topic Re: High CPU usage with latest 3345 Graphic drivers update in Graphics

High CPU usage with latest 3345 Graphic drivers update

DEdoh — Mon, 16 Dec 2013 14:35:58 GMT

Hi,

I recently updated my drivers from version 3257 to 3345 and noticed an important increase in the CPU usage of my system.

I have an 8 cores system and one of those cores goes up to more than 85% usage, by doing nothing! With version 3257, it was around 3%!

Here is the log from sys_analyzer containing the informations of my system:

Graphics Devices:

Name Version State

Intel(R) HD Graphics 4600 10.18.10.3345 Active

System info:

CPU: Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz

OS: Microsoft Windows 7 Enterprise

Arch: 64-bit

Any idea on what could be the cause?

I've attached a screenshot of the resource manager with the different CPUs in my machine. It shows that CPU 0 is very high

Thanks!

Re: High CPU usage with latest 3345 Graphic drivers update

Allan_J_Intel1 — Tue, 17 Dec 2013 17:06:16 GMT

Hi dedohbed,

Based on the picture I see CPU Usage as 9%. I just ran windows installation and got video drivers 3345, I am afraid I did not get any high processor high usage. The driver was tested with some other computers and it seems to run fine.

Allan

Re: High CPU usage with latest 3345 Graphic drivers update

WHilt — Wed, 18 Dec 2013 16:25:00 GMT

Hi!

It's a definitive bug in the driver...

I installed a new system with Windows 7 Home Premium 64bit (Gigabyte GA-H87-D3H newest bios, Intel Core i7-4770).

CPU-Usage abolutely flatlined.... then the first and only thing i installed was the 3345-Driver.

Suddenly an increase on Core0 to 60-80% Usage, permanently! Sluggish System behaviour.

Processmonitor => Usage comes from interrupts/dpc

Uninstalled driver.... immediately flatlined again.

Installed older driver.... still flatlining and everything running smoothly.

Re: High CPU usage with latest 3345 Graphic drivers update

JFort7 — Thu, 19 Dec 2013 08:58:26 GMT

Hi,

Since i7-4770 has 4 cores with hyper-threading (8 threads in total), close to 85% CPU usage in one core (thread) would result in roughly 10.625% (85%/8) total CPU usage in resource manager.

I also have similar high CPU usage problem with the newest 3345 driver under Win 7 64 bit which I suspect is due to some problem related to the HDMI part of the driver. The Intel HDMI audio device is gone missing from the device manager with the newest 3345 driver whereas the old 3257 driver does not have this problem.

Re: High CPU usage with latest 3345 Graphic drivers update

Allan_J_Intel1 — Thu, 19 Dec 2013 14:56:08 GMT

Hi all,

Reply with information of the graphics report and DirectX diagnostic report as well. I will send the information to our engineering department and check if they have this issue already reported with the latest video drivers.

Graphics report:

http://www.intel.com/support/graphics/sb/cs-009476.htm http://www.intel.com/support/graphics/sb/cs-009476.htm

DirectX report:

click Start, click Run, type dxdiag in the Open box, and then click OK.

Allan.

Re: High CPU usage with latest 3345 Graphic drivers update

DEdoh — Thu, 19 Dec 2013 16:22:39 GMT

Hi Allan,

Here are the reports you asked for.

Thanks!

Re: High CPU usage with latest 3345 Graphic drivers update

JVane11 — Fri, 20 Dec 2013 20:56:00 GMT

I have the same problem on MSI H87-G43 (BIOS V2.6) mainboard with i5-4440 (Intel HD 4600).

Any newer driver than 3257 (I tested 3345, 3325, 3277) causes 10-12% CPU used

by system interrupts. I also find out that when put system on sleep (S3) and

wake up again the CPU peak flattened, after restart the peaks appears again. I

also noticed that with all newer drivers the Intel Display Audio disappeared from

device manager.

Windows 7 64bit

Re: High CPU usage with latest 3345 Graphic drivers update

MSchm21 — Fri, 20 Dec 2013 21:45:55 GMT

I had this interrupt usage problem as well but with driver 3257, since driver 3325 the problem disappeared for me. The cause was USB EHCI 8C2D, GPU driver was in conflict somehow.

Re: High CPU usage with latest 3345 Graphic drivers update

JE_1 — Fri, 17 Jan 2014 15:03:41 GMT

i have the same Problem !

i have a asrock h87 pro4 and installed HD4600 Driver from CD , it was 9.18.10.3186 Driver.

than i updated the Graphics Driver from Intel Support Detect. 10.18.10.3345

after Reboot i see High CPU 25% without any work on the PC, Process Explorer show High System Interrups .

i deinstalled this Driver and go back to the Mainboard CD Driver , the High System Interrups was gone...

Will Intel fix this Problem ?

Re: High CPU usage with latest 3345 Graphic drivers update

MSchm21 — Sun, 19 Jan 2014 03:05:13 GMT

Do you have Windows 7 or 8/8.1? Amazing that Intel isn't in a hurry to fix that.

Re: High CPU usage with latest 3345 Graphic drivers update

JE_1 — Sun, 19 Jan 2014 13:41:16 GMT

Windows 7 / 64

Re: High CPU usage with latest 3345 Graphic drivers update

RBrug1 — Mon, 20 Jan 2014 06:20:05 GMT

I am having the same problem of CPU usage with a Shuttle PC SH87R6, i3-4330, and the 3345 graphics drivers.

Re: High CPU usage with latest 3345 Graphic drivers update

NMish3 — Thu, 30 Jan 2014 20:50:52 GMT

I can confirm it too. I have the same specs as HansCZ. Also I launched LatencyMon to see which driver is generating interrupts. It seems that USBPORT.SYS is the culprit on my system. But I didn't have this problem until I updated the drivers. When idle cpu usage is 12~18% on core 0. Going to sleep and back kicks cpu usage down to 0%.

Re: High CPU usage with latest 3345 Graphic drivers update

NMish3 — Fri, 31 Jan 2014 09:12:20 GMT

Downgrading the GPU drivers solved the issue. Driver version 3257 fixes it.

Re: High CPU usage with latest 3345 Graphic drivers update

EEgon — Sun, 23 Feb 2014 18:13:11 GMT

oops... here is another thread...

well downgrading is not an option, since i get a blackscreen then... Its up to Intel to fix this. Hopefully asap.

kind regards

Re: High CPU usage with latest 3345 Graphic drivers update

GGroo — Thu, 10 Dec 2015 01:30:12 GMT

So I was doing some debugging of an application I'm working on and this wonderful bug springs up while I'm doing some work around my persistence engine. The logical conclusion was that the changes I made had had some profound performance impact on the persistence scheme we had, so I started debugging only to discover (some two hours later) that this !@*# ing thing was the cause of my problems. It was only sometime later that I actually cranked open the performance tab on task manager and noticed that the performance problems had nothing to do with my application or the changes I was making.

What an annoying waste of a couple hours.