Hi Guys, another newbie sort of question.

My Panasonic SDR-H280 saves my clips as MPEG-2 to it's hard drive. I note that using either Panasonic's VideoCam Suite, or SU12.1 "Capture" to transfer files to the hard drive, and then "Make Movie" that there is a noticeable loss in video quality than if I play video directly from the camera on to our television using a "video In" connection. I believe that MPEG-2 is a 'lossy' codec. I would like to understand the best method of transfering files to minimise losses. Basic stuff I guess, but disappointing to see the loss in 'sharpness' when you are trying trying to preserve whatever you do have. I have 'searched' the forum without an answer - perhaps I'm not phrasing the search criteria correctly.

 Mark

I note that using either Panasonic's VideoCam Suite, or SU12.1 "Capture" to transfer files to the hard drive, and then "Make Movie" that there is a noticeable loss in video quality than if I play video directly from the camera on to our television using a "video In" connection

That shouldn't happen. How are you judging the quality? You need to compare like for like - and view both the camera output and the DVD on the same TV.

Pinnacle "Capture" import of mpeg-2 files is a straight digital file transfer. No quality should be affected. If you "Make Move" to a DVD disc using "Best Quality" and don't add any effects, the render should be a straight stream copy of the orginal file. OK, it is converted to a VOV file, but this doesn't involve any recoding.

If you are comparing the output of your camera on a TV, to the quality of a make movie iperation on a computer screen, then you will see a percieved change of quality as your computer screen as a higher resolution, and therefore you can see more faults in the original picture.

Pinnacle "Capture" import of mpeg-2 files is a straight digital file transfer.

Agreed.

No quality should be affected. If you "Make Move" to a DVD disc using "Best Quality" and don't add any effects, the render should be a straight stream copy of the orginal file. OK, it is converted to a VOV file, but this doesn't involve any recoding.

Not always the case.  While I agree that the capture isn't dropping in quality, there is no guarantee that Studio isn't forced to re-render when making the output, and if a re-render occurs, rendering MPEG2 > MPEG2 is almost ALWAYS going to look worse.  Reason:  All of the little subtle "tricks" that MPEG2 uses to cram so much information in such a small space comes back to bite you (same-same with AVCHD rerendering, too).  Little subtle things, like the "checkerboard pattern" of MPEG2 macroblocks, can be "noticed" by the encoder and preserved during the re-render while other detail is dropped, enhancing the (for want of a better word) "macroblockiness" of the output.

One easy way to tell if Smartrendering (a.k.a "Minimal MPEG Rerendering" is happening:  If the render is significantly faster than real-time, then the "copying the MPEG2 stream directly to the VOB file" action is occurring.  If the render is slightly faster to much slower than real-time, then it's fully rerendering, and the quality *will* likely be visibly affected.

Not always the case.  While I agree that the capture isn't dropping in quality, there is no guarantee that Studio isn't forced to re-render when making the output, and if a re-render occurs, rendering MPEG2 > MPEG2 is almost ALWAYS going to look worse.

Agreed, but how much worse? If the OP is comparing what he/she sees from their camera on their TV with what they are seeing on full screen preview of the same mpeg-2 file on a computer monitor, then as we both know, it's apples and oranges. It's going to look worse - possibly quite a lot worse.

Let's assume that Direct Stream Copy/"mimimal"/Smart render isn't happening. That's one generation of mpeg-2 recode. Let's also assume that there aren't any Progressive vs. Interlace gotchas or the OP isn't using "most video on disc" - so you are getting a second generation 6000>8500 Mbps recoded mpeg-2 file. How much of a drop in quality is that? Enough for a casual observer to say that the quality is considerably reduced? Or just enough so a critical observer could notice a increase in movement artifacts?

What I failed to mention, is that the original mpeg-2 file might be using a different mpeg-2 codec, which would cause a re-code anyway. Even then I don't see why a huge loss in quality is inevitable - although the audio is bound to be 1/3 of a frame OOS

I think we need to return to the OP and ask how they are measuring quality loss and how severe they judge it to be. It's all a matter of degree.

jjn:

I think we need to return to the OP and ask how they are measuring quality loss and how severe they judge it to be. It's all a matter of degree.

This seems to be the part where it will all unfold.

Thanks guys, while I appreciate your response I'm not sure that I'm any the wiser

I am comparing the camera and DVD output on the same TV (analogue). My "Make Movie" settings are "Best Quality" the rest are default settings for DVD. I agree whole heartedly that quality loss is subjective. At a guess I would think that my camera is meant for 'the casual user' so output isn't going to be brilliant from the outset. If I could get the video quality in my movie equal to that of direct projection from the camera I would be a happy camper. On a scale of 1 - 10. 1 being abysmal quality and 10 being a dead ringer of a copy, I would deem my movie output to be in the order of 7 - 8. I must add that this perceived drop in quality can vary somewhat from one clip to another, evenly lit scenes faring better than brilliantly sunlit ones.

The movie in this case is a series of video scenes that have fades and dissolves, some zoom and some speed adjustments, some sections of clips have been cut and deleted. Pretty much normal stuff I would have thought. This is my first foray into video, previously and through several incarnations of Studio I have used stills with very pleasing results. I have quite often read items in the forum about people's woes with Studio and thought, smugly, thank heavens I don't have the issues that some of those poor buggers do!

I guess I was asking for a 'rule of thumb' to transfer videos from my camera to DVD output without loss. I think you are telling me that in either of the methods I have tried that there 'shouldn't' be losses.

 Mark

I guess I was asking for a 'rule of thumb' to transfer videos from my camera to DVD output without loss. I think you are telling me that in either of the methods I have tried that there 'shouldn't' be losses.

There will be, but nothing that would be easy to notice.

OK, thank you Marc!

Mark

Let us know how things go.

Ozyboy:

Thanks guys, while I appreciate your response I'm not sure that I'm any the wiser

I am comparing the camera and DVD output on the same TV (analogue).

At the risk of decending into pedantry, are the two souces connected to the TV in the same manner? A S-video hook-up should give you a better picture than a composite one, for example.

The movie in this case is a series of video scenes that have fades and dissolves, some zoom and some speed adjustments

All of these are going to trigger a re-code.

I guess I was asking for a 'rule of thumb' to transfer videos from my camera to DVD output without loss. I think you are telling me that in either of the methods I have tried that there 'shouldn't' be losses.

unless you are getting  "Smart render/direct stream copy" - render time considerly fast than real time as Bittmann says - then there will be losses. However, a score of 8 out of 10, even by the most subjective standards, seems a little on the low side to me for these losses.

Perhaps something else is going on - the mpeg-2 codec from your camera isn't the same as Studio's and additional losses occur?

Try to achieve a smart render of some video, and then compare the DVD with the camera. They really should look the same.

Thanks JJN, for your response and patience, feel free to stoop to pedantry, I thrive on descriptions using single syllables and lots of pictures. I'm not only a newbie to the forum but also to video editing per se. I obviously have a lot to learn, guys like yourself are my only mentors.

Unsure about what you term "smart render". If I were to take a video clip, put it on the Time Line, do nothing else with it, then "Make Movie". If I've got it right, you're saying that it should look pretty much the same as viewing the raw clip from the camera. Yes, I'm using composite connections in both viewing scenarios.

The owner's manual for the camera only states that compression is MPEG-2, no further detail. If I were to contact Panasonic, what are the key words I should use asking about the codec?

Mark