Hi, I am using BT878A and bttv-0.7.91 driver. I try to use only time-decimation part of your patch. But I get very strange result. For example, if I got the following images sequences: I1 I2 I3 I4 I5 I6 I7 I8 I9 I10 I11 I12 I13 I14 I15... Then the I7 will be same as I5 and I13 will be the same as I11... That's the case the small value of btv->decimation. If btv->decimation is bigger, then I5 will the same as I3 and I9 will be the same as I7... Does anyone has the same situation? Thanks! Regards, Kevin ----- Original Message ----- From: "Benoit PAPILLAULT" <bpapillault@xxxxxxxxxxxxx> To: <video4linux-list@xxxxxxxxxx> Sent: Monday, May 06, 2002 8:24 PM Subject: Fw: Re: Re: temporal decimation in bttv-0.7.91 driver [patch] > > Here is the patch. It has been generated with (I put a local cvs with 0.7.91 inside): > > $ cvs diff -u -r1.1 bttv-0.7.91 > bttv-decimation.diff > $ gzip bttv-decimation.diff > > Few notes about the decimation patch: > > - I only modified the RISC program generated by make_vrisctab(), mainly by inverting > odd and even frame capture. > - I added a small RISC disassembler which can be activated by the option "dis_risc=1" > in your /etc/modules.conf > > Few notes about the BTTV_MAX patch, generated by: > $ cvs diff -u -r decimation-ok bttv-0.7.91 > bttv-max.diff > $ gzip bttv-max.diff > > - I modify the Makefile to generate 16 /dev/video? entries and alias in /etc/modules.conf > - I apply modifications already mentionned by Gerd Knorr > - I only test for video (no audio, no tuner, ...). > > Benoit, now using 16 cards at the same time! > > PS: For the moderators, disregard my old email. I think using gzip should be ok this time. > >
