Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

CML Camera Evaluation 2015

David. This is a a no-win situation, and we can adjust to correct the view of interest, if we so wish (there are enough exposures)

It is good advice to rate the SKIN-OLPF lower than the LowLight, as they basically share the same range.

And we change the OLPF to change the Characteristics.


It is a good thing to have in mind when looking at the test, but doesn't make it less relevant. (I'd have been of the same opinion if he had exposed both for 800, and the critic was that that didn't adjust for the light transmission difference...)

It's a "choose your position" situation, and when goldielock/kippertie/underwater OLPF's comes in, we'll have even more POV's.

Should we test them with the same exposure, and thus not get an opinion on their wekness/strength "out the gate", or should we expose them as different cams.

I see good logics for the 2nd approach, as we will usually not use them the same way.




:)
 
I see all sides on this. I'm not arguing the test is flawed, just searching for clarity on the differences in lighting and exposure because those slight differences make apples to apples comparisons not possible.

As a small example, you and I show up to a set and you have the LLO and I have the STH. We aren't going to light the set differently for each OLPF, so it would be nice to see them in the same light.

I don't think Geoff did a bad test, the files are very helpful. It just makes certain types of comparisons not possible, like how it is the LLO vs STH in daylight in the same conditions.

The sensor is excellent, the DR is there through a range of ISO ratings with either OLPF.

Battistella
 
If you want to see how they perform under the same light conditions, change the comparison parameter by one stop. That is what I like about this, we have both datasets as a postulate of the "from over to under" approach to all cameras.
 
David and Gunleik, if a skilled DP, not particularly familiar with the red eco system read your lasts posts he will say, interesting stuff but nothing that will help me on set making better pictures. Geoff's test shows that the Epic Dragon exists now in 2 flavours. What irritates me more - not really because I understand the cause - is the interesting finding in post #40 that the STH OLPF will create less DR on the highlight side which is conflictive to the intended purpose.

This is what I meant in post #47. We are only seeing the first snow flake of an breaking avalanche of Dragon/OLPF/noise/ISO/RAW/DragonColor1&2/RedGamma4/RedLogFilm/5KHD/6K/DEB/its all Metadata confusion hitting the forums here.

---

Ok. Personally I like the idea of changeable OLPFs, from time to time I like tinkering with my dragonized Epic. But I've got the feeling that the LL OLPF exists only because of a miscarried marketing claim saying that IOS 2000 is easily achievable with the new dragon sensor and a clientele that wants to see in the dark and takes this advertisement literally. Anyway, it is like it is.

However, the CML tests shows that the Epic Dragon holds up very well well in the surrounding camera market, something that matches petty much my own experiences, at least with the Alexa.

Hans
 
I see all sides of this and the potential confusion it could create. Even with 10 different film stocks you had similar discussions and confusion popping up all over the place. Tests had to be done, etc and that was kinda expected. Now, folks want answers but don't really want to do the testing. They want to be told what is the best way to go. But take a DP like Christopher Probst, who posts here a fair bit. He doesn't go out the door without knowing exactly what the machine is going to do or how it is going to perform. I appreciate that kind of work ethic and commitment to craft.

I get that not everyone can do that because of schedules and budgets, etc.

Personally, I get pretty confused about all the various models of ALEXA camera right now. Last time I checked there were like five of them and more different models on the way. For me, that could create a similar kind of confusion if I was going to and Alexa job tomorrow.

It's interesting to say that the tools are just tools, but it's not entirely true. As you become more familiar and intimate with the tools, how they work for you changes. If you looked at the hockey sticks of a team of hockey players, they might be the same brand, but you could see the personality and individuality of each person in that "tool". It's up to each person to see where it works best for them.

Relating that back to this discussion. There might be many people who say, "give me any stick and I can play hockey with it." but will they be at their best?

Battistella
 
David,

Love the hockey analogy. FYI...I'm using Bauer sticks with wooden blades with a pretty flat curve for my backhand shot. In other words, I have currently set up my Epic Dragon with STH. :)
 
Justin Carrig said:
STH Dragon daylight at +4 - Her face is clipping
RED_1.2.1_result.jpg


DIT: It seems the difference in highlight performance seen here is actually result of different ISO/Light levels used between them and the consequences of that (STH Daylight had a stop more light being set at 400 ISO, and so clipped the sensor at +4).

