Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
Please, support PV!
It allows to keep PV going, with more focus towards AI, but keeping be one of the few truly independent places.
Cluster X: Series 1 - moon - ЅріzZ - nebula - drewnet - Slipstream
  • 1008 Replies sorted by
  • I have 16gb sdxc 95mbs card will that be able to handle the New DrewNET patch?

    Thanks

    i plan on testing Moon and New drewnet here soon

  • @GravitateMediaGroup I will do a quick doodle search and will get back with you ; ) For me, it is based on testing with this particular camera and the results I have been getting in the process for the last two years. This is subjective. For example, we all have different threshold and tolerances for artifacts, noise, etc. And the different picture styles give different results( which are interpreted differently by each person) including post(grading,etc). I did use extensively standard before but evaluating the results, for me the eye candy comes with smooth.

  • @jclmedia Drewnet is a lower bit setting so there shouldn't be problems with this kind of card. Its the san disk extreme pro right? I have used it for Sedna and all Intras. Maybe you should test it for spanning just to be on the safe side if you need longer takes.

  • Here's an old test I conducted (this was EX Tele modes) showing the profiles and amount of bitrate each utilises. It still stands today as a test. The results showed (Ive put these in order below to show amount of avg bitrate each profile used for the same picture);-

    • Nostalgic - 80160 kbps
    • Smooth - 77315 kbps
    • Nature - 75641 kbps
    • Dynamic - 75448 kbps
    • Vibrant - 74243 kbps
    • Standard - 74119 kbps
    • Cinema - 64180 kbps

    Clearly the GH2 pushes the encoder around regarding the chroma/luma differences in each profile. I choose Standard because its baseline without any additions/subtractions. Whatever floats your boat - use it! Here's the video (in no particular order of bitrate size used);-

  • From 74-80 there is no way the naked eye could tell the difference.

    Also, this still doesn't increase or decrease DR between profiles.

  • The difference between cinema and nostalgic(the extremes?) is striking in cam. Less pronounced for standard and smooth but still clear if you look carefully enough(in cam) For the post, things get subjective as I mentioned earlier.

  • Does anyone else think it's funny that Driftwood is doing such excellent work, and there's people filling this thread with arguments about cards and profile settings and other stuff... and when someone asks about something that pertains more to the hack settings (like benefits of the intra, etc...), they get told to look elsewhere?

    Hopefully not starting anything bad... just wanted to offer a perspective.

  • @luxis After a brief search it comes down to properly exposing for whatever the subject or situation is, following a histogram, and don't believe what the cameras LCD is showing you. I guess this is why expensive monitors are a big deal.

    Either way. Back on topic. Moon 5 & 3 are pretty spectacular. Lol

  • @danielvy and to anyone else who wants to know why Intra is best for movement

    A very good explanation of GOP sizes and why any GOP with i frames over a half second apart is not so good for fast action/scene changes is here;-

    http://documentation.apple.com/en/compressor/usermanual/index.html#chapter=18%26section=5%26tasks=true

    An extra longer GOP (greater than classic 12/15 GOP) is ok for statics but as soon as you start moving the camera and following fast action the predictive frames will get pretty muddy.

    and a good point about 24p v 23.97(8) states;-

    "...If your source video has a frame rate of 24.00 fps rather than 23.98 fps, Apple Compressor (and no doubt FCPX) skips one out of every 1000 source frames. If the 24p source video is 23.98 fps, Compressor transcodes all source frames, without skipping (or repeating) any of them..."

  • Nostalgic - 80160 kbps Smooth - 77315 kbps Nature - 75641 kbps Dynamic - 75448 kbps Vibrant - 74243 kbps Standard - 74119 kbps Cinema - 64180 kbps

    ...what can be done in grading to simulate any picture style is doing so after subsampling and after compression. If the "flatter" picture styles are a fraud then that's negating the work and research of scores of people affecting not just this camera but all cameras. I guess Technicolor and other pros are just "croaking smack"?

    No, it doesn't take a genius to figure out that shooting with the lifted profile means that in the grading process the "flatter" shooter is pulling noise and compression artifacts down as they grade, making them less visible regardless of any noise-reduction being done. Gravitate is pushing his noise up.

    Plus, those numbers up there don't lie. That's more information making its way through the compression. Even if you crush some of it down, it's disingenuous to downplay an extra 5K-6K when the simple fact of pursuing these patches, and Driftwood's continued experiments and revisions, is an act of clawing and straining for every last bit that can make it onto the card. Well, that's my PV.

  • We know not what they (Pany) do with their profiles. We see with our eyes, we measure with YUV software and we form an opinion. In the grade I happen to prefer 'Standard' profile. A lot of people prefer smooth and Nostalgic. Im not saying anyone is wrong or right.

