Z790 Godlike Bios advice

sean.w.thompson

New member
PRIVATE E-2
Joined
Feb 18, 2019
Messages
11
Hi. Sorry if this is posted in the wrong area. I need some straight simple advice. I have just put together a pc using the above mobo with a i9 Raptor 14th gen refresh. The pc boots but the bios is ancient. I have reviewed the forums and have a clear understanding of the bios update process but I am confused with the intel me update.

Do I install the me update identified in the MSI website below the bios, and do I do this before or after the bios update.

Also, MSI centre is stating that I dont have intel me installed but I can find the driver in Device Manager? I have looked on the web for the answer which has just made me more confused. I need to hear from the experts...

Thanks in advance for any help.

Sean
 
I recommend updating to the second newest BIOS, for your board that's the v17 from https://www.msi.com/Motherboard/MEG-Z790-GODLIKE/support

Reason being, there are some reports about bugs in the latest version, so the v17 is the "known good" one for now. If there is a v19 in the future, you can update to that then.
The BIOS update itself is done via M-Flash in the BIOS, it's rather straightforward and should be done before Windows is installed, see here.

The ME firmware update is completely independent from the BIOS Update. You can update to BIOS v17, and then just run the ME FW update in Windows, it will bring the ME FW to the newest version. Takes 2-3 minutes and a reboot.

The ME FW is only loosely connected to the BIOS version. Most BIOS Updates will bring a newer ME FW with them (one-step update), i don't know why they split it into two steps this time, but it doesn't matter much, nor is it critically important to update the ME FW. The BIOS is more important to update.

MSI Center i usually stay away from, see here. You can install the relevant drivers yourself. If you want to find the ME in the device manager, it is called "Intel (R) Management Engine Interface" (under System Devices).
 
I recommend updating to the second newest BIOS, for your board that's the v17 from https://www.msi.com/Motherboard/MEG-Z790-GODLIKE/support

Reason being, there are some reports about bugs in the latest version, so the v17 is the "known good" one for now. If there is a v19 in the future, you can update to that then.
The BIOS update itself is done via M-Flash in the BIOS, it's rather straightforward and should be done before Windows is installed, see here.

The ME firmware update is completely independent from the BIOS Update. You can update to BIOS v17, and then just run the ME FW update in Windows, it will bring the ME FW to the newest version. Takes 2-3 minutes and a reboot.

The ME FW is only loosely connected to the BIOS version. Most BIOS Updates will bring a newer ME FW with them (one-step update), i don't know why they split it into two steps this time, but it doesn't matter much, nor is it critically important to update the ME FW. The BIOS is more important to update.

MSI Center i usually stay away from, see here. You can install the relevant drivers yourself. If you want to find the ME in the device manager, it is called "Intel (R) Management Engine Interface" (under System Devices).

Thanks. That was very easy to follow. I will update to the second newest bios and then install intel me. Sound advice. I already have windows installed. If I install the bios first, will Windows load up ok to enable me to update me? I read somewhere that my m.2 might not be recognised if me is not updated first?
 
You should normally always update the BIOS first before installing Windows, because otherwise the Windows activation can be lost sometimes. Windows can then act as if you changed the hardware and can ask to be activated again. Maybe try the "fix" as mentioned here, which is more of a workaround. I did it that way but it might have been coincidence that my Win11 was still activated after the BIOS update.

The ME FW has no bearing on your M.2 SSD being recognized or not. I did the ME FW weeks after i updated to my board's equivalent of the v17 BIOS, and i had no problems whatsoever from it.
 
You should normally always update the BIOS first before installing Windows, because otherwise the Windows activation can be lost sometimes. Windows can then act as if you changed the hardware and can ask to be activated again. Maybe try the "fix" as mentioned here, which is more of a workaround. I did it that way but it might have been coincidence that my Win11 was still activated after the BIOS update.

The ME FW has no bearing on your M.2 SSD being recognized or not. I did the ME FW weeks after i updated to my board's equivalent of the v17 BIOS, and i had no problems whatsoever from it.
That's great. Thanks for all your help. I will update tomorrow and, fingers crossed, my windows activation will be good. I am surprised that the 14th gen is running so smoothly. My ram is good at 6400 too. Only problem seem to be my chip only boosting to i13900 spec. The bios I am running now dates back to.... I think it was June.

Again, cheers.

Sean
 
With the GODLIKE as your board, i assume you got the 14900K. If left to its own devices (meaning, running without any power limits), it can cause a headache even for high-end AIO watercoolers, because the power draw is simply off the scale under full load. Hence, it will experience thermal throttling (downclocking to prevent damage from too high temperature). There are some countermeasures against all that. If you want, we can take a look at that later.
 
