Computer restarts when playing games

[matiasl9815]

4 minutes ago, Mindovermaster said:

Ever try going back on a previous driver? See if that works any better?

 

If you had a spare GPU, I would try that out. Would rule out that the GPU was the problem. It might not be getting hot, but other parts of the card might be damaged.

Tried a lot of previous versions of the driver

sadly I don't have a spare gpu

[sikhwarrior]

7 minutes ago, matiasl9815 said:

Tried a lot of previous versions of the driver

sadly I don't have a spare gpu

Going to sound crazy but try without the gpu and just use the built-in one ( on the processor). Have you updated the game in case there is a patch that could be causing the random reboot.

[Mindovermaster Moderator]

9 minutes ago, sikhwarrior said:

Going to sound crazy but try without the gpu and just use the built-in one ( on the processor). Have you updated the game in case there is a patch that could be causing the random reboot.

AFAIK, that Ryzen 3 1300x doesn't have an APU. Onboard graphics will not work.

[matiasl9815]

15 minutes ago, Mindovermaster said:

AFAIK, that Ryzen 3 1300x doesn't have an APU. Onboard graphics will not work.

that's right it doesn't have it

[DevTech]

21 hours ago, DevTech said:

BIOS

-------

 

- make sure "auto overclock" is off

- then make sure all timings are slowed down a bit

- make sure voltages are set correct

- update mobo BIOS

 

- turn ON "load line calibration" or similar in BIOS

 

- if you didn't previously overvolt the CPU, try a small CPU overvolt to increase stability but keep it small

 

 

 

Another thing to try:

 

- List all programs on computer and temp uninstall ones that might install a "software device driver"

 

- List all devices on mobo and computer and go to each one's manufacturer website and download latest drivers

 

Recent versions of Windows have been changing the way drivers work and any OLD drivers are prime suspects for weird issues..

 

-----------------------------------------

 

Cut n paste of my previous suggestions due to possible reading comprehension issue.

 

PEOPLE: stop the endless repeats of useless advice! He has a specific problem so think about the specifics! THINK!

 

 

[DevTech]

6 hours ago, briangw said:

Thanks for these tips. My son bought a CyberPowerPC on Black Friday and the exact same thing is happening but only with Battlefield 5. He does have 8 GBs of RAM which he will be increasing shortly. It's a Ryzen CPU. BIOS looks fine although there is an update to it that allows for memory compatibility but you have to load it in the BIOS vs. a neat installer package for Windows (ugh, why can't they write it to install it in Windows like my Mobo?). Mobo is MSI B450m Bazooka. After reading some reviews on Newegg, it seems many people are having issues with the PC rebooting with BF 5 so I'm hoping we can correct this with a RAM upgrade.

Both mobos are probably having the same old as the hills problem: some component is simply BORDERLINE.

 

All sorts of things can be borderline, but usually unexpected reboots without a BSOD are VOLTAGE RELATED.

 

Voltage problems are 99.99% due to motherboard, NOT the PSU, since it is the motherboard on board PSU that takes 12 VOLTS from the PSU and turns it into > 40 AMPS of car battery type current to the CPU.

 

If that voltage glitches, you have "Mysterious Reboot"

 

High end motherboards have special circuits usually called LOAD LINE CALIBRATION to counter voltage droops on the fly with a dynamic circuit. This needs to be turned ON in the BIOS since it uses more energy and so default is OFF.

 

If the motherboard does not have this circuit or even if it does but it is just crappy, then you can OVERVOLT the CPU by manually setting the voltage up by about 1% to 5% but it is vital to make sure that mobo "AUTO-OVERCLOCK" or "AUTO-TURBO" etc is turned off since in rare cases those stupid options also do an AUTO-OVERVOLT.

 

So if you OVERVOLT your CPU a bit you increase system stability a lot at the expense of making your CPU COOLER work a bit harder.

 

While in the BIOS, you  have the option to slow down the RAM a bit and even the CPU to increase SYSTEM STABILITY. RAM that is 10% slower will be completely not noticed, but CPU reduction might a bit so 5% is something to try there.

