I did notice a very thin line at the bottom of the frame and at the right.
I'll do some more
adjust with the h v offsets. But its absolutely acceptable.
#if 0
static const unsigned char init[] = {
0x00, 0x00, /* 00 - ID byte */
0x01, 0x00, /* 01 - reserved */
/*front end */
0x02, 0xd0, /* 02 - FUSE=3, GUDL=2, MODE=0 */
0x03, 0x23, /* 03 - HLNRS=0, VBSL=1, WPOFF=0, HOLDG=0, GAFIX=0,
GAI1=256, GAI2=256 */
0x04, 0x00, /* 04 - GAI1=256 */
0x05, 0x00, /* 05 - GAI2=256 */
/* decoder */
0x06, 0xf6, /* 06 - HSB at 13(50Hz) / 17(60Hz) pixels after end
of last line */
0x07, 0xdd, /* 07 - HSS at 113(50Hz) / 117(60Hz) pixels after end
of last line */
0x08, 0xc8, /* 08 - AUFD=1, FSEL=1, EXFIL=0, VTRC=1, HPLL=0, VNOI=0
*/
0x09, 0x01, /* 09 - BYPS=0, PREF=0, BPSS=0, VBLB=0, UPTCV=0, APER=1
*/
0x0a, 0x80, /* 0a - BRIG=128 */
0x0b, 0x47, /* 0b - CONT=1.109 */
0x0c, 0x40, /* 0c - SATN=1.0 */
0x0d, 0x00, /* 0d - HUE=0 */
0x0e, 0x01, /* 0e - CDTO=0, CSTD=0, DCCF=0, FCTC=0, CHBW=1 */
0x0f, 0x00, /* 0f - reserved */
0x10, 0x48, /* 10 - OFTS=1, HDEL=0, VRLN=1, YDEL=0 */
0x11, 0x1c, /* 11 - GPSW=0, CM99=0, FECO=0, COMPO=1, OEYC=1,
OEHV=1, VIPB=0, COLO=0 */
0x12, 0x00, /* 12 - output control 2 */
0x13, 0x00, /* 13 - output control 3 */
0x14, 0x00, /* 14 - reserved */
0x15, 0x00, /* 15 - VBI */
0x16, 0x00, /* 16 - VBI */
0x17, 0x00, /* 17 - VBI */
};
#endif
static const unsigned char init[] = {//this are the reset values from the
datasheet
0x00, 0x00, /* 00 - ID byte */
0x01, 0x00, /* 01 - reserved */
/*front end */
0x02, 0xc0, /* 02 - FUSE=3, GUDL=2, MODE=0 */
0x03, 0x23, /* 03 - HLNRS=0, VBSL=1, WPOFF=0, HOLDG=0, GAFIX=0,
GAI1=256, GAI2=256 */
0x04, 0x00, /* 04 - GAI1=256 */
0x05, 0x00, /* 05 - GAI2=256 */
/* decoder */
0x06, 0xeb, /* 06 - HSB at 13(50Hz) / 17(60Hz) pixels after end
of last line */
0x07, 0xe0, /* 07 - HSS at 113(50Hz) / 117(60Hz) pixels after end
of last line */
0x08, 0x08, /* 08 - AUFD=1, FSEL=1, EXFIL=0, VTRC=1, HPLL=0, VNOI=0
*/
0x09, 0x01, /* 09 - BYPS=0, PREF=0, BPSS=0, VBLB=0, UPTCV=0, APER=1
*/
0x0a, 0x88, /* 0a - BRIG=128 */
0x0b, 0x44, /* 0b - CONT=1.109 */
0x0c, 0x44, /* 0c - SATN=1.0 */
0x0d, 0x04, /* 0d - HUE=0 */
0x0e, 0x01, /* 0e - CDTO=0, CSTD=0, DCCF=0, FCTC=0, CHBW=1 */
0x0f, 0x00, /* 0f - reserved */
0x10, 0x40, /* 10 - OFTS=1, HDEL=0, VRLN=1, YDEL=0 */
0x11, 0x1c, /* 11 - GPSW=0, CM99=0, FECO=0, COMPO=1, OEYC=1,
OEHV=1, VIPB=0, COLO=0 */
0x12, 0x01, /* 12 - output control 2 */
0x13, 0x00, /* 13 - output control 3 */
0x14, 0x00, /* 14 - reserved */
0x15, 0x00, /* 15 - VBI */
0x16, 0x00, /* 16 - VBI */
0x17, 0x00, /* 17 - VBI */
};
I noticed that there is an if statement that was not in the older version.
The if statement contains if 0 to initialize int char[].
I'm not sure the value of "if 0", is this an error status? In the older
version after the "if 0" condition there is another declaration for
the char[]. It seems to be a reset of the saa7111a chipset. Maybe if we have
the reset in there and the reset will activate the colour.
I'm just scare of breaking the PAL stuff to get the NTSC stuff working. Also,
maybe different devices like web cams and TV cards will require different device
information. The driver could get really messy. Fix one problem, create
another problem with another device that is attached. Or add another device and
it
breaks the other supported devices. It would be nice to probe the NT1003 or
NT1004 for the information.
I was thinking about that. But I was concerned that the changes that you made
were necessary. So I did not want to just move the image over to get rid of
the green line. I noticed in some of the functions that there were register
values that were different.
Does your PAL devices show colour?
Dwaine.
Jörg Heckenbach <[EMAIL PROTECTED]> on 04/10/2001 10:09:05 AM
Please respond to [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
cc: (bcc: Dwaine Garden/ISD/WCBO)
Subject: Re: [linux-usb-devel] New Device added to USBVISION code
Hi Dwaine,
the picture looks good. I'll take the values from your site.
[EMAIL PROTECTED] wrote:
>
> I have better input values for NTSC video. Also, I have identified a
problem
> in the saa7111-new that seems to result in no colour.
Give me some time to check my last changes in saa7111-new.c
>
> If I use the old version of saa7111-new I do get colour, but with a green line
> on the left hand side of the video output.
Maybe we can delete the green line by increasing the h-offset in
usbvision_set_output.
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
http://lists.sourceforge.net/lists/listinfo/linux-usb-devel
