Why 2k instead of 1080 24P? answer here!

Home Page forums Archive Rumors/Tips Why 2k instead of 1080 24P? answer here!

  • This topic has 6 replies, 5 voices, and was last updated 16 years ago by Anonymous.
Viewing 7 posts - 1 through 7 (of 7 total)
  • Author
    Posts
  • #199688
    Anonymous
    Guest

    from the discreet forum

    “Lately i’ve been quoting a film conform job. My client wants a 2k film conform to then prin it out to film.

    Whille thinking about the workflow for such a huge job in 2K…”

    answer: get Scratch!! is going to make you change your mind. I manage my way in a private demo at IBC, painless and affordable 2k finally!

    ditto

    #208771
    Anonymous
    Inactive

    Hi guys,
    we are working mainly with DI, but i can tell you why, you should use
    2K files (10 bit log) instead of 1080 24p (Panasonic HD D5-8bit linear) !!!

    Simply because of the headroom in color correction, i mean DI (D5)
    is 8bit 4:2:2 and compressed on the HD tape. (painful during comping
    bluscreen plates, cause of the compression artefacts and less colorspace)

    If you only want to print out, f.e.Arrilaser, DI is enough, cause the cinema
    release print is almost 8bit, but if you have a lot of keying and comp work
    i would prefer 2K files.

    I mean we screen every day dailies in our own digital cinema (barco digital projection system) and there are a lot of pros and contras, but my opinion
    is that Digital Intermediate is good enough for the audience in the cinema.

    How many people really see the difference???, even our directors somtimes
    don’t know whether is film or digital projection…

    If you know how to handle and final grade, i bet knowone is able to see the difference…….

    Regards,
    Mark

    #208773
    Cueto
    Participant

    OK, agreed. The panasonic D5 HD is a 4:2:2 (YUV) colorspace system and it is compressed 4:1 (360 Mb data to tape).

    We were considering using the new sony HDCAM-SR, which is capable of recording to tape at 4:4:4 (RGB). This deck is capable of 593 Mbps which on 4:2:2 colorspace represents about 2.7:1 compression. on 4:4:4 is more keen to 4:1 compression.

    that is why we were considering transfering ourmaterial to HDCAM-SR 4:4:4 and taking it from there to create a DI.

    Since we won’t be cutting back and forth between scaneed material and unprocessed film, i Don’t think the diference would be easy to apreciate.

    Any thoughts?

    Thanx

    Frank Cueto
    Reaktor Post

    #208772
    Anonymous
    Inactive

    If the colour grading you’re going to achieve is quite heavy you’ll find many artifacts such as dithering, banding, highlight cropping, bigger contrast already insert by the scaled down color space. If you can, I think the best way is to crop to the HD format your scan (I don’t know if it’s possible directly during the scan) but saving the files into Hard disks in DPX format who preserve much more information (due to its nature).

    HTH

    Francesco

    #208774
    Anonymous
    Inactive
    Anonymous wrote:
    from the discreet forum

    answer: get Scratch!! is going to make you change your mind. I manage my way in a private demo at IBC, painless and affordable 2k finally!

    ditto

    ASSIMILATE looked like it might develop into something good at some point but was still quite wet behind the ears.

    Most noticeable was the lack of secondary color correction.

    Very Cyborgish.

    Chris

    #208769
    Anonymous
    Guest
    Mindmachine wrote:
    Hi guys,
    we are working mainly with DI, but i can tell you why, you should use
    2K files (10 bit log) instead of 1080 24p (Panasonic HD D5-8bit linear) !!!

    Simply because of the headroom in color correction, i mean DI (D5)
    is 8bit 4:2:2 and compressed on the HD tape. (painful during comping
    bluscreen plates, cause of the compression artefacts and less colorspace)

    If you only want to print out, f.e.Arrilaser, DI is enough, cause the cinema
    release print is almost 8bit, but if you have a lot of keying and comp work
    i would prefer 2K files.

    I mean we screen every day dailies in our own digital cinema (barco digital projection system) and there are a lot of pros and contras, but my opinion
    is that Digital Intermediate is good enough for the audience in the cinema.

    How many people really see the difference???, even our directors somtimes
    don’t know whether is film or digital projection…

    If you know how to handle and final grade, i bet knowone is able to see the difference…….

    Regards,
    Mark

    we mainly shoot directly to disk via HD_SDI out of the HDCAM, get a nice uncompressed 10 bit (but YUV422) Quicktime for FX shots and the rest goes HDCAM. Better than manymanymany 35mm 2ks i’ ve seen 🙂

    main advantages are speed, cost, flexibilty and directors can feedback while shooting without having to wait for “the tent” .

    #208770
    Anonymous
    Guest

    > Whille thinking about the workflow for such a huge job in 2K…”

    Theory is theory, but in practice, 24p color grading provides the same results than 2K. We have tested the complete process for the same film work but using two different applications and workflows, working at different color dephts in the final part of the work:

    Scanner: C-Reality

    Digital Intermediate:

    System1 – Luster, working with DPX files at 2K RGB 10 bit, from the start to the end of the work, sending DTF2 with 10 bit DPX files to the film laboratory to print the film.

    System2 – Mistika ( before named “Jaleo”), Importing 10 bit DPX files while applying a first colour correction in RGB 4:4:4 10 bit, then converting to HD 24p until the end of the work, and sending a D5 HD 24p tape to the film laboratory

    The final results in film projection where impossible to distinguish (probably because film projection is equivalent to no more than 7 bit!). Finally, we have decided to work in Mistika, because it provides a true realtime workflow, and because it really provides all benefits from the HD technology in terms of production speed. We think that “Quality” is not only a mathematical issue, in fact it is more related with the amount of “human” time that you can dedicate to achieve it.

    In this sense, bit deph is not as important as having more time in order to freely experiment and refine the results. At the end, if your workflow is faster you will have more time to adjust the details, and the final quality will be higher.

    The real question is this: Before or after, the colour depth is always going to be reduced, so the key point is to chose the right point in the production chain for doing this reduction,

Viewing 7 posts - 1 through 7 (of 7 total)
  • You must be logged in to reply to this topic.
Copy link
Powered by Social Snap