Jump to content
Corsair Community

Gigabyte GA-Z87-D3HP and CMZ16GX3M4A1600C9 - Crashes, Lockups, BSOD


chrispy550

Recommended Posts

Hello,

 

I recently purchased a new system.

 

My motherboard is as stated in the title:

 

Gigabyte GA-Z87-D3HP

 

Memory is Corsair Vengeance 16GB in (4 x 4GB configuration)

CMZ16GX3M4A1600C9

 

My processor is QuadCore Intel i5-4670K Haswell.

 

I have created a ticket on Corsair's support website. It stated I would see a reply in 1 business day. It has been 5 days as of the creation of this thread.

 

First, the problem:

The system will Blue Screen at desktop if all 4 memory modules are in the system and with the BIOS setting forcing the RAM into the advertised 1600mhz mode. If I simply activate the XMP profile in BIOS it remains at 1333mhz despite the BIOS showing 1600mhz. System will blue screen at desktop in forced 1600mhz mode and system is unstable at 1333mhz mode (games crash, blue screens, lockups, hangs, hiccups, looping sound, etc).

 

Went into BIOS, attempted to enable a BIOS feature called "Enhanced RAM Stability Mode". This appeared to help at first, but inevitably the same result occurred. System can boot at 1333mhz with all 4 modules in, but it continues to blue screen, hang, crash, etc.

 

Removed two sticks of RAM from system - this brought me down to 8GB total RAM in system. This initially seemed to work and I thought my problem was bad RAM slots in the motherboard. I returned the board, assuming it was defective. Got new board. Installed everything. Same problem. To me this ruled out bad RAM slots, but its always possible to get two defective boards in a row.

 

On the current board I have removed 2 sticks again, so total RAM in system is 8GB. I have moved them into each of the two dual channel slots to test.

I have now discovered that:

If the system is in "Enhanced RAM Stability Mode" (BIOS setting) and only 2 of my 4 memory modules are in I can run the system stable at 1333mhz AND forced 1600mhz at stock timings and voltage.

 

If, at any time, I turn off "stability mode" or I put all 4 memory sticks into the system, it returns to blue screens and lock ups.

 

My RAM part number CMZ16GX3M4A1600C9 is on my board's compatibility list.

 

I ran MemTest for 12 hours with all 4 sticks in system, no errors.

 

I do not believe the RAM to be defective, but it is a possibility because I am a layman at this and not an expert. I have explained all of this in detail to Corsair in my ticket and received no response.

 

I am running Windows 7 Enterprise 64-bit.

 

All my other hardware is available in my profile, but if you can't see it or need to know, ask. I have tried reinstalling Windows as well. At this point my belief is that (despite being on my mobo's compatibility list) the 4 slots of RAM that I have are not stable/compatible with my system/hardware. I can't explain it, but I believe it to be true. I have asked Corsair for an exchange for different RAM, preferably in a 2 x 8GB configuration, no response.

 

Thank you for taking the time,

-Chris

Link to comment
Share on other sites

  • Replies 56
  • Created
  • Last Reply

Chris , please make sure you have the latest BIOS for your board

 

Then all you should have to do is enable the XMP profile in your BIOS. This will set the memory and also adjust other necessary system voltages. Do not use "stability mode" Jus XMP. If you are unfamiliar with that setting please refer to your manual for the location .

 

My RAM part number CMZ16GX3M4A1600C9 is on my board's compatibility list.

 

I ran MemTest for 12 hours with all 4 sticks in system, no errors.

The memory is compatible and since it will pass memtest , that would suggest just a BIOS setting at fault. You may have to manually adjust your memory voltage to 1.55 or 1.6v if XMP does not cure your issues on it's own.

I have asked Corsair for an exchange for different RAM, preferably in a 2 x 8GB configuration, no response.

Unless you bought them directly from the Corsair store the best they can do is exchange it for the exact same kit. Although a 2x8 kit would have been less strain on your CPU's memory controller, it should still run your 4x4 kit . They are indeed 100% compatible with that board.

Link to comment
Share on other sites

Thank you for your reply.

 

I am running the latest BIOS - flashed it twice since purchasing and RMA'ing.

