Lag Spikes, Slutters And Low Usage Issue From New Graphics Card While Gaming! Please Help!

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015
Hello everyone.... I hope all my brothers here on Tomshardware.com are fine..... So lets jump towards the topic without wasting time :)

_________
MY BUILD:
---------

Intel Xeon E3-1245 V3
Cooler Mater Hyper 212 LED TURBO Cooler
Asus Q87M-E
8GB RAM
500GB Seagte HDD ##########NOTE########## HDD's health is 0% and performance on 100%
Msi Gaming X GTX 1060 6GB
Antec VP450 PSU
Cooler master MB600L Case

______
Issue:
------

I recently purchased a MSI GAMING X GTX 1060 6GB without box and with one week checking warranty for 28500PKR. I got back home and placed the card in my build and noticed that my PSU don't have a 8-pin connector! So i purchased a 6 to 8-pin LEGIT connector. Ran the card and everything was fine:) next day which is today when i turned my pc on, i noticed a very unusual behaviour from my GPU! I am going to express this in points.

1. I started getting slutters while gaming and sudden lag spikes

2. Downloaded MSI Afterburner Monitored my GPU and CPU usage and found that whenever there are fast moves or heavy vegetation the GPU Usage drops down to 46% or even less!

3. GPU and CPU usage continously keed fluctuating.... get stable sometimes like when driving on an open highway with less things around

4. I launched GPU-Z and went on "sensors" tab and then started playing Vampyr and here's the screenshot after 5 mins of gameplay:
HzLVTQ5.gif


5. Temps never exceed 61*C

If you guys need anything else..... FEEL FREE TO ASK. Please help.. I am really worried!
 
Solution
1 - the 300W PSU will still be providing some power to the GPU via the PCIe slot.

2 - IT DOES NOT MATTER if you are using the Antec to only connect the to the GPU's 8-pin connector. The 6-pin connector and wiring from the Antec IS NOT DESIGNED to carry twice as much as its rated load to the GPU.

The 6 pin cable is specified to provide only up to 75W power. The 8-pin connector on GPU expects to be able to pull as much as 150W from that connector.

Your test is therefore not valid nor useful for diagnosing this.
Sounds like your GPU probably isn't getting enough juice. A 6 pin PCIe power cable is unable to provide the same amount of power as an 8 pin. While the adapter will fit, if your GPU calls for an 8pin and you only have a 6pin you won't be providing enough power so the GPU won't be able to run at full speed.
 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015


But it was running fine yesterday!
 
Yesterday it might not have needed as much power, but today it needed more. How old is the PSU? Recommended PSU is 387W for your setup. If the powersupply is over 2 years old capacitor aging could make it such that it is only able to supply ~80% of its rated amount. Therefore the GPU could be spiking for more power and it just isn't able to give it what it needs.
 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015


My psu is one months used!

Did u saw the GPU-Z ss ?
 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015


Few of my friends have a PC. And they are still using Core 2 Duo and have no idea of Modern Gaming!
 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015
#############NOTE###########
-------------------------------------------------


I ran a benchmark on UNIGINE SUPERPOSITION and there was no USAGE ISSUE through the whole benchmark!!!! GPU Usage remained at constant 99%! Then why I am having such an issue with games?

 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015


Hey Brother! Check out this detailed experiment conducted by me. So I tested FEW GAMES and SUPERPOSITION BENCHAMRK. Here are the results: Please go through them thoroughly :)

NOTE: All games were running on 1280x1024 resolution (my native)


##########VAMPYR ON MAXED OUT GRAPHICS SETTING#############
VwS8M3i.png

PAFhRIH.gif



GPU-Z Logs:
GPU Core Clock [MHz] =1949.0
GPU Memory Clock [MHz]= 2003.4
GPU Temperature [°C] = 52.0
Memory Used [MB] =2213
GPU Load [%]=25
Memory Controller Load [%]= 12
Video Engine Load [%]=0
Bus Interface Load [%]=1
Power Consumption [W]=55.3
Power Consumption [% TDP]=27.7
PerfCap Reason []=4
VDDC [V]=1.0620
CPU Temperature [°C]= 58.0
System Memory Used [MB]= 4433






##########Need for Speed Most Wanted 2012 ON MAXED OUT GRAPHICS SETTINGS #############

hFBKPxM.png

AUcaeXu.gif



GPU-Z Logs:
GPU Core Clock [MHz] =1936.0
GPU Memory Clock [MHz]= 2003.4
GPU Temperature [°C] = 54.0
Memory Used [MB] =1438
GPU Load [%]=51
Memory Controller Load [%]= 21
Video Engine Load [%]=0
Bus Interface Load [%]=0
Power Consumption [W]=91.2
Power Consumption [% TDP]=53.0
PerfCap Reason []=4
VDDC [V]=1.0500
CPU Temperature [°C]= 50.0
System Memory Used [MB]= 2741