 

 

 

[matiasl9815]

1 hour ago, DevTech said:

- update mobo BIOS

 

- turn ON "load line calibration" or similar in BIOS

 

- if you didn't previously overvolt the CPU, try a small CPU overvolt to increase stability but keep it small

 

 

 

Another thing to try:

 

- List all programs on computer and temp uninstall ones that might install a "software device driver"

 

- List all devices on mobo and computer and go to each one's manufacturer website and download latest drivers

 

Recent versions of Windows have been changing the way drivers work and any OLD drivers are prime suspects for weird issues..

 

-----------------------------------------

 

Cut n paste of my previous suggestions due to possible reading comprehension issue.

 

PEOPLE: stop the endless repeats of useless advice! He has a specific problem so think about the specifics! THINK!

 

 

Hello, no comprehension issue here, I answered you back, I updated the bios, didn't found neither load line calibration or something similar in my bios. I don't know how to overvolt, I searched all day and found nothing useful. I did reinstall drivers downloading them from the manufacturer site, after installing a nvidia driver from the geforce app it seems like it got fixed but i don't know yet so I'm testing it now

[DevTech]

1 hour ago, matiasl9815 said:

Hello, no comprehension issue here, I answered you back, I updated the bios, didn't found neither load line calibration or something similar in my bios. I don't know how to overvolt, I searched all day and found nothing useful. I did reinstall drivers downloading them from the manufacturer site, after installing a nvidia driver from the geforce app it seems like it got fixed but i don't know yet so I'm testing it now

Allrighty-then, there will be a BIOS screen that sets the voltage of all components, sometimes you need to turn a setting on the main CPU settings from "AUTO" to "MANUAL"

 

Then you look at the VID of your CPU since it varies from silicon batch to batch and use that as the starting voltage. If your chip's default VID is 1.0 Volts for example then you would try 1.05 Volts and monitor both the increase in temps and the results to see if it is worthwhile.

 

Sometimes you can also do the same for the GPU.

 

Just double-check all your entries - for example people in a rush with ZERO attention span (all Facebook and Instagram users near as I can tell) might set CPU to 1.5 instead of 1.05 and will fry it.

 

Sizzle. Sizzle. Snap. Crackle. Pop.

 

 

 

[matiasl9815]

On 12/17/2018 at 11:36 PM, DevTech said:

Allrighty-then, there will be a BIOS screen that sets the voltage of all components, sometimes you need to turn a setting on the main CPU settings from "AUTO" to "MANUAL"

 

Then you look at the VID of your CPU since it varies from silicon batch to batch and use that as the starting voltage. If your chip's default VID is 1.0 Volts for example then you would try 1.05 Volts and monitor both the increase in temps and the results to see if it is worthwhile.

 

Sometimes you can also do the same for the GPU.

 

Just double-check all your entries - for example people in a rush with ZERO attention span (all Facebook and Instagram users near as I can tell) might set CPU to 1.5 instead of 1.05 and will fry it.

 

Sizzle. Sizzle. Snap. Crackle. Pop.

 

 

 

These are the options that I got

[DevTech]

17 minutes ago, matiasl9815 said:

These are the options that I got

https://www.reddit.com/r/Amd/comments/7tpdax/normal_vcore_on_ryzen_3_1300x/

 

1. OK, so you want a high quality CPU Cooler, preferably something like a Corsair H100, just to eliminate heating as an issue so you can get just the right voltage to max your STABILITY

 

2. Stay at 1.4 Volts or under, 1.4 Volts is the upper end of the range so it might produce lots of heat, hence need for cooling.

 

3. Set "Overclock Tuner" from AUTO to MANUAL as I previously stated.

 

4. You want to avoid "offset mode" which increments voltage and can make calculations dangerously confusing. You just want to set an EXACT voltage and have nothing else muck with that.

 

5. So you did in fact have "Load Line Calibration" in fact a few settings in that area. Set it all to "Extreme" which does a SOLID CLAMP on your voltage at the expense of more heat, so make sure you have a fan blowing on mobo near the CPU socket if you use integrated water block on the CPU since otherwise there is usually no air flow in that area.

