Forum Replies Created

Page 16 of 196
  • Joseph Owens

    March 7, 2017 at 4:20 pm in reply to: Export DCP fail in Davinci

    [Shermen Chan] “Timeline length: 1hr 27mins
    File size: 1.5GB”

    In preamble:
    What (tiny) file would this be referring to?
    Is your installation of easyDCP licensed?

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

  • If the magenta is quite distinct, you could likely do a secondary isolation and then blur the heck out of it. That would be the down and dirty.

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

  • Joseph Owens

    February 9, 2017 at 4:48 pm in reply to: DNG to dark?

    It used to be that if you had no IRE spread, either all 0’s or all 1’s… indistinguishable from a blank sheet of paper.

    What species of DNG are you working with? If you have any kind of metadata describing exposure index, maybe you can pump those numbers to see if there is anything down there in the inky abyss.

    Then its all about noise suppression and banding. You are trying to scale a 100 foot cliff with a ten foot ladder.

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

  • Joseph Owens

    February 4, 2017 at 6:27 pm in reply to: Lens and Raw corrections in Timeline-clips vs Clips

    [Adriano Castaldini] “Is there any difference between applying those corrections directly on the clips instead of doing the same on timeline-clips?”

    It probably doesn’t matter much if they are relatively static compositions. If there is a lot of motion in the shot, or camera reframing, it might be a better strategy to deal with the specific trim selection in the edit/color timeline.

    There would be a definite advantage in getting this kind of adjustment out of the way as soon as possible. Either in the browser or the first or second adjustment node, depending on the application of a colorspace LUT. If you intend on doing much tracking or stabilization, you do need to set the geometry before invoking any of those other grade-related operations, as they are not hierarchically linked or parented. That is, if you have performed a stabilization or tracked a window to follow an object in the frame — if you re-define the geometry with some other operation like lens correction or a stabilization, the window track will not automatically take that into account.

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

  • Joseph Owens

    February 3, 2017 at 5:55 pm in reply to: Which format/codec for best performance?

    Ive experienced 1080 ProRes playing back just fine as a single stream from a USB2 drive. That’s probably not a good benchmark, there are a lot of factors.

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

  • Joseph Owens

    February 2, 2017 at 5:41 pm in reply to: Which format/codec for best performance?

    [Paul Meyer] “fastest performance in DaVinci Resolve? TIFF, OpenEXR, DPX, PNG, ProRes 4444?

    I feel like the answer should be out there already, but I couldn’t find it using google…”

    Do you really mean “optimal” performance?

    You are talking about loading the system with a variety of sequential formats and a Quicktime .mov container, with a related bitrate/compression disparity.

    With a strong enough system, you would not notice any difference between any of these formats, but low confidence in your specific instance makes me think you should probably stick with the ProRes option. 444 is probably also overkill, unless there is a theatrical or other high-bandwidth deliverable… also dependent on whether using a larger capacity media type makes sense with respect to your source material. One tollhouse cookie in a container ship is still just one tollhouse cookie.

    Unless your MacPro is especially lame, Resolve will likely still be faster than the read/write cycle of your storage, irrespective of format. Larger files will simply take longer to be retrieved from your drives, and teraflops of processing will be wasted in idle mode.

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

  • [Eric Santiago] “I don’t believe this was the case years ago.”

    It likely had something to do with the project database. The evolved versions allow more flexibility. One thing to be very careful about is inadvertently introducing operations in Studio that are unsupported in Resolve-Free. That used to just crash on opening in the non-license version.

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

  • Joseph Owens

    February 1, 2017 at 7:18 pm in reply to: Davinci Resolve vs Lumetri and Media Encoder?

    [David Lah] “I want to know if Resolve uses different processing algorithms when making color changes. E.g. if I drop blacks, will Lumetri crush them before Resolve does?”

    No idea. Mostly because I am not familiar with Lumetri. What I can state is that Resolve performs all of its calculations in 32-bit float which should infer that its numerical approach is more likely able to accommodate almost any range of values that you throw at it. Resolve, as an evolution of the original daVinci system, was and pretty much remains one of the few grading applications that can operate in constant-luminance, inferring that an RGB re-balance will not adversely affect brightness/contrast impression of the resulting image. Its a subtle distinction, but massively important and more appreciated as experience with grading grows.

    [David Lah] “if I drop or lift the same slider in Lumetri and Resolve will Resolve process it better and preserve crushing or clipping.”

    Not sure what you mean by “preserving” crushing or clipping. To me, those are destructive consequences. If you are hoping that somehow pushing the limits of the exposure range in one part of a grade can be retrieved in a concatenated later process… that’s not how it works.

    As far as Render goes… about all that I have observed is that the Rule of []Good []Fast []Cheap applies.
    You get to pick 2 boxes, and no, you can’t check one of them twice.
    If I was going to introduce one other factor, clients have commented very favourably on renders and compressions that I have exported out of Apple Compressor4, which is coming close to shading the above Rule.

    In the matter of *visually lossless* and *lossy* codecs, I do notice differences in exported media from AME as opposed to Resolve. A national broadcaster sent me some feedback not long ago indicating a preference away from Media Encoder. “too many short cuts” was the informal explanation, which does seem to underline Fast& Cheap. Excludes Good.

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

  • Joseph Owens

    January 31, 2017 at 3:53 pm in reply to: Conform hell…

    [Marc Wielage] “getting everybody to agree with those standards will be problematic.”

    If not proprietary in some cases.

    And then there is politics. This is received and experiential knowledge spanning from a time when there were only three standards for television production, and they just keep multiplying.

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

  • Joseph Owens

    January 30, 2017 at 7:50 pm in reply to: Conform hell…

    [Hector berrebi] “Did you try to use it for anything other than basic stuff? “

    I’ve been grading with daVinci (Renaissance, 8:8:8) since 1994, started NLE grade with Silicon Graphics “Final Touch” (purchased by Apple and re-released as “COLOR”). Switched back to DaVinci with release 9 (multi-layer) and basically snuck a peek at SpeedGrade when it came in the Crackerjack box with Premiere. Closed it and never opened it again.

    Haven’t been interested in anything else. Previewed Baselight. Couldn’t afford the $125,000 license, or the economy offering …another plugin for layering on top of an NLE timeline…. really???. Pablo, Mistika… always interesting to get opinions when demoing at shows….

    jPo, CSI

    “I always pass on free advice — its never of any use to me” Oscar Wilde.

Page 16 of 196

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