@pershoot
The only reason i changed to NvmExpressDxe_4 is to restore the full functionality of the module like it was in v3 module but keep the improvements of the vSmall module. The _small version was already an equivalent of NvmExpressDxe_4 but some fonctionality were removed to make it small. Those are restored in the ordinary v4 version of the module. Differrences are things that usually no one ever use. Like be able to unload a driver in EFI Shell and some debugging support. Things like that. For more details, see this post about the changes between the _small module versus the old v3 module. As for the speed between both (Small and v4), it is supposed to be the same code that manage it. For more on that part, see this thread.
Nothing else in the bios was touched in the last update and nothing in this bios was ever changed related to the memory. If you get more problems with that, i suggest to test your memory just to be sure they don’t have any problems. Windows 10 got some memory test when you can access the boot menu but they are rather generic. Memtest86 or Memtest86+ are even better programs to test memory and test them more deeply.
I installed V14 yesterday, all seems operational … if anything v14 seems smoother than v12 …
Booting is fast and the bios seems to do everything to evaluation specs :
Power and powerpack function = operational & all setting available.
CPU functions all work (cpu is quiet on cool and quiet)
all the tabs on the hardware page seem to function, The fan controls work (on monitor page)
Overclock functions (DP)
V14
evaluated by DT
@DukeThrust
Concerning your comment about the spectre/meltdown patch, it all depends on the CPU you have and which microcode your PC is using and also which variant of the vulnerability you look at. The Bios has all the latest microcodes installed and you can easily verify that by using the latest version and database of MC Extractor. But that being said, up to date, of the 11 microcodes that are in this Bios, only 2 of them have received some mitigation patch from AMD. Those 2 are the microcodes 600F12 and 600F20.
Those 2 received the mitigation patch for CVE-2017-5715 [branch target injection], this one is totally fixed with that patch.
They did NOT receive any mitigation patch for CVE-2017-5754 [rogue data cache load] (AKA Meltdown) for the simple reason that all AMD processors are naturally immune to this one and don’t require any correction at all.
They did NOT receive any mitigation patch for CVE-2018-3639 [speculative store bypass] because this one was discovered later than the others and i expect that a new patch from AMD will come out for that later in the future.
You can test yourself all your vulnerability by installing the latest version of the “Speculationcontrol” module in PowerShell. You can google it if you don’t already have it.
As an example, my CPU is a FX-8150 and is using the 600F12 microcode and this is the report i get with the v14 Bios:
For more information about the output below, please refer to https://support.microsoft.com/en-in/help/4074629
AMD CPU detected: mitigations for branch target injection on AMD CPUs have additional registry settings for this mitigation, please refer to FAQ #15 at https://portal.msrc.microsoft.com/en-us/…isory/ADV180002
Hardware support for branch target injection mitigation is present: True
Windows OS support for branch target injection mitigation is present: True
Windows OS support for branch target injection mitigation is enabled: True
Speculation control settings for CVE-2017-5754 [rogue data cache load]
Hardware requires kernel VA shadowing: False
Speculation control settings for CVE-2018-3639 [speculative store bypass]
Hardware is vulnerable to speculative store bypass: True
Hardware support for speculative store bypass mitigation is present: False
Windows OS support for speculative store bypass mitigation is present: True
Windows OS support for speculative store bypass mitigation is enabled system-wide: False
BTIHardwarePresent : True
BTIWindowsSupportPresent : True
BTIWindowsSupportEnabled : True
BTIDisabledBySystemPolicy : False
BTIDisabledByNoHardwareSupport : False
KVAShadowRequired : False
KVAShadowWindowsSupportPresent : True
KVAShadowWindowsSupportEnabled : False
KVAShadowPcidEnabled : False
SSBDWindowsSupportPresent : True
SSBDHardwareVulnerable : True
SSBDHardwarePresent : False
SSBDWindowsSupportEnabledSystemWide : False
This cover the hardware part.
As for the software part (OS), be aware that even if CVE-2017-5715 [branch target injection] is showing as enabled for an AMD CPU, it is not completely protected because this mitigation include 3 scenarios to be protected and only 2 of them are enabled by default in Windows for AMD. The last one must be manually enabled by adding the related registry keys for those who want that additional protection. See the link and the FAQ#15 in the spoiler for all the details.
Also be aware that even if the hardware and software patch would be present for CVE-2018-3639 [speculative store bypass], the sofware patch is not enable by default in a Windows client. Only the Windows Server version have this patch enabled by default. This protection can be enable also by adding the related registry keys for those who want it.
My last comment and warning is to state the obvious: For each additional protection that is enabled in the OS by modifying the registry, there will be an additional performance penalty of the system. So only add a protection if you absolutely need it and don’t mind the cost in performance.
