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 591

post #5901 of 10420
Quote:
Originally Posted by Dasboogieman View Post

Ahh I see, the Zotac VRM design looks similar to their 1080 (from what I can glean on the Official page), just more phases. If it matters, the MOSFETs on that edition had a weird 2 phase low side + 1 phase high side (i.e. 3 MOSFETs per VRM phase) setup. However the mosfets themselves are pretty cheapo individually (i.e. low efficiency) which means they need some seriously beefy cooling to get their optimal output. The 2 by 1 setup means your ultimate current capability is limited by the spike durability of the lone high side FET so thats the one that is going to blow if something goes wrong. As for how to get 16 + 2 phases, I have no idea how they're achieving that. 6 x 3? maybe? but triplers really weird but chances are, unless they got permission to use a different voltage reegulator, the actual regulation should be on par with the Aorus.

Also, I re-listened to the PCB breakdown again, I was wrong. The ASUS has a 5 x 2 +2 setup not a 6 x 1 +2. However, it is also very likely ASUS still has the best voltage regulation because their MOSFETs are godly, high efficiency and run at the same 300khz that all the others so far are running at, though MSI come a close second.
Aorus are running at 600 khz redface.gif. hell, i run my mobo's phases at 670 khz. is that bad?
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
post #5902 of 10420
Update your drivers to the latest hotfix, mine did this too and that fixed it.
Quote:
Originally Posted by Jsunn View Post

I am seeing some strange behavior on my 1080Ti FE.
I'll try to outline what I am seeing as best as I can.

I had my card mounted vertically using the Lian-Li riser card, you can see how I did it here:
Custom Vertical GPU mount Lian-Li PC-O11

I noticed that my benchmark scores were about 2,000-3,000 points lower in Fire Strike, and I randomly seemed to get a random hang upon reboot. The video posted below shows the behavior I was getting.

Then, somewhere along the line, upon boot, the BIOS and BIOS load screen (where you press F2 or Del to enter BIOS or UEFI) switched to my secondary monitor.

I thought that it might be that the bends in the riser cable were too tight, so I did some experimenting and thought I fixed it, I had more gradual bends and made sure that the cables weren't folded back and touching each other. But then I got the weird boot hang again. mad.gif

I decided to return to stock for now, but then I got it again. (the video posted is with the GPU installed in the motherboard, without the PCIe riser cable. )

I completely reset my BIOS, and everything seems to be back to normal, but I am still concerned that there might be something wrong. Can I bend the cable too sharply and cause problems?

Has anyone seen this behavior before? Is this a sign of something more serious?

Thanks,
Jason
post #5903 of 10420
Quote:
Originally Posted by SlimJ87D View Post

Sure I can test it tomorrow. But I don't think that's true about cpu and ram not affecting you at 4K, people says it "doesn't matter" because it literally makes like 2 to 4 fps of a difference which is a few percentages which makes people wave it off as it doesn't matter.

What are we doing here? We're epeening literally about a 0.009% performance difference and 1 or 2 frames.

Even if I test without a curve, we don't have the same hardware and OS where literally getting 0.5 fps gets you 100 points more and 0.0009% difference in overall score.
cpu and ram DEFINITELY influence fps at any resolution, including 4k. More draw calls = more fpa. yes, it's a lot less of a difference than in a cpu bound resolution. But, for instance, i get about ~120fps in BF1 in 1440p on my i-7 5775c at stock 3.3ghz with 1333mhz ram. But at 4.3Ghz and 3.7ghz uncore and 2133mhz Cas-9 ram, i get ~144+ fps. ANd who thinks a ryzen 4 core stock has the same fps in 4k as a 5.0Ghz 7700k? C'mon, guys, let's stop with this old argument. That like the guys who kept saying you'd get no fps gains in QHD going from a 4790k to a 5775c. Well, everyone that did it like myself, Benny, palotte, and a few others ALL report 10+ fps gains, just from a cpu drop in.


