Due to my own customer demand for answers I've had to spend time over the past few days further debugging Gigabytes mess. Here is my report;
Missing options / features in F22 (AX370 Gaming 5)
Overclocking & memory related
CLDO_VDDP control CLDO_VDDP voltage Last present: Firmware F9f
BankGroupSwap Last present: Firmware F9f
BankGroupSwapAlt Last Present: Never
Channel Interleaving Rank Interleaving
Both present but need exposing through the use of AMIBCP
General useful options missing
AMD PSP Last present: Never
Spread Spectrum and related options USB MSI (Message Signal Interrupt) MsiDis in HPET Chipset Power Saving Features USB PHY Power Down C1E Support
All present, but needs exposing with AMIBCP. Disabling USB MSI works around the bug some (all?) USB keyboards have of not waking from sleep if allowed to enter sleep mode while in the UEFI.
Timer Tick Tracking Clock Interrupt Tag EHCI Traffic Handling Fusion Message C Multi-Core Fusion Message C State
All present but need exposing with AMIBCP. Timer Tick and Clock Interrupt are particularly useful for pre-emptive multitasking and could potentially resolve SMT anomalies in certain applications.
Most likely goes without saying but F9f despite having plenty of issues of its own is about the most reliable firmware revision to stick with at this time later revisions are just a half baked mess.
One of my ram module too died few days back. Out of nowhere. I don't want to believe it was the motherboard but then memory doesn't die for no reason.
It usually dies due to heat degredation due to lack of sideways cooling because systems often have non-traditional CPU fans these days, in combo with poor heat sinks which act more like insulators than aids to cooling in some of these more tarted up 'gaming' strips that are popular. That or the old standby of static damage.
If your mobo killed it, it would usually be due to over-volting due to a firmware bug. Usually the readout from a voltage checker app would confirm this. Have you seen ram over-volting going on here?
Thanks for the reply mate. Ram was running 2400 Mhz on XMP1 profile. I believe the voltage was 1.35. And I have tried to maintain good airflow in my case. Have multiple fans with proper airflow. I am not sure what happened there. System was shut down properly one day and the next day it started with a bootloop. One stick works and the other gets stuck at 0d. Other than dead module could it be due to something else? With both sticks Ram does get detected in the BIOS but half of it is unusable as system reserved from within Windows.
It usually dies due to heat degredation due to lack of sideways cooling because systems often have non-traditional CPU fans these days, in combo with poor heat sinks which act more like insulators than aids to cooling in some of these more tarted up 'gaming' strips that are popular. That or the old standby of static damage.
If your mobo killed it, it would usually be due to over-volting due to a firmware bug. Usually the readout from a voltage checker app would confirm this. Have you seen ram over-volting going on here?
Thanks for the reply mate. Ram was running 2400 Mhz on XMP1 profile. I believe the voltage was 1.35. And I have tried to maintain good airflow in my case. Have multiple fans with proper airflow. I am not sure what happened there. System was shut down properly one day and the next day it started with a bootloop. One stick works and the other gets stuck at 0d. Other than dead module could it be due to something else? With both sticks Ram does get detected in the BIOS but half of it is unusable as system reserved from within Windows.
Get a program that monitor the volts, dont just go by whats in bios make sure you are plugging the RAM in tightly, touch the frame of the case with power supply plugged into wall to ground yourself out before removing parts. if you have a working ground, power supply doesn't have to be on to ground the case. your ram speed maybe incorrect, ram controller on CPU could be bad or you damaged pins on the board, clean the sticks with alcohol and Q tip and insert then before it dries with power off
RAM is stacked so tight i use a fan that attached to the ram to use XMP
The motherboard only provides power and a connection to the the CPU so if only one is bad it's probably not the board if volts don't go over 1.5, try each stick one at a time and make sure you are plugging them in to correct matching color of the insert or you won't get dual channel and a XMP profile could cause one to malfunction.
High voltage bug also on F22b!!!! Be very careful, my cpu(Ryzen 5 1600) just received 1.68Volts on F22b firmware!!!! Fortunately it survived, i stopped the system from the power supply directly the moment i saw the voltage in cpu-z. To reproduce it (if you dare ) just add +0.16 V voltage dinamic offset, then without changing that to zero put manual voltage to 1.35. Also set LLC to Extreme.
