Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 15:11:26
(permalink)
Edome Great result Sajin, what were the clocks?
Same as my best run. 2063/13000.
|
Edome
iCX Member
- Total Posts : 271
- Reward points : 0
- Joined: 2017/10/14 13:31:21
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 15:19:34
(permalink)
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 15:23:07
(permalink)
|
Edome
iCX Member
- Total Posts : 271
- Reward points : 0
- Joined: 2017/10/14 13:31:21
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 15:30:14
(permalink)
Thank you. If the slider allows for +50%, then it's 450W.
|
Edome
iCX Member
- Total Posts : 271
- Reward points : 0
- Joined: 2017/10/14 13:31:21
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 15:34:36
(permalink)
Your minimum voltage during Graphics Test 2 Extreme is 1.063V? Could you quickly do 1 run with stable 1950MHz core/+500 memory and post the graphics score & especially Graphics Test 1 fps. Then I could compare it to my result with the same clocks and stock bios, and we can finally see how much faster the Kingpin PCB is clock-to-clock. Thank you!
post edited by Edome - 2017/11/28 15:56:10
|
the_Scarlet_one
formerly Scarlet-tech
- Total Posts : 24581
- Reward points : 0
- Joined: 2013/11/13 02:48:57
- Location: East Coast
- Status: offline
- Ribbons : 79
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 16:02:25
(permalink)
CptSpig CptSpig --- 7980XE@5.0 -- GTX TitanXp@2114/1620 --- 6180 TSE on chilled water to 8c. http://www.3dmark.com/spy/2763626
Happy?
Happy to see your run, sure. But the whole comparison was the 5960x paired with a 1080ti Founders, versus the K|ngp|n that Sajin has. Your 7980xe and Titan Xp better perform better since it has more GPU cores, and your have an extra 10 cores/20 threads on your cpu. Can't really compare your scores in a thread labeled " Fastest air/water cooled 1080 ti's on time spy extreme". That would be like someone saying "Fastest Ford Fiesta at the drag strip" and someone saying "look at the times my Ferrari is pulling"... This is my best run yet: The_scarlet_one--5960x@4.747--single 1080ti 2100/11906-- https://www.3dmark.com/spy/2829766 (not chilled.. Maybe I should hook up my phase change and get the CPU to -50?)
|
CptSpig
SSC Member
- Total Posts : 736
- Reward points : 0
- Joined: 2011/11/21 11:12:11
- Location: Sunny California
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 16:12:36
(permalink)
the_Scarlet_one
CptSpig CptSpig --- 7980XE@5.0 -- GTX TitanXp@2114/1620 --- 6180 TSE on chilled water to 8c. http://www.3dmark.com/spy/2763626
Happy to see your run, sure. But the whole comparison was the 5960x paired with a 1080ti Founders, versus the K|ngp|n that Sajin has. Your 7980xe and Titan Xp better perform better since it has more GPU cores, and your have an extra 10 cores/20 threads on your cpu. Can't really compare your scores in a thread labeled "Fastest air/water cooled 1080 ti's on time spy extreme". That would be like someone saying "Fastest Ford Fiesta at the drag strip" and someone saying "look at the times my Ferrari is pulling"... This is my best run yet: The_scarlet_one--5960x@4.747--single 1080ti 2100/11906--https://www.3dmark.com/spy/2829766 (not chilled.. Maybe I should hook up my phase change and get the CPU to -50?)
You missed the point. It's not a comparison I was being sarcastic. You wanted a score posted with the CPU-Z and GPU-Z data for verification. Time to lighten up laugh a little. Hook up the phase change see what she can do.......
post edited by CptSpig - 2017/11/28 16:16:20
|
the_Scarlet_one
formerly Scarlet-tech
- Total Posts : 24581
- Reward points : 0
- Joined: 2013/11/13 02:48:57
- Location: East Coast
- Status: offline
- Ribbons : 79
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 16:16:07
(permalink)
CptSpig You missed the point. It's not a comparison I was being sarcastic. You wanted a score posted with the CPU-Z and GPU-Z data for verification. Time to lighten up laugh a little.
I figured you were, but I won't be happy until you send me your chiller.. I will send you my phase change because I haven't used it in a while. :-) Seriously, you should take a glance in the marketplace section.. There is some good stuff that would you really unlock ever more potential for that 18 core.
|
CptSpig
SSC Member
- Total Posts : 736
- Reward points : 0
- Joined: 2011/11/21 11:12:11
- Location: Sunny California
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 16:22:34
(permalink)
the_Scarlet_one
CptSpig You missed the point. It's not a comparison I was being sarcastic. You wanted a score posted with the CPU-Z and GPU-Z data for verification. Time to lighten up laugh a little.
