Certain games crashing.

[Djw_]

I recently built a new computer and my computer automatically set my CPU clock speed to 5700MHZ. I've tried some different settings in BIOS but I don't really know what I doing, and the settings have cause the PC to fail to boot or have BIOS freeze. When trying to run certain games they crash on start or in game, with a couple stating video memory errors. The games include Fortnite, The Finals, and Apex Legends, and at one point COD but I reset the computer and COD worked. I've ruled out the GPU as I've tried a different GPU in the computer and same issue, it's also not integrated graphics as I turned them off, reapplied thermal paste as the temp got pretty high when under load, ran windows mem diagnostic, ran a couple of different benchmarks which have either crashed, or state that the CPU use is 0 or no there is no CPU to test, taken out RAM one at a time to test each stick, and completely reset computer none of which has worked. I'm almost positive the issue is with the CPU being set to to too high of clock speed, but I don't know which settings to adjust in BIOS.

Specs:

Motherboard: Gigabyte z790 Aorus Elite AX

CPU: Intel Core i9-14900k

GPU: Geforce RTX PNY OC 4090

RAM: Corsair Vengance 2x16 gb

PSU: Super flower 1600 W Titanium

CPU cooler: Liquid Freezer II 360

SSD: Samsung 970 EVO Plus 1TB, Samsung 990 PRO 1TB

[Djw_]

16 minutes ago, Djw_ said:

I recently built a new computer and my computer automatically set my CPU clock speed to 5700MHZ. I've tried some different settings in BIOS but I don't really know what I doing, and the settings have cause the PC to fail to boot or have BIOS freeze. When trying to run certain games they crash on start or in game, with a couple stating video memory errors. The games include Fortnite, The Finals, and Apex Legends, and at one point COD but I reset the computer and COD worked. I've ruled out the GPU as I've tried a different GPU in the computer and same issue, it's also not integrated graphics as I turned them off, reapplied thermal paste as the temp got pretty high when under load, ran windows mem diagnostic, ran a couple of different benchmarks which have either crashed, or state that the CPU use is 0 or no there is no CPU to test, taken out RAM one at a time to test each stick, and completely reset computer none of which has worked. I'm almost positive the issue is with the CPU being set to to too high of clock speed, but I don't know which settings to adjust in BIOS.

Specs:

Motherboard: Gigabyte z790 Aorus Elite AX

CPU: Intel Core i9-14900k

GPU: Geforce RTX PNY OC 4090

RAM: Corsair Vengance 2x16 gb

PSU: Super flower 1600 W Titanium

CPU cooler: Liquid Freezer II 360

SSD: Samsung 970 EVO Plus 1TB, Samsung 990 PRO 1TB

CPU was thermal throttling, I set the max W for CPU in BIOS to match the W for the cooler on the website and the game started up just fine.

[MrSimplicity]

To be honest I don't know what you mean by W, watts? In which case I wouldn't touch those settings. I would just tone down the overclock to between 5.0GHz and 5.4Ghz. I would also recommend resetting everything to default. If you don't know what you're doing the only thing that should be messed with in the BIOS is setting the proper RAM speed.

 

Also I know the 14900K is a hot chip but it shouldn't be thermal throttling. It'll hit its thermal limit but then it should down clock the CPU to stay within its parameters. So either the cooler isn't making good contact or something in your BIOS is setting a hard limit preventing the CPU from ramping down. My best guess is the way the overclock is set that it's trying to keep the CPU at 5.7GHz which is causing the instabilities under load.

 

So either reset everything to default which is highly recommended. Or set the core clock multiplier to 50 and call it a day.

[Djw_]

15 hours ago, MrSimplicity said:

To be honest I don't know what you mean by W, watts? In which case I wouldn't touch those settings. I would just tone down the overclock to between 5.0GHz and 5.4Ghz. I would also recommend resetting everything to default. If you don't know what you're doing the only thing that should be messed with in the BIOS is setting the proper RAM speed.

