Ryzen bsod

[Entermaker]

I bought my system about 3 months ago and since then I always have some kind of problem with it.

First I had to RMA my processor because of the segfault problem.

I'm also experiencing system crashes (bsod) since I put together the system, even after the replaced processor.

 

The error message is always the same:

Bug Check String: DRIVER_IRQL_NOT_LESS_OR_EQUAL

Caused by Driver: pci.sys

Caused by Address: pci.sys+ce9b

Crash Address: ntoskrnl.exe+6f980

 

The bsod happens when I play games.

This makes me think that the problem is caused by the gpu, however if I reduce the memory frequency the bsod gets more rare.

Last time I ran the memory at 2400MHz when it happened again.

 

Can someone help me find out which component causes the problem?

 

System:

CPU: Ryzen 7 1700 (stock)

GPU: Sapphire RX 560 4G

Mobo: ASRock X370 Gaming K4

Memory: G.Skill Ripjaws v 3200MHz (2x8)

PSU: Seasonic G Series 650W

[Jaxder]

Run a few tests:

memtest86

your preferred GPU stress test

your preferred CPU stress test

 

 You could probably run everything in Aida64 though.

 

[Entermaker]

I have already ran memtestx86, CineBench and Superposition Benchmark all of them had no problem.

Is there anything else I can try?

[just_dave]

Try reseating the CPU

[Entermaker]

Ok, I'm gonna try these out.

[johndms]

2 hours ago, Entermaker said:

I have already ran memtestx86, CineBench and Superposition Benchmark all of them had no problem.

Is there anything else I can try?

Can you run 10 passes of test #6 in Memtest86? If it's ran with All CPUs in Parallel option checked, it should take about 20 minutes. Updating the bios to the most recent version can often improve stability. With the CPU left at stock frequency, manually set the cpu core voltage to 1.35v and see if there are any issues.

[Entermaker]

Thanks for the replies.

 

The bios is up to date and the system passes all tests.

 

I spent a lot of time tuning the memory frequencies and it turned out that the system can go up to 2933MHz, but in some cases it refuses to boot: once it works, once it reboots.

 

The memory seems to be very unstable. In some cases (sometimes even with 2666MHz) Windows freezes after the boot logo, the screen goes to black or green and the motherboard's debug LED shows "00".

Based on the manual it refers to a CPU problem, but it works fine (I just got it back from RMA).

 

Do you have any idea which component causes the instability?

I think it's the memory or the motherboard.

[Oxydoreduction]

42 minutes ago, Entermaker said:

Thanks for the replies.

 

The bios is up to date and the system passes all tests.

 

I spent a lot of time tuning the memory frequencies and it turned out that the system can go up to 2933MHz, but in some cases it refuses to boot: once it works, once it reboots.

 

The memory seems to be very unstable. In some cases (sometimes even with 2666MHz) Windows freezes after the boot logo, the screen goes to black or green and the motherboard's debug LED shows "00".

Based on the manual it refers to a CPU problem, but it works fine (I just got it back from RMA).

 

Do you have any idea which component causes the instability?

I think it's the memory or the motherboard.

For the moment I also got odd problems with my Ryzen system. Screen just goes black and usb peripherals just shut down. I'd bet also for a motherboard oe ram problem. Do you have any other ram sticks to test it?

[doddlark]

Ive had my 1700x for just under a month now and since I received it have been having the same issues, games crashing randomly, sometimes with an "XXXX.exe (XXXX being the game) has been blocked from accessing graphics hardware" or more recently with blue screens with that same driver not less or equal. Ive ran AIDA64 stressing CPU, FPU and cache and i get "Warning Hardware failure detected! Test stopped", also I ran memtest a few times, the first time I ran it I got a plethora of errors but now nothing? (nothing has changed with my ram speed/timings)

Time to send my system back to where I bought it?

Ryzen 1700x (stock)
Zotac 1080ti AMP Extreme
ASRock X370 Professional Gaming
G.Skill TridentZ 4x16gb 3200mhz @3066 16-16-16-36
Corsair RM750X

[Oxydoreduction]

44 minutes ago, doddlark said:

Ive had my 1700x for just under a month now and since I received it have been having the same issues, games crashing randomly, sometimes with an "XXXX.exe (XXXX being the game) has been blocked from accessing graphics hardware" or more recently with blue screens with that same driver not less or equal. Ive ran AIDA64 stressing CPU, FPU and cache and i get "Warning Hardware failure detected! Test stopped", also I ran memtest a few times, the first time I ran it I got a plethora of errors but now nothing? (nothing has changed with my ram speed/timings)

Time to send my system back to where I bought it?

