Overclock.net - An Overclocking Community - Reply to Topic

Thread: NEW!!! DRAM Calculator for Ryzen™ 1.5.1 (overclocking DRAM on AM4) + MEMbench 0.7 (DRAM bench) Reply to Thread
Title:
Message:

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in


  Additional Options
Miscellaneous Options

  Topic Review (Newest First)
06-17-2019 11:50 AM
herericc
Quote: Originally Posted by rul3s View Post
Hi guys!
I've found that HCI also encounters memory errors when vcore is so close to the edge of crashing, but not enought to chash, I explain.
Doing manual and fixed OC on my 2600x, I've found that [email protected] was stable on Realbench OCCT, but sometimes playing I've had some crashes, specially on PUBG.
My RAM config was [email protected] as Ryzen DRAM calculator says, with 1.40 on dram and 1.025 on vsoc. Well, so running HCI was giving me 1 or 2 errors on 400-500%.
Tried rising vdimm and also vsoc until 1.5 on vdimm and 1.20v on vsoc, but same experience, my only way to not hving errors was going down to [email protected]

After some research and test I finally tried to go rise vcore to 1,3125 and errors where solved, and now running [email protected] with vdimm on 1.38 and vsoc on 1.00v.

So, in my experience try to add a little bit more of vcore if experiencing very few errors on HCI.
That's good to know - I've noticed that when I do get errors in DRAM BENCH from the calculator they tend to be a streak of errors on a certain thread. Could be that particular thread on the processor is a little bit unstable due to PBO + undervolt... I'll have to try getting stability on RAM with PBO on stock settings and then work my way back from there.
06-17-2019 09:10 AM
rul3s Hi guys!
I've found that HCI also encounters memory errors when vcore is so close to the edge of crashing, but not enought to chash, I explain.
Doing manual and fixed OC on my 2600x, I've found that [email protected] was stable on Realbench OCCT, but sometimes playing I've had some crashes, specially on PUBG.
My RAM config was [email protected] as Ryzen DRAM calculator says, with 1.40 on dram and 1.025 on vsoc. Well, so running HCI was giving me 1 or 2 errors on 400-500%.
Tried rising vdimm and also vsoc until 1.5 on vdimm and 1.20v on vsoc, but same experience, my only way to not hving errors was going down to [email protected]

After some research and test I finally tried to go rise vcore to 1,3125 and errors where solved, and now running [email protected] with vdimm on 1.38 and vsoc on 1.00v.

So, in my experience try to add a little bit more of vcore if experiencing very few errors on HCI.
06-17-2019 08:03 AM
NightAntilli
Quote: Originally Posted by Keith Myers View Post
I'm not impressed. I beat most all those scores with my lowly 2700X and 3466Mhz CL14 RAM.

https://browser.geekbench.com/v4/cpu/12627947
Remember that the infinity fabric cuts its speed in half if the memory clock exceeds 3800 MHz.

06-16-2019 11:20 PM
hurricane28
Quote: Originally Posted by 1usmus View Post
194 are only a few weeks



to avoid conflicts with the new windows scheduler (1903) you will have to use AGESA Combo-AM4 1.0.0.1 or newer



AGESA Combo-AM4 1.0.0.1

Spoiler!


here are the results AGESA PinnaclePI 1.0.0.4
https://www.techpowerup.com/reviews/..._Guide/10.html
Good job man, looks like a lot of work.

Seems like 3466 MHz is the sweetspot for high CPU and RAM overclocks. Running this for months without problems.

Thnx for your hard work and keep it up.
06-16-2019 04:44 PM
Keith Myers
The new Linux 5.0 kernel is FAST

I have to keep reigning in my impulse to backport to the new Linux 5.0 kernel on my Ubuntu 18.04.2 LTS main daily driver partition. I have that in Disco Dingo on my development partition and it is FAST.
I see I am going to have to see what the Geekbench 4 scores are for my daily driver on that partition.
06-16-2019 01:28 PM
AlphaC
Quote: Originally Posted by Pilotasso View Post
that 2700X was benchmarked on linux 4.3.2 and cannot be compared with 3600X on V 4.3.4 windows directly.

this is my 2700X on V 4.3.4 windows.
https://browser.geekbench.com/v4/cpu/13550563
And then you start seeing a difference between 2700X and 3600X.
Exactly.


I have tested my R7 1700X @3.9 with 3466C14 in Ubuntu 18.04LTS with Linux 5.0 backported and obtained 5150ish single core
06-16-2019 08:25 AM
nick name
Quote: Originally Posted by Pilotasso View Post
that 2700X was benchmarked on linux 4.3.2 and cannot be compared with 3600X on V 4.3.4 windows directly.

this is my 2700X on V 4.3.4 windows.
https://browser.geekbench.com/v4/cpu/13550563
And then you start seeing a difference between 2700X and 3600X.
Here's mine. My multi-core is running at 4.3GHz and single at 4.35GHz. Windows 1903.

https://browser.geekbench.com/v4/cpu/13554145

That 3600 is still putting up a higher score at its 4.2GHz single than I am at 4.35GHz so that's nice to see. I wish I knew what its multi-core speed was.

And looking at the 16 core score running at 4.3GHz single-core is about 500 points higher than my 4.35GHz. And that one was running 4133MHz RAM.
06-16-2019 01:35 AM
Pilotasso
Quote: Originally Posted by Keith Myers View Post
I'm not impressed. I beat most all those scores with my lowly 2700X and 3466Mhz CL14 RAM.

https://browser.geekbench.com/v4/cpu/12627947


that 2700X was benchmarked on linux 4.3.2 and cannot be compared with 3600X on V 4.3.4 windows directly.

this is my 2700X on V 4.3.4 windows.
https://browser.geekbench.com/v4/cpu/13550563
And then you start seeing a difference between 2700X and 3600X.
06-16-2019 01:03 AM
gupsterg
Quote: Originally Posted by 1usmus View Post
AGESA Combo-AM4 1.0.0.1

Spoiler!


here are the results AGESA PinnaclePI 1.0.0.4
https://www.techpowerup.com/reviews/..._Guide/10.html
Thank you .
06-15-2019 10:33 PM
Keith Myers
Quote: Originally Posted by nick name View Post
Ditto on the beating of scores. What excites me is the possibility of going lower on my 3600MHz kit's timings. Or the possibility of the same timings, but up to 3666 or 3733. That would be ideal.
Yes, I have a G. Skill 3600 CL16 kit that I have never been able to get past 3466. It will be nice to actually achieve the rated XMP values. I'm fairly certain the optimized I/O die will do most of the work. Even on X470 boards that don't have the new X570 motherboard substrate and optimized RAM trace layout.
This thread has more than 10 replies. Click here to review the whole thread.

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off