So the results here are not so much showing a difference between OLPF performance.

This could be seen as a demonstration of the effect of adjusting and exposing for ISO and the effect that has on shifting the dynamic range around: You can work at higher ISO's for better highlight performance, and work at lower ISO's and add light for better shadows/noise performance.
Click to expand...


The different ISO, and the compensated light level in the daylight tests, simply reflects RED's reccomended ISO with the relevant OLPF's.
 
Brian Boyer said:
Okay...after further inspection, what I thought was the case...isn't. It all comes down to the labeling for the ISO 400 tungsten shots versus the ISO 800 ones.

Given people's tendency to read the directions last, I think the assumption that all "Stop Whatevers" are created equal is going to cause a bit of confusion. Perhaps the discrepancy should be mentioned more prominently.
Click to expand...


Guys, I wrote very clearly that the tungsten and daylight tests were different, for simple reasons we couldn't increase the tungsten light beyond what we had so I thought I made it very clear by showing the first clip in the series "correctly" exposed and then running the series as usual. I made this clear in the notes. ALL 400 ISO cmaeras were treated this way.

If people can't manage to amke the correlation between the 2 then maybe they should be looking for a job elsewhere?
 
Jake Wilganowski said:
Clear as mud
Click to expand...


I'll try very simply.

RED, you know the company that makes the cameras, told us to rate the LL at 800 ISO and the Skin at 400 ISO. In reality this i,as has been said, a metadata change.

We tried to expose the cameras "correctly" for their rating, ie 400 ISO lit to 100FC and 800 ISO lit to 50FC for a T5.6 exposure.

We were not able to do this for the tungsten and noted this clearly so people could make a reasonable judgement from the data provided.

I have no idea what the metadata shows, I'm telling you what we exposed for, if the metadata shows something else then somebody screwed up but that screwup has no effect whatsoever on the actual results.

I wonder why I do this at times...
 
I'm amazed that all you're talking about is DR what about the significantly different colour response?

That's what the new chart was for!

What about the huge difference in some cameras between tungsten and daylight colour response!
 
Geoff it's all appreciated and not a stab at you.
It's just (even though you have made things clear) just a discussion on how to make the examples as widely useable as possible.

And I agree. The devil is in the colors. Not in the DR.

Thanks again for sharing this and also updating things in this thread.
It is only appreciated.

G
 
Geoff,

I certainly appreciate the tests and I'm better understanding the method as explained by you it makes perfect sense. I had not read the notes as was basing things on what I was reading in this thread.

I think I pointed out quite early the vast differences in color response.

Thanks for the work you did in this test. It's always appreciated. I had not appreciated that you were exposing the STH at 400 based on reds recommendation.

Battistella
 
Last edited:
geoffboyle said:
Guys, I wrote very clearly that the tungsten and daylight tests were different, for simple reasons we couldn't increase the tungsten light beyond what we had so I thought I made it very clear by showing the first clip in the series "correctly" exposed and then running the series as usual. I made this clear in the notes. ALL 400 ISO cmaeras were treated this way.

If people can't manage to amke the correlation between the 2 then maybe they should be looking for a job elsewhere?
Click to expand...

Come on, Geoff. I understand you went through a lot of effort and invested a good portion of your time and expertise to make this test happen. However, none of that necessarily means you shouldn't at least entertain the thought that your writeup couldn't benefit from some further editing for clarity.

And speaking of jobs...you should understand the role of editing in this business.

Obviously, the explanation is clear to you and I'll admit that, on the whole, it's clear enough to piece things together correctly. I also think it could be made clearer for those of us who haven't yet realized we're in the wrong line of work. For example:


"Light levels in the daylight setting were adjusted to optimise exposure for 400 or 800 in daylight, in tungsten everything was lit for 800 and that's why some frames indicate +1 stop as the iris was adjusted to compensate for the difference."


Not only is that a run-on sentence, it seems to contradict the following run-on sentence, which is found on a different page:


"Exposed at T5.6

On the tungsten day we kept this level for all cameras, this meant that the 0-Stop reading on the 400 ISO cameras is actually -1-Stop"

This is important and probably should've also been mentioned on the first page where people can start downloading images.


The highlighted portion of the first quote, without the benefit of the second quote, could be misconstrued to mean you compensated for the 1 stop discrepancy by opening up the iris on the ISO 400 cameras and calling that "Stop 0."