Ryzen 1700x (stock)
Zotac 1080ti AMP Extreme
ASRock X370 Professional Gaming
G.Skill TridentZ 4x16gb 3200mhz @3066 16-16-16-36
Corsair RM750X

Have you tried to let the ram frequency to 2133 to see if it could solve anything (I doubt it will honestly)?

Maybe test with another graphics card if you have one but this really makes me think of a motherboard problem.

 

 

[doddlark]

9 minutes ago, Oxydoreduction said:

Have you tried to let the ram frequency to 2133 to see if it could solve anything (I doubt it will honestly)?

Maybe test with another graphics card if you have one but this really makes me think of a motherboard problem.

 

 

Running the same test at the moment @ 2133mhz 18-18-18-38
I did also try with 2 dimms instead of loading out all 4 and in game still got crashes (didn't run aida64 with 2 dimms, but Im going to assume since its still crashing that it would show the same result)

I dont have another graphics card unfortunately as Ive been a mac dude for the last 4 years or so haha

should I just get ready to send the system back?

[Oxydoreduction]

1 minute ago, doddlark said:

Running the same test at the moment @ 2133mhz 18-18-18-38
I did also try with 2 dimms instead of loading out all 4 and in game still got crashes (didn't run aida64 with 2 dimms, but Im going to assume since its still crashing that it would show the same result)

I dont have another graphics card unfortunately as Ive been a mac dude for the last 4 years or so haha

should I just get ready to send the system back?

Yep, I think you'll have to send it back sadly.

 

[doddlark]

1 minute ago, Oxydoreduction said:

Yep, I think you'll have to send it back sadly.

 

blaaaaaaaaaaaaaaaast it. So am I just unlucky with the board/chip I got or is something not compatible here?

[johndms]

 

Blah, I'm quoting the wrong person.. fixing that. Get your own thread, people.

 

 

 

[Oxydoreduction]

38 minutes ago, doddlark said:

blaaaaaaaaaaaaaaaast it. So am I just unlucky with the board/chip I got or is something not compatible here?

Nah, just unlucky I guess but hey, we're 2 to have the roughly the same issue.

That's already my second board by the way.

Still have a spare DDR4 stick to try but there's not much hope on my side either.

[johndms]

@Entermaker Too many damn people posting THEIR problems and it's getting confusing trying to follow YOURS. They should make their own threads.

14 hours ago, Entermaker said:

Thanks for the replies.

 

The bios is up to date and the system passes all tests.

 

I spent a lot of time tuning the memory frequencies and it turned out that the system can go up to 2933MHz, but in some cases it refuses to boot: once it works, once it reboots.

 

The memory seems to be very unstable. In some cases (sometimes even with 2666MHz) Windows freezes after the boot logo, the screen goes to black or green and the motherboard's debug LED shows "00".

Based on the manual it refers to a CPU problem, but it works fine (I just got it back from RMA).

 

Do you have any idea which component causes the instability?

I think it's the memory or the motherboard.

If you run the custom Memtest86 test I described earlier, all 10 passes of JUST test #6 and it showed no errors upon completion, then your ram is probably not the problem.

 

It sounds like your CPU may be the issue. You say it's at stock frequency? Try leaving it at stock, but manually enter 1.35v as the CPU Core Voltage. Run Aida64's System Stability Test stressing CPU/FPU/Cache while keeping an eye on the temperatures. You may need to click the Preferences button and add "CPU Diode" to your Aida64 monitoring window. The "CPU" reading likely isn't the one you should be watching. CPU Diode (on my ASRock board) is the same temperature reading you'll get in Ryzen Master and it's the Tdie temperature you'll see in HWiNFO64. Your CPU Diode temperature shouldn't be going above 80c. See if you can run a test for at least 1 to 2 hours.

[Entermaker]

Unfortunately I don't have any other videocard or memory, so I can' test it with different ones.

 

Yes it passed memtest86 test #6 10 times.

 

Yes, the CPU is at stock. It has never been overclocked.

