Matt Sandström
Forum Replies Created
-
thanks, but i’m talking about something that has been working fine for 10 years since version 1.0 and just recently (pretty much after i installed the proapp update two weeks ago?) stopped working. i’m what you would call a final cut expert and that’s why i feel a bit silly asking this. 🙂
/matt
-
if it’s anamorphic it will display correctly on both widescreen and normal tv sets when playing from dvd. letterboxed video is double letterboxed (sides too), or worse squeezed, on widescreen displays. (yes, i know you can change the tv settings to zoom but few do/bother)
make sure you use high quality scaling to create the anamorphic since new lines will be created. compressor with frame controls at high quality should do a good job. in either case make sure it’s progressive scaling and not interlaced.
/matt
-
[CaptainMench] “Remember, we’re talking about filming in anamorphic:
we are? i thought this was an fcp forum. 😉 ok, fair enough, it doesn’t matter. the facts are the same.1) I’m saying only 75% of the CCD is being used to fill 100% of the image. How is that not an image loss?
that’s a design flaw of your camera. i think you should call your camera manufacturer to complain, not use it to draw conclusions. 🙂 the rest of the discussion becomes academic since this seems to be where the misunderstanding lies. there are plenty of ways to gather footage in native 16:9 including many dv cameras, all higher end sd cameras, all hd cameras, film, cgi, and so on. sorry if you’re camera isn’t one of them.3) Letterbox is 720×480 and when editing stays 720×480. Anamorphic is 720×480 but when editing it goes to 853ishx480.
no, anamorphic is 720×480. whatever fcp might do to display the footage on your computer monitor obviously doesn’t change the quality.3) Any way you look at it, some pixel is being distorted. Distortion has to lead to quality loss. Just to what degree?”
i’ve no idea what way you’re looking and i don’t understand what you mean by distortion. they were always rectangular, and rectangular they will remain. converting them to square pixels is what would distort them. keeping them anamorphic doesn’t./matt
-
16:9 is not stretched, it’s squeezed. It’s not a letterboxed image stretched to full height, it’s a wide image squeezed to a narrower width. In the case of converting 4:3 material you have to stretch it though, hence the 25% loss of “quality” i mentioned. /matt
-
if you’re converting to uncompressed it doesn’t matter when you do the conversion, but if you’re editing in dvcpro and adding any kind of effects in final cut you’re losing a generation if you convert before. i use compressor for conversion since i find it does a better job in many ways than qt pro.
/matt
-
[CaptainMench] “Might not be too much a quality loss… remember footage shot in anamorphic is a quality loss in itself”
the “quality loss” is exactly 25%, no more, not less. how much this shows depends on your scaler. i like to use compressor with frame controls for any conversions which usually creates better results than fcp. after effects also does a good job.
in what way is anamorphic quality loss in itself? it’s 720×480/576 pixels just like 4:3. that’s only a quality loss compared to the native width of 864/1024.
/matt
-
thanks. good to know there’s nothing “wrong” with my setup, and doing it the way you suggest isn’t too much work after all.
/matt
-
yes, i use this filter all the time and never got rejected. i always check the levels on the scopes though and use the 3-way cc to get close and only let the filter take care of any peaks. and i’m in the pal world and i hear it’s less strict…
/matt
-
yes, that’s not it either. it seem like the problem is that i’ve set the speed of the master clip. when i change the speed on the timeline i don’t get this problem, it seems. however making the clip on the timeline independent doesn’t help either. once it’s “contaminated” by the master clip’s setting nothing changes this behaviour.
can you replicate this? change the speed of a master clip to say 97%, blending off, add it to a timeline, remove the speed attribute and roll the out point a few frames. what’s the speed of the clip now? for me it often ends up at 100.14% or so, suggesting that it’s a rounding error.
/matt
-