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

Newegg Daily Deals

Reply
 
Thread Tools
Old 12-15-05, 01:56 PM   #13
Dr. Tyrell
Registered User
 
Dr. Tyrell's Avatar
 
Join Date: Aug 2004
Posts: 100
Default Re: 8174 on Toshiba Satellite with ACPI "legacy free"

Okay, I altered modprobe.conf with softedids=1 and getfromROM=1
also did what you asked in xorg.conf and commented anything that might interfere.

Looks like 8174 DOES read edid from these toshibas, go figure.
But, like always in the past, detects only 1372 horiz resolution, not 1400, so I get that black space at right.
And continues to detect max dot clock of 108Mhz, and i cannot get any program or website to create 1400x1050x60Hz with 108Mhz!

Could it be the Toshiba bios?
Could it be Nvidia drivers?

I must also mention that console switching is still only possible with vga=normal in lilo.conf
Attached Files
File Type: zip nvidia-bug-report.zip (28.0 KB, 105 views)
Dr. Tyrell is offline   Reply With Quote
Old 12-15-05, 02:04 PM   #14
netllama
NVIDIA Corporation
 
Join Date: Dec 2004
Posts: 8,763
Default Re: 8174 on Toshiba Satellite with ACPI "legacy free"

The EDID in the bug report you posted shows:

(--) NVIDIA(0): Detailed Timings:
(--) NVIDIA(0): 1372 x 1050 @ 60Hz
(--) NVIDIA(0): Pixel Clock : 108.00 MHz
(--) NVIDIA(0): HRes, HSyncStart : 1372, 1448
(--) NVIDIA(0): HSyncEnd, HTotal : 1560, 1688
(--) NVIDIA(0): VRes, VSyncStart : 1050, 1051
(--) NVIDIA(0): VSyncEnd, VTotal : 1054, 1066
(--) NVIDIA(0): H/V Polarity : -/-

I'm afraid I don't see anything wrong here.

-Lonni
netllama is offline   Reply With Quote
Old 12-15-05, 07:39 PM   #15
Dr. Tyrell
Registered User
 
Dr. Tyrell's Avatar
 
Join Date: Aug 2004
Posts: 100
Default Re: 8174 on Toshiba Satellite with ACPI "legacy free"

(--) NVIDIA(0): 1372 x 1050 @ 60Hz
(--) NVIDIA(0): Pixel Clock : 108.00 MHz

BIG problems there.
Should be 1400x1050 and 224 MHz

All Nvidia Linux proprietary drivers since 5336 have misdetected this.
Xorg's nv driver detects properly.

And I must say once again that because of that 108.00 MHz I'm therefore limited to modelines calculated with 108.00, and therefore cannot get all the possible modes of this display.

In fact, I can only get 1400x1050 properly, and at 56 Hz. I lost count but have tried more than 40 modelines from different sources and the two noted in my xorg.conf are the only ones that give near-full refresh at 1400.

The "black bar" issue is referred to many many times in this forum, and now that I am limited to that 108 I am very very limited.

What's causing that 108? Toshiba or you guys?

The 1372 can be dealt with IFF the 108 can be raised.

Thanks again and I think this thread is relevant for lots of people.
Dr. Tyrell is offline   Reply With Quote
Old 12-15-05, 07:45 PM   #16
netllama
NVIDIA Corporation
 
Join Date: Dec 2004
Posts: 8,763
Default Re: 8174 on Toshiba Satellite with ACPI "legacy free"

The 108Mhz pixelclock is a hardware limitation.

You state that drivers since 1.0-5336 have misdetected the native resolution and pixelclock. Please install one of the drivers before 1.0-5336, start X with the following command:
startx -- logverbose 6
and then post your X log.

Thanks,
Lonni
netllama is offline   Reply With Quote
Old 12-15-05, 08:34 PM   #17
Dr. Tyrell
Registered User
 
Dr. Tyrell's Avatar
 
Join Date: Aug 2004
Posts: 100
Default Re: 8174 on Toshiba Satellite with ACPI "legacy free"

I was unclear on that, I meant that the Clock is now misdetected, the 1378x1050 has been a problem all along, hence the need for modelines.

