[SOLVED] Micro-stutter with all processes

ajdemoulpied

Commendable
Jan 15, 2019
31
0
1,530
So I recently upgraded to a RTX 2070, with that I am not sure if that is causing my issue. Here is the deal, everything I do and I mean everything I am getting consistent micro-stuttering, if that is a word. Even simply watching Netflix, micro-stutter when the camera pans or something of the sort. While playing GTA V I have no issues generally until a couple hours in then it will either have the micro-stutter issue or just freeze up all together and kick me out. AC Odyssey seems to be sluggish and choppy, even while limiting FPS to 50, and Forza 7 runs perfectly during game play like during a race, but any cinematics or cut scenes run like crap and stutter.

Nothing is over heating, GPU hits 68 C max, CPU hits 49 C max, during benchmark with Time Spy on 3D Mark. Here are some odd readings though. During the bench mark, the GPU never gets over 9% load and the CPU averaged in the high 70's and maxed out here and there. When 3D Mark tried to load the second graphics test, the program dumped and kicked me out. Crazy thing as well, there are uniform load spikes from the CPU and GPU in unison as I am just sitting here typing this out. RAM stays consistent, not fluctuation. I have updated all drivers, and even went as far as completely restoring windows to try and fix this issue, nothing has worked. Like I had mentioned earlier, I started having issues when I upgraded to a 2070 from a 1050Ti. I also stepped up monitors from a 60hz 1080p to a 144hz 1440p.

Here is what I am thinking, either something is going out like the CPU, or my power supply can't handle the upgrade, PC specs are below. One thing I also noticed is the stuttering got worse, significantly worse to the point where program windows stutter while being dragged across the screen, once I hooked my old monitor so I could see a live feed of the read outs while bench marking. I included anything I could think of in case anyone agrees that it may be a low PSU output issue.

Any help would be beyond grateful before I just cave and start buying components until it goes away, not how I would like to spend money right now.

PC Specs:
MOBO Gigabyte Gaming 7
i5 6600K OC'd by MOBO to 4.2Ghz, air cooled by a Hyper 212
16GB G. Skill Ripjaws 3200Mhz
RTX Gigabyte 2070 running at 1890Mhz core clock and 1750Mhz Memory clock
EVGA 650W 80+ Gold PSU
Windows runs off a 970 EVO NVMe
For games I have 850 EVO and a 860 EVO SSD's
Have additional WD Blue 7200 RPM 1TB HDD
Fan's- 5 120mm Corsair RGB, 1 140 Corsair non-RGB

I am sure I have left something out so please ask away. This could not be more frustrating.
 
Solution
Doubtfully. I think it's definitely a software issue at this point. If your games are working okay for now, then at least that's half the problem solved. Honestly, I'm not sure what would cause sfc or DSIM to fail unless the OS is bugged. I mean technically it could be something is unstable hardware related, but if you're running base clocks on everything we can only assume software.

Try running memtest 86. (RAM)
Try running Prime 95 (CPU)

A fresh copy of windows from the Microsoft webpage is really the only sure way that your OS is completely clean.

ajdemoulpied

Commendable
Jan 15, 2019
31
0
1,530
UserBenchmarks: Game 99%, Desk 92%, Work 59%
CPU: Intel Core i5-6600K - 86.2%
GPU: Nvidia GeForce RTX 2070 - 105.9%
SSD: Samsung 970 Evo NVMe PCIe M.2 250GB - 124.4%
SSD: Samsung 850 Evo 250GB - 82.6%
SSD: Samsung 860 Evo 1TB - 85.4%
HDD: WD Blue 1TB (2012) - 93.6%
RAM: G.SKILL F4 DDR4 3200 C16 2x8GB - 70.6%
MBD: Gigabyte GA-Z170X-Gaming 7

Results from UserBenchmark, apparently most of the hardware is running way below expectations. According to the results it is running in the 23rd percentile. mind you this is the second time I had to run this bench because the first time, my PC crashed and I got a DPC_WATCHDOG_VIOLATION error with a bluescreen of death. I had to manually restart.

After restarting I adjusted my SATA Controller driver. It was iaStorA.sys, I manually changed it to storahci.sys and on another restart, it ran the test just fine and gave me the above results. I double checked to make sure this wasn't a fluke by changing it back, and it crashed again. Not sure it this fixed my issue or not as it takes time after a restart for the skipping to happen.
 

ajdemoulpied

Commendable
Jan 15, 2019
31
0
1,530


I had stepped everything down due to the issues that were coming up, though perhaps it was being pushed to hard. Here are results from everything being overclocked. CPU at 4.4, RAM set to 3200 with the XMP profile, and a 115 and 450 Mhz increase on the core and memory clock respectively for the GPU. The GPU and SSD's are still under performing. I will run the lantencymon here next.