High voltage bug also on F22b!!!! Be very careful, my cpu(Ryzen 5 1600) just received 1.68Volts on F22b firmware!!!! Fortunately it survived, i stopped the system from the power supply directly the moment i saw the voltage in cpu-z. To reproduce it (if you dare ) just add +0.16 V voltage dinamic offset, then without changing that to zero put manual voltage to 1.35. Also set LLC to Extreme.
Hi again, try disabling C-states, reboot, load OS. Reboot, and enable hte C-states in AUTO. Work for my to obtain a normal CPU Vcore. I set the Vcore manually.
Right now I'm Offline I have made 90 posts
I joined May 2017
I've received 13 likes
My gender is Male
I currently have no messenger contact information filled out.
I currently have no social network information filled out.
When RGB Fusion is running, should the POST code readout on the board change to either A0 or AD? I ended up uninstalling it again just to be safe, but I can confirm that the extremely high CPU usage from previous versions is gone (yay!!!). Also, I think this may be a bug with the current F22 bios revision, but the CPU RGB header does not appear to follow the exact same setting as the other RGB leds and headers on the board. Not sure if anyone else has seen this, but to my eye the output tends to be whiter than the other ones.
In regards to the high voltages people have been reporting on recent bioses. I have not seen higher than 1.456V at idle in F22 (which is the same value HWiNFO64 has been reporting since I got this board last summer).
That value is only there when idling at the desktop. When a load like Cinebench or CPU-Z's benchmark is running, the voltage goes straight to something right around or below 1.2V. I have my CPU voltage set to auto, with CPB left to auto as well.
If it helps, I also have XMP turned on and have set DDR4 voltage to 1.37V (which reads as 1.38V in Windows) and the VSOC voltage to 1.125V (which reads as 1.116V). My idle temps seem okay (29C as I am typing this). If I run a game for a few hours and come back to HWiNFO64 the max temp I see is usually somewhere between 50C and 60C, but the average is usually a bit lower than that.
Any ideas on how to limit that upper (XFR?) voltage value? Since it is the out of the box Auto setting, is it safe, or is this just how Gigabyte (or Ryzen in general) boards handle idle?
My Signature... Steam ID - Arekkusu [http://steamcommunity.com/id/icypwnage]
Main System -:- (CPU) AMD Ryzen R7 3800X [@ Stock w/PBO] [w/ Corsair H115i] - (GPU) Nvidia GTX 1070 FE x2 (SLI) - (PSU) Corsair RM850x (MB) Gigabyte AX370 Gaming K7 [BIOS F50a] - (MEM) G.Skill F4-3200C14D-32GTZR [@ 3200MT/s] - (OS) Windows 10 Pro [1909]
That's 3 people who reported Boot Loops including myself with F22 on the K7.
Other people reporting Hpet Issues.
I have an ongoing discussion about the motherboard in Gigabyte E-Support.
I let them know about my issue and you guys should let them know as well.
Again for me it happens when I overclock the CPU then boot loop. Any amount of overclock.
I tried using their so-called "support" on behalf of one of my customers who purchased several systems and to call what they offer "support" is a joke. They asked me to test a beta firmware and let them know if the problem was resolved so I went the extra mile and comprehensively tested the beta firmware and compiled a detailed report clearly describing every bug I found and how to reproduce it. I never heard back from Gigabyte. That was over 2 months ago now.
They keep telling me to perform Tests for them too.
They recently wanted me to "test" F22 cpu overclock with no ram overclock (meaning running the ram at default).
The problem is I never overclock without overclocking the ram. This test has no value to me. Default Ram settings is not even rated settings for most DDR4.
Usually they try to get me to do additional tests and I keep refuting it as useless and not in my best interests. Then they talk to the engineers.
I tried using their so-called "support" on behalf of one of my customers who purchased several systems and to call what they offer "support" is a joke. They asked me to test a beta firmware and let them know if the problem was resolved so I went the extra mile and comprehensively tested the beta firmware and compiled a detailed report clearly describing every bug I found and how to reproduce it. I never heard back from Gigabyte. That was over 2 months ago now.
They keep telling me to perform Tests for them too.
