TR Forums

SoM · Mon Feb 26, 2018 11:38 pm

was time to dust off the PC, removed the GPU, blew out the case, put everything back, works OK.

decided to once again try intel 530 gfx since i'm planning on selling the 1070 but wasn't impressed as expected so i reinstalled the 1070 and win10 took almost 5min to load. normal is 15 seconds

updated to latest 391.01 gfx drv just incase. also noticed my disk transfer rate is constantly doing read/write with high spikes i've never had before, also re-seated the GPU few times, check all connections etc..

i'm going for a walk

Mon Feb 26, 2018 11:55 pm

100 KB/sec is not high at all. That's like the speed of a floppy disk.

What's the boot device in this system?

SoM · Tue Feb 27, 2018 12:07 am

samsung 950pro

just happened to be on 100, it fluctuates, now it's 7mb-10mb

path of exiles takes a bit to load, never happend
quake champions loads slowly, but get black screen before seeing character, game responds just black screen, alt-tab works.

Tue Feb 27, 2018 12:15 am

7-10MB is not high either, especially for a SSD.

What does your CPU usage look like?

K-L-Waster · Tue Feb 27, 2018 7:51 am

I'd reseat the connector for the SSD (sorry, I'm drawing a blank at the mo' as to whether that's a SATA or an M2 SSD... but same suggestion regardless).

SoM · Tue Feb 27, 2018 9:51 am

just brew it! wrote:
7-10MB is not high either, especially for a SSD.

What does your CPU usage look like?

only FF 58.0.2 running, the CPU should be idle at 800mhz for one

SoM · Tue Feb 27, 2018 9:52 am

K-L-Waster wrote:
I'd reseat the connector for the SSD (sorry, I'm drawing a blank at the mo' as to whether that's a SATA or an M2 SSD... but same suggestion regardless).

you could read my sig

SoM · Tue Feb 27, 2018 10:08 am

just rebooted, win10 booted fast, 15sec.

DancinJack · Tue Feb 27, 2018 10:53 am

So everything seems fine to me? CPU's, especially newer ones, are gonna clock up and down all the time. It's not likely just gonna sit at 800MHz even with just FireFox running.

edit: for instance, my 6700K just went from 1.2GHz to 3.33GHz (according to Task Manager) and the only thing running was Chrome with two tabs of just text/image web pages.

SoM · Tue Feb 27, 2018 1:05 pm

did another dust off, and also cleaned the contacts on the GPU/RAM/M.2

win loads fast, tried running Quake Champions, after the bethesda game launcher the game starts but sits at a black screen and i've noticed the fans don't even turn on the 1070,

edit; tried Just Cause 3 and GPU fans don't turn and it's choppy, GPU not warming up.

DPete27 · Tue Feb 27, 2018 1:50 pm

Based on the variables you listed, it seems the graphics driver would be my first suspect.
Whenever something like this happens, always try to get back to last known working condition (hardware and software), then change single variables at a time until the problem presents itself again.

Please think through your process:
1) Removed GPU - did you uninstall the Nvidia drivers to transition to IGP, or did you leave the [known working] Nvidia driver installed?
2) Put everything back - Everything? The only thing you said you removed was the GTX1070.
3) Wasn't impressed - no boot time issues yet at this point?
4) Reinstalled the GTX1070, took 5 minutes to load Windows - No other hardware/software changes since step 3?
5) Updated gfx drivers - did this fix anything?
6) Disk transfer spikes - Did you notice this before updating the gfx driver?
Any Windows updates during this process?

The fans on your GTX1070 (and most/all modern GPUs) are semi-passive. They won't spin until the GPU is at a certain temp. (Assuming a decent heatsink) If you're running a game/load that's not terribly demanding, the fan may not ever spin. If you have good airflow through your case, or the side of the case is still off, the fan may not spin. Etc etc.

SoM · Tue Feb 27, 2018 3:04 pm

1. no, just removed GPU, like before, no problem
2.yes, reconnect 2 6x2 pin cables and case cables
3.boot is back to normal 15sec

i've rolled back my driver to 390.77 no improvement.

yes i know about the fans spinning/not spinning, they should spin at POST and turn off once i'm all loaded. Quake Champions is a heavy game, even the bethesda game launcher can put some stress on the card starting the fans, they should be spinning at this point, their not.

here's a quote from someone on the geforce forums but does not pertain to me

You'll definitely need to RMA it or return it and get another. A fan not spinning at all is most likely a faulty controller on a hardware level. Just to confirm, on first boot before Windows even loads, it still does not spin right? If it DOES then perhaps we can troubleshoot. If not, RMA.

