Tape vs Firewire Direct Recording?

[Kevin Spahr]

I always assumed that what was recorded onto a tape would provide exactly the same quality as recording the same thing through the firewire connection - did any one else ever compare them?

I set my DVC30 up and connected it to FCP via firewire and captured that video and then I captured the same video from the tape (through the same firewire connection) and when you compare them, the firewire connection has darker blacks and more vivid colors and maybe even a little better resolution! I did not change the capture method in FCP - but I think I might experiment with different codecs in FCP.

I assumed that the data coming out of the firewire cable was the same as the data being written to the tape.

Can anyone enlighten me? I can see the difference very plainly - I don't think I'm a 'nutter'.

According to this, using a direct to disk recording method would not only speed up ingesting - it would also provide better quality.

[Noah Kadner]

You're just seeing the standard sort of gamma shift you get in FCP. The quality and actual gamma of the original signal is identical. Also you can do the same thing in camera if you reset IRE to 0 from 7.5. Recording direct to hard drive is obviously faster than shooting tape and ingesting tape. But it has a couple of drawbacks- no tape archive to fall back on and the encumbrance of carrying your computer and hard drives around.

And that said, the signal would be identical if you setup everything correctly. DV is what it is, the compression happens at the moment of recording image, not much you can do to undo that save for getting a higher image quality camera.

Noah

[Kevin Spahr]

I don't think that's what is going on because both images are were captured with FCP, with the same settings in FCP, and the difference is seen while viewed in FCP. The only difference in the images were that one was captured live and the other was what was recored on the tape. No changes were made on the camera or in FCP. How would FCP cause a gamma shift in one image and not the other - how would FCP know if the image is live coming in through the firewire cable or coming off of a tape through the same firewire cable?

I tend to think that FCP is not the variable in this equation (since nothing is changed in FCP) - that leaves the data path or processing of the data in the camera.

I asked this question because I thought that since the tape stored the images as digital data and the data from the tape would be transferred via firewire directly to the computer and since the path is all digital there should be very little difference between them. Maybe the camera does some additional processing either going onto or off of the tape. Some where something is causing some loss in quality. I would love to see a block diagram of the data path in the DVC30 - that might explain things.

I plan to do a more precise test when I get some free time...

I originally was interested in the quality differences between Panasonic's Frame-Mode and standard interlaced mode. But I was surprised by the differences in the live video as compared to the taped video (visible in both modes btw).

[Kevin Spahr]

I exported these images from FCP as tiffs and then zoomed 200% to make the differences easier to see.

The one on the left is live the one on the right was taped. The quality of the diagonal lines is interesting - less combing effect.