User panel stuff on forum
  13 posts on 1 page  1
General Discussion
2016-11-02, 04:34
News Writer
912 posts

Registered:
Jan 2006
Anyone got one of these yet? If so, how is the performance in QW and QuakeLive? Some timedemos of the demos that come with nQuake would be great.

Looking at getting this 120hz laptop and I am keen to see the GTX 1060 performance first (they use the same chips for both desktop and laptop now... just clocked differently...

(Edited 2016-11-02, 07:03)
2016-11-02, 07:03
News Writer
912 posts

Registered:
Jan 2006
So I went in to an MSI showroom today and they let me install nQuake on the MSI GS73VR laptop in order to benchmark it.... unfortunately it only got an underwhelming 400fps+ on the e1m2 4on4 demo that comes with nQuake which is perfect for real-world testing.

I was scratching my head thinking... how can this be? The CPU is an i7-6700HQ with a single core speed of 2.6Ghz (3.1Ghz with boost) and it has a GeForce GTX 1060 graphics card which on paper should hit 1000fps+ without any issues.

Surely the default config of nQuake can't be that bad, right? Wrong.

I come home and get on my desktop which has an i7-6700K CPU and a GeForce GTX 980ti and my config from QuakeCon (which uses nQuake and requires no additional files apart from the cfg)....
nQuake default: 460fps!
QuakeCon config: ~1600fps

Yes, you read that correctly... With the default nQuake config you get around 30% of the FPS compared to that of an optimised config.

To make sure that this wasn't a fluke, I tested my other computer which is an i7-5500U with a GeForce GTX 770 (Laptop with external GPU)...
nQuake default: 289fps
QuakeCon config: 785fps

The good news is that there is still hope for a 1001fps/120hz gaming laptop. They've given me permission to take in any config or game that I want to benchmark on the Laptop so I will be doing exactly that tomorrow. I'll be taking in my QuakeCon config and also a copy of CPMA with an updated client/config/demo.
2016-11-02, 09:16
Member
133 posts

Registered:
Dec 2008
I believe there is just a lock on max fps in driver.
And In general ezquake-gl with cl_maxfps 0 is a good test of quality of video card.
On my current pc 1st video card became broken after experiment with cl_maxfps 0.
New video card also cannot handle it. It reboots in several seconds.

So I think in your case it is safety restriction.
100 fps should be enough for everybody! ;-)
May be it is not directly restriction on fps but restriction on power consumed by some unit in videocard.

By the way my notebook (not pc) with nvidia has similar restriction. It does not allow big fps.
Even worse! I forced intel card for quake there to make gameplay smoother.
2016-11-02, 10:48
Member
6 posts

Registered:
Oct 2012
I'm running nQuake (not with default config though) on a MacBook Pro 15" 2015 (AMD R9 M370X) and get somewhere between 1500-2000 FPS when doing so in macOS, which is cranked up to 2000-2200 when running nQuake in windows 10 with the latest AMD drivers (Apple drivers suck balls).

I can't believe the new line of Nvidia GPUs would perform worse.

(Edited 2016-11-02, 12:26)
2016-11-02, 11:29
News Writer
912 posts

Registered:
Jan 2006
Bal0n wrote:
I'm running nQuake (not with default config though) on a MacBook Pro 15" 2015 (AMD R9 M360X) and get somewhere between 1500-2000 FPS when doing so in macOS, which is cranked up to 2000-2200 when running nQuake in windows 10 with the latest AMD drivers (Apple drivers suck balls).

I can't believe the new line of Nvidia GPUs would perform worse.


Any chance you can try with a completely vanilla nQuake install and config? Would be a great comparison...
2016-11-02, 12:26
Member
6 posts

Registered:
Oct 2012
dirtbox wrote:
Bal0n wrote:
I'm running nQuake (not with default config though) on a MacBook Pro 15" 2015 (AMD R9 M360X) and get somewhere between 1500-2000 FPS when doing so in macOS, which is cranked up to 2000-2200 when running nQuake in windows 10 with the latest AMD drivers (Apple drivers suck balls).

I can't believe the new line of Nvidia GPUs would perform worse.


Any chance you can try with a completely vanilla nQuake install and config? Would be a great comparison...


Sure, will look into it if I can find the time today!
2016-11-02, 20:44
News Writer
912 posts

Registered:
Jan 2006
se-sss wrote:
And In general ezquake-gl with cl_maxfps 0 is a good test of quality of video card.


I would have to disagree with you there. This simply shows the max fps the card can get in a single room but it doesn't measure it against anything constant. It is completely dependent on what you are looking at. The only way to accurately measure a card is by doing a timedemo using the same demo and same config.
2016-11-02, 22:15
Member
280 posts

Registered:
Jan 2015
Can I hijack the thread just for a moment? Why am I getting 3-20 fps drops regardless of the maxfps I use? I remember someone asking to disable thread optmization on nvidia settings, i did it already and the problem persists.
dev
2016-11-03, 02:19
News Writer
912 posts

Registered:
Jan 2006
andrestone wrote:
Can I hijack the thread just for a moment? Why am I getting 3-20 fps drops regardless of the maxfps I use? I remember someone asking to disable thread optmization on nvidia settings, i did it already and the problem persists.

No you may not Andre - get Process Lasso to see what process is spiking up your CPU
2016-11-03, 05:58
News Writer
912 posts

Registered:
Jan 2006
I went back to the store with my config and the FPS only increased from ~400fps up to 580fps.

As the demo being used is of an intensive 4on4 demo on e1m2 (which will get the lowest FPS due to all of the SNG action) this means that the laptop would be suitable for anyone that is happy to play with 616fps which it would be able to handle in most QW games. In a 1on1 timedemo on aerowalk it was around 650fps.

My next test will be on a 120hz Laptop running a Geforce GTX 1070 however this is a huge price jump.
2016-11-03, 07:27
Member
32 posts

Registered:
Sep 2016
Did you try with ezquake 2.2? I get about 8000 fps in timerefresh at povdmm4 with 2.2 but only 2000 with 3.0.1, and on dm4 ammo room its 6000 vs 1600.
2016-11-03, 08:37
News Writer
912 posts

Registered:
Jan 2006
Repast wrote:
Did you try with ezquake 2.2? I get about 8000 fps in timerefresh at povdmm4 with 2.2 but only 2000 with 3.0.1, and on dm4 ammo room its 6000 vs 1600.

Have you tried a timedemo (not timerefresh)?
2017-10-23, 07:59
Member
40 posts

Registered:
Feb 2009
Dirtbox, I have a similar MSI laptop (GE72VR), and I realize the same problem, however I think is not about laptop power. I used to have much slower machines with 780M or even 460M, and both were faster 460M made solid 500FPS stable, 780M over 1000FPS, but I have used 720FPS rock solid. I guess this issue is coming from the freaking intel hd cpu vga... I have 400-500fps which is crap comparing the power of the 1060GTX. In tests (such as 3dmark, heaven) it outperforms easily the 980M, which made 1000+FPS... I have not found the solution, but I hardly believe it is about vga power.... of course with other games (overwatch, bf1) it has solid fps, so it could be some ezquake thingy...
/say_team $R fuck yeah!
  13 posts on 1 page  1