Go Back   nV News Forums > Linux Support Forums > NVIDIA Linux

Newegg Daily Deals

Reply
 
Thread Tools
Old 04-15-08, 04:43 PM   #97
pawels133
Registered User
 
pawels133's Avatar
 
Join Date: Mar 2008
Posts: 201
Default Re: 8400M GS : Low performences

Just stop trolling right now. Lol closed source driver outperforms open ones, because there's no open documentation. It's pointless to talk with trolls like you. CU
__________________
Not everybody can smoke and drink... xd
pawels133 is offline   Reply With Quote
Old 04-15-08, 04:57 PM   #98
NvFuchs
Registered Fox
 
NvFuchs's Avatar
 
Join Date: Oct 2007
Location: Confoederatio Helvetica
Posts: 530
Default Re: 8400M GS : Low performences

Quote:
Originally Posted by pawels133
Just stop trolling right now.
"An Internet troll, or simply troll in Internet slang, is someone who posts controversial and usually irrelevant or off-topic messages in an online community, such as an online discussion forum, with the intention of baiting other users into an emotional response[1] or to generally disrupt normal on-topic discussion"
Wikipedia

Let me check: my posts are neither controversial, nor irrelevant, nor off-topic but filled with facts (bench results) and polite asking for you to post (compareable) results as well.

Quote:
Lol closed source driver outperforms open ones, because there's no open documentation. It's pointless to talk with trolls like you. CU
So I guess I'll neither get the options you used nor some Intel results of you. Fine for me, but I'll use this thread as a "this guy has a big mouth with nothing behind it" for your next attacks against nvidia or any other forum user.

Thanks in advance for your understanding.

Fuchs
NvFuchs is offline   Reply With Quote
Old 04-15-08, 05:23 PM   #99
pawels133
Registered User
 
pawels133's Avatar
 
Join Date: Mar 2008
Posts: 201
Default Re: 8400M GS : Low performences

Are you nvidia guy? Do you know how to help resolve my problem with nvidia drivers? If you don't you're just trolling and attacking people who's got problems with their cards. Do you need my xorg.conf, lspci, dmesg? Now, about my above test. I ran it few times and results was +/- 1sec. What settings do you want?

Quote:
Compared to?
Then what are the nv and vesa results,
if 27 seconds (+/- half the time vesa needs)
are very slow?
Are you joking?! Binary drivers result - 27 seconds (but it's 34 seconds using newest driver; using your xorg it was 36 seconds) is VERY low. As far as I know nv is in some way supported by nvidia and it's worst than vesa...

Quote:
polite as ever, and yes, of course it is fanboy bull****
Just naming things as they are.
__________________
Not everybody can smoke and drink... xd
pawels133 is offline   Reply With Quote
Old 04-15-08, 05:40 PM   #100
NvFuchs
Registered Fox
 
NvFuchs's Avatar
 
Join Date: Oct 2007
Location: Confoederatio Helvetica
Posts: 530
Default Re: 8400M GS : Low performences

Quote:
Originally Posted by pawels133
Are you nvidia guy?
No.

Quote:
Do you know how to help resolve my problem with nvidia drivers?
I allready did, here as an example:
http://www.nvnews.net/vbulletin/show...48#post1603648

But I can't solve every single problem you have.
You also wont expect a car mechanic to solve all your problems
over the internet when doesn't even know the car you got.

Quote:
If you don't you're just trolling and attacking people who's got problems with their cards.
Re-read the quote I gave you about what trolling is.
And no, I am not attacking, I am trying to help,
but I also tell people that not every single fault is nvidias fault.
Read the past threads, many things where BIOS settings,
faulty packages (pink shadows in compiz) or similar things.

Quote:
Do you need my xorg.conf, lspci, dmesg? Now, about my above test. I ran it few times and results was +/- 1sec. What settings do you want?
I just want to know which settings you used.
If you used 1000 runs the numbers are fine, in my opinion.

