• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

The Last straw (GTX 970 cry for help!)

Yeah I saw that one too, plus some others which seemed to have the same issues, only it wasn't the 32gb causing the problem, but that all memory slots were filled. I'm still not convinced.
 
implosiveturnip said:
Yeah I saw that one too, plus some others which seemed to have the same issues, only it wasn't the 32gb causing the problem, but that all memory slots were filled. I'm still not convinced.
Click to expand...

That's what I was thinking too, that it was related to all slots being used rather than the amount of memory. Afaik, using more slots changes the timings a little ... how that would affect a GPU though I'm not sure, maybe it's also PCI-E which is affected and the 970 being particularly sensitive to that.
 
On purchasing this system I too had random crashing which turned out to be RAM-related. Memtest was useless for finding memory errors. Prime95 quickly found errors on my RAM.
 
Sounds more like a mobo issue to me - removing sticks of RAM may cure it but that doesn't make the RAM itself the culprit. Particularly if other people are experiencing the same with all banks populated, suggests another issue. If the RAM worked with the 670 then it works - if the system starts failing memtest with a graphics card swap then it's most likely mobo or IMC, second most likely PSU (though I doubt PSU given it still fails with just a single 970).

Remember memtest and similar programs are not only testing the memory, they're testing the memory controller and the mobo too.
 
Last edited:
8-Pack reported that if you populated all your RAM slots eg 4 x 4gb whilst overclocking cpu it can cause instabilites with ivy bridge or haswell. Not in everone's case but I see your CPU is overclocked so it could be this.

However I have 4 x 4gb, CPU O/C, GPU O/C, MSI GTX 970 and mine is fine. From what you have said removing 2 x 4gb has proven the theory its the fully populated ram. I would think on the whole most people buy 2 x 4gb for 8gb or 2 x 8gb for 16gb etc which is why the problem is not that common.

Really you should set all your devices to the base line, eg stock CPU, Stock GPU, correct memory ram speeds with all 4 slots populated and test this first before eliminating anything else.
 
Might be luck of the draw, since if your processor has a slightly weaker IMC it could be just enough to tip it over the edge.
 
varkanoid said:
Really you should set all your devices to the base line, eg stock CPU, Stock GPU, correct memory ram speeds with all 4 slots populated and test this first before eliminating anything else.
Click to expand...

Already tried that, it's in the OP. You're right about OCing with 4 slots though, I originally had 4.8 stable with only 8gb installed, but wanted the 16gb I paid for and 4.5 was the highest I could get with all channels populated, which was fine because I got lower temps and 4.5 is no slouch tbh!

In the meantime I'll be able to bang the cards in my secondary system as I'll have a PSU for it tomorrow! Problem is finding time to test the buggers as I'm decorating our new house at the moment.


I appreciate all the input guys! Will be leaving memtest running after I leave for work in the morning so I will report back how well this goes. I will be leaving the rest of the system at stock for memtest, but the RAM on xmp.

The memory for those interested is a 4x4gb kit from Avexir, 2400mhz C10 @1.65v

I'm keeping my fingers crossed it's just some crappy Nvidia driver issue that just happens to conflict with something in my particular setup. Anyone know if there's a driver update due?
 
In your situation I'd have admitted defeat a long time ago and just re-installed Windows.

If you have a spare disk lying around then perhaps even do a secondary test install of Windows onto that just to test if it solves the problem.

Was in this situation myself the other day with flamin' GFWL crashing and preventing any game which used it from running. Managed to get it working in the end but I know my install is due for a nuke and rebuild.
 
Vertigo1 said:
In your situation I'd have admitted defeat a long time ago and just re-installed Windows.

If you have a spare disk lying around then perhaps even do a secondary test install of Windows onto that just to test if it solves the problem.

Was in this situation myself the other day with flamin' GFWL crashing and preventing any game which used it from running. Managed to get it working in the end but I know my install is due for a nuke and rebuild.
Click to expand...

It was the first thing I did after I realised I was having the issue in every game!

On another note, I woke up late this morning so I didn't have chance to set memtest running before I left. :o
 
Still no crash. Card has been running full tilt for around 6 hours now, maybe more. I'll test the second card at some other time. As for now, it's time for bed.

Memtest is on pass four on my memory now, so that doesn't appear to be the culprit at all, and so i'm guessing my board is fine too.

