DVB interlace issue in GK

[Funkdafied]

I made a recording of a 576i broadcast (Australian TV), demuxed it and cut out the ads.. DGIndex reports it as PAL-interlaced, but I can't find any horizontal lines. When I open the project in GK, set the bits/(pixel*frame) value to say 0.157 and do a compressability check, it comes up as 91.3%! This can't be right.. so what's going on? (I'm going for 1/2 CD XviD, btw).. Could it have something to do with the source being interlaced or maybe its 50fps?
How do i fix this problem?

[manono]

Hi-

DGIndex reports it as PAL-interlaced...

It just reports how it was encoded, how it is flagged. That is not the same thing as the source being interlaced. If you don't see any interlacing/combing, then it's not interlaced.

When I open the project in GK, set the bits/(pixel*frame) value to say 0.157 and do a compressability check, it comes up as 91.3%! This can't be right..

Why not?

[Funkdafied]

DGIndex reports it as PAL-interlaced...

It just reports how it was encoded, how it is flagged. That is not the same thing as the source being interlaced. If you don't see any interlacing/combing, then it's not interlaced.Yup, I understand that part...


When I open the project in GK, set the bits/(pixel*frame) value to say 0.157 and do a compressability check, it comes up as 91.3%! This can't be right..

Why not?As for why not.. well.. because the resolution was practically the same as the source... and it just seemed way too high. If I lowered the res at all, it would go over 100% of the compressability check, and if I increased the res, it would be larger than the source.. also, the values I was getting didnt seem to fit with the recommendations in the doom9 guide.

[manono]

Hi-

Well, different sources compress differently. I'll admit it's unusual for a high res AVI to compress that well given that b/p*f, but it's certainly not unheard of. Assuming you did everything correctly, maybe it just compresses well. If you want to get that percentage down, you can do a number of things, such as lower the intended file size, upgrade the audio (AC3 instead of MP3), leave off the B-Frames, don't filter, use a sharper resizer (LanczosResize instead of BicubicResize, for example), or if XviD, use a high bitrate matrix. There are all kinds of things you can do. Doom9's recommendations are just guidelines. Once you run a compress test, those guidelines go out the window. I don't see anything wrong with making the resolution larger than the source, if you decide to go that route.

[jggimi]

What is the length of your content? You mention an output size of 350MB (1/2 CD), but if its only a few minutes long, that might account for your confusion over compress test results. But, since we haven't seen any logs, this is just conjecture.

For more information from us, we need more information from you. :logfile:

[Funkdafied]

It was a good 40 minutes or so..

I'm going to do a bit of testing then I'll get back to you :)