I'm having some weird issues where i have no answer to so i thought i ask here.
The issues im having are that i replaced my RAM from G.Skill FlareX to G.Skill Trident Z Neo and exchanged my Samsung 950 Pro for Samsung 970 EVO PLUS and all hell broke loose really.
This means that i was stable before with 4.4 GHz Ryzen 3600, 3800 MHz CL16 RAM with 1900 fclk but now its very erratic and i get this Event ID 19 error with the following error when testing with OCCT:
A corrected hardware error has occurred.
Reported by component: Processor Core
Error Source: Unknown Error Source
Error Type: Bus/Interconnect Error
Processor APIC ID: 0
The details view of this entry contains further information.
Something tells me that its the fclk but it was stable before but cannot get it to be stable now for the life of me.
Here is my hardware:
Asus crosshair vii hero wifi with BIOS 3103 now but i tried with the latest with the newest AGESA but it simply doesn't work right so i reverted back to 3103.
Nvidia RTX 2060
Samsung 970 EVO PLUS
G.Skill Trident Z Neo F4-3600C16D-16GTZN.This suppose to be Ryzen optimized RAM but its not what i see as when i load up DOCP only half of the settings are actually applied.. It worked fine on my FlareX 3200 MHz kit though. This new G.Skill RAM should be better than my FlareX and testing RAM with TM5 i find no errors but when i stressing with OCCT i get this EVENT ID 19 error.
This only happens with OCCT and in some games as far as i know and its very random. 2 days ago i tested for 6 hours without any issue but when i tried yesterday i got this WHEA EVENt ID 19 error and the whole PC became unstable..
I am very curious as to what can cause this erratic behavior and how to solve it.
I really hope there is an sollution or explanation for this as i cannot use my system anymore.
if you need more info plz let me know and i will provide all the info i can.
Your RAM config and FCLK are first suspects.
Set your RAM to 3200, default (auto) timings. FCLK 1600. See if error persists.
I used to have 3667/1866 clocks, but got WHEA 18. Setting 3200/1600 solved it completely.
Well that's the weird thing, these problems happened after i switched RAM and SSD.
Before it was perfectly stable at 1900 fclk and 3800 MHz RAM...
My current RAM is 3600 MHz DOCP stock which also can cause issues ive read on the internet.
What an cluster**bleep** of problems with Ryzen man.. sorry for my language but this is just very strange and weird.
Similar issue. But I've got the Whea 19 in idle state only.
My RAM allows 3600MHz only via XMP, so If activate it I give Whea 19 and random resets. In case the basic Jedec 2400 or manual raise to 3200Mhz I have no issues. After lots of days spent I probably caught the culprit: global c states. Even in raise to 3600mhz I have not Whea and no resets with CStates disabled. Maybe it is the voltage issue, i dont know
+ MBD ROG Strix B550-F Gaming WiFi, BIOS 2006
+ CPU Ryzen 5 3600
+ RAM HyperX Fury 8GBx2 DDR4 3600MHz HX436C17FB3K2/16
+ SSD Samsung 970 PRO 512 GB MZ-V7P512BW
+ GPU ASUS geforce gtx 1650 super oc 4gb
+ PSU Corsair RM650x
Forgot to note:
Last chipset drives are installed, last Windows updates are installed, last stable bios is installed (2006).
No issues with CStates under the "official" 3200Mha memory speed
I solved it for now!
It was too low voltage.... apparently 3800 MHz requires 1.450v on the RAM in order to be stable on both my FlareX and trident Z neo kit..
I was too scared to push this high voltage but up to 1.5v is acceptable for 24/7 so im good for now.
But AFAIK, XMP specifies voltage, if this kit does not work at this frequency and 1.35V it should be not used in this hi-freq kit, or higher voltage must be programmed in XMP.
I guess it is the RAM vendor fail. They try raise RAM-IC as high as possible and sell the faulty sticks, isn't it?