 

XMP setting yields no results. There is only "profile 1" under the XMP setting. When I enable that, it claims to be taking the RAM from 1333 to 1600 in BIOS however using CPU-Z and/or AIDA64 in Windows reports the RAM to be unchanged and still running at stock 1333mhz.

 

I have not yet tried changing the voltage setting directly. I assumed the XMP profile would do that automatically. Also, I bought this RAM because it advertised 1600mhz and 1.50V. It won't run stable at 1333mhz and 1.50V, I find it hard to believe it will run at 1600mhz and 1.50, 1.55 or 1.60V but I will try it, provided it won't harm the system.

 

That is discouraging to hear that they will likely not exchange. I don't believe my RAM to be defective and it's supposedly compatible, just not with my system. I will continue to try to work the problem.

 

UPDATE: Changing the stock voltage to 1.55V and enabling the XMP profile as you describe actually allows the PC to boot to desktop AND in 1600mhz mode, which is a first for me (with all modules in). Remains to be seen if it is stable or not.

 

UPDATE 2: Problem persists. Computer can boot to desktop and run for a time at 1600mhz in XMP profile and manually set to 1.55V, but eventually blue screens and/or crashes/locks-up.

 

UPDATE 3: Went up to 1.60V. Same problem. Can still boot, but crashes within minutes of any program. RAM advertised at 1.50V at stock and XMP.

Link to comment
Share on other sites

I'm in the same boat with the same RAM, same board and same CPU. The RAM has been tested to death previously as the first batch I bought for a previous build had to be RMAed.

 

My problem isn't quite as severe as the machine has been BSODing about once/twice per day in normal use. I've tried the 1.55v tweak (although the motherboard utility shows 1.536v) and will continue to monitor. BIOS is the current highest non-Beta (F4). Memory is currently clocked at 1333.

Link to comment
Share on other sites

I'm in the same boat with the same RAM, same board and same CPU.

 

Sorry to hear you're having the same problem. At the same time, I am relieved to see someone with my same hardware having the same or similar issue.

 

At this point my options are dwindling. I was happy to see I could run the 4 sticks at 1600mhz, but its a moot point if it won't run stable. If Corsair doesn't swap out the sticks for ones that are actually compatible with this hardware, I will need to RMA this mobo AGAIN and this time get a different board.

Link to comment
Share on other sites

Chrispy, as I stated earlier this memory is 100% compatible with that MB/CPU.

 

I have not yet tried changing the voltage setting directly. I assumed the XMP profile would do that automatically. Also, I bought this RAM because it advertised 1600mhz and 1.50V. It won't run stable at 1333mhz and 1.50V, I find it hard to believe it will run at 1600mhz and 1.50, 1.55 or 1.60V but I will try it, provided it won't harm the system.

The problem with voltages comes in to play when you populate all your slots. It puts more strain on the CPU's memory controller than it would running just two sticks. So often you need to make some minor tweaks.

 

There is also the strong possibility that you have a MB that under volts the memory bus, which happens quite a bit. That has been proven here on these forums many time. So it's not that the memory won't run at 1.5v. There are other factors that come into play. They all run and pass memtest at 1.5v so the memory should be fine.

XMP setting yields no results. There is only "profile 1" under the XMP setting. When I enable that, it claims to be taking the RAM from 1333 to 1600 in BIOS however using CPU-Z and/or AIDA64 in Windows reports the RAM to be unchanged and still running at stock 1333mhz.

Which information in CPUz are you looking at? With XMP enabled look at the memory tab in CPUz. It should read 800mhz. DDR stands for "double data rate" so you would take the 800mhz and multiply by 2 for en effective speed of 1600mhz. If your looking at tne SPD tab that information is static and never changes. That just list the differnt JDEC profiles the memory is programed with. If your seeing something different could you please post a screenshot?

 

Both of you are more than welcome to request an RMA and have the memory replaced. Which might be a better idea than swapping boards again.

 

But generally they will not change memory kits. You can only RMA for the same kit. Unless you bought it from the Corsair store directly, then you can return them and get another. If not then you would need to take it up with your reseller.

