First installment of SCCE shoot out to be released today...

tom.wong said:
i've seen first hand how the highlights stay white on a ki mini off a f3, while it goes that mucky yellow/green on the xdcam.
Click to expand...

You are the first person I've ever heard see the same thing I see on XDCAMs, and not just the F3.

I thought it was the camera head though, not the codec. Never thought to compare SDI out vs XDCAM codec for color!

When I've been telling people I mainly want S-Log because it does away with the "XDCAM look", they usually look at me like I'm nuts.
 
I've noticed that mucky yellow/green too. I saw some profiles from Allistair that addressed the colors, but I didn't like the other tweaks.
Just shot a bunch of stuff with XDCam and KiPro. I'm gonna compare the two for color. Generally, they look close to the same to me unless I start really yanking them around in grading.
 
Barry_Green said:
Don't know why you'd say that. I was organizing six-way shootouts back in 2006, years before the DSLRs came out....
Click to expand...


By these I meant the Zacuto / SCCE shoot out the one the OP is talking about with DSLRs and pro cameras... not all the shootouts of the world :)
 
Last edited:
Thats because i didnt hit 5+ stops over.... ;)

I used ya know..... Lights, to make the image.
 
hmm well very interested in knowing what every camera was recorded onto and what bitrate, codec, and more was used... is this info anywhere?
 
Yeah once again I'm with Giuseppe. Its kind of pointless if they don't tell us what the camera settings were. How can you evaluate overexposure without knowing if you're comparing say Slog to a 709 gamma with who knows what if any knee? Is this info somewhere? All in all though very interesting , informative and a great video by Zacuto
 
Last edited:
Lenilenapi said:
Yeah once again I'm with Giuseppe. Its kind of pointless if they don't tell us what the camera settings were. How can you evaluate overexposure without knowing if you're comparing say Slog to a 709 gamma with who knows what if any knee? Is this info somewhere?
Click to expand...

It's not a secret. I'm fairly certain a quick phone call can lead you to all the information you desire. It's a test afterall... the information was recorded and shall be given. Don't think they will release/sell the video and not tell you the control used in the test. :)
 
If its one thing I've learned over the past 3 meetings with Sony, is that gamma curves, knee levels, log space, can hugely effect what was being shown, and greatly cause lots of issues in post if not worked with correctly. I'm not trying to argue if it was done right or not. Everyone looks like they did a great job. But without this information right up front, it kinda doesn't help me evaluate anything. a 709 is completely different from a C-Log or S-log. Also for the RED MX which doesnt have anything baked it at all... what did they do for that?

Just a lot of unanswered questions that would really put into perspective what we are seeing. Without that information you really cant take any of the images of a digital camera and evaluate anything. Also would love to see the info on the IE rating's and how they came up with meatering at 800, considering all those digital cameras have pretty much been proven are NOT actually 800, even though the camera says it is. So that could make the difference between something clipping and something not, even a 1/3 of a stop can make a difference between seeing a line, and not seeing it.

I'm not trying to be the grumpy old man, I'm just saying its kinda silly to me to post such a vast comparison with no data what-so-ever of what we are seeing. Those are the things that actually effect the image the most. Without knowing the color space thats being worked in between each camera, how could you evaluate anything? Its pure science, not a matter of opinion.

as far as clipping is concerned, without that information right up front, the test really doesn't tell me much other than the expensive cameras handled the way you'd think they should, and the cheaper ones blew out sooner. But without knowing what the settings where, it very much could have been a different ball game with each camera.

compression rates, bit rates, codecs, gammas, without knowing any of this, its just footage and not useful as an evaluation. If they were all REC709 then tell us, if they were all tweaked to have a *cine* gamma curve that was similar to each other, then tell us...

Again I really respect the amount of time an effort it takes for all these people to come together and do a test such as this, but wihtout the fundamentals explained and this being webcasted all over the world, without such explanations only feeds more into misunderstandings of what really effects these images.
 
Last edited:
and before I get attacked about the whole SCCE thing.. this is directed towards Zacuto's putting these videos online for the world to see, not the SCCE. Im sure the SCCE has logged every bit of information they possibly could with these cameras. But I say shame on zacuto for not putting this information in the videos or at least on the web text. Its so hugely important to know these values before you can evaluate any kind of footage you're seeing. Leaving that data out really negates the whole process that the SCCE did to create a quality and equal test.
 
RED Just looks like RedColor and RedGamma with proper white balance attached. It's behaving near-to what I would expect it to straight out of camera if these are the settings, balanced for a WYSIWYG method of working.

If it comes down to settings, then the RED MX is the one that loses out the most, being that the brunt of its magic comes out in post.

