Overclock.net › Forums › Graphics Cards › NVIDIA › NVIDIA GTX 1080 Ti Owner's Club
New Posts  All Forums:Forum Nav:

NVIDIA GTX 1080 Ti Owner's Club - Page 770

post #7691 of 10722
Hey thanks for reply! I also thought that because the sound was so mechanical, but turning the fans up or down manually and even stopping them with my finger didn't edit the noise at all, and the temperatures have never gone over 74c so that can't be causing the crashes too =/ I mean I most certainly could be wrong is there any way I can test it more definitively?
post #7692 of 10722
Quote:
Originally Posted by TheBoom View Post

I thought my Amp Extreme had no issues but I was wrong.

I didn't notice until now that my fans were constantly spinning even though GPU-Z, MSI AB and Firestorm all show that the fan is at 0%.

And this is at 45c idle temps. Custom curves do nothing same as auto or manual.

Anyone ever encounter something like this before?

I really hope it's a software issue tho I'd hate to have to RMA again.
I also have a Zotac Extreme and I've also observed some strange behaviour with its automatic fan control.
But I don't think it's hardware fault.
From what I have noticed, the fans start spinning at 50°c and are locked at 850rpm during 0% to 30%.
Thus preventing stop/start during low %.
Then they increase with the % (above 30%) as temperature rises as usual.
After the load, temp decreases and the fans stop at 40°c.
And you don't have control during idle, if try to set even 100%, the fans won't move.
But this is not the worse, because once I have noticed during benchmark when temperature was 75° that the fans rpm was at 0!
I checked and indeed they were stopped and forgot to start!
It can be really dangerous if you don't notice it!
I think Zotac has tried to do something smart to prevent start/stop behaviour during low load but has done something strange and buggy that I've never seen before...
post #7693 of 10722
Quote:
Originally Posted by Luckbad View Post

@TheBoom See my post below. Grab Zotac FireStorm and disable the fan freeze crap so you can actually control the card with custom curves.
Thanks!
I've switched this setting to OFF but I still don't have total control on the fans.
They are spinning at a constant 850rmp during idle, even if I choose to set 0%!
They are still "locked" below 30% at 850rmp.
I've never seen such behaviour with all my previous cards...
post #7694 of 10722
Ok, I have enough of those cards. I have no idea how do they work and why they do work and why they don't and when...

SAME Afterburner Settings, same curv
e:

1. Yesterday- played for 5 hours straight, temps 58-61. Fan speed 55%- Didn't go below 2050mhz on clock in Witcher 3 at all.
2. Today- played for an hour, Temp 63- crash. Ok, well, seems like temp is the key.
3. Played with higher fan speed- 70% to keep temps lower (after reboot ofc)- Clocks didn't break 57C. Downclocked to 2038 (same settings and curve!).
4. I think to myself- ok, maybe higher fan speed causes more power consumption (didn't get any prefs or higher TDP but whatever) and clocks downclock.
5. Ok, back to 55% Fan speed (same as pt 1 ) to confirm my theory. Card didn't yet break 58C- downclock to 2025.....

EVERYTHING ON SAME CURVE AND AB SETTINGS!!

I give up, I have enough. I have no idea how those stupid 1080Ti chips work... mad.gif
Tomb
(19 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 5775C 4.2 Ghz 1.36V Asus VII Hero ASUS GTX 1080 Ti STRIX OC 11 GB  G.Skill DDR3 16GB (2x8GB) TridentX 2133MHz 
Hard DriveHard DriveCoolingOS
SSD HyperX Savage 480GB SATA3 Seagate 3.5'', 2TB, SATAIII Corsair H105 Windows 10 Pro 
MonitorKeyboardPowerCase
XB271HU 1440p 165Hz Corsair K70 LUX RGB EVGA Supernova G2 1000 PHANTEKS Enthoo Evolv ATX Tempered Glass- Black. 
MouseMouse PadAudioAudio
Razer DeathAdder Chroma EXCO Scropion Athletics Edifier R1800BT SENNHEISER HD598 SPECIAL EDITION - BLACK 
OtherOtherOther
LC-Power LC-CFC-1 Fan Controller Xbox One One Gears of War 4 Crimson Omen specia... Playstation 4 Pro + Licensed PS4 Pad stand 
  hide details  
Reply
Tomb
(19 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 5775C 4.2 Ghz 1.36V Asus VII Hero ASUS GTX 1080 Ti STRIX OC 11 GB  G.Skill DDR3 16GB (2x8GB) TridentX 2133MHz 
Hard DriveHard DriveCoolingOS
SSD HyperX Savage 480GB SATA3 Seagate 3.5'', 2TB, SATAIII Corsair H105 Windows 10 Pro 
MonitorKeyboardPowerCase
XB271HU 1440p 165Hz Corsair K70 LUX RGB EVGA Supernova G2 1000 PHANTEKS Enthoo Evolv ATX Tempered Glass- Black. 
MouseMouse PadAudioAudio
Razer DeathAdder Chroma EXCO Scropion Athletics Edifier R1800BT SENNHEISER HD598 SPECIAL EDITION - BLACK 
OtherOtherOther
LC-Power LC-CFC-1 Fan Controller Xbox One One Gears of War 4 Crimson Omen specia... Playstation 4 Pro + Licensed PS4 Pad stand 
  hide details  
Reply
post #7695 of 10722
Quote:
Originally Posted by Slackaveli View Post

im tellin you, try to turn off HEDT in bios.

