Asus Prime X399-A - Overclock.net - An Overclocking Community

Forum Jump: 

Asus Prime X399-A

Reply
 
Thread Tools
post #1 of 180 (permalink) Old 11-13-2017, 10:28 AM - Thread Starter
New to Overclock.net
 
Quijano's Avatar
 
Join Date: May 2017
Posts: 6
Rep: 0
Since I can’t find any Asus x399-a dedicated thread I will start one.

I put together my 1950x build today and ran into some memory issues with it.

I m using a gskill ripjaws F4-3200c14q-32gvr Kit.

If I use A1, B1 & A2, B2 slots all stick get recognized, I can set docp profile and it will boot with 3200mhz.

However if I use any of the C & D slots the system won’t boot. It will start up and after couple of seconds shut down again and then start rebooting. This goes on for a long time. I had to force a shutdown.

I use the 0403 bios.

Has anyone expierenced the same problem?
Quijano is offline  
Sponsored Links
Advertisement
 
post #2 of 180 (permalink) Old 11-15-2017, 02:14 AM
New to Overclock.net
 
kryton13's Avatar
 
Join Date: Nov 2017
Posts: 1
Rep: 0
Made an account just to respond to this.

I have had similar but not identical issues with this board.

Am using 2 identical 32gb kits (sorry i cannot remember the exact model right now, can update later) and when both are installed the system will sometimes get stuck in the same type of post cycle noted above. Interestingly cutting all power to the system and then starting it again allows the system to post, although it still takes far too long ~ 30 seconds, and then it boots into windows 10 and is suddenly all happy.

The system was not experiencing these issues with just one of the 32gb kits installed.

I had the C and D sockets populated with one kit when it was working fine - it was then using all the A and B sockets with the other kit that caused this problem.

Another, possible related issue, is that the RAM won't go to the 2666 frequency it should be at, if i force it to that then the system won't post at all. I have not tried upping voltages yet.

I have also been using the 0403 bios.

I have currently not had time to do a full check stick by stick to see if the RAM is good or not, have done visual inspection of tracks etc though.
kryton13 is offline  
post #3 of 180 (permalink) Old 11-15-2017, 05:58 AM - Thread Starter
New to Overclock.net
 
Quijano's Avatar
 
Join Date: May 2017
Posts: 6
Rep: 0
Update:

I took apart the whole system, removed the CPU, but the CPU back in. But the rest back in and now it works. I assume one of the pins didnt connect proberly/in the right place. Not sure how this can happen, considering the socket construction. But at least it works now as intended.
Quijano is offline  
Sponsored Links
Advertisement
 
post #4 of 180 (permalink) Old 12-05-2017, 08:34 AM
New to Overclock.net
 
Tamalero's Avatar
 
Join Date: Dec 2016
Posts: 322
Rep: 5 (Unique: 5)
Quote:
Originally Posted by kryton13 View Post

Made an account just to respond to this.

I have had similar but not identical issues with this board.

Am using 2 identical 32gb kits (sorry i cannot remember the exact model right now, can update later) and when both are installed the system will sometimes get stuck in the same type of post cycle noted above. Interestingly cutting all power to the system and then starting it again allows the system to post, although it still takes far too long ~ 30 seconds, and then it boots into windows 10 and is suddenly all happy.

The system was not experiencing these issues with just one of the 32gb kits installed.

I had the C and D sockets populated with one kit when it was working fine - it was then using all the A and B sockets with the other kit that caused this problem.

Another, possible related issue, is that the RAM won't go to the 2666 frequency it should be at, if i force it to that then the system won't post at all. I have not tried upping voltages yet.

I have also been using the 0403 bios.

I have currently not had time to do a full check stick by stick to see if the RAM is good or not, have done visual inspection of tracks etc though.

Having a similar problem as well.

Using Corsair Domminator Platinum(4x8 32GB 3000Mhz kit) on latest bios (0403). Got a loop after hitting stage 14 and after stage 15, it freezes for a sec then reboots making a loop.
It seems it does not like 4 modules populating the silver slots.

Interestingly, it boots to bios at default memory speed (not the rated ones) a few times.