  • @burnetrhoades if this is the case, how come technicolor can't come up with a picture style equal or greater than a RED or BMCC dynamic range?

    latitude and DR are different, but in the same family. technicolor is misleading by saying it increases DR because this simply is not true, or all cameras should be able to produce the same DR, but they can't because it's a hardware ability, not a "camera setting" I've done some more research and it all comes back to the same thing "properly expose"

    technicolor is a company seeking customers, company's have been known to say things that's aren't 100% true to draw in more customers ; ) I have seen several other websites selling their "canon picture styles" and I have seen one very boldly admit something along the lines of "this will not, and can not boost your dynamic range, we aren't trying to sell you lies" they just admitted that it will give a flat baseline image, and others that will give a more cinematic look was crushed blacks and so on.

  • if this is the case, how come technicolor can't come up with a picture style equal or greater than a RED or BMCC dynamic range?

    Seriously?

    Maybe a new release is any day away "Now With More Witchcraft!"

    LOL

    technicolor is a company seeking customers, company's have been known to say things that's aren't 100% true to draw in more customers ; )

    Their profile is free. It requires no custom software. Back when I cared to stay informed about what was going on in EOS land I'm pretty sure they released it to the public before they had their "look suite".

  • they offer "one" for free. others cost. it's to bait you in, then convince you to buy. https://www.technicolorcinestyle.com/cinestyle-looks/

    there will NEVER be a picture style on any camera that will ever give an 8bit camera the same DR as a 12+ bit camera lol

  • Thanks to @driftwood , Vitaliy and everyone involved, i´ve learn a lot in this forums, really amazing.

  • there will NEVER be a picture style on any camera that will ever give an 8bit camera the same DR as a 12+ bit camera lol

    Only you are making that comparison. You're stating something obvious like you're teaching us all something.

  • @burnetrhoades eh, I was starting to get the impression that some people here actually thought a picture profile/style magically created more DR. Good thing you have a understanding that this isn't possible.

  • Yeah. But I see the value in manipulating the DR that's there. This whole hack/patch thing and pushing these cameras to places Panasonic (or Canon for that matter) never dreamed of when they were designing them is a fight for every little bit you can get. It's like Pacino's great half-time speech near the end of Any Given Sunday about fighting for every inch.

  • What was this thread about again? How to make chicken soup right?

  • so.....do you think you gain more DR with a patched camera over a non patched camera?

  • @vicharris yeah, you're right lol

    @burnetrhoades anything else you need to say, just PM it

  • @driftwood just curious, and I'm sure this has been answered 500 times, but is 244mb/s the MAX for a gh2, or has it been pushed further and it just isn't safe, or 95mb/s cards can't properly handle anything over that?

  • @driftwood "We see with our eyes......A lot of people prefer smooth and Nostalgic. Im not saying anyone is wrong or right." -Amen! back to topic:"chicken soup!" ; ) Nick, is there any change of movement handling from Moon trial 3 and trial 5? And you suggest using 24L in 80% is this for stability purposes or does it simply work better with the L setting? I just tried it with the ex tele/80%/24H on a 32/45 san disk and it worked without issues at least with 160iso.

  • Based on a couple threads over at EOSHD I ran an ISO noise test on Moon trial 3 and found some surprising results. I'd seen previous tests where other patches had an ISO320 that was not very good at all and clearly inferior to ISO400. This patch brings ISO320 back into the fold and ISO1250 is as good or better than ISO800 but ISO640 looks to be for all practical purposes just as good as (if not better than!) ISO160 and ISO200.

    FULL COLOR http://i49.tinypic.com/2wc1dp5.png

    RED ONLY http://i48.tinypic.com/ju7fh1.png

    GREEN ONLY http://i48.tinypic.com/14c78yv.png

    BLUE ONLY http://i50.tinypic.com/2uzdk5d.png

    LUMA ONLY http://i47.tinypic.com/2554zlw.png

    ...to create these I added an 'Exposure' node in CS6-AE. Shooting profile was 'Smooth' and -2,-2,0,-2

  • Did some low-light playing around with 1 camera on intravenus V1 and 1 on moon TV5, as i did not see difference in the lower iso's, 1250 and 1600 was indeed a tad better in moon than IV, 3200 seemed the same. when i have some time if will turn fw's around to see if it's the patch or camera sample.

    @BurnetRhoades any clue to why iso1000 stands out so much ? looks like a diff exposure than the rest.

    Also a note and some advice to people on gh2 and higher iso: Any higher than 3200 and i get bands in the footage with both patches, so not sure about noise > 3200, but that's more of a GH2 issue in general as different GH2's show different banding issues with same patch or unpatched, for anyone wondering why the stock upper limit is 3200, it's there for a reason.... use extreme iso for experiments only.

This topic is closed.
← All Discussions
Start New Topic

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In with Google Sign In with OpenID

Sign In Register as New User

Tags in Topic

Top Posters