Dual Core with Windows 7 sometimes starts if... sometimes not

[radge]

Hello, I have a Dual Core from a family member... with Windows 7 -32 bits, 4Gb of ram... Asrock motherboard I think... it's a PC that works great for what it's used for, Outlook and browsing, the problem is that sometimes when it starts up after the BIOS screen it gets stuck with a black screen and the "cursor" blinking and doesn't go any further or it shows the black background with the Windows 7 logo and doesn't go any further.

I don't think it's the Windows or the hard drive because I've formatted it like 20 times and I'm not exaggerating...
And I've tried it with several sata2 hard drives I have of different capacities... so I would say it's the "ram" or the motherboard... if it's a hardware problem.

Right now it has Windows 7 "stripped down" with Outlook, antivirus and not much else.
Any ideas what I can do? I'm really annoyed about this... because curiously, whenever I go, it turns on the first time... (sometimes it happens at first thing in the morning when it's cold, but other times if you shut it down at noon and open it in the afternoon... that's when it fails)

P.D.: it's connected to a UPS... I don't think it's a voltage problem... could it be a bad capacitor on the motherboard?

Thanks

[whoololon]

Man, for starters you can run memtest.
What strikes me is this @radge:

I don't think it's the windows or the hard drive because I've formatted it like 20 times and I'm not exaggerating…

Because if it were the motherboard or the memory it would fail before completing the installation, and even more so… did it work right after formatting?
In any case, start by checking the memory to rule it out.

[radge]

@whoololon:

Man, for starters you can run the memtest.
What strikes me is this: Because if it were a motherboard or memory issue, it would fail before completing the installation, and even more so… did it work right after formatting?
In any case, start by checking the memory to rule it out.

The thing is that every time I fix it, I take the PC home, transfer the data, format another hard drive... install Windows... all its updates... restart.. install Office and transfer its data.
This, with the nonsense, is at least 5 or 6 hours during which I restart the PC a couple of times, then I give it to him and on the 2nd or 3rd day... it starts doing weird things.

Also, at his house I have it plugged into my old UPS, I lent it to him to rule out if it was a spike issue or something like that... in theory the UPS regulates this...

regards

[whoololon]

I've had that problem, in fact, I'm currently using a second-hand Acer laptop until I can buy a new one, and it's happened to me more than once. With WXP, because for 32 bits and 3GB of RAM it works great, but more or less the same, C2duo at 1.6 and it would get stuck in the middle of the boot... and it was a problem with the hard drive (I have 2 of 160GB installed).
In fact, I also suspected the memory, but since the memory check takes longer, I started with the hard drive, whose check takes less time. And it found damaged sectors that, if the OS doesn't detect them or they appear after installation, it will install and operate on them with the consequent loss of data and stability.
@radge:

And I've tried it with several sata2 hard drives that I have of different capacities...

But as you indicate that you've installed several drives and the same thing happens with all of them, I reiterate my previous message and I would recommend that you run a memory check.

[MaxLG]

DO NOT discard the hard drive.
I had the same problem with a laptop that I still don't know how it works.

What happens to it (in my laptop) is that sometimes the head doesn't touch the disk, according to what I was told at a computer repair shop next to me that repairs PCs. The solution I found for my laptop is to give it a few taps where the hard drive is, and it unlocks for me. I don't know if it's the same for your PC, but I had the same problem as you but with Windows XP, and it would be a bit difficult to give a few taps to the hard drive lol.

[Fassou]

I'll tell you a curiosity, in case it serves you for new tests.

I had a case where a mouse was detected as a boot drive, and if you left it connected before booting, many computers would get stuck waiting, until they froze.

In some, disabling USB boot options would fix it, but in others, if you didn't leave the mouse disconnected when turning on or restarting, they would get stuck in BOOT.

Best regards!

[MaxLG]

@Fassou:

I'll tell you something interesting, in case it's useful for new tests.

I had a case where a mouse was detected as a boot drive, and if you left it connected before booting, many computers would get stuck waiting.

In some, disabling USB boot options would fix it, but in others, if you didn't leave the mouse disconnected when turning on or restarting, they would get stuck in BOOT.

Cheers!

Wow, that's a good one

[Kernel1.0]

@MaxLG:

Do not discard the hard drive.
I had the same problem with a laptop that I still do not know how it works.

