[PC-BSD Testing] PCBSD 8.0-RC-x64 speed issues with nvidia.

Ian Robinson fitchkendall at gmail.com
Mon Feb 1 15:31:29 PST 2010

>Message: 2
>Date: Mon, 01 Feb 2010 08:25:22 -0500
>From: Kris Moore <kris at pcbsd.com>
>Subject: Re: [PC-BSD Testing] PCBSD 8.0-RC-x64 Speed Issues
>To: testing at lists.pcbsd.org
>Message-ID: <4B66D642.7010705 at pcbsd.com>
>Content-Type: text/plain; charset=ISO-8859-1; format=flowed

>On 01/31/2010 16:52, Ian Robinson wrote:
>> PCBSD 8.0-RC-x64 -- Speed issues:
>> On 1/31/10 in Testing Digest, Vol 36, Issue 69, I mentioned screen
>> lag, web page loading delays, and screen corruption.
>> I am running PCBSD 8.0-RC-x64 in the Live Mode off a USB stick.  It
>> runs faster than the installation to the hard disk.  I don't have any
>> screen corruption.
>> The hard disk installation uses NVidia's 64 bit binary driver. The
>> Live Version has used "nv".
>> I wonder wonder if that NVidia 64-bit binary driver is having a
>> conflict with something in the 8.0-RC?
>That sounds feasible, you can test this easily by switching from the
>"nvidia" driver to "nv" on your installed system, does that make a huge
>FWIW, I'm using nvidia-64bit here, and it works fine on my 9800GT + duel
>head, no corruption that I can see. Do you have 3d effects turned off BTW?
> Kris Moore
>PC-BSD Software
>http://www.pcbsd.com <http://www.pcbsd.com/>

Still using the nvidia driver.  I'm not sure what you meant by turning off
3d effects -- I'm not running Compiz.

I eliminated most of the Desktop Effects, but it makes a really huge
difference to turn off compositing.

Initially, I tried to turn off compositing but suffered several crashes in
the Desktop module (System Settings > Desktop) each time I tried.
Eventually I  got it turned off.  Turning off compositing results in
Immediate speed increases, no menu or screen contamination, and no lag in
programs like Gwenview, Firefox, or Opera.

Ian Robinson
Salem, Ohio
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.pcbsd.org/pipermail/testing/attachments/20100201/2c02e721/attachment.html>

More information about the Testing mailing list