 

@bigfatron

My problem isn't quite as severe as the machine has been BSODing about once/twice per day in normal use. I've tried the 1.55v tweak (although the motherboard utility shows 1.536v) and will continue to monitor. BIOS is the current highest non-Beta (F4). Memory is currently clocked at 1333.

Let me know how this turns out for you. You can go as high as 1.6 if you need to.

Link to comment
Share on other sites

peanutz94,

 

I want to thank you for at least helping with the situation.

 

I understand what you are saying about my mobo and the possibility that its under-volting because I have 4 modules. To attempt to remedy this I have tried running the XMP profile and manually setting the voltage to 1.55V and later to 1.60V. Both of those voltages allowed me to boot to desktop and at 1600mhz as well, as I stated in my updates above. Unfortunately, it didn't solve the problem since all the crashing and blue screens persist.

 

I also understand that 800mhz = 1600mhz. That is what CPU-Z reveals so I know I am at 1600mhz when I activate XMP and use 1.55V or 1.60V. I also have a program called AIDA64 that actually directly shows 1600mhz. I've also seen the SPD tab and understand that it's static.

 

I have now requested an RMA from Corsair directly. I have been approved, however I have the STRONG suspicion that my RAM is NOT defective and if they choose to send me the same modules I am currently running I believe I will have the same problem and be without a computer for longer. I have requested to this representative that I be offered a different and comparable Corsair product, but there's no indication of whether or not that will be possible. Unfortunately, the reseller is newegg and they will not replace outside of 30 days.

Link to comment
Share on other sites

I want to thank you for at least helping with the situation.

 

I understand what you are saying about my mobo and the possibility that its under-volting because I have 4 modules. To attempt to remedy this I have tried running the XMP profile and manually setting the voltage to 1.55V and later to 1.60V. Both of those voltages allowed me to boot to desktop and at 1600mhz as well, as I stated in my updates above. Unfortunately, it didn't solve the problem since all the crashing and blue screens persist.

NP, You just never know the level of a users knowledge from a couple of forum posts. So i thought i would throw that information out there.

 

I have the STRONG suspicion that my RAM is NOT defective and if they choose to send me the same modules I am currently running I believe I will have the same problem and be without a computer for longer

I don't believe there is anything wrong with them either. But you might want to try calling CS on Monday and see about an advanced replacement. That way you will only be down as long as it takes to change the modules. You would also be able to discuss changing kits. They are closed on the weekends.

 

Let us know how you make out.

Link to comment
Share on other sites

Just a quick update on my own predicaments. The 1.55v volt nudge has been OK thus far, although i'd left XMP off so the RAM is at 1333. Tried sticking XMP on and it BSOD'ed on me twice fairly soon after (including one time with a further increment to 1.6v).
Link to comment
Share on other sites

Well, to both of you I'm kind of stumped. There should be no reason that memory shouldn't run just fine. But since both of you have the same exact board, I'm wondering if it isn't something that would need addressed in a BIOS update?

 

Or possibly some thing to do with Haswell's new power structure having all basic power functions basically happening in the CPU itself?

 

And there is a SLIM possibility that the memory just doesn't like that particular board. (basically a compatibility issue) but the memory is tested on those chipsets and again, there is no real technical reason they shouldn't run .

 

Sorry the best I can do know is wait to see what RamGuy has to say. He may have a little more insight to the new platforms.

Link to comment
Share on other sites

Thank you again for helping. I am going to go through with this RMA soon, but I am going to push hard for a different part number. For the above reasons stated, I think the only option for me is going to be new RAM despite my current ones being on this board's compatibility list.

 

I also considered the possibility that Haswell and it's power saving is coming into play, but don't know how to remedy.

 

I will update with anything relevant.

Link to comment
Share on other sites

Well, to both of you I'm kind of stumped. There should be no reason that memory shouldn't run just fine. But since both of you have the same exact board, I'm wondering if it isn't something that would need addressed in a BIOS update?

 

Or possibly some thing to do with Haswell's new power structure having all basic power functions basically happening in the CPU itself?

 

And there is a SLIM possibility that the memory just doesn't like that particular board. (basically a compatibility issue) but the memory is tested on those chipsets and again, there is no real technical reason they shouldn't run .

