GPU slower than old laptop??

WTF, cpu on bmw benchmark 14min, gpu 11 min??? my old laptop 7min??
am I missing something!!
I just got my alienware pc now!! it should be faster??

tile size set for cpu 32
gpu 512

Hi.
We can not say anything if you do not mention anything about what hardware (CPU and GPU) has each machine.

are you using all the cores for CPU?

are you using a savin’ energy mode?

Are you using less samples on the laptop ?

We’re only shooting blind at this point, we need to know what GPU model you are using (as the difference between say a GTX 710 and a GTX 780 are huge), and what features are being used in the scene (since they can have a far greater impact on speed than if you just used the CPU).

We also would like the know if your drivers are up to date as well and whether you’re using an official Blender release build or the buildbot.

were can I get this info, all I find is Gforce gpu??

LOL. This is funny. :smiley:

Basically something is hogging my resources, its unlikely its a simple setting, recent test reveal the my cpu is same speed, and NVidia card 2 the speed of my 6 year old laptop. ive turned on all my 8 processors in the boot setting made no difference in render time. the cpu is only 2ghz but the gpu is the worry. The drivers are updated. I feel I’m just going to have to live with this.

What is your operating system? Windows? which version?

Usually when you open the nvidia settings, there gives information about the model of your graphics card.

I’m using win10. Its ulikely it has much to do with operating systems. As I said this is an old problem.

You haven’t yet said anything concrete. Just that you are running win 10. Download gpu-z and cpu-z and put real info here.

Ok thanks for the tip, according to the web https://www.techpowerup.com/gpudb/2536/geforce-gtx-860m
geforce-gtx-860m.
I’m starting to think I got screwed with an old gpu. My fault should have done more research.

~

Attachments


Alienware will always give you overpriced PC’s.

The 860m 2Gb verison uses the Maxwell architecture, but only has 640 CUDA cores, which isn’t much.