New AM5 BIOS updates now available

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

   AGESA 1.3.0.0a introduced bank refresh modes and defaults to "fine granularity 
or mixed mode" when set to auto. This is not ASRock specific. I believe some 
brands have released updates to force the default to "normal" mode when auto is 
selected, as a workaround for the issue. The problem is caused by RAM manufacturers
not board manufacturers or AMD. That said, a BIOS level tweak can be implemented
as a workaround. I have no information as to whether ASRock will do so however.
The default, as set by AMD in their AGESA, is intended to provide optimum performance 
by utilizing the best secondary timings the RAM supports (TRFC1/2 
and TRFCSB). On some kits these timings are improperly inputted or are simply
incorrect, KingBank being one of the manufacturers with the issue. 

In prior BIOS versions (3.xx and earlier) these incorrect timings were ignored. 

There is a much better explanation here:

Sorry if I wasn't clear enough in my description but this is the issue I was 
referring to. 

Edited by Xaltar - Yesterday at 8:55am

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
Xaltar Members Profile Send Private Message Find Members Posts Add to Buddy List Moderator Group Joined: 16 May 2015 Location: Europe Status: Offline Points: 35243	Post Options Post Reply Quote Xaltar Report Post Thanks(0) Quote Reply Topic: New AM5 BIOS updates now available Posted: Yesterday at 8:53am
	AGESA 1.3.0.0a introduced bank refresh modes and defaults to "fine granularity or mixed mode" when set to auto. This is not ASRock specific. I believe some brands have released updates to force the default to "normal" mode when auto is selected, as a workaround for the issue. The problem is caused by RAM manufacturers not board manufacturers or AMD. That said, a BIOS level tweak can be implemented as a workaround. I have no information as to whether ASRock will do so however. The default, as set by AMD in their AGESA, is intended to provide optimum performance by utilizing the best secondary timings the RAM supports (TRFC1/2 and TRFCSB). On some kits these timings are improperly inputted or are simply incorrect, KingBank being one of the manufacturers with the issue. In prior BIOS versions (3.xx and earlier) these incorrect timings were ignored. There is a much better explanation here: Sorry if I wasn't clear enough in my description but this is the issue I was referring to. Edited by Xaltar - Yesterday at 8:55am


chanel303 Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: Yesterday Status: Offline Points: 30	Post Options Post Reply Quote chanel303 Report Post Thanks(0) Quote Reply Posted: Yesterday at 5:55am
	you are misleading the user. In firmware version 4.10, the usual asrock refresh mode does not turn on under any circumstances, this is a bios problem.

Xaltar Members Profile Send Private Message Find Members Posts Add to Buddy List Moderator Group Joined: 16 May 2015 Location: Europe Status: Offline Points: 35243	Post Options Post Reply Quote Xaltar Report Post Thanks(0) Quote Reply Posted: 22 Mar 2026 at 4:21am
	This is likely caused by a known issue with KingBank RAM. KingBank has an error in it's firmware that causes an issue with 4.xx BIOS versions. The issue is that "Mixed Mode" in previous versions was set to normal by default but on the newer 4.xx BIOS the default is "auto/default" which = enabled. Check under: OC Tweaker-> DRAM-> Change Refresh Mode = set to Normal See if that resolved your POST issue. It also explains your previous issues after updating. You can open a support ticket with KingBank if you like, they may be able to help resolve the issue at the firmware level. That said, flashing RAM firmware can result in bricking the RAM and needing to replace it so the workaround may be the safer bet. Mixed mode is the superior setting but owing to the issue in KingBank RAM's firmware it won't work as intended and causes the issue you described. Edited by Xaltar - 22 Mar 2026 at 4:27am