 

Sorry the best I can do know is wait to see what RamGuy has to say. He may have a little more insight to the new platforms.

 

This particular RAM is on their supported list if I remember right. You may also be correct regarding the BIOS thing, although i'm not desperate enough at this stage to update to a beta BIOS.

Link to comment
Share on other sites

This particular RAM is on their supported list if I remember right. You may also be correct regarding the BIOS thing, although i'm not desperate enough at this stage to update to a beta BIOS.

 

I've been on the beta BIOS for days now and it hasn't helped the problem any, so don't bother updating, yet, doubt it'll help.

Link to comment
Share on other sites

I am also experiencing random BSOD with an H87 Gigabyte board, I thought it was the GTX 660 video card at first because I could get it to crash all the time from sleep. I returned the card and I'm using the integrated card now and I'm still getting BSOD crashes, they vary from memory errors to page faults but they aren't as often now. So I thought it was the memory and ran memtest for 12+ hours, not a single issue with the memory. I really wanted it to be a memory problem. Then I uninstalled all the old video drivers from nvidia, crash still.

 

What I found on newegg is a suggestions to disable C3 and C6 states in the BIOS, this came directly from Gigabyte and was in reference to the Z87 board. I'm currently trying this but my crashes take a few days before they popup now. If you're out of ideas maybe try disabling C3 and C6 and see if it helps. I have no idea if it will help or not, but I do think something is amist with the Z87 gigabyte boards and certain memory configurations. Another suggestion was running it with just a single stick of ram, again someone on newegg said it solved their BSOD, thereby pointing to a memory controller issue. I think I saw someone else say they swapped out to an Asus board and all was well.

 

 

I know computers are very complex and the issues I have may not be even related to your issues. Just sharing what I know from researching this the past couple days.

Link to comment
Share on other sites

I am also experiencing random BSOD with an H87 Gigabyte board, I thought it was the GTX 660 video card at first because I could get it to crash all the time from sleep. I returned the card and I'm using the integrated card now and I'm still getting BSOD crashes, they vary from memory errors to page faults but they aren't as often now. So I thought it was the memory and ran memtest for 12+ hours, not a single issue with the memory. I really wanted it to be a memory problem. Then I uninstalled all the old video drivers from nvidia, crash still.

 

What I found on newegg is a suggestions to disable C3 and C6 states in the BIOS, this came directly from Gigabyte and was in reference to the Z87 board. I'm currently trying this but my crashes take a few days before they popup now. If you're out of ideas maybe try disabling C3 and C6 and see if it helps. I have no idea if it will help or not, but I do think something is amist with the Z87 gigabyte boards and certain memory configurations.

 

I know computers are very complex and the issues I have may not be even related to your issues. Just sharing what I know from researching this the past couple days.

This is kind of what i was suspecting. But when ever errors or BSOD's show up, everyone wants to blame the memory . But all three of you have Gigabyte boards and I really believe the issue lies with them. But lets see how this works out. There has not been on e issue with the new chipsets and any other brand board..at least that I know of here in the forums.

Another suggestion was running it with just a single stick of ram, again someone on newegg said it solved their BSOD, thereby pointing to a memory controller issue. I think I saw someone else say they swapped out to an Asus board and all was well.

If it would turn out to be a memory controller problem , then it would boil down to a bad CPU or MB since that is where the memory controller resides. They moved those from the MB to the CPU with the launch of the first gen I series CPU's But since running only a single stick of ram solved that users issues , then that board is bad. There is nothing else it could be. It also defeats the purpose of dual channel effectively cutting memory performance in half.

 

Just out of curiosity did either bigfatron or chrispy get in contact with gigabyte?

 

@bigfatron

This particular RAM is on their supported list if I remember right. You may also be correct regarding the BIOS thing, although i'm not desperate enough at this stage to update to a beta BIOS.

Exactly! That means that this kit has been specifically tested on that board. And also another reason why I believe you both have bad MB's. But again, lets see how this plays out and please keep us updated if you can.

Link to comment
Share on other sites

I just wanted to report back my findings with the C3, C6 suggestions I saw on newegg for the Z87 Gigabyte board. Even though I'm running the H87 I think there is some overlap (my manual even says ga-Z87 on it), but because the issue also pops up on the H87 from my personal experience and other comments I've seen on newegg referring to BSODs.