99.99% sure it's just Windows being a see-you-next-tuesday. Which makes me happy in some roundabout kind of way as I really didn't want to RMA anything!

2 things to note though: the PC I'm testing the cards in only has 2 of four DIMM slots populated. And I noticed about halfway through this evenings gaming session, that after I alt-tabbed out and opened chrome to make my last post, after tabbing back to the game there was every now and then some odd delay laggy thing going on. I'd left chrome open when going back in game, whereas it wasnt open before.

I had noticed previously that the GPU process in chrome was causing my cards to lock at boost clock and voltage when not in game. it's only just dawned on me this might be a problem. I'm thinking the times where it didn't crash for ages before are when I've disabled the GU process, gamed for some time, then opened a new instance of chrome thus making the GPU process start back up; making my games crash after tabbing back into them. Hmmmm...

Again, I'll keep updating, but now it's time for bed :)
 
**NECRO-UPDATE**

Sorry for digging my old thread up, but I thought I'd just offer some closure on this, and how it panned out for me.

I've been buying a/moving house so things have been quite hectic and I'd not had much chance to tinker with my kit until recently.

I have ended up putting both cards back in my main system, and I'm not having any issues now. That being said, I'm not opening chrome whilst I game as I'm using the machine solely to stream via steam in-home streaming down to my living room, (as I don't have a desk to sit at and game the proper way since moving :( )

I'm pretty sure either keeping chrome closed or an Nvidia driver update has solved my issues. I'd also changed my games storage drive from a 2 disk RAID 0 on the Intel chipset, to a single large SSD + a single 360gb mechanical drive for some media. I don't know if moving from the RAID had any impact on the issue - I wouldn't have thought so, but it's something to consider perhaps.

Anyway, thanks for everyone's help, and I hope this thread can give anyone else that's having similar issues point in the right direction.
 
Interestingly, I've just posted in the Overclocking section.

My machine has started to crash, well power off completely mid BF4. It was rock solid until I changed from my old 7970 to a GTX970. I was thinking that the crash could be PSU related, as the machine just dies completely and the PSU is old (6 years) but now reading this post - I'm not sure. Like you, I hav 4x4G sticks.

I'll pull two out when I get home from work and see if that changes anything. But, I've just picked up a new PSU just in case - which I think I'll leave sealed.
 
Awahwah said:
Interestingly, I've just posted in the Overclocking section.

My machine has started to crash, well power off completely mid BF4. It was rock solid until I changed from my old 7970 to a GTX970. I was thinking that the crash could be PSU related, as the machine just dies completely and the PSU is old (6 years) but now reading this post - I'm not sure. Like you, I hav 4x4G sticks.

I'll pull two out when I get home from work and see if that changes anything. But, I've just picked up a new PSU just in case - which I think I'll leave sealed.
Click to expand...

Wife's PC was suffering from restarts every 30-40mins regardless of gaming or sat there in Windows, replaced her 6yr+ old PSU problem stopped immediately.
 
mattcrane said:
Wife's PC was suffering from restarts every 30-40mins regardless of gaming or sat there in Windows, replaced her 6yr+ old PSU problem stopped immediately.
Click to expand...

Do you think that the change in GPU has shown a weakness in my old PSU or pushed it over the edge? Does sound logical but the GTX970 'should' depending on what review you read draw less wattage than a 7970 clocked at 1200Mhz
 
implosiveturnip: IF you fancy seeing if the problem still exists with Chrome... is there an option [in chrome] to change how it produces the browser graphics?

Where you can tell the browser to use your CPU or your GPU to make the browser graphics?

Also, do you have ULPS disabled?
 
Sounds really random, but try this : set your PCI-E lanes to run in Gen 2.0! I had this issue on Z77 motherboards including my current one, games would just crawl to a stand and lock up the entire PC when running lanes in PCI-E 3.0 setting to 2.0 just worked and you won't notice the bandwidth change worth a try
 
Last edited:
Threepwood said:
implosiveturnip: IF you fancy seeing if the problem still exists with Chrome... is there an option [in chrome] to change how it produces the browser graphics?

Where you can tell the browser to use your CPU or your GPU to make the browser graphics?

Also, do you have ULPS disabled?
Click to expand...

Yuo mean hardware acceleration?

I was going to mention this too, hopefully solves the problem with crashing while chrome is open.
 
You must log in or register to reply here.
Back
Top Bottom