[Tig] DI workflow, capabilities and requirements

Christopher Noellert cnoellert
Sun Mar 27 20:40:16 BST 2005

On 3/26/05 5:29 PM, "Steve Shaw" <digital.praxis at virgin.net> wrote:

> Thanks to Pete Challinger of Edifis for supporting the TIG.
> --
> Hi Greg,
> Backs from the pub, after a few pints of 'wife beater', so accept any
> grammatical errors...
> The calibrated dailies approach is potentially cheaper than the traditional
> model as the colourist tends to be the most expensive part. Also, self
> calibrating transfer engines will be cheaper than telecines as they require
> less functionality. I'm impressed by the Cintel dataMill as the present
> likely system of choice, and its book price is a lot less than traditional
> telecines. It can also scan film at SD, HD 2K and 4K, all calibrated to
> cineon LOG specs. However, you don't need to scan dailies and the final high
> resolution on the same system as any transfer devise that works to cineon
> specs will be well within acceptable tolerance if calibrated correctly.

I agree that the DataMill looks very interesting on paper and It will be
interesting to see what their Ditto thing is all about, but my chief concern
about a machine like the DataMill (which is the same it is for the Mill2) is
the stability/drift of a tube machine on such long runs as you do for

The 444 and 4k options for DataMill put it up also in a price range where
you might get more use out of a Shadow, a TCS box and a used 888... Of
course the Shadow's not a Data scanner, but we're talking about dailies and
rushes workflow.  The CCD virtually ensures that your material will have
minimal drift and the TCS will calibrate you up against the stock (provided
it's Kodak) using the kc reader.  When you're not scanning dailies you can
do a music video.  If you don't want to use the luts in the TCS and need
something even more correct you can run the signal through the Truelight box
calibrated for the printstock you're going to use on the job and you're even
closer to what's what.

My 2 cents.
> Due to costs the present approach is still to scan for dailies/offline at HD
> or below, then re-scan for the online at 2K (4K?) using a pull list from the
> offline. We are all waiting for speed and storage capacity to allow full 2K
> scanning for dailies that can be used for the offline and online, but that's
> some time away, although for Silence Becomes You we did obviously capture
> all material in full resolution on digital media - 25TB in total, so we are
> getting there.

I can say like this... For Super16 and lower budget 35 work why not rush
directly to SR.  Calibrate up the Spirit/Shadow (I'm boycotting Cintel on
this one) against the TCS box.  Then rush labroll down to SR without passing
through the TCS so that you get the log encoded signal in 444 down to the
SR.  At the same time route the signal that's going to the SR into the
Truelight box with a lut for the printstock, recorder profile and interneg
profiles you're using on the job and downconvert that signal to a digibeta
or DVCAM for offline.

Make one SR for each Digibeta and label them both accordingly.  When you've
got the final EDL, capture and conform from the SR tapes, grade and shit it
out to the recorder.  Doesn't get easier/cheaper than that.  If the client
doesn't want to pay for the SR tapes, blank 'em and use them for the next

I find all the talk about disks and san and nas kind of funny.  Nothing is
cheaper or faster than tape. For the time being.

I know, "The future, the future" but now we're talking about "the present,
the present".


Chris Noellert
Frithiof Film to Video. AB
Sturegatan 58 
114 36  Stockholm

Tel: +46 8 545 678 78
Fax: +46 8 545 678 79
Mob: +46 735 32 00 03
AIM: cmnoellert

More information about the Tig mailing list