Since that wasn't what you did and the iris was kept at T5.6 for everything that day, I don't see where you actually compensated for the specific ISO 800 versus ISO 400 difference with the iris (as opposed to the opening and closing of the iris to achieve the various exposures for all cameras), like the first quote implies you did.

Maybe I'm misreading it but I've been racking my brain and I can't figure out what else it could mean.

It seems the only compensation for the difference was made after the fact on the website via the placement of the reference frames in conjunction with a somewhat disjointed (IMHO) explanation in the notes.


Please don't feel you have to explain the process yet again. I get it and explained it to someone else in an earlier post. What I don't understand is what you meant by the portion I highlighted in the first quote from the website. It's not really that big a deal for me.

I'm merely pointing it out to suggest your explanation may not be as clear as you think it is and you needn't get upset when people say they don't understand what you wrote.

Behind many great works of literature there was an editor/proofreader/trusted resource who correctly told the author something needed changing. Try not to take it so personally.
 
Last edited:
Sidney L. Plaut said:
i have a deep respect for Geoff who goes through this every year, fully knowing that in the end, he will always come out as the bad guy:)
Click to expand...

I'm used to being the bad guy!

At a film festival recently where my most recent film was shown in a Q&A with the festival organiser I was asked why my film looked like a movie when another project shot in the same locations at the same time with the same kit looked like a TV soap.

I replied "because I was a complete pain in the arse for the entire shoot and post" bear in mind that the audience included the cast and crew...

But it was an accurate reply, try and do things right an you are bound to upset a lot of people.


The tests have been done the same way publicly for around 16 years now and privately for a further 25 or so, they show me what I need to know, if they're useful to other people then great.
If they can't read them or disagree with them then they can do their own and publish them for our comment, oh, hang on, nobody else does this. Every other "test" that I see isn't about finding out what the cameras inherent characteristics are, they're all about proving something.


Am I clear in the way I describe what we do? probably not but anyone who knows their craft should be able to use them.
 
I'm amazed that all you're talking about is DR what about the significantly different colour response?
Click to expand...

+1

..this is very true but colour does often become more subjective for people. DR feels to most to be more black an white;-) ... Not since people played with reversal (positive) stock has anyone embraced less dynamic range.
 
Michael Lindsay said:
+1

..this is very true but colour does often become more subjective for people. DR feels to most to be more black an white;-) ... Not since people played with reversal (positive) stock has anyone embraced less dynamic range.
Click to expand...

I think colors/fidelity might have become more important when DR approaches "ok/good".

To me it is all more and more about fidelity these days.
 
Gunleik Groven said:
I think colors/fidelity might have become more important when DR approaches "ok/good".

To me it is all more and more about fidelity these days.
Click to expand...

Absolutely right. But then the interesting discussion starts: what is good fidelity, what is not so good fidelity?

I know quite a few who think that the old Panasonic Varicam was the benchmark for colours in video-land back then. Many think that the new Varicam does a lot right in this regard. Personally I found the colour rendition of Panasonic's old Varicam horrible, and thus I'm not fond of testing the new one. I also was never happy with the colour rendition of Sony cameras. I once had F5 footage in my studio and remember well saying me, oh Sony! Vibrant, somehow right, boring, and not particular nice in the highlights. Nothing that drives a great creative grading session. Arri footage is much more inspirational for me, for instance. And the Dragon is also a sensor that makes great pictures possible.

I found then and today that the fidelity that comes from well exposed film stock is hardly to beat, especially if we talk about skin tones. In early 2008 I shot my first project digital with the RedOne. It was a commercial for cinematic release. I was appalled how difficult it was to create acceptable skin tones. But digital cameras developed and now 7 years later we are in a stage that I can say, good by film - without scarifying something that I will miss.

---

If we talk about fidelity I think we are talking about personality. The usual suspects deliver tons of usable fidelity today. Does the digital neg. contain personality or is ist a well engineered picture of the environment? I'm not looking for the "right" colours, I'm searching for the interesting look that matches the intention of the photography. If you look at macbeth charts that were shot with film and compare them to video shot colour charts, the film charts seem to be off. Personally I find the "offness" to be right. Probably a personal acclimatisation over the years. In the moment I have a talented intern under my wings. He hates any texture/grain/noise, for him this is flawed photography. Fidelity and quality is very subjective.

Hans
 
You must log in or register to reply here.
Back
Top