ASRock.com Homepage
Forum Home Forum Home > Technical Support > AMD Motherboards
  New Posts New Posts RSS Feed - X370 K4 Bios Update Fail to post after
  FAQ FAQ  Forum Search Search  Events   Register Register  Login Login

X370 K4 Bios Update Fail to post after

 Post Reply Post Reply Page  12>
Author
Message Reverse Sort Order
parsec View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 04 May 2015
Location: USA
Status: Offline
Points: 4996
Post Options Post Options   Thanks (0) Thanks(0)   Quote parsec Quote  Post ReplyReply Direct Link To This Post Topic: X370 K4 Bios Update Fail to post after
    Posted: 15 Apr 2017 at 9:10am
Originally posted by Windows98 Windows98 wrote:

Originally posted by Windows98 Windows98 wrote:

Welp... 2.10 bios is out. Should I give it a spin? 2.00 surely didn't go well...

Flipped a coin, coin said to do it. 

So I did.

Went fine. Big smile

Typing this on my Ryzen box right now. Wonder what the deal with 2.00 was...

Still seeing odd stock clock speeds, btw. I noticed if you change the minimum processor state from 90% to some other value (like 10%) the clock speeds will fall by half. hwinfo reflects ~2100 mhz.

Kind of interesting. 




That's the AMD processor's Cool n Quiet feature enabled, and you might have the C6 C State power saving option enabled.

Under low load conditions, with those options enabled and the Power Plan configured as you said, the CPU will switch to a lower speed to save power. You might see even lower core speeds with the Minimum Processor state below 10%.

Even with the Minimum Processor state set to 90%, or using the AMD Ryzen Power plan, I see the VCore and core power usage changing to low values, which means most of the cores are in the C6 C State.
Back to Top
parsec View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 04 May 2015
Location: USA
Status: Offline
Points: 4996
Post Options Post Options   Thanks (0) Thanks(0)   Quote parsec Quote  Post ReplyReply Direct Link To This Post Posted: 15 Apr 2017 at 9:03am
Originally posted by debaser debaser wrote:


Hello, I noticed you have a fan for cooling VRM area, may I ask you how did you place it?



Good eyes, it was not hard but also not simple. The PC case you are using will make a difference if is possible to do or not.

This picture is of a different mother board in the same PC case I'm using for my Ryzen system. I'm using the same custom fan mount I made to hold an 80mm fan to cool the VRM heatsink.

It's a 2" L bracket from a hardware store, with the fan held on to it by a zip tie. The L bracket has holes in it, with a #6x32 screw with a plastic spacer connecting the L bracket to a magnet in the round chrome plated thing on the far right. The magnet is very strong, and hold the fan in place on the mount. The magnet thing is from a surplus store, so matching it will be difficult. I've changed the fan mount a little since this picture was taken, adding a second screw and magnet.

You can barely see the VRM heatsink, it's the red part you can see behind the fan. That's an ASRock 970 Performance /3.1 that I used with an FX-9590.

That case is a Cooler Master HAF XB "Test Bench and LAN box" PC case. It is nothing like a standard PC tower case, and just by luck allows me to mount the fan near the VRM heatsink.

https://www.newegg.com/Product/Product.aspx?Item=N82E16811119265&ignorebbr=1

Honestly, this fan does not make that much difference, since this case has a 200mm fan in the top cover, blowing air down onto the board. If you can insure some air movement across the VRM heatsink areas, they won't over heat. When I was stress testing my 1700X at 3.9GHz, the CPU temperature was higher than the VRM temperature.











Back to Top
Windows98 View Drop Down
Newbie
Newbie


Joined: 11 Apr 2017
Status: Offline
Points: 22
Post Options Post Options   Thanks (0) Thanks(0)   Quote Windows98 Quote  Post ReplyReply Direct Link To This Post Posted: 15 Apr 2017 at 7:18am
Originally posted by Windows98 Windows98 wrote:

Welp... 2.10 bios is out. Should I give it a spin? 2.00 surely didn't go well...

Flipped a coin, coin said to do it. 

So I did.

Went fine. Big smile

Typing this on my Ryzen box right now. Wonder what the deal with 2.00 was...