Tried it last night, no go man. I still get a slight micro stutter in my games. Like I tested witcher out more last night and when in the field it's about 95% smooth, but there are a few occasionsal hiccups. That's not to bad however until I ran into novigrad and it'd like the card is truly struggling to keep up with textures or something. It's not the worst stuttering I've ever seen been it's there, know what I mean?

It can go for stretches where there's no stuttering, but then it'll eventually happen and you turn the camera and it hangs for a split second but still might be hard for a non Pc gamer to notice.

I'm just really disappointed.. Because the Temps,, clocks all hold nicely but I still get the stutters. 60 is locked every game and I'm always over 2038 in games for stretches..but that micro stutter drives me nuts.

What makes me mad is this is the first time I jumped from evga, so it obviously makes me dislike gigabyte gpus now even more which sucks because I hate being that way. It would just figure I get a gpu that looks go run well but when you're playing it just hangs a bit.

Unless, do these cards still have slight issues pushing max 4k and a little stuttering here and there is to be expected? Or should I have zero stuttering outside maybe once or twice a session in well optimized games?
post #7696 of 10722
Quote:
Originally Posted by Gandyman View Post

Hey thanks for reply! I also thought that because the sound was so mechanical, but turning the fans up or down manually and even stopping them with my finger didn't edit the noise at all, and the temperatures have never gone over 74c so that can't be causing the crashes too =/ I mean I most certainly could be wrong is there any way I can test it more definitively?

What mode of payment did you use to pay for the card? If PayPal you should ask them (the retailer) if they would rather deal with PayPal disputes or simply take the card back.

Another option would be to dispute t with the credit card company and see how they handle it.

You could also buy another card from the scamy place and return the old faulty card as the one you had just purchased. Not sure how and if they track the serials they sell but I wouldn't think twice about doing this to places that refuse returns for what THEY precieve is a non issue. Not in 2017. Not on a product worth $700+ Places like this make me sick and if it was me, I would simply say the card is broken and doesn't work. Would not be past me to break the card on purpose by clipping a power connector to prove to them that it is in fact broken.

Where exactly did you buy the card so we know never to buy there? What is their policy?

The best bet may be to RMA the card with the manufacturer if you can get them to agree to an RMA. It may take a lot of explaining and proving but ultimately you would get a factory replacement. If they have a cross shipping option, that may be the best bet. You pay for the card they ship but get your money back when they receive the faulty card.
Edited by KraxKill - 4/29/17 at 9:07am
post #7697 of 10722
Quote:
Originally Posted by Benny89 View Post

Ok, I have enough of those cards. I have no idea how do they work and why they do work and why they don't and when...

SAME Afterburner Settings, same curv
e:

1. Yesterday- played for 5 hours straight, temps 58-61. Fan speed 55%- Didn't go below 2050mhz on clock in Witcher 3 at all.
2. Today- played for an hour, Temp 63- crash. Ok, well, seems like temp is the key.
3. Played with higher fan speed- 70% to keep temps lower (after reboot ofc)- Clocks didn't break 57C. Downclocked to 2038 (same settings and curve!).
4. I think to myself- ok, maybe higher fan speed causes more power consumption (didn't get any prefs or higher TDP but whatever) and clocks downclock.
5. Ok, back to 55% Fan speed (same as pt 1 ) to confirm my theory. Card didn't yet break 58C- downclock to 2025.....

EVERYTHING ON SAME CURVE AND AB SETTINGS!!

I give up, I have enough. I have no idea how those stupid 1080Ti chips work... mad.gif

I've given up on the curve. I reverted back to the old method of cranking the voltage and power slider, and then adjusting clocks that way. Since then I haven't had any stability issues reported in gpu z and gaming feels smoother. My timespy scores have also increased. I got 2038/6000 stable with 1.093 max on one card and 1.062 on the other. Never doing the curves again, it's almost impossible to get a curve to behave well in SLI when the cards are different.
Foxbox
(20 items)
 
