Jump to content
  • Announcements

    • Allan

      Forum Rules   04/13/2018

      Hello folks! As some things are being fixed, we'll keep you updated. Per hour the Forum Rules don't have a dedicated "Tab", so here is the place that we have our Rules back. New Users Lounge > [READ] - InsanelyMac Forum Rules - The InsanelyMac Staff Team. 
Sign in to follow this  
shiftysamurai

[Build Log] Shifty G5/mATX Conversion

136 posts in this topic

Recommended Posts

Ok, I've tried with the following pieces all taken out, one at a time. I've tried every combination that I can think of.

 

Memory, tested each stick by itself (memtest found no errors in memory)

2 different motherboards

3 video options (7850, 5750, Intel HD 4000)

2 HDDS

1 SSD

 

The only thing I haven't tried is a new CPU and/or a PSU. But neither of those make that much sense to me considering the symptoms.

 

I run games under full settings, I run FurMark on high, I run Prime95, and everything works as it should.

 

But I run a random racing game and/or some Media Center software (that will run on an Xbox!) and I get lockups and crashes. Just for giggles, I ran this all on my old PC, which only really tells me that, if it's hardware, and I think it is, then one of the following is to blame.

 

CPU (I guess it's possible, but it works fine under high load in Prime95)

Motherboard (two in a row with the same symptoms?)

RAM (2 faulty sticks that pass memtest alone and together?)

PSU (Seems like the only piece I haven't booted without)

 

So... Maybe I'll try to get the IT department to look at it at my work. I hate saying this, but it's kind of depressing even looking at the thing right now. I'm $200 over budget, got the damned Intel stock cooler on the cheap board and it doesn't want to come off... It's just been a nightmare of a week.

 

I've spent at least 60 hours doing everything I could after work for the last week (work - PC - bed - repeat) and I still have no idea what it could be.

 

Just want to be done with it and have a stable PC. Maybe spend some time with the wife and dog.

 

I've had build problems before, but never one that has been this hard to track down.

 

PS: PC also locks on updates sometimes.

Share this post


Link to post
Share on other sites

when you say you ran all of this on your old PC, you mean the software? have you tried using some of the new components in your old rig? like new ram and gpu in the old rig or new hdd/ssd ram in the old rig? something like that? And I know PSUs don't seem to be the issue, but just remember that if something is out of tolerance by just enough, it can cause wonky behavior on your machine. try the PSU from the old rig and see what you get.

Share this post


Link to post
Share on other sites

I guess you already checked for bent CPU pins?

 

My setup just randomnly froze when I had 4x8GB sticks installed. Taking out 2 of them made it work again. So I started to check and found out that one of my pins in the area of the DDR3-controller was bent. The whole machine worked completely fine with just 2x8GB installed. I didn't manage to bend the pin back, so I'm just going with "only" 16GB. And it's plenty anyways...

Share this post


Link to post
Share on other sites

Thanks for the advice, ElAwesome. I'll take a look at the boards in the morning. I don't know how likely it is that both of these boards is bad, but it's worth taking a look. I was sure it was the motherboard, but now I'm thinking RAM or PSU.

 

I also thought it might just be software, but I got a lockup while in the BIOS today and Memtest gave me an "Unexpected Halt" which doesn't always point to RAM as the problem (RAM errors usually look a bit different).

 

Once again, I am unsure of the cause of my problem.. :/

Share this post


Link to post
Share on other sites

Have you tried updating the BIOS?

Maybe something got corrupted there (although it is very unlikely that this happened twice)

 

Download on another PC the latest BIOS for your board and then flash it, it could solve your problems. If Memtest gives you an error, it means that there is something related with RAM. If you say your RAM works, then you need to check your CPUs DDR3-controller and the chipset. You got another board, errors occured on both boards, that means you can probably exclude bent pins and chipset failure. I'd rather check your CPU if you see any physical damage and take it to a store and ask if you could test it on a board there. I suspect your CPU is somewhat bad.

