MSI Launches The GTX 980 Ti GAMING 6G

So should I replace my corsair hx1000 psu ? the buzz/coil noise comes from it 100% certain ( used water cooling tube to check) anything higher than 60 fps or so makes noise even with one card, which psu do you guys recommend that doesn't make noise at high fps ?

I think it being like 5 years old doesn't help..

You can never go wrong with Seasonic, but probably the best PSU for the dollar and performance is the 1300W G2, was for $165 after rebate the last I checked on egg.
 
The new Super Flowers are superior to the most recent Seasonics.
Haven't really seen anything ground-breaking from Seasonic in a while...

EVGA G2 series is the way to go.
 
I think this is the very highest I can do in Valley with throttling between 1554 and 1541 and only a couple dips below 1541

Oh man, you got me by 3 points. :p

29z6l3m.jpg
 
The G2 is made by Super Flower not Seasonic. It's a top oem for power supplies. I just replaced my PSU with the 850w G2 and its fixed my instability and crashes. After a while, capacitors just go. My old PSU was fine with the 670 but the 980 TI just choked it.

I wonder if it's worth it to give it more volts over stock to try for that magic 1500 barrier?
 
So should I replace my corsair hx1000 psu ? the buzz/coil noise comes from it 100% certain ( used water cooling tube to check) anything higher than 60 fps or so makes noise even with one card, which psu do you guys recommend that doesn't make noise at high fps ?

I think it being like 5 years old doesn't help..

Is it modular? Are you able to test a different pci connector rail? My PSU has 4. But yeah its probably getting on, i bought my bro a Tt 750w gold PSU and it started buzzing after 2 years then blew. Of course that was an RMA issue though, 2 years with an i5 and 1 550ti should of held longer.
 
I got a EVGA G2 1300 believe it was like $150 AR last year been great replaced a Corsair 850 that I used in another system now.
 
So should I replace my corsair hx1000 psu ? the buzz/coil noise comes from it 100% certain ( used water cooling tube to check) anything higher than 60 fps or so makes noise even with one card

PSUs are causing much of this coil whine. People just don't realize it, don't want to admit it, or get giggles out of blaming the cards.
 
Is it modular? Are you able to test a different pci connector rail? My PSU has 4. But yeah its probably getting on, i bought my bro a Tt 750w gold PSU and it started buzzing after 2 years then blew. Of course that was an RMA issue though, 2 years with an i5 and 1 550ti should of held longer.

It's semi modular but has two 500w rails, tried to swap some cables but only made the melody change tune..

Thanks for the tips guys, was eyeing the corsair ax1200i but it seems the evga 1300w g2 is also nice and nearly half the price, gonna grab that one !
 
Oh man, you got me by 3 points. :p

29z6l3m.jpg

At 1537 I scored 4420, a bump to 1557 and more on memory yielded a 4200 score. Regardless, 1537+volts I'm throttling for some reason in Witcher 3, volts or TDP.. even though 109% isn't being touched. These benchies are pointless sadly.. makes me think the reviews standards of stable is skew too. No throttling on either setting btw w/my G10. These 980Ti's are... hard to understand haha.

I scored 4374 on my daily overclock of +0mV/+135/+200 for 1477/3703 for an interesting reference. I'm very happy with my 980Ti from MSI, great purchase after G10, before I would 80C+ 100% fan speed in witcher3.
 
Last edited:
At 1537 I scored 4420, a bump to 1557 and more on memory yielded a 4200 score. Regardless, 1537+volts I'm throttling for some reason in Witcher 3, volts or TDP.. even though 109% isn't being touched. These benchies are pointless sadly.. makes me think the reviews standards of stable is skew too. No throttling on either setting btw w/my G10. These 980Ti's are... hard to understand haha.

I scored 4374 on my daily overclock of +0mV/+135/+200 for 1477/3703 for an interesting reference. I'm very happy with my 980Ti from MSI, great purchase after G10, before I would 80C+ 100% fan speed in witcher3.
That's probably because you have the texture filtering in the Nvidia control panel on the default setting of quality instead of high quality.
 
