Overclock.net › Forums › Intel › Intel Motherboards › ASUS North America - ASUS Z170 Motherboards Q&A Thread
New Posts  All Forums:Forum Nav:

ASUS North America - ASUS Z170 Motherboards Q&A Thread - Page 330

post #3291 of 5737
Quote:
Originally Posted by pegnose View Post

So I had these hard locks, randomly it seems, but rather during gaming than idle. I remembered that it wasn't there in the beginning, when I hadn't used the ASM1061 for software raid1 of my data partition. So I moved the disks to the Intel controller and did a resync of the raid array. Because I could provoke a hard lock during resync on the ASM1061 before with load (within 2-5h), I retried this:



This is:
- 3 threads of Prime95 800k (memory) with 7 GB tested (not in-place)
- 3 threads of HCI memtest (remaining memory)
- Furmark with dynamic background, post processing, and 8x AA on 1280x720 windowed
- 1 thread CPU burner (comes with Furmark)
- and, of course, the raid resync

So I had high load on CPU (100%), Ram (93%), disk (30% read on disk 1; 60-70% write on disk 2 of the raid array). I could do this for 7h straight, until the resync was done. So I finally was happy: the problem was the darn ASM1061, I would be good now.

Then I rebooted, logged in, started Firefox (very quickly, before waiting until everything was up), and my system froze. Q-Code "04". Never had that one before.

If I could copy text from the manual, I would tell you that it says "PCH initialization before microcode loading". But that applies only during boot, no?

The good thing is: the array is resyncing again. So I can retry with load. Haha.


EDIT: After that I used the tool "wdidle3" to disable the intellipark feature of the 2 Sata HDDs. After that on the reboot at least one the the WD Green wouldn't start. You know the rhythmic "clack, clack"? I had that before with one drive and I had RMAed it because it threw a read error on SMART test. But this one is either ok or new.

Could this be PSU having not enough power to operate 2 WD Green 3TB? You see my other specs in my signature.


If that screenshot is anything to go by it looks like you're testing your page file rather than your memory.


HCI needs to be configured correctly if ran from within Windows in order for it to be a worthwhile test of stability. Using task manager you can allocate enough, whilst leaving the OS enough to breathe. One instance needs to be allocated per thread, with the desired amount divided between each instance of HCI. Also best to not run other tests whilst the instances are running as the memory is obviously being allocated elsewhere, and not tested accordingly, including Prime.
Edited by Silent Scone - 1/24/16 at 10:22am
post #3292 of 5737
Quote:
Originally Posted by Silent Scone View Post

If that screenshot is anything to go by it looks like you're testing your page file rather than your memory.


HCI needs to be configured correctly if ran from within Windows in order for it to be a worthwhile test of stability. Using task manager you can allocate enough, whilst leaving the OS enough to breathe. One instance needs to be allocated per thread, with the desired amount divided between each instance of HCI.

Sorry, you either didn't read or look carefully enough. It is NOT about Ram testing. Also, my system drive (with the pagefile) showed little to no activity. So please, either read what I have written, or don't.

In short: I just meant to stress my system during the Raid rebuild.
post #3293 of 5737
Quote:
Originally Posted by pegnose View Post

Sorry, you either didn't read or look carefully enough. It is NOT about Ram testing. Also, my system drive (with the pagefile) showed little to no activity. So please, either read what I have written, or don't.

I looked well enough to know what you were running wasn't really going to isolate any problem in particular, but I was trying to be nice about it.
post #3294 of 5737
Quote:
Originally Posted by pegnose View Post

Sorry, you either didn't read or look carefully enough. It is NOT about Ram testing. Also, my system drive (with the pagefile) showed little to no activity. So please, either read what I have written, or don't.
Quote:
Originally Posted by Silent Scone View Post



I looked well enough to know what you were running wasn't really going to isolate any problem in particular, but I was trying to be nice about it.

a) You were not very nice.
b) I have written that I was able to freeze my system more or less reliably before by putting load to it during raid rebuild. And that was what I did.

But it doesn't matter, because it wasn't working. The more important question is: it really seems that my PSU has issues with my WD eco drives (5400 rpm). I only can't really understand why that is. It can deal with high load on CPU, mobo, and GPU. I can quench up to 450 W out it for hours, it is a 700 W PSU. But it has problems starting my two Sata HDDs? How so? Only, there is no other good explanation, unless my repeated freezes wrecked two brand new drives in a couple of months, which I can't really believe either.

EDIT: I am sorry if I come across impolilte. But this really gets to me and I can't figure it out. And that is hardly ever seen in computer things.
Edited by pegnose - 1/24/16 at 10:32am
post #3295 of 5737
Quote:
Originally Posted by pegnose View Post


a) You were not very nice.
b) I have written that I was able to freeze my system more or less reliably before by putting load to it during raid rebuild. And that was what I did.

But it doesn't matter, because it wasn't working. The more important question is: it really seems that my PSU has issues with my WD eco drives (5400 rpm). I only can't really understand why that is. It can deal with high load on CPU, mobo, and GPU. I can quench up to 450 W out it for hours, it is a 700 W PSU. But it has problems starting my two Sata HDDs? How so? Only, there is no other good explanation, unless my repeated freezes wrecked two brand new drives in a couple of months, which I can't really believe either.

EDIT: I am sorry if I come across impolilte. But this really gets to me and I can't figure it out. And that is hardly ever seen in computer things.

Hello

You would do well to listen to what @Silent Scone and a handful of others here have to offer when seeking advice. Something to keep in mind is their systems are stable with most having an impressive overclock. On the other hand you are here looking for help because you system is not stable.
post #3296 of 5737
Quote:
Originally Posted by Praz View Post

Hello

You would do well to listen to what @Silent Scone and a handful of others here have to offer when seeking advice. Something to keep in mind is their systems are stable with most having an impressive overclock. On the other hand you are here looking for help because you system is not stable.