T
With the GODLIKE as your board, i assume you got the 14900K. If left to its own devices (meaning, running without any power limits), it can cause a headache even for high-end AIO watercoolers, because the power draw is simply off the scale under full load. Hence, it will experience thermal throttling (downclocking to prevent damage from too high temperature). There are some countermeasures against all that. If you want, we can take a look at that later.
That would be great. I have slightly overclocked my last cpu but have not attempted undervolting. I will post my bios update results tomorrow. At present, Dead Space on 4k with everything on full is giving me max temps of 75. That could be due to the fact my bios is not set for the 14900k. I have a Kraken z73 aio with Corsair in a push / pull config on the rad. 1 rear exhaust. 3 bottom intakes. 3 side intakes and 3 front intakes. My GPU is a MSI 4090 Suprim X on air. Again, thank you for your time and assistance. You are a credit to this forum. Maybe, in time, I can offer others the same assistance you have given to me.
 
Games will not nearly tax this CPU to the fullest, because most games won't use more than six cores (some will use eight, but that's already rare). And those cores won't even be fully loaded, because the GPU is doing the brunt of the work, the CPU is just doing legwork for the GPU, so to speak. In most games, you won't see more than 150W CPU power draw, that can easily cooled by your Kraken (meanwhile, the GPU draws 450W).

It's only when you create fully multithreaded CPU load (full load on all cores) that the CPU power draw shoots into the stratosphere. And for those occasions, most people will need power limits, because this CPU is extremely difficult to keep cool with that kind of power draw and heat. So how you want to set it is so that your cooling is always on top of the situation, the CPU temps are meant to stay below 90°C in every scenario. Mid-80s is the best, because then you still have a good 10° of headroom for higher ambient temperatures, so you can always avoid thermal throttling.

We can see about all that tomorrow or whenever. Thanks also for your kind words, always glad to help.
 
Just to let you know, I followed your advice and all went well. Windows remained activated, too. Been looking at load line undervolting.... Any experience with this? I had a quick look into the bios and couldn't find a reference.
 
Good. Now, there's two things i would do.

1) Test which power limits you need to set for your cooling. As i said, because of the 14900K's outrageously high power draw when running without power limits in place, most people (even with high-end coolers) will have to - and should - set power limits for it, in order to prevent so-called thermal throttling. Thermal throttling is a mechanism that prevents the CPU from dying of overtemperature, it acts when approaching 100°C CPU temperature and tries save itself from overheating. But it's not good to rely on that, because if it comes to that and your cooling is already maxed out (fans at full speed, good airflow through the system), it means the CPU generates more heat than the cooler can get rid of, so thermal throttling has to step in to prevent the worst. In essence, it's an emergency mechanism, not something that should be able to happen for daily use. You want power limit throttling to step in before thermal throttling ever has to.

What you'd do is, you eyeball a number that your cooler might be able to handle comfortably (for your cooler you could set 250W for a test), you set those power limits in the BIOS, then you check how high the temperatures get under fully multithreaded load. If they are now in the mid-80°C range, that's good, you found your cooler's potential and you have some headroom left for higher ambient temperatures in the summer. If you still run into thermal throttling, you set the limits a bit lower.

To test the limits, you'd check your sensors with HWinfo64 during a Cinebench run. Run it and open "Sensors", then expand all sensors by clicking on the little <--> arrows on the bottom, also expand the columns of the sensors a bit so everything can be read. Make it three big columns of sensors (or four, if the screen resolution is high enough). In the end, it should be a screenshot with all the sensors visible at once, like this:

yes.png


Make sure your power plan in Windows is on "Balanced". Do nothing on the PC for a while (couple minutes), so the "minimum" baselines for the values are established. After that time in idle, then produce full CPU load with Cinebench (either R23 or R24), and after completing a 10 minute run, when the CPU temperatures have stabilized at the highest level, take a screenshot of the sensor window and tell me the Cinebench score. We shall see if the power limits are working well for your cooling or if they need adjusting.

Even with 250W or slightly lower limits and at safe temperatures, you shouldn't lose much performance compared to unlimited. That is because above 250W, the performance left to gain will be in the low- to mid-single-digits, but at much higher power draw. So these last performance gains are to be dismissed anyway, they are highly ineffective. So you are improving the calculation efficiency by limiting the power draw a bit, because there will be less energy spent for the job to finish.