Well im prepared for this beast i think. I set my rig back to a more power friendly mode given the horsepower of this card should be great. My 2400 RAM doesnt work unless i have high consistency in voltage, but 2133 runs great ar any level. My 3960x is at 4.4ghz, eist enabled again and getting 27c idle, 50c in gta v with the 7970. I got anxious and done some ocing on the 7970 again lol, damn Gigabyte for locking the voltage to 1.256, i know it can do higher.

Hope its not a sign of bad things to come from the MSI, but my exhaust fan died, well the CHA2 header died, worked again on the CHA3 header. Damn im getting anxious, only half a day til arrival hopefully.
 
I put the G10 and H90 on my card but I am having some issues. Before with the stock cooler I was stable at 1466 for hours. Now, I will get a driver crash after about an hour with the same overclock. My core temps in AB cap out around 60-61C with my custom fan curve.

My setup is: the H90 is exhausting out the back fan port with a Noctua A14 PWM fan. I have 2 more of the noctua in the front of my case doing intake. On the G10 bracket I have a Noctua A9 92mm fan that I can run 100% without a sound on the VRM. I placed copper heatsinks on the memory midplate I have the H90 pump plugged directly into the PSU via molex but I have tried it on a motherboard fan header.

I thought maybe the VRM were overheating so I ghetto installed a spare corsair 140mm fan into the bottom of the case blowing onto the card but still the same behavior. The temp probe I had laying around was not working so I have a new one coming in tomorrow that I am going to use to check VRM temps. I just have a hard time believe the VRM are getting worse airflow now that with the stock cooler. Could this all be because I am not seated very well on the GPU core even though temps are good? I can clock up to 1550 and it runs fine for about 2 minutes then driver crash like it is overheating. The backplate also gets pretty hot but the VRM and memory midplates are warm but not scorching.

I had the H90 hooked up to a motherboard head and it showed ~1500 rpm at max....so I am guessing it is not the h90 pump?? I have tried with all fans at 100% but while it may prolong the crash it eventually happens. Could really use some help on this one as it is my first time using the G10.
 
It's semi modular but has two 500w rails, tried to swap some cables but only made the melody change tune..

Thanks for the tips guys, was eyeing the corsair ax1200i but it seems the evga 1300w g2 is also nice and nearly half the price, gonna grab that one !

Personally, I have an XFX 1200W PSU. Zero coil whine anywhere. If there is coil whine, it's definitely due to the video card or other component (in my case, my old EVGA GTX 570 which coil whined).
 
I put the G10 and H90 on my card but I am having some issues. Before with the stock cooler I was stable at 1466 for hours. Now, I will get a driver crash after about an hour with the same overclock. My core temps in AB cap out around 60-61C with my custom fan curve.

My setup is: the H90 is exhausting out the back fan port with a Noctua A14 PWM fan. I have 2 more of the noctua in the front of my case doing intake. On the G10 bracket I have a Noctua A9 92mm fan that I can run 100% without a sound on the VRM. I placed copper heatsinks on the memory midplate I have the H90 pump plugged directly into the PSU via molex but I have tried it on a motherboard fan header.

I thought maybe the VRM were overheating so I ghetto installed a spare corsair 140mm fan into the bottom of the case blowing onto the card but still the same behavior. The temp probe I had laying around was not working so I have a new one coming in tomorrow that I am going to use to check VRM temps. I just have a hard time believe the VRM are getting worse airflow now that with the stock cooler. Could this all be because I am not seated very well on the GPU core even though temps are good? I can clock up to 1550 and it runs fine for about 2 minutes then driver crash like it is overheating. The backplate also gets pretty hot but the VRM and memory midplates are warm but not scorching.

