2018 Hardware Thread

OldMike65 said:

Yes and I posted those results, fact still remains that the ONLY 370 ASUS board that Did NOT need the voltage set higher was the Apex X .....That is why all the other Asus boards did not make their QVL list.
Here is the question from a user, and the Answer from SL


Question that was asked:
How do the 370 Maximus boards hold up to 9700K/9900K? I notice the APEX is there, but the rest are omitted, which is curious to me seeing that, as far as I can tell, the power delivery capabilities are roughly the same, just that 390 has more efficient VRM.

Answer from Silicon Lottery:
The way they measure core voltage under load is too far off other motherboards, so they didn't make the QVL this time. You would need to set a higher voltage than we tested with to match our results. Apex isn't affected as much. Elmor made a comment about this here: Overclock.net - An Overclocking Community - View Single Post - Coffee Lake Refresh Binning

Well, you posted an attachment that some people might not read and might only take away the impression that the X boards aren't good enough (I have seen this in other forums), and are thinking they are getting great core volts based on their experience with X, so I thought it important to elucidate.

OldMike65 said:

Still it does look like to run the 9900K with the better VRM's, for the Best cooling, we would need a Asus 390 board.

For sure. And you can trust their core voltage more.

Asus rog user's have complained that when they use high LLC, the board would deliver much more voltage under load than they had set in BIOS. Well the board wasn't actually delivering more, only incorrectly reporting more, so they have finally fixed that (they never responded to user complaints with the facts though).

essenbe said:

@Geneo, how much do you think that has to do with SVID? I know everyone doesn't have this problem, but this board does. If I enable SVID, my voltage goes extremely high and it doesn't matter if it is manual or offset. Based on my fan speed, I think it is actually giving it that nuch voltage. I've never left it there long enough to check to be sure.

If I run 4.8 with SVID disabled here is my CPUz



If I boot into BIOS and do nothing but enable SVID, save and exit, here is my CPUz.

Hi Steve,

It should not have anything to do with SVID, but I think I know why that happens.

First, this measurement of Vcore is not accurate on the X boards (excepting the Apex accoring to SL) because there is a resistance the measurement of vcore isn't taken into account that results in a positive vcore offset when current flows through it, the more current the larger the offset, so this became more apparent with the 9xxx processors which draw 50% more current than the 8xxx.

I believe the SVID issue is due to the way the BIOS sets the LLC. if you turn on SVID, it applies Intel's load-line, which is conservative. I had this happen when I first booted up my Code X and saw I was unexpectedly at 1.5v. If you set the IA DC and AC load- line slope values to .001 instead of auto in the BIOS, this should not happen.