How to set power limits? You go here in the BIOS and enter 250 for Long and Short duration power limit:

MSI_SnapShot_14.png


You can also set the other stuff like Intel C-State, C1E Support, Intel Speed Shift all to Enabled, and set Package C State Limit to C10. All that lowers power consumption in idle.

2) You can see i modified "CPU Lite Load" there too. That's a setting that has to be found out for each specific CPU by doing stability testing. By lowering it, you'd essentially shave off the generic headroom that MSI like to add on VCore (CPU core voltage) and adapt it to your specific CPU sample. This is a good way to lower power consumption in all load states. Check this post and the one i linked there under the "CPU Lite Load" link. Furthermore, check this post for some additional settings to lower the power consumption without any performance loss.

In my opinion, "CPU Lite Load" is one of the best undervolting methods on Intel MSI boards. That is basically the load line undervolting you're looking for. One thing about it though: The higher the CPU model, the less you can usually lower CPU Lite Load. Because the simple fact is, for the high frequencies of an i7/i9, you need a pretty high VCore (CPU core voltage) to be able to push it that high. So the safety margin that you can shave off using CPU Lite Load is smaller than on a mid-range CPU, which runs at turbo frequencies that are much more in the silicon's original comfort zone. If you can lower the CPU Lite Load on a i7/i9 by a couple modes, and it stays perfectly stable in stress tests, that would already be a decent improvement. On my i5-13500 meanwhile, i could go all the way down to Mode 3. But that's only because they are not pushing that model to the limit from factory, like they do on the i7/i9.

You need to verify the stability at each step (using Prime95 Small FFTs or OCCT for example). Each step brings down VCore for all load states a bit, so then the CPU can clock higher at the same power draw, making the performance with full load at the 250W power limit improve a little, and of course it will also save power in all load states below the power limit.
 
Last edited:
Good. Now, there's two things i would do.

1) Test which power limits you need to set for your cooling. As i said, because of the 14900K's outrageously high power draw when running without power limits in place, most people (even with high-end coolers) will have to - and should - set power limits for it, in order to prevent so-called thermal throttling. Thermal throttling is a mechanism that prevents the CPU from dying of overtemperature, it acts when approaching 100°C CPU temperature and tries to limit it to that to prevent damage. But it's not good to rely on that, because if it comes to that and your cooling is already maxed out (fans at full speed, good airflow through the system), it means the CPU generates more heat than the cooler can get rid of, so thermal throttling has to step in to prevent the worst. In essence, it's an emergency mechanism, not something that should be able to happen for daily use. You want power limit throttling to step in before thermal throttling ever has to.

What you'd do is, you eyeball a number that your cooler might be able to handle comfortably (for your cooler you could set 250W for a test), you set those power limits in the BIOS, then you check how high the temperatures get under fully multithreaded load. If they now they in the 80°C range, that's good, you found your cooler's potential and you have some headroom left for higher ambient temperatures in the summer. If you still run into thermal throttling, you set the limits a bit lower.

To test the limits, you'd check your sensors with HWinfo64 during a Cinebench run. Run it and open "Sensors", then expand all sensors by clicking on the little <--> arrows on the bottom, also expand the columns of the sensors a bit so everything can be read. Make it three big columns of sensors (or four, if the screen resolution is high enough). In the end, it should be a screenshot with all the sensors visible at once, like this:

yes.png


Make sure your power plan in Windows is on "Balanced". Do nothing on the PC for a while (couple minutes), so the "minimum" baselines for the values are established. After that time in idle, then produce full CPU load with Cinebench (either R23 or R24), and after completing a 10 minute run, when the CPU temperatures have stabilized at the highest level, take a screenshot of the sensor window and tell me the Cinebench score. We shall see if the power limits are working well for your cooling or if they need adjusting.

Even with 250W or slightly lower limits and at safe temperatures, you shouldn't lose much performance compared to unlimited. That is because above 250W, the performance left to gain will be in the low- to mid-single-digits, but at much higher power draw. So these last performance gains are to be dismissed anyway, they are highly ineffective. So you are improving the calculation efficiency by limiting the power draw a bit, because there will be less energy spent for the job to finish.

How to set power limits? You go here in the BIOS and enter 250 for Long and Short duration power limit:

MSI_SnapShot_14.png


You can also set the other stuff like Intel C-State, C1E Support, Intel Speed Shift all to Enabled, and set Package C State Limit to C10. All that lowers power consumption in idle.

