Odd boot behavior on RVE not stable

[RAP66]

Hi all,

My first RVE using 1701 was very unreliable. It would randomly lock up; browsing the web, playing games etc. Just before it would lock all of the system fans would spin up and any inputs by keyboard/mouse would get very choppy. I've tried reinstalling the OS, updating drivers, resetting the bios to "normal settings" etc. Nothing would prevent the system from locking. So I figured that the board was bad so I purchased a second MB. System was OC'd to 4.4g using the 5-way optimization. I do not use the OC panel. The bios would give code 40 or 66.

The new MB works better, BUT this one I need to reboot 2-3x before I get to the desktop. I did not reinstall the OS (I probably should have). During the boot-up I get code 40 most of the time and get to the login screen. At this point I reboot using the OS and the bios will post AA and everything works as it should. Now this system does lock up sort of. Just like before the system gets choppy and sluggish but so far has not done a total lockup. I tell the pc to reboot and during the post I get code 96 but no logon screen. At this point IO need to hit the reset button a few times to get the system working and the bios will post code AA. Sometimes I need to hard start the system in order to get it to boot up. This MB is OC'd to 4.2 using the 5-way optimization. On t he plus side this MB does not lock up like the previous board while playing games.

I'm stumped! I have never had a system behave like this. What do you suggest? I would like to get this system reliable, stable and still be able to OC it to take advantage of the components I have.

Thanks

Rich

[Qwinn]

EDIT: Sorry, wrong thread.

[stumpy6317]

Hawkstorm wrote:
Firstly if you have the 3.1 board the first thing you should know is that they are different and each board has its own Bios revisions.

The highest revision for the 3.1 board is 1502.

Do not goto the wrong support page at Asus, goto the 3.1 support page and get drivers etc from there for the USB 3.1 board, DON'T flash the Bios with 1701 from the other board, its not meant for it if it is a 3.1 board and has 1701 on it then someone has flashed it to the wrong bios.

I think you have your facts wrong the latest is in fact 1701. I have the 3.1 version having originally bought previous version both I have installed 1701 (latest) bios with no problems. Here is a post about this question.
https://rog.asus.com/forum/showthread.php?75393-Replacement-board
Oh yeah there is this one.
https://rog.asus.com/forum/showthread.php?78196-Rampage-3-1-bios-or-rampage-bios&highlight=1701
The boards are the same just a newer revision. The bios are the same, the drivers are the same. Nothing is different apart from the 3.1 add-on card in the box. This card can also be bought for the original version.

[Hawkstorm]

Everyone has favourite tests me I prefer OCCT.

As for using other test you have to be careful depending on your overclock you don't damage you CPU by pulling too much wattage though it and generating too much heat.

Which is why I like OCCT you can leave it running and it has a Thermal cutout, it monitors temps accurately too.

[RAP66]

I just purchased this board and it was shipped with 1701 installed. I bought it thinking that the board was "updated" and would be more stable than the board I removed. This board is more stable and does not lock up as often.

That said. I have been trying to run the memtest program as described and I have not been able to get 1 complete run. I was going to start the program and let it work all day while I did stuff around my house but for some reason my system would reach a point then reboot!? I tried to sit and watch the computer to see where it bombs but I cannot tell you how "entertaining" that is.

So I kinda gave up on it.. I left the timings and voltages and continued to run my system as-is. So far it has only locked up once but I believe that is my fault for too much multitasking: browsing the web, forums, downloads & watching youtube.

I'll continue to try running some sort of memory test and hopefully everything will clear up.

Thanks for everyones input.

-Rich

[Chino]

Rebooting on its own probably points to an unstable memory overclock. I know how fun HCI Memtest can be especially for users who have insane amount of RAM. LOL

By the way, are you running it as I have detailed in my DDR4 guide?

As a sidenote, there is an even better alternative to HCI Memtest but you'd need a copy of Linux. The Google Stress App does an excellent job of isolating the memory from the rest of the system for some hardcore testing. So you can get away with running a 1 hour pass. 😛

[RAP66]

Chino wrote:
Rebooting on its own probably points to an unstable memory overclock. I know how fun HCI Memtest can be especially for users who have insane amount of RAM. LOL

By the way, are you running it as I have detailed in my DDR4 guide?

As a sidenote, there is an even better alternative to HCI Memtest but you'd need a copy of Linux. The Google Stress App does an excellent job of isolating the memory from the rest of the system for some hardcore testing. So you can get away with running a 1 hour pass. 😛

DDR4 guide?? I did not know about that... Can you provide a link?

-Rich

[Chino]

RAP66 wrote:
DDR4 guide?? I did not know about that... Can you provide a link?

-Rich

It's in my signature. 😛 I have a section on HCI Memtest.

[vmanuelgm]

A good way to test cpu and memory stability is 30 passes of Intel Burn Test in very high.

The cpu gets hot (but without avx2 which directly burns), but you will know if it is stable with memory settings.

[Qwinn]

Actually I read Raja say in a post somewhere that the problem of AVX instructions drawing too much power was a known problem in Haswell chips which was corrected in Haswell-E. I know I haven't duplicated the problem on my 5930k, though I've also not been brave enough to try Prime95 yet.

[Hawkstorm]

Don't run small fft's on it if you do.

OCCT has AVX capable LINPAC and it also comes with the thermal cutout I'd stick with that, Prime95 doesn't so don't leave it unattended and monitor you temps and powerdraw closely if you do use it but I'd say steer clear of it.

And the vcore you will need to be stable under Prime95 will be quite a bit higher and generate quite a bit more heat too.