########################CSGO OFFLINE WITH BOTS MAXED OUT SETTINGS###############

6MzJytp.png

zFwoPAU.gif


GPU-Z Logs:
GPU Core Clock [MHz] =1949.0
GPU Memory Clock [MHz]= 2003.4
GPU Temperature [°C] = 48.0
Memory Used [MB] =716
GPU Load [%]=31
Memory Controller Load [%]= 16
Video Engine Load [%]=0
Bus Interface Load [%]=3
Power Consumption [W]=57.5
Power Consumption [% TDP]=28.7
PerfCap Reason []=4
VDDC [V]=1.0500
CPU Temperature [°C]= 56.0
System Memory Used [MB]= 3889

#####################BATMAN ARKHAM ORIGINS MAXED OUT#################[b

UxQ3YPN.png

AHk85wa.gif



GPU-Z Logs:
GPU Core Clock [MHz] =1936.0
GPU Memory Clock [MHz]= 2003.4
GPU Temperature [°C] = 54.0
Memory Used [MB] =992
GPU Load [%]=13
Memory Controller Load [%]= 6
Video Engine Load [%]=0
Bus Interface Load [%]=0
Power Consumption [W]=44.5
Power Consumption [% TDP]=22.2
PerfCap Reason []=4
VDDC [V]=1.0500
CPU Temperature [°C]= 52.0
System Memory Used [MB]= 2873


#####################UNIGINE SUPERPOSITION 1080P HIGH###############

2gYfiUD.gif


GPU-Z Logs:
GPU Core Clock [MHz] =1923.5
GPU Memory Clock [MHz]= 2003.4
GPU Temperature [°C] = 61.0
Memory Used [MB] =3112
GPU Load [%]=99
Memory Controller Load [%]= 50
Video Engine Load [%]=0
Bus Interface Load [%]=0
Power Consumption [W]=125.6
Power Consumption [% TDP]=62.8
PerfCap Reason []=4
VDDC [V]=1.0500
CPU Temperature [°C]= 54.0
System Memory Used [MB]= 4946


#####################UNIGINE SUPERPOSTION CUSTOM 720P LOWEST################

oIGxWDh.gif






GPU-Z Logs:
GPU Core Clock [MHz] =1949.0
GPU Memory Clock [MHz]= 2003.4
GPU Temperature [°C] = 52.0
Memory Used [MB] =788
GPU Load [%]=52
Memory Controller Load [%]= 32
Video Engine Load [%]=0
Bus Interface Load [%]=1
Power Consumption [W]=74.7
Power Consumption [% TDP]=33.6
PerfCap Reason []=4
VDDC [V]=1.0620
CPU Temperature [°C]= 52.0
System Memory Used [MB]= 4869














 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015


Did you saw that when I tested UNIGINE SUPERPOSITION ON 1080P HIGH the GPU started working fine!
 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015


It always performs fine WHEN RUNNING UNIGINE SUPERPOSTION 1080P HIGH benchmark
 

King_V

Illustrious
Ambassador
As mentioned, don't ignore the advice you've already been given.



This is a recipe for disaster. The 6-pin cable from the PSU is only designed to carry a maximum of 75W. An 8-pin is used by the video card if there is an expectation of more than 75W being required from that connector.

The PSU is not going to be able to reliably push more than 75W along that 6-pin cable if the video card demands it, and those wires are not rated to carry in excess of 75W.

Yes, it seems to work fine when running unigine, but that is irrelevant.

You are risking damaging your components, and risking an electrical fire. You are right to be worried, but you are worried for the wrong reason.

If you had two separate 6-pin connectors on your PSU, then you could use an adapter to convert both 6-pin connectors to a single 8-pin connector.

If there are still problems after powering the card correctly, then there could be a separate issue.
 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015


I dont think HDD (i have tested on a SSD), RAM( I have tried using 16GB) or 6 to 8pin is causing the issue! I have a new info to give.... Let me know what you think... So IF I STRESS TEST MY GPU, IT GOES FINE BUT IF I STRESS TEST MY GPU AND CPU TOGETHER, GPU STARTS THROTTLING! I used FurMark for GPU and CPU-Z for CPU For Stress tests and used GPU-Z to monitor SENSORS. Any possible possible conclusion now?

 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015


I dont think HDD (i have tested on a SSD), RAM( I have tried using 16GB) or 6 to 8pin(A friend is also using it to power a ASUS STRIX RX 480 8GB) is causing the issue! I have a new info to give.... Let me know what you think... So IF I STRESS TEST MY GPU, IT GOES FINE BUT IF I STRESS TEST MY GPU AND CPU TOGETHER, GPU STARTS THROTTLING! I used FurMark for GPU and CPU-Z for CPU For Stress tests and used GPU-Z to monitor SENSORS. Any possible possible conclusion now?

 
That right there says one of two things.
1. You don't have proper airflow in the case and the CPU and GPU are getting too hot and having to throttle.
2. You don't have a big enough power supply to feed all the components at once so as soon as they are pushed to max power draw they are being starved which causes them to throttle.
Of the two I think the PSU not being powerful enough is the most likely cause.
 

taimoorbaig382

Reputable
Jun 15, 2017
615
3
5,015


1. Max CPU Temp: 52°C AND Max GPU Temp: 61°C

2. Yes it can be PSU but:

I also did another experiment:

Borrowed a 300W psu from my friend

Powered my whole system from that 300W psu EXCEPT GPU

Powered GPU through my ANTEC VP450 PSU

Still i am experiencing the same issue..... Any more conclusions now? Thank you very much Jeremy