Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Activity Forums AJA Video Systems AVI / QT Codex for Xena

  • AVI / QT Codex for Xena

    Posted by Carl Skaff on May 18, 2008 at 5:21 am

    Hi All.
    I’ve been searching the web for answers but can’t find it so I thought I’ll ask you…

    I have a brand new Xena 2K installed and running Premier and Machina.
    My goal is to capture uncopressed 8 & 10bit QT, AVI, DPX, Tiff & Tga from either SD or HD 4:2:2/RGB4:4:4 from my Spirit Telecine.
    (the files needs to be cross-platform compatible on Mac, PC & Linux. For apps like: FinalCut, Premiere, AfterEffect, Avid, Shake, Combustion, Nuke, Flame, Flint, Smoke, Luster, Resolve, Nucoda, BaseLight… Not saying that the QT needs to be able to import on an Avid, but if it imoprts it, it can’t be any gamma/gammut problems.)

    In the past we used to record to HDSR and then ingest to a FCP in “uncopressed” using a AJA Kona3.

    My Q is:
    When in Machina, ther is a lot of format options to chose from but I can’t understand what the diference is between then are. Does anyone know what they all mean?
    Here’s a list of the ones:

    AVI files in the following Subtypes:
    8-Bit YUV 4:2:2 – ‘2vuy’
    8-Bit YUV 4:2:2 – ‘2Vuy’
    8-Bit YUV 4:2:2 – ‘uyvy’
    8-Bit YUV 4:2:2 – ‘v210’
    8-Bit RGB 4:4:4 – ‘bgr’
    8-Bit RGB 4:4:4 – ‘bgr’ T2B
    10-Bit RGB 4:4:4 – ‘R10k’
    10-Bit RGB 4:4:4 – ‘r10k’
    10-Bit Log RGB 4:4:4 – ‘R10g’

    QuickTime files in the following Subtypes
    8-Bit YUV 4:2:2 – ‘2vuy’
    8-Bit YUV 4:2:2 – ‘2Vuy’
    8-Bit YUV 4:2:2 – ‘v210’
    8-Bit RGB 4:4:4 – ‘rgb’
    8-Bit RGB 4:4:4 – ‘bgr’
    10-Bit RGB 4:4:4 – ‘R10k’
    10-Bit RGB 4:4:4 – ‘r10k’
    10-Bit Log RGB 4:4:4 – ‘R10g’

    DPX Sequences in the following Subtypes
    10-Bit RGB 4:4:4 (BE) T2B
    10-Bit RGB 4:4:4 (BE) B2T
    10-Bit RGB 4:4:4 (LE) T2B
    10-Bit RGB 4:4:4 (LE) B2T
    10-Bit Log RGB 4:4:4 (BE) T2B
    10-Bit Log RGB 4:4:4 (BE) B2T
    10-Bit Log RGB 4:4:4 (LE) T2B
    10-Bit Log RGB 4:4:4 (LE) B2T

    TIFF Sequences in the following Subtypes
    8-Bit RGB 4:4:4(BE) T2B
    8-Bit RGB 4:4:4(BE) B2T
    8-Bit RGB 4:4:4(LE) T2B
    8-Bit RGB 4:4:4(LE) B2T

    TGA Sequences in the following Subtypes
    8-Bit RGB 4:4:4 T2B
    8-Bit RGB 4:4:4 B2T

    Kevin Christopher replied 17 years, 11 months ago 4 Members · 11 Replies
  • 11 Replies
  • Ramona Howard

    May 18, 2008 at 10:01 pm

    Carl,

    Let me say jump on the bandwagon and record to DPX. Save yourself some frustration of all the codec nightmares. DPX is supported widely and is the choice in the film-world…..

    We have become very experience in this workflow (the other half of our customers are VFX) and I think you are going to find out that the off-the-shelf or free apps are not going to give you everything you need for a Telecine environment. It is fairly demanding.

    I can certainly refer you to some professionals in this arena (that are not in your area, so no competition issues) that can enlighten you since your going down this path.