I figured you were, but I won't be happy until you send me your chiller.. I will send you my phase change because I haven't used it in a while. :-) Seriously, you should take a glance in the marketplace section.. There is some good stuff that would you really unlock ever more potential for that 18 core.
Your LN2 pots? I am happy with my chiller I can actually play games and bench mark while sitting down. I don't use the chiller for gaming I have a 360 rad with six fans for all that stuff. seriously you have to admit Sajin does a really good job with all his cards on air.
|
the_Scarlet_one
formerly Scarlet-tech
- Total Posts : 24581
- Reward points : 0
- Joined: 2013/11/13 02:48:57
- Location: East Coast
- Status: offline
- Ribbons : 79
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 16:30:02
(permalink)
CptSpig Your LN2 pots? I am happy with my chiller I can actually play games and bench mark while sitting down. I don't use the chiller for gaming I have a 360 rad with six fans for all that stuff. seriously you have to admit Sajin does a really good job with all his cards on air.
Lol, yeah, need to off load those pots. I haven’t been able to use them in a long time. My rig is still on 2x480mm fans for the cpu and 2x480mm + one 360mm for my GPUs. It’s quiet for sure. I wish I could get a chiller for benching or get my phase change installed and not have any issues, but I have been too lazy to hook it back up. His GPU is flat out awesome. Everything is my system, aside from gpu vram, is clicked higher and doesn’t compare to his gpu score and combined score.
|
CSN7
iCX Member
- Total Posts : 325
- Reward points : 0
- Joined: 2014/10/18 21:20:27
- Location: Germany
- Status: offline
- Ribbons : 6
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 17:29:14
(permalink)
|
Edome
iCX Member
- Total Posts : 271
- Reward points : 0
- Joined: 2017/10/14 13:31:21
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 17:56:09
(permalink)
Sajin's Graphics Test 1 fps is just out of this world (35.57 FPS, 2063/13000MHz) :) My results (330W MSI Trio stock bios power limit is enough for Test 1 not to cause clock reduction): pcie 3.0 8x 2100MHz 1093mV: 33.03 +900 mem 33.26 +920 mem 33.27 +940 mem 33.31 +960 mem 33.32 +960 mem 32.98 +1000 mem pcie 3.0 16x 2100MHz 1093mV: 33.46 +960 mem 33.01 +960 mem 32.88 +1000 mem 2063MHz 1075mV 32.92 +1000 mem 2063MHz 1075mV Sajin has 40MHz lower core clock, same memory clock but his fps is still 6% faster than my best score :) I think this confirms that the Kingpin PCB is simply faster.
|
Edome
iCX Member
- Total Posts : 271
- Reward points : 0
- Joined: 2017/10/14 13:31:21
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 18:31:49
(permalink)
Undervolted 1080ti SLI is just insane! The most awesome thing about this ultimate setup is that the power consumption is actually slightly less (total system ~ 420W) than with a max overclocked (XOC bios) single card (~ 500W). 9016 Graphics score, which is 75% more than my best single-card result. This 75% increase comes with 10x less noise as the fans can be used in whisper-quiet RPM (~1300 max) for the heat load is divided between the two cards. And the price of this was only 115% of a single Titan XP. Note that overclocked 1080ti SLI is only 10% faster (Graphics Score 10074), while the power consumption is doubled(!) causing serious fan noise. Full-size image: https://i.imgur.com/fRp96Ma.jpg
post edited by Edome - 2017/11/28 18:42:53
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 18:39:00
(permalink)
Edome Your minimum voltage during Graphics Test 2 Extreme is 1.063V? Could you quickly do 1 run with stable 1950MHz core/+500 memory and post the graphics score & especially Graphics Test 1 fps. Then I could compare it to my result with the same clocks and stock bios, and we can finally see how much faster the Kingpin PCB is clock-to-clock. Thank you!
Yes. Run with clocks as requested...
|
CSN7
iCX Member
- Total Posts : 325
- Reward points : 0
- Joined: 2014/10/18 21:20:27
- Location: Germany
- Status: offline
- Ribbons : 6
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 19:00:38
(permalink)
Sajin can you run the benchmark on the 'normal' or 'standard OC' bios?
|
Edome
iCX Member
- Total Posts : 271
- Reward points : 0
- Joined: 2017/10/14 13:31:21
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 19:09:15
(permalink)
Thanks Sajin! Time Spy Extreme Graphics Test 1 locked to 1950MHz/12000MHz (use AB V/F curve) results: Sajin Kingpin: 33.35 FPS My results: MSI: 30.45 FPS Palit: 30.24 FPS Kingpin is 10% faster! It's like Titan XP :) Someone with a normal 1080ti, post your result so we can compare :)
post edited by Edome - 2017/11/28 19:15:56
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 19:14:58
(permalink)
Snipes7 Sajin can you run the benchmark on the 'normal' or 'standard OC' bios?