Still seeing odd stock clock speeds, btw. I noticed if you change the minimum processor state from 90% to some other value (like 10%) the clock speeds will fall by half. hwinfo reflects ~2100 mhz.

Kind of interesting. 




Edited by Windows98 - 15 Apr 2017 at 7:23am
Back to Top
debaser View Drop Down
Newbie
Newbie


Joined: 31 Mar 2017
Status: Offline
Points: 35
Post Options Post Options   Thanks (0) Thanks(0)   Quote debaser Quote  Post ReplyReply Direct Link To This Post Posted: 15 Apr 2017 at 2:35am
Originally posted by parsec parsec wrote:

It's much to early to come to conclusions about what is happening with the Ryzen P-States, etc,  for many reasons. One important one is, the monitoring programs don't have Ryzen right yet. Even HWiNFO needs some work. CPU-Z cannot deal with Ryzen's RAID, it locks it up before it's done with its detecting phase.

Plus I've found that the UEFI/BIOS option settings for VCore and multipliers seem to affect what is shown in monitoring programs. I wish HWiNFO could show the P-State levels and percentages like it does with the Intel C-State levels and percentages.

Here's my HWiNFO example:



This is with the core multipliers set to 37, and the VCore set to Offset, and the offset value left on Auto. You can see the CPU frequency never changes, like yours. But the VCore can drop to a super low 0.384V, and yes it is stable, I just took that screenshot a few minutes ago. Also check the minimum core power values, below 0.1W. I'm using the Ryzen power plan, Windows 10. C n Q enabled, C6 enabled, and the P States expanded in their screen in the UEFI.

I have seen the 1.5V+ VIDs, with VCore set to Auto. Get rid of Auto if you want that gone. Just a VID though, so if like an Intel processor, not the true VCore.

With my Intel systems, Z270 and i7-7700K for example, if I used the Windows High Performance plan, the CPU frequency would stay high, and never drop to 800MHz, with all C states down to C7 enabled. But the C State monitoring in HWiNFO would show I was in C7 90+% of the time.

IMO, to early to say exactly what is going on precisely. Our tools are not good enough, and we are still learning Ryzen.



Hello, I noticed you have a fan for cooling VRM area, may I ask you how did you place it?

Back to Top
Windows98 View Drop Down
Newbie
Newbie


Joined: 11 Apr 2017
Status: Offline
Points: 22
Post Options Post Options   Thanks (0) Thanks(0)   Quote Windows98 Quote  Post ReplyReply Direct Link To This Post Posted: 15 Apr 2017 at 2:09am
Welp... 2.10 bios is out. Should I give it a spin? 2.00 surely didn't go well...
Back to Top
ASRock_TSD View Drop Down
ASRock_Official
ASRock_Official


Joined: 20 Mar 2015
Status: Offline
Points: 8727
Post Options Post Options   Thanks (0) Thanks(0)   Quote ASRock_TSD Quote  Post ReplyReply Direct Link To This Post Posted: 13 Apr 2017 at 11:42am
Dear Happy2b1,

Greetings, this is ASRock TSD.


We test Fatal1ty X370 Gaming K4 from BIOS P1.60, L1.61 to L1.63 with instant flash and system can boot up properly.

With this model, we could boot up properly with two G.Skill F4-3200C16Q-16GRK memory modules without error after updating BIOS.

We will keep testing and take care of updating BIOS issue.


Thanks!

All the best,
ASRock TSD

Back to Top
parsec View Drop Down
Moderator Group
Moderator Group
Avatar

Joined: 04 May 2015
Location: USA
Status: Offline
Points: 4996
Post Options Post Options   Thanks (0) Thanks(0)   Quote parsec Quote  Post ReplyReply Direct Link To This Post Posted: 11 Apr 2017 at 12:39pm
It's much to early to come to conclusions about what is happening with the Ryzen P-States, etc,  for many reasons. One important one is, the monitoring programs don't have Ryzen right yet. Even HWiNFO needs some work. CPU-Z cannot deal with Ryzen's RAID, it locks it up before it's done with its detecting phase.

Plus I've found that the UEFI/BIOS option settings for VCore and multipliers seem to affect what is shown in monitoring programs. I wish HWiNFO could show the P-State levels and percentages like it does with the Intel C-State levels and percentages.