VVi3ard Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 22 Mar 2026 Status: Offline Points: 30	Post Options Post Reply Quote VVi3ard Report Post Thanks(0) Quote Reply Posted: 22 Mar 2026 at 3:16am
	Video and photo: https://drive.google.com/drive/folders/16Z132HFweaXrygJH4KnkHwe78OsvymyM?usp=sharing CPU: 9700X AMD64 Family 26 Model 68 Stepping 0 RAM: FPM5GM9505 (KingBank Dark Star Blade DDR5 UDIMM 6800MHz Dark Heatsink (Hynix M-die C32) 2x24 ?? (@6000, manual timings ZenTimings_Screenshot.png) MB: ASRock B850 Pro RS Video: Nvidia 4070 Power: Be Quiet! Pure Power 12 850W Atx 3.1 80Plus Gold (Bp004Eu) Age: 9600X - ~7 month. 9700x ~4 month. First, my 9600X stopped working. I put the PC into sleep mode before bed, and in the morning it wouldn't turn on again, with a CPU post code. I tried a lot of things, including FlashBack and clearing the CMOS, but nothing helped. I bought a 9700X, installed it, and everything worked right away. Later, BIOS 4.03 came out, and I updated it via FLASH INSTANT, but the PC wouldn't boot; it had a black screen and froze on the CPU. I spent three hours trying to get it to work (swapping memory, clearing CMOS, and trying to flash version 4.03 via FlashBack. Sometimes, after clearing CMOS, an image would randomly appear. I could even enter the BIOS or even start the system booting. But then everything would freeze, display random errors, or just freeze). I was very upset because I thought the 9700X was broken too. But then I decided to try reverting BIOS back to 3.40 via FlashBack, and lo and behold, everything worked correctly right away. The first boot was slow (about 40 seconds), but then everything worked fine. Then version 4.07.AS01 was released, and I didn't install it. Today I saw that version 4.10 was released and decided I should try installing it. And I ended up having exactly the same problems as before, only the system booted intermittently. Immediately after the update, the system froze on DRAM. I powered it off and back on. Same thing. I powered it off, cleared the CMOS, powered it on, and the image appeared. I was even able to enter the BIOS, clicked "Load EFI Defaults," and when I exited, I clicked "Save," after which the system froze. Powering it on again results in a black screen and CPU (it's completely unclear where it is, because the first two lights are on at the same time). I've tried many options. Immediately after clearing the CMOS, the system boots, passes all POST codes, but then randomly it either freezes on the boot screen, or sometimes it starts loading but immediately crashes with the ACPI_BIOS_ERROR 0xA5 error. After the freeze and a second reboot (with a power outage), everything freezes on the white VGA LED. This continues until I clear the CMOS. I also tried re-flashing 4.10 using FlashBack, with the same result. Sometimes I can enter the BIOS after flashing, but sometimes I get a black screen. As soon as I flash 3.40 via FlashBack, everything starts working correctly. I tried flashing 3.40, resetting the settings, saving, and then entering the BIOS again to flash 4.10. It still freezes on boot after flashing 4.10 (or it enters the BIOS but freezes when trying to save the BIOS settings). I don't know what to do anymore; I have no warranty :( And there are no service centers in my country. Video and photo: https://drive.google.com/drive/folders/16Z132HFweaXrygJH4KnkHwe78OsvymyM?usp=sharing

Xaltar Members Profile Send Private Message Find Members Posts Add to Buddy List Moderator Group Joined: 16 May 2015 Location: Europe Status: Offline Points: 35243	Post Options Post Reply Quote Xaltar Report Post Thanks(0) Quote Reply Posted: 22 Mar 2026 at 1:53am
	Good luck, the battle is half the fun


JamesL Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 20 Mar 2026 Status: Offline Points: 30	Post Options Post Reply Quote JamesL Report Post Thanks(0) Quote Reply Posted: 21 Mar 2026 at 10:10am
	Thanks brother! I?�ve already tried two Ryzen 9 9700X CPUs, and both ended up with the exact same result. The current CPU I?�m using can run DDR5 16G?2 at 8400 C34 with tight timings and pass all stability tests on another motherboard, and it can even boot at 8600 MT/s (maybe 8800 too). So right now I?�m not sure if it?�s still an IMC limitation of the CPU, a motherboard compatibility issue, or just some BIOS settings/ODT configurations that I haven?�t nailed down yet. I?�ll keep experimenting further. Appreciate the help!

