Thread: RED Dynamic Range & Latitude Above and Below 18% Gray

Reply to Thread
Page 3 of 4 FirstFirst 1234 LastLast
Results 21 to 30 of 31
  1. #21  
    Senior Member Karim D. Ghantous's Avatar
    Join Date
    Oct 2011
    Location
    Melbourne AU
    Posts
    1,708
    I think it's worth pointing out that Monstro has more DR than film. I don't know of any other sensor with that feature. Film still has an edge as far as highlights go, which is not surprising.
    Good production values may not be noticed. Bad production values will be.
    Pinterest
    | Flickr | Instagram | Martini Ultra (blog)
    Reply With Quote  
     

  2. #22  
    Senior Member
    Join Date
    Dec 2010
    Location
    Toronto & Vancouver
    Posts
    3,743
    Quote Originally Posted by Phil Holland View Post
    Most that hire me to do these tests, including other camera companies, would likely say I don't do confusing or inflated, I am literally showing the data as I have been during the 2000s and 2010s. I'll take the criticism, but the point of making these graphics is to simplify really. I'm literally putting numbers on it so you can count samples, but if I just did the Photoshop gradient and slapped a couple numbers on there it looks like people would be content too.
    Obviously I appreciate what’s been done here, your in-depth results/analysis, and your explanations (no such thing as an “over-explanation”). But I’m also reading all of it to improve my understanding (and to compare to what I see in my own RED images) to help insure I’m getting the best results.

    My concern is that most camera agnostic shooters probably aren’t reading all of the details in RU threads. They’re looking at the chart and thinking, “Okay ISO3200 is even over/under!” (which is what these charts are intended for). So if/when the typical results don’t match the chart (e.g. someone shooting Weapon 6k/Dragon-X with STH under ISOcal1 at ISO3200, which would be unusably noisy, a stop darker, and closer to -8/+6 rather than even -7/+7), they’re understandably gonna scoff at these types of RED charts/paper claims (which could possibly be why RED doesn’t put out official ones — too many variables.)

    Hell, it’s probably why the rest use photochop gradients... It’s much more pliable/accurate as a simple/quick reference if there’s a gradient chart that’s “-7+/+7+ at ISOcal2 3200 for all DSMC2*” (*noise/additional DR dependent on OLPF and Sensor combos). Can’t fault numbers that conservative (safe bet) and it’s encompassing of all DSMC2 cameras (the actual results will likely be better).

    Again, my 2cents.
    Reply With Quote  
     

  3. #23  
    Moderator Phil Holland's Avatar
    Join Date
    Apr 2007
    Location
    Los Angeles
    Posts
    11,062
    Quote Originally Posted by Karim D. Ghantous View Post
    I think it's worth pointing out that Monstro has more DR than film. I don't know of any other sensor with that feature. Film still has an edge as far as highlights go, which is not surprising.
    Film = 13 stops according to Kodak.

    If you work with it flawlessly you can get 15 stops out of it on Vision 3, lower speed stock in particular. It's a "burning thing" and how light penetrates the medium.

    I've said it once or twice over the last several years. We've passed film on a technical merit for a while now. And whether you subscribe to my thoughts or Yedlin's regarding the whole thing via emulation, there's nearly no real reason to work with film outside of the love of the medium and it's aesthetic, which is a valid reason to honestly work with anything and I personally still love film deeply. Those who are getting the most out of film these days are even exploring larger format for the aesthetic tied to that as well as the increased resolution and cleaner image when pushed through post which helps it compete against modern premium digital cinema cameras. That's where much of my OG testing on Dragon was focused back on regular DSMC.

    Remember I didn't have the same to say back in the days of MX however. Things have evolved a lot since then.

    Here's something circa 2017 that I never posted here via Dragon 8K VV, back with my older skinny logo no less.



    A lot of testing with actual film and large format film was done between 2015-2017 in my world for studios and filmmakers, particularly IMAX peeps late 2017/2018. But I think this particular exploration is likely "done" at this point. Everybody sort of gets it now.

    If you are looking for a more film like image without emulation, give ISO 1280-1600 a whirl for a while. Even 3200 for the brave.