Quote:
Are you joking?! Binary drivers result - 27 seconds (but it's 34 seconds using newest driver; using your xorg it was 36 seconds) is VERY low.
Says who? Compared to other (1000 runs) results here in the
forums and on the authors page it is rather fast.
For 100 runs it would be low, yes.

Fuchs
NvFuchs is offline   Reply With Quote
Old 04-15-08, 06:35 PM   #101
txf
Registered User
 
Join Date: Feb 2008
Posts: 272
Default Re: 8400M GS : Low performences

Quote:
Originally Posted by pawels133
I did some tests:
gtk-engines-qt, 173.08. Have you anything smart to say now (but without nvidia fanboy bull**** please)? Is 33.94secs alright using nvidia binary driver? LOL. Using 169.12 it was 27 secs which is very slow. I'll do some testing with Intel card too which performs much better in KDE4.
would you keep it civil please, there is no reason to be inflammatory. I would love to see your xrenderbenchmark results from intel in the entirety (I suspect they accelerate more operations) as I lost the results I had

My xrenderbenchmark results using ipp=2 no glyphcache (not relevant anyways)
Code:
X Server from: The X.Org Foundation, Release: 10400090
	Xrender version: 0.10
---------------------------------------------
Test: PictOpClear
		 Plain............................................ Time: 0.022 sec.
		 Plain With Alpha................................. Time: 0.016 sec.
		 Transformation................................... Time: 0.018 sec.
		 Transformation/Bilinear filter................... Time: 0.018 sec.
Test: PictOpSrc
		 Plain............................................ Time: 0.019 sec.
		 Plain With Alpha................................. Time: 0.020 sec.
		 Transformation................................... Time: 0.013 sec.
		 Transformation/Bilinear filter................... Time: 0.017 sec.
Test: PictOpDst
		 Plain............................................ Time: 0.000 sec.
		 Plain With Alpha................................. Time: 0.000 sec.
		 Transformation................................... Time: 0.000 sec.
		 Transformation/Bilinear filter................... Time: 0.000 sec.
Test: PictOpOver
		 Plain............................................ Time: 0.017 sec.
		 Plain With Alpha................................. Time: 0.027 sec.
		 Transformation................................... Time: 0.020 sec.
		 Transformation/Bilinear filter................... Time: 0.017 sec.
Test: PictOpOverReverse
		 Plain............................................ Time: 0.000 sec.
		 Plain With Alpha................................. Time: 0.000 sec.
		 Transformation................................... Time: 0.000 sec.
		 Transformation/Bilinear filter................... Time: 0.000 sec.
Test: PictOpIn
		 Plain............................................ Time: 0.017 sec.
		 Plain With Alpha................................. Time: 0.020 sec.
		 Transformation................................... Time: 0.017 sec.
		 Transformation/Bilinear filter................... Time: 0.017 sec.
Test: PictOpInReverse
		 Plain............................................ Time: 0.000 sec.
		 Plain With Alpha................................. Time: 0.024 sec.
		 Transformation................................... Time: 0.000 sec.
		 Transformation/Bilinear filter................... Time: 0.000 sec.
Test: PictOpOut
		 Plain............................................ Time: 0.019 sec.
		 Plain With Alpha................................. Time: 0.020 sec.
		 Transformation................................... Time: 0.018 sec.
		 Transformation/Bilinear filter................... Time: 0.017 sec.
Test: PictOpOutReverse
		 Plain............................................ Time: 0.020 sec.
		 Plain With Alpha................................. Time: 0.024 sec.
		 Transformation................................... Time: 0.018 sec.
		 Transformation/Bilinear filter................... Time: 0.017 sec.
Test: PictOpAtop
		 Plain............................................ Time: 0.020 sec.
		 Plain With Alpha................................. Time: 0.027 sec.
		 Transformation................................... Time: 0.020 sec.
		 Transformation/Bilinear filter................... Time: 0.017 sec.
