PRO Z690-A DDR4/WIFI Owners Thread

[kogos121f0297]

...i had on drive causing problems today... When it was plugged in, mobo got stuck on white debugging led and wouldn't boot. I had to connect it on USB and clean it with diskpart... then it worked with no issues..

About XMP, have you tried to reset the bios? Maybe memory training is not working without bios reset!

 

[mitsosmu153e02d4]

...i had on drive causing problems today... When it was plugged in, mobo got stuck on white debugging led and wouldn't boot. I had to connect it on USB and clean it with diskpart... then it worked with no issues..

About XMP, have you tried to reset the bios? Maybe memory training is not working without bios reset!

the drive was busted.
XMP kept on using 1.25v on the memory sticks. Had to manually set it to 1.35 and worked as intended

 

[lasercodb150502d5]

Anyone try this? Or one similar?

 

[SmiimS]

I haven't yet, but the so-called "washer mod" helps decreasing temperatures somewhat in my case, with an i7-12700K and Noctua D15 cooler. It mostly depends on the CPU and how/if it deforms after securing the loading mechanism.

 

[lasercodb150502d5]

I haven't yet, but the so-called "washer mod" helps decreasing temperatures somewhat in my case, with an i7-12700K and Noctua D15 cooler. It mostly depends on the CPU and how/if it deforms after securing the loading mechanism.

I Interesting. thought from all the internet chatter that every cpu got warped by the lm? I haven't checked mine yet.

 

[SmiimS]

Most do, but not to the same extent, and in many cases there's no temperature gain to be obtained by either installing the washers (up to 1mm height) or these frames. It can be easily checked by observing the thermal paste footprint after removing the cooler. If (assuming it was previously homogeneously spread) residual paste appears to have a greater thickness on the center of the heat spreader, it means that the surface is bending and becoming concave.

 

[raimondkieftenbel157602de]

Anyone try this? Or one similar?
View attachment 159641

And I think that people are waiting for someone else too try it.
I am

 

[kogos121f0297]

I tried it but seen no difference on 12700k + Arctic liquid freezer II 280. Fan curves were the same for with and without this frame, and tried cinebench r23 and prime95. Had the same temps (+-1/2C at cases, which is in marginal error)... I left it on, but i'm gonna remove it in couple of days that i'll be opening my case again.

It's not even really something most people should bother with. I'd recommend, ignore it, unless you have issues and have no idea why...

There are 2 lga1700 socket manufacturers... this board has the good quality. I don't remember their names, but i read it on reddit sometime ago.

 

[raimondkieftenbel157602de]

I thought so,maybe yours was not that bad?

I have no issues with my d15,at least no overheating temps ect
But betseen core temps there is an diference, with 5 to 10 c.

Can I ask you how are the temps with the Artic freezer 280mm.
Is your cpu overclocked?

Thanxs

 

[SmiimS]

On my configuration using 1 mm washers with a Noctua NH-D15S cooler I see about 5 °C difference with prime95 SSE and CPU Package power around 200-210W at temperatures around 90°C. However there is still residual concavity, so thermal transfer might not be optimal yet.

See photos below, before and after installing the washers on my motherboard, decreasing pressure from the CPU's integrated heat spreader (IHS). It's clear how the thermal paste footprint improves and the IHS becomes flatter. I think the "bending corrector frame" could improve things further. I have the supposedly good loading mechanism from Lotes, but that does not matter as the problem has been observed with both manufacturers (the other one being Foxconn).

At the moment I'm mostly using the CPU with a close to standard power limit (125W), and temperatures aren't an issue in this use-case. When overclocking or using the processor without power limits however, 5-10 °C lower temperatures can help avoiding thermal throttling. Or perhaps it could also help taking advantage of Thermal Velocity Boost when used.

 

[kogos121f0297]

It's not overclocked most of the times. I got the fan curves set like this (see attached).
On office work load, it never goes higher than 35C. On Cinebench r23 it goes up to 75C max (actually, it immediately goes that high and stays there, no matter how long the test is gonna take). On prime95 unrealistic scenarios, it went to 79C max.

I tried overclocking with the CPU boost motherboard option and intel's extreme utility... didn't have the time to go into OC more, but will do the following weeks. However, i've seen a minor temp increase, like 5C max on idle, and 2-3C max on heavy load.

I don't use the "dot/rice" method to apply thermal paste. I spread it on the IHS evenly... i have no idea if that makes any difference, but thought of mentioning it.