    Somewhere between here and there people either prefer cleaner images or don't mind the visual aesthetic of image noise/grain/texture when it's not "too busy". Many these days actually do a grain pass or even a Noise Reduction pass prior to adding grain. Whether they do the rest of the emulation is likely up to their workflow. i.e. Gate Weave, Base Bias, Halation, Print Emulation, etc.
    Phil Holland - Cinematographer - Los Angeles
    ________________________________
    phfx.com IMDB
    PHFX | tools

    2X RED Monstro 8K VV Bodies and a lot of things to use with them.

    Data Sheets and Notes:
    Red Weapon/DSMC2
    Red Dragon
    Reply With Quote  
     

  4. #24  
    Senior Member Aaron Lochert's Avatar
    Join Date
    Oct 2013
    Location
    Tucson, AZ
    Posts
    1,015
    Quote Originally Posted by Phil Holland View Post
    Here's something circa 2017 that I never posted here via Dragon 8K VV, back with my older skinny logo no less.
    Is that print emulation for philmColor or was that a separate project?
    Reply With Quote  
     

  5. #25  
    Moderator Phil Holland's Avatar
    Join Date
    Apr 2007
    Location
    Los Angeles
    Posts
    11,062
    Quote Originally Posted by Aaron Lochert View Post
    Is that print emulation for philmColor or was that a separate project?
    You've got some goodies in there related this. I can't do full grain emulation, gate weave, proper halation simulation, and other things via LUTs. But look-wise, yes, you've got the jazz.

    Shortly you'll have even more jazz.
    Phil Holland - Cinematographer - Los Angeles
    ________________________________
    phfx.com IMDB
    PHFX | tools

    2X RED Monstro 8K VV Bodies and a lot of things to use with them.

    Data Sheets and Notes:
    Red Weapon/DSMC2
    Red Dragon
    Reply With Quote  
     

  6. #26  
    Senior Member
    Join Date
    Apr 2007
    Location
    New York, NY
    Posts
    1,013
    Ever run those Xylas through Imatest? I'd be curious about the quantitative analysis for SNR=1 and SNR=2 etc... The 16.5/17stops always seems a bit meaningless, as do all RED's DR specs since they say Gemini/Helium/Dragon(incl. DSMC1) all have 16.5+ and Monstro 17+ and anyone who uses these cameras can see that makes no sense that they are somehow all the same. I assume this is just how many chips are at all visible, with no consideration of how incredibly noisy they are?

    Still cannot figure out why the middle grey chip would be 1 and not 0, as a quick view of the chart seems to say Monstro has 7 stops over at ISO800 which is known not to be true, right?
    Noah Yuan-Vogel | noahyv.com
    Reply With Quote  
     

  7. #27  
    Senior Member
    Join Date
    Apr 2007
    Location
    New York, NY
    Posts
    1,013
    Quote Originally Posted by Karim D. Ghantous View Post
    I think it's worth pointing out that Monstro has more DR than film. I don't know of any other sensor with that feature. Film still has an edge as far as highlights go, which is not surprising.
    You don't? And aren't highlights part of DR? What about Arri? They still have more DR than Monstro, especially the LF which appears to have ~1.5stops more than Monstro. (https://agdok.de/de_DE/kameratest2019)
    Noah Yuan-Vogel | noahyv.com
    Reply With Quote  
     

  8. #28  
    Senior Member Karim D. Ghantous's Avatar
    Join Date
    Oct 2011
    Location
    Melbourne AU
    Posts
    1,708
    Quote Originally Posted by Noah Yuan-Vogel View Post
    You don't? And aren't highlights part of DR? What about Arri? They still have more DR than Monstro, especially the LF which appears to have ~1.5stops more than Monstro. (https://agdok.de/de_DE/kameratest2019)
    The LF doesn't have 1.5 stops over the Monstro - not at the Monstro's (nominally) native ISO. (The A7S did better than I expected, which means that the new model could be amazing). And I think you know what I'm talking about regarding highlights. It's not the same thing as DR in the real world. Neither the ARRI nor the RED can capture light sources properly, particularly at night. Even Super 8 can do that.

    Quote Originally Posted by Phil Holland View Post
    I've said it once or twice over the last several years. We've passed film on a technical merit for a while now.
    Not in every way. When that happens, we'll know. ;-) BTW the film emulation is quite convincing. I think you overdid it with the purple fringing though! It might not match film perfectly, but then again, why not let digital be its own thing. E.g. gate weave should be eliminated from film in the DI, so I don't want it introduced in digital. Same with distortion.