Test: PictOpAtopReverse
		 Plain............................................ Time: 0.000 sec.
		 Plain With Alpha................................. Time: 0.024 sec.
		 Transformation................................... Time: 0.000 sec.
		 Transformation/Bilinear filter................... Time: 0.000 sec.
Test: PictOpXor
		 Plain............................................ Time: 0.019 sec.
		 Plain With Alpha................................. Time: 0.024 sec.
		 Transformation................................... Time: 0.018 sec.
		 Transformation/Bilinear filter................... Time: 0.017 sec.
Test: PictOpAdd
		 Plain............................................ Time: 0.023 sec.
		 Plain With Alpha................................. Time: 0.023 sec.
		 Transformation................................... Time: 0.021 sec.
		 Transformation/Bilinear filter................... Time: 0.020 sec.
Test: PictOpSaturate
		 Plain............................................ Time: 11.151 sec.
		 Plain With Alpha................................. Time: 11.675 sec.
		 Transformation................................... Time: 12.435 sec.
		 Transformation/Bilinear filter................... Time: 14.814 sec.
Test: PictOpDisjointClear
		 Plain............................................ Time: 0.215 sec.
		 Plain With Alpha................................. Time: 0.215 sec.
		 Transformation................................... Time: 0.672 sec.
		 Transformation/Bilinear filter................... Time: 0.663 sec.
Test: PictOpDisjointSrc
		 Plain............................................ Time: 0.015 sec.
		 Plain With Alpha................................. Time: 0.019 sec.
		 Transformation................................... Time: 0.607 sec.
		 Transformation/Bilinear filter................... Time: 0.881 sec.
Test: PictOpDisjointDst
		 Plain............................................ Time: 0.000 sec.
		 Plain With Alpha................................. Time: 0.000 sec.
		 Transformation................................... Time: 0.000 sec.
		 Transformation/Bilinear filter................... Time: 0.000 sec.
Test: PictOpDisjointOver
		 Plain............................................ Time: 31.606 sec.
		 Plain With Alpha................................. Time: 50.712 sec.
		 Transformation................................... Time: 50.273 sec.
		 Transformation/Bilinear filter................... Time: 50.134 sec.
Test: PictOpDisjointOverReverse
		 Plain............................................ Time: 49.690 sec.
		 Plain With Alpha................................. Time: 49.634 sec.
		 Transformation................................... Time: 50.132 sec.
		 Transformation/Bilinear filter................... Time: 49.496 sec.
Test: PictOpDisjointIn
		 Plain............................................ Time: 49.315 sec.
		 Plain With Alpha................................. Time: 49.993 sec.
		 Transformation................................... Time: 50.611 sec.
		 Transformation/Bilinear filter................... Time: 52.630 sec.
Test: PictOpDisjointInReverse
		 Plain............................................ Time: 53.264 sec.
		 Plain With Alpha................................. Time: 51.271 sec.
		 Transformation................................... Time: 49.777 sec.
		 Transformation/Bilinear filter................... Time: 50.391 sec.
Test: PictOpDisjointOut
		 Plain............................................ Time: 49.622 sec.
		 Plain With Alpha................................. Time: 49.860 sec.
		 Transformation................................... Time: 50.089 sec.
		 Transformation/Bilinear filter................... Time: 50.039 sec.
you're getting 12 seconds where I'm getting 50...are you using a powemizer hack?

there are others but nvidia have stated that they are not accelerated as they are not commonly used operations. It could be the reason certain functions are slow or not (I need to recompile qt without debugging symbols before seeing how kde4 handles).

@fuchs:
You're getting 12 seconds where I'm getting 50...are you using a powemizer hack? during those operations my perf level has sunk down to 0...if I peg it at 2 it's closer to your results.