    Please feel free to email me off-list.

    ramona@spectsoft.com

    Cheers,
    Ramona
    Rave now supports 4K
    http://www.spectsoft.com

  • Carl Skaff

    May 19, 2008 at 6:36 am

    Yes DPX is for shure the way to go.
    But our plan is to start delivering dailies/onelights as QT to clients for offline. And FCP doesn’t do to well with dpx if an non-experiesed client is running it. (as a lot of client are)

    But even for DPX it has a lot of choises: like BE / LE / B2T / T2B???
    Whats what I need to find out. And the weird codecs for QT’s.

    But thanks anyway for the quick answer.
    I have been looking into a more expencive DDR appl for this; QuickClipPro from Drastic. Its a bit better then Machina cus you can controll it from the daVinci 2K+. But for my use Machine should do just fine. No need to go big if I don’t use the fancy tools.

    /carl

  • Jeff Brown

    May 19, 2008 at 3:27 pm

    Carl,
    The abbreviations I believe are:
    BE : Big Endian
    LE : Little Endian
    B2T: Bottom To Top
    T2B: Top To Bottom

    These are all variations in the way the frame data is written to the file; the order of the data within the file. It would all be the same data in the end. An analogy: each frame is a book; you can choose to order the pages front-to-back or back-to-front, or to read left-to-right or right-to-left. The story would be the same.
    In the end, it should not matter other than your external software (FCP, Premiere, After Effects, etc.) probably renders one “flavor” of DPX, one “flavor” of TGA, and so on, so that would be a good one to standardize on.

    -jeff

  • Ramona Howard

    May 19, 2008 at 4:02 pm

    Carl,

    feeling a bit long winded here, sorry…..

    Actually FCP does just fine with DPX frames using a little app called Glue Tools.

    DPX is not as difficult as you might think or it might seem. Most apps write out a standardized file that is cross compatible (or at least they should, I know we do). It is all in how the data is packed in the header. The biggest point is with DPX, you get bit for bit (both log and lin) of the image, plus all the metadata you need.

    I hope you are looking beyond the bells and whistles for a DDR. It is about how solid the product runs with the devices, remember your replacing a tape deck that has a proven track record.

    Machine control (RS422) seems so easy……yet has been a tricky thing to get right. Doing this in slave mode is even more tricky….. Just remember not all products work the same way just because they use the same protocol. Most standards are left open for interpretation and thus you get a varying level of accuracy in how it works.

    We have been developing Rave for about 8 years. It took so long because it is on Linux and we wrote everything from the ground up, drivers, libraries, machine control, etc…The point is: success can be achieved with the right approach to the problem. We have stepped into many studios and replaced DDRs that have been around for years because of the machine control. Many colorist prefer a Rave vs XXX due to how well it responds. We are native DPX………..

    Sorry for the diversion. DPX you will see is becoming the de-facto standard in the film world, I think you will even see this from a few unexpected parties (sorry can’t spill the beans). Other codecs seem to be struggling with getting this bit for bit thing down so I jumped in to hopefully help you look at the problem another way.

    Best of luck on you project.

    Cheers,
    Ramona
    Rave – The only DDR that also gives you access to the source code

  • Carl Skaff

    May 19, 2008 at 5:56 pm

    Thanks Jeff. Thats the info I’ve been looking for.
    Now its just al the QT flavours to figuer out…

    Ramona…
    Dude, I know if anyone that DPX is “da shit”.
    Im a colorist working on Resolve and daVinci 2K+.
    Resolve work nativly in dpx and thats all we use if we are doing anything between systems like our Smoke, Flint…

    But its kind of hard to say to a Client that needs his 2hrs of onelight deliverd on a FireWire drive as uncompressed QT. And then try to convince his that dpx is better. It might be, but his workflow and understanding is for shure not supporting dpx.

    All I need from a DDR is, not even remote, to crash record while capturing ext LTC. It must be able to record native dpx, tiff, QT, AVI etc etc.

    And to my big surprise Machine (together with a LTC>RS422 from Adrienne Electronics Corporation called AEC-mBOX-2, only $395) was able to do it.
    It also need to run on my WindowsXP/OSX dualboot MacPro conected to my BrightSAN. Cus we also use that box to copy stuff to and from the san over USB/FW400/FW800/ESATA.

