• 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.

NVIDIA GTX 970 OWNERS THREAD

I did yeah I could not get over 3.5gb even in full screen, ultra textures etc.... It sat at 4gb on my 980 and does on the 290 I have now as well.

I think there was something up with the card from the get go and the yellow hatching is it giving up. I had issues with the DP randomly not displaying anything as well...
 
I did yeah I could not get over 3.5gb even in full screen, ultra textures etc.... It sat at 4gb on my 980 and does on the 290 I have now as well.

I think there was something up with the card from the get go and the yellow hatching is it giving up. I had issues with the DP randomly not displaying anything as well...

I think if Nvidia released new drivers a lot of these problems with 970 or 980 would go away.
 
Had an MSI 970 working nicely for a few weeks now (upgrade from a R9 290) until today, when attempting to play any game made my system reboot. Tried a few things but it kept doing it until the last couple of times when on power up I get the following message:

"Please power down and connect the PCIe Power Cable(s) for this graphics card"

Checked the cables and they seem to be ok. Eventually it booted back into Windows again.

Has anyone else had this problem? I'm guessing it's either the card or the PSU (Corsair 650w thats a fair few years old now)?

EDIT: Swapped the power supply out and normal service seems to be resumed - phew! Loving this card by the way, very much prefer it over the 290! Can't believe how quiet these things are!
 
Last edited:
hey sorry to but in here guys but had the MSI 970 in the shopping cart for a while now but havent bought as ive been seeing these messages and videos about the awful coil whine its giving off in some cases. Now is this a problem with MSI or is it gonna be a matter of going for a different brand of the 970 range. cheers!
 
My MSI seems to have developed a fault, every now and then only one of the fans spins, the other one you can see is trying, as it twitches. Until I gently poke it with my finger does it then start to spin.

That's the problem people are having which the BIOS update tried to fix. Isn't a problem for me as I run my fan at a constant speed but for those on auto it's a nightmare.

I had a bug with my MSI last night where one of the fans was stuck at 100% speed after gaming. I had to shut the PC down and power back up to get it to go to 0% again in windows. I tried manually setting the fan speed with afterburner but it would not budge. I had already upgraded to the latest bios. Hoping it is just a one off and not something that is going to happen frequently.

MSI and fan issue shocker, surprised they didn't blame it on bit mining and high temps. My view of the company went from MSI rules to MSI disgust me after they came out with that. Yet people still keeping buy them, must be the cheap deals?

ps I'm going to save up for a Gigabyte or EVGA :D
 
Last edited:
Leave heaven looping for a while when you can or just play some games for a lengthy period. Certain games will work fine others will probably crash at those clocks. Diablo 3 seems to hate an overclock, can play for 2 hours and it will be fine, then go on the next day and get a driver reset within half an hour. Games like BF4, ARMA 3 and Planetside 2 don't crash at all.

To be honest Diablo 3 doesn't even need an overclock so I may just switch to stock clocks while playing that, could also be the fact of new drivers on a new card but I'd have to check if other people have had crashes on 9 series cards. Max bench stable and max game stable can sometimes be a fair distance apart.

hmm i'm wondering if it's related to drivers too...i asked about this on the 980 thread as i have the same problem on a 980 g1, driver resets in diablo 3 while 'overclocked'. thing is, the card isn't actually overclocking when running d3, it mostly sits at 1100 with brief excursions up to 1200, sometime as low as high 900s. it never gets into 1300 territory, let alone the 1450+ it will do for hours in heaven and valley. ie, it's actually running at stock speeds (i haven't o/c'd the mem at all).

i wonder if it's a bug related to the driver when an overclock is applied, regardless of the actual speed that the core is running at. the kicker is i've seen this behaviour on two different cards, but i need to run it more on the other card to be sure.
 
I've been collecting some data on my overclocks and when crashes occur, and 99% of the time they are related to power states - or rather power state lag, let me explain.