Tue Feb 27, 2018 4:26 pm

As a last resort, you might try Display Driver Uninstaller to get to a completely blank slate before reinstalling the latest graphics-card drivers: http://www.guru3d.com/files-details/dis ... nload.html

You can also use the sensor readouts in GPU-Z to get a sense of how your card's clocks, power limits, etc. are behaving under load. That information might allow you to figure out what the card isn't doing, at least: https://www.techpowerup.com/gpuz/

MOSFET · Tue Feb 27, 2018 4:32 pm

That's the first time in years I've seen someone use the combined CPU view in task Manager. YEARS. Per-thread FTW.

Can you get a GPU fan to spin manually? And you're positive the PSU-GPU is hooked back up solidly?

You messed up the equilibrium by cleaning. Don't do that ! Monitor, monitor, monitor (utilities, not screens)

SoM · Tue Feb 27, 2018 4:53 pm

Jeff Kampman wrote:
As a last resort, you might try Display Driver Uninstaller to get to a completely blank slate before reinstalling the latest graphics-card drivers: http://www.guru3d.com/files-details/dis ... nload.html

You can also use the sensor readouts in GPU-Z to get a sense of how your card's clocks, power limits, etc. are behaving under load. That information might allow you to figure out what the card isn't doing, at least: https://www.techpowerup.com/gpuz/

well it's hard to do that since i can't load any heavy 3D game to put the card under stress and i hold my fingers on the backplate and it's cool while trying to load the game, now i'm also experiencing 2d performance issues. takes 15 sec to open a 35kb jpg

gpu temp is 44-46C while quake tries to load.

@MOSFET
yes, they spin freerly by me blowing at them, made sure connections are tight

edit, make that 48c idle gpu

Tue Feb 27, 2018 5:02 pm

Have you checked in Event Viewer to see if there's anything odd going on (e.g. disk errors)?

Dieter · Tue Feb 27, 2018 5:03 pm

Try a different drive, removing all but one? When one of my secondary drives was failing (with no obvious errors), my system (i7, 32GB, SSD, etc) was slow and would take minutes to do almost anything. Pulling the failing drive put everything back to normal.

With the bad drive, CPU would usually show normal, and sometimes it would show 100% usage on the failing drive, but not always.

Why dusting would cause this, who knows. Like others said, maybe a cable got knocked loose, but it usually won't hurt to try another drive.

Good luck!

SoM · Tue Feb 27, 2018 5:27 pm

just brew it! wrote:
Have you checked in Event Viewer to see if there's anything odd going on (e.g. disk errors)?

yes, checked it last night, nothing

SoM · Tue Feb 27, 2018 5:28 pm

Dieter wrote:
Try a different drive, removing all but one? When one of my secondary drives was failing (with no obvious errors), my system (i7, 32GB, SSD, etc) was slow and would take minutes to do almost anything. Pulling the failing drive put everything back to normal.

With the bad drive, CPU would usually show normal, and sometimes it would show 100% usage on the failing drive, but not always.

Why dusting would cause this, who knows. Like others said, maybe a cable got knocked loose, but it usually won't hurt to try another drive.

Good luck!

cables checked/re-checked

i don't have a spare M.2 SSD

DPete27 · Tue Feb 27, 2018 5:30 pm

MSI Kombustor/Furmark (or similar stress testing program) is what I'd use if I'm concerned about load testing a GPU. Much lighter/easier to load than a game.
http://www.guru3d.com/files-details/msi-kombustor-(furmark)-7-beta.html

GPU backplates are rarely thermally conductive to any components on the board.

Again, maybe I'm wrong, but none of this sounds like a problem caused by GPU hardware.

SoM · Tue Feb 27, 2018 7:02 pm

managed to turn the fans on with kombustor, quake loaded to main menu in 5 min

back on 390.77 clean install after DDU

testing...

Tue Feb 27, 2018 7:16 pm

Checked for malware?

SoM · Tue Feb 27, 2018 8:20 pm

just brew it! wrote:
Checked for malware?

started a full scan 10min before u posted, will get back to you on this.

next thing i'll try is my 1070's 2nd bios after scan

SoM · Tue Feb 27, 2018 8:54 pm

900037 files scanned, 0 threat

try 2nd bios

SoM · Tue Feb 27, 2018 9:33 pm

after the scan finished i did a shutdown (it was fast, like a switch) waited 10 seconds, press power, boot in 15sec and right off the bat i can see the system very responsive, i proceed to launch quake champions, loads fast, fans spin from the start, did a full game of deathmatch on my regular ULTRA settings 1440p and it's smooth, everything is as it was before i cleaned the PC.

wtf was that than ?

i didn't use the 2nd bios

synthtel2 · Tue Feb 27, 2018 10:44 pm

Maybe LatencyMon could reveal something interesting? It gives a lot of useful info beyond just the DPC latency numbers.