What happens to it (in my laptop) is that sometimes the head does not touch the disk according to what I was told at a computer repair shop next to me where they repair PCs. The solution I found for my laptop is to give it a few taps where the hard drive is, and it unlocks for me. I do not know if it is the same for your PC, but I had the same problem as you but with Windows XP, and it will be a bit difficult to give a few taps to the hard drive lol.

Change the store…:facepalm:

[Franziskaner]

that with a hammer will surely unscrew the heads ;D;D the shame will be the tower type "golf ball"
I hope it's not pretty, although in the worst case it will be an original MOD, the title could be : niall mackenzie. famous when he raced in 500 for "dented the exhaust pipes with a hammer, because he said that this way the bike performed better "
(sarcasm)
regards

[radge]

Well for now the computer has been running a memtest for 12 hours… now I understand why I had never done it before and a couple of errors have come up… tomorrow I will show screenshots

regards

[MaxLG]

@Kernel1.0:

Cambia de tienda…:facepalm:

Sisi,cambia de tienda pero lo de los golpecitos funciona xD Y por logica en mi caso es por el disco duro

[Sylver]

@MaxLG:

Yes, change the store but the tapping thing works xD And logically in my case it's because of the hard drive

You'll tell us how long it lasted... :sisi:

[radge]

For now the memtest… not finished goes like this... what do you think?

[whoololon]

I think you already figured it out: that module is broken.

[radge]

@whoololon:

I think you already know: that module is broken.

I'm saying this from memory because I don't have the PC in front of me, but it has 1 module of 2gb and then 2 in dual channel of 1gb... how do I know which one of the 3 is it? I don't see it in the memtest... or when the test is finished will it tell me or something?

regards

[whoololon]

It can't be done. (At least I don't know how.)
In fact, memtest is left running until it finds an error. In fact, every time I've done one to the end it's been because I thought the RAM was faulty and after the test it turned out to be fine.
The procedure when an error is found is simple, remove the modules and test them one by one. The moment it fails, you stop the program, turn it off, unplug it, discharge the static and replace the module. You turn it back on and run memtest again, like that with all of them.
Yes, it's a pain. ;D

[MaxLG]

@Sylver:

Ya nos contarás cuánto te ha durado… :sisi:

De momento lleva unos 6 meses

Lo de los modulos, no van numerados los zocalos?(creo que se llama asi :P)

[garfield]

You can try one thing, because the memory at the price it is… increase the voltage, and if I remember correctly leave it at test 5, I think the sequence was 1 - 3 - 5 or something like that. Krampak will surely remember. It was to get the lazy ddr400 to go higher, or at least to work at lower voltage. At first the errors that the memories gave were brutal, and when they reached less than ¿100? you did the same thing again but increased fsb. Many managed 20mhz more in our ddr.
What you do have to keep in mind is that if you increase voltage, as the memories are going to be working at full capacity put a fan on top of the memories.
Let's see if I find the message.

I found it:
Burn-in memories

Don't take it to the letter. If the ddr2 works at 1.8, don't even think about putting it at 2.5v, put it at 1.9v for example and do the burn-in. DO NOT skip any recommendation, and if you don't understand anything ask.
And if the module ends up being ruined, as it was already broken nothing will happen. That is, under your responsibility. I know how far I can go. If you don't feel capable don't do it.

[radge]

@garfield:

You can try one thing, because memory at the price it is… increase the voltage, and if I remember correctly leave it at test 5, I think the sequence was 1 - 3 - 5 or something like that. Krampak will surely remember. It was to get the lazy ddr400 to go higher, or at least to work at less voltage. At first the errors that the memories gave were brutal, and when they reached less than ¿100? you did the same thing but increased fsb. Many managed 20mhz more in our ddr.
What you have to keep in mind is that if you increase voltage, as the memories will be working at full capacity put a fan on top of the memories.
Let's see if I find the message.

I found it:
Burn-in memories

Don't take it to the letter. If the ddr2 works at 1.8, don't you dare put it at 2.5v, put it at 1.9v for example and do the burn-in. DO NOT skip any recommendation, and if you don't understand anything ask.
And if the module ends up breaking, as it was already broken nothing will happen. That is, under your responsibility. I know how far I can go. If you don't feel capable don't do it.

Thank you very much… but to do this I prefer to see if the 2gb module is cracked take it off and leave it running with only 2gb more than enough.. or if one of the 2 1gb modules has failed take those off and leave the 2gb...

Anyway the memtest has been going for 30H this never ends or what????
Attached screenshot

regards

[Kernel1.0]

You have already been told that it NEVER ends, it goes on indefinitely.