Praz, you are totally right. I am looking for help. And I am very happy if somebody actually decides to help me. But what I don't like is if somebody just disregards my effort and thinking so blatantly. I explained very well why I did what I did. The reason was that I suspected that the ASMedia Sata chip somehow failed on rebuild during general system load. Why? I don't really know. I am kinda helpless at the moment.

Then what I would very well be able to accept is if somebody tells me that my hypothesis is crap because he or she has more knowledge about the underlying hardware. That would be totally fine by me. I would really like to learn more about my board!

But if someone tells my that I can't find out anything by what I am doing, when getting a differenciated diagnostically valuable outcome was explicitly NOT the goal... I wrote it above.

I just wanted to see if my system could do a rebuild under general, undifferenciated load. Because it couldn't in the past. And that was what I tested.


EDIT: Apologies again, @Silent Scone, I wasn't very friendly. But if you are honest to yourself, you weren't, either.
Edited by pegnose - 1/24/16 at 11:14am
post #3297 of 5737
Quote:
Originally Posted by pegnose View Post


a) You were not very nice.
b) I have written that I was able to freeze my system more or less reliably before by putting load to it during raid rebuild. And that was what I did.

But it doesn't matter, because it wasn't working. The more important question is: it really seems that my PSU has issues with my WD eco drives (5400 rpm). I only can't really understand why that is. It can deal with high load on CPU, mobo, and GPU. I can quench up to 450 W out it for hours, it is a 700 W PSU. But it has problems starting my two Sata HDDs? How so? Only, there is no other good explanation, unless my repeated freezes wrecked two brand new drives in a couple of months, which I can't really believe either.

EDIT: I am sorry if I come across impolilte. But this really gets to me and I can't figure it out. And that is hardly ever seen in computer things.
Quote:
Originally Posted by pegnose View Post

Praz, you are totally right. I am looking for help. And I am very happy if somebody actually decides to help me. But what I don't like is if somebody just disregards my effort and thinking so blatantly. I explained very well why I did what I did. The reason was that I suspected that the ASMedia Sata chip somehow failed on rebuild during general system load. Why? I don't really know. I am kinda helpless at the moment.

Then what I would very well be able to accept is if somebody tells me that my hypothesis is crap because he or she has more knowledge about the underlying hardware. That would be totally fine by me. I would really like to learn more about my board!

But if someone tells my that I can't find out anything by what I am doing, when getting a differenciated diagnostically valuable outcome was explicitly NOT the goal... I wrote it above.

I just wanted to see if my system could do a rebuild under general, undifferenciated load. Because it couldn't in the past. And that was what I tested.


EDIT: Apologies again, @Silent Scone, I wasn't very friendly. But if you are honest to yourself, you weren't, either.


I didn't say anything that could have been deemed offensive. You were loading HCI incorrectly so therefore it had no purpose being open. I will reply to you one last time in attempt to stop your from going off-track.

Retest the memory within Windows with HCI with the correct configuration. Your previous run going by your post this morning, it was not clear if the test had completed successfully, or what coverage the system had covered. Saturating the system with multiple tests is more likely to produce errors that may not even be related to stability. You said you would be ok with someone telling you that you are doing it wrong, and you are.
post #3298 of 5737
Quote:
Originally Posted by Silent Scone View Post


I didn't say anything that could have been deemed offensive. You were loading HCI incorrectly so therefore it had no purpose being open. I will reply to you one last time in attempt to stop your from going off-track.

Retest the memory within Windows with HCI with the correct configuration. Your previous run going by your post this morning, it was not clear if the test had completed successfully, or what coverage the system had covered. Saturating the system with multiple tests is more likely to produce errors that may not even be related to stability. You said you would be ok with someone telling you that you are doing it wrong, and you are.

Thank you for bearing with me, Silent Scone. My nerves are worn off, I am really sorry.

Let me report, that I had a) run memtest86 successfully for a whole day. b) HCIs memtest with a coverage of 650% without error or hard lock. For that I ended all other programs, divided the remaining memory (~14GB) by 8 and started this number of threads in the desktop version. In addition, I achieved a coverage of ~110% in the boot CD version. Only this low, because this version is really slow.

Long story short: I really believe that my memory is stable. Or did I do another mistake?


EDIT: The thing is: with my unconventional setup I really wanted (!) to produce an error. But my system was stable for 7h. Until after the following reboot.
post #3299 of 5737
Quote:
Originally Posted by pegnose View Post

Thank you for bearing with me, Silent Scone. My nerves are worn off, I am really sorry.

Let me report, that I had a) run memtest86 successfully for a whole day. b) HCIs memtest with a coverage of 650% without error or hard lock. For that I ended all other programs, divided the remaining memory (~14GB) by 8 and started this number of threads in the desktop version. In addition, I achieved a coverage of ~110% in the boot CD version. Only this low, because this version is really slow.

Long story short: I really believe that my memory is stable. Or did I do another mistake?


EDIT: The thing is: with my unconventional setup I really wanted (!) to produce an error. But my system was stable for 7h. Until after the following reboot.


As Praz has already mentioned, and from what you are telling us at this point it's best to re-test at stock and see if you are able to replicate the issue
post #3300 of 5737
Quote:
Originally Posted by Silent Scone View Post

As Praz has already mentioned, and from what you are telling us at this point it's best to re-test at stock and see if you are able to replicate the issue

Replicate what exactly?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel Motherboards
Overclock.net › Forums › Intel › Intel Motherboards › ASUS North America - ASUS Z170 Motherboards Q&A Thread