What is wrong with my Ryzen?

[XBloodyR]

Im running my Ryzen 1700 at 3.7 Ghz on 1.2625V with stock cooler it idles at about 35-45 degrees with an ambient of 25 degrees. Now i did a realbench stress test with 8GB of memory tested and it crashed after 10 minutes and topped out at 86 degrees, I am using HWmonitor it shows about the same temps as ryzen master usually does. Isnt that kind of high?
I have msi b350 tomahawk with beta bios 1.62 and running trident z 3200 cl16 on 2933 cl14 could the issue actually be that the ram is unstable, also I dont trust the temp in this bios could it be just a false temp?

 

[Seanie280672]

Im running my Ryzen 1700 at 3.7 Ghz on 1.2625V with stock cooler it idles at about 35-45 degrees with an ambient of 25 degrees. Now i did a realbench stress test with 8GB of memory tested and it crashed after 10 minutes and topped out at 86 degrees, I am using HWmonitor it shows about the same temps as ryzen master usually does. Isnt that kind of high?
I have msi b350 tomahawk with beta bios 1.62 and running trident z 3200 cl16 on 2933 cl14 could the issue actually be that the ram is unstable, also I dont trust the temp in this bios could it be just a false temp?

That temp is a bit high, idle temps are normal, 10oC above ambient is normal, is 86oC the max or is it still going up ? but I would say your instability is the RAM, also, that bios is beta, it may not be fully stable yet.

Trident Z 3200mhz CL16 is Hynix Chips, im not even sure if they are capable of CL14, try CL15 at a later date, for now use stock timings to rule out the ram being the issue, if you can pass stress tests, then you know it was the ram.

 

[XBloodyR]

So i tried again with 2666 16 18 18 36 and it kind of passed the test but when it said stopping stress test after the 15 mins it crashed again my monitor got no input anymore like in the previous crashes. The temperature topped out at 85 degree this time it doesent go higher than that.
I should also mention many people are having problems with the RGB of this RAM the control software seems to change the SPD of the ram. I cant get one stick to light up anymore so i wanted to sent them back anyway the reason i mention this is because both sticks are reporting CRC errors with the Thaiphoon burner software this might also cause instability?

Edit: Crashing when it is stopping the benchmark seems to be a nvidia driver related issue so you can say it actually passed the test.

 

[Seanie280672]

So i tried again with 2666 16 18 18 36 and it kind of passed the test but when it said stopping stress test after the 15 mins it crashed again my monitor got no input anymore like in the previous crashes. The temperature topped out at 85 degree this time it doesent go higher than that.
I should also mention many people are having problems with the RGB of this RAM the control software seems to change the SPD of the ram. I cant get one stick to light up anymore so i wanted to sent them back anyway the reason i mention this is because both sticks are reporting CRC errors with the Thaiphoon burner software this might also cause instability?

Edit: Crashing when it is stopping the benchmark seems to be a nvidia driver related issue so you can say it actually passed the test.

I would get rid of that beta bios to start and flash the latest official bios, that could be the root of all of your problems.

First of all, dont use any software to control the LEDs on your RAM, I didnt realise you had the RGB version, un-install it completely, it changes the LED colours and pattern by modifying the SPD data stored on the sticks, and yes this does corrupt the SPD data, you can fix it with Thaiphoon, I fixed mine, but you will have to purchase it at $16 to do that, or send them back, when you get the new sticks, just live with rainbow effect or get the regular non rgb sticks, once you've uninstalled the software, run Thaiphoon again and see if you still get the CRC error message.

2nd, I would say your CPU isnt stable yet, maybe it requires a little more voltage.

3rd, I read all the time about BSOD's and people having problems with the nVidia drivers, are you using the latest ?

To find out for sure if the CPU is causing the crashes, try stress test with IBT, set it for 50 passes, standard, that will test CPU only, if you want to then include ram, set it to 20-25 passes max setting, max will take about 6 hours to run, standard takes about 15 mins: http://www.techspot.com/downloads/4965-intelburntest.html

 

[XBloodyR]

I am using the latest drivers. So i flashed the bios to the latest official one and it did not change anything really it crashed again at the end then i set the cpu voltage to 1.275V and ran the intel burn test on standard 50 passes and it crashed after about 8 passes and got up to 89 degrees which kind of scared me wanted to stop it soon anyway. How is it possible that i cant even handle 1.275V other people do 1.35V and 3.8 OC with the stock cooler.. I have 5 case fans 2 intake front and 3 exhaust 2 of them on top 1 on back all running at 100% at above 75 degree also the cpu cooler runs at 100% at 80 degree do i really have to lower the OC to 3.65?

Edit: I changed the VCore to 1.2875 NB Voltage to 1.1 and the LLC to Mode 2 = 20% did 5 passes in IBT it rose to 92 degrees in just 5 passes.. Now with realbench it did not crash this time but the temps got crazy hot the max was 94 degree.

 

[Seanie280672]

I am using the latest drivers. So i flashed the bios to the latest official one and it did not change anything really it crashed again at the end then i set the cpu voltage to 1.275V and ran the intel burn test on standard 50 passes and it crashed after about 8 passes and got up to 89 degrees which kind of scared me wanted to stop it soon anyway. How is it possible that i cant even handle 1.275V other people do 1.35V and 3.8 OC with the stock cooler.. I have 5 case fans 2 intake front and 3 exhaust 2 of them on top 1 on back all running at 100% at above 75 degree also the cpu cooler runs at 100% at 80 degree do i really have to lower the OC to 3.65?

Edit: I changed the VCore to 1.2875 NB Voltage to 1.1 and the LLC to Mode 2 = 20% did 5 passes in IBT it rose to 92 degrees in just 5 passes.. Now with realbench it did not crash this time but the temps got crazy hot the max was 94 degree.

You are getting there, the NB voltage helps a lot, but dont forget, the NB is in the CPU too, so raising the voltage affects the CPU temps too.

If you have any thermal paste, it might be worth trying to remount the cooler, see if you can get those temps down a little, failing that, you can always pick up another cooler fairly cheap.

 

[XBloodyR]

Im going to get some good thermal paste then. What do you think is 94 degree max acceptable for stress test? seems like a scary temp to me.

 

[Seanie280672]

Im going to get some good thermal paste then. What do you think is 94 degree max acceptable for stress test? seems like a scary temp to me.

Defo a bit on the high side, they throttle at 95, so you are right on the edge there, I wouldnt really allow mine to go above 75-80oC in all fairness.