No point as the stock normal, oc, & ln2 vbioses throttle the card hard once 117% PT is hit.
post edited by Sajin - 2017/11/28 19:42:34
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 19:17:31
(permalink)
Edome Thanks Saijin! Time Spy Extreme Graphics Test 1 locked to 1950MHz/12000MHz (use AB V/F curve) results: Sajin Kingpin: 33.35 FPS My results: MSI: 30.45 FPS Palit: 30.24 FPS Kingpin is 10% faster! It's like Titan XP :) Someone with a normal 1080ti, post your result so we can compare :)
Wonder what I could do with an xP. CPU doesn't affect the 4k graphic tests at all since the tests are gpu bound.
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 19:36:34
(permalink)
|
Edome
iCX Member
- Total Posts : 271
- Reward points : 0
- Joined: 2017/10/14 13:31:21
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 19:51:11
(permalink)
Sajin Wonder what I could do with an xP.
You've tried other 1080ti cards so you can confirm the kingpin PCB/VRM really gives a performance boost? You don't use any software tweaks like a secret nvidia profile inspector setting for 3dmark, windows clean boot etc? :)
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 19:55:28
(permalink)
I'd say my card works great. Will you get the same on different hardware? Maybe. Maybe not. If there are any nvidia inspector tweaks that work with 3dmark please let me know them so I can boost my score even higher. Windows clean boot isn't a tweak.
|
Edome
iCX Member
- Total Posts : 271
- Reward points : 0
- Joined: 2017/10/14 13:31:21
- Status: offline
- Ribbons : 1
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 19:58:52
(permalink)
Thanks for the info, I'm just trying to eliminate all the other factors. I haven't found any software tweaks, I've tried Nvidia profile settings and shutting down e.g. Afterburner during the benchmark, no effect. So I'd say there's a 95% chance that either the Kingpin VRM is superior or the memory timings in that bios are better. That leaves 5% for the chance that your other components like the motherboard is somehow especially fast :)
|
HeavyHemi
Insert Custom Title Here
- Total Posts : 15664
- Reward points : 0
- Joined: 2008/11/28 20:31:42
- Location: Western Washington
- Status: offline
- Ribbons : 135
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 20:33:52
(permalink)
the_Scarlet_one
CptSpig CptSpig --- 7980XE@5.0 -- GTX TitanXp@2114/1620 --- 6180 TSE on chilled water to 8c. http://www.3dmark.com/spy/2763626
Happy?
Happy to see your run, sure. But the whole comparison was the 5960x paired with a 1080ti Founders, versus the K|ngp|n that Sajin has. Your 7980xe and Titan Xp better perform better since it has more GPU cores, and your have an extra 10 cores/20 threads on your cpu. Can't really compare your scores in a thread labeled "Fastest air/water cooled 1080 ti's on time spy extreme". That would be like someone saying "Fastest Ford Fiesta at the drag strip" and someone saying "look at the times my Ferrari is pulling"... This is my best run yet: The_scarlet_one--5960x@4.747--single 1080ti 2100/11906--https://www.3dmark.com/spy/2829766 (not chilled.. Maybe I should hook up my phase change and get the CPU to -50?)
The strange thing is I beat your GPU score but you utterly decimate my CPU score. Those saying the CPU plays no role in the score are fooling themselves. https://www.3dmark.com/spy/2740261
EVGA X99 FTWK / i7 6850K @ 4.5ghz / RTX 3080Ti FTW Ultra / 32GB Corsair LPX 3600mhz / Samsung 850Pro 256GB / Be Quiet BN516 Straight Power 12-1000w 80 Plus Platinum / Window 10 Pro
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 20:37:54
(permalink)
HeavyHemi Those saying the CPU plays no role in the score are fooling themselves.
The time spy extreme tests are gpu bound. Yes, the cpu does affect the final total score.
|
HeavyHemi
Insert Custom Title Here
- Total Posts : 15664
- Reward points : 0
- Joined: 2008/11/28 20:31:42
- Location: Western Washington
- Status: offline
- Ribbons : 135
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 20:49:59
(permalink)
Sajin
HeavyHemi Those saying the CPU plays no role in the score are fooling themselves.
The time spy extreme tests are gpu bound. Yes, the cpu does affect the final total score.
Claiming a CPU doesn't affect 4K graphic GPU (benching) scores at all isn't true. While they are very highly biased toward the GPU, your CPU, particularly clocks has a impact on the GPU score.
EVGA X99 FTWK / i7 6850K @ 4.5ghz / RTX 3080Ti FTW Ultra / 32GB Corsair LPX 3600mhz / Samsung 850Pro 256GB / Be Quiet BN516 Straight Power 12-1000w 80 Plus Platinum / Window 10 Pro
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 21:14:05
(permalink)
HeavyHemi
Sajin
HeavyHemi Those saying the CPU plays no role in the score are fooling themselves.
The time spy extreme tests are gpu bound. Yes, the cpu does affect the final total score.
Claiming a CPU doesn't affect 4K graphic GPU (benching) scores at all isn't true. While they are very highly biased toward the GPU, your CPU, particularly clocks has a impact on the GPU score.
https://youtu.be/JonBbRx8Nuw?t=12m17s Listen until 12:47. Guess kingpin must be fooling himself.
|
HeavyHemi
Insert Custom Title Here
- Total Posts : 15664
- Reward points : 0
- Joined: 2008/11/28 20:31:42
- Location: Western Washington
- Status: offline
- Ribbons : 135
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 21:16:45
(permalink)
Sajin
HeavyHemi
Sajin
HeavyHemi Those saying the CPU plays no role in the score are fooling themselves.
The time spy extreme tests are gpu bound. Yes, the cpu does affect the final total score.
Claiming a CPU doesn't affect 4K graphic GPU (benching) scores at all isn't true. While they are very highly biased toward the GPU, your CPU, particularly clocks has a impact on the GPU score.
https://youtu.be/JonBbRx8Nuw?t=12m17s Listen until 12:47. Guess kingpin must be fooling himself.
Yes must be...I prefer just using facts https://www.3dmark.com/compare/spy/2831177/spy/2740261#All I changed here was CPU clocks. From 4.5ghz OC to the default which was at 4.0 during most of the test. I mean it's not like I been around and doing this for well over a decade.
EVGA X99 FTWK / i7 6850K @ 4.5ghz / RTX 3080Ti FTW Ultra / 32GB Corsair LPX 3600mhz / Samsung 850Pro 256GB / Be Quiet BN516 Straight Power 12-1000w 80 Plus Platinum / Window 10 Pro
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 21:19:45
(permalink)
HeavyHemi
Sajin
HeavyHemi
Sajin
HeavyHemi Those saying the CPU plays no role in the score are fooling themselves.
The time spy extreme tests are gpu bound. Yes, the cpu does affect the final total score.
Claiming a CPU doesn't affect 4K graphic GPU (benching) scores at all isn't true. While they are very highly biased toward the GPU, your CPU, particularly clocks has a impact on the GPU score.
https://youtu.be/JonBbRx8Nuw?t=12m17s Listen until 12:47. Guess kingpin must be fooling himself.
Yes must be...I prefer just using facts https://www.3dmark.com/compare/spy/2831177/spy/2740261# All I changed here was CPU clocks. From 4.5ghz OC to the default which was at 4.0 during most of the test. I mean it's not like I been around and doing this for well over a decade.
lol. your graphics tests barely changed.
|
the_Scarlet_one
formerly Scarlet-tech
- Total Posts : 24581
- Reward points : 0
- Joined: 2013/11/13 02:48:57
- Location: East Coast
- Status: offline
- Ribbons : 79
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 21:22:01
(permalink)
I think you two are talking about different things. Heavy is talking about the cpu score affecting the over all score. Sajin is saying the cpu and cpu speed is not holding the graphics score itself back.
By downclocking, heavy shows that the overall score drops a good bit, but his graphics score only drop 70 points overall, which is well within a margin of error.
Both of you are correct.
|
Sajin
EVGA Forum Moderator
- Total Posts : 49128
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: Fastest air/water cooled 1080 ti's on time spy extreme
2017/11/28 21:25:39
(permalink)
the_Scarlet_one I think you two are talking about different things. Heavy is talking about the cpu score affecting the over all score. Sajin is saying the cpu and cpu speed is not holding the gpu score itself back.
Both of you are correct.
I said the cpu doesn't affect 4k graphic tests in post #78. I also said the cpu does affect the final total score in post #84.
|