They recently wanted me to "test" F22 cpu overclock with no ram overclock (meaning running the ram at default).Â
The problem is I never overclock without overclocking the ram. This test has no value to me. Default Ram settings is not even rated settings for most DDR4.
Usually they try to get me to do additional tests and I keep refuting it as useless and not in my best interests. Then they talk to the engineers.Â
It's to rule out RAM when troubleshooting. All DDR4 RAM is supposed to work correctly at jdec speeds (default), so if you get no errors or crashes with jdec RAM and default CPU, but errors or crashes with jdec RAM and OC'd CPU, then your RAM is probably not the cause (probably, because some RAM instability may be caused by a CPU OC due to more frequent memory access by the CPU).
arekkusu, the LEDs connected to the header may have slightly different resistance, which would cause them to be a little brighter or darker. I think there is a way to calibrate rgbfusion, but I don't remember where I read that.
I tried using their so-called "support" on behalf of one of my customers who purchased several systems and to call what they offer "support" is a joke. They asked me to test a beta firmware and let them know if the problem was resolved so I went the extra mile and comprehensively tested the beta firmware and compiled a detailed report clearly describing every bug I found and how to reproduce it. I never heard back from Gigabyte. That was over 2 months ago now.
They keep telling me to perform Tests for them too.
They recently wanted me to "test" F22 cpu overclock with no ram overclock (meaning running the ram at default).
The problem is I never overclock without overclocking the ram. This test has no value to me. Default Ram settings is not even rated settings for most DDR4.
Usually they try to get me to do additional tests and I keep refuting it as useless and not in my best interests. Then they talk to the engineers.
To be honest with the comprehensive detailed report I gave Gigabyte these are the issues that should have been resolved a long time ago;
1. Saved profiles not storing all information (CPU multiplier, RAM timings, etc) 2. 2400 and 2666 memory dividers causing a no POST when XMP is enabled (F4-3200C15D-16GVK, Samsung B-Die) 3. SPD \ manually set memory timings not being displayed in the UEFI for at least half of the parameters 4. Some LED preset colours (orange particularly) are not the colour they are supposed to be. 5. High vcore SoC voltage when left at "Auto" when CPU overclocking 6. Disabling GearDown causes a no POST when XMP is enabled 7. There needs to be a key-in method for RGB Fusion because the mouse DPI setting and / or polling rate in the UEFI is not sensitive enough to make using the colour wheel practical.
To my knowledge and testing none of these problems have been fixed although I gave up testing the saved profiles issue compared to everything else thats nothing at all. I reported these problems to GB at the time of beta firmware F9f. On top of all of those problems you can now add with F22;
8. Disabling HPET results in a 0d POST code when restarting Windows (after further testing also happens when saving changes and exiting the UEFI after the first initial reboot disabling HPET) 9. P-State overclocking is broken yet again, POST code 85 with +200MHz (R5 1600) 10. XMP is now completely broken with the same F4-3200C15D-16GVK kit as I used months ago. (Yes I set it aside, I had a feeling back then GB products were going to cause me headaches with customers) 11. MANY missing options within the UEFI that either used to be there but no longer are (CLDO_VDDP, BankGroupSwap, Channel & Rank Interleaving, etc, for example), or should be there not just for flexibility but in some cases needed to resolve bugs or act as potential workarounds for certain issues (USB MSI, Timer Tick Tracking, Clock Interrupt Tag, Spread Spectrum, etc).
Ironically it's all the options GB have removed or hide from the user that function perfectly. I know this because I'm a very hands on person I get involved with things I don't sit back in my big comfy chair and let everyone else do the work as such I have become very adept at many things after all the years I've been involved with technology now. If I say theres a problem, you can take that to the damn bank with the knowledge that I have thoroughly tested to narrow down or fully isolate the cause and if there is a workaround I've also found that too. You can also take it to the bloody bank when I say All of the options that are missing are contained in the main firmware image but hidden from the user, or are present in the AMD CBS file GB just have not bought them forward in to the main firmware image, and if their too lazy to do that they can just remove the hidden / suppress tag in the AMD CBS file and they will then be directly displayed in the AMD CBS menu. One of my adept skills is firmware so yes GB, I know what you have told me is BS. GB can EASILY fix at least 8 of the 11 issues I have found.
So there you have it, the firmware really has got worse over time.