RAM usually never breaks if you don't overclock it, and do check your voltages in BIOS. If they are all fine, you PSU should be doing fine. But if you say that you were able to stress test the computer, I'd say your PSU works properly. Otherwise you wouldn't have got that far.

 

I hope I can help you!

 

But as said before, the firs thing I would do is upgrade to the latest BIOS.

Share this post


Link to post
Share on other sites

Hey ElAwesome, I'm running the latest versions of each board's BIOS.

 

I did run Intel's Processor Diagnostic (64-bit) on the "cheaper" board (not my ASRock Z77pro4-m) as the damned Intel heatsink and fan do not want to allow themselves to be removed....

 

 

 

--- IPDT64 - rev 1.24.0.0 ---

 

--- Start Time: 03/31/2013 11:59:08---

 

--- Skipping Config ---

 

--- Reading CPU Manufacturer ---

Expected --> GenuineIntel

Detected --> GenuineIntel

Found --- Genuine Intel Processor ---

--- Temperature Test ---

Temperature Test Passed!!!

 

Temperature = 69 degrees C below maximum.

 

--- Reading Brand String ---

Detected Brand String:

Intel Core i5-3570K 3.40GHz

Brand String Test Passed!!!

 

--- Reading CPU Frequency ---

Expected CPU Frequency is --> 3.40

 

Detected CPU Frequency is --> 3.40003

CPU Frequency Test Passed!!!

 

--- FSB NOT Supported on this Processor ---

 

--- Running Base Clock test ---

 

Detected Base Clock --> 100

Base Clock test Pass ---

--- Running Floating Point test ---

 

Million Floating Points per Second, MFLOPS --> 446.4

Floating Point Test Pass ---

--- Running Prime Number Generation Test ---

 

Operation Per Second--> 5.98121e+006

Prime Number Generation Test Pass ---

--- Reading Cache Size ---

- Detected L1 Data Cache Size --> 4 x 32

- Detected L1 Inst Cache Size --> 4 x 32

- Detected L2 Cache Size --> 1024

- Detected L3 Cache Size --> 6144

 

Cache Size Test Passed!!!

 

--- Determining MMX - SSE capabilities ---

 

--- CPU FEATURES DETECTION FOR ---

--- MMX SSE ---

MMX - MMX Supported --> Yes

SSE - SSE Supported --> Yes

SSE2 - SSE2 Supported --> Yes

SSE3 - SSE3 Supported --> Yes

SSSE3 - SSSE3 Supported --> Yes

SSE4.1 - SSE4.1 Supported --> Yes

SSE4.2 - SSE4.2 Supported --> Yes

--- MMX SSE - capabilities check complete ---

MMX Test Result --- PASS

SSE Test Result --- PASS

SSE2 Test Result --- PASS

SSE3 Test Result --- PASS

SSSE3 Test Result --- PASS

SSE4.1 Test Result --- PASS

SSE4.2 Test Result --- PASS

MMX SSE Testing Passed !!

--- Determining AVX AES PCLMULQDQ capabilities ---

 

--- CPU FEATURES DETECTION FOR ---

--- AVX/AES/PCLMULQDQ ---

AVX - Advanced Vector Extensions Supported --> Yes

AVX OS Support - AVX Operating System Supported --> Yes

AES - Advanced Encryption Standard Supported --> Yes

PCLMULQDQ - Polys Carry-Less Multiply Supported --> Yes

--- AVX AES PCLMULQDQ capabilities check complete ---

AVX Compare Test Result --- PASS

AES Test Result --- PASS

PCLMULQDQ Test Result --- PASS

AVX AES PCLMULQDQ Testing Passed !!

--- Reading Memory Size ---

 

Detected Memory Size is --> 8.00GB

 

--- Integrated Memory Controller Stress Test ---

 

--- Integrated Memory Controller Stress Test Pass!!! ---

 

Integrated Memory Controller Test Pass!!!

