Hi - please help - I have been at this for two years now and it is driving me nuts!!!
My personal preference of Video Editing and DVD Authoring software is Vegas Video 4.0 and Vegas DVD Architecht 1.0.
This is important why? Read on!
I use a Sony DV Camcorder and capture DV (using Vegas Capture) to an .AVI file at 720 x 576 x 24 , 25 fps, with 16 bit audio (PAL DVD).
The video stream in the source material once captured is crisp and clear when played back using (for example) PowerDVD.
However - when the file is rendered using Vegas 4.0 the images and video stream in the MPEG2 file are no longer sharp or crisp i.e it is almost as though a 'soften effect' has been applied to the source material.
In frustration I have tried TMPGenc as well as ULead Video Studio and Adobe Premiere but all with much the same results.
After much experimentation it would appear that this problem has something to do with interlacing / deinterlacing. Is this possible? If so why? If so - what is the solution?
I have played back a DVD that I created on a set top DVD player and TV and although the images / video stream appear to be slightly clearer or crisper they are still not to my satisfaction.
What can I do now?
The only way that I can get images that I feel are close to the original is by applying a 'sharpen' effect to the source material prior to rendering. This (I am told) actually results in a loss of information and that is why the edges appear to be sharper. Is this correct?
Another question - I have an AOpen DVDRW which uses DVD+R/+RW. What is the difference between DVD+R/+RW and DVD-R/-RW and why are there two types. I believe that DVD-R is the DVD Forum Standard and AOpen is a member of the forum so why do they only produce a DVD+R/+RW drive. From what I gather Pioneer seems to be the only DVD-R/-RW drive available.
+ Reply to Thread
Results 1 to 6 of 6
-
-
you should not be deinterlacing DV source material , it may look better on your PC , but (much) worse on your tv.
you may be looking at the interlacing on a pc which has a progressive screen , you should use the monitor out function of vegas to preview on an external PAL monitor ..
you are compressing an image when encoding to mpeg2 , some loss is going to happen , though without knowing your settings - its hard to make much judgement .. mpeg2 can be made that looks just about the same as the source (specially a DV source) . -
Thank you very much for the prompt response.
After posting my original message I searched these forums and found reference to updated MainConcept files.
I downloaded the latest version of DVD Architect and this was supposed to update the MainConcept files and version but even after installing this update the version number is still v1.0 build 798 so I have sent and email to support@mainconcept.com to ask them what the latest version should be. I will post any replies from them.
You say that I should not be deinterlacing DV. I agree with you but the strange thing is that before opening my DV AVI source file I obtain my Vegas Project Properties straight from the file i.e. I match the project properties of the source file to my project before opening the file and when I do this the deinterlace settings in the Vegas Project Properties dialog changes to 'Lower Field First' which tells me that either the source is in fact interlaced (I don't see how or why) or Vegas is not correctly picking up the fact that it is DV and is not interlaced. Any ideas. The only other thing that I can think of is that Vegas Video Capture automatically deinterlaces during capture (the reason I say this is that you have the option to deinterlace still images before saving them to disk although there is no option to deinterlace (or not) captured video).
Sorry - just to be clear - is DV supposed to be interlaced or non-interlaced? -
I'm not an expert, but have baffled myself with de-interlacing and interlacing to the extent that I believe I have understood the topic. So, here is my feedback:
All video material you see (camcorder video, DVD content, TV broadcast, etc), is all interlaced. This is how things started and now, although technology is available to have progressive scanning compatibility issues make it slow to implement.
It is true that a non-interlaced (i.e. progressive) video looks better on the PC. I have also found that most MPEG-2 encoders encode better non-interlaced video than they do with interlaced (for any given bitrate, non-interlaced material produces better image quality - less artifacts).
However, de-interlacing the video is difficult and as some postings and web pages "prove" is almost impossible without some loss in clarity. You will get either jagged edges in panning motion or blurry edges, depending on the algorithm you choose to use.