    Perhaps this is a discussion for another time, but some people have modern thinking about film (as a deliberate choice, not merely a reluctance to change), while others have old-fashioned thinking about digital (big cameras - including DSLRs, a strange obsession with large sensors, a false belief that sensors have fixed colour signatures). Strange world.
    Good production values may not be noticed. Bad production values will be.
    Pinterest
    | Flickr | Instagram | Martini Ultra (blog)
    Reply With Quote  
     

  9. #29  
    Moderator Phil Holland's Avatar
    Join Date
    Apr 2007
    Location
    Los Angeles
    Posts
    11,062
    Quote Originally Posted by Karim D. Ghantous View Post
    Not in every way. When that happens, we'll know. ;-) BTW the film emulation is quite convincing. I think you overdid it with the purple fringing though! It might not match film perfectly, but then again, why not let digital be its own thing. E.g. gate weave should be eliminated from film in the DI, so I don't want it introduced in digital. Same with distortion.
    I'm happy about letting film be. When we started doing film emulation in the early 2000s it was to match digital to film best as possible. This was certainly expanded upon as film VFX counts rose as well as many different cameras being used often on larger budget productions with film often being the hero medium to match to.

    Eventually digital became the primary medium industry wide, sort of 2010s through now thing, and the expanded conversation I have mostly with DPs, Directors, Colorists, Companies, Studios, the Mirror, the Plants Outside is what exactly should a motion picture look like. Many still are after the film look, which is why so many different paths to get there exist. Some are also very happen to explore new visual horizons and see where things land on a project to project basis. I fall into every side of that conversation because I have clients who want the cleanest image possible all the way through "I need this to look like film". No one way, which is certainly fine by me.

    In terms of the Halation model used in the emulation, that is the "medium" strength. That particular algorithm has none, weak, medium, strong, and overt settings. In the case of the Mamiya, it does indeed have CA which can be seen in the source image, which wasn't removed in this case but rather fed upon by the Halation model itself where appropriate. The thing that sold this model was actually the "sparkle" found on the rims of the car's front wheel, which I'm still pretty proud about when you see it projected. Strangely that sparkle note I got from the big D led to some side work on a glisten effect, but that's more of a VFX thing really. Luckily with this emulation you can literally dial everything in or toggle it on/off. Also things like grain and halation are tied to format size to be accurate. Didn't make anything other than a vertical gate weave model, but horizontal is certainly a thing, but just not what people think about commonly when they feel it outside of a 70mm projection which runs sideways.

    http://www.phfx.com/temp/phfx_REDDra...ionExample.jpg

    For me personally, color and tonal response as well as potential lab processes are the most interesting thing about the film look. I don't care to add gate weave too often. Grain, pretty much never these days if I don't have to. I don't remove it much either. Some of the other features though I find are cool and fun to explore, but still rare for me to finish out a project that way if it's not required for some reason.
    Phil Holland - Cinematographer - Los Angeles
    ________________________________
    phfx.com IMDB
    PHFX | tools

    2X RED Monstro 8K VV Bodies and a lot of things to use with them.

    Data Sheets and Notes:
    Red Weapon/DSMC2
    Red Dragon
    Reply With Quote  
     

  10. #30  
    Quote Originally Posted by Noah Yuan-Vogel View Post
    Ever run those Xylas through Imatest? I'd be curious about the quantitative analysis for SNR=1 and SNR=2 etc... The 16.5/17stops always seems a bit meaningless, as do all RED's DR specs since they say Gemini/Helium/Dragon(incl. DSMC1) all have 16.5+ and Monstro 17+ and anyone who uses these cameras can see that makes no sense that they are somehow all the same. I assume this is just how many chips are at all visible, with no consideration of how incredibly noisy they are?

    Still cannot figure out why the middle grey chip would be 1 and not 0, as a quick view of the chart seems to say Monstro has 7 stops over at ISO800 which is known not to be true, right?
    If you look about halfway down this page Luigi did a test using the Imatest.
    http://www.reduser.net/forum/showthr...17#post1839117
    Reply With Quote  
     

Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts