calling all technical gurus – just curious

[Ben Holmes]

You’d be best off going firewire in from the DV deck to the Mac. That way, there’s no conversion or data loss with the ingest. What you see is what you got. Whether or not you prefer the ‘look’ of other methods is another question entirely…. But SVHS shouldn’t be a ‘better’ option.

Ben

Editec Broadcast Editing Ltd
EVS & FCP specialists for live OB operations.

Producer/Director “The Supercar Run” now available for international distribution from http://www.electricsky.com

[Rennie Klymyk]

[Ben Holmes] “But SVHS shouldn’t be a ‘better’ option”

Yes I agree. Sometimes the s-vhs comming from a digital feed looks realy good but it is still chroma all mixed on 2 wires and luma on the other 2 wires. You lose the rgb component value. DV inspite of the 5:1 compession is still componet over firewire at least.

“everything is broken”

[Walter Biscardi]

First off, I’m assuming you mean S-Video, not S-VHS. S-VHS is a videotape format while S-Video in a video I/O.

If you are mixing DV footage with DigiBeta, I would bring it in via S-Video to make it all match. As you have observed, bringing it in via S-Video actually results in a better image in an 8bit timeline over bringing in the footage via Firewire, then rendering it in an 8bit timeline.

Always better to digitize all the material into the same format during ingest rather than doing all the renders.

Walter Biscardi, Jr.
https://www.biscardicreative.com
HD Editorial & Animation for Food Network’s “Good Eats”
HD Editorial for “Assignment Earth”

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

[Alan Lacey]

I’m really interested to hear you say that Walter, you obvioulsy know through experienc, any ideas why that should be the case?

Alan PALland

[Ben Holmes]

I suppose the difference here is that you are importing a good quality S-Video (component Y-C) signal via an uncompressed codec, much in the same way that the component HD outputs on an HDV deck, or the HD-SDI outs on a Canon HDV Cam are superior. Better to let the deck process the video and colourspace than the computer.

Note that we’re not talking about mixing material here, just working in DV. Is this still a better option? Does this say something about the processing in FCP? You’d think dropping the DV stream ‘digitally’ into an uncompressed timeline would be better, but is this a question of processing?

Interesting.

Ben

Editec Broadcast Editing Ltd
EVS & FCP specialists for live OB operations.

“The Supercar Run” now available for international distribution from http://www.electricsky.com

See you at IBC – we’re there with our FCP/replay truck…

[Glenn Chan]

FCP doesn’t apply any chroma smoothing when it decompresses DV. If you had taken the SDI output off a deck instead, most decks will smooth the chroma (into 4:2:2 color space, not 4:4:4) and then that will go into FCP.

WIth firewire ingest, you could apply the ?color? smoothing 4:1:1 filter to smooth out the chroma. With material ingested via SDI, you could apply the color smoothing 4:2:2 filter.

2- If you want to mix SDI ingested material and DV, sometimes it’s better to ingest the DV via SDI so you don’t have to muck around with the differences between the 720×486 and 720×480 frame sizes.

[Rennie Klymyk]

[Ben Holmes] “Note that we’re not talking about mixing material here, just working in DV. Is this still a better option? Does this say something about the processing in FCP? You’d think dropping the DV stream ‘digitally’ into an uncompressed timeline would be better, but is this a question of processing?”

I don’t work have digi-beta but for DVCAM I use a dsr-11 and a dsr-85 (no sdi card). For ingesting into fcp I would opt for the dsr-85 component -AJA IO -fcp. I would think by powering down the AJA IO and ingesting from the dsr-11 via firewire I would get the same component signal from the tape but with unbalanced audio. Because s-video is reduced to chroma and luminance I’ve always stayed away from that option. Since the advent of component betacam we’ve always been able to capture the real world through our lens to seperate ccds keeping the primary colors seperate ultimately till broadcast and these days right to delivery on the viewers component tv. With today’s dsp and equipment like Kim’s A510 these technical attributes do seem to blur ( or should I say meld).

“everything is broken”

[Glenn Chan]

I believe betaSP (and Digital Betacam and DVCAM) converts the signal into a luma component and two color difference signals/components (chroma).

On tape, more bandwidth is given towards luma than for chroma. This is to optimize the signal for our visual system. We see less color resolution than brightness resolution. Video doesn’t fully seperate out the color from the brightness information- a little portion of brightness information is found in the chroma, and a little portion of the color information found in the luma. This is most visible in the dark band between the green and magenta bars in color bars. In practice it’s not a big deal.
*The problem stems from the way luma and chroma are formed… the order of operations between gamma correction and forming the weighted sum is reversed. This lowers cost / improves performance (and may improve S/N ratio; not sure).
If ‘ stands for gamma correction, then the video formula is:
Rec. 601: Luma (Y

[Kim Rowley]

Thanks to all for your input. It’s all pretty interesting. I would have thought as Rennie, that ingesting DV straight through firewire was “better” looking. My evaluation is my eye at this point, not waveforms or mathematical equations – but I was surprized by my “real world” results. The bottom line is I think I’ll go ahead as always when I have to mix DVCam and Beta footage which is also the workflow Walter also suggested. If the job instead originates in DVCAM I’ll use common sense and probably go firewire to save on disk space. My DVCAM VTR does not have component outputs. When I chose the DSR 25 I did so knowingly. The need I have for the DVCAM at the present didn’t warrent the added cost of the component possibility. And the fact that this machine -like the DSR 11- reads and records NTSC and PAL was a more important feature for me. There’s always a trade-off somewhere. The important thing I think is knowing what you’re trading and making thought-out decisions.

Thanks Walter too for the subtle correction of my terminology. I did indeed mean S-Video output.

Dual 2.7 GHz G5, 4GB RAM, ATI Radeon 9650, Xserve RAID, AJA IO, 2 20″ Cinema Display, FCP 5.03, OS X10.4.3

[Walter Biscardi]

[Kim Rowley] “My evaluation is my eye at this point, not waveforms or mathematical equations – but I was surprized by my “real world” results.”

And that’s really the only way to go in my opinion. Let everyone else delve deep into the whole technical and mathmatical reasons for why something is better. At the end of the day, your clients don’t do math, they simply look at your monitor.

[Kim Rowley] “Thanks Walter too for the subtle correction of my terminology. I did indeed mean S-Video output.”

Not a problem, a lot of people do that. Quite honestly S-Video is just as good as component to the eye so you didn’t lose out on anything by not getting component on your deck.

Walter Biscardi, Jr.
https://www.biscardicreative.com
HD Editorial & Animation for Food Network’s “Good Eats”
HD Editorial for “Assignment Earth”

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