TL:DR - cpu and ram still affect performance at any resolution.
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
post #5904 of 10420
Quote:
Originally Posted by KedarWolf View Post


Less input lag limiting frame rates with no Vsync and Rivatuner and G-Sync on. smile.gif

 

Yes, I agree. But that's actually not what I was asking about, no pun intended :)

 

My question was:

 

Why then would you consider enabling VSYNC in NVCP part of the "optimal g-sync settings for single gpu/monitor" tagline? Or am I missing something here? The box specifically says GSYNC ON + VSYNC NVCP ON + VSYNC IN-GAME OFF + FPS LIMIT.

post #5905 of 10420
Quote:
Originally Posted by madpete76 View Post

Update your drivers to the latest hotfix, mine did this too and that fixed it.

Thank you! I will give it a try and post back.
I will also get everything plugged back up again and see if it works with the vertical GPU. Too many variables to try and track down the cause.
This weekend I installed the Win 10 latest update (hotfix addresses issues cause by this), the installed the riser card.
Edited by Jsunn - 4/18/17 at 11:57am
Glass Prism
(8 items)
 
  
CPUMotherboardGraphicsRAM
i7 7700k ASUS ROG Maximus IX Code NVidia GTX 1080Ti Founders Edition G. Skill TridentZ 
Hard DriveCoolingOSCase
Samsung 850 Pro EK-Kit X360 Windows 10 Pro Lian-Li PC-O11 
  hide details  
Reply
Glass Prism
(8 items)
 
  
CPUMotherboardGraphicsRAM
i7 7700k ASUS ROG Maximus IX Code NVidia GTX 1080Ti Founders Edition G. Skill TridentZ 
Hard DriveCoolingOSCase
Samsung 850 Pro EK-Kit X360 Windows 10 Pro Lian-Li PC-O11 
  hide details  
Reply
post #5906 of 10420
Quote:
Originally Posted by Slackaveli View Post

cpu and ram DEFINITELY influence fps at any resolution, including 4k. More draw calls = more fpa. yes, it's a lot less of a difference than in a cpu bound resolution. But, for instance, i get about ~120fps in BF1 in 1440p on my i-7 5775c at stock 3.3ghz with 1333mhz ram. But at 4.3Ghz and 3.7ghz uncore and 2133mhz Cas-9 ram, i get ~144+ fps. ANd who thinks a ryzen 4 core stock has the same fps in 4k as a 5.0Ghz 7700k? C'mon, guys, let's stop with this old argument. That like the guys who kept saying you'd get no fps gains in QHD going from a 4790k to a 5775c. Well, everyone that did it like myself, Benny, palotte, and a few others ALL report 10+ fps gains, just from a cpu drop in.


TL:DR - cpu and ram still affect performance at any resolution.

That's what I'm trying to tell him, but I think he just wants the crown with the highest core clock lol. We're literally comparing 0.5 FPS difference here, and this is where ram and CPU can make a difference.

If I literally got 1 more FPS on my superposition I would get an extra 200 points. I just down clocked my CPU, lost 0.5 fps and lost like 100 points on the test.

To non-Turbo nerds, 0.5 FPS is ridiculous subject to obsess about.

I need at 5775C! Just need to find a used one biggrin.gif
Edited by SlimJ87D - 4/18/17 at 12:25pm
post #5907 of 10420
Thread Starter 
Quote:
Originally Posted by kevindd992002 View Post

Quote:
Originally Posted by KedarWolf View Post

Less input lag limiting frame rates with no Vsync and Rivatuner and G-Sync on. smile.gif

Yes, I agree. But that's actually not what I was asking about, no pun intended smile.gif

My question was:

Why then would you consider enabling VSYNC in NVCP part of the "optimal g-sync settings for single gpu/monitor" tagline? Or am I missing something here? The box specifically says GSYNC ON + VSYNC NVCP ON + VSYNC IN-GAME OFF + FPS LIMIT.