 

Chipset : Intel® B75 Express Chipset

Chipset Stepping : C1

No device connected to PCI-to-PCI Bridge Controller

2 device connected to PCI Express Port 0

1 : AMD Radeon HD 7800 Series

2 : High Definition Audio Controller

No device connected to PCI Express Port 1

1 device connected to PCI Express Port 6

1 : Realtek PCIe GBE Family Controller

1 device connected to SATA Controller 1

1 : KINGSTON SV300S37A60G ATA Device

2 device connected to SATA Controller 2

1 : ATAPI iHAS124 W ATA Device

2 : ST31500341AS ATA Device

3USB devices detected

1 : Xbox 360 Wireless Receiver for Windows

2 : USB Composite Device

3 : USB Mass Storage Device

--- Platform Controller Hub Test Completed!!! ---

 

..

..Query 2..

..

--- Querying for Intel® Integrated Graphics Device (IGD) ---

..Did not detect Intel® Integrated Graphics Device (IGD).

..This could be that the test platform is using a discrete graphics card.

..To test IGD, verify that there is an onboard video connector,

..shutdown the test platform, remove the graphics card, and

..move video cable from graphics card to onboard video connector

..Skipping Intel® Integrated Graphics Device Presence Test..

..

--- CPU Load ---

 

--- Load Level = 8

 

CPU Load Passed!!!

 

--- Temperature Test ---

Temperature Test Passed!!!

 

Temperature = 45 degrees C below maximum.

 

--- Test End Time: 03/31/2013 12:03:08---

 

 

 

 

Anyhow, CPU passed all tests, and my RAM passed Intel's (rather weak) memory/CPU stess test. I will rerun Memtest86+ to see if I can replicate the error, but it's not looking like a "Bad Memory" error.

 

Will edit to add results of Memtest run.

Share this post


Link to post
Share on other sites

Removing the stock heatsink is actually simple - just make sure you turn the screws against the direction indicated and then carefully lift it. Lift it straight upwards, otherwise it won't come off. But I guess that isn't any news to you, I thought I'm gonna share that just in case..

 

 

Make sure Memtest uses every bit of available RAM!

Also go ahead and to the same test of the AsRock board, to make sure the chipset and the socket are working properly.

 

 

Share this post


Link to post
Share on other sites

Removing the stock heat-sink is actually simple - just make sure you turn the screws against the direction indicated and then carefully lift it. Lift it straight upwards, otherwise it won't come off. But I guess that isn't any news to you, I thought I'm gonna share that just in case..

 

The stock Intel cooler sucks for removal. I broke mine. Google it - about half of them break when they come off. I ended up breaking two of the four spring loaded posts and had to use a knife and pliers to get the damn thing out. Thank the 8 pound, 6 ounce tiny infant Jesus that I had a replacement CPU coming.

Share this post


Link to post
Share on other sites

I got it out once, this next time is probably going to be with wire cutters. 2 of the white posts won't come out...

 

Also: 14 hour Memtest reports no errors. Maybe it's bad form, but I'm just wanting something to be broken in a bit more of an obvious way than this. It's like chasing a memory of a perfume trail from a ghost.

Share this post


Link to post
Share on other sites

If memtest tells you after 14h that everything is fine, memory is probably fine.

 

This is some really strange stuff. I'd somehow suspect the CPU... this is the only thing that you didn't test so far. Maybe you can ask for a replacement CPU?

Share this post


Link to post
Share on other sites

Hmmm I really don't know at this point...

I'd probably get in contact with Intel's support, because the CPU is the only thing you didn't replace so far. RAM works, you've got two mobos working, voltages of the PSU are fine... this is really strange. Do you get any kind of error message when the PC locks up on Windows?

Share this post


Link to post
Share on other sites

I just get the random (and vague):

 

Windows Kernel event ID 41 error "The system has rebooted without cleanly shutting down first"

 

With no more information in the log (that I can see).

Share this post


Link to post
Share on other sites

Okay, so I've been googling this error now...

 

It means that your kernel didn't get enough power. Now that doesen't necessarily lead to a broken PSU, it can also be because overheating.

You switched the cooler quite often, didn't you? That means that the thermal paste probably isn't working good anymore.