 

Also I know the 14900K is a hot chip but it shouldn't be thermal throttling. It'll hit its thermal limit but then it should down clock the CPU to stay within its parameters. So either the cooler isn't making good contact or something in your BIOS is setting a hard limit preventing the CPU from ramping down. My best guess is the way the overclock is set that it's trying to keep the CPU at 5.7GHz which is causing the instabilities under load.

 

So either reset everything to default which is highly recommended. Or set the core clock multiplier to 50 and call it a day.

Didn't overclock anything the motherboard automatically set it to 5.7 GHZ. Resetting to default sets clock speed to 5.7 GHZ

[Djw_]

Just now, Djw_ said:

Didn't overclock anything the motherboard automatically set it to 5.7 GHZ. Resetting to default sets clock speed to 5.7 GHZ

I don’t know what to change in BIOS to change overclock settings and I found a video about “undervolting” the CPU and just did what he said.

[Djw_]

15 hours ago, MrSimplicity said:

To be honest I don't know what you mean by W, watts? In which case I wouldn't touch those settings. I would just tone down the overclock to between 5.0GHz and 5.4Ghz. I would also recommend resetting everything to default. If you don't know what you're doing the only thing that should be messed with in the BIOS is setting the proper RAM speed.

 

Also I know the 14900K is a hot chip but it shouldn't be thermal throttling. It'll hit its thermal limit but then it should down clock the CPU to stay within its parameters. So either the cooler isn't making good contact or something in your BIOS is setting a hard limit preventing the CPU from ramping down. My best guess is the way the overclock is set that it's trying to keep the CPU at 5.7GHz which is causing the instabilities under load.

 

So either reset everything to default which is highly recommended. Or set the core clock multiplier to 50 and call it a day.

I’ll try to set the core clock multiplier to 50 so see if that fixes the problem as well.

[Djw_]

My computer BSOD when set to 50

[MrSimplicity]

9 hours ago, Djw_ said:

My computer BSOD when set to 50

Honestly, The only reason a CPU would cause a BSOD is if it thermal grenading itself (too hot and can't cool down) or if something is just inherently wrong with it which would normally cause a no-post issue. Also just actually realized you went balls to the wall with your build.

 

I'm more curious now why you figured you needed a 1600 watt power supply. Along with it being a brand I've never heard of, I'm hoping you're on a 240v grid cuz that could be your issue also. 120v is very finnicky, if the circuit you're plugged into is only a 10 or 15 amp fuse then the PSU is probably trying to draw more power than the circuit can give it, more surprising that you haven't blown the fuse. Especially when opening games, everything will ramp up for the initial loading, which can cause the crashing hence why undervolting worked. Two things I would suggest if you think/know your cooler is properly seated. 1. get a UPS (uninterrupted power supply) you'll need about a 2000va or 2400va or 2. downgrade your PSU to 1000-1200w. I highly encourage you to stick with a known reliable brand EVGA, ASUS, SeaSonic etc.

 

That or start contacting the retailer you bought everything from. Your Mobo might even be the culprit and just wants to throw everything at your hardware. Gigabyte has gone down hill as far as QC goes. I'm an Asus kind of guy myself but MSI and even AsRock are better options. You can do all that or just go back to undervolting. I would also try to walk you through how to properly setup a stable OC that wouldn't crash, probably even at 5.7, but I know absolutely nothing about Gigabyte's BIOS to even begin trying.

 

Also random question. When you open up task manager, is your CPU just sitting pegged at 5.7? or is it idling normally at 3-4GHz?

[Bjoolz]

5 hours ago, MrSimplicity said:

I'm more curious now why you figured you needed a 1600 watt power supply. Along with it being a brand I've never heard of, I'm hoping you're on a 240v grid cuz that could be your issue also.

Super Flower is very well known as a manufacturer, but not so much for their own brand. They still make EVGA's "higher quality" PSUs I think. 