I had the H90 hooked up to a motherboard head and it showed ~1500 rpm at max....so I am guessing it is not the h90 pump?? I have tried with all fans at 100% but while it may prolong the crash it eventually happens. Could really use some help on this one as it is my first time using the G10.

Double checked your powers good? Dodgy cable maybe? And just curious, does that h90 go any higher than 1500 on the pump? My shotty Tt pump always goes 2600rpm. 1500 seems a bit low but if you say temps are good then meh.
 
Double checked your powers good? Dodgy cable maybe? And just curious, does that h90 go any higher than 1500 on the pump? My shotty Tt pump always goes 2600rpm. 1500 seems a bit low but if you say temps are good then meh.

Yeah power is good. I just took out the card and tightened the screws for the AIO it I seem to have dropped about 5-7C so maybe it was not tight enough. I can see the pcb bending slightly on the card now. Going to run it for the next couple days and see how it reacts.

I actually got the temp sensor in today so I put it under the VRM plate as close as possible to the actual vrm modules. At stock settings it read 53C, given I am not directly on the VRM I feel comfortable saying it is probably under 60C which should not trigger crashes to desktop.



My H110 and H90 both say ~1500-1550 rpm in asus fan xpert when I had them plugged into the motherboard ports.
 
That's probably because you have the texture filtering in the Nvidia control panel on the default setting of quality instead of high quality.

You mean the initial screen of the spinning logo? Let me see if that has an effect.
 
You mean the initial screen of the spinning logo? Let me see if that has an effect.
No. Its in the manage 3D settings and then global settings. The fact that you were not sure where its at pretty much confirms its on the default setting.:p

Most people never touch that so they end up with tiny bit higher scores in some benchmarks.
 
Whew, yeah I set it to performance and my score rocketed up to almost 4,900. This is at 1577Mhz, +500 on memory. 71.1% ASIC.

 
Last edited:
I'm thinking about installing a 3rd party cooler for this. AIO watercooler is a no go due to crappy pump noise, regardless of the model and brand. Accelero Xtreme iv may be the candidate, since it can bring even a titan x to low 50's.
 
Thanks to Dumb-Hopeless-Lost global mail im waiting another day for my card. (Says fuck 5x)
 
Uhh, the pump noise from my H75 is..well barely anything. The fans are louder.
 
I'm thinking about installing a 3rd party cooler for this. AIO watercooler is a no go due to crappy pump noise, regardless of the model and brand. Accelero Xtreme iv may be the candidate, since it can bring even a titan x to low 50's.

Yeah im not sure what you mean here, sounds like you got a shit pump. Even my shotty Tt pump doesn't make a sound lol (Rad fans whine like a babys rattle though over 50% lmao). That cooler looks nice, while I don't have any experience with it it does says it's rated to 300w so I can't imagine it can be that great in comparison to what I've seen with the AIB coolers. Personally I'm really happy I went with the MSI, bought it while it was 669 at BHPhoto, and here in Aus the stockers are charging 1100 for it (In AU whereas I paid under 900). Hopefully tomorrow I'll get it in the morning if Dumb-Hopeless-Lost get there shit together. Been sitting at the Brisbane sorting center since Sunday and there are 20 flights a day up here from there.
 
Does anyone know the criteria for throttling with these 980 Ti's? I hit 1500 at +87 volts, but it seems to throttle back after 30 minutes. The temperature is 75 so it's fine. I may reapply my custom BIOS of 120% power limit to see if if it's throttling due to 109% limit.
 
Does anyone know the criteria for throttling with these 980 Ti's? I hit 1500 at +87 volts, but it seems to throttle back after 30 minutes. The temperature is 75 so it's fine. I may reapply my custom BIOS of 120% power limit to see if if it's throttling due to 109% limit.

You answered your own question, the MSI's are TDP limited no matter what BIOS you throw at it, so if you increase the power limit, the the TDP hits sooner. There is a big discussion about this at OCN
 
Last edited:
You answered your own question, the MSI's are TDP limited no matter what BIOS you throw at it, so if you increase the power limit, the the TDP hits sooner. There is a big discussion about this at OCN

Do you have the link? 120% is well under the TDP of 75+150+150. 120% of 275 is only 330.
 
