Boot problem, don't know what is wrong

[J0kador]

My PC boot problem

 

The Problem

About 75% of the time, when my PC boots, there is a problem. Everything is slow, all the animations stutter, the sound stutters, the mouse stutters and has tremendous lag, it's like the computer doesn't have enough processing power to even use windows.

 

I have put an extensive description of the problem and when it occurs neatly in a document here.

 

Thanks in advance,

 

J0kador

[tbroadwater2010]

My PC boot problem

 

The Problem

About 75% of the time, when my PC boots, there is a problem. Everything is slow, all the animations stutter, the sound stutters, the mouse stutters and has tremendous lag, it's like the computer doesn't have enough processing power to even use windows.

 

I have put an extensive description of the problem and when it occurs neatly in a document here.

 

Thanks in advance,

 

J0kador

So to be clear, its not an issue of the BIOS logo its an issue when windows loads itself?

[Naeaes]

This sounds awfully lot like a RAM compatibility and/or timing issue to me. What settings have you input for the RAM in the BIOS? Did you verify that they stuck using CPU-Z or similar?

 

Found your RAM SKU# in the PCPP post you linked. The correct timings for GB38GB1600C9DC look to be 9-9-9-28 and the speed 1600MHz and apparently 1.60 Volts. None of this is standard. If you haven't forced those settings on, they'll definitely be off. Also, at least according to the QVL at Gigabyte website your kit isn't a known good one. Doesn't bode well in my books...

[J0kador]

This sounds awfully lot like a RAM compatibility and/or timing issue to me. What settings have you input for the RAM in the BIOS? Did you verify that they stuck using CPU-Z or similar?

 

Found your RAM SKU# in the PCPP post you linked. The correct timings for GB38GB1600C9DC look to be 9-9-9-28 and the speed 1600MHz and apparently 1.60 Volts. None of this is standard. If you haven't forced those settings on, they'll definitely be off. Also, at least according to the QVL at Gigabyte website your kit isn't a known good one. Doesn't bode well in my books...

I have made 2 screenshots of my cpuz readings, one of them is with the (only) XMP profile enabled, and one without (as given in the window's title bar). I did not force the timings or voltage in the bios.

[J0kador]

So to be clear, its not an issue of the BIOS logo its an issue when windows loads itself?

Well by talking about the boot logo, i tried to say that the issue is apparent before even windows loads.

[Naeaes]

I have made 2 screenshots of my cpuz readings, one of them is with the (only) XMP profile enabled, and one without (as given in the window's title bar). I did not force the timings or voltage in the bios.

So it's almost spot on. Usually I'd say go with the XMP but 1T is kind of tight for 1600MHz. Go check the voltage in BIOS too. It's not visible in CPU-Z.

[J0kador]

So it's almost spot on. Usually I'd say go with the XMP but 1T is kind of tight for 1600MHz. Go check the voltage in BIOS too. It's not visible in CPU-Z.

Okay, i went in to the bios, turned off the xmp profile, then set the voltage, frequency, and timings according to your post earlier from the GeIL website. And out of the 3 times booting 2 we're bad. I will soon try a different set of RAM sticks. I will post my findings in this thread. So you can follow it if you wish.

[J0kador]

Today I've installed a different set of RAM (2x4GB CMV4GX3M1A1333C9), and so far I've done 4 reboots. With the other RAM it would have gone wrong by now, but I'm not taking any chances. I am going to try it out for at least a week before I call the problem (and the thread) solved.

 

I will update this thread with further developments.

[J0kador]

When I booted this morning it went wrong again. I am not sure if it's the RAM that's wrong now. Although neither my original set, nor the second set are on the official supported list. The problem is that I do not own a set of RAM on that list to experiment with.

 

I may be wrong, but doesn't the fact that 2 different sets of RAM display the exact same faulty behavior tell me that the problem doesn't have to do with the RAM?

[Naeaes]

When I booted this morning it went wrong again. I am not sure if it's the RAM that's wrong now. Although neither my original set, nor the second set are on the official supported list. The problem is that I do not own a set of RAM on that list to experiment with.

 

I may be wrong, but doesn't the fact that 2 different sets of RAM display the exact same faulty behavior tell me that the problem doesn't have to do with the RAM?

I wouldn't call it certain before testing a kit from a QVL. This Corsair kit is very basic JEDEC kit. And the other one was X.M.P. Both are kind of best case scenarios on my book. If anything should work then these should. But for whatever reason they're not listed. It sucks that no-one keeps a list of bad RAM and the reasons they don't work. But it wouldn't be very good marketing, I guess. Anyway, if it's not the fact that they're incompatible, then what is it? A bad CPU?

 

Testing the RAM won't cost you. From my experience, if they don't present errors in Memtest86+ after 10+ hours, they're not broken but might still be incompatible. Even a single error indicates a broken RAM if it repeats (usually every time the same region of RAM is tested). If not, it might be a glitch. And all errors all the time might indicate a bad CPU or motherboard or RAM. Incompatible RAM can work fine in it or it can crash the program so that there's no text but only the colored boxes.

Memtest86+ doesn't stress the system at all so stuff like power or heat issues won't present problems in it. Also the random nature or RAM makes it so that one or two passes of the test don't mean anything. It has to pass several times. I always had time constraints at work so I ran it over night. My threshold was either ten passes or however many passes makes in ten to twelve hours.

The original Memtest86 is also rearing it's head. They've been out of the picture for several years in between so I don't use it as often. But it is a similar test.