Do you need Kona2 for DVCPROHD or IO is good?

[Andy Edwards]

If you need to do a bunch of conversions, then the Kona2 is the way to go. If you have all your footage on Panasonic varicam tapes, then just rent the AJHD-1200a deck and import all your footage via firewire directly into FCP without the AJA IO or Kona2. I did this last year just after NAB as the Kona2 was not shipping. I had all my HDCAM tapes dubbed to 1080i60 on Panasonic tapes and rented the deck. I had the entire 30 minute documentary edited on a single firewire drive in DVCPRO HD.

Andy

[Walter Biscardi]

Varicam footage IS DVCPro HD so whether you come in Firewire or SDI, it’s the same quality.

HDCAM is more of an uncompressed format, though not totallly, it’s something like 3:1 compressed but generally you capture that at native 1080i uncompressed.

The Kona 2 features both upconversion (SD to HD) and downconversion (HD to SD) on the card which is fantastic when you’re working with the SD formats as you note. Really what I would do in your situation since it’s all SD material, is capture it at 8 or 10 bit uncompressed SD, edit the piece, then use the Kona 2 to upconvert the material to HD.

Or, you can simply capture in DVCPro 50 which is a much better format than DV or DVCAM for graphics and just stay in the SD world which I’m pretty sure you can do with the Io.

Walter Biscardi, Jr.
Creative Genius, Biscardi Creative Media
https://www.biscardicreative.com

Now in Production, “The Rough Cut,” https://www.theroughcutmovie.com

“I reject your reality and substitute my own!” – Adam Savage, Mythbusters

[Graeme Nattress]

Comparing HDCAM and DVCproHD to 4:4:4 uncompressed HD (all 1080) I came up with the following figures:

HDCAM is 135mbps, which is a 11:1 compression over fully uncompressed.
DVCproHD is 100mbps, which is a 14:1 compression over fully uncompressed.

However,

Of the 6220800 pixels across the 3 components of fully uncompressed,

HDCAM records 2592000 of them, or about 42% of them.
DVCproHD records 2764800 of them, or about 45% of them.

But if you look at the ratio between recorded luma pixels and recorded chroma pixels,

HD 4:4:4 YCbCr for every 1 luma pixel there are 2 chroma pixels.
HDCAM for every 1 luma pixel there are 0.66 chroma pixels.
DVCproHD for every 1 luma pixel there are 1 chroma pixels.

So yes, HDCAM is less compressed than DVCproHD, but it also records less of the pixels from the uncompressed image, although it does this by recording more luma resolutio and a lot less chroma resolution. We can now look at the number of bits per pixel that is used to compress the image:

HDCAM uses 1.82 bits per pixel and
DVCproHD uses 1.26 bits per pixel

Because both formats balance the compression parameters very differently, especially the lower spatial luma resolution recorded by DVCProHD, it’s very hard to compare them directly just looking at the figures. A lower spatial resolution filtered down from the original 1920 to 1280 should be easier to compress because a) it’s smaller and b) will have a higher degree of super-sampling and hence have less sharp edges which cause DCT codecs bother. Also, the lower spatial resolution of DVCproHD might correlate better to the limitations of lens resolution on the camera, and this will especially be the case as we get more 1/3″ chip cameras using the DVCproHD codec.

Hope this is enlightening and if anyone has more facts or figures to add, that’d be great.

Graeme

– http://www.nattress.com – Film Effects for FCP