G-Sync on, Vsync in nvidia control panel and game off, Rivatuner 1 FPS lower then your refresh rate. smile.gif
    
CPUMotherboardGraphicsRAM
Intel Core i7 5960X Asus Sabertooth X99 Gigabyte 1080 Ti Founders Edition Corsair LPX 
Hard DriveHard DriveCoolingOS
1.2TB Intel 750 PCI-E SSD 2TB Western Digital Black EKWB Predator 360 With One Titan X In Loop. Windows 10 Anniversary Edition 
MonitorKeyboardPowerCase
Acer XB280HK 4K Gsync QNIX QX2710 LED 27" WQHD 2560x1440 Corsair AX1500i Thermaltake Core X9 
MouseMouse Pad
MadKatz R.A.T. 8 Zowie 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel Core i7 5960X Asus Sabertooth X99 Gigabyte 1080 Ti Founders Edition Corsair LPX 
Hard DriveHard DriveCoolingOS
1.2TB Intel 750 PCI-E SSD 2TB Western Digital Black EKWB Predator 360 With One Titan X In Loop. Windows 10 Anniversary Edition 
MonitorKeyboardPowerCase
Acer XB280HK 4K Gsync QNIX QX2710 LED 27" WQHD 2560x1440 Corsair AX1500i Thermaltake Core X9 
MouseMouse Pad
MadKatz R.A.T. 8 Zowie 
  hide details  
Reply
post #5908 of 10420
Quote:
Originally Posted by KedarWolf View Post

G-Sync on, Vsync in nvidia control panel and game off, Rivatuner 1 FPS lower then your refresh rate. smile.gif

But what if you have OCD and don't like odd numbers?
post #5909 of 10420
Quote:
Originally Posted by SlimJ87D View Post

But what if you have OCD and don't like odd numbers?
The struggle is real... biggrin.gif
post #5910 of 10420
Quote:
Originally Posted by Dasboogieman View Post

Because some games are much more sensitive to overclocks than others. In fact, theres little correlation between how games are loading the GPU and how stable an overclock is. I personally found the FFXIV Free benchmark to be a really good indicator of stability (just loop the 4k at max settings), if you make it through an hr its truly "stable. Hell, I remember Ghost Recon Future Soldier used to crash on AIB cards unless you get rid of the factory overclock.

Lastly, get ur GPU colder, mine needs 100% fan speeds to keep temps under 70 otherwise my 2050 clocks are no longer stable

Oh its cold enough, never gets over 39degs. Shes under a big loop.
Taint3d
(20 items)
 
  
CPUMotherboardGraphicsRAM
4970k 4.8 EVGA z97 Classified EVGA 1080 Ti G.Skill 2400 
RAMRAMRAMHard Drive
G.Skill 2400 G. Skill 2400 G. Skill 2400 Samsung 840 Pro 
Hard DriveHard DriveCoolingOS
Samsung 840 Pro Samsung 850 Evo Custom Koolance/EK Watercooler Windows 10 
MonitorKeyboardPowerCase
Acer Predator X34 G-Sync @ 100hz Corsair Vengeance K70 EVGA supernova 1300 g2 Rosewill Blackhawk Ultra 
MouseMouse PadAudioOther
R.A.T 7 Corepad SoundBlaster Z Philips SPH 9500 Headphones 
  hide details  
Reply
Taint3d
(20 items)
 
  
CPUMotherboardGraphicsRAM
4970k 4.8 EVGA z97 Classified EVGA 1080 Ti G.Skill 2400 
RAMRAMRAMHard Drive
G.Skill 2400 G. Skill 2400 G. Skill 2400 Samsung 840 Pro 
Hard DriveHard DriveCoolingOS
Samsung 840 Pro Samsung 850 Evo Custom Koolance/EK Watercooler Windows 10 
MonitorKeyboardPowerCase
Acer Predator X34 G-Sync @ 100hz Corsair Vengeance K70 EVGA supernova 1300 g2 Rosewill Blackhawk Ultra 
MouseMouse PadAudioOther
R.A.T 7 Corepad SoundBlaster Z Philips SPH 9500 Headphones 
  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