  
CPUMotherboardGraphicsGraphics
i7 5820K @ 4.5 AsRock X99 Extreme4 GTX 1080Ti GTX 1080Ti 
RAMHard DriveHard DriveHard Drive
Corsair Vengeance LPX 16GB DDR4 DRAM 3000MHz Samsung 850 EVO 500GB SSD Samsung 850 EVO 500GB SSD Samsung 850 EVO 1TB SSD 
CoolingOSMonitorMonitor
Corsair H110 Windows 10 Pro OLED55C6P Dell S2417DG 
KeyboardPowerCaseMouse
Corsair K65 Corsair HX1200i  Corsair 780T Logitech G502 Proteus  
Mouse PadAudioAudio
OCZ HW-K550  Sennheiser PC 373D 
  hide details  
Reply
Foxbox
(20 items)
 
  
CPUMotherboardGraphicsGraphics
i7 5820K @ 4.5 AsRock X99 Extreme4 GTX 1080Ti GTX 1080Ti 
RAMHard DriveHard DriveHard Drive
Corsair Vengeance LPX 16GB DDR4 DRAM 3000MHz Samsung 850 EVO 500GB SSD Samsung 850 EVO 500GB SSD Samsung 850 EVO 1TB SSD 
CoolingOSMonitorMonitor
Corsair H110 Windows 10 Pro OLED55C6P Dell S2417DG 
KeyboardPowerCaseMouse
Corsair K65 Corsair HX1200i  Corsair 780T Logitech G502 Proteus  
Mouse PadAudioAudio
OCZ HW-K550  Sennheiser PC 373D 
  hide details  
Reply
post #7698 of 10722
Quote:
Originally Posted by Luckbad View Post

@TheBoom See my post below. Grab Zotac FireStorm and disable the fan freeze crap so you can actually control the card with custom curves.

Thanks for this.

+rep
Quote:
Originally Posted by Foxrun View Post

I've given up on the curve. I reverted back to the old method of cranking the voltage and power slider, and then adjusting clocks that way. Since then I haven't had any stability issues reported in gpu z and gaming feels smoother. My timespy scores have also increased. I got 2038/6000 stable with 1.093 max on one card and 1.062 on the other. Never doing the curves again, it's almost impossible to get a curve to behave well in SLI when the cards are different.

Exactly what I thought. I'm back to good ol manual oc too.



Ok no go on the fans. They're still stuck at 870 RPM on idle. Tried setting manual and curve in both Firestorm and AB and nothing.
Edited by TheBoom - 4/29/17 at 10:05am
post #7699 of 10722
Quote:
Originally Posted by Foxrun View Post

I've given up on the curve. I reverted back to the old method of cranking the voltage and power slider, and then adjusting clocks that way. Since then I haven't had any stability issues reported in gpu z and gaming feels smoother. My timespy scores have also increased. I got 2038/6000 stable with 1.093 max on one card and 1.062 on the other. Never doing the curves again, it's almost impossible to get a curve to behave well in SLI when the cards are different.

I only use the curve to control my top desired voltage bin if the voltage the card is pulling is a bin below where I want to be.

For example my card, even with the voltage slider all the way to the right (Afterburner beta) at +100% wants to run 2100 at 1.440-1.050v and is reluctant at selecting the 1.063v bin when temps are low. Thus, I pull my 1.063v bin down to meet 2100 in order to be able to use 1.063v there.

As far as the rest of the curve goes, it seems that the less you touch the better. I would also re-check the curve after reboot to make sure the desired max bin is still being selected appropriately. Afterburner tends to move the curve a fraction on reboot. At least sometimes from what I noticed.
post #7700 of 10722
Quote:
Originally Posted by pez View Post

That's not a terrible idea, but I'm not that bent on it, and ya know...morals redface.gif.
well, yeah. im not 100% what issue you have w/ your card. If just clocks (and you can get ~2Ghz), then, yeah. I wouldnt do it personally. I try not to abuse the system so it's there for me when i do need it. Caveat: company like newegg do exchange foe SAME item only. They could at least do a store credit where if you dont like the model you could add a few dollars and try a better one. Them limiting to same model kind of excuses the return for bad clocks morally in my eyes. Ya know? Shady is as shady does. In other words, if all im getting on a return is another same brand/model gpu, it better be able to do at least 2Ghz and not crash. And it better not have coil whine!
Edited by Slackaveli - 4/29/17 at 10:07am
EVA3.1
(9 items)
 
  
CPUMotherboardGraphicsRAM
Intel Broadwell i7-5775c MSI Z97m gaming Aorus Gtx 1080Ti Corsair Dominator Platinum 
Hard DriveCoolingOSMonitor
Adata ssd Asetek 240mm Win10 Samsung 8 series suhdtv  
Mouse
Logitech G602 
  hide details  
Reply
EVA3.1
(9 items)
 
  
CPUMotherboardGraphicsRAM
Intel Broadwell i7-5775c MSI Z97m gaming Aorus Gtx 1080Ti Corsair Dominator Platinum 
Hard DriveCoolingOSMonitor
Adata ssd Asetek 240mm Win10 Samsung 8 series suhdtv  
Mouse
Logitech G602 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: NVIDIA
Overclock.net › Forums › Graphics Cards › NVIDIA › NVIDIA GTX 1080 Ti Owner's Club