua NF-A14 | 1x Corsair AF140L |

Link to post
Share on other sites

32 minutes ago, B.Rms said:

I may be missing something here, but how can I get faster results on Davinci when the settings give me a lower score on cinebench?

 

Its possible that due to the speed bump da vinci is able to more quickly provide the CPU with files that it needs to export hence the faster time. I would not worry too much about the cinebench score if your actual work with the system is faster.

I will recommend an NHu12s (or an NHd15 (maybe)) for your PC build. Quote or @ me @Prodigy_Smit for me to see your replies.

PSU Teir List | Howdy! A Windows Hello Alternative 

 

 

Desktop :

i7 8700 | Quadro P4000 8GB |  64gb 2933Mhz cl18 | 500 GB Samsung 960 Pro | 1tb SSD Samsung 850 evo

Laptop :

ASUS G14 | R9 5900hs | RTX 3060 | 16GB 3200Mhz | 1 TB SSD

Link to post
Share on other sites

Cinebench is a tile based renderer, and doesn't need the cores to communicate very fast.

 

Fast ram on AMD Ryzen CPUs improves core to core speed, and thus will improve performance in every program that isn't a tile based renderer (so basically everything that is isn't cinebench and a couple other things)

I WILL find your ITX build thread, and I WILL recommend the SIlverstone Sugo SG13B

 

Primary PC:

i7 8086k - EVGA Z370 Classified K - G.Skill Trident Z RGB - WD SN750 - Jedi Order Titan Xp - Hyper 212 Black (with RGB Riing flair) - EVGA G3 650W - dual booting Windows 11 and Fedora Linux - Black and green theme, Razer brainwashed me.

Draws 400 watts under max load, for reference.

 

PSU tier list

How many watts do I need?

PSU misconceptions, protections explainedgroup reg is bad

Link to post
Share on other sites

1 hour ago, Prodigy_Smit said:

if your actual work with the system is faster.

Well, that's exactly my thoughts.

 

1 hour ago, Fasauceome said:

Cinebench is a tile based renderer, and doesn't need the cores to communicate very fast.

 

Fast ram on AMD Ryzen CPUs improves core to core speed, and thus will improve performance in every program that isn't a tile based renderer (so basically everything that is isn't cinebench and a couple other things)

Sounds logical. So typing all the settings manually for the RAM according to Thaiphoon and leaving everything else as optimized defaults is a valid way to setup the BIOS? Because it's the only way I found to make both the CPU and RAM run at max stock speeds without overclocking.

Threadripper 3970x | ASUS ROG Strix TRX40-E Gaming | Corsair HX1200i | 64 GB DDR4 Ripjaws-V 3600Mhz CL18-22-22-42 | Gigabyte RTX 3090 Xtreme Waterforce | Obsidian Series 750D Airflow | 250GB WD Blue SSD SN500 NVMe M.2 | 2x Samsung 860 Evo 1 TB Aorus 8 TB NVMe PCIe SSD | Nocuta NH-U14S TR4-SP3 | 3x Corsair ML120 Pro RGB | 1x Corsair ML140 Pro RGB | 1x Noctua NF-A14 | 1x Corsair AF140L |

Link to post
Share on other sites

1 hour ago, B.Rms said:

Sounds logical. So typing all the settings manually for the RAM according to Thaiphoon and leaving everything else as optimized defaults is a valid way to setup the BIOS? Because it's the only way I found to make both the CPU and RAM run at max stock speeds without overclocking

Yep

I WILL find your ITX build thread, and I WILL recommend the SIlverstone Sugo SG13B

 

Primary PC:

i7 8086k - EVGA Z370 Classified K - G.Skill Trident Z RGB - WD SN750 - Jedi Order Titan Xp - Hyper 212 Black (with RGB Riing flair) - EVGA G3 650W - dual booting Windows 11 and Fedora Linux - Black and green theme, Razer brainwashed me.

Draws 400 watts under max load, for reference.

 

PSU tier list

How many watts do I need?

PSU misconceptions, protections explainedgroup reg is bad

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×