[MrSimplicity]

10 hours ago, Bjoolz said:

Super Flower is very well known as a manufacturer, but not so much for their own brand. They still make EVGA's "higher quality" PSUs I think. 

good to know. Well still, if he's on a 120v grid a 1600w PSU is definitely gonna give him issues depending on what size fuse is on the circuit he's plugged into. He could definitely pull that if it's a 20amp circuit and it's the only thing on it. 

[Djw_]

15 hours ago, MrSimplicity said:

Honestly, The only reason a CPU would cause a BSOD is if it thermal grenading itself (too hot and can't cool down) or if something is just inherently wrong with it which would normally cause a no-post issue. Also just actually realized you went balls to the wall with your build.

 

I'm more curious now why you figured you needed a 1600 watt power supply. Along with it being a brand I've never heard of, I'm hoping you're on a 240v grid cuz that could be your issue also. 120v is very finnicky, if the circuit you're plugged into is only a 10 or 15 amp fuse then the PSU is probably trying to draw more power than the circuit can give it, more surprising that you haven't blown the fuse. Especially when opening games, everything will ramp up for the initial loading, which can cause the crashing hence why undervolting worked. Two things I would suggest if you think/know your cooler is properly seated. 1. get a UPS (uninterrupted power supply) you'll need about a 2000va or 2400va or 2. downgrade your PSU to 1000-1200w. I highly encourage you to stick with a known reliable brand EVGA, ASUS, SeaSonic etc.

 

That or start contacting the retailer you bought everything from. Your Mobo might even be the culprit and just wants to throw everything at your hardware. Gigabyte has gone down hill as far as QC goes. I'm an Asus kind of guy myself but MSI and even AsRock are better options. You can do all that or just go back to undervolting. I would also try to walk you through how to properly setup a stable OC that wouldn't crash, probably even at 5.7, but I know absolutely nothing about Gigabyte's BIOS to even begin trying.

 

Also random question. When you open up task manager, is your CPU just sitting pegged at 5.7? or is it idling normally at 3-4GHz?

Task manager says base speed is 3.2 GHz and idles around 1-3 GHz but in bios and some other programs they say 5.7 GHz and will fluctuate from 1 GHz up to 6 GHz. When Undervolting it seems the clock speed stays between 1.1-4.4 and occasionally jumps to 6 GHz. The reason I got the 1600 W was because it was $300 compared to some other PSUs that were 1200 W and almost twice the cost. I'll look into getting a UPS, and contact Gigabyte since there is very little on the internet in terms of the issue that I have.

[Djw_]

48 minutes ago, MrSimplicity said:

good to know. Well still, if he's on a 120v grid a 1600w PSU is definitely gonna give him issues depending on what size fuse is on the circuit he's plugged into. He could definitely pull that if it's a 20amp circuit and it's the only thing on it. 

I'm on a 120 v.

[MrSimplicity]

23 hours ago, Djw_ said:

Task manager says base speed is 3.2 GHz and idles around 1-3 GHz but in bios and some other programs they say 5.7 GHz and will fluctuate from 1 GHz up to 6 GHz. When Undervolting it seems the clock speed stays between 1.1-4.4 and occasionally jumps to 6 GHz.

Okay,sounds like you either have a faulty CPU (less likely) or a faulty motherboard (more likely). I'm just curious what happens if you get a UPS cuz that solved a lot of issues for me back in the day. Just remember you'll need a 2000va+ being you have a 1600w PSU. Mine is only 1500va but I also only have a 1000w PSU. Clean power is one helluva drug for a computer.

 

Unfortunately I'm also still hung up on the chance your CPU Cooler isn't seated properly. You haven't commented yet saying that you are 100% certain that it is. Also for a sanity check make sure that the pump/cpu block isn't the high point in the loop. That would causes worse problems for you.

 

If you are certain it's not the cooler and you can get a large enough UPS (which is nice to have regardless) then you'll have to look into the CPU or Motherboard.

