i had just bought my gtx 1080 and replaced it with my gtx 780. once i got the 1080 installed i downloaded the latest driver off of geforce experience. after that i went into MSI Afterburner to overclock and i bumped the core clock to +180 and +400 on memory. i then ran some benchmarks and then after i played some games like Rainbow 6 Siege and using Heaven 4.0 benchmark, i had gotten the error that the display driver stopped working and had successfully recovered. this only happened after i finished and closed out my game and im not sure if thats normal when i have an unstable overclock. whats a safe core clock and memory clock to use without bumping the voltage at all?

my boost was 2088mhz when running heaven 4.0

i tried 120mhz and no crashes so far. it was starting to get annoying when my computer randomly restarted when i had the +180 OC. is that normal if computer restarts because of unstable OC? also my driver had an error even before. i had +50 core clock which was like 1800mhz boost and it crashed until i reinstalled 368.39 from website and reoverclocked to +120

MikeBGeyer
Pushing power target to 120 with fan speed toped out and +225 on clock plus +500 mem clock on the FE edition seems to be a pretty standard top out. I have seen many posts and benchmarks with those numbers. Anything above was unstable. I also have been running this configuration without issue.

Is that with stock voltage?

My room is 60f does that count for nitrogen?

thebest21445
i had just bought my gtx 1080 and replaced it with my gtx 780. once i got the 1080 installed i downloaded the latest driver off of geforce experience. after that i went into MSI Afterburner to overclock and i bumped the core clock to +180 and +400 on memory. i then ran some benchmarks and then after i played some games like Rainbow 6 Siege and using Heaven 4.0 benchmark, i had gotten the error that the display driver stopped working and had successfully recovered. this only happened after i finished and closed out my game and im not sure if thats normal when i have an unstable overclock. whats a safe core clock and memory clock to use without bumping the voltage at all?

I use +200GPU / +500MEM / +100mV / Power Limit 120%/ Custom fan curve to prevent temps above 70ºC (85% fan rotation at this temp). MEM can go up to +650MHz, but performance degrades.
 
In games it goes about 2076-2088MHz range on the GPU clock, no throtling at all (MSI AfterBurner 4.3.0 Beta 4 hardware monitoring).
 
Also uninstalled EVGA Precision X OC, these 6.0.0 and 6.0.1 versions are too much unstable, crashing every time. 

I did +235 core and +500 memory with 85% fan and 120% power. Stable. Nvidia FE. I think it boosts to 2126.
5,642 Firestrike Ultra

thebest21445
actually my problem was the actual gtx 1080. packaging had some damage and it might had damaged the gpu while shipping or packaging. no matter what i do, even going into regedit and doing all that file stuff, reinstalling graphics driver for like 8 times already and still nothing. going to send this back and buy a ftw card by evga or maybe classified

 

once i got the 1080 installed i downloaded the latest driver off of geforce experience. after that i went into MSI Afterburner to overclock and i bumped the core clock to +180 and +400 on memory. i then ran some benchmarks and then after i played some games like Rainbow 6 Siege and using Heaven 4.0 benchmark, i had gotten the error that the display driver stopped working and had successfully recovered.

 
First of all, uninstall GeForce Experience.  Follow the whole procedure of reinstalling your graphics drivers for the new card.  http://forums.evga.com/Basic-troubleshooting-steps-Windows-clean-boot-clean-driver-install-debug-mode-etc-m2486769.aspx .  Don't just install a new driver.  Don't use GeForce Experience.  And don't regedit random stuff because someone online said that it was a good idea.
 
Second of all, remove your overclock and test the card at stock speeds after doing a clean driver install.
 

jgonz
99.9 % of GTX 1080 hit 2 Ghz with little to no effort, that is 267 Mhz above advertised range. 
That is pretty darn good for what these cards are designed to do.

Yep, it could be better, but I´m very happy with 2025-2088MHz with no throtling at all.

EVGA GTX Hybrid

jeppouza

thebest21445
i had just bought my gtx 1080 and replaced it with my gtx 780. once i got the 1080 installed i downloaded the latest driver off of geforce experience. after that i went into MSI Afterburner to overclock and i bumped the core clock to +180 and +400 on memory. i then ran some benchmarks and then after i played some games like Rainbow 6 Siege and using Heaven 4.0 benchmark, i had gotten the error that the display driver stopped working and had successfully recovered. this only happened after i finished and closed out my game and im not sure if thats normal when i have an unstable overclock. whats a safe core clock and memory clock to use without bumping the voltage at all?

I use +200GPU / +500MEM / +100mV / Power Limit 120%/ Custom fan curve to prevent temps above 70ºC (85% fan rotation at this temp). MEM can go up to +650MHz, but performance degrades.
 
In games it goes about 2076-2088MHz range on the GPU clock, no throtling at all (MSI AfterBurner 4.3.0 Beta 4 hardware monitoring).
 
Also uninstalled EVGA Precision X OC, these 6.0.0 and 6.0.1 versions are too much unstable, crashing every time. 

Agreed Precision X OC is just not up to scratch atm. MSI AfterBurner for now.
Man, I wish I could overvolt to get a bit more but I'm too scared!

Don´t be afraid, I think FE can only go up to 1.08v with the actual voltage add , depending on the chip ASIC (which is bugged in GPU-Z BTW), my FE can´t pass from 1.05v I think, I´ll confirm this later.
 

EVGA GTX Hybrid

jeppouza

EVGA GTX Hybrid

jeppouza

thebest21445
i had just bought my gtx 1080 and replaced it with my gtx 780. once i got the 1080 installed i downloaded the latest driver off of geforce experience. after that i went into MSI Afterburner to overclock and i bumped the core clock to +180 and +400 on memory. i then ran some benchmarks and then after i played some games like Rainbow 6 Siege and using Heaven 4.0 benchmark, i had gotten the error that the display driver stopped working and had successfully recovered. this only happened after i finished and closed out my game and im not sure if thats normal when i have an unstable overclock. whats a safe core clock and memory clock to use without bumping the voltage at all?

I use +200GPU / +500MEM / +100mV / Power Limit 120%/ Custom fan curve to prevent temps above 70ºC (85% fan rotation at this temp). MEM can go up to +650MHz, but performance degrades.
 
In games it goes about 2076-2088MHz range on the GPU clock, no throtling at all (MSI AfterBurner 4.3.0 Beta 4 hardware monitoring).
 
Also uninstalled EVGA Precision X OC, these 6.0.0 and 6.0.1 versions are too much unstable, crashing every time. 

Agreed Precision X OC is just not up to scratch atm. MSI AfterBurner for now.
Man, I wish I could overvolt to get a bit more but I'm too scared!

Don´t be afraid, I think FE can only go up to 1.08v with the actual voltage add , depending on the chip ASIC (which is bugged in GPU-Z BTW), my FE can´t pass from 1.05v I think, I´ll confirm this later.
 

Thanks, I might have go as I've never OV'd a GPU before, I'll take it slow.  Yeah I did see GPU-Z is not showing the ASIC quality anymore.

Sorry, my mistake, voltages here are going above 1.05v on the GPU core and GPU-Z does not support ASIC read for Pascal yet:

 

orlando2542
I have the same +185 clock  on MSI but +550 on memory
 

I can go further with memory too, the max achieved were +625MHz on the slider, but I´ve noticed performance degradation instead.