Will be testing. Might be a bit worse tome since my memories are probably Hynix and not Samsung.
Tamalero is offline  
post #5 of 180 (permalink) Old 12-05-2017, 05:25 PM
New to Overclock.net
 
Tamalero's Avatar
 
Join Date: Dec 2016
Posts: 322
Rep: 5 (Unique: 5)
Scratch my latest post..


I decided to do stick by stick training of the memory.

So after NOT running AT ALL on 4 modules (4 x 8) on 2666 and above, aka the dreaded Error 14-15 loop. I decided to remove all but one module of memory.
This memory ran fine at 2666 in auto settings.
Then added the second module, and both ran fine.(did not run previously at 2666 or anything higher, caused the 14 15 loop too)
Then added 4 and ran fine at 2666. (would not boot to anything above 2400Mhz with the 14 15 loop before)
All looked fine.. so decided to get back to bios and upped the stakes by setting the speed to 2933.

Got 3 quick Reboots Then I got a forced in to the BIOS (no memory reset errors).
Then noticed that the auto ram settings and timings had changed automatically from what was before . All while using AUTO mode in the memory timings section.
The changes were the basic timings from CL15 to CL16.

Decided also to adjust some settings. First turned on geardown as per other suggestions I've seen online. and Rate 1T.

Booted successfully and arrived to windows with no issue running the memories at. at 2933 CL16.


Once I get my box ready I will be stress testing.
Maybe later I will try to see if it boots at 3200 (this kit is 3000Mhz CL15. Most probably Hynix memory.)
Tamalero is offline  
post #6 of 180 (permalink) Old 01-10-2018, 01:18 PM
New to Overclock.net
 
Tamalero's Avatar
 
Join Date: Dec 2016
Posts: 322
Rep: 5 (Unique: 5)
New Bios update for ASUS PRIME X399 A is now available in the ASUS website

https://www.asus.com/us/Motherboards/PRIME-X399-A/HelpDesk_BIOS/

It says "improves performance". So I guess this patch is the new AGESA?

Because these patch notes are very cryptic.
Tamalero is offline  
post #7 of 180 (permalink) Old 01-10-2018, 03:49 PM
New to Overclock.net
 
The L33t's Avatar
 
Join Date: Mar 2017
Location: Portugal
Posts: 275
Rep: 16 (Unique: 13)
I doubt it has the new AGESA, that is usually mentioned. The Zenith had it and then was removed due to some issues, I'd say the same issues would be present on the Prime as well. The only brand rocking the new AGESA for TR is still ASRock. I think the new AGESA probably does have some issues that ASRock managed to avoid somehow?!, because none of the other vendors are touching it for some reason and I do not believe it is just "laziness". Time will tell.

Performance could be a number of things, usually optimizations withing the BIOS and some BIAS built in for some specific tasks.
The L33t is offline  
post #8 of 180 (permalink) Old 01-10-2018, 07:16 PM
New to Overclock.net
 
Tamalero's Avatar
 
Join Date: Dec 2016
Posts: 322
Rep: 5 (Unique: 5)
Quote:
Originally Posted by The L33t View Post

I doubt it has the new AGESA, that is usually mentioned. The Zenith had it and then was removed due to some issues, I'd say the same issues would be present on the Prime as well. The only brand rocking the new AGESA for TR is still ASRock. I think the new AGESA probably does have some issues that ASRock managed to avoid somehow?!, because none of the other vendors are touching it for some reason and I do not believe it is just "laziness". Time will tell.

Performance could be a number of things, usually optimizations withing the BIOS and some BIAS built in for some specific tasks.

Noone of the prime patches says anything about AGESA or at all, they just mention "stability" "performance" etc..

So I guess you could be right.

*edit*

Other motherboards just got AGESA 1.0.0a (new ratios and support for new Pinnacle Ridge processors).

*edit 2*

Tried updating my bios with the latest 407.
A total nightmare, couldn't even get it to be stable at 2666 Mhz ram in stock settings. :|

*edit 3*
Tried Flashing 3 times, redownloaded the bios as well.. and...