With ipp=1 my results are a lot slower almost 2 times sometimes more. Have you tried setting it at 4 (gpu accesible sysmem).
txf is offline   Reply With Quote
Old 04-15-08, 06:47 PM   #102
NvFuchs
Registered Fox
 
NvFuchs's Avatar
 
Join Date: Oct 2007
Location: Confoederatio Helvetica
Posts: 530
Default Re: 8400M GS : Low performences

Quote:
@fuchs:
You're getting 12 seconds where I'm getting 50...are you using a powemizer hack? during those operations my perf level has sunk down to 0...if I peg it at 2 it's closer to your results.
Small script which causes powermizer to stay on level 2
as long as the notebook is connected to a AC power source.
So yes, I'd call this hack as well, but best thing until nvidia
makes it configurable otherwhise.

Posted it here about twice, but:

Code:
#!/bin/sh

while true; do
    
    powerstate=`cat /proc/acpi/ac_adapter/AC/state | awk '{print $2}'`
    
    if [ $powerstate = "on-line"  ]; then
       nvidia-settings -q all > /dev/null
    fi
    sleep 25;
done
You might have to change it according to your
/proc/acpi/ac_adapter/AC/state
(or even change this path)

hth,

Fuchs
NvFuchs is offline   Reply With Quote
Old 04-16-08, 06:10 AM   #103
pawels133
Registered User
 
pawels133's Avatar
 
Join Date: Mar 2008
Posts: 201
Default Re: 8400M GS : Low performences

Quote:
Says who? Compared to other (1000 runs) results here in the
forums and on the authors page it is rather fast.
For 100 runs it would be low, yes.
Is it something like this: gtkperf -a -c 1000 (I hope it's not and better don't ask why ).
__________________
Not everybody can smoke and drink... xd
pawels133 is offline   Reply With Quote
Old 04-16-08, 06:16 AM   #104
NvFuchs
Registered Fox
 
NvFuchs's Avatar
 
Join Date: Oct 2007
Location: Confoederatio Helvetica
Posts: 530
Default Re: 8400M GS : Low performences

Quote:
Originally Posted by pawels133
Is it something like this: gtkperf -a -c 1000 (I hope it's not and better don't ask why ).
According to gtkperf --help that'd be:

-c number of counts, so: 1000, in your case
-a automatic run the tests

you could configure -c in the gui as well, but as I expected:
your results are from a 1000 run test, mine are from the
default, which is 100. So your results are, in my opinion
and compared to others, just fine.

Fuchs
NvFuchs is offline   Reply With Quote

Old 04-16-08, 06:28 AM   #105
pawels133
Registered User
 
pawels133's Avatar
 
Join Date: Mar 2008
Posts: 201
Default Re: 8400M GS : Low performences

Thanks, but those are my results using "gtkperf -a -c 1000":

GtkPerf 0.40 - Starting testing: Wed Apr 16 11:58:28 2008

GtkEntry - time: 2.35
GtkComboBox - time: 70.70
GtkComboBoxEntry - time: 60.07
GtkSpinButton - time: 13.54
GtkProgressBar - time: 9.56
GtkToggleButton - time: 20.84
GtkCheckButton - time: 22.02
GtkRadioButton - time: 29.55
GtkTextView - Add text - time: 136.38
GtkTextView - Scroll - time: 42.45
GtkDrawingArea - Lines - time: 9.12
GtkDrawingArea - Circles - time: 12.45
GtkDrawingArea - Text - time: 25.57
GtkDrawingArea - Pixbufs - time: 4.17
---
Total time: 458.80

My previous result was using 'clean' gtkperf so it was something like "gtkperf -a -c 100". I've reinstaled system and installed drivers from nvidia site. I didn't change any options.

