Please list your CPU cooler and PSU model. 193A means 193 Ampere, so, just as it says on the tin. The maximum you can set might be 255A, or maybe 4096A if you're lucky.
You have several potential stumbling blocks with your overclocking. First of all, you pair the highest-end CPU of that generation with the lowest-end Z390 board from MSI. This does not go well together at all. Please observe the following chart with a rating for the respective board's CPU VRM power delivery capabilities:
All MSI boards but the Z390 GODLIKE use cheap discrete MOSFETs for the CPU VRM, but the Z390-A PRO has especially cheap MOSFETs, not enough of them, and they are only partly heatsinked.
This wouldn't be a big issue with a lower-end CPU. But you are using the 9900K, the first CPU that Intel pushed to the limit (the 8700K before was much more tame, but by the time of the 9th gen, AMD really brought good competition). And now you want to overclock the 9900K. But the cheap Z390 board models are getting really uncomfortable when running the 9900K at full pelt even without any OC. The VRM will heat up significantly, because the MOSFETs lack in efficiency and there are not enough of them to spread the thermal load, which might lead to VRM throttling or even shutdowns if you attempt to OC significantly.
To reiterate, the MSI Z370-A PRO has a bottom of the barrel 4+2-phase VRM design with the cheapest MOSFETs they could get. This type of board was not designed with a 9900K in mind.
This can be seen in most reviews of cheaper Z390 boards. For example the
Z390 Tomahawk review:
When fully loaded under Prime95 small-FFTs, the Z390 Tomahawk spikes to around 241W and then drops to around 140W. It also drops to around 3.40 GHz. And that’s not realistic since the CPU is only throttling to around 4.5GHz in several of our benchmarks. And so, we disabled thermal throttling and saw system power draw jump up to 330W.
Unfortunately, we forgot that temperatures over 115 °C were out of range for our thermistor. And so, temperatures kept climbing until the board shut down and refused to boot back up, instead showing a CPU error. No shooting flames, smoke geysers or welding arcs as we’ve seen in previous boards, just a quiet power down without recovery, and a burned thermistor.
Testing with MSI’s monitoring software allowed us to find the sweet-spot, where the voltage regulator would hover around 100 °C at full load at the CPU’s rated 4.70 GHz when running Prime95 small-FFTs. But that’s when we found out that the voltage regulator heat sink temperature was around 40° lower than its MOSFET (transistor) temperature. Ouch.
Thus, we get efficiency scores that mean absolutely nothing. The board can’t power our CPU to full load without cranking down its core voltage, but doing so isn’t fair to any of the other products.
Having found the one voltage the lets us reach the standard 4.70 GHz Intel Turbo Boost frequency for eight loaded cores…without overheating the voltage regulator, there was no room left to push the CPU frequency upward. Our 1.20V setting is the be-all-end-all, you’ll get 4.70 GHz at the voltage regulator’s limit.
And so the low price that looks like a great deal in our price-to-performance table is a terrible deal for anyone who would like to use the Core i9-9900K at its full potential. We wish we were able to recommend such an inexpensive board to Core i5-9600K users, but we simply didn’t have the spare CPU to verify that the board is even capable of pushing that processor’s performance potential.
ASRock Z370 Extreme4,
https://www.tomshardware.com/review...ntel-9th-gen-core-atx-motherboard,5978-4.html
Moreover, the Z390 Extreme4 bombs the PCMark Creative Test, and the voltage regulator thermal throttling we noticed in our power test would be the most-likely cause.
How does a board go about losing our MS Word and Excel benchmarks? We find it a little hard to believe that these could be strenuous enough to cause voltage regulator thermal throttling, but the benchmarks indicate that some type of throttling has occurred.
Indeed, voltage regulator temperatures were terrible, and these are just the temps we measured before throttling kicked down the CPU frequency. Meanwhile, the directly-competing MSI MAG Z390 Tomahawk reached its thermal limit and throttled before we could even get an accurate measurement. It appears this ASRock board isn't alone in its issues with the i9-9900K.
Simply putting the Core i9-9900K’s stock 8-core-loaded Intel Turbo Boost ratio in the Overclocking chart would be too generous for the Z390 Extreme4, given that it couldn’t hold that frequency. The MAG Z390 Tomahawk could, but only after we undervolted our CPU.
To be fair, ASRock's mid-priced Z390 Xtreme4 isn't the only board we've tested that doesn't properly support the Core i9-9900K.
Gigabyte Z390 Gaming X, actually the best-performing "budget" Z390 board they tested,
https://www.tomshardware.com/reviews/gigabyte-z390-gaming-x-atx-motherboard,6250-4.html
The Z390 Gaming X’s highest thermal reading (111°C VRM temperature) came from its monitoring software, and we had to lower our ambient temperature to 16° just to make it run long enough to get its highest reading. We tried even lower temperatures for the MSI and ASRock boards, but still needed to add a fan.
The Z390 Gaming X ran hotter, so hot that we wouldn’t recommend anyone run Prime95 small FFTs on a bone-stock Core i9-9900K using nothing more than the fans of a front-mounted radiator to cool its voltage regulator. Not that an average user would, but testing in that manner proved that the Z390 Gaming X really needs a fan to be very close to its CPU socket just to keep the voltage regulator from throttling back the CPU under heavy loads.
Gigabyte’s Z390 Gaming SLI is the only sub-$160 motherboard we’ve tested that could support the full power load of our $520 Core i9-9900K. On the other hand, a representative from one of the competing board makers said that they didn’t think anyone buying a $500 CPU would use a low-priced Z390 board.
They don't design these cheap boards' VRM sections to cope with the 9900K, simply because they know how big of a mismatch this is. Someone building a PC on a budget will get a cheap board and cheap CPU. Someone who wants a lot of power and has a bigger budget will get a high-end board and high-end CPU. Only rarely do people mix a high-end board with a cheap CPU and vice versa, and then they can run into problems.
You see here, this is my old board, the Z390 ACE. It used 12x 4C029N + 12x 4C024N MOSFETs, so, significantly more parts as on the Z390-A PRO, better quality, and fully heatsinked.
Now observe the temperatures with 225W CPU load:
If you have seperate MOSFETs with high switching losses and pull 200+W through the CPU, you can get dangerously high temperatures around the socket. On the Z390-A PRO it's obviously worse.
This is why, for the Z490 boards, only the cheapest models kept the seperate MOSFET solution:
https://www.anandtech.com/show/15723/the-intel-z490-motherboard-overview/2
For the Z590 boards, there was not a single model (!) still using the MOSFET solution:
https://www.anandtech.com/show/16347/the-intel-z590-overview/2
They all went to Powerstages,
https://www.gigabyte.com/MicroSite/312/images/PowIRstage.html
This brings us again to the 9900K. Its high power consumption and the borderline temperature significantly limit the overclocking options. With an air cooler, the prospects are very small. Maybe with a top-end air cooler, you have a slight chance. Ideally, it has to be a very good all-in-one water cooler or a custom water cooler to OC without hitting temperature limits. At least 500 MHz OC would be necessary to achieve a two-digit performance increase - that is unrealistic, unless you have a golden sample. You can maybe push it to +400 MHz all-core for a single-digit performance increase, however, power consumption will shoot up by +50% or so. So you massively lose efficiency and this becomes hard to cool.
Lastly, about plugging in the PCIe 6-pin power on the bottom of the board, i explain here why it's not necessary:
https://forum-en.msi.com/index.php?...with-samsung-980-pro-ssd.370985/#post-2101599
And yes, it has nothing to do with the CPU VRM.
