I´m using the software Magix Movie Edit Pro 2014 Plus for editing both old VHS-tapes (which I have converted to avi-files) and new mts-files.
To improve the quatliy (sharpness, color etc) I sometimes use Virtual Dub and VD-filters and sometimes the tools which Magix offers.
When I recently was about to edit an old "VHS-avifile" using Magix, I suddenly came up with the (wild) idea to drag the imported avifile into several timelinetracks so that the files are "layers above each other". Probably I am wrong but I think that the sharpen is better when I play the video.
Is this an illusion due to my hopes?
Try StreamFab Downloader and download from Netflix, Amazon, Youtube! Or Try DVDFab and copy Blu-rays! or rip iTunes movies!
+ Reply to Thread
Results 1 to 28 of 28
Thread
-
-
If Magix works like any other mult-timeline nle then ,yes, your eyes deceive you. Nle's can read from multiple timelines but if the content overlaps then only the highest is read.
-
Thanks. I see, but it would be nice if there were a software which could "merge" the clips in order to improve the quality as if you paint a wall serveral times with the same color!
-
Merging exact copies of the same video will give you an exact copy of the original video -- except for possible rounding errors.
You can get better results by blending multiple caps together. Or using the median of several caps. -
If you can control transparency (as on most NLEs) you can "merge" the clips. But if they are the exact same file there will be no difference whatsoever. If you use transparency and add different effects and transfer modes to each layer you may be able to emphasize or deemphasize certain aspects of the image -- but I doubt magix is that sophisticated.
-
Of course you can blend video tracks to enhance the output. This is what they do in HDR photography. The idea is to bring out shadows in one track, and bring out highlights in the other, then composite together.
Try different composite levels and blend modes to see what works best. It works pretty well. Even if you simulate it with copies of the same track.Last edited by budwzr; 20th Oct 2014 at 12:42.
-
Yes, but in HDR imaging, each track/layer is composed of a different version of the image.
Example:
1. Layer 1 has good highlights but no shadow or midtone detail.
2. Layer 2 has good midtones, but no shadow or highlight detail.
3. Layer 3 has good shadows, but no midtone or highlight detail.
Combine them together and you get an extended capacity of detail at the cost of a larger filesize with higher display capability requirements - usually so high that a standard display cannot accurately depict the output. So then you do something like tonemapping to squeeze the dynamic range into a smaller spectrum that the display IS capable of. Or wait and get an $$$ HDR display.
Again, your simulation improvements can really only be done when there is something DIFFERENT between the tracks/layers. Usually, this entails some kind of masking. And understand that a simulation is NEVER as good as the real thing.
WITHOUT utilizing differences between layers, the result of various compositing/blending should just be a less-than-100% application of whatever process was applied to a single layer.
@Mållgan, Your best bet is to follow jagabo's suggestion.
Scott -
Yes, but we're not trying to do HDR, just steal the rough principle. We just want to grab the best parts of two ends of the dynamic range, and meld them. The midtones would remain the same essentially.
A little tweak of the levels, just a smidge mind you, could improve things a lot. I don't think this conflicts with what jagabo is saying. He's saying to use the capture machine to obtain another copy with different levels, and I'm saying you can probably do it in Post if the source isn't too bad.
The OP already said it looked better by simple compositing. The OP is thinking outside the box, and stumbled onto compositing secrets of the pros.Last edited by budwzr; 20th Oct 2014 at 15:55.
-
I understand (at least I think I understand!)
.
But what about the DB83-reply? (... can read from multiple timelines but if the content overlaps then only the highest is read). Does the content not overlap if I follow jagabo's suggestion? -
[Deleted] See my next post.
Last edited by budwzr; 20th Oct 2014 at 21:50.
-
Db83 is correct when all/top layer is 100% opaque and blend mode is normal, the other suggestions are possible when those 2 requisites are not true.
@budwzr, we aren't necessarily disagreeing, but I am not sure you caught the nuance of the need for segmental differences between the layers in order to provide a level of influence beyond what could already be done by incremental processing of the same layer.
Scott -
Last edited by budwzr; 20th Oct 2014 at 16:45.
-
Look at the nose to see the differences. The composite is just a tad more richer. Not a lot because the source is good from the beginning.
-
Last edited by budwzr; 20th Oct 2014 at 21:52.
-
-
Now I have tried with four tracks and different efffectsettings in each track. The result is a file which has better quality than the inputfil (an old VHS-avifile) as far as I can observe but not "super" quality.
But what is really happening processing the same input file in, for example, four tracks like this? As far as I know a (avi)file consists of frames and each frame consists of pixels (I give them here pixel number 1,2,3 and so on). Then, when I drag the same input file to four differents tracks and change some of the effectsettings in each track and finally let the program produce an outputfil (for example a mp4-file), my question is if the pixel number 1 of the outputfile, is a "summary" of pixel 1 from each of the four input files? That is, is the pixel 1 of the outputfile a new "pixellayer" including the "best" from pixel 1 of the four inputtrackpixels? Or what is happening? -
....my question is if the pixel number 1 of the outputfile´s frame 1, is a "summary" of pixel 1 from each of the four input files (pixel 1)? That is, is the pixel 1 of the outputfile´s frame 1 a new "pixellayer" including the "best" from pixel 1 of the four inputtrackpixels´s frame 1? Or what is happening?
-
The output of your editor depends on what you told it to do with the four "stacked" tracks. It's probably just using one of them for the output. You can easily test this by using four different videos. In any case, the editor has no idea what the "best pixel" is and could not select on that basis.
-
Strange if the output consists of just one of the tracks. What´s the reason then to use several tracks with different settings in these several tracks?
-
You TELL the editor what to do with those tracks. You don't just add a bunch of tracks and hope it does something. Do you put a whole bunch of ingredients in a bowl and hope a cake comes out?
-
??" hope the cake come out". No, not at all. "Hope it does somenting"? I just wondered what the output file consists of in relation to the four "input" tracks.
"Now I have tried with four tracks and different efffectsettings in each track. ......Or what is happening?" -
A typical use of using several tracks is to put separate videos on to each - or even parts of one.
You can then slightly overlap them and place a transition filter between them.
Also, you may just want to apply certain filters to part of a video so you just split it up.
It ultimately comes down to what your NLE can do. No idea of the capabilities of Magix. -
What happens is whatever you tell the editor to do with those four tracks. The default is probably to use only the topmost track and ignore the others.
Budwzr is doing you a huge disservice by discussing advanced HDR techniques when you don't have any idea what you're doing. -
Jesus.
I know exactly what I'm doing but not what is happening efter I have done what I'm doing. And no one else seems to know what's happening, that is, the relation between the four inputtracks and the outputfile. -
-
You have to look in the Magix Help File and see if track compositing/blending is supported. Whatever I suggested is based on a concept. If you, or your NLE, is/are incapable, then someone else might pick up on the concept, and benefit.
The relation between the tracks is determined by the opacity level, order, and blend mode, of each track.
I thought the OP's initial "epiphany" of doubling up the tracks to increase levels was a sign of a budding, although neophyte, tendency towards more advanced ideas and techniques. But unfortunately that's not panning out.Last edited by budwzr; 21st Oct 2014 at 11:36.