    Although I haven’t got my cheep-ass-DDR setup fully working yet (since I’m bissy Grading) but if I run into problems I might look into your product. Feel free to email me som info and list-priceing. (no internal storage needed, we connect to our SAN)

    So if anyone knows of the diferent QT-flavours please let me know what ther mean.
    I’m planing to do one of each and compare them in FCP and Premiere (thats mostly what my clients will use) and compare them to the same material ingested on that workstation.

    /carl

  • Ramona Howard

    May 19, 2008 at 6:13 pm

    Carl,

    Nice to know you actually know what your talking about.

    DPX vs QT on the customer end….oh yes we fight that battle too. Many house switch because they really do see the benefit, eventually. Keep up the fight 🙂

    Adrienne is a great company, we use their LTC board (wrote our own driver to tie into the AJA board so we always get accuracy) good choice…..

    and Bright…..also another good choice, we support that one also.

    Best of luck,
    Ramona
    Rave – Uncompressed made easy

  • Jeff Brown

    May 20, 2008 at 2:54 pm

    I have a couple clients using FCP; the real annoyance has been that FCP doesn’t directly support file sequences (stupid!)– has that changed recently? Many Mac users seem to thing QuickTime is the be-all and end-all for footage; they don’t grok the benefits of file sequences.

    -jeff

  • Ramona Howard

    May 20, 2008 at 3:58 pm

    Jeff,

    Watch the boards 🙂

    Quicktime is great but I would much rather work with file sequences where at all possible.

    During an edit if I need to replace a small sequence, I simply drop in those frames rather than replace the whole clip.

    During a render if a few frames go bad, I would rather re-render just those frames rather than the whole sequence.

    wrapped formats such as QT, AVI and others are great for delivery but while I am working I want full control over every last frame….

    Everyday more and more of the “video” side is catching on to how the “film” side works and hopefully another couple years at it, there won’t be a difference.

    To grok (pronounced /ˈgrɒk/) is to share the same reality or line of thinking with another physical or conceptual entity. In Heinlein’s view of quantum theory, grokking is the intermingling of intelligence that necessarily affects both the observer and the observed.

    Some people just don’t want to think. Honestly, i have days like that 🙂

    Cheers,
    Ramona

  • Kevin Christopher

    June 3, 2008 at 3:41 pm

    From Your List:

    QuickTime files in the following Subtypes
    8-Bit YUV 4:2:2 – ‘2vuy’ — Apples New 8 bit YUV File
    8-Bit YUV 4:2:2 – ‘2Vuy’ — Original Apple 8 Bit YUV File back in the cinevision days
    8-Bit YUV 4:2:2 – ‘v210’ — Not 8 bit, but 10 bit YUV File
    8-Bit RGB 4:4:4 – ‘rgb’ — Simple 8 bit RGB File (Red, Green, Blue)
    8-Bit RGB 4:4:4 – ‘bgr’ — Same RGB File Written backwards
    10-Bit RGB 4:4:4 – ‘R10k’ — 10 bit linear RGB File
    10-Bit RGB 4:4:4 – ‘r10k’ — 10 bit linear RGB File
    10-Bit Log RGB 4:4:4 – ‘R10g’ — 10 bit Logrithmic RGB File

    Hope this clears some things up.

    If you have any other codec questions take a look at:
    https://www.fourcc.org/codecs.php

    Kevin

  • Carl Skaff

    June 3, 2008 at 6:29 pm

    Cool.
    Thats interesting.
    Although that brings som folow-up questions:

    If the ‘v210’ is in fact 10bit, I gues that one is better then ‘2vuy’?
    (if my source is an uncompressed 4:2:2 SDI signal from a Telecine)

    Whats the diference betwen ‘R10k’ and ‘r10k’

    I gues that if I’m creating a QT from a DualLink RGB 4:4:4 SDI signal then I should select ‘R10k’ / ‘r10k’
    And if I have the source as YUV 4:2:2 SDI then I should select ‘v210’

    Am I thinking correct?

    /carl

Page 1 of 2

We use anonymous cookies to give you the best experience we can.
Our Privacy policy | GDPR Policy