UserBenchmarks: Game 112%, Desk 116%, Work 74%
CPU: Intel Core i5-6600K - 100.5%
GPU: Nvidia GeForce RTX 2070 - 118.7%
SSD: Samsung 970 Evo NVMe PCIe M.2 250GB - 171.5%
SSD: Samsung 850 Evo 250GB - 85.5%
SSD: Samsung 860 Evo 1TB - 92.1%
HDD: WD Blue 1TB (2012) - 95.2%
RAM: G.SKILL F4 DDR4 3200 C16 2x8GB - 102.9%
MBD: Gigabyte GA-Z170X-Gaming 7

 

ajdemoulpied

Commendable
Jan 15, 2019
31
0
1,530


According to latencymon, as of right now my system " appears to be suitable for handling real-time audio and other tasks without dropouts". I'll see if this holds true as the stuttering starts to happen once the system has been running a bit.

UPDATE: After running a race in Forza 7, the results cam back that the system is having trouble with the highest measured interrupt being 5826. AC Odyssey actually crashed and gave me a reading of 9449.
 


its half right...:) this report has no issues
keep it running in background until u get stutters, then post back
 

ajdemoulpied

Commendable
Jan 15, 2019
31
0
1,530


Here is another report after playing around in AC Odyssey, having a crash, and also running the open GL test in CineBench where the stutter is very prevalent. Every second in perfect timing the image stops for a micro second and then continues on. It is quite literally like a clock ticking. It does this non-stop.

https://pastebin.com/zPsJDfgu

 


its showing up nvidia drivers
run DDU in safe mode https://www.guru3d.com/files-details/display-driver-uninstaller-download.html
remove your current nvidia drivers with it, reboot, then install new driver
 

ajdemoulpied

Commendable
Jan 15, 2019
31
0
1,530



Alright, did that and it did fix the problems. After running Cinebench, 3DMark, and Userbench I did not notice any issues at all. I even started to pull better scores. And here comes the but... I am still getting some crashing in games and now as I was sitting here working on some stuff I started to notice a slight flicker every 10 seconds or so. Random and very slightly, if you blinked you would miss it. I am really starting to think the GPU is having some problems. Maybe time to contact Gigabyte and have them check it? Still under warranty. Although I do still have my old 1050ti so maybe I can throw that in and see if anything changes, but I don't know how well it will run 1440p and changing the monitor and resolution could throw in to many variables. I am certainly lost on this one. I do really, REALLY appreciate the help up this point though, truly.

UPDATE: And now all of a sudden videos will randomly not load. Sound will be there, but nothing will show and the web page starts showing double and will repeat images as I scroll. The only way to make it work properly again is to restart the whole computer.
 
crashing in games is one big variable, any error logs from game or windows event viewer?
flicker on monitor dunno, could be cable, could be gpu, could be some grounding issue
can u try different web browser to see if you will have same/similar issues with videos?
 

ajdemoulpied

Commendable
Jan 15, 2019
31
0
1,530


I did think about messing with the voltages. At idle, the CPU will pull about.86, under heavy load and stress testing it will average 1.26 or so and fluctuate a little. Not sure if that is good or not as the MOBO does all that.
 

ajdemoulpied

Commendable
Jan 15, 2019
31
0
1,530


Uplay will send crash reports, but that is about it. No difference in browsers, using Chrome, but I did notice that everything is pretty much fine until the GPU gets stressed, like playing a game or something of the sort. After that everything goes down hill and the computer does not want to run up to par. I did notice last night that ther GPU was not seated all the way. I had to bend the metal backplate where the thumbscrews go in in order for the card to fully seat. I popped my old card in to see if it was the same and it was. Could this have done something to the MOBO over time with it not being all the way in? The side with the little snap would be fully seated, but as it worked it's way towards the back of the MOBO, it would be slightly angled. Fully seating it has made no difference by the way in terms of the issues.
 
nvidia driver seems to be okay

acpi driver seems to have some hickups, also hardware interrupts (CPU 0) are way to high

try to reset CMOS, dont change anything in bios, just boot order (if needed), and reinstall chipset drivers
see if those interrupts gets normal then
 

ajdemoulpied

Commendable
Jan 15, 2019
31
0
1,530


I cleared the CMOS, let her boot up (did not make any changes to the BIOS including and overclocks) and checked each chipset line item for driver updates. They all said they were good expect for an Intel Xeon E3 1200/1500 v5/6 gen PCIe controller(x16). After that, I checked for a widows update, there were none, and restarted again. I have lantencymod running again and I will post another log later in the day after it runs different processes and what not.