I must apologize for my English and my tone in previous posts. I was in some way desperate.
__________________
Not everybody can smoke and drink... xd
pawels133 is offline   Reply With Quote
Old 04-16-08, 07:21 AM   #106
NvFuchs
Registered Fox
 
NvFuchs's Avatar
 
Join Date: Oct 2007
Location: Confoederatio Helvetica
Posts: 530
Default Re: 8400M GS : Low performences

Okay, this is far too slow.
Sane values should be +/-

fuchs@thinkfox src $ ./gtkperf -a -c 1000
GtkPerf 0.40 - Starting testing: Wed Apr 16 13:15:24 2008

GtkEntry - time: 0.24
GtkComboBox - time: 16.25
GtkComboBoxEntry - time: 13.89
GtkSpinButton - time: 1.25
GtkProgressBar - time: 2.27
GtkToggleButton - time: 4.58
GtkCheckButton - time: 5.61
GtkRadioButton - time: 8.46
GtkTextView - Add text - time: 23.33
GtkTextView - Scroll - time: 19.05
GtkDrawingArea - Lines - time: 7.06
GtkDrawingArea - Circles - time: 10.28
GtkDrawingArea - Text - time: 4.45
GtkDrawingArea - Pixbufs - time: 0.14
---
Total time: 116.86

Take again a look at my Xorg.conf here
http://www.thinkwiki.org/wiki/Instal..._R61#xorg.conf

And try several InitialPixmapPlacement values.
Here 1 works fine, but usually 2 should perform better.

Then please tell me what distribution,
what version of X11 you are using,
whether this is x86 or x86_64,
what driver version etc.

Then I might be able to help you.
(No guarantee on this, though.
as a) I am not a nvidia dev and
b) I don't have physical access to your machine)


And, of course, apology for the tone accepted,
I know how desperate you must feel when a high end,
expensive GPU has a performance this low.
But this is, in many cases, just not nvidias fault.
(not in all, though. But bugs usually get fixed)

No apologies needed for your english,
english is not everybodys native language,
it is mine neither. As long as everybody understands what you try to say it is fine.

Fuchs
NvFuchs is offline   Reply With Quote
Old 04-16-08, 08:44 AM   #107
pawels133
Registered User
 
pawels133's Avatar
 
Join Date: Mar 2008
Posts: 201
Default Re: 8400M GS : Low performences

Quote:
Take again a look at my Xorg.conf here
http://www.thinkwiki.org/wiki/Instal...61#xorg.co nf

And try several InitialPixmapPlacement values.
Here 1 works fine, but usually 2 should perform better
I used options from your xorg and I did tests with different InitialPixmapPlacement values. It's strange, but there wasn't almost any difference in performance +/- 3secs (there was some improvement with 169.12 and I'll do tests with this driver too). I did only gtkperf -a -c 100.

Quote:
Then please tell me what distribution,
what version of X11 you are using,
whether this is x86 or x86_64,
what driver version etc
I'm using Kubuntu 8.04 right now, but I used earlier versions before, Ubuntu and Xubuntu 7.10, Fedora 8 and Archlinux and performance was comparable. I'll install Archlinux and do some more testing.
X.Org X Server 1.4.0.90, nvidia driver 173.08.

When I typed: nvidia-settings --glxinfo there was an error:

ERROR: Error fetching GLX Information: Unknown Error

Maybe there are some library conflicts. When drawing box by a cursor on desktop, xorg cpu usage is almost 100%. Direct rendering is enabled and 3D acceleration works. I'm not using laptop so I suppose that powermizer doesn't affect performance.

@NvFuchs

Thanks for your patience.
__________________
Not everybody can smoke and drink... xd
pawels133 is offline   Reply With Quote
Old 04-16-08, 09:00 AM   #108
Kirurgs
Registered User
 
Join Date: Jul 2007
Posts: 165
Default Re: 8400M GS : Low performences

Just to compare results: T61, C2D 2.2, NVS 140M, F8 x86_64, 173.08, only "tweak" I have is NvAgp=0, as I have PCIE card...
Findings: ipp=0 + gc=1 = fastest for me

========================================
$ nvidia-settings -a InitialPixmapPlacement=0 -a GlyphCache=1

Attribute 'InitialPixmapPlacement' (xxx:0.0) assigned value 0.