Can confirm, Latest bios is extremely unstable. Cant even get memories to 2660.
Programs do weird stuff too, saying the block rate is moving and the multipliers fixed (memtest, aida64, only hwinfo gets the correct readings).
Also constant crashes on both video card, blue screens on everything stock ( memory pool, memory management, DCP protected).

Not to mention that memory training would insist into trying to shove very tight timings to my memories (even more than the older 403 bios)
Tamalero is offline  
post #9 of 180 (permalink) Old 01-14-2018, 01:46 AM
Meddling user
 
gupsterg's Avatar
 
Join Date: Jan 2015
Location: Lurking over a keyboard
Posts: 6,733
Rep: 730 (Unique: 342)
@Tamalero

If you open HWINFO without ticking sensors/summary you'll be able to see AGESA version in motherboard section and CPU microcode/SMU firmware version in CPU section. I have noted sometimes the UEFI has incorrect string for AGESA.

Post up a screenshot of Thaiphoon Burner, be interesting to know what RAM IC, etc you have.

Ryzen Master I didn't use on Ryzen, as it pretty much had no use to for me, but on ThreadRipper I have found it displays ProcODT value. It helped me to know what was [Auto] ProcODT and then had a starting point where to go with it when gunning for higher RAM MHz. In the past when played with ProcODT it can mean at post a smooth ride or total hell, does also have affect on stability in OS.

CLDO_VDDP setting has also helped resolve post issues, this will move a memory hole if you have encountered one. When you encounter one usually it is full post fail. As ThreadRipper has 2 IMC it could also be one IMC enters a hole and other doesn't. Why I say that is I was in a situation where intermittently 1/2 of the installed RAM was being hardware reserved, after playing with CLDO_VDDP it has resolved issue so far.

CAD Bus has not helped me with post process when pushing RAM but more with stability of it in OS.

SOC/VDIMM I would tune as needed rather than use [Auto] and let it be set based on "autorules".
gupsterg is offline  
post #10 of 180 (permalink) Old 01-14-2018, 11:19 AM
New to Overclock.net
 
Tamalero's Avatar
 
Join Date: Dec 2016
Posts: 322
Rep: 5 (Unique: 5)
Quote:
Originally Posted by gupsterg View Post

@Tamalero

If you open HWINFO without ticking sensors/summary you'll be able to see AGESA version in motherboard section and CPU microcode/SMU firmware version in CPU section. I have noted sometimes the UEFI has incorrect string for AGESA.

Post up a screenshot of Thaiphoon Burner, be interesting to know what RAM IC, etc you have.

Ryzen Master I didn't use on Ryzen, as it pretty much had no use to for me, but on ThreadRipper I have found it displays ProcODT value. It helped me to know what was [Auto] ProcODT and then had a starting point where to go with it when gunning for higher RAM MHz. In the past when played with ProcODT it can mean at post a smooth ride or total hell, does also have affect on stability in OS.

CLDO_VDDP setting has also helped resolve post issues, this will move a memory hole if you have encountered one. When you encounter one usually it is full post fail. As ThreadRipper has 2 IMC it could also be one IMC enters a hole and other doesn't. Why I say that is I was in a situation where intermittently 1/2 of the installed RAM was being hardware reserved, after playing with CLDO_VDDP it has resolved issue so far.

CAD Bus has not helped me with post process when pushing RAM but more with stability of it in OS.

SOC/VDIMM I would tune as needed rather than use [Auto] and let it be set based on "autorules".

Here you go!

HWINFO AGESA line:
AMD AGESA Version: ThreadRipperPI-SP3r2-0.0.6.0
Microcode Update Revision: 8001129
SMU Firmware Revision: 4.25.92.0








Now, the issue of these RAM modules, on start or using automated mode, it will use 15,15,15,36,50 which is unstable.
It does not load the XMP stable until the third reboot (15-17-17-35-53, but keeps the 1.2V profile).

I'm running now at 3066 on Bios 0403. 0407 is currently the newest. (memtest stable, prime stable and OCCTP stable)

The latest bios is a no go (nothing is stable even at 2100/stock cpu).

Thanks for helping me man!
Tamalero is offline  
Reply

Quick Reply
Message:
Options

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



Currently Active Users Viewing This Thread: 5 (0 members and 5 guests)
 


Forum Jump: 

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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off