Contribute
Register

Asus X299 - Support

Status
Not open for further replies.
@CaseySJ

i feel this is the post of the last tentative as we don't really get any clue (I know that the MB is a tough one...) After discussing with my friend he is almost on the verge of buying a Z490 and an Intel i710700 to get the same config as mine, stable, thanks to your outstanding tutorial.
any comment from your side?
Thanks so much in advance ;-)

You're on a pretty old BIOS and used an older EFI as a starting point. I would try to get updated on both and check your BIOS settings and cooling. Did you follow @kgp's guide for recommended bios settings? You may have to tweak avx offsets and stuff to get stable.

Also full specs, BIOS settings, and copy of EFI help too.
 
Hey guys, so since I installed the NVME drives it’s been insanely fast, but I’m noticing random instability with my hack suddenly.

I had handbrake open and it wasn’t running anything and I was just doing a couple of things and the machine KP’d while not being pushed at all. The log showed it was handbrake that caused it. Later in the day, had the Speedtest app opened as I had just ran it due to some internet weirdness, was doing something in chrome for work and again, KP, that time is was Speedtest.

So far, I’ve disk checked all my volumes, reset the NVRAM (F11 at the clover boot screen) and it seems fine. But- I have a weird feeling it’s a heat issue with the NVME. My Vega 64 is installed in the slot basically above it. Is there any way I can move the video card to a different slot on my Asus WS X299/10G and it still work and get the full speed? I have everything running solid otherwise including TB3 with my Titan ridge.

I am probably going to pick up a pair of NVME heat sinks but I worry the clearance won’t be there because that Vega 64 is there and it’s a chonker lol- so even a low profile heat sink may not fit on that slot.
 
Last edited:
Hey guys, so since I installed the NVME drives it’s been insanely fast, but I’m noticing random instability with my hack suddenly.

I had handbrake open and it wasn’t running anything and I was just doing a couple of things and the machine KP’d while not being pushed at all. The log showed it was handbrake that caused it. Later in the day, had the Speedtest app opened as I had just ran it due to some internet weirdness, was doing something in chrome for work and again, KP, that time is was Speedtest.

So far, I’ve disk checked all my volumes, reset the NVRAM (F11 at the clover boot screen) and it seems fine. But- I have a weird feeling it’s a heat issue with the NVME. My Vega 64 is installed in the slot basically above it. Is there any way I can move the video card to a different slot on my Asus WS X299/10G and it still work and get the full speed? I have everything running solid otherwise including TB3 with my Titan ridge.

I am probably going to pick up a pair of NVME heat sinks but I worry the clearance won’t be there because that Vega 64 is there and it’s a chonker lol- so even a low profile heat sink may not fit on that slot.

Look up the Sage specs to see how slots are allocated but I believe x8 vs x16 is marginal anyway. If you haven’t used the U.2 slot you can also get a u.2 to m.2 2.5 adapter and it seems to work fine (little expensive tho)

https://www.amazon.com/gp/product/B0716K357W/?tag=tonymacx86com-20
https://www.amazon.com/gp/product/B01HR0PKJQ/?tag=tonymacx86com-20
 
Look up the Sage specs to see how slots are allocated but I believe x8 vs x16 is marginal anyway. If you haven’t used the U.2 slot you can also get a u.2 to m.2 2.5 adapter and it seems to work fine (little expensive tho)

https://www.amazon.com/gp/product/B0716K357W/?tag=tonymacx86com-20
https://www.amazon.com/gp/product/B01HR0PKJQ/?tag=tonymacx86com-20
So I do have some free 16x slots. Moving the card won’t affect anything in MacOS right? I mean it shouldn’t ... lol.
 
So I do have some free 16x slots. Moving the card won’t affect anything in MacOS right? I mean it shouldn’t ... lol.

Motherboards are limited by PCIe lanes.

Just because a motherboard like SAGE/10G has all x16 slots doesn't mean each one will run at that speed.

Taking a look at the manual, see below which slots are actually capable running at x16.

PS. No, moving around PCIe slots won't affect macOS at all. Just make sure you keep the Thunderbolt 3 PCIe card in Slot 2 as instructed in the manual, although some users have success in different slots, I would keep it there. It's probably related to PCIe routing through CPU or PCH that ASUS recommends. Also this motherboard has PLX switches for PCIe, so it will automatically switch to the correct speeds based on what PCIe card you use.

Screen Shot 2021-01-10 at 4.41.46 PM.png
 
Motherboards are limited by PCIe lanes.

Just because a motherboard like SAGE/10G has all x16 slots doesn't mean each one will run at that speed.

