ASRock.com Homepage
Forum Home Forum Home > Technical Support > AMD Motherboards
  New Posts New Posts RSS Feed - Fatality X399 Professional Gaming powers down
  FAQ FAQ  Forum Search Search  Events   Register Register  Login Login

Fatality X399 Professional Gaming powers down

 Post Reply Post Reply
Author
Message
ondro727 View Drop Down
Newbie
Newbie


Joined: 03 May 2019
Location: Slovak republic
Status: Offline
Points: 4
Post Options Post Options   Thanks (0) Thanks(0)   Quote ondro727 Quote  Post ReplyReply Direct Link To This Post Topic: Fatality X399 Professional Gaming powers down
    Posted: 03 May 2019 at 4:56pm
I have 4 identical render machines:

Mobo - Fatality X399 Professional Gaming
CPU - TR 2950x
RAM - GSkill RipjawV 3200MHz CL16 (4x16B)
GPU - dual RTX 2080Ti (non SLI)
HDD - 1TB NVMe SSD + 2x4TB SATA SSD
PSU - Seasonic Prime 1300W Gold
OS - Win10Pro 64bit

The issue I am having is that under heavy GFX load (machines are used for photogrammetry calculations which are both CPU and GPU heavy) the PCs shut down completely (not clean Windows shut down but sudden complete power off). They don't lock up on OS level, but shut down and can't be powered back on unless disconnected from power and reconnected back. This is not an isolated issue, since all 4 render machines exhibit the same bahaviour. The PSU is more than sufficient, the temperatures are well within threshold (CPU 55-60C, GPU 70C, VRM 50-60C). Other than the shutdowns the system doesn't exhibit any other unusual behaviour. CPUs are not using Precision Boost Overdrive and are set to default values - in fact whole BIOS is set to default, except for memory (XMP selected) and fan profiles (full speed). In this state the render machines are unusable at the moment. I have 6 other machines (for CPU work) which are almost identical, but have AMD TR 2990wx CPUs and NVidia 1660Ti GPU (single) - they do not behave like that. Any suggestion what may cause the boards to shut down/power off? Thank you in advance.
Back to Top
hudloom View Drop Down
Newbie
Newbie


Joined: 06 Jun 2019
Status: Offline
Points: 3
Post Options Post Options   Thanks (0) Thanks(0)   Quote hudloom Quote  Post ReplyReply Direct Link To This Post Posted: 06 Jun 2019 at 9:33pm
I can confirm the exact same behaviour. My System runs rock stable but under heavy load e.g. gaming (i am on 4k) it happens the machine powers down like someone pulls the plug. This can happen after 30min, 4 hours or not at all... If you pull the plug and reconnect it starts again.

Heat is not an issue, Afterburner shows me 75 Deg on GPU and 60 Deg on CPU.

I already replaced the power supply as I expected a hardware issue but same behaviour applies after changing the PSU.

Fatal1ty B450 Gaming K4
Ryzen 2600
GSkill Aegis RAM
RTX 2080Ti
bequite Straight Power 550
Win10
Back to Top
Xaltar View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 16 May 2015
Location: Europe
Status: Offline
Points: 27473
Post Options Post Options   Thanks (0) Thanks(0)   Quote Xaltar Quote  Post ReplyReply Direct Link To This Post Posted: 06 Jun 2019 at 10:49pm
Hudloom, your issue is not the same, you are on a different platform and not using an adequate PSU for your setup.

Your PSU is completely wrong for a gaming system. BeQuiet's Straight Power lineup are multi rail PSUs. What is happening with your system is your PSU is getting overloaded by your 2080ti. You need a decent PSU with a single 12v rail of at least 50A.

BeQuiet are not a bad brand but their PSUs tend to be garbage for gaming systems. This isn't true of all their PSUs but practically ALL of their budget models are multi rail. I do not recommend them.



