pastrychef's Asus ROG Strix Z370-G Gaming (WI-FI AC) build w/ i9-9900K + AMD 6600 XT

[pastrychef]

Looking at that case, I suspect that a negative pressure setup would work better for the open cooled GPU, particularly because if means reversing the airflow from back to front.

Without any case fans at all, the air from the GPU (and CPU) would rise to the top of the case (between the PSU and disks), and as it cooled, it would spill over the top and fall behind (under) the MB. You would end up with internal airflow rising in the big part of the case and falling in the small gap under the MB - especially if the GPU and CPU fans were stopped.

With your current positive pressure setup, the cool air from the front will be split. Most will be directed through the CPU cooler and exhausted through the back, but a significant amount will pick up the hot air from the GPU and blow it round the case, probably following the natural air flow path, so you end up heating the MB from below and then have hot air mixing with the air going through the CPU cooler - bad all round.

That natural motion will happen whenever a fan can't prevent it. In your case, most air from the GPU will rise in that area. The exhaust fan will only deal with the air from the CPU cooler and the front case fans, which means the GPU heated air will be forced back into the general airflow in a position where it will come through the CPU cooler. On the way, it will be heating the MB.

If you reverse the case fans, the CPU gets pure external air, so should have the best temps possible. The GPU heated are will be drawn to the front of the case, and even if it spills over the top, that air will also be drawn to the front before it can heat anything else.

With the closed blower cooler, you have no GPU heat remaining in the top of the case, so the air moving round the case helps cool the disks and MB instead of heating them, resulting in good cooling all round. Because the GPU air is being exhausted at the back, you want your current front to back airflow to prevent the GPU air being sucked back in again.

Hmm... Interesting... I didn't think that the positive pressure could be detrimental... I always thought that compressing more air in to the case would cool the air and help all components.

I can see how using the two front case fans to suck out the heat from the case might help. However, don't you think that the fans on the video card would still force the hot air generated from the video card downward towards the CPU cooler?

I'm trying to envision in my head how changing direction of the airflow and creating a negative pressure situation would work. I think for it to be effective, I'd have to reverse all the case fans and the two CPU fans. I'd also need to rethink dust management. Alternatively, I think I can do a quick, albeit less efficient, test simply by unplugging the two front fans.

Reversing all the fans will require a considerable amount of time due to how tightly everything is packed in right now. I'd also have to saw off a corner of the rear fan so that the Wi-Fi/Bluetooth card to fit. So, this test will probably have to wait until the next time I pull everything out to mess with my Z390 motherboard again.

 

[pastrychef]

What is the equivalent full size ATX Intel motherboard for this build?

Yes, as scottkendall stated, the Z370-E is probably closest. But I think all the Asus Z370 motherboards are extremely similar and would work almost equally as well.

 

[scottkendall]

I'm trying to envision in my head how changing direction of the airflow and creating a negative pressure situation would work. I think for it to be effective, I'd have to reverse all the case fans and the two CPU fans. I'd also need to rethink dust management. Alternatively, I think I can do a quick, albeit less efficient, test simply by unplugging the two front fans.

I get what he is saying to you basically right now the fans all pull the air threw the CPU cooler and out the back if you reverse the flow of air the fan in the front should pull the heat from the GPU out of the case before the CPU cooler picks up the heated air.

Edit: Though I get it that it is a crap ton of work...

 

[nib01]

What are the advantages and disadvantages between mATX and ATX mobo as far as gaming and editing PC.

Does it mean that using ATX over mATX it gives you better performance because of the size and components? Thanks!

 

[pastrychef]

What are the advantages and disadvantages between mATX and ATX mobo as far as gaming and editing PC.

Does it mean that using ATX over mATX it gives you better performance because of the size and components? Thanks!

Performance-wise I don't think there's any measurable difference between the different motherboard sizes, assuming all other things are equal.

I just like the smaller size of mATX. The only advantage that ATX offers is more PCI-e slots, but I don't have that many PCI-e cards to install. I just have a video card and a 10GBase-T card.

Also, Z370 really doesn't have enough lanes for all those PCI-e slots on ATX motherboards. That's why you see some motherboards do things like switch M.2 slots to 2 lanes when certain PCI-e slots are used or other such tricks. Other ATX motherboards use PCI-e multiplexers to add the extra PCI-e slots.

If there were truly that many lanes available, mATX motherboards will have three or four x16 slots, but we never see that. On this particular motherboard, there are only enough PCI-e lanes leftover to give us two x1 slots, not even an x4 slot.

You might wonder why not just go with an ITX motherboard then... Because, all the ITX motherboards I've ever seen only have two RAM slots.

 

[aussiepilot]

Hmm... Interesting... I didn't think that the positive pressure could be detrimental... I always thought that compressing more air in to the case would cool the air and help all components.

I can see how using the two front case fans to suck out the heat from the case might help. However, don't you think that the fans on the video card would still force the hot air generated from the video card downward towards the CPU cooler?

I'm trying to envision in my head how changing direction of the airflow and creating a negative pressure situation would work. I think for it to be effective, I'd have to reverse all the case fans and the two CPU fans. I'd also need to rethink dust management. Alternatively, I think I can do a quick, albeit less efficient, test simply by unplugging the two front fans.