Taking a look at the manual, see below which slots are actually capable running at x16.

PS. No, moving around PCIe slots won't affect macOS at all. Just make sure you keep the Thunderbolt 3 PCIe card in Slot 2 as instructed in the manual, although some users have success in different slots, I would keep it there. It's probably related to PCIe routing through CPU or PCH that ASUS recommends. Also this motherboard has PLX switches for PCIe, so it will automatically switch to the correct speeds based on what PCIe card you use.

View attachment 504585
Perfect! Yeah I checked the manual too and realized that slot 3 was open when I posted the last message. Should be all good, I'll go all surgery in a few hours to try to fix this. @izo1 do you have an NVME heat sink that works and fits right on this board on your NVME drive(s)? I'm using both slots, so ideally if I order one, I want to put the heat sinks on both preferably.
 
Perfect! Yeah I checked the manual too and realized that slot 3 was open when I posted the last message. Should be all good, I'll go all surgery in a few hours to try to fix this. @izo1 do you have an NVME heat sink that works and fits right on this board on your NVME drive(s)? I'm using both slots, so ideally if I order one, I want to put the heat sinks on both preferably.

Just an FYI, asus recommends slot 1 then slot 5. It should be fine in slot 3 but it may change how the plx chip allocate the lanes so you may run into some issues. You may have to experiment shuffling your pcie cards around :lol:

For example, I had nothing in any of the slots except slot 6 and 7. The gpu worked fine in slot 7 but my wifi/bt card did not work in slot 6. It worked fine in slot 5.
 
Perfect! Yeah I checked the manual too and realized that slot 3 was open when I posted the last message. Should be all good, I'll go all surgery in a few hours to try to fix this. @izo1 do you have an NVME heat sink that works and fits right on this board on your NVME drive(s)? I'm using both slots, so ideally if I order one, I want to put the heat sinks on both preferably.

Not using any heatsinks and haven't for years. My system is fully watercooled, but I run my fans at very low speeds, so ambient temps do go high a bit than usual.

The 970 Pro NVME is in the slot that's laying flat on the motherboard and the 970 EVO Plus is in the the slot that's sticking out.

I haven't seen temps go berserk on either one, although the 970 Pro is a tad older and runs a bit hotter than the EVO.
 
Not using any heatsinks and haven't for years. My system is fully watercooled, but I run my fans at very low speeds, so ambient temps do go high a bit than usual.

The 970 Pro NVME is in the slot that's laying flat on the motherboard and the 970 EVO Plus is in the the slot that's sticking out.

I haven't seen temps go berserk on either one, although the 970 Pro is a tad older and runs a bit hotter than the EVO.
Yeah, I saw the one below the video card spike to around 86 C and want to mitigate any long-term issues. Typically it idles around 49 (while the other is around 44 in the vertical slot that sticks out) - so I'm sure it's just because it's trapped under that Vega 64 monstrosity I have. Once I push the drives, they get toasty.
 
Hey all, so the NVME's are working great. As per @izo1 & @djlild7hina I moved all the cards around and made room.

I sustain around 1.4 GB/s after the DRAM cache is full (so usually around 75 gigs into a large copy) which is perfectly normal. Before that, it copies between the 2 drives around 2.8 GB/s. Very happy!

SO everything is fine right? ..... well .........

My achilles-heel is OBS.

So I'm back to this. Right now, I've disabled GPU encoding entirely and it seems to have stopped it.... somewhat.

With it on, after it reboots randomly, I checked, no kernel panic log, so I ran:

Code:
log show --predicate 'eventMessage contains "Previous shutdown cause"' --last 24h

The code is spat back was: 5 which means: Shutdown was initiated normally, from the macOS Shut Down menu or other shutdown command such as /sbin/shutdown. This does not indicate an issue with your system.

And it's completely random when I'm running anything that uses my GPU heavily (I had Zoom running w/ OBS and it randomly reboots the machine).

So today, I tried to upscale a video using Topaz Video Enhance AI and the same thing happens. Unlike OBS, I can make it crash the exact same way doing this process. Even if I lower the GPU memory usage and bind the job strictly to the CPU.

Now, it's not heat, I run my daily workloads and it pins my 10980XE at 100% and runs all 18 cores for hours and hours and no crash. So cooling isn't my issue (my tJunction is set to 82 C in the BIOS anyway- and with logging I'm not approaching it when the machine crashes).

Is it something w/ my Vega 64 and the better question being: if I can't find any kernel panic logs, how can I determine what's causing this? It's the last piece of the puzzle to an otherwise perfect machine.
 
Status
Not open for further replies.
Back
Top