Attribute 'GlyphCache' (xxx:0.0) assigned value 1.

$ gtkperf -a -c 1000
GtkPerf 0.40 - Starting testing: Wed Apr 16 15:45:24 2008

GtkEntry - time: 0.16
GtkComboBox - time: 13.18
GtkComboBoxEntry - time: 11.69
GtkSpinButton - time: 1.48
GtkProgressBar - time: 3.58
GtkToggleButton - time: 5.10
GtkCheckButton - time: 4.40
GtkRadioButton - time: 5.60
GtkTextView - Add text - time: 37.71
GtkTextView - Scroll - time: 15.66
GtkDrawingArea - Lines - time: 4.18
GtkDrawingArea - Circles - time: 4.68
GtkDrawingArea - Text - time: 3.52
GtkDrawingArea - Pixbufs - time: 7.09
---
Total time: 118.04

Quitting..
========================================

Comparison gc vs nogc:

========================================
$ nvidia-settings -a InitialPixmapPlacement=0 -a GlyphCache=0

Attribute 'InitialPixmapPlacement' (xxx:0.0) assigned value 0.

Attribute 'GlyphCache' (xxx:0.0) assigned value 0.

$ gtkperf -a -c 1000
GtkPerf 0.40 - Starting testing: Wed Apr 16 15:51:50 2008

GtkEntry - time: 0.00
GtkComboBox - time: 1.34
GtkComboBoxEntry - time: 1.18
GtkSpinButton - time: 0.16
GtkProgressBar - time: 0.36
GtkToggleButton - time: 0.54
GtkCheckButton - time: 0.44
GtkRadioButton - time: 0.58
GtkTextView - Add text - time: 0.49
GtkTextView - Scroll - time: 0.29
GtkDrawingArea - Lines - time: 0.88
GtkDrawingArea - Circles - time: 0.47
GtkDrawingArea - Text - time: 11.86
GtkDrawingArea - Pixbufs - time: 0.86
---
Total time: 19.47

Quitting..
$ nvidia-settings -a InitialPixmapPlacement=0 -a GlyphCache=1

Attribute 'InitialPixmapPlacement' (xxx:0.0) assigned value 0.

Attribute 'GlyphCache' (xxx:0.0) assigned value 1.

$ gtkperf -a -c 100
GtkPerf 0.40 - Starting testing: Wed Apr 16 15:52:15 2008

GtkEntry - time: 0.00
GtkComboBox - time: 1.31
GtkComboBoxEntry - time: 1.15
GtkSpinButton - time: 0.16
GtkProgressBar - time: 0.36
GtkToggleButton - time: 0.53
GtkCheckButton - time: 0.43
GtkRadioButton - time: 0.57
GtkTextView - Add text - time: 0.48
GtkTextView - Scroll - time: 0.28
GtkDrawingArea - Lines - time: 0.36
GtkDrawingArea - Circles - time: 0.47
GtkDrawingArea - Text - time: 0.38
GtkDrawingArea - Pixbufs - time: 0.73
---
Total time: 7.22

Quitting..
========================================

That's about 3 times better...
Kirurgs is offline   Reply With Quote
Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


Similar Threads
Thread Thread Starter Forum Replies Last Post
8800 GT and Ubuntu - Plymouth low resolution legluondunet NVIDIA Linux 3 06-24-12 04:22 PM
Video: Achieving Ultra-low Latency in the Cloud: How Low Can We Go? News Archived News Items 0 06-18-12 05:40 AM
Sluggish performance 8400M GS Ubuntu 12.04 Unity 3D bgneal NVIDIA Linux 1 05-24-12 07:16 PM
OCZ Releases Low Profile Version of Vertex 3 News Archived News Items 0 05-18-12 10:30 PM
Low OpenGL Performance Instinct NVIDIA Linux 10 08-08-02 04:56 PM

All times are GMT -5. The time now is 05:26 AM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 1998 - 2014, nV News.