So, I overclock to 1500mhz boost, have gpu-z open and at desktop voltage is 0.85v and clocks are 150mhz. Launch 3dmark and I see the clocks go up to 1126mhz and volts to 1.05v. When the bench starts, the clocks go up to 1500mhz and the volts briefly go to 1.075v then to the full 1.21v. Bench completes no problem. If I clock above this, when the bench starts, the clocks go up to 1510mhz for example but the volts still step to 1.075v then to the full 1.21v. More often than not though the bench freezes at 1510mhz and 1.075v, causing a driver hang and reset. If the bench gets past the first couple of seconds the volts get all the way to 1.21v and the bench completes ok, I've benched up to 1600mhz successfully as long as it gets past the first couple of seconds.

A game I play a lot is Planetside 2 and the only crashes I see when overclocked are of the same variety. I'll be playing quite happily but as it doesn't require a lot of gpu power it'll often be running at 1126mhz and running 1.05v. Occasionally though there will be a big explosion on screen and the card will upclock to 1500mhz boost due to the load but the power state and volts will lag behind occasionally causing a driver reset or hard lock up at 1.075v again. This is very repeatable.

So, it seems my card is quite capable of running at a high mhz, but when the voltage lags behind due to the powerstates, it's causing me instability. as obviously 1.075v isn't enough to push 1500mhz or above.

I'm hoping the above gets fixed with a driver update and/or modded bios. I thought EVGA precision had a function called kboost which fixed clocks/voltages so I could test this but the option cannot be clicked when I try it. If I run a small 3d app in the background that gets the voltages up before a 3dMark bench run, it completes no problem, thereby confirming it's the transition in powerstates that causes the hang/crash.
 
Last edited by a moderator:
interesting. i was wondering a similar thing, as it's one scenario that would be different between the o/c and non/oc configs when the core is actually running at the same clock under both configs (as in d3, for me). as you say, it's the core clock it might be trying to transition too under demand that is different.
 
So my curiosity got the better of me and I decide to take a look at the card - since I installed the MSI 970 I haven't really taken the side off my case, just plugged it in and set about benching, adjusting settings and playing games

It looks as though my fans are behaving as they should (if I force them to ramp up for a bit then cut the benchmark off they stop again as they are meant to)... But I notice I do have pretty bad coil whine

However when the side of the case is on I can't hear a thing (it's an Antec P183)... sounds absolutely whisper quiet to my ears and when benching even once it hits ~68-70 degrees which is the absolute highest I've seen it go, the fan is barely above about 50% and I also can't hear it at all (compared to my old 670 where I'd frequently hear the fan blast into the 80% plus range)

So my question is this - given that I can't really hear it... should I care? Is it damaging the card or the rest of my PC? If it's just an unfortunate side effect of operation but ultimately not harming anything then I'm happy to just leave it alone
 
So my curiosity got the better of me and I decide to take a look at the card - since I installed the MSI 970 I haven't really taken the side off my case, just plugged it in and set about benching, adjusting settings and playing games

It looks as though my fans are behaving as they should (if I force them to ramp up for a bit then cut the benchmark off they stop again as they are meant to)... But I notice I do have pretty bad coil whine

However when the side of the case is on I can't hear a thing (it's an Antec P183)... sounds absolutely whisper quiet to my ears and when benching even once it hits ~68-70 degrees which is the absolute highest I've seen it go, the fan is barely above about 50% and I also can't hear it at all (compared to my old 670 where I'd frequently hear the fan blast into the 80% plus range)

So my question is this - given that I can't really hear it... should I care? Is it damaging the card or the rest of my PC? If it's just an unfortunate side effect of operation but ultimately not harming anything then I'm happy to just leave it alone

who wants a whinny card :eek::confused::confused:
 
who wants a whinny card :eek::confused::confused:

I take your point but I can't hear it at all... if I take the side of the case off and get down right next to it with my ear almost touching the card sure, but with the side on it's totally inaudible. In fact if I'd never looked in this thread I expect I might well have never even known
 
I've been collecting some data on my overclocks and when crashes occur, and 99% of the time they are related to power states - or rather power state lag, let me explain.