Xaltar Members Profile Send Private Message Find Members Posts Add to Buddy List Moderator Group Joined: 16 May 2015 Location: Europe Status: Offline Points: 35243	Post Options Post Reply Quote Xaltar Report Post Thanks(0) Quote Reply Posted: 21 Mar 2026 at 2:23am
	I am afraid I haven't been active in the overclocking scene for quite some time. I don't have much to offer suggestion wise. What you describe does make sense from an overclocking perspective however. It's always easier to overclock a single RAM module than a pair. The fact that your B2 channel isn't able to match the A2 channel tells me that your IMC's A channel is more capable than channel B. If this is the case then it's highly likely that you won't be able to push it any further in dual channel regardless of settings. You would need another CPU with a stronger IMC. Back when I used to chase overclocking records I would often go through over 20 CPUs to find one that could handle the settings needed to achieve my goal. From what I have seen of DDR5 clocks on AM5, 8000 MT/s is pretty decent but I am guessing by the fact you picked up an OCF board that you are not looking to settle for decent. If you have access to more CPUs to try that would be my first stop. If not I have a feeling you have hit the limit of the IMC on that CPU. My guess would be that you will find stability probably at around 7600 MT/s in dual channel. Combining both channels never yields the same clocks as the weakest channel due to the added load and heat going through the IMC. Typically you can expect between 80% to 90% of the maximum stable clock of the weakest channel when using dual channel. I am sure this isn't what you wanted to hear but it's all I have to offer. There may be other forum members that have more recent experience and possibly some tricks I am unaware of. You can additionally ask ASRock themselves by opening a support ticket: https://tw.asrock.com/events/tsd.asp Though I suspect they won't be able to help much if at all. Good luck


JamesL Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 20 Mar 2026 Status: Offline Points: 30	Post Options Post Reply Quote JamesL Report Post Thanks(0) Quote Reply Posted: 21 Mar 2026 at 12:32am
	I?�m encountering critical overclocking stability issues with my ASRock X870E Taichi OCF (BIOS 4.10) paired with an AMD Ryzen 9 9700X and Biwin Origin Code 48GB?2 DDR5 memory. Here?�s a detailed breakdown: Memory Overclocking (8000+ MT/s): ??A2 Channel (2 DIMMs): Boots and tests stable at 8400 MT/s without issues. ??B2 Channel (2 DIMMs): Fails to boot even at 8000 MT/s (no POST). ??Dual Channel (A2+B2): Cannot boot at 8400 MT/s; struggles heavily to pass stability tests even at 8000 MT/s. ??Memory Refresh Mode: The option to set Normal mode is locked/unavailable in BIOS. CPU BCLK Overclocking: ??CPU base clock (BCLK) can only be stable at 105 MHz or lower. Any value above 105 MHz triggers error code 22 on POST. Memory Divider Overclocking: ??Attempting to reach 8400 MT/s via memory divider results in an immediate error code C5. Has anyone else experienced similar issues with this CPU/motherboard/BIOS/memory combination? Any suggestions for troubleshooting or BIOS settings to resolve these stability problems? Optional Short Version

Xaltar Members Profile Send Private Message Find Members Posts Add to Buddy List Moderator Group Joined: 16 May 2015 Location: Europe Status: Offline Points: 35243	Post Options Post Reply Quote Xaltar Report Post Thanks(0) Quote Reply Posted: 19 Mar 2026 at 3:27pm
	The X600 ITX contained in the DeskMeet has already received the AGESA 1.3.0.0a update so the X600 STX in the DeskMini should receive an update too.


Xaltar Members Profile Send Private Message Find Members Posts Add to Buddy List Moderator Group Joined: 16 May 2015 Location: Europe Status: Offline Points: 35243	Post Options Post Reply Quote Xaltar Report Post Thanks(0) Quote Reply Posted: 18 Mar 2026 at 6:08pm
	Interesting, I don't think I have seen a single report of boot issues (00 code or DRAM/CPU LEDs lit) on x600 based systems. Given they were designed for thermal management not outright performance it could be that the update isn't needed. I will see what I can find out.