2) You can see i modified "CPU Lite Load" there too. That's a setting that has to be found out for each specific CPU by doing stability testing. By lowering it, you'd essentially shave off the generic headroom that MSI like to add on VCore (CPU core voltage) and adapt it to your specific CPU sample. This is a good way to lower power consumption in all load states. Check this post and the one i linked there under the "CPU Lite Load" link. Furthermore, check this post for some additional settings to lower the power consumption without any performance loss.

In my opinion, "CPU Lite Load" is one of the best undervolting methods on Intel MSI boards. That is basically the load line undervolting you're looking for. One thing about it though: The higher the CPU model, the less you can usually lower CPU Lite Load. Because the simple fact is, for the high frequencies of an i7/i9, you need a pretty high VCore (CPU core voltage) to be able to push it that high. So the safety margin that you can shave off using CPU Lite Load is smaller than on a mid-range CPU, which runs at turbo frequencies that are much more in the silicon's original comfort zone. If you can lower the CPU Lite Load on a i7/i9 by a couple modes, and it stays perfectly stable in stress tests, that would already be a decent improvement. On my i5-13500 meanwhile, i could go all the way down to Mode 3. But that's only because they are not pushing that model to the limit from factory, like they do on the i7/i9.

You need to verify the stability at each step (using Prime95 Small FFTs or OCCT for example). Each step brings down VCore for all load states a bit, so then the CPU can clock higher at the same power draw, making the performance with full load at the 250W power limit improve a little, and of course it will also save power in all load states below the power limit.
That's great. Will try it out tomorrow and see what I find.
 
Sorry for the delay in getting back--been away for a week visiting family. I have followed your clear instructions and I will run a bench test today and report back with the screen shots. Do you think it would be worth while doing a slight offset on the vcore also? And do you have any advice on the best way to apply it... As a complete newbie I am presuming I would find the offset vcore in the bios and input something around -0.100 to go with my 250w limit to my power draw?
 
Well, all was good. My first 10 minute run (balanced power plan) hit 90+ after stabilising out but this was due to my fan curve being conservative (I have a thing about noise pollution) I changed my fan curve so that they would reach full speed when the CPU hit 85c. They are Corsair SP fans and have good air pressure. I also set my water pump to full.

On the second and third runs, this resulted in a much better prolonged temperature of 84c (give or take a few degrees either way). Not too bad as it is literally freezing here in the northeast of England and the central heating has been constantly running.

The only downside from all this was the noise from the cooling; but, in real world scenario, when I use the pc for 4k gaming at 120fps and light video editing, it is unlikely that the CPU will reach this level of activity and require the fans to operate at such speeds.

Thanks to your help, I can rest easy in the knowledge that my CPU should have sufficient cooling (and an adequate safety margin) in the worst case scenarios. And due to the power limits now in place, I will be saving electricity costs in the long run.

My main concerns were minimising heat and reducing energy waste/costs whilst maintaining running efficiency. The board and CPU were presents from the son. Excellent gifts, but they may not have been my personal preferred choice. Anyway, being stuck on the z390 platform for years, I was well pleased :) and the board, like my first, was MSI.

I can confirm after a heavy afternoon of gaming, all settings on max, that I had zero loss of efficiency with the reduction in wattage. In fact, monitoring the CPU Core Temp programme, I was lucky to see my pc spiking above 55w. Gaming temperatures never exceeded 64c

On a final note, I have yet to spot my cpu boosting above 5.6 (I have the frequency set up to display on my taskbar) I am led to believe that i14900ks should boost up to 6.0 under low, core count tasks such as gaming? I dont have the knowledge to know if this is due to the power limit or not and what knock-on effect the limit may have on other bios settings.

Have a good weekend and again, thank you for your time and help.

Sean
 
Those are impressively low temperatures (if we can call it low) for that kind of power draw, your cooling packs a punch. Yeah, this is only to test the cooling capabilities, such kind of full load is rare in daily use. But when push comes to shove, you want your cooling to be up to the task even with the highest load.

On a final note, I have yet to spot my cpu boosting above 5.6 (I have the frequency set up to display on my taskbar) I am led to believe that i14900ks should boost up to 6.0 under low, core count tasks such as gaming?

Only up to four-core load. A lot of modern games use six cores, some eight. Or there might be half load on a bunch of cores. Also, whenever you have a frequency displayed for the entire CPU, you don't know if it takes the E-cores into account, which would drop the average. So it's not so clear-cut. You could test with Cinebench Single for example, and then check the individual core clocks in HWinfo to see if any P-cores hit 6 GHz. It certainly won't have to do with any power limits if they don't, because with up to four cores fully loaded you cannot reach anywhere close to 250W.
 
Back
Top