HDV File Size?

GaryV wrote on 5/8/2008, 9:56 AM

I currently have an SD mini-DV camera. I find that when I capture the footage, via firewire, I end up with ~13GB .AVI file for a 1 hour mini-DV tape.

I am considering upgrading to a mini-DV based HDV camera (eg. Canon HV20/30 or Sony HC9). I understand that I can capture 1 hour of HDV on a mini-DV tape with these cameras. When I capture this footage to my computer (via firewire), how large are the resultant files (.m2t I think instead of .avi) ?

Is it also ~13 GB for a 1 hour HDV mini-DV tape when captured to my hard disk?

Thanks in advance,
Gary

Comments

johnmeyer wrote on 5/8/2008, 10:25 AM

Is it also ~13 GB for a 1 hour HDV mini-DV tape when captured to my hard disk?Yes, it is exactly the same size.

However, if you later create uncompressed versions, or render to some intermediate like Cineform, those files will be 4x larger than their DV counterparts. Also, internally on your computer, each frame must be "unpacked" in order to be displayed. Since HDV has 1440x1080 pixels (that's 1,555,200 pixels) compared to NTSC DV which has 720x480 (345,600) each frame will take 4.5 times as much RAM. Most people won't notice, unless their computer doesn't have much RAM. In that case, what worked for DV will have problems with HDV.

In general, however, the transition should be pretty smooth, except that timeline performance when editing the native m2t files is not as fast as DV. To somewhat get around this, if you capture or convert to Cineform AVI, the difference between editing DV and HDV is far less noticeable, but you'll need 60 GBytes/hour storage instead of 13 GBytes/hour.

GaryV wrote on 5/8/2008, 11:03 AM

Thank you very much John. I appreciate your help.
Cheers,
Gary

Terry Esslinger wrote on 5/8/2008, 1:37 PM

if you capture or convert to Cineform AVI
John is that basically the same as when using Gearshift? Which is preferred?

johnmeyer wrote on 5/8/2008, 2:28 PM

Gearshift is used for proxies, not intermediates.

What does this mean? There are three ways to edit HDV:

1. Put directly on the Vegas timeline the m2t files you captured from tape. Playback performance is slow, especially once you add fX, compositing, etc. However, for simple projects (cuts-only with a few fX, titles, etc.) this is the way to go. I have an old (5.5 years old) computer and this works fine with Vegas 7.0d.

2. Convert to Cineform intermediates. These are high-def AVI files. You will still be editing in high-def, with all the advantages of being able to see exactly what you are going to get. If you need to do multiple renders or advanced fX, you will suffer virtually no further degradation, something that can happen when dealing directly with the m2t files. This is because the compression for m2t is a lot more lossy and therefore multiple generations lose a lot more quality. The playback performance with Cineform is faster than m2t, but not as fast as good "old fashioned" DV files. Also, they take 60 GBytes for each hour as compared to 13 GBytes for DV or m2t files.

3. Convert to a low-res proxy, and then just before rendering, switch back to the m2t files. Like #2, this involves taking an extra step to render another set of files from your original m2t files. However, unlike Cineform, these files are low resolution, typically standard definition DV files. You then edit with these files, apply your fX, etc. Since the resolution and possibly the colorspace (depending on what file format you use for the proxy) is different from the original HD file, you cannot do really critical corrections, especially color. You really can't do green screen and other things that depend on exact color and pixel placement. However, for most work, they are adequate. You get fantastic playback performance on the timeline, and not much additional disk space. Just prior to rendering, you replace the proxy (which is a word that basically means "substitute") with the original m2t file and then render. All your cuts and corrections are preserved, but they now operate on the original file instead of the proxy. Since you may have created dozens or hundreds of these proxies, this last step of doing the substitution -- as well as the initial step of creating all the proxies -- can be tedious. Fortunately, they are lots of commercial scripts that will do this for you with the push of one button.

So, if you have cuts-only, use the m2t.

If you have to do serious color corrections, green screens, etc. and still want timeline performance, use Cineform.

If you have an old computer or just want the fastest possible timeline performance, use proxies.

If you purchase the full version of Cineform, and have lots of disk space, you can actually capture directly to this format. With a fast computer and plenty of disk space, this is definitely the way to go.