Here's my HWiNFO example:



This is with the core multipliers set to 37, and the VCore set to Offset, and the offset value left on Auto. You can see the CPU frequency never changes, like yours. But the VCore can drop to a super low 0.384V, and yes it is stable, I just took that screenshot a few minutes ago. Also check the minimum core power values, below 0.1W. I'm using the Ryzen power plan, Windows 10. C n Q enabled, C6 enabled, and the P States expanded in their screen in the UEFI.

I have seen the 1.5V+ VIDs, with VCore set to Auto. Get rid of Auto if you want that gone. Just a VID though, so if like an Intel processor, not the true VCore.

With my Intel systems, Z270 and i7-7700K for example, if I used the Windows High Performance plan, the CPU frequency would stay high, and never drop to 800MHz, with all C states down to C7 enabled. But the C State monitoring in HWiNFO would show I was in C7 90+% of the time.

IMO, to early to say exactly what is going on precisely. Our tools are not good enough, and we are still learning Ryzen.


Back to Top
Windows98 View Drop Down
Newbie
Newbie


Joined: 11 Apr 2017
Status: Offline
Points: 22
Post Options Post Options   Thanks (0) Thanks(0)   Quote Windows98 Quote  Post ReplyReply Direct Link To This Post Posted: 11 Apr 2017 at 11:03am
Here's a screencap of HWInfo. I like it because it shows current, minimum, and maximum recorded values for however long the program has been running. In this shot, it's been running for about 2 1/2 hours, so I'd assume it's a safe bet that this isn't one of those weird uptime issues you describe. 



You can plainly see the XFR has kicked in on certain cores, but the minimum clock speed never falls below 3691.9mhz.


As for the CPU-Z vcore thing, I am 99.99999999% sure it's wrong. It appears to be reading that value from somewhere (because it flutters within ~0.02V as I leave the application open) but it never deviates +/-  0.02V if I throw a stress test at it or not. 

It's probably just reading the wrong sensor. 

HWInfo appears to pull the vcore down appropriately, though it does show a "max" of 1.5V which is kind of concerning...



Edit: In looking at first screenshot, I did some digging on the weird VID values and found that it too is a bug that needs fixed in the microcode. 


Edited by Windows98 - 11 Apr 2017 at 11:05am
Back to Top
wardog View Drop Down
Moderator Group
Moderator Group


Joined: 15 Jul 2015
Status: Offline
Points: 6447
Post Options Post Options   Thanks (0) Thanks(0)   Quote wardog Quote  Post ReplyReply Direct Link To This Post Posted: 11 Apr 2017 at 10:42am
Originally posted by Windows98 Windows98 wrote:

Originally posted by wardog wardog wrote:

I believe that kit is made of Hynix based ICs.

That's fine, but it doesn't run *at all* on BIOS v 2.00. I can run it just fine @ 2666 on 1.64. 

I knew my kit wasn't on the QVL, so after I updated my BIOS and it died, I went out and bought the Corsair kit that everyone got with their review samples and it behaved the same exact way. Error 07. 

Originally posted by wardog wardog wrote:

That concerns me. What have or haven't you done to confirm CnQ is or isn't functioning?
My past several builds have been Intel iX builds, so my gut is telling me to look at two things: clock speed and voltage. If Ryzen operates in a different manner (or reports it in a different manner) then that would explain my inability to see it. 

I've noticed voltage is kind of a mixed bag with monitoring apps. CPU-Z is constantly reporting my chip @ 0.64V and it never changes even if I throw a load at it. I just chalked this up to CPU-Z being wrong. HWInfo and AIDA64 seem to report it accurately. 

However, nearly *every* monitoring app shows my clock speeds at ~3692mhz, which indicates that:

1: My processor has a random ~+100mhz clock over the base clock for some reason.
2: It isn't falling into the lower p-states. 