23 hours ago, Djw_ said:

The reason I got the 1600 W was because it was $300

There's 2 things you don't cheap out on in a system. Graphics and Power. You didn't get a cheap GPU that's for sure. Though the PSU is a different story. I'm not saying it's a bad PSU but I can't atest to it's reliability or brand honesty since I've never heard of it. Back in the day I had 2x 980ti's and I threw in a cheap $100 1000w Rosewill 80+ bronze PSU with them and within 4 months it fried itself. My 1000w now was $400 but that had more to do with it being an Asus ROG branded product. It's an amazing PSU and even saved me from my own stupidity and a major shortage.

[Djw_]

4 hours ago, MrSimplicity said:

Okay,sounds like you either have a faulty CPU (less likely) or a faulty motherboard (more likely). I'm just curious what happens if you get a UPS cuz that solved a lot of issues for me back in the day. Just remember you'll need a 2000va+ being you have a 1600w PSU. Mine is only 1500va but I also only have a 1000w PSU. Clean power is one helluva drug for a computer.

 

Unfortunately I'm also still hung up on the chance your CPU Cooler isn't seated properly. You haven't commented yet saying that you are 100% certain that it is. Also for a sanity check make sure that the pump/cpu block isn't the high point in the loop. That would causes worse problems for you.

 

If you are certain it's not the cooler and you can get a large enough UPS (which is nice to have regardless) then you'll have to look into the CPU or Motherboard.

There's 2 things you don't cheap out on in a system. Graphics and Power. You didn't get a cheap GPU that's for sure. Though the PSU is a different story. I'm not saying it's a bad PSU but I can't atest to it's reliability or brand honesty since I've never heard of it. Back in the day I had 2x 980ti's and I threw in a cheap $100 1000w Rosewill 80+ bronze PSU with them and within 4 months it fried itself. My 1000w now was $400 but that had more to do with it being an Asus ROG branded product. It's an amazing PSU and even saved me from my own stupidity and a major shortage.

Right now I'm at my college dorm, but I'll be back home on Thursday if I need to do anymore testing with it. I've looked into the UPS which are 2000va, do you have any recommendations? The prices that I see are everywhere from 200-2000+ for the same kva. I mounted the radiator at the top of the case as that is the only place it would fit so I don't believe that the airbubbles could be trapped. I also replaced the thermal paste to make sure, I'll take a picture when I get back of the inside of the case, but I'm 90% sure that the CPU Cooler is correctly installed.

You can check out the reviews from NEWEGG for the PSU, I'm 80% sure that the PSU is not the problem.
 https://www.newegg.com/super-flower-leadex-titanium-sf-1600f14ht-1600w/p/1HU-024C-00014?Item=9SIAMNPB0A4085&nm_mc=AFC-RAN-COM&cm_mmc=afc-ran-com-_-PCPartPicker&utm_medium=affiliate&utm_campaign=afc-ran-com-_-PCPartPicker&utm_source=afc-PCPartPicker&AFFID=2558510&AFFNAME=PCPartPicker&ACRID=1&ASID=https%3a%2f%2fpcpartpicker.com%2fproduct%2f2BcRsY%2fsuper-flower-leadex-titanium-1600-w-80-titanium-certified-fully-modular-btx-power-supply-sf-1600f14ht&ranMID=44583&ranEAID=2558510&ranSiteID=8BacdVP0GFs-1d4htjSz9Cmfc8h3E7uJZQ

[MrSimplicity]

Personally I don't trust newegg anymore. It's basically just become Amazon and bloated with a pile of Chinese crap and poor customer service. Luckily I have a micro center near me that I can drive to when I need stuff. If you got one near campus or home, I highly recommend just going there even an hour drive is worth it. I think both of my UPS's are 1500va and were about $215 a piece.

 

Looking at higher end UPSs it looks like you might have an issue there. The best value I found was a $800 2200va rack mounted one. So you might be better off figuring out what's with your CPU/Mobo first. You could try getting a 1500va for $200 and see what happens.