 

 

Anyways, it crashed still, although I was getting different error codes this time...I kept getting "CRITICAL_OBJECT_TERMINATION" or F4 errors. Never got those before, I then went on to turn off the C3, C6, C1E, and the EIST features. This also resulted in a crash.

 

What I found very strange is I could sort of make it crash with those settings off by running prime95....then stopping it. The crash would happen AFTER running prime95 when the voltages start to drop and the temp goes down, hard to say if it was a coincidence or not, I have no idea and didn't feel like figuring it out or watching voltages. So I renabled the C3, C6, C1E, and the EIST and I'm back to square one sort of....

 

 

When I was having the sleep issues with the video card, a lot of people said "its the ram!" then some other people said "it's the ram timings!". I tried to drop down from 1600 to 1333 on my 2x8gb corsair ram, but it would never stick for some reason. So I went with the "it's the video card"...since memtest ran for 12+ hours with no errors.

 

 

I've gone back to the original idea of maybe it's the timings, and was actually able to get the memory set at 1333 9-9-9-24 which is the SPD rating for my ram (instead of 1600 10-10-10-27). What I have discovered 100% for a fact is the memory settings are not sticking unless you turn the computer OFF. That's why when I first tried the timings idea before it never worked, I always did a "save & restart" from BIOS but the machine never turned off and the settings never locked in....they saved yes....but they didn't turn on. I didn't clue into this until I couldn't set the tRAS setting on the second channel, it would not stick, then I did a "shutdown" in windows by accident....bam....tRAS set. So I know for a fact on the Gigabyte H87 you have to turn off the machine for the memory settings to hold, or at the very least it's a random bug and turning it lock in the settings.

 

 

So this will be my final test before returning the motherboard, I don't mind running at the SPD settings for the ram if it works. But I'm just tired of trying to figure what's wrong, and there is very little support in terms of BIOS updates on the H87 vs the Z87 for gigabyte. The Z87 has update after update and even beta BIOSs, lots of people complaining that the BIOS updates still don't fix their issues, but at least you can see they are trying to fix it. But on the H87 I have lost hope as there has not been a single update.....which I understand since it's just a basic board and probably not enough people use it. So if the SPD settings which I finally got working fail, I will upgrade to a more popular board that has more BIOS updates and fixes.

 

I'll report back what I find, maybe it will help someone else. Hopefully the memory setting bug I found helps out someone, it might be fixed in the Z87 BIOS since they are 3 revisions ahead of me, but the Gigabyte H87 F3 BIOS....you have to power the machine down after you change your memory settings for them to actually take effect.

 

I also ran memtest 4.3.0 BETA last night, again on the 1600 setting.....no errors...again. My theory is these boards H87 and Z87 were just released too early with not enough testing, and everyone that bought them are now doing free R&D for gigabyte so they can make the BIOS what it should have been in the beginning.

Link to comment
Share on other sites

I've gone back to the original idea of maybe it's the timings, and was actually able to get the memory set at 1333 9-9-9-24 which is the SPD rating for my ram (instead of 1600 10-10-10-27). What I have discovered 100% for a fact is the memory settings are not sticking unless you turn the computer OFF. That's why when I first tried the timings idea before it never worked, I always did a "save & restart" from BIOS but the machine never turned off and the settings never locked in....they saved yes....but they didn't turn on. I didn't clue into this until I couldn't set the tRAS setting on the second channel, it would not stick, then I did a "shutdown" in windows by accident....bam....tRAS set. So I know for a fact on the Gigabyte H87 you have to turn off the machine for the memory settings to hold, or at the very least it's a random bug and turning it lock in the settings.

But the only problem there is that your CPU has a native 1600mhz memory controller. All memory will default to the SPD 1333mhz when first installed just to be sure it will boot successfully. But with your CPU and chipset 1600mhz shouldn't be an issue at the advertized settings.

 

I also ran memtest 4.3.0 BETA last night, again on the 1600 setting.....no errors...again. My theory is these boards H87 and Z87 were just released too early with not enough testing, and everyone that bought them are now doing free R&D for gigabyte so they can make the BIOS what it should have been in the beginning.

