Dramatic loss of image quality with FCPX?

P

Picture

Member
I’m coming to the end of my 1 month free trial of FCPX and ready to buy,
but I just did a comparison of some short clips processed in various ways to check image quality.


AVCHD re-wrapped in ClipWrap to .MOV = No loss of image quality.
AVCHD processed in QuickTime to .MOV = No loss of image quality.
AVCHD processed in FCPX to .MOV = Definite loss of image quality.


The same goes for time lapse movies from photo’s - only worse!


I assumed I was doing something wrong, and tried dozens of input/output options in FCPX,
but in terms of finished image quality they are no better than iMovie.


Is this normal? or am I missing some hidden FCPX trick for retaining image quality?


Any help/advice/links?


Thanks.
 
Last edited:
FCP X does not degrade any image quality in the actual file, but can in the Viewer window. Check a few things. In FCP X preferences, in the Playback tab, what is "Playback Quality" set to? I'll assume it's also set to "Use original or optimized media" selected, also.

What Mac/graphics card are you using?

If in serious doubt, export (Cmd+E) a QT of the Timeline and open it in QT Player. It should look fine.
 
Hello Ben, thanks for the response. I think I recognise your logo from one of the many forums or blogs I visited looking for other people who had this same problem - no-one has, so I’m obviously doing something wrong?


The problem is not with the viewer or playback,
but the finished export - a ProRes 422 HQ .MOV.


Compared to the conversion in MPEG streamclip or ClipWrap, the FCPX looks soft and mushy, see attached jpeg.

110408 moonset test.jpg

cropped screen grabs from 1080P movs.
 
Is that a center crop? Regardless, all 3 look the same to me except the FCPX one has a different gamma.
 
Camera: Hacked GH1
Center(ish) crop see jpg.


I’ve just had a visit from a film maker friend (I’m coming from an art/science/photography perspective)
and he says there IS a slight difference in detail, but he would not have noticed it if I hadn’t pointed it out
in a side by side comparison.


But even if I accept that I’m being way too critical, and I should change
the title of the thread to “Slight loss of image quality in FCPX”
Slight loss is still some loss, and I’m looking for NO loss.


I also accept that most of the time in video a slight loss of detail can have large aesthetic benefits -
My old Leica lens would never win a technical battle with any modern lens,
but side by side in print, or on screen, I’m drawn into the Leica lens image every time
- despite or maybe because of it’s “flaws”.


But that’s not the point, I want lossless.


I think I’ll have to go into an Apple store and have a “one to one” with an expert,
but thanks for your input, I think I’ve narrowed it down to one of three possibilities:


1. (and most likely) I’m making a silly mistake somewhere along the line.
2. I should get a life and stop examining individual frames at 200%.
3. Editing programs are not truly lossless because they don’t need to be - which doesn’t sound right, that can’t be true.


Ps this is not a FCPX bash, it would be perfect for me if I could figure out how I’m loosing detail.


What does “Probably 2-pop, fcp.co, or FB.” mean?

110408 moonset test FF 02.jpg
 
Turn off the auto correction in FCP-X. And turn off the color correction that appears to be on the clip.
 
arniepix - I tried turning off the auto correction, and tried several other combinations for input and output,
all with the same result.


I was just about to do another test but my trial period is up. Damn!


Has anyone tried importing a 1920 x 1080 photo, then exporting it as a short .mov,
and then compared a frame of the mov with the original photo?
Am I right in assuming they should be identical - no loss of detail in the mov frame compared to the original photo?


I’ll try it as soon as I get the time to go to an Apple store, but if anyone has already done it I’d be interested in the results.
 
I'm not seeing any loss of quality in anything, but then, I'm not using my computer screen, which is the worst way to judge image quality. I'm running it all out to an FSI broadcast monitor. No quality loss at all.
 
Picture said:
Has anyone tried importing a 1920 x 1080 photo, then exporting it as a short .mov,
and then compared a frame of the mov with the original photo?
Am I right in assuming they should be identical - no loss of detail in the mov frame compared to the original photo?
No, you are not correct in assuming this. Importing, transcoding, exporting, etc. are all types of processing on the image and will have some impact on the quality when compared to an original un-retouched photo. The format of the still capture and its' compression technique will also play a role.
Click to expand...
Click to expand...
 
“… some impact on the quality…”
Yes, but it seemed to be much more noticeable in FCPX.
In my side by side comparison with After Effects, MpegStreamclip,
and ClipWrap, the FCP mov had the worst image quality.


It was almost certainly user error, but I spent most of the trial period
trying all combinations and suggestions, and still ended up with mush, so I gave up.
I never got to stage two with FCPX.


Stage 1 - What the … !!!
Stage 2 - Oh, I see … ?
Stage 3 - Cool :-)


Thanks all for the input, it was appreciated even though it came to nothing.
Shame Apple don’t do “do-overs” for dummies, I wouldn’t mind having another try.
 
You must log in or register to reply here.
Back
Top