I debugged a really weird one for a friend lately - a 7700HQ+1050Ti machine had ceased being able to hold 60 fps in Rainbow Six Siege sometime in January, going more like 20-30 fps in heavy scenes regardless of settings. Other games seemed mostly fine. Boot and load times were probably damaged too, but that data is iffy. Clocks, thermals, resource utilization figures, graphics drivers, drive health, and the rest of the usual suspects (including game-specific ones) all turned up completely normal. LatencyMon provided the first real lead, showing a variety of core Windows drivers taking 2 to 6 millis (trimodally if you include normality as a mode) way too often regardless of CPU utilization. I can't name names off the top of my head, but these were the sorts of drivers that should be the most bulletproof ones on the system, and whatever was going on was hitting at least a half dozen of them. On a hunch, we did a particularly aggressive pass of disabling startup items and services, and everything cleared right up. We didn't go on to figure out which specific entry was tied to the problem, but the bulk of the iffy-looking things were provided by Intel or Asus.

I've been wondering if some otherwise benign software wasn't interacting badly with some Spectre mitigation or other. The timetable lines up, and if that was it, whatever it was is unlikely to be the only one out there. If so, I could imagine a worse case of it looking a lot like this.

SoM · Tue Feb 27, 2018 11:23 pm

does this tell you anything ?
looks normal to me

_________________________________________________________________________________________________________
CONCLUSION
_________________________________________________________________________________________________________
Your system appears to be suitable for handling real-time audio and other tasks without dropouts. 
LatencyMon has been analyzing your system for  0:05:12  (h:mm:ss) on all processors.


_________________________________________________________________________________________________________
SYSTEM INFORMATION
_________________________________________________________________________________________________________
Computer name:                                        DESKTOP-0KJ5OCU
OS version:                                           Windows 10 , 10.0, build: 16299 (x64)
Hardware:                                             ASUSTeK COMPUTER INC., Z170-A
CPU:                                                  GenuineIntel Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz
Logical processors:                                   8
Processor groups:                                     1
RAM:                                                  32706 MB total


_________________________________________________________________________________________________________
CPU SPEED
_________________________________________________________________________________________________________
Reported CPU speed:                                   4008 MHz
Measured CPU speed:                                   1 MHz (approx.)

Note: reported execution times may be calculated based on a fixed reported CPU speed. Disable variable speed settings like Intel Speed Step and AMD Cool N Quiet in the BIOS setup for more accurate results.

WARNING: the CPU speed that was measured is only a fraction of the CPU speed reported. Your CPUs may be throttled back due to variable speed settings and thermal issues. It is suggested that you run a utility which reports your actual CPU frequency and temperature. 



_________________________________________________________________________________________________________
MEASURED INTERRUPT TO USER PROCESS LATENCIES
_________________________________________________________________________________________________________
The interrupt to process latency reflects the measured interval that a usermode process needed to respond to a hardware request from the moment the interrupt service routine started execution. This includes the scheduling and execution of a DPC routine, the signaling of an event and the waking up of a usermode thread from an idle wait state in response to that event.

Highest measured interrupt to process latency (µs):   94.275461
Average measured interrupt to process latency (µs):   2.332836

Highest measured interrupt to DPC latency (µs):       90.698614
Average measured interrupt to DPC latency (µs):       0.575145


_________________________________________________________________________________________________________
 REPORTED ISRs
_________________________________________________________________________________________________________
Interrupt service routines are routines installed by the OS and device drivers that execute in response to a hardware interrupt signal.

Highest ISR routine execution time (µs):              104.718313
Driver with highest ISR routine execution time:       HDAudBus.sys - High Definition Audio Bus Driver, Microsoft Corporation

Highest reported total ISR routine time (%):          0.084346
Driver with highest ISR total time:                   dxgkrnl.sys - DirectX Graphics Kernel, Microsoft Corporation

Total time spent in ISRs (%)                          0.085699

ISR count (execution time <250 µs):                   70518
ISR count (execution time 250-500 µs):                0
ISR count (execution time 500-999 µs):                0
ISR count (execution time 1000-1999 µs):              0
ISR count (execution time 2000-3999 µs):              0
ISR count (execution time >=4000 µs):                 0


_________________________________________________________________________________________________________
REPORTED DPCs
_________________________________________________________________________________________________________
DPC routines are part of the interrupt servicing dispatch mechanism and disable the possibility for a process to utilize the CPU while it is interrupted until the DPC has finished execution.