Do some prime95 and play some intensive games while keeping an eye on the temperaure.

 

If it stays normal, you need to check your PSU. Use this PSU on your old PC or the one from your old PC on the new one - either way check if it works now.

 

http://support.microsoft.com/kb/2028504

Share this post


Link to post
Share on other sites
You switched the cooler quite often, didn't you? That means that the thermal paste probably isn't working good anymore.

 

I've switched it about 5 times, but I never use the same thermal compound... I easily could have messed it up more than once, but I would like to think that at least ONE out of five would have been good/even coverage.

 

I've also gamed the {censored} out of some of my titles (BioShock and TombRaider!) with no issue, heat stayed at 35-40c at some extreme settings. The GPU temp went up a bit though. (79c in Furmark). :D

 

The only really crappy part is that I can't connect more than one SATA drive to my old PSU, as it's a BIT on the old side (350w P4 PSU from about 1999), and there is no way I would be able to power my GPU. This thing has a 20-pin and a 4-pin, so I could boot up the system, but all I would be able to run is CPU, RAM, and one SATA drive.

Share this post


Link to post
Share on other sites

I recently was using an Intel 80486dx2 rig, and had a look inside. It's from about that era (1999), and I must say I'm surprised that such a PSU is even going to power modern hardware :D

The CPU was passively cooled, and I somehow felt a little stupid with my custom watercooling loop on my rig that had cost me half a fortune.

 

Your PSU isn't really high quality - it's moe from Corsair's lower range. Your temps are very good by the way, with the stock cooler on my GTX680 I hit about 90 degrees while running BF3 for some hours on Ultra in 1440p resolution. I don't know if there is any way to test a powersupply in software - but I found this quote that might get your attention:

 

"However, a failing power supply can often be the cause of problems you might not expect like random lockups, spontaneous reboots, and even some serious error messages." - some random internet page

Share this post


Link to post
Share on other sites

I just get the random (and vague):

 

Windows Kernel event ID 41 error "The system has rebooted without cleanly shutting down first"

 

With no more information in the log (that I can see).

If memtest tells you after 14h that everything is fine, memory is probably fine.

 

This is some really strange stuff. I'd somehow suspect the CPU... this is the only thing that you didn't test so far. Maybe you can ask for a replacement CPU?

when you say you ran all of this on your old PC, you mean the software? have you tried using some of the new components in your old rig? like new ram and gpu in the old rig or new hdd/ssd ram in the old rig? something like that? And I know PSUs don't seem to be the issue, but just remember that if something is out of tolerance by just enough, it can cause wonky behavior on your machine. try the PSU from the old rig and see what you get.

 

looks like you might could try a different PSU... I hope it works so then I can say "I told ya so!"

 

In all seriousness, troubleshooting a finicky problem like this is the worst.. I agree that it would be better if something was obviously smoking or what not. Hell - sometimes its a combination of things... the perfect storm of random errors. You are on the right track... its seems as though the CPU and PSU are your last two items that you haven't swapped out. You picked up an el-cheapo MB for testing... try an el-cheapo PSU - at least try it before you try to RMA the CPU.

Share this post


Link to post
Share on other sites

The El Cheapo mobo is going to be returned for an El Cheapo PSU. :D

 

I will gladly listen to some "I told you so"s if that means this PC is working right. I mean, come on, I still have a lot of finishing to do with the bak panel, but I'm not gonna spend a lot of time on that until it's actually running stable.

 

Note to self: One more thing to consider. Grounding. Make sure to check it.

Share this post


Link to post
Share on other sites

yup - good call on the grounding as well... although that might be an indication of a flaky PSU as well...

 

... oh, and I think I'm gonna make a brand called El-Cheapo© so I can start making some money!!

Share this post


Link to post
Share on other sites
... oh, and I think I'm gonna make a brand called El-Cheapo© so I can start making some money!!

 

We'll be partners. Everything made in Brazil. Don't know if that will save us money.

 

I HAVE been shocked by my case a few times, but it's metal and the floor is carpet. :/

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×