On the yellow cast: I see it ALL the time in a lot of cameras. AF-100, F3, and especially the FS100. Can't stand it. I also noticed that the F3 seems to have a reddish cast and the Af100 leans toward green. Both in this video and in my limited experience with both real world. Sure you can tune it properly.

Awesome little shootout! No discrepancies from me. Pretty much shows exactly what I thought it would...
...And, I wonder how many people still think you can't tell the difference on the internet.
 
Barry_Green said:
They did. The same external recorder was used for all cameras.
Click to expand...

where is this information? also how is it possible that the same external recorder was used for all cameras, and if so what does that say of RED's footage, if it was recorded externally instead of internally thats a huge crutch to the MX then....

I saw you were chewing me out for "not knowing for sure" if zacuto had anything to do with the cameras. THEY said Bob picked the cameras, but how do WE know that Bob wasn't politely asked by someone to add the DSLR's ??? my point is NO one knows. So please dont chew me out saying I dont know, when clearly the only conclusion then could be no one knows for sure but the man himself. I always try to be civil on here, but I like to not budge when I see something amiss.

I question zacuto's coverage because I see missing information and inconstancy with how they do their visual analysis.... Just like watching the news, everyone puts their own twist on the truth... Reporting it well is what makes the credible from the "FOX" channel.

I bring out points in a way that I hope brings to peoples attention something that they didnt know was a big deal. But I also question why things are done in the manor that they are done. If no one questions, what happens? we all become 'yes men'?

What you see is not really what you get in the digital world. If you dont know what you're doing, what you see is not at all what you get. Knowing where a white point is set, knowing what codec does what to an image, knowing the gamma curve's point at 800% or 1000% or any of the numbers you wanna run at... my point is this is way more important than just looking at the footage.

I met with 2 engineers from Sony, one who worked very closely with the F35. He was explaining that a lot of the film guys who are now working digital still get a ton of things wrong. Very well respected DPs and colorists when thrown into the digital world, are still making very large mistakes when working with the materials they are getting. There were some great stories about the TV show Justified and how they worked with the F35 and for the first week were getting horrible results, even though it seemed they were doing everything right. It takes one person to miss one calculation to ruin or preserve the digital image.

uncompressed vs. 10:1 or anything else, it all effects what you're seeing depending on what you do with it. Gamma curves greatly effect the clipping point. I just dont see Zacuto talking about that EVER in any of their tests. I'm not trying to attack them, I'm trying to bring something to light. The lack of information is what creates the problems. Understanding if something was brought back into a 709 space or left in a cine gamma state, all of this is not talked about at all, yet to evaluate you must understand what happens.

A lot of people I talk to dont even know what a gamma curve was even doing, or why it was invented in the first place. I'd just like to see the hard work of the people doing this test, to show they werent just 700 people involved in a test with no hard data. I know its there and I'd like for Zacuto bring to the public the huge importance of this info.

So thats why I question them, and I question their motives in the documentary of whats shown.
 
TimurCivan said:
Thats because i didnt hit 5+ stops over.... ;)

I used ya know..... Lights, to make the image.
Click to expand...


compare it with the xdcam on board, everything is yellow/green. :P lights or not, i'm claiming the xdcam does it.

however upon further review, extreme clip points start gaining the yellow, even on the pro res, but it doesn't start happening until it gets insane. and many times less severe than the xdcam. skin tones stay spot on too.
 
Kholi said:
RED Just looks like RedColor and RedGamma with proper white balance attached. It's behaving near-to what I would expect it to straight out of camera if these are the settings, balanced for a WYSIWYG method of working.

If it comes down to settings, then the RED MX is the one that loses out the most, being that the brunt of its magic comes out in post.

On the yellow cast: I see it ALL the time in a lot of cameras. AF-100, F3, and especially the FS100. Can't stand it. I also noticed that the F3 seems to have a reddish cast and the Af100 leans toward green. Both in this video and in my limited experience with both real world. Sure you can tune it properly.


Awesome little shootout! No discrepancies from me. Pretty much shows exactly what I thought it would...
...And, I wonder how many people still think you can't tell the difference on the internet.
Click to expand...



i'm pretty sure they were using redgamma2 and redlog film to develop the image. Jim Jannard himself specified to SCCE to do this. Alexa and F35 were all log developed, the RED MX got the same chance with their log curve. either way, i think the comparison was pretty awesome. I really love the station setups, and the mix of the charts. It may not have been perfect, but I really admire the time, effort, and thought put into it. Just shows what a great community this is when it comes to learning and teaching.
 
You must log in or register to reply here.
Back
Top