They are all the same basic chipset family . The only differnce being a few extra bells and whistles if you get my drift. But I have to agree with you and your findings.

 

And please guys I'm not saying this because I'm a "fanboy" or anything like that. Just giving you my complete honest feelings on the issue. I really don't think this is a memory issue but more of a problem that lies with Gigabyte.

.you have to power the machine down after you change your memory settings for them to actually take effect.

Actually , you should clear your CMOS before changing settings. Just powering down may not be enough to flush out the old setting stuck in the BIOS.

Link to comment
Share on other sites

Just out of curiosity did either bigfatron or chrispy get in contact with gigabyte?

 

I haven't as yet although I fully intend to. They want the s/n off the motherboard for the support ticket and I haven't come round to popping the lid off to find it yet. I'll report back with any updates as and when.

Link to comment
Share on other sites

Just out of curiosity did either bigfatron or chrispy get in contact with gigabyte?

 

No. I had assumed that since they have the RAM on their boards compatibility list that they must test these? But that is appearing to not be the case.

 

Exactly! That means that this kit has been specifically tested on that board. And also another reason why I believe you both have bad MB's. But again, lets see how this plays out and please keep us updated if you can.

 

As I stated in my original thread, this is my second one of these boards. When I first started working through this problem my determination was that I had a bad motherboard and I promptly RMA'd it for another. As we know, it is entirely possible to receive two defective motherboards in a row, but I don't believe it to be a 'defective' issue. It is seeming like they didn't do enough testing before releasing these boards. I really like Gigabyte boards, but I am strongly considering RMA'ing this board AGAIN and going to something more established so I can just have a working computer.

 

Corsair has declined sending me a new part number for RAM and instead is only going to send me new sticks of the same part number, which I believe will be a waste of time - I don't think my RAM is defective.

 

 

One last thing to report: I too attempted the C3, C6, EIST "fix". I have a friend that is very knowledgeable helping me through this problem since I first started having it. One of his suggestions was that it is a power saving issue with Haswell. But unfortunately, just as you discovered yourself, trying to remedy the problem by disabling those power saving BIOS features proved fruitless as well.

Link to comment
Share on other sites

Raised a ticket and it was suggested that I try the beta BIOS (F5C at the time of writing). I have not long since applied this and the system is currently running with RAM at 1600MHz. I will report back in a day or two with an update on stability.

 

EDIT: clearly no help with stability and also caused an issue with the USB 3 (kept coming up with 'This device can perform faster') so restored the previous BIOS. Further updates to follow when I get them...

Link to comment
Share on other sites

I'll just comment on what I tried, aka lowering my memory to 1333, it was stable for over 24hrs...which was a first! I was doing prime95 a number of times, putting the pc to sleep, waking it up (it usually hates that), I couldn't get a BSOD....I was pretty happy. Then 5 min ago....I notice I plugged a device into the USB and it didn't pick it up.....then my internet stopped working. I kept waiting for a BSOD, nothing, I couldn't open "explorer.exe" to check my system, I had one still in the taskbar...all my drives vanished, no ssd drive, no 1TB drive...but my DVD drive stayed online for some reason. Then some random memory popup, but no BSOD. I had to reset the machine..it wouldn't BSOD for some reason, of course it works on reboot....for how long? that's up to the Gigabyte gremlin to decide. But I'm done with this Gigabyte H87 board, I'm going to see if I can get a refund on it, I'll be moving onto the top of the line ASUS H87 board since I don't overclock...unless I see a Z87 on sale....because I now know Z87 boards get more BIOS updates, which is nice since all these boards are new.

 

 

It will probably be a few weeks until I can post back with what I find out on the Asus side of things. If it takes a while to get a reply for a refund, I might test out a single stick of ram just for kicks.

 

 

Also if anyone is wondering why I didn't try Gigabyte support, I've seen mixed results from doing that. One they might just replace the board, but maybe it's just the BIOS that needs fixed...which is what I'm leaning towards....and second..sometimes it takes up to a week to get a reply from them...I don't feel like playing email tag with them. All these options close the window on me getting a refund, so I'll cash out now instead of playing the waiting game with Gigabyte and see what an Asus board does. I'll report back what I find, hopefully we all figure this out and find a solution I know how frustrating this is.

