[Mjpeg-users] Mjpeg-users] Another problem with very old VHS tapes

2006-11-08 Thread Michael Hanke
Hi, I have another capturing problem with very old VHS tapes. I am using a Canopus ADVC-55 for capturing to PAL-DV. Even if my tapes are very old, I do not have any problem with lost frames, drops outs etc. My problem seems to be a synchronization issue of the individual scan lines. In some fra

Re: [Mjpeg-users] Mjpeg-users] Another problem with very old VHS tapes

2006-11-08 Thread Steven M. Schultz
On Wed, 8 Nov 2006, Michael Hanke wrote: > I have another capturing problem with very old VHS tapes. I am using a Canopus > ADVC-55 for capturing to PAL-DV. Even if my tapes are very old, I do not have > any problem with lost frames, drops outs etc. My problem seems to be a That is beca

Re: [Mjpeg-users] Mjpeg-users] Another problem with very old VHS tapes

2006-11-08 Thread stefan
> You might try putting the data thru the new and improved deinterlacer > and creating a progressive stream. Maybe that will 'blend' the two > fields together into a more stable image. > I fear that this will not help much, as the new deinterlacer core really needs motion fo

Re: [Mjpeg-users] Mjpeg-users] Another problem with very old VHS tapes

2006-11-09 Thread Michael Hanke
Am Donnerstag, 9. November 2006 00:51 schrieb stefan: > > You might try putting the data thru the new and improved deinterlacer > > and creating a progressive stream. Maybe that will 'blend' the two > > fields together into a more stable image. > > I fear that this will not help much,

Re: [Mjpeg-users] Mjpeg-users] Another problem with very old VHS tapes

2006-11-09 Thread stefan
Michael Hanke schrieb: > Do you have more ideas? > probably not really ... Here is what I would do for a start: 1. Divide the frame (the two fields...) into 5 regions: Top-left, Top-right, Bottom-left, Bottom-right, Center. Let all these regions be blocks of 1/5th of the image-si