Highest DPC routine execution time (µs):              544.032435
Driver with highest DPC routine execution time:       nvlddmkm.sys - NVIDIA Windows Kernel Mode Driver, Version 390.77 , NVIDIA Corporation

Highest reported total DPC routine time (%):          0.051152
Driver with highest DPC total execution time:         nvlddmkm.sys - NVIDIA Windows Kernel Mode Driver, Version 390.77 , NVIDIA Corporation

Total time spent in DPCs (%)                          0.076638

DPC count (execution time <250 µs):                   318661
DPC count (execution time 250-500 µs):                0
DPC count (execution time 500-999 µs):                1
DPC count (execution time 1000-1999 µs):              0
DPC count (execution time 2000-3999 µs):              0
DPC count (execution time >=4000 µs):                 0


_________________________________________________________________________________________________________
 REPORTED HARD PAGEFAULTS
_________________________________________________________________________________________________________
Hard pagefaults are events that get triggered by making use of virtual memory that is not resident in RAM but backed by a memory mapped file on disk. The process of resolving the hard pagefault requires reading in the memory from disk while the process is interrupted and blocked from execution.


Process with highest pagefault count:                 none

Total number of hard pagefaults                       0
Hard pagefault count of hardest hit process:          0
Highest hard pagefault resolution time (µs):          0.0
Total time spent in hard pagefaults (%):              0.0
Number of processes hit:                              0


_________________________________________________________________________________________________________
 PER CPU DATA
_________________________________________________________________________________________________________
CPU 0 Interrupt cycle time (s):                       4.243152
CPU 0 ISR highest execution time (µs):                104.718313
CPU 0 ISR total execution time (s):                   2.142083
CPU 0 ISR count:                                      70264
CPU 0 DPC highest execution time (µs):                544.032435
CPU 0 DPC total execution time (s):                   1.664608
CPU 0 DPC count:                                      242086
_________________________________________________________________________________________________________
CPU 1 Interrupt cycle time (s):                       0.229190
CPU 1 ISR highest execution time (µs):                3.228293
CPU 1 ISR total execution time (s):                   0.000173
CPU 1 ISR count:                                      104
CPU 1 DPC highest execution time (µs):                37.340319
CPU 1 DPC total execution time (s):                   0.024186
CPU 1 DPC count:                                      8721
_________________________________________________________________________________________________________
CPU 2 Interrupt cycle time (s):                       0.241836
CPU 2 ISR highest execution time (µs):                2.285928
CPU 2 ISR total execution time (s):                   0.000068
CPU 2 ISR count:                                      45
CPU 2 DPC highest execution time (µs):                77.533683
CPU 2 DPC total execution time (s):                   0.028102
CPU 2 DPC count:                                      10706
_________________________________________________________________________________________________________
CPU 3 Interrupt cycle time (s):                       0.285579
CPU 3 ISR highest execution time (µs):                2.261477
CPU 3 ISR total execution time (s):                   0.000056
CPU 3 ISR count:                                      36
CPU 3 DPC highest execution time (µs):                37.09980
CPU 3 DPC total execution time (s):                   0.027797
CPU 3 DPC count:                                      11051
_________________________________________________________________________________________________________
CPU 4 Interrupt cycle time (s):                       0.279540
CPU 4 ISR highest execution time (µs):                2.803643
CPU 4 ISR total execution time (s):                   0.000068
CPU 4 ISR count:                                      43
CPU 4 DPC highest execution time (µs):                47.511477
CPU 4 DPC total execution time (s):                   0.030619
CPU 4 DPC count:                                      11127
_________________________________________________________________________________________________________
CPU 5 Interrupt cycle time (s):                       0.237513
CPU 5 ISR highest execution time (µs):                2.564870
CPU 5 ISR total execution time (s):                   0.000038
CPU 5 ISR count:                                      26
CPU 5 DPC highest execution time (µs):                44.255988
CPU 5 DPC total execution time (s):                   0.046537
CPU 5 DPC count:                                      11621
_________________________________________________________________________________________________________
CPU 6 Interrupt cycle time (s):                       0.290253
CPU 6 ISR highest execution time (µs):                0.0
CPU 6 ISR total execution time (s):                   0.0
CPU 6 ISR count:                                      0
CPU 6 DPC highest execution time (µs):                50.360279
CPU 6 DPC total execution time (s):                   0.046117
CPU 6 DPC count:                                      11720
_________________________________________________________________________________________________________
CPU 7 Interrupt cycle time (s):                       0.261852
CPU 7 ISR highest execution time (µs):                0.0
CPU 7 ISR total execution time (s):                   0.0
CPU 7 ISR count:                                      0
CPU 7 DPC highest execution time (µs):                42.411677
CPU 7 DPC total execution time (s):                   0.047973
CPU 7 DPC count:                                      11630
_________________________________________________________________________________________________________

synthtel2 · Tue Feb 27, 2018 11:35 pm

That does all look normal. It isn't having the problem at the moment though, right? There not being much to debug at the moment, I guess it's just an idea to stow in case it flares up again.

IIRC we were on the drivers tab, looking at execution times (sorted by the max execution time column).

TR Forums

sluggish performance

sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Re: sluggish performance

Who is online