Link to comment
Share on other sites

chunky,

 

I am with you.

 

Even as I currently await my RMA RAM from Corsair I am looking for an ASUS board since I still have the option to return this Gigabyte mess.

 

I like Gigabyte a lot and my last build ran flawlessly with it for years. It's unfortunate to see what happened with these particular boards. It is probably a very small issue with our exact hardware's, but never-the-less I would have preferred a bit more extensive testing when they placed our RAM on their board's list of compatible hardware.

 

I suppose this comes with the territory when you upgrade when a new generation comes out. This is my first time waiting for a new generation of processor/socket and buying/building right away. There are headaches.

Link to comment
Share on other sites

Also if anyone is wondering why I didn't try Gigabyte support, I've seen mixed results from doing that.

 

I must admit its been a mixed bag thus far. They've been responding fairly quickly but there is little indication that the suggestions are based on expecting results rather than 'just try everything' (like they suggested trying changing some settings but weren't saying what to set them to and so on). Things are currently at a bit of an impasse as i'd pointed out the USB 3 issues that their beta BIOS was causing. They'd also suggested checking the chip socket although i'll need to first get some more thermal paste for remounting the heatsink, and getting it off in the first place is a fairly hefty task in its own right (its the BeQuiet one that was bundled with the board).

 

The RAM i'm not convinced is the issue as I've run Memtest for about an hour solid with no issues. I'll do a longer run later. The only RAM I have available to swap it out with is only 1333 stuff anyway. Plus the comedy-oversized heat spreaders mean the chore of stripping the fan off in order to get the RAM in/out of slot 1.

 

One other thing I should be able to rule out shortly is power. The PSU I've got is a few years old and admittedly cutting it a little fine (middle of the road Thermaltake 500W PSU and PSU calculators reckon i'm drawing about 400W). I needed a new one anyway (i'm passing on my old i5 750 machine to someone else so this PSU will go back into that) and that should be in the machine by this time tomorrow. Is this a viable suspect for BSOD behaviour?

Link to comment
Share on other sites

One other thing I should be able to rule out shortly is power. The PSU I've got is a few years old and admittedly cutting it a little fine (middle of the road Thermaltake 500W PSU and PSU calculators reckon i'm drawing about 400W). I needed a new one anyway (i'm passing on my old i5 750 machine to someone else so this PSU will go back into that) and that should be in the machine by this time tomorrow. Is this a viable suspect for BSOD behaviour?

Even drawing 400w out of 500w PSU shouldn't cause a blue screen. If your voltages were low to start with , sure it possible or are dropping more than 5% under load. You could rule that out easy enough by stressing the system and watch your voltages as it run's. If the 5v or 3.3v rail drop real low that will usually cause a crash or blue screen.

 

If there is one thing I've learned about computers is that EVERYTHING is a viable suspect for a blue screen! LOL

 

However, i don't believe power is the issue either.

 

Let us know how you make please!

Link to comment
Share on other sites

Even drawing 400w out of 500w PSU shouldn't cause a blue screen. If your voltages were low to start with , sure it possible or are dropping more than 5% under load. You could rule that out easy enough by stressing the system and watch your voltages as it run's. If the 5v or 3.3v rail drop real low that will usually cause a crash or blue screen.

 

If there is one thing I've learned about computers is that EVERYTHING is a viable suspect for a blue screen! LOL

 

However, i don't believe power is the issue either.

 

Let us know how you make please!

 

Having just re-run the numbers then its throwing up a minimum wattage of 401W. Although I didn't enter an ageing figure for that (must be 3 years old), so add 20% as a conservative figure and that's taken us to a new minimum of 481W and a recommended wattage (531W) in excess of the supply (Thermaltake TR2-500). EDIT: a bit of googling of the part in question suggests the 500W rating was optimistic for this PSU anyway

 

As an experiment I've popped out the graphics card, enabled the integrated graphics and am currently running a Prime95 test on it. It will be interesting to see how that goes. Then hopefully the new PSU lands on Thursday.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...