 

I really wish I had a better answer for you. CPU is probably fine since typical errors keep it from posting. You could look into seeing if there's an update for your Mobo BIOS. Oddly enough PSU issues don't cause immediate failures, it'll usually work until under load and you get a BSOD or it'll eventually just quit. Next time you get a BSOD try to remember what the error code is. Or even get a picture of it if you can. Sorry that I'm leaning so hard against the PSU, all the evidence just seems to be pointing towards it.

[Bjoolz]

On 3/31/2024 at 1:09 AM, Djw_ said:

My computer BSOD when set to 50

You only BSOD when messing with the CPU settings right? If you have had other BSODs: Go to C:\Windows\Minidump and check if you have any minidump files. If you do, go back to the Windows folder and copy the Minidump folder itself to the Downloads folder (You can use the desktop if you don't have OneDrive syncing files). Zip the copied folder and attach it to a post. Please follow the instructions to the letter as Windows doesn't like you messing with files in this location.

[Djw_]

10 hours ago, MrSimplicity said:

Personally I don't trust newegg anymore. It's basically just become Amazon and bloated with a pile of Chinese crap and poor customer service. Luckily I have a micro center near me that I can drive to when I need stuff. If you got one near campus or home, I highly recommend just going there even an hour drive is worth it. I think both of my UPS's are 1500va and were about $215 a piece.

 

Looking at higher end UPSs it looks like you might have an issue there. The best value I found was a $800 2200va rack mounted one. So you might be better off figuring out what's with your CPU/Mobo first. You could try getting a 1500va for $200 and see what happens.

 

I really wish I had a better answer for you. CPU is probably fine since typical errors keep it from posting. You could look into seeing if there's an update for your Mobo BIOS. Oddly enough PSU issues don't cause immediate failures, it'll usually work until under load and you get a BSOD or it'll eventually just quit. Next time you get a BSOD try to remember what the error code is. Or even get a picture of it if you can. Sorry that I'm leaning so hard against the PSU, all the evidence just seems to be pointing towards it.

I wish I had micro center near me, but I’m in Phoenix and there aren’t any, otherwise I probably would have gotten my parts from there. Motherboard BIOS is up to date, the most recent was March 22, I’m almost possible it isn’t to do with software as all my drivers are up to date. I don’t have another PSU that is powerful enough to run the computer on. I’m not sure if you’ve checked out the exact specs for the PSU if maybe it isn’t compatible or if there are any issues, I’ve looked at them but don’t really know what they mean exactly. If you want to check it out here it is.  https://www.super-flower.com.tw/en/products/leadex-titanium-1600w-20221201134111

 

5 hours ago, Bjoolz said:

You only BSOD when messing with the CPU settings right? If you have had other BSODs: Go to C:\Windows\Minidump and check if you have any minidump files. If you do, go back to the Windows folder and copy the Minidump folder itself to the Downloads folder (You can use the desktop if you don't have OneDrive syncing files). Zip the copied folder and attach it to a post. Please follow the instructions to the letter as Windows doesn't like you messing with files in this location.

Yeah I’ve only had BSOD when messing with CPU settings and I had to reset BIOS/CMOS to get my computer to boot after trying to set the multiplier to 50.

[MrSimplicity]

On 4/2/2024 at 2:39 PM, Djw_ said:

If you want to check it out here it is.

Everything always looks good on paper. Ultimately I'm not saying that it's the PSU's faulty or anything like that. All of your problems just seem to be PSU related. Like I said before, being on 120v can be massive problem for such a high wattage PSU. Especially if your Mobo is asking for too much power. Or just having dirty power on the grid.

On 3/31/2024 at 9:22 PM, Djw_ said:

When Undervolting it seems the clock speed stays between 1.1-4.4 and occasionally jumps to 6 GHz.