I think the washer mod will help people that have a pure flat cold plate, which is not the case in most cases... But as i said, i never really pushed it to the limit to see any difference. When i dive into oveclocking, it might make a difference... can't really tell.

On the other hand, if it is a wide spread issue, how come motherboard manufacturers don't correct this on production level?

 

[SmiimS]

Some manufacturers probably started doing something already, for example by using little to no torque for the T20 Torx screws that secure the independent loading mechanism (ILM) to the motherboard, which was seemingly the case for my MSI PRO Z690-A. The more the screws are tightened, the worse the ILM issue becomes. However this is fundamentally a design problem from Intel that cannot be easily solved without a completely new system.

To some extent it might also be a CPU manufacturing issue, since it has been observed to vary using the same motherboard and ILM. A more rigid CPU package, or one that is initially convex (to only become flat or very slightly concave after installation) could possibly help.

See this data from Derbauer using different CPUs with the same motherboard and his own "contact frame":

 

[lasercodb150502d5]

...However there is still residual concavity, so thermal transfer might not be optimal yet.

...It's clear how the thermal paste footprint improves and the IHS becomes flatter.

Do you think the CPU IHS flattens out after the ILM pressure is released? Or is it "bent" for good?

 

[SmiimS]

Most of the deformation is dynamic, but there's also a static component to it, i.e. after a few hundred hours it might permanently get slightly concave:

Alder Lake’s cooling problem straightened out by 5 degrees! - Simple ILM-Mod for Intel’s LGA-1700 socket | Practice | igor´sLAB

Intel's Alder Lake CPUs run hot, very hot. A few weeks ago, Igor had already identified the bending of the CPU and motherboard due to the LGA1700 socket as a potential cause for this. Today I'

www.igorslab.de

 

[lasercodb150502d5]

Just put together a build MSI Edge with a lot less pressure on the lga 1700 loading mechanism than my December Pro A ddr4 wifi. Wonder if some suppliers are on the the warping and trying some mods. bios was mid December and it wasn't lotes. Maybe the hype will be gone with the wind.,

 

[SmiimS]

The ILM from Lotes on my MSI PRO Z690-A has a slight bend, perhaps that's also what causes excessive pressure on the center of the CPU.

 

[kogos121f0297]

I am soooo tempted to try that again on my z690 pro wifi + 12700k...
I already done the tests without the mod today. I also seen on youtube that the sticky washers that are included in arctic liquid freezer II can do the work... this guy also mentioned in the comments that he had best performance with these washers, and i got plenty of those around.
(

)

These are my temps now, running a 10mins cinebench r23, stock frequencies, only xmp on, with NO mod.

Now, all i need is to find sometime away from work to open up the case and test the mod... maybe during weekends

According to alred lake die topology for 12700k + 12900k + 12900ks, core 5 and core 7 are close to the middle of the die... and those are the hotter cores for me. However, core 4 and core 6 are also in the middle, but on the other side... and these are running cooler. So, i'm not sure if it's bended or its just silicon lottery on those 2 cores, or lottery on the side of the cpu. To be honest, most probably it's just normal, because if it was bended, i should be seeing high temps on all "middle" cores, including core 4 + core 6, and i hear that a temp difference of +-5C between cores is normal.

 

[lasercodb150502d5]

View attachment 159641
Changed out i5 12600k with ILM and put on 12700K w above frame. I5 didn't look bent. Temps slightly lower at idle with I7 and new frame. Even without the bending/heat issue, the frame feels easier and more manageable than the clamping ILM. Anyone try it yet?

 

[romanvilla159402e4]

Hi, new to pc building. I bought this motherboard and i7 12700k. ¿What would be the correct order of installation of bios, windows and drivers? ¿Is there a problem with internet and this board; if so how to solve it? ¿Is better to go to windows 11 or 10?
Thanks in advance

 

[lasercodb150502d5]

Hi, new to pc building. I bought this motherboard and i7 12700k. ¿What would be the correct order of installation of bios, windows and drivers? ¿Is there a problem with internet and this board; if so how to solve it? ¿Is better to go to windows 11 or 10?
Thanks in advance

Windows 11, drivers (come with Windows), Bios (update if necessary). Windows 11 is optimized for Alder Lake (so is 10 but consensus is 11 interfaces better). Then it's recommended to update drivers through MSI and Intel, graphics card from company site (you didn't specify) and monitor driver if any. I have gig download from my internet provider, and the built-in Lan goes to 1400 mbps, so I don't find any problem with internet. I'm new to pc building as well, so this should be up-to-date instructions because it's what I just used in Dec 2021. Good luck. (What GPU & cooler are you using?)