Reversing all the fans will require a considerable amount of time due to how tightly everything is packed in right now. I'd also have to saw off a corner of the rear fan so that the Wi-Fi/Bluetooth card to fit. So, this test will probably have to wait until the next time I pull everything out to mess with my Z390 motherboard again.

Whilst the GPU fans blow the air downwards (in your case), the heat sink (base of the card), blocks the air from continuing down. It has to come out sideways and then move in the direction of lowest local pressure - which is likely to be back up again, partly due to the GPU fans and partly due to the relative air temperatures.

There are possibly other options that could help without revering the fans, but I suspect those would need you to cut/drill holes in the case. Probably not something to do as an experiment with no idea if it will actually help.

 

[pastrychef]

Whilst the GPU fans blow the air downwards (in your case), the heat sink (base of the card), blocks the air from continuing down. It has to come out sideways and then move in the direction of lowest local pressure - which is likely to be back up again, partly due to the GPU fans and partly due to the relative air temperatures.

There are possibly other options that could help without revering the fans, but I suspect those would need you to cut/drill holes in the case. Probably not something to do as an experiment with no idea if it will actually help.

As the weather gets warmer and I spend more time indoors, I'll probably give it a try soon. At the same time, I'll swap out the motherboard and test out my Z390 motherboard again since there have been some reports of NVRAM working on Z390. Maybe sometime next month...

 

[kentval]

It might be an NVRAM issue. Try replacing AptioMemoryFix-64.efi with OsxAptioFix2Drv.efi + EmuVariableUefi-64.efi in /EFI/CLOVER/drivers64UEFI/.

In short: how?
This (can't restart/shutdown) is my remaining problem as well.
I'm using Clover 4934 (apparently 4937 is available). Looking at the package installer in Pacifist, I see EmuVariableUefi-64.efi but the OsxAptioFix2Drv.efi is not present. There is a OsxAptioFix3Drv.efi in the package.
So:
1) is version 3 OK to use instead of version 2 of OsxAptioFix-Drv.efi?
2) how do I actually replace them? Is there an installer that will let me select and replace just these drivers, or do I need to do more. I didn't see anything in CloverConfigurator, and I'm not sure how to use the Clover package installer do do this surgery. Can I extract them with Pacifist, and then just drag-n-drop the drivers into the active EFI?

Does this also require replacing the NVRAM file?

Once again, my ignorance shows thru.

 

[pastrychef]

In short: how?
This (can't restart/shutdown) is my remaining problem as well.
I'm using Clover 4934 (apparently 4937 is available). Looking at the package installer in Pacifist, I see EmuVariableUefi-64.efi but the OsxAptioFix2Drv.efi is not present. There is a OsxAptioFix3Drv.efi in the package.
So:
1) is version 3 OK to use instead of version 2 of OsxAptioFix-Drv.efi?
2) how do I actually replace them? Is there an installer that will let me select and replace just these drivers, or do I need to do more. I didn't see anything in CloverConfigurator, and I'm not sure how to use the Clover package installer do do this surgery. Can I extract them with Pacifist, and then just drag-n-drop the drivers into the active EFI?

Does this also require replacing the NVRAM file?

Once again, my ignorance shows thru.

OsxAptioFix2Drv.efi was removed from the Clover installer a while back when AptopMemoryFix and OsxAptioFix3Drv were working well. This was before Z390 was released. Fast forward to present, many Z390 users have found OsxAptioFix2Drv to work best with Z390 but it hasn't been added back to the Clover installer package...

1. No, OsxAptioFix3Drv and OsxAptioFix2Drv are different. I suggest you try OsxAptioFix2Drv first. Just use Google to find OsxAptioFix2Drv. You can probably find it in one of the older Clover installers.
2. You can just go to /EFI/CLOVER/drivers64UEFI/ to add or delete whichever files you need.
3. Yes, extracting with Pacifist should be fine.

If you haven't been using EmuVariableUefi and the RC scripts, you should not have any NVRAM file. When NVRAM is working right, anything that needs to be written to NVRAM will be written to NVRAM. If you are using emulated NVRAM, things that should be written to NVRAM will be written to a file.

 

[kentval]

OsxAptioFix2Drv.efi was removed from the Clover installer a while back when AptopMemoryFix and OsxAptioFix3Drv were working well. This was before Z390 was released. Fast forward to present, many Z390 users have found OsxAptioFix2Drv to work best with Z390 but it hasn't been added back to the Clover installer package...

1. No, OsxAptioFix3Drv and OsxAptioFix2Drv are different. I suggest you try OsxAptioFix2Drv first. Just use Google to find OsxAptioFix2Drv. You can probably find it in one of the older Clover installers.
2. You can just go to /EFI/CLOVER/drivers64UEFI/ to add or delete whichever files you need.
3. Yes, extracting with Pacifist should be fine.

If you haven't been using EmuVariableUefi and the RC scripts, you should not have any NVRAM file. When NVRAM is working right, anything that needs to be written to NVRAM will be written to NVRAM. If you are using emulated NVRAM, things that should be written to NVRAM will be written to a file.

Bingo! Thanks once again.
(My mobo requires an emulated NVRAM...)