Now that I'm thinking too (even though unlikely) there's another setting in the CPU section of the BIOS that coincides with the multiplier. I forgot what it was called but it'll be a number around 100-105 or it's set to auto. If it's set to a ridiculous number even just above 105 can cause issues that look like instable overclocks (which essentially it is). Simply put, it's a secondary multiplier. i.e. 50 x 100 = 5000 = 5GHz, 57 x 100 = 5700 = 5.7GHz, 57 x 105 = 5985 = 5.985GHz (~6GHz), 50 x 105 = 5250 = 5.25GHz. In which case can just mean that your CPU is just bad or at least there are a couple cores that can't reach past 5GHz stably. TM may randomly report 6GHz but that could literally be on just one single core and you have 24 of them. I'm actually surprised that it passed QC. Though unfortunately the CPU is technically only spec'd at 3.2GHz. So I would still undo the undervolting but set the multiplier to 44 or 45 and set the secondary to 100 instead of auto or whatever it's set to.

On 4/2/2024 at 1:46 AM, Djw_ said:

I'll be back home on Thursday

Did you ever get a chance to see what the BSOD message was? This would give a major insight into what your computer is saying that the problem is.

[Djw_]

On 4/7/2024 at 4:48 AM, MrSimplicity said:

Everything always looks good on paper. Ultimately I'm not saying that it's the PSU's faulty or anything like that. All of your problems just seem to be PSU related. Like I said before, being on 120v can be massive problem for such a high wattage PSU. Especially if your Mobo is asking for too much power. Or just having dirty power on the grid.

Now that I'm thinking too (even though unlikely) there's another setting in the CPU section of the BIOS that coincides with the multiplier. I forgot what it was called but it'll be a number around 100-105 or it's set to auto. If it's set to a ridiculous number even just above 105 can cause issues that look like instable overclocks (which essentially it is). Simply put, it's a secondary multiplier. i.e. 50 x 100 = 5000 = 5GHz, 57 x 100 = 5700 = 5.7GHz, 57 x 105 = 5985 = 5.985GHz (~6GHz), 50 x 105 = 5250 = 5.25GHz. In which case can just mean that your CPU is just bad or at least there are a couple cores that can't reach past 5GHz stably. TM may randomly report 6GHz but that could literally be on just one single core and you have 24 of them. I'm actually surprised that it passed QC. Though unfortunately the CPU is technically only spec'd at 3.2GHz. So I would still undo the undervolting but set the multiplier to 44 or 45 and set the secondary to 100 instead of auto or whatever it's set to.

Did you ever get a chance to see what the BSOD message was? This would give a major insight into what your computer is saying that the problem is.

Sorry I haven’t responded been busy recently and haven’t been back home for a couple weeks, but is the CPU setting you’re talking about CPU Base clock or something else? When I click on the CPU Base Clock the lowest it goes is 80.0 MHz. I’ll take a picture of my BIOS and maybe you can tell me what to do, but right now I’ll just keep the under volt since last time I tried to change things in BIOS it bluescreened and didn’t boot.

[MrSimplicity]

So the base clock is where it's supposed to be. The CPU clock ratio is the main number to change. It looks like that is set to 32 but it's still saying you're running at 5700MHz for some reason.  So IDK if you can take them off auto? Especially the base clock. But if your clock ratio is set to 32, it shouldn't be hitting 5700MHz for any reason. It should never go over 3200MHz. You could try also setting the Efficiency cores to the same number.

 

Lastly, If your core voltage is set to 1.2V then that's honestly where it should be to begin with. So IDK where that was to begin with. I think 1.35V is the max nowadays. So I think if you just leave the "undervolt" and set the clock ratio back to like 44 or even 50, it should be just fine. But as it stands you're still hitting 5.7GHz for some reason and if that's the case I would honestly just not even worry about it anymore if it's been running just fine. If you run into more issues down the line then there's definitely a bigger issue to look into.

 

It also wouldn't hurt to try contacting manufacturers though and asking them. Also if at any point you do get another BSOD try to remember the error code it gives you so we can have something better to go off of.