Phoenix48
I just update my motherboard with the modded version (stable orochi 15.0.0.0) and the pc is more stable and the temeprature lows 10 degrees.
How is the situation with the modded bios with orochi 15.0.0.7??
@Fexthor24
Still trying to make it work with the v1.5.0.2 because it is easier (kind of…) before trying to work again with v1.5.0.7.
That being said:
Here is an other try:
BIOS with AGESA v1.5.0.2 updated to v7 Beta
Changelog:
[Changed] - In short: Everything updated the same as the stable v14
[Changed] - AmdAgesaDxeDriver changed to v1.5.0.2 but nothing else related to the AGESA. (AmdProcessorInitPeim is original in this version)
Be warned that any program analyzing the Bios that report the AGESA version by pulling the number from the AmdProcessorInitPeim module will still report v1.5.0.0. Those pulling the version number from AmdAgesaDxeDriver will report v1.5.0.2.
The goal of this version is getting reports to know if all the same problems as in v6.1x are still there or not (see Beta in first post), and to know if new ones appears.
BIOS with AGESA v1.5.0.2 updated to v7 Beta:
Command Rate gets locked at 1T (when attempting to switch to 2T, 2T is shown on next boot in to bios screen but 1T is still shown as current setting). Can’t really boot with it and any type of higher DDR clock other then optimized defaults (I run 4x8GB modules).
Hi Phoenix48
You can ask - is it working on 15 bios?
@Ozi
Not sure i understand your question right.
If you are asking if i am working on the next stable version 15, then not yet, because there was nothing new lately that could have been updated in the Bios. So the stable v14 is still up to date for now.
If you are asking if the memory settings feature related to post #106 is working in v15 then there is no v15 Bios yet. But yes, it is working perfectly in the stable v14 Bios.
I will wait for the 15th build, I hope that there will be good and useful changes …
I have the same computer configuration as you have.
Is this bios for amd fx 9590 too?
Do you think it supports it? is a CPU with a high energy value that requires the motherboard, I know that the formula-Z supports it but apart from the difference of Digi + I do not seem very different as cards
@pipes80
You almost said it all yourself… Yes, the Formula and the Bios can support the FX-9590. BUT the Digi+ is different. To say it simply, the Digi+ of the Formula is not as resistant as the Digi+ 2 version of the Formula-Z. And this means that even if it can support it, i would not recommend overclocking it to the extreme because the card may not support it.
Bothom line, you can use the FX-9590 on the Formula without any problem as long as you don’t overclock like crazy and overload the Digi+.
Also a warning to state the obvious: It takes a liquid system to cool the FX-9590 and also protect the motherboard from the heat it will produce, but that is the same if used with the Formula-Z. No air system will do anywhere with that CPU.
hi, can any one implement nvme for that motherboard, or it is possible?
@rednik
This bios mod already add support for nvme, see this line in the first post:
[Added] - -----> NvmExpressDxe_4 (New feature) Added support for boot from SSD with PCIE
@Phoenix48 i have an issue with the last stable version, inside the bios config, the names of the connected HDD, SSD or ODD doesn’t appear (not in the boot settings) in the SATA settings where you can actuvate or deactivate the SATA ports, and other thing (not an issue) i heard a rumor about the AMD AHCI ROM and the OROM AMD RAID MISC, about an update, is that true?? And the microcodes for the FX8300 are the last ones?
Pd: Sorry for my english, it’s been a while…
@Fexthor24
Your drives should be listed in the Bios without problem. One exception i know of is if you have Windows boot in UEFI on a GPT formated disk. In that case your drive should appear under a name like "PATA disk". Another thing to try is to press F8 at startup and look if you see your drive there.
No, there is no new update for AMD AHCI or AMD RAID OROM at this time. They are the latest.
And yes, the microcode for the FX8300 is the microcode 600F20 and it is the latest one.
There is no way to this motherboard to support PCI Express 3.0?? And i have another question? Is possible to this motherboard support a bridgeless crossfire (2 RX560 cards for example) for the subject of the pci express 3.0
@Fexthor24
Unfortunately, the PCIe version is a hardware feature and can’t be changed without changing the motherboard. As for the bridge aspect in a bridgeless Crossfire, this only depends on the ability of the video cards to support it. If my memory is right, RX290 and up can support bridgeless Crossfire. So with the Formula, this would mean that your 2 cards in Crossfire would run in bridgeless mode at the maximum speed of PCIe 2.0 x16/x16 if they are both installed in the slots that support x16 speed.
Phoenix48
5 months have passed since the 14th version of the BIOS mod - I would like to know if the work on the new version is underway?