Jump to content

GTX 980 SLI LOW USAGE

Hey guys,i jst got my spanking new 980 sli setup and ran unigine heaven to check my scores.I got a score of 2100 with sli (1440p res and all settings maxed except for 3D).then i ran it without sli and got a score of 1900.Some thing was off so i ran heaven again while monitoring my gpu usage,without sli, my single 980 had a usage of 99%,With sli both cards stay at 95% but sometimes drop to the 80s.I tried overclocking my cpu (i5 4690k) to 4ghz(max i could do without increasing my voltage),and i still didnt see any performnce improvement,plz guys any help would be awesome.i have a corsair 650 watt 80+gold psu(forgot the name dont know if its CS or CX)


 


EDIT:i just check my cpu usage when running heaven and saw that it was at 30-35% so im gussing its not the bottle neck?


 


>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>EDIT<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<


YH SO OKAY SRY IT SEEMS IT WAS 1.3 VOLTS AND NOT 3.00 VOLTS IM AM SO FREAKING SRY GUYS :P


 


 


EDIT 2: I FINALLY DID IT I GOT MY CPU TO 4.5 GHZ AFTER 9 HOURS OF FIGHTING AN ENDLESS BATTLE WITH MY FREAKING H100I NO PERFORMANCE INCREASE WATSOEVER A LIL HELP PLZ.


Link to comment
Share on other sites

Link to post
Share on other sites

95-80% for SLI is not low lol. SLI does not scale perfectly...

Spoiler

CPU: AMD R7 5800X | CPU Cooler: Corsair H115i PRO | Motherboard: MSI B550-A PRO | Memory: G.Skill RIPJAWS V DDR4 3200mhz 64GB | GPU: EVGA RTX 3080Ti FTW3 Ultra | PSU: Seasonic Prime 1300w | OS Drive: Samsung 850 EVO SSD 500GB | Games Drive: Samsung 850 EVO SSD 1TB | Media Drive: 2x WD Blue HDD 1TB in Raid 0 | Media Drive: 2x WD Black HDD 2TB in Raid 1 | Case: Corsair Obsidian 750D | Monitor 1: ASUS PB287Q UHD | Monitor 2: ASUS PG278Q WQHD

 

Link to comment
Share on other sites

Link to post
Share on other sites

95-80% for SLI is not low lol. SLI does not scale perfectly...

There is no such thing as 100 percent yield ever...were seriously expecting that?

Everything you need to know about AMD cpus in one simple post.  Christian Member 

Wii u, ps3(2 usb fat),ps4

Iphone 6 64gb and surface RT

Hp DL380 G5 with one E5345 and bunch of hot swappable hdds in raid 5 from when i got it. intend to run xen server on it

Apple Power Macintosh G5 2.0 DP (PCI-X) with notebook hdd i had lying around 4GB of ram

TOSHIBA Satellite P850 with Core i7-3610QM,8gb of ram,default 750hdd has dual screens via a external display as main and laptop display as second running windows 10

MacBookPro11,3:I7-4870HQ, 512gb ssd,16gb of memory

Link to comment
Share on other sites

Link to post
Share on other sites

There is no such thing as 100 percent yield ever...were seriously expecting that?

dude i know that but my scores r trash :-( i got 1500 unigine heaven@1440p while other ppl r getting scores like 3k-3.5k

Link to comment
Share on other sites

Link to post
Share on other sites

I'd recommend overclocking your CPU a bit more with a tiny bit more voltage. 4.3-4.5 ghz should be achievable if you have atleast a halfway decent cooler.

A quad core can definitely hold 2 980's back in some scenarios.  My i5 4690 that I'm temporarily using actually bottlenecks my 780 TI classified in some games I see it shoot up to 95-100% cpu usage in heavy scenarios.

Stuff:  i7 7700k @ (dat nibba succ) | ASRock Z170M OC Formula | G.Skill TridentZ 3600 c16 | EKWB 1080 @ 2100 mhz  |  Acer X34 Predator | R4 | EVGA 1000 P2 | 1080mm Radiator Custom Loop | HD800 + Audio-GD NFB-11 | 850 Evo 1TB | 840 Pro 256GB | 3TB WD Blue | 2TB Barracuda

Hwbot: http://hwbot.org/user/lays/ 

FireStrike 980 ti @ 1800 Mhz http://hwbot.org/submission/3183338 http://www.3dmark.com/3dm/11574089

Link to comment
Share on other sites

Link to post
Share on other sites

I'd recommend overclocking your CPU a bit more with a tiny bit more voltage. 4.3-4.5 ghz should be achievable if you have atleast a halfway decent cooler.

A quad core can definitely hold 2 980's back in some scenarios.  My i5 4690 that I'm temporarily using actually bottlenecks my 780 TI classified in some games I see it shoot up to 95-100% cpu usage in heavy scenarios.

my cpu usage is ar 30-35% throughout the benchmark and i jst cant get my 4690k to 4.5 even if i get the voltage up it fails prime 95 instantly and,if i put in even more damage the temperature crashes it,my cooler is a corsair h100i

Link to comment
Share on other sites

Link to post
Share on other sites

my cpu usage is ar 30-35% throughout the benchmark and i jst cant get my 4690k to 4.5 even if i get the voltage up it fails prime 95 instantly and,if i put in even more damage the temperature crashes it,my cooler is a corsair h100i

 

 

What voltage were you using?

Stuff:  i7 7700k @ (dat nibba succ) | ASRock Z170M OC Formula | G.Skill TridentZ 3600 c16 | EKWB 1080 @ 2100 mhz  |  Acer X34 Predator | R4 | EVGA 1000 P2 | 1080mm Radiator Custom Loop | HD800 + Audio-GD NFB-11 | 850 Evo 1TB | 840 Pro 256GB | 3TB WD Blue | 2TB Barracuda

Hwbot: http://hwbot.org/user/lays/ 

FireStrike 980 ti @ 1800 Mhz http://hwbot.org/submission/3183338 http://www.3dmark.com/3dm/11574089

Link to comment
Share on other sites

Link to post
Share on other sites

What voltage were you using?

well i had to keep putting it up till the text became red(above 3.00 i think)

Link to comment
Share on other sites

Link to post
Share on other sites

well i had to keep putting it up till the text became red(above 3.00 i think)

 

 

Which voltage were you even changing that went up to 3.0?

Stuff:  i7 7700k @ (dat nibba succ) | ASRock Z170M OC Formula | G.Skill TridentZ 3600 c16 | EKWB 1080 @ 2100 mhz  |  Acer X34 Predator | R4 | EVGA 1000 P2 | 1080mm Radiator Custom Loop | HD800 + Audio-GD NFB-11 | 850 Evo 1TB | 840 Pro 256GB | 3TB WD Blue | 2TB Barracuda

Hwbot: http://hwbot.org/user/lays/ 

FireStrike 980 ti @ 1800 Mhz http://hwbot.org/submission/3183338 http://www.3dmark.com/3dm/11574089

Link to comment
Share on other sites

Link to post
Share on other sites

Which voltage were you even changing that went up to 3.0?

cpu core voltage

Link to comment
Share on other sites

Link to post
Share on other sites

cpu core voltage

 

 

If you put 3 volts through it you would of killed it instantly, are you sure it wasn't 1.30?

Stuff:  i7 7700k @ (dat nibba succ) | ASRock Z170M OC Formula | G.Skill TridentZ 3600 c16 | EKWB 1080 @ 2100 mhz  |  Acer X34 Predator | R4 | EVGA 1000 P2 | 1080mm Radiator Custom Loop | HD800 + Audio-GD NFB-11 | 850 Evo 1TB | 840 Pro 256GB | 3TB WD Blue | 2TB Barracuda

Hwbot: http://hwbot.org/user/lays/ 

FireStrike 980 ti @ 1800 Mhz http://hwbot.org/submission/3183338 http://www.3dmark.com/3dm/11574089

Link to comment
Share on other sites

Link to post
Share on other sites

If you put 3 volts through it you would of killed it instantly, are you sure it wasn't 1.30?

i got the clockspeed up to 4.5 and the voltage to 1.3 first,my pc insta crashed so increased voltage to 2.00 and prime 95 failed insta failed,so got voltage upto 3.00 once as a last effort and it crashed again so i jst quit overclocking.

 

my mobo is an msi z97 gaming 5 and the onboard overclocking utility overclocks  my cpu to 4ghz and i was able to run p95 for 2 hrs then i stopped.Should i use it?

Link to comment
Share on other sites

Link to post
Share on other sites

 

i got the clockspeed up to 4.5 and the voltage to 1.3 first,my pc insta crashed so increased voltage to 2.00 and prime 95 failed insta failed,so got voltage upto 3.00 once as a last effort and it crashed again so i jst quit overclocking.

 

my mobo is an msi z97 gaming 5 and the onboard overclocking utility overclocks  my cpu to 4ghz and i was able to run p95 for 2 hrs then i stopped.Should i use it?

If you were really oblivious enough to try to put 3v as core voltage on purpose, then you should probably not overclock for now. Your story is so wacky it almost seems like a troll. Even the most diehard overclockers I've seen in my overclocking thread only got up to like 1.6v and even that is utter insanity. What makes you think you can take 1.3v and then just double it and everything will be ok?

In Placebo We Trust - Resident Obnoxious Objective Fangirl (R.O.O.F) - Your Eyes Cannot Hear
Haswell Overclocking Guide | Skylake Overclocking GuideCan my amp power my headphones?

Link to comment
Share on other sites

Link to post
Share on other sites

i got the clockspeed up to 4.5 and the voltage to 1.3 first,my pc insta crashed so increased voltage to 2.00 and prime 95 failed insta failed,so got voltage upto 3.00 once as a last effort and it crashed again so i jst quit overclocking.

 

my mobo is an msi z97 gaming 5 and the onboard overclocking utility overclocks  my cpu to 4ghz and i was able to run p95 for 2 hrs then i stopped.Should i use it?

Umm... yeah.... reset your bios back to default please, then read a few OCing guides before you attempt to OC again. You are putting your hardware at serious risk with that kind of reckless abandon. I cant even find the nerve to go beyond 1.35V, let alone anywhere near what you have yours set at. Let us be thankful you are not running LinX, or any hardcore AVX stress tests, because i can assure you that your chip would not survive the outcome.

 

Take voltage down to something a bit safer, like 1.25v or 1.30v, and slowly raise your CPU multiplier. Stress test it after every 2 multiplier increments until you crash. When you do, back down x1 on the multiplier, then test that. If it is still unstable, go back to the last stable OC, and be satisfied. If that does not cut it for you, invest in a hardcore cooling solution, pump voltage to 1.35V and go for broke. Anything beyond that is suicide, and i wish you the best of luck.

My (incomplete) memory overclocking guide: 

 

Does memory speed impact gaming performance? Click here to find out!

On 1/2/2017 at 9:32 PM, MageTank said:

Sometimes, we all need a little inspiration.

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Did you reinstall your graphics drivers when you put the new card in?

 

Are you running any sort of frame limiter while you benchmark?

4K // R5 3600 // RTX2080Ti

Link to comment
Share on other sites

Link to post
Share on other sites

Umm... yeah.... reset your bios back to default please, then read a few OCing guides before you attempt to OC again. You are putting your hardware at serious risk with that kind of reckless abandon. I cant even find the nerve to go beyond 1.35V, let alone anywhere near what you have yours set at. Let us be thankful you are not running LinX, or any hardcore AVX stress tests, because i can assure you that your chip would not survive the outcome.

 

Take voltage down to something a bit safer, like 1.25v or 1.30v, and slowly raise your CPU multiplier. Stress test it after every 2 multiplier increments until you crash. When you do, back down x1 on the multiplier, then test that. If it is still unstable, go back to the last stable OC, and be satisfied. If that does not cut it for you, invest in a hardcore cooling solution, pump voltage to 1.35V and go for broke. Anything beyond that is suicide, and i wish you the best of luck.

 

If you put 3 volts through it you would of killed it instantly, are you sure it wasn't 1.30?

 

 

 

 

If you were really oblivious enough to try to put 3v as core voltage on purpose, then you should probably not overclock for now. Your story is so wacky it almost seems like a troll. Even the most diehard overclockers I've seen in my overclocking thread only got up to like 1.6v and even that is utter insanity. What makes you think you can take 1.3v and then just double it and everything will be ok?

 

WELL IT WAS 1.3 VOLTS AND NOT 3.00 VOLTS IM SO SRY GUYS

Link to comment
Share on other sites

Link to post
Share on other sites

Going over 1.35v is perfectly fine, and not even remotely close to suicide. Chips don't just spontaneously combust and explode. Chip death usually occurs anywhere from 1.75-1.9v depending on the chip and what cooling you have.

If you have very good cooling, 1.4v is fine for 24/7 use, and benchmarking sessions past 1.5 volts are perfectly fine if you have the cooling to deal with the heat. You'll also need a good motherboard with a good set of phases.

Stuff:  i7 7700k @ (dat nibba succ) | ASRock Z170M OC Formula | G.Skill TridentZ 3600 c16 | EKWB 1080 @ 2100 mhz  |  Acer X34 Predator | R4 | EVGA 1000 P2 | 1080mm Radiator Custom Loop | HD800 + Audio-GD NFB-11 | 850 Evo 1TB | 840 Pro 256GB | 3TB WD Blue | 2TB Barracuda

Hwbot: http://hwbot.org/user/lays/ 

FireStrike 980 ti @ 1800 Mhz http://hwbot.org/submission/3183338 http://www.3dmark.com/3dm/11574089

Link to comment
Share on other sites

Link to post
Share on other sites

Going over 1.35v is perfectly fine, and not even remotely close to suicide. Chips don't just spontaneously combust and explode. Chip death usually occurs anywhere from 1.75-1.9v depending on the chip and what cooling you have.

If you have very good cooling, 1.4v is fine for 24/7 use, and benchmarking sessions past 1.5 volts are perfectly fine if you have the cooling to deal with the heat. You'll also need a good motherboard with a good set of phases.

well i jst tried 4.4ghz at 1.2 volts and p95 was running but my 3 of my cores hit thermal limit,is this normal as i have a h100i and i expected better cooling im not even able to push 4.5 even though i have a h100i

Link to comment
Share on other sites

Link to post
Share on other sites

Did you reinstall your graphics drivers when you put the new card in?

 

Are you running any sort of frame limiter while you benchmark?

nope no fps limiter i also checked my fps during benchmarks at i hit 100-120 fps sometimes

Link to comment
Share on other sites

Link to post
Share on other sites

well i jst tried 4.4ghz at 1.2 volts and p95 was running but my 3 of my cores hit thermal limit,is this normal as i have a h100i and i expected better cooling im not even able to push 4.5 even though i have a h100i

Always remember overclocking performance varies wildly even in the same batch. Also check your thermal paste if you haven't already.

Everything you need to know about AMD cpus in one simple post.  Christian Member 

Wii u, ps3(2 usb fat),ps4

Iphone 6 64gb and surface RT

Hp DL380 G5 with one E5345 and bunch of hot swappable hdds in raid 5 from when i got it. intend to run xen server on it

Apple Power Macintosh G5 2.0 DP (PCI-X) with notebook hdd i had lying around 4GB of ram

TOSHIBA Satellite P850 with Core i7-3610QM,8gb of ram,default 750hdd has dual screens via a external display as main and laptop display as second running windows 10

MacBookPro11,3:I7-4870HQ, 512gb ssd,16gb of memory

Link to comment
Share on other sites

Link to post
Share on other sites

Always remember overclocking performance varies wildly even in the same batch. Also check your thermal paste if you haven't already. 

Well nw im on 4.4 ghz@1.220 vcore and aida 64 has been running for 30 mins with no crashes,can i do something to get my core clock upto 4.5?

Link to comment
Share on other sites

Link to post
Share on other sites

Going over 1.35v is perfectly fine, and not even remotely close to suicide. Chips don't just spontaneously combust and explode. Chip death usually occurs anywhere from 1.75-1.9v depending on the chip and what cooling you have.

If you have very good cooling, 1.4v is fine for 24/7 use, and benchmarking sessions past 1.5 volts are perfectly fine if you have the cooling to deal with the heat. You'll also need a good motherboard with a good set of phases.

 

I disagree somewhat. I mentioned investing in hardcore cooling solutions beyond 1.35v because of one specific reason: Running heavy AVX benches on these chips can generate far more heat than any real world usage scenario. It would be far worse if he had an i7, but luckily, he does not. I honestly would not attempt 1.4V on 90% of the air coolers available on the market, simply because they are not that great for that kind of thermal dissipation. That, and if you are going to throw $90 into a superb air cooling solution, you could just throw another $10-$20 into that budget and get a pretty great closed loop liquid cooling solution. 

 

Everyone has their line drawn in the sand, and it differs per person, but i draw mine at 1.35v personally. I've seen people on OCN run at 1.45V and say its perfectly fine to do so for a long time, i just personally think that the end does not justify the means. Eventually you hit that wall where you have to pump far more voltage just to get a slight clock change, and its not worth it to me.

 

Also, i did not say chips would explode. I said running LinX at 3.0V would put an end to that CPU. Which, unless it safely hit its TJ Max and safely shut itself down, would have been the outcome. Luckily, he was only running 1.3V, so none of the things i said applied to him. 

My (incomplete) memory overclocking guide: 

 

Does memory speed impact gaming performance? Click here to find out!

On 1/2/2017 at 9:32 PM, MageTank said:

Sometimes, we all need a little inspiration.

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

q

 

 

Booting above 2v will kill the chip, even on Ln2 people's chips are dying past 1.8v-1.9v. When I said very good cooling, I'm not talking about high end air coolers whatsoever, talking only about the highest end AIO's on the market, and custom loops.

And whilst running high voltages 24/7 like that, I'd never recommend running stress tests unless it's just once or twice to double check it's stable.

I only go over 1.4v for benching, usually don't run more than 1.4 for 24/7, even though I have a few friends that have ran 1.45v 24/7 for months and never had a single issue.

 

It's all personal preference though really, unless that jump from 1.35 to 1.45 is netting you a few hundred mhz, I wouldn't be running it.  IE: if 1.35v got me to 4.7 and 1.45v got me to 5, I'd do 1.45 no problem.

 

But if it was only like 4.8 to 4.9, nope.

Stuff:  i7 7700k @ (dat nibba succ) | ASRock Z170M OC Formula | G.Skill TridentZ 3600 c16 | EKWB 1080 @ 2100 mhz  |  Acer X34 Predator | R4 | EVGA 1000 P2 | 1080mm Radiator Custom Loop | HD800 + Audio-GD NFB-11 | 850 Evo 1TB | 840 Pro 256GB | 3TB WD Blue | 2TB Barracuda

Hwbot: http://hwbot.org/user/lays/ 

FireStrike 980 ti @ 1800 Mhz http://hwbot.org/submission/3183338 http://www.3dmark.com/3dm/11574089

Link to comment
Share on other sites

Link to post
Share on other sites

I disagree somewhat. I mentioned investing in hardcore cooling solutions beyond 1.35v because of one specific reason: Running heavy AVX benches on these chips can generate far more heat than any real world usage scenario. It would be far worse if he had an i7, but luckily, he does not. I honestly would not attempt 1.4V on 90% of the air coolers available on the market, simply because they are not that great for that kind of thermal dissipation. That, and if you are going to throw $90 into a superb air cooling solution, you could just throw another $10-$20 into that budget and get a pretty great closed loop liquid cooling solution. 

 

Everyone has their line drawn in the sand, and it differs per person, but i draw mine at 1.35v personally. I've seen people on OCN run at 1.45V and say its perfectly fine to do so for a long time, i just personally think that the end does not justify the means. Eventually you hit that wall where you have to pump far more voltage just to get a slight clock change, and its not worth it to me.

 

Also, i did not say chips would explode. I said running LinX at 3.0V would put an end to that CPU. Which, unless it safely hit its TJ Max and safely shut itself down, would have been the outcome. Luckily, he was only running 1.3V, so none of the things i said applied to him. 

 

Booting above 2v will kill the chip, even on Ln2 people's chips are dying past 1.8v-1.9v. When I said very good cooling, I'm not talking about high end air coolers whatsoever, talking only about the highest end AIO's on the market, and custom loops.

And whilst running high voltages 24/7 like that, I'd never recommend running stress tests unless it's just once or twice to double check it's stable.

I only go over 1.4v for benching, usually don't run more than 1.4 for 24/7, even though I have a few friends that have ran 1.45v 24/7 for months and never had a single issue.

 

It's all personal preference though really, unless that jump from 1.35 to 1.45 is netting you a few hundred mhz, I wouldn't be running it.  IE: if 1.35v got me to 4.7 and 1.45v got me to 5, I'd do 1.45 no problem.

 

But if it was only like 4.8 to 4.9, nope.

guys guys guys i GOT MY  4690K TO 4.5 GHZ MY TEMPS R LOOKING REALLY GOOD NEVER WENT ABOVE 75C (STILL RUNNING NW).IM GONNA LET IT RUN FOR 12 HOURS THEN ILL RUN HEAVEN AGAIN AND CHECK MY SCORE,THANKS TO U ALL I LUV U GUYS SO MUCH

 

P.S I HAVE A H100i(HAD TO REINSTALL IT AND NW SHITS LOOKING REAL GOOD)

Link to comment
Share on other sites

Link to post
Share on other sites

Booting above 2v will kill the chip, even on Ln2 people's chips are dying past 1.8v-1.9v. When I said very good cooling, I'm not talking about high end air coolers whatsoever, talking only about the highest end AIO's on the market, and custom loops.

And whilst running high voltages 24/7 like that, I'd never recommend running stress tests unless it's just once or twice to double check it's stable.

I only go over 1.4v for benching, usually don't run more than 1.4 for 24/7, even though I have a few friends that have ran 1.45v 24/7 for months and never had a single issue.

 

It's all personal preference though really, unless that jump from 1.35 to 1.45 is netting you a few hundred mhz, I wouldn't be running it.  IE: if 1.35v got me to 4.7 and 1.45v got me to 5, I'd do 1.45 no problem.

 

But if it was only like 4.8 to 4.9, nope.

 

Yeah, i am hitting a huge wall on my G3258. I can hit 4.3 ghz on 1.18V, but to get 4.4ghz, i have to go to over 1.28v, and it just does not seem worth it for that extra 100mhz. I would say running LinX for 5 passes at half of your memory capacity is enough to justify semi-stability in a system. Once you hit a point where LinX causes a blue screen, start a Prime95 blend test assuming your thermals are fine (under 80C for 24/7, this again varies per person, but under 80C is universally considered okay). I am sure i don't have to tell you how to do this, i am just giving out how i do things in case anyone else is wondering how to test stability quickly before dumping hours into prime95 after each and every multiplier change. LinX is probably the most stressful stress test i have encountered, and it gets my chips hotter than any other stress test. Its also very quick to run, compared to the several hours required in prime95. So passing LinX as a preliminary test is enough to let me go through several different multiplier adjustments before i have to hunker down with Prime95 for hours. 

 

 

guys guys guys i GOT MY  4690K TO 4.5 GHZ MY TEMPS R LOOKING REALLY GOOD NEVER WENT ABOVE 75C (STILL RUNNING NW).IM GONNA LET IT RUN FOR 12 HOURS THEN ILL RUN HEAVEN AGAIN AND CHECK MY SCORE,THANKS TO U ALL I LUV U GUYS SO MUCH

 

P.S I HAVE A H100i(HAD TO REINSTALL IT AND NW SHITS LOOKING REAL GOOD)

 

GRATS! Super happy to hear you are making progress in your OC, and are no longer having any issues. I hope you can get to the bottom of your GPU usage, as an overclocked i5 should be enough to handle those bad boys. Just understand that SLI can scale poorly on certain games, so do not expect 100% usage in both cards. 

My (incomplete) memory overclocking guide: 

 

Does memory speed impact gaming performance? Click here to find out!

On 1/2/2017 at 9:32 PM, MageTank said:

Sometimes, we all need a little inspiration.

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×