This is your PSU, in red you see the power is split between 4 rails. Totaling up the 4x18A would give you a total power of about 800w so in this case, it isn't 4 actual rails, it's 40A split between 4 rails to a maximum of 18A per rail. 40A = 480w of total power of which, no single device can draw more than 18A (216w). In other words, as far as your GPU is concerned, it may as well be a 220w PSU. Worse still, if the motherboard and GPU are both sharing the same rail that is 220w split between them.



This is a proper 550w PSU for comparison. As you can see, the entire 45A 12v power rating is available on a single rail. Now, as I said, you want a PSU with a 50A +12v rating which is about 600w.

Edited by Xaltar - 06 Jun 2019 at 11:23pm
Back to Top
ondro727 View Drop Down
Newbie
Newbie


Joined: 03 May 2019
Location: Slovak republic
Status: Offline
Points: 4
Post Options Post Options   Thanks (0) Thanks(0)   Quote ondro727 Quote  Post ReplyReply Direct Link To This Post Posted: 06 Jun 2019 at 10:55pm
Tried several other solutions - changed PSU (for EVGA 1600W), reset RAM do JDEC specs (2133MHz). But nothing helped so far. No response from ASRock support at all... So much for the "pro" branding of products. Seems to me the ASRock is not a way to go with serious deployment.
Back to Top
Xaltar View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 16 May 2015
Location: Europe
Status: Offline
Points: 27473
Post Options Post Options   Thanks (0) Thanks(0)   Quote Xaltar Quote  Post ReplyReply Direct Link To This Post Posted: 06 Jun 2019 at 11:09pm
I have sent you a PM Ondro727. Hopefully we can get your problem sorted.
Back to Top
hudloom View Drop Down
Newbie
Newbie


Joined: 06 Jun 2019
Status: Offline
Points: 3
Post Options Post Options   Thanks (0) Thanks(0)   Quote hudloom Quote  Post ReplyReply Direct Link To This Post Posted: 07 Jun 2019 at 12:37pm
Hi Xaltar,

thank you for your detailed contribution, sounds plausible. I suspected the PSU but I thought about a hardware failure, but replacing with the same model did not help.

However, I did some reading here in the forum on the first 20 pages or so and did see that there were some issues due to overheating and shut downs.

So I changed the fan monitoring to tctl and did go on playing for about 4h in total yesterday. The PC runs now much cooler, looks like the temp afterburner showed me was not the real CPU temp (still at 55 C but the case is much cooler by touching)

I will investigate further through the weekend and post results.
Back to Top
Xaltar View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 16 May 2015
Location: Europe
Status: Offline
Points: 27473
Post Options Post Options   Thanks (0) Thanks(0)   Quote Xaltar Quote  Post ReplyReply Direct Link To This Post Posted: 07 Jun 2019 at 1:28pm
The primary byproduct of an overloaded PSU is heat, both within the PSU itself and any components it powers. The air coming out the back of your PSU should never be more than warm, if it is hot, your PSU is struggling. Typically if a PSU is overloaded you get voltage droop, low voltage = high current = heat.

TCL exists for more rapid fan control response so if you are running hot, it is a good idea to use it. With that said however, your PSU is utter garbage for that system. If you search "Be Quiet" on the forums here you will see dozens of posts where their PSUs turned out to be behind a broad spectrum of issues on gaming setups. They are not the only brand guilty of branding multi rail PSUs in a confusing way but they are one of the more popular budget options. I have even made a sticky on the topic.
Back to Top
hudloom View Drop Down
Newbie
Newbie


Joined: 06 Jun 2019
Status: Offline
Points: 3
Post Options Post Options   Thanks (0) Thanks(0)   Quote hudloom Quote  Post ReplyReply Direct Link To This Post Posted: 09 Jun 2019 at 6:54am
Probably going for a Corsair HX750, this should do the job :)
Back to Top
 Post Reply Post Reply
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.04
Copyright ©2001-2021 Web Wiz Ltd.

This page was generated in 0.092 seconds.