I can see why GB would ask you to test what they have but without specific details I couldn't say if what they are asking you to do is a waste of time. All I can say is that GB asked me to do things I already said I had tested so they were blatantly stalling or they are so inept they have people who are absolutely clueless acting as support reading from scripts.
Right now I'm Offline I have made 90 posts
I joined May 2017
I've received 13 likes
My gender is Male
I currently have no messenger contact information filled out.
I currently have no social network information filled out.
To be honest if everything Stevenx1x mentioned in his post got fixed, I would have no issues with the board anymore.
Looking back on it now, the BIOS has been the most lacking aspect of this board.
Also, per the second post on the first page of this Thread, the current list of BIOS revisions are labeled with F at the beginning. Is there any reason we are still running Beta BIOS revisions? I realize that AMD CPUs are less popular than Intel ones, but this is supposed to be Gigabyte's top end AM4 board (which is why I got this one over some of its competition), isn't it supposed to be the best supported one?
My Signature... Steam ID - Arekkusu [http://steamcommunity.com/id/icypwnage]
Main System -:- (CPU) AMD Ryzen R7 3800X [@ Stock w/PBO] [w/ Corsair H115i] - (GPU) Nvidia GTX 1070 FE x2 (SLI) - (PSU) Corsair RM850x (MB) Gigabyte AX370 Gaming K7 [BIOS F50a] - (MEM) G.Skill F4-3200C14D-32GTZR [@ 3200MT/s] - (OS) Windows 10 Pro [1909]
Mar 29, 2018 10:18:06 GMT -8 | amd6string likes this
Has the fastboot resume from standby fan bug been fixed yet?
If not, can we have the option to have 'last HDD only' enabled regardless of fastboot setting?
Care to elaborate on this bug? I'm trying to build a comprehensive bug report list that I'm going to send to GB. Don't be surprised if they stonewall, but at least it can categorically then be said that they have had a comprehensive bug report along with how to replicate each bug sent to them. I've got about 13 on the list so far.
Has the fastboot resume from standby fan bug been fixed yet?
If not, can we have the option to have 'last HDD only' enabled regardless of fastboot setting?
Care to elaborate on this bug? I'm trying to build a comprehensive bug report list that I'm going to send to GB. Don't be surprised if they stonewall, but at least it can categorically then be said that they have had a comprehensive bug report along with how to replicate each bug sent to them. I've got about 13 on the list so far.
Basically, after I put my PC to sleep and resume, all the fans in my system behave incorrectly either by stopping or spinning up to max if I have fastboot set to 'enabled' or 'ultra fast'. A cold or warm reboot fixes it or just disabling fast boot in the BIOS. Super strange, but has happened in every BIOS revision since I bought the mobo in July last year.
Right now I'm Offline I have made 22 posts
I joined January 2018
I've received 2 likes
My gender is Male
I currently have no messenger contact information filled out.
I currently have no social network information filled out.
I flashed back F7a (with AGESA: SummitPI-AM4 1.0.0.6) on my AX370-K7 board
- HPET turned OFF works fine . - Fast Boot works as intended, with HiberBoot enabled.
- UEFI GOP works correctly (not falling back and having UI elements streched while Windows is booting)
Windows 10 options aren't exposed from F10 and onwards. I believe "Other OS" gets defaulted and Fast Boot cannot work properly.
About LEDs, I have three RGB fans from Coolermaster (Air Balance) and colors from onboard LEDs and fans are mostly in sync. I don't have much to complain regarding this one.
About voltages, SoC voltage was always overvolting if left on Auto. Gigabyte says it's normal, but it's actually pretty abusive, and 1.1v should be more than sufficient for 3.9Ghz CPU and 3200Mhz DDR4 CL16.
Can Gigabyte get their ducks in order and fix these obvious issues, along with what's been reported by the other users on this thread?
