Luke Neumann
May 18th, 2009, 11:26 AM
All right...here is my current process.
1. Film on a Sony HDR FX1000 in 1080i.
2. Import via Firewire in .M2T using HDV Split.
3. Edit in ULead VideoStudio11 in an MPEG-2 Timeline (1920x1080, 35mb/s etc.)
4. Burn to DVD at the highest resolution (720x480, 9800 kb/s)
I know that I could have better software to edit in...but my question is...is the only place I am losing quality here in steps 2 and 4? If I am losing quality elsewhere or there is a simple task or step that I am missing to improve final DVD quality...please let me know. Thanks
Tripp Woelfel
May 18th, 2009, 06:20 PM
The potential for loss really only exists in steps 3 and 4. Step 2 doesn't create any loss at all. You're simply copying, bit for bit, the data from tape to your disk. The fact that it happens in real time is merely a coincidence.
There are a lot of ways to mess up an HDV to DVD workflow and it goes beyond "loss". Mess up the fields or deinterlace in a less than optimal manner and your footage will look degraded. Proper conversion to SD is critical.
Also, with HDV, there's not a lot of space to play color games with since it's just 4:2:0 (I think). Push the color around hard or manhandle the input white and black points and banding and other gremlins rear their ugly heads. You're not really losing a lot of image data, but it can noticeably degrade the image quality.
The other huge variable in the quality game is your transcoding software. For example, prevalent opinions on the Adobe Media Encoder range from so-so to complete rubbish. My view is that there are better tools out there. Many are quite dear, but others like TMPGEnc are not and deliver quite good quality.
So... you asked a simple question, but there are no real simple answers.
Luke Neumann
May 18th, 2009, 06:27 PM
I kind of figured there wouldn't be too much more to do with what I have now...upgrading software or camera being the next step.
Lutz Dieckmann
June 8th, 2009, 04:24 PM
Hi,
you create the dataloss in step 1. When shooting with I formats you create dataproblems inside the encoding. And you encode/decode everytime you render something on the timeline. When you transform to SD you have another encoding.
I shoot and edit with 25p and the results on the DVD with VBR (6 to 8 MBit) are stunning.
Best
Lutz Dieckmann
Bryan Sellars
June 8th, 2009, 04:48 PM
Hi Luke,
Your camera is upper field first and Ulead Video Studio 11 is the same as Corel Video Studio Pro X2 that I use and it defaults to lower field first when you make a DVD so always change the field order to upper if you are using interlace video, I don't think the encoder does a bad job but probably not as good as TMPGEnc that Tripp mentions.
Bryan