If you want to make a DVD out of a DV capture, you must encode as interlaced and make sure you don't encode with the wrong field order. The original (DV) video stream will have each frame in two "blocks", the first block may have odd scan lines and the second even scan lines. Or it may be the other way round. This is what is called Top field first or Bottom Field first.
Why is this important? Because if the captured material is in top-field-first sequence (TFF), encoding with a bottom field first setting will cause blurry edges as well. I recommend that you make a short clip with your camera where you do a horizontal panning motion, capture it and encode it with the two alternative interlace sequence options. View it on the TV set and you will see which setting is the correct one.The more I learn, the more I come to realize how little it is I know. -
Originally Posted by dpaterso
install vegas 4.0c and latest DVDA .. that is latest version of encoder ..
mini DV is interlaced bottom field first always
you dont have to match anything -- just use the NTSC (or PAL) DV template in vegas -- end of story .. -
Hi,
Thanks for all the help.
After trying all of the above and still not getting results I decided to go back to basics.
I have had a good, hard, objective look at my source footage and have found that the source is not quite as crisp and clear and sharp as I thought it was and I think what is happening is that the rendering process is just accentuating the 'soft' images and making them much worse.
Which opens up another can of worms!
I just purchased this Sony DV Camcorder (TRV245E) (about a week ago) hoping that this would solve the problem (I was using a much cheaper Mitsubishi Analog Camera before and was getting the same 'soft' images). I thought it was my capture card and spent much money and hours on perfecting the analog capture i.e. no dropped frames, compression, etc. etc. and it is disappointing to find out that after all of this the results that I get with either camera are the same!!!
So - I have decided that the only thing that I can do is to use the DV Camera but apply a 'sharpen' effect to the source material prior to rendering. My question - is this an acceptable way of doing things?
The only thing that this new camera has bought me is the fact that I can sharpen the images without noticeable artefacts appearing (the Mitsubishi is old and seems to have a lot of 'noise' or 'speckles' even when no tape was being played / recorded and the 'sharpen' effect only accentuated these 'speckles'. The Sony (because it is new I would imagine) at least does not exhibit these 'speckles' so the sharpen effect does not accentuate these imperfections).
I have (hopefully) attached two .JPG pictures to this message - one sharpened and one original - any comments would be appreciated.
Where to now? Has anyone got any recommendations on cameras? Is it necessary to purchase a 3 CCD DV Camera in order to get any sort of quality without having to doctor the source?
I have noticed that sharpening the image also appears to affect the colours in the source i.e. the colors of the sharpened picture are closer to what they should be. Could this be a color bleed problem?
What are your thoughts on using the camera features like Image Stabliser (or Steadyshot as Sony calls it) and Automatic Exposure and Auto Focus?
I know that there are a lot of issues and questions in this post - sorry - but - like I said - I have been at this for about two years now (and first started testing stuff like Adobe Premiere Version 1 with FAST Capture Boards about 10 years ago on a 486) and would like to produce something good after all of this!!!
PS - I could only upload the original image - the sharpened one is (slightly) too large - but - if you look at the source image you will see that 'soften effect' that I am talking about. The 'sharpen' effect seems to correct the colors somewhat (the original is slightly blueish) and also makes the letters of the registration number more defined as well as the horizontal lines of the grille of the vehicle.
Similar Threads
-
proprietary codec used by Sharp for MPEG2
By ghostma in forum Software PlayingReplies: 17Last Post: 16th Nov 2011, 17:11 -
Merging MPEG2 stream with its AC3 audio... or converting MKV to MPEG2
By cwt in forum Video ConversionReplies: 2Last Post: 5th May 2010, 14:18 -
Audio Stream Swap in WMV video files
By a1s2d3f4 in forum Newbie / General discussionsReplies: 17Last Post: 15th Sep 2008, 10:57 -
need MPEG2 Transport stream and Program stream profiler
By afarun in forum ProgrammingReplies: 3Last Post: 23rd Aug 2007, 01:06 -
How is done the sync between audio and video in a MPEG2-VOB file stream
By rmanal in forum EditingReplies: 0Last Post: 22nd Aug 2007, 09:33