 

6. You don't show the options for "Power Phase Control" but set that to the highest number which simply means the number of voltage regulators the mobo will turn on for you - more = better quality power, yeah more heat.

 

NOTE: if you just adjust the load line settings to max, it probably fixes your problem right there, but mobos that automatically adjust CPU VOLTAGE, I find to be personally very creepy and I always turn that off.

 

NOTE: your mobo was already automatically overclocking your CPU so you might want to manually do a reasonable overclock by finding a guide for a Ryzen 3, but never bother overclocking RAM, in fact slow RAM down a bit to eliminate headaches. Staying away from RAM makes overclocking a breeze and speeding RAM gives close to ZERO real world advantage.

 

 

 

 

 

 

[Slugsie]

My PC recently developed a fault whereby it would stutter sometimes when playing video, and randomly stop when under load playing games. After various other trouble shooting attempts I ran the Windows 10 RAM tester and it turned out I had a faulty RAM stick (2x 8GB, one of which wasn't working quite right). Replaced under warranty and running flawlessly since. Worth a shot?

[Mindovermaster Moderator]

6 minutes ago, Slugsie said:

My PC recently developed a fault whereby it would stutter sometimes when playing video, and randomly stop when under load playing games. After various other trouble shooting attempts I ran the Windows 10 RAM tester and it turned out I had a faulty RAM stick (2x 8GB, one of which wasn't working quite right). Replaced under warranty and running flawlessly since. Worth a shot?

Yeah, that's just ONE thing that can happen. There's like 1000 more variables. Until he tries something, we may never know.

[DevTech]

54 minutes ago, Slugsie said:

My PC recently developed a fault whereby it would stutter sometimes when playing video, and randomly stop when under load playing games. After various other trouble shooting attempts I ran the Windows 10 RAM tester and it turned out I had a faulty RAM stick (2x 8GB, one of which wasn't working quite right). Replaced under warranty and running flawlessly since. Worth a shot?

He already tried that. It would not have been my first choice anyways...

 

[Mindovermaster Moderator]

12 minutes ago, DevTech said:

He already tried that. It would not have been my first choice anyways...

 

And that would create a stutter, not a shut down.

[DevTech]

48 minutes ago, Mindovermaster said:

Yeah, that's just ONE thing that can happen. There's like 1000 more variables. Until he tries something, we may never know.

Well true, but that is what a diagnostic technique is all about.

 

You don't shotgun a million options, you narrow down more and more based on symptoms and performing tests from already eliminated trees of the diagnostic forking, just adds a lot of noise.

 

Random reboots without an entry in the Event Log or a BSOD error code are almost always VOLTAGE fluctuations.

 

And VOLTAGE fluctuations are usually NOT the PSU (counter-intuitive)

 

And there are other less likely issues such as a branch of the diagnostic tree which almost NEVER gets any attention that is "Software Device Drivers" which are "pretend" device drivers installed by applications you might have installed that are needed to get access to low level PC stuff that are NOT part of the Windows API such as Core Temp readings etc.

 

These "pretend" device drivers can seriously de-stabilize a computer!

 

And starting with 1803, Microsoft has created a new security model for device drivers based on sandboxing them into a Hyper-V virtual environment which for sure is going to FREAK OUT old obsolete device drivers. The timing of this one explains perhaps some of the hardware issues where the computer suddenly seems to be erratic after a few months of usage... Maybe not many, but some for sure...

 

 

 

[DevTech]

6 minutes ago, Mindovermaster said:

And that would create a stutter, not a shut down.

I'm pretty sure you know this but sometimes your response speed to posts is admirably way too crazy-fast which makes thinsg fun and move-along at a nice clip and...

 

Sometimes fail to be precise... LOW RAM will cause gaming stutter, #2 cause after #1 insufficient GPU Performance but BAD RAM will almost always lead to a BSOD...

 

[Mindovermaster Moderator]

1 minute ago, DevTech said:

I'm pretty sure you know this but sometimes your response speed to posts is admirably way too crazy-fast which makes thinsg fun and move-along at a nice clip and...

