Very depressed about AF100 test...

The contrary is much more difficult or near impossible if the highlight are blown or shadows are already crushed. If you look at ungraded best DR camera like the Alexa the image are milky and very very flat.

The un-graded shots with the Alexa from the Zacuto tests had a huge dynamic range. They didn't look
milky at all. Unless you use a compression curve (like Log/SLog), not using the dynamic range
of a camera is simply throwing the bits away.
 
Last edited:
Just for correction, better dynamic range means flatter image and in this case your argument is not valid, pop, contrast is the contrary of better DR. I am not saying that this test is good or bad (They had a Panasonic guy at the test if I am not mistaken), but just saying that better DR is the exact opposite of what you said. In this case, with just a little curve adjustment in post the d7000 image would be as contrasty or pop as the Af100. The contrary is much more difficult or near impossible if the highlight are blown or shadows are already crushed. If you look at ungraded best DR camera like the Alexa the image are milky and very very flat.


"better dynamic range means flatter image" is an oversimplification. Any image can be made "flatter". Dynamic range describes the range of light intensity that a camera can record while still maintaining detail.

Dynamic range is limited by the darkest tone where texture can no longer be discerned and the highest full well saturation before highlights get blown out. A flat image can still have poor dynamic range; images that pop and have nice contrast can have high dynamic range if there is good depth and detail at all those brightness levels.

Here is an image with high dynamic range. It "pops", it has good contrast, and it's not flat:

6793531327_ef9a2bb026_o.jpg
 
"better dynamic range means flatter image" is an oversimplification. Any image can be made "flatter". Dynamic range describes the range of light intensity that a camera can record while still maintaining detail.

Dynamic range is limited by the darkest tone where texture can no longer be discerned and the highest full well saturation before highlights get blown out. A flat image can still have poor dynamic range; images that pop and have nice contrast can have high dynamic range if there is good depth and detail at all those brightness levels.

Here is an image with high dynamic range. It "pops", it has good contrast, and it's not flat:

I went into the oversimplification because I did not want to go into the read noise to full well capacity etc. I wanted to be a bit more clearer and not dwell into the overall technical details but more on this particular example where from the resized image the flatter D7000 image had better perceived DR (You can see more in the shadows of the faces and the highlights are not blowing as on the faces in the AF100 image). It is true that a flatter image does not necessarily show better DR if it does not have texture/detail in those highlights and shadows. The image above does say anything in the sense that by the direction of the light, it is very early morning in at least Autumn and even if the sun is out, its power can still be very low during this time.

A test of the Alexa log, Sony f3 slog and the Canon 60D. How many people would take the ungraded Alexa footage, washed-out etc ( if you did not tell them that it was the Alexa) but the amount of info is just phenomenal with so much DR. The Sony F3 is also very very nice. It is the interior shot that really show where DR is so much important. Where the Alexa and F3 still can expose the whole scene correctly, the 60D is completely blown for the outside shots.

 
when did this thread go off topic? i immediately noticed the AF100 'lips' separated on the three persons. it even looked better than the Alexa. More contrast yes but the color differentiation is there. Also Isn't 4:2:2 the limit for green screen? how does the D7000 manage in green screen?
 
Last edited:
I went into the oversimplification because I did not want to go into the read noise to full well capacity etc. I wanted to be a bit more clearer and not dwell into the overall technical details but more on this particular example where from the resized image the flatter D