Re: Attempt to solve "ERR: SCR moves backwards, remultiplex input."
|
View:
New views
3 Messages
—
Rating Filter:
Alert me
|
 |
Re: Attempt to solve "ERR: SCR moves backwards, remultiplex input."
by Lawrence D'Oliveiro-2
::
Rate this Message:
Back on 2nd April 2004, philou <philippe.sarazin@fr...> wrote:
> Like some of you, I had to face some trouble with a recurrent error > message in dvdauthor (v 0.6.10), "ERR: SCR moves backwards". > > This message occurs when the SCR (System Clock Reference) goes backward > in the data stream that I'm trying to author. After a lot of search on > the net, I think that there's a case where this is not an error as > dvdauthor reports it : the SCR can sometimes be reset to zero in the > middle of a VOB. I have decided to incorporate this patch. Well, I figure trying to recover and continue can’t be worse than dying with an error. :) ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ Dvdauthor-developer mailing list Dvdauthor-developer@... https://lists.sourceforge.net/lists/listinfo/dvdauthor-developer |
|
|
Re: Attempt to solve "ERR: SCR movesbackwards, remultiplex input."
by Martin Crossley
::
Rate this Message:
Hi Lawrence -
I did quite a bit of work on this issue back in Jun 2008: see http://www.mail-archive.com/ivtv-users@.../msg08234.html. Unfortunately for some reason the source code of the patches themselves seems to have vanished from that site, so I'll dig it out from archive and sent it to you. I was taking a lot of input from a hardware MPEG encoder card, that triggered this problem *a lot*. I wrote a bit about the causes of the issue here (see in particular "Part 2: how to patch dvdauthor"). My post here: http://old.nabble.com/Submission-of-proposed-patch-to-dvdauthor-0.6.14-(affects-dvdvob.c)-p17695878.html explains a bit more what the patches are doing The patches that I developed are probably a bit long in the tooth these days (they are against version 0.6.14) but I think that they might be helpful for you to have a look at. One warning though: while I found that the DVDs produced via this patch work absolutely fine with my players here (Sony, Denon), I have had reports from some friends that they don't work in certain older kit. Cheers - and good luck MartinC PS: for the benefit of the list, here is another document I wrote a while ago, on the subject of how to use the Gimp to do anti-aliased menu text with dvdauthor. It's well worth the effort! http://ivtvdriver.org/index.php/Howto:anti-aliased_menus_and_buttons_with_dvdauthor ----- Original Message ----- From: "Lawrence D'Oliveiro" <ldo@...> To: <dvdauthor-developer@...> Sent: Sunday, November 08, 2009 4:52 AM Subject: Re: [Dvdauthor-developer] Attempt to solve "ERR: SCR movesbackwards, remultiplex input." > Back on 2nd April 2004, philou <philippe.sarazin@fr...> wrote: > >> Like some of you, I had to face some trouble with a recurrent error >> message in dvdauthor (v 0.6.10), "ERR: SCR moves backwards". >> >> This message occurs when the SCR (System Clock Reference) goes backward >> in the data stream that I'm trying to author. After a lot of search on >> the net, I think that there's a case where this is not an error as >> dvdauthor reports it : the SCR can sometimes be reset to zero in the >> middle of a VOB. > > I have decided to incorporate this patch. Well, I figure trying to > recover and continue can’t be worse than dying with an error. :) > > ------------------------------------------------------------------------------ > Let Crystal Reports handle the reporting - Free Crystal Reports 2008 > 30-Day > trial. Simplify your report design, integration and deployment - and focus > on > what you do best, core application coding. Discover what's new with > Crystal Reports now. http://p.sf.net/sfu/bobj-july > _______________________________________________ > Dvdauthor-developer mailing list > Dvdauthor-developer@... > https://lists.sourceforge.net/lists/listinfo/dvdauthor-developer > ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ Dvdauthor-developer mailing list Dvdauthor-developer@... https://lists.sourceforge.net/lists/listinfo/dvdauthor-developer |
|
|
Re: Attempt to solve "ERR: SCR movesbackwards, remultiplex input."
by Martin Crossley
::
Rate this Message:
Here is a copy of my patch against version 0.6.14 (taken from
http://www.mail-archive.com/ivtv-users@.../msg08199.html): The only file that needs to be patched is dvdvob.c MartinC ============================================================= --- src/dvdvob_dist.c 2008-05-30 22:07:22.000000000 +0100 +++ src/dvdvob.c 2008-05-30 22:09:43.000000000 +0100 @@ -240,25 +240,51 @@ buf[2] == 1 && buf[3] == 0xba ) { + int offs = (buf[17] == 0xbb) ? 24 : 0; writescr(buf+4,readscr(buf+4)+tsoffs); // video/audio? // pts? - if( buf[14] == 0 && - buf[15] == 0 && - buf[16] == 1 && - (buf[17]==0xbd || (buf[17]>=0xc0 && buf[17]<=0xef)) && - (buf[21] & 128)) + if( buf[offs+14] == 0 && + buf[offs+15] == 0 && + buf[offs+16] == 1 && + (buf[offs+17]==0xbd || (buf[offs+17]>=0xc0 && buf[offs+17]<=0xef)) && + (buf[offs+21] & 128)) { - writepts(buf+23,readpts(buf+23)+tsoffs); + writepts(buf+offs+23,readpts(buf+offs+23)+tsoffs); // dts? - if( buf[21] & 64 ) { - writepts(buf+28,readpts(buf+28)+tsoffs); + if( buf[offs+21] & 64 ) { + writepts(buf+offs+28,readpts(buf+offs+28)+tsoffs); } } } } +int find_gop(unsigned char *buf) +{ + if (buf[14] == 0 && + buf[15] == 0 && + buf[16] == 1 && + buf[17] == 0xbb && + buf[38] == 0 && + buf[39] == 0 && + buf[40] == 1 && + buf[41] == 0xe0) + { + int i = 42; + while (i < 1024) + { + if (buf[i] == 0 && + buf[i+1] == 0 && + buf[i+2] == 1 && + buf[i+3] == 0xb8) + return 1; + i += 4; + } + } + return 0; +} + static int mpa_valid(unsigned char *b) { unsigned int v=(b[0]<<24)|(b[1]<<16)|(b[2]<<8)|b[3]; @@ -548,7 +574,7 @@ scanvideoptr(va,buf+i,thisvi,cursect,vsi); } if( !va->vd.vmpeg ) - vobgroup_set_video_attr(va,VIDEO_MPEG,"mpeg1"); + vobgroup_set_video_attr(va,VIDEO_MPEG,"mpeg2"); // if the mpeg version changed, then rerun scanvideoframe, because // scanvideoptr updates the aspect ratio in the sequence header if( mpf != va->vd.vmpeg ) { @@ -824,6 +850,7 @@ int FindVobus(char *fbase,struct vobgroup *va,int ismenu) { unsigned char *buf; + unsigned char *buf_copy = (unsigned char*) malloc(2048); int cursect=0,fsect=-1,vnum,outnum=-ismenu+1; int vobid=0; struct mp2info { @@ -837,6 +864,7 @@ int i,j; int hadfirstvobu=0; pts_t backoffs=0, lastscr=0; + int generate_vobu=0,copy_packet=0,tsoffs=0; struct vob *s=va->vobs[vnum]; int prevvidsect=-1; struct vscani vsi; @@ -869,8 +897,9 @@ } buf=writegrabbuf(); - i=fread(buf,1,2048,vf.h); - if( i!=2048 ) { + if (copy_packet == 1 ) { + memcpy( buf, buf_copy, 2048); + } else if( 2048 != (i=fread(buf,1,2048,vf.h)) ) { if( i==-1 ) { fprintf(stderr,"\nERR: Error while reading: %s\n",strerror(errno)); exit(1); @@ -986,6 +1015,38 @@ writeundo(); continue; } + + if( fsect == -1 ) { + char newname[200]; + fsect=0; + if( outnum >= 0) + sprintf(newname, "%s_%d.VOB", fbase,outnum); + else + strcpy(newname,fbase); + writeopen(newname); + } + + // we should get a VOBU before a video with GOP + if( (generate_vobu == 1 || hadfirstvobu == 0) && + copy_packet == 0 && find_gop(buf) ) + { + // create VOBU + generate_vobu = 1; + copy_packet = 1; + memcpy( buf_copy, buf, 2048); + + buf[41] = 0xbf; + buf[42] = 0x03; + buf[43] = 0xd4; + buf[44] = 0x81; + memset( buf+45, 0, 2048-45); + buf[1026] = 1; + buf[1027] = 0xbf; + buf[1028] = 0x03; + buf[1029] = 0xfa; + buf[1030] = 0x81; + } else if (copy_packet == 1) + copy_packet = 0; if( buf[0]==0 && buf[1]==0 && buf[2]==1 && buf[3]==0xba ) { pts_t newscr=readscr(buf+4); if( newscr < lastscr ) { @@ -997,16 +1058,15 @@ backoffs=newscr; } transpose_ts(buf,-backoffs); - if( fsect == -1 ) { - char newname[200]; - fsect=0; - if( fbase ) { - if( outnum>=0 ) - sprintf(newname,"%s_%d.VOB",fbase,outnum); - else - strcpy(newname,fbase); - writeopen(newname); - } + if( !hadfirstvobu && buf[0]==0 && buf[1]==0 && buf[2]==1 && buf[3]==0xba ) + { + int offs = (buf[17] == 0xbb) ? 24 : 0; + if( buf[offs+14] == 0 && + buf[offs+15] == 0 && + buf[offs+16] == 1 && + (buf[offs+17]==0xbd || (buf[offs+17]>=0xc0 && buf[offs+17]<=0xef)) && + (buf[offs+21] & 128)) + tsoffs = readpts(buf+offs+23); } if( buf[14] == 0 && buf[15] == 0 && @@ -1051,7 +1111,7 @@ printvobustatus(va,cursect); vsi.lastrefsect=0; vsi.firstgop=1; - } else { + } else if (generate_vobu == 0 || copy_packet == 1) { fprintf(stderr,"WARN: System header found, but PCI/DSI information is not where expected\n\t(make sure your system header is 18 bytes!)\n"); } } @@ -1369,6 +1429,7 @@ } } writeclose(); + free(buf_copy); printvobustatus(va,cursect); fprintf(stderr,"\n"); free(crs); ----- Original Message ----- From: "Martin Crossley" <martin@...> To: "Lawrence D'Oliveiro" <ldo@...>; <dvdauthor-developer@...> Sent: Sunday, November 08, 2009 7:22 PM Subject: Re: [Dvdauthor-developer] Attempt to solve "ERR: SCR movesbackwards,remultiplex input." > Hi Lawrence - > > I did quite a bit of work on this issue back in Jun 2008: see > http://www.mail-archive.com/ivtv-users@.../msg08234.html. > Unfortunately for some reason the source code of the patches themselves > seems to have vanished from that site, so I'll dig it out from archive and > sent it to you. > > I was taking a lot of input from a hardware MPEG encoder card, that > triggered this problem *a lot*. > > I wrote a bit about the causes of the issue here (see in particular "Part > 2: > how to patch dvdauthor"). > > My post here: > http://old.nabble.com/Submission-of-proposed-patch-to-dvdauthor-0.6.14-(affects-dvdvob.c)-p17695878.html > explains a bit more what the patches are doing > > The patches that I developed are probably a bit long in the tooth these > days > (they are against version 0.6.14) but I think that they might be helpful > for > you to have a look at. > > One warning though: while I found that the DVDs produced via this patch > work > absolutely fine with my players here (Sony, Denon), I have had reports > from > some friends that they don't work in certain older kit. > > > Cheers - and good luck > MartinC > > PS: for the benefit of the list, here is another document I wrote a while > ago, on the subject of how to use the Gimp to do anti-aliased menu text > with > dvdauthor. It's well worth the effort! > > http://ivtvdriver.org/index.php/Howto:anti-aliased_menus_and_buttons_with_dvdauthor > > > > > ----- Original Message ----- > From: "Lawrence D'Oliveiro" <ldo@...> > To: <dvdauthor-developer@...> > Sent: Sunday, November 08, 2009 4:52 AM > Subject: Re: [Dvdauthor-developer] Attempt to solve "ERR: SCR > movesbackwards, remultiplex input." > > >> Back on 2nd April 2004, philou <philippe.sarazin@fr...> wrote: >> >>> Like some of you, I had to face some trouble with a recurrent error >>> message in dvdauthor (v 0.6.10), "ERR: SCR moves backwards". >>> >>> This message occurs when the SCR (System Clock Reference) goes backward >>> in the data stream that I'm trying to author. After a lot of search on >>> the net, I think that there's a case where this is not an error as >>> dvdauthor reports it : the SCR can sometimes be reset to zero in the >>> middle of a VOB. >> >> I have decided to incorporate this patch. Well, I figure trying to >> recover and continue can’t be worse than dying with an error. :) >> >> ------------------------------------------------------------------------------ >> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 >> 30-Day >> trial. Simplify your report design, integration and deployment - and >> focus >> on >> what you do best, core application coding. Discover what's new with >> Crystal Reports now. http://p.sf.net/sfu/bobj-july >> _______________________________________________ >> Dvdauthor-developer mailing list >> Dvdauthor-developer@... >> https://lists.sourceforge.net/lists/listinfo/dvdauthor-developer >> > > > ------------------------------------------------------------------------------ > Let Crystal Reports handle the reporting - Free Crystal Reports 2008 > 30-Day > trial. Simplify your report design, integration and deployment - and focus > on > what you do best, core application coding. Discover what's new with > Crystal Reports now. http://p.sf.net/sfu/bobj-july > _______________________________________________ > Dvdauthor-developer mailing list > Dvdauthor-developer@... > https://lists.sourceforge.net/lists/listinfo/dvdauthor-developer > ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ Dvdauthor-developer mailing list Dvdauthor-developer@... https://lists.sourceforge.net/lists/listinfo/dvdauthor-developer |
| Free embeddable forum powered by Nabble | Forum Help |