Yeah, you too... 

 

But, yeah, faulty RAM can do a number of odd things. Why there is a program called Memtest.  

[DevTech]

I guess it's my turn at not being precise:

 

BAD RAM:

 

1. Will BSOD if executing Kernel-Level code

 

2. Will crash app if executing app code

 

3. Will BSOD is used to store important dynamic system data (OS call parameters etc)

 

4. Will cause insane to debug on app reading bad data

 

All those possibiliies assume a double bit error which can defeat Parity Checks.

 

So way more complicated than I implied...

 

[Slugsie]

13 hours ago, Mindovermaster said:

Yeah, that's just ONE thing that can happen. There's like 1000 more variables. Until he tries something, we may never know.

Very true, essence of diagnosis is try things one at a time until something makes a difference.

12 hours ago, DevTech said:

He already tried that. It would not have been my first choice anyways...

 

Neither was it my first choice. I'd already ruled out a Windows corruption (fresh install) and a HD/SSD issue. My next step was a RAM check which came up faulty.

12 hours ago, Mindovermaster said:

And that would create a stutter, not a shut down.

I was getting both. Stutters while playing videos. Random outright resets while playing games.

[Astra.Xtreme]

12 hours ago, DevTech said:

I guess it's my turn at not being precise:

 

BAD RAM:

 

1. Will BSOD if executing Kernel-Level code

 

2. Will crash app if executing app code

 

3. Will BSOD is used to store important dynamic system data (OS call parameters etc)

 

4. Will cause insane to debug on app reading bad data

 

All those possibiliies assume a double bit error which can defeat Parity Checks.

 

So way more complicated than I implied...

 

This is my thought as well.  I've seen bad RAM be a pain many times, and it typically shows a kernal power log in the Event Viewer.

 

Try running the PC on a single RAM stick and see if one of them is the culprit.

[DevTech]

2 hours ago, Slugsie said:

Very true, essence of diagnosis is try things one at a time until something makes a difference.

Neither was it my first choice. I'd already ruled out a Windows corruption (fresh install) and a HD/SSD issue. My next step was a RAM check which came up faulty.

I was getting both. Stutters while playing videos. Random outright resets while playing games.

 

1 hour ago, Astra.Xtreme said:

This is my thought as well.  I've seen bad RAM be a pain many times, and it typically shows a kernal power log in the Event Viewer.

 

Try running the PC on a single RAM stick and see if one of them is the culprit.

OMG - we were discussing how RAM was probably NOT the issue for the OP.

 

All that is just going to confuse him.

 

OP - ignore that stuff and focus on your mobo voltage settings...

 

[Mindovermaster Moderator]

26 minutes ago, DevTech said:

OP - ignore that stuff and focus on your mobo voltage settings...

 

How do you know his voltage settings are the issue here? That's just one assumption.

 

OP, if I were you, I would do a factory reset on your BIOS/UEFI and start from the ground up...

[Astra.Xtreme]

2 hours ago, DevTech said:

 

OMG - we were discussing how RAM was probably NOT the issue for the OP.

 

All that is just going to confuse him.

 

OP - ignore that stuff and focus on your mobo voltage settings...

 

I highly doubt it's a voltage issue.  That would be a massive design defect that hardly anybody would have the knowledge to fix.

[Slugsie]

4 hours ago, DevTech said:

 

OMG - we were discussing how RAM was probably NOT the issue for the OP.

 

All that is just going to confuse him.

 

OP - ignore that stuff and focus on your mobo voltage settings...

 

Or OP could just set the Memory Diagnostic to run just before going to bed tonight and quickly rule out one possible cause.

[DevTech]

9 minutes ago, Slugsie said:

Or OP could just set the Memory Diagnostic to run just before going to bed tonight and quickly rule out one possible cause.

Or he could surround his computer with garlic bulbs to ward away Vampires...

 

That is simply NOT how one employs a diagnostic process or else Sherlock Holmes would still be wandering around in the moors surrounded by howling hounds of the Baskervilles...

 

In any case, people keep failing to actually READ this thread where the OP stated early on that he performed a RAM test.