I have been running HWInfo (v5.5) while I've been typing this post. The minimum reported core speed on all cores is 3,692.7mhz, which is approx 100mhz higher than my base clock speed should be (3,600mhz). I can see that one of the cores did kick into XFR, but that state also has a +100mhz bump 4,091mhz instead of ~3,991mhz. 

per the latest AMD Ryzen blog post, monitoring apps will evidently show last reported clock the core was on prior to it entering it's low P-state...so even if the apps are misreporting the various P-states on my chip, they should show the last "step" before the core goes into one one of those low-power states. In my case, all of the cores are reporting a minimum clock of 3692.52MHZ ...not the 3200-3400 reported in the post above. 

I am also running the latest Ryzen Chipset drivers and the Ryzen power plan (the latter is linked in the blog post above).

I've tried toying with the "minimum processor state" options in Power Settings and that makes no difference either (that I can tell).

AS SUCH...

I have been operating under the assumption that this would all get "fixed" in a microcode update from AMD...hence giving me another reason to update to the latest BIOS. 

Edit: I just rebooted my PC and checked out BIOS to make sure I hadn't inadvertently set a custom ratio on the CPU. Everything is on auto except my RAM (I loaded an XMP profile, backed the clock speed from 3000 -> 2666).



Tonight I'll take a see at what you describe. Yet if any app, CPU-z or other, shows 0.64v rest assured it read that voltage from sometime since your last boot into Windows.

I have noticed Ryzen seem to take a longer time than I'm used to in settling down after a (re-)boot. Minutes actually. Sure you're not "waiting long enough' to notice this?




Back to Top
Windows98 View Drop Down
Newbie
Newbie


Joined: 11 Apr 2017
Status: Offline
Points: 22
Post Options Post Options   Thanks (0) Thanks(0)   Quote Windows98 Quote  Post ReplyReply Direct Link To This Post Posted: 11 Apr 2017 at 7:56am
Originally posted by wardog wardog wrote:

I believe that kit is made of Hynix based ICs.

That's fine, but it doesn't run *at all* on BIOS v 2.00. I can run it just fine @ 2666 on 1.64. 

I knew my kit wasn't on the QVL, so after I updated my BIOS and it died, I went out and bought the Corsair kit that everyone got with their review samples and it behaved the same exact way. Error 07. 

Originally posted by wardog wardog wrote:

That concerns me. What have or haven't you done to confirm CnQ is or isn't functioning?
My past several builds have been Intel iX builds, so my gut is telling me to look at two things: clock speed and voltage. If Ryzen operates in a different manner (or reports it in a different manner) then that would explain my inability to see it. 

I've noticed voltage is kind of a mixed bag with monitoring apps. CPU-Z is constantly reporting my chip @ 0.64V and it never changes even if I throw a load at it. I just chalked this up to CPU-Z being wrong. HWInfo and AIDA64 seem to report it accurately. 

However, nearly *every* monitoring app shows my clock speeds at ~3692mhz, which indicates that:

1: My processor has a random ~+100mhz clock over the base clock for some reason.
2: It isn't falling into the lower p-states. 

I have been running HWInfo (v5.5) while I've been typing this post. The minimum reported core speed on all cores is 3,692.7mhz, which is approx 100mhz higher than my base clock speed should be (3,600mhz). I can see that one of the cores did kick into XFR, but that state also has a +100mhz bump 4,091mhz instead of ~3,991mhz. 

per the latest AMD Ryzen blog post, monitoring apps will evidently show last reported clock the core was on prior to it entering it's low P-state...so even if the apps are misreporting the various P-states on my chip, they should show the last "step" before the core goes into one one of those low-power states. In my case, all of the cores are reporting a minimum clock of 3692.52MHZ ...not the 3200-3400 reported in the post above. 

I am also running the latest Ryzen Chipset drivers and the Ryzen power plan (the latter is linked in the blog post above).

I've tried toying with the "minimum processor state" options in Power Settings and that makes no difference either (that I can tell).

AS SUCH...

I have been operating under the assumption that this would all get "fixed" in a microcode update from AMD...hence giving me another reason to update to the latest BIOS. 

Edit: I just rebooted my PC and checked out BIOS to make sure I hadn't inadvertently set a custom ratio on the CPU. Everything is on auto except my RAM (I loaded an XMP profile, backed the clock speed from 3000 -> 2666).



Edited by Windows98 - 11 Apr 2017 at 8:08am
Back to Top
 Post Reply Post Reply Page  12>
  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 0.297 seconds.