So, I overclock to 1500mhz boost, have gpu-z open and at desktop voltage is 0.85v and clocks are 150mhz. Launch 3dmark and I see the clocks go up to 1126mhz and volts to 1.05v. When the bench starts, the clocks go up to 1500mhz and the volts briefly go to 1.075v then to the full 1.21v. Bench completes no problem. If I clock above this, when the bench starts, the clocks go up to 1510mhz for example but the volts still step to 1.075v then to the full 1.21v. More often than not though the bench freezes at 1510mhz and 1.075v, causing a driver hang and reset. If the bench gets past the first couple of seconds the volts get all the way to 1.21v and the bench completes ok, I've benched up to 1600mhz successfully as long as it gets past the first couple of seconds.

A game I play a lot is Planetside 2 and the only crashes I see when overclocked are of the same variety. I'll be playing quite happily but as it doesn't require a lot of gpu power it'll often be running at 1126mhz and running 1.05v. Occasionally though there will be a big explosion on screen and the card will upclock to 1500mhz boost due to the load but the power state and volts will lag behind occasionally causing a driver reset or hard lock up at 1.075v again. This is very repeatable.

So, it seems my card is quite capable of running at a high mhz, but when the voltage lags behind due to the powerstates, it's causing me instability. as obviously 1.075v isn't enough to push 1500mhz or above.

I'm hoping the above gets fixed with a driver update and/or modded bios. I thought EVGA precision had a function called kboost which fixed clocks/voltages so I could test this but the option cannot be clicked when I try it.
Give Gigabyte OC Guru 2 a go , that has an option to offset the voltage minimum & maximum values.
 
I take your point but I can't hear it at all... if I take the side of the case off and get down right next to it with my ear almost touching the card sure, but with the side on it's totally inaudible. In fact if I'd never looked in this thread I expect I might well have never even known

If you can't hear it, why would you worry?

I'd only be concerned about coil whine if it was audible and annoying, under normal use.
 
Coil whine is just an unlucky by product of most electrical devices that contain chokes etc. It's just a fact of life.

While I feel for the folks that have really bad coil whine, in which case it's definitely RMA time... I think personally that a lot of folks are only hearing it because they've opened the case door on their PC and stuck their ear right next to the card. Pretty much all electrical devices (not just PC's) can make a noise under the right conditions.

As an example... one guy in another thread was complaining about coil while and getting hot under the collar about it. He then later admits that it's only audible when rendering at over 150 FPS, under this it's totally silent. A this point I was some what lost for words. Though the OP still seemed convinced that he had a "problem".

Another guy goes to the problem of recording the noise, case door open and mic right next to the card and I'll be damned (even with the sound turned up on my PC) that I could even hear anything other than a very small buzz. And when he moves back about a foot, totally inaudible over his case fans.

Not trying to dis-respect folks who have a serious problem here... just think some folks need to stand back and take a breath.

The problem with coil whine in graphics cards (and it's been said so many times... that I feel rather like I'm shouting into the wind)... is that you can take a card with apparent coil whine and put it into another system with a different PSU/mobo (and yes, the motherboard can contribute to it) and it could be a lot quieter, or even totally silent. Which is why RMAing a graphics card for coil whine is a tricky business (obviously not always). As if it tests OK, you are then lumbered with testing costs and two way delivery to get your card back.

As to why the 970/980's "appear" worse than other cards, not quite sure myself. But this sort of issue can somewhat be a self fulfilling prophecy, so to speak.

Take the MSI 970/980 gaming cards, my take is that these are pretty well built. Never had an issue with any MSI card before (and I've had a few), anecdotal evidence I know.

Not trying to put any folks down with real issues (as there quite obviously are a fair few) but I also think a fair few folks seem to be getting worried about nothing.
 
^^ Thanks for the response, that's what I was looking for :)

And to Kelt - I wasn't worrying because I could hear it, just wanted to be sure that the fact it was doing at at all wasn't indicative of some sort of problem with the card like for example it was going to deteriorate really quickly over time or something like that...
 
I take your point but I can't hear it at all... if I take the side of the case off and get down right next to it with my ear almost touching the card sure, but with the side on it's totally inaudible. In fact if I'd never looked in this thread I expect I might well have never even known


if your happy with it there is no problem then :rolleyes: ;)

it only matters to the people that can hear it and it annoys in the long run...:(

i guess there is even lots who just think its the norm :eek:
 
Back
Top Bottom