And that makes the third brand of card ive bought that is voltage locked. 2x 7970, one gigabyte and one (well 2) asus and now an MSI Nvidia card.

Will be here within an hour or 2 if AUS-terity-Post is to be believed.
 
Whew, yeah I set it to performance and my score rocketed up to almost 4,900. This is at 1577Mhz, +500 on memory. 71.1% ASIC.


Man, how are you guys getting those scores? I have a MSI 980 Ti running +145 / +400 (67.2% Asic) and my scores are consistently 1400 less than those, average FPS of 80 or so. Rest of my rig are in specs, but running an i7-875k at 4GHz. I'm not thinking the faster core you are running is going to make up that difference in frame rate, so I'm guessing I'm messing something up somewhere...
 
Man, how are you guys getting those scores? I have a MSI 980 Ti running +145 / +400 (67.2% Asic) and my scores are consistently 1400 less than those, average FPS of 80 or so. Rest of my rig are in specs, but running an i7-875k at 4GHz. I'm not thinking the faster core you are running is going to make up that difference in frame rate, so I'm guessing I'm messing something up somewhere...
Maybe read what he just said? He changed the Nvidia control panel settings to performance which degrades graphics quality.
 
I have to go shopping with the mrs, but i did a quick video of the unboxing that I'll upload a bit later. My card is slightly bent at the end as well, just enough so that the rubber contacts (i believe they're rubber?) Dont touch those sinks at the end.

One thing ive noticed coming from AMD is that this nvidia card scales to my VG278H perfectly and it almost seems i have a little more real estate than the 7970. Im not sure why this is tbh.
 
Just quickly tested it before i go on default oc mode from the gaming app, in the little gpuz render test the card boosts to 1392/1775 @65c but only 35% fan which i cant hear at all. Might just do a quick test of the fan at 100% and see how it sounds. Also ASIC quality is 74.4%

Edit - well the fan thing aint working in afterburner but it got to 61% over 5 minutes in the little render test. Cant hear a thing even when i put my ear next to it. Those heat pipes on the side got fucking hot but, was reading 60c but it was almost untouchable. Have to go now, ill have a play later. Also forgot nvidia still on pci 2.0 lol not that it makes any difference for the current gens.
 
Last edited:
Those heat pipes on the side got fucking hot but, was reading 60c but it was almost untouchable.

Yep, my observations too. The first time I touched it it seared my fingers so I almost got blisters. I was quite surprised it was so hot...but that means it's doing it's job, no? :p
 
Maybe read what he just said? He changed the Nvidia control panel settings to performance which degrades graphics quality.

Maybe I need to be more explicit, with the same settings (ie Texture filtering at High Performance), I get 1400 less. If I run it with Quality, I get 1400ish less than the people running it on Quality.
 
Last edited:
Everybody fucking hits 1500 except me lol.

I am right there with you buddy. The best I can do is 1477 with +27mV and even that doesn't hold after about an hour of gaming/stressing. Sometimes you just don't get lucky with the silicon lottery lol. My ASIC is 65.4%. My temps are stable around 55C at load and I pretty much stick to 1466/7500. Doesn't matter how much voltage I throw at it, 1500 just isn't in my card. My sensor on VRM gets to 54C after 2 hours of Heaven benchmark running and holds off there.

At least my system is nice and quiet with the G10 and everything. Also I installed mine a little different without the use of the NZXT backplate. I used the G10 screws with some washers to hold the screws off the MSI backplate. After I had it tightened down all the way, there hasn't been any trouble except when I tried to get over 1500.
 
Forgive me for asking this question (again,) but I can't make a decision. Is this is the quietest air cooled 980 Ti on the market, or does that distinction belong to a different one?
 
Back
Top