I've had some stability problems for 1-2 months, which possibly caused by RAM seating(But memtest always passing, and memory corruption BSOD/SEL entry occuring after a week or two). So after fully reseating all DIMM modules I decided to try BIOS update again(This time to v1.5), but was prepared for it to not work well for me.
I used MicrocodeExtractor tool to check the microcode version in Smicro 1.5 BIOS, featuring AGESA 1.0.0.8, and found that they bundled newer microcode which has Inception vulnerability fix inside(Which, is bundled officially by AMD only in AGESA 1.0.0.9). All well - Updated BMC and BIOS with full reset on both and got code 91 again like in previous time.
I tested all possible settings in BIOS including new options(Used riser cable with PCI-E USB card to not pull/push into the PCI-e clot on the mobo btw each setting test), to no avail. Restored my settings for this board and removed the PCI-e USB card and dropped the GPU in - It worked with this BIOS.
Added secondary PCI-E USB card which worked during previous 1.4 try(For one Boot only), and noticed it had solder points which can touch the Smicro radiator and suspected it was the cause of it not working more than once during previous try. So I slot-switched it with my Raid controller and it works just fine. PERR errors are not happening anymore on boot with this BIOS, while 1.1a got me errors on different slots every boot.
Both USB cards have ASmedia controller with a Via hub(Working one is USB 3.0, problematic is USB 3.1). I've order another one by Startech which features the same newer ASMedia 3.1 controller, but without hub - It used two controllers directly servicing 2 ports each - didn't tested it yet, because stability test was more important.
The system runs well and stable for more than 2 weeks now and this is great. The only downside is that I feel CPU temps got up 5-7 degrees with this BIOS...
One air cooler announced and available for purchase:
EDIT: Sry, this one seems only to match the smaller SP6/sTR5 socket...
Which proofs of rumors that TR Pro 7000 will end up in a SP6 socket instead of SP5...
Another one, this time AIO only announced:
Some ppl suspect it will not work good top mounted in horizontal position because pump is in the radiator and not on the waterblock. Interesting ...
I used MicrocodeExtractor tool to check the microcode version in Smicro 1.5 BIOS, featuring AGESA 1.0.0.8, and found that they bundled newer microcode which has Inception vulnerability fix inside(Which, is bundled officially by AMD only in AGESA 1.0.0.9). All well - Updated BMC and BIOS with full reset on both and got code 91 again like in previous time.
I tested all possible settings in BIOS including new options(Used riser cable with PCI-E USB card to not pull/push into the PCI-e clot on the mobo btw each setting test), to no avail. Restored my settings for this board and removed the PCI-e USB card and dropped the GPU in - It worked with this BIOS.
Added secondary PCI-E USB card which worked during previous 1.4 try(For one Boot only), and noticed it had solder points which can touch the Smicro radiator and suspected it was the cause of it not working more than once during previous try. So I slot-switched it with my Raid controller and it works just fine. PERR errors are not happening anymore on boot with this BIOS, while 1.1a got me errors on different slots every boot.
Both USB cards have ASmedia controller with a Via hub(Working one is USB 3.0, problematic is USB 3.1). I've order another one by Startech which features the same newer ASMedia 3.1 controller, but without hub - It used two controllers directly servicing 2 ports each - didn't tested it yet, because stability test was more important.
The system runs well and stable for more than 2 weeks now and this is great. The only downside is that I feel CPU temps got up 5-7 degrees with this BIOS...
One air cooler announced and available for purchase:
EDIT: Sry, this one seems only to match the smaller SP6/sTR5 socket...
Which proofs of rumors that TR Pro 7000 will end up in a SP6 socket instead of SP5...
Arctic Announces the Freezer 4U-M CPU Cooler for Servers
Another one, this time AIO only announced:
SilverStone Intros XE360-SP5 AIO Liquid CPU Coolers for AMD Socket SP5
SilverStone introduced the XE360-SP5, an all-in-one liquid CPU cooler for AMD Socket SP5, making it fit for servers and workstations based on the EPYC "Genoa" and "Genoa-X" processors, and its upcoming Ryzen Threadripper HEDT processors, assuming AMD sticks to this socket infrastructure. The...
www.techpowerup.com
Some ppl suspect it will not work good top mounted in horizontal position because pump is in the radiator and not on the waterblock. Interesting ...
Last edited: