[Craig Ricker]

Hmmm the proxy quality setting in mpeg stream clip does nothing. They always come out at 19.7mbps. So there is no way to get anything even lower? I thought 3mbps would be awesome, as then nothing under the sun would bog the comp down and I can just simply relink at the end?

So nothing that keeps 1080 size but can go even lower in bit rate?

(the relinking thing with go pro is not really that bad – just make two copies, 1 at full res, 1 at “crap”, once you edit with the crap from its location, it shouldnt ever try link to the full cause it can find the crap in the exact location it was in last)

Also is it fine to just relink to the orginal h.264’s or is it better to relink to a pro res lt? Just thinking whats faster
1. link to original h.264’s with a slow final online render out to pro res lt ready for compression for web
or
2. convert h.264’s to pro res lt upfront and then just link to them, with a therefore quicker render to pro res lt on export

[Craig Ricker]

Yeah but u just can’t see the quality so it counts for nothing. I’m wondering if shooting 1080 and sticking in 720 sequence may be the goer. Is there a quality lose in fcp for the fact it will use the basic motion scale setting to resize? Like a loss big enough to justify just shooting straight 720.

[Craig Ricker]

Quote
“First because, at the same data-rate, a picture at 720p25 has more info/pixel than a 1080p25.
More quality.
Then because you will be processing tons of info that will be trashed when you downscale.
And third because downscaling is not just discharging or averaging pixels. The most you downscale the image, the most complicated will get to calculate the new values and depending of the filters used the results will vary very much.
FCS doesn’t shine at all on this.
So shoot 1080 if you need to zoom all that picture, if not, shoot 720. You still have room to re-frame and zoom and pan a little.
rafael”

1. From my tests shooting in 720p it is indeed the same bit rate of 35Mbps but I see no noticeable difference between it and the 720p. Is there a hidden benefit to the higher bit rate per pixel?

2. So would we not be better of shooting in the 1080p in a 720p sequence so we can use pans and zooms. Does FCP cause a high lose of quality when it automatically uses the scale function to reduce the 1080p to fit into a 720p sequence.

3. If using the scale function in fcp to do pans and zooms do we create poor quality as the scale travels through uneven divisions of the original i.e as you go from 80 to 100% its gonna pass through 87.73% which is not an even factor.

4. Our final product is gonna be SD (DV PAL) on tape for broadcast and DVD authoring, but a 360p and a 720p for online viewing. So just trying to confirm your thoughts on the quality difference between down rezzing from 1080 to 720 in fcp and then to sd pal via compressor.

[Craig Ricker]

Cheers, I choose to go with compressor. I do seem to get some distortion on some of my graphics, 98% looks ok though. ALso get sime slight jagged edges on video foogtage when viewed on a CRT? I have compressor settings set to the highest quality in all options (frame controls included). Im also converting it to apple pro res and then putting it into a dv pal timeline. Found the quality to be much better this way.

I’ve heard talk over 720p25 being better to shoot in if converting to SD at the end, rather than shooting in 1080p? Whats the reasons for this?

[Craig Ricker]

Thanks for the reply.

Been trying to work out what your code is doing. I’m kinda noob to code.
Do you think you could explain what each line is doing, just so I can learn how it works.

What does the % sign mean?

[Craig Ricker]

So just to help my understanding.

What is the difference between 25psf and 50i besides the 50 instances in time and the 25 instances in time respectively.

They both have 50 fields, so if I tell FCP that my 25psf footage has a field order of “none”, what is FCP doing with the fields? Playing both at once and holding them there for a whole frame?

What happens when I put my sequence back to DVCAM with field order set to “none”. Does final cut do the same thing and duplicate the fields on both upper and lower, so that it is back to 50i on the tape?

[Craig Ricker]

Brilliant! Learning heaps, thanks!

[Craig Ricker]

Thanks for the replies guys.

Ok so I want to make it a little more complicated and I wonder if this is possible.
What I would like is this.

Object A follows object B using
timeToStop = 6.5;
if (time < timeToStop)
{thisComp.layer(“Shape Layer 1”).transform.position.valueAtTime(time-1)}
else{
}
At 6.5 Object A continues on, and I would like Object B to follow but I would like it to gradually (say over the next 1 second) add 5 to the x value of the position of Object B.

So at 6.5 seconds the following would be true
Object A position [0,0]
Object B position [0,0]

And at the end of that next second (7.5) the following would be true
Object A position [5,10]
Object B position [10,10]

How can I go about gradually incrementing the X position of object B whilst still having it follow Object A?