tictoc

I can't read your post, do you want all the TC folders to PM @BWG or @BWG.

Without a BMC, initial setup will probably be a bit painful, with a whole bunch of connecting and disconnecting. Once you have everything set up definitely save the profile, and then never update the BIOS. As far as cooling goes, I would swap the out the AIO for a something like a NH-C14S if you have the clearance. If not, you could punch some holes in the side of the case and mount the AIO in the bottom of the chasis. With the AIO out of the way you will have proper front to back airflow, and with 3-4k+ RPM fans you can probably keep the GPUs from throttling. It's not going to be quiet, but it can be done. If that's not an option, then I would hunt around on ebay for some 980ti/titan heat-sinks. https://www.ebay.com/itm/154408987068

Make sure you have Above 4G encoding enabled in the BIOS. That board should allow you to save BIOS profiles to a USB stick. Then It's just a matter of loading the profile from the USB if you have to clear the CMOS.

That is looking great. I didn't know that MM was still making cases. I sold my MM case more than a decade ago. It is awesome to see an OG enthusiast company still around. Now you've got me configuring a custom case for my server. lol

Just realized that if I swap the switch and the patch panel in the rack, I can keep the wires from crossing and not have to label the patch panel 2,1,4,3,6,5... Now we have achieved maximum OCD relief.

These p177x units are a real ppd killer on AMD gear. 2M'ish average ppd down to 440k. ?

It would be nice to see the raw points. Tooltips would make it easy to see the comparison side by side, w/o having to add a column to one of the tables.

For all you cable management lovers, here's a little preview of my new 12U network/power rack:

Niche board with a niche use case, but every other Threadripper board I've ran worked without a hitch. I'm not sure why this doesn't work on this board. I don't care about using NVMe RAID mode or booting the OS off the drives (RAID will be handled by the OS), just looking for what has become a pretty standard option on most server motherboards. Right now I don't really need this to work, but I figured I might as well try to iron out all the possible kinks before I load the hardware in the box.

Piled on a few more MHz for May.

After you watch that video you are then forced to go down the "Vocal Coach Reacts to Nightwish" rabbit hole. *Edit* They are an epic band to see live, even if you're not into that style of metal.

I have been back-and-forth with ASRock, testing a number of beta BIOS. The same issue still exists with x8x8 working, but x4x4x4x4 just acts like x16. It is going up the chain to a BIOS engineer, so it seems as if there might be some progress in the near future on the bifurcation issues.

You could go browns with dampers. That is pretty quiet and what I use at work. If you're not bottoming out the keys, it is not too loud.

The handicapping is what makes it interesting. I have been crushed at work, so no time to start fiddling around with different GPUs. Hopefully things slow down and I can start testing some other GPUs next week.

RDNA2 GPUs have Vulkan Ray Tracing via the latest AMDGPU-Pro driver. 6900XT Metro Exodus - https://www.youtube.com/watch?v=B4r183JwhL0 NVIDIA has been releasing Vulkan Ray tracing enabled drivers since December.

NVIDIA has gotten much better over the last few years of disclosing vulnerabilities. Prior to that, I'm sure there were just as many issues, but they were probably just quietly fixed with new drivers. Link to NVIDIA's Security Bulletin Page: https://www.nvidia.com/en-us/security/ Link to the old archives: https://www.nvidia.com/en-us/security/archive/

If I had an ASRock x370 instead of an ASUS x370, I's go this route for sure. I have one of the bugged 1700s, that I never RMA'd. It works fine for most tasks, but it will crash with heavy compiling.

Even in a test/demo I do not like losing. ? I might have to actually OC the Radeon VII that I'm running for TC.

The board does have some limits on the number of addressable peripherals. On the previous BIOS I tested with all the PCIe slots populated and PCIE 2 split into two slots with a passive splitter. In order for all the devices to be recognized I had to disable onboard audio and the USB 3 header. I've ran into this issue on other platforms, and depending on what all you need there can be some work-arounds with some things disabled. Your testing seems pretty thorough, but I'll give it a go with an ASRock Ultra Quad M.2 card. Choice of colors is pretty standard for a server motherboard. I don't think I've ever seen an ASRock Rack board that is anything other than green with white, blue, or black RAM and PCIe slots.

The -2M should probably be given, but for now I'm prepared to let him be happy at the top of the leaderboard as a reward for getting things fired up. All jokes aside, I don't see any issues, and thanks for taking the time to help.

Flashed to a beta bios with additional PCIe bifurcation options. I've yet to test the additional options other than double checking the link speed. No odd splits like x4x4x8 or lower like I have on some intel platforms, but 4x or 8x NVMe for a smoking fast db server is now possible.

Other than the very obvious 1.2M mixup in FPN and TFA's point totals everything looks good to me.

Speaking of other accounts. Did anyone ever cash in their FLDC?

Currently the GPU portion of this is just 3D acceleration of OpenGl. Someone did file a request on the WSLg repo for OpenCl support. https://github.com/microsoft/wslg/issues/39 It was possible to do this before, but it was not a seamless experience like what Microsoft is doing here. It is still a bit hacky with what amounts to a nested VM inside of WSL accessed over RDP. Other than that, I'll just keep my thoughts to myself on the whole MicrosoftLinux narrative.