 

I will try out different ram configurations and do every test again as you described.

 

[johndms]

1 hour ago, Entermaker said:

I will try out different ram configurations and do every test again as you described.

Leave your ram alone. It's not the problem.

[Entermaker]

I think you are right.

 

I set the ram speed to 2933MHz and started Aida64. In some cases no failure for an hour, but once it took just 58 seconds to get a failure.

To get hardware failure I don't even have to stress the memory.

 

But there is must be a connection between the videocard and the problem. Because the only way I get bsod is when playing games.

 

So its the CPU or the GPU or the motherboard.

 

Do you know a way to find out?

 

I don't know if it matters but when I installed my audio driver I got a crash (but the driver is working properly).

One more thing: very rarely but sometimes Microsoft Management Console crashes.

[johndms]

6 minutes ago, Entermaker said:

I don't know if it matters but when I installed my audio driver I got a crash (but the driver is working properly).

One more thing: very rarely but sometimes Microsoft Management Console crashes.

What crashes is probably irrelevant. If the CPU is having problems, random things will crash. Refresh my memory, what CPU settings are you using? Stock or overclocked, what voltage?

 

EDIT: And did you run that Memtest at 2933, or another frequency? You can't change a memory overclock without testing stability, or that'll cause more problems. Leave the memory at 2133 and let's figure out the CPU.

[Entermaker]

All CPU settings are stock, including the voltages.

 

I will try out one more thing: I will run Aida64, but this time I will stress the CPU/FPU/Cache only.

[Santo22]

1 hour ago, Entermaker said:

I think you are right.

 

I set the ram speed to 2933MHz and started Aida64. In some cases no failure for an hour, but once it took just 58 seconds to get a failure.

To get hardware failure I don't even have to stress the memory.

 

But there is must be a connection between the videocard and the problem. Because the only way I get bsod is when playing games.

 

So its the CPU or the GPU or the motherboard.

 

Do you know a way to find out?

 

I don't know if it matters but when I installed my audio driver I got a crash (but the driver is working properly).

One more thing: very rarely but sometimes Microsoft Management Console crashes.

that's a weird dip there toward the end of this test and then it goes back up. I would assume that's where something failed. Do you know what the voltage did when this dip happened? Temps look to be ok. Also, can you check your device manager and see if any of your driver's have an exclamation mark next to them? 

[Entermaker]

I ran Memtest86 multiple times with different memory speeds, including 2933MHz. It passed everything.

 

The weird dip that can be seen is because I ran multiple tests in a row. When the firs test ended(failed) the CPU usage went down and I started the test again, it went up.

 

All drivers are installed correctly, there is no exclamation mark.

 

All test are done with 2933Mhz.

I tested only the CPU, and then only the GPU for 6 hours. There was no crash.

 

I noticed one more thing: when I do a combined test(CPU+GPU) I can hear some kind of noise probably from the PSU.

 

At this point I think that the problem is the motherboard or the PSU. Unstable voltages?

[johndms]

Just now, kk027 said:

Why do you have a Ryzen 7 and a 560?

who cares? fix his problem or go away.

[johndms]

32 minutes ago, Entermaker said:

I ran Memtest86 multiple times with different memory speeds, including 2933MHz. It passed everything.

 

The weird dip that can be seen is because I ran multiple tests in a row. When the firs test ended(failed) the CPU usage went down and I started the test again, it went up.

 

All drivers are installed correctly, there is no exclamation mark.

 

All test are done with 2933Mhz.

I tested only the CPU, and then only the GPU for 6 hours. There was no crash.

 

I noticed one more thing: when I do a combined test(CPU+GPU) I can hear some kind of noise probably from the PSU.

 

At this point I think that the problem is the motherboard or the PSU. Unstable voltages?

It looks like you've eliminated the CPU, Memory, and GPU individiually. The noise may need further investigation, but one would think a 650w SeaSonic PSU would be just fine.

 

I'd try something like Unigine Valley and see if that can run for an hour or two. Valley should put a simulated gaming load on the system. Max GPU and a fair amount of CPU usage. At this point, I'm trying to force a BSOD.

 

..Continued. Your temperatures look fine, but you may be right, perhaps the combined power draw of the CPU+GPU is causing problems. Do you have another PSU to test with? 450-500W should be plenty for that build.