The clock has been 224MHz up thru 7667, the most recent driver i used.
Now, 8174 claims 108MHz, and now refuses to allow modelines calculated above that, even with IgnoreEDID=1.

So, since prior Nvidia drivers have acted the same, I attached 7667 in two flavors, EDID used with no modelines and EDID ignored with a known-good modeline from gtf.

You'll notice that 7667 with EDIDs thinks things are too fast for 110MHz,and there's the rub. For 7667 WITHOUT EDIDs and a good modeline IS ALLOWED.

I reiterate that Xorg's nv driver works perfectly, in all resolutions.

Also, I have not tried AgpGart, and probably will not.

Thanks tons for the super quick and pointed replies
Attached Files
File Type: zip nvidia-bug-report_NOmodelines.zip (22.9 KB, 127 views)
File Type: zip nvidia-bug-report_YESmodeline.zip (22.6 KB, 126 views)
Dr. Tyrell is offline   Reply With Quote
Old 12-15-05, 08:50 PM   #18
netllama
NVIDIA Corporation
 
Join Date: Dec 2004
Posts: 8,763
Default Re: 8174 on Toshiba Satellite with ACPI "legacy free"

There was a bug in the pixelclock detection up through 1.0-7667 where it was mis-reported (sometimes as high as 400Mhz on some systems).

Do you have an X log that indicates that the 'nv' driver is actually running 1400x1050@60Hz and also detecting 1400x1050@60Hz from the EDID?

Is there a reason why you haven't tried agpgart?

Thanks,
Lonni
netllama is offline   Reply With Quote
Old 12-15-05, 09:40 PM   #19
Dr. Tyrell
Registered User
 
Dr. Tyrell's Avatar
 
Join Date: Aug 2004
Posts: 100
Default Re: 8174 on Toshiba Satellite with ACPI "legacy free"

Not used agpgart because a) don't want to, your glx seems better, b) haven't gotten around to it.

Attached is the xorg log from nv driver. With it my screen goes to ALL resolutions without a hitch.

Note the lines in xorg.0.log.....
"(II) NV(0): Mode "1600x1200" is larger than BIOS programmed panel size of 1400 x 1050. Removing."
"(--) NV(0): Panel size is 1400 x 1050
(II) NV(0): Panel is LVDS"

I highlight in blue the very important part.
And you'll also note the valid modelines...there's a ton of them, and all clock speeds work.

Yes, i see that the max 350MHz exists, but no modelines even attempt it.
Attached Files
File Type: zip Xorg.0.log_nv_logverbose5.zip (7.0 KB, 108 views)
File Type: zip Xorg.0.log-8174.logverbose5_YesEdids.zip (6.4 KB, 118 views)
File Type: zip Xorg.0.log-8174.logverbose5_NoEdids.zip (6.2 KB, 110 views)
Dr. Tyrell is offline   Reply With Quote
Old 12-15-05, 11:24 PM   #20
Dr. Tyrell
Registered User
 
Dr. Tyrell's Avatar
 
Join Date: Aug 2004
Posts: 100
Default Re: 8174 on Toshiba Satellite with ACPI "legacy free"

"The 108Mhz pixelclock is a hardware limitation." , whaaa?

Also, I attached proper logverbose'd logs above.

*footnote: I must say, even with this annoying and slightly creepy bug, everything including GL apps run just fine, a bit smoother I'd say, and even with Composite and transparency in KDE everything is smooth and perhaps a bit faster than before. Good job so far Nvidia!
Dr. Tyrell is offline   Reply With Quote

Old 12-16-05, 04:47 PM   #21
Dr. Tyrell
Registered User
 
Dr. Tyrell's Avatar
 
Join Date: Aug 2004
Posts: 100
Default Re: 8174 on Toshiba Satellite with ACPI "legacy free"

Just hoping to remind Toshiba legacy free users, NVREG_MOBILE=1
Dr. Tyrell 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
Toshiba Satellite 1955/S801, GeForce4 440 Go azaroth NVIDIA Linux 2 01-01-03 08:19 PM

All times are GMT -5. The time now is 07:00 AM.


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