I won't even try to set up Secure Boot from F10 firmwares and onward as this will probably be an utter disaster because CSM doesn't turn off, and pure UEFI boot is a requirement for SecureBoot.
greggg wrote: Apr 19, 2022 8:53:19 GMT -8 Tokenbad. You got to call the dirty buggers to get anything done. I did and was on hold for 2.5 hours before they talked to me.
slyhambo wrote: Apr 15, 2022 21:26:44 GMT -8 256 chars is not enough to ouline B450 Pro-WiFi Bios update to F40
bigdave wrote: Apr 15, 2022 7:35:10 GMT -8 techbox, I have only seen collisions occur with RAID drives.
techbox wrote: Apr 12, 2022 5:25:38 GMT -8 Here is what I found once I was in Disk Management,"Disk 1 Offline the disk is offline because it has a signature collision with another disk that is online" I right clicked it to change the letter and never even got the opportunity, kicked right on as D
kshipper wrote: Apr 9, 2022 17:11:14 GMT -8 Token: I guess no ones knows for sure. I suggest that you follow the format of the original email you first sent and received to obtain a RMA. This means you have to quote a case number in the subject line or your email could be filtered.
tokenbad wrote: Apr 5, 2022 18:01:55 GMT -8 How long does the gigabyte rma stuff take. I have sent in my card they got it back on feb 26th..says repaired back on march 5th..now no update even after trying to reach them through email.
kubi74 wrote: Apr 3, 2022 18:07:46 GMT -8 Help Gigabyte
jwsg wrote: Apr 2, 2022 10:44:57 GMT -8 memovsvs if youve bricked the m/b by changing secure boot key settings then try - remove battery for clear CMOS, trigger dualBIOS if you have it, use QFLASH+ to re-flash if you have it, try onboard video if you have an iGPU
memovsvs wrote: Apr 1, 2022 9:18:08 GMT -8 I need help. Changed CSM and secure boot and now the computer just doesn't do anything. Is there a way to reset bios? I have tried removing battery and clearing cmos with no result
jetema wrote: Mar 27, 2022 17:21:40 GMT -8 Just got an Aorus x570S Elite Ax Rev 1.1, when I turn it on you can hear the fans ramp up for a few seconds then 4 quick beeps followed by 3 more, nothing on screen. Saw some info saying 4 beeps is memory error, replaced the memory with same results,ideas?
jwsg wrote: Mar 27, 2022 11:10:53 GMT -8 vermillionkingpin Can you be more specific? You cant have Secure boot with CSM enabled. You must have a GPT format drive before disabling CSM. People have bricked their m/b changing secure boot settings the wrong way. Change things one step at a time!
vermillionkingpin wrote: Mar 27, 2022 10:07:17 GMT -8 Anyone having issues with their z390 not booting after enabling CSM, Intel PTT and secure boot?
bigdave wrote: Mar 25, 2022 13:32:51 GMT -8 abasssaial, Last time I had a c1 error on my Z690 it was a memory issue. Make sure that your RAM is on the Gigabyte's QVL. I used Q Flash Plus to flash the BIOS and that fixed the issue for me.
abasssalal wrote: Mar 24, 2022 1:09:55 GMT -8 I recently upgraded my pc with a z690 Aorus Ultra, 12700k CPU, and 2x16 XPG DDR5 6000MHz. For the first 5 days,the PC was working normally after that when I was power on my pc, the pc not Booting and the motherboard starts flashing red, bringing up code C1*
r3volt97 wrote: Mar 20, 2022 23:07:52 GMT -8 Hi guys ,
i've got serious issues with my Red Devil RX 6800 not displaying anything at all - vga led on most of the time . Tested the gpu in another rig and it worked , now back to X570 Master - gave it a shot with a different gpu ( GTX 970 ) through a
kshipper wrote: Mar 20, 2022 12:57:07 GMT -8 hairyscot: I assume you have a problem? We don't have to own/use your board to be helpful. Maybe you could post about your problem on the forums on the left under Intel Motherboards?
hairyscot wrote: Mar 18, 2022 2:58:58 GMT -8 Anyone have a Gigabyte B560 HD3 motherboard system?
jwsg wrote: Mar 5, 2022 5:58:10 GMT -8 foot5001 Be very careful with secure boot key settings its possible to brick m/b. To enable secure boot I suggest: first get PC booting OK with CSM off, then toggle standard to custom, custom to standard accepting factory defaults, then enable sec boot
jwsg wrote: Mar 5, 2022 5:55:34 GMT -8 foot5001 Disabling CSM requires GPT format boot drive and UEFI compliant GPU - for the beeps check your GPU is compliant.
) just add +0.16 V voltage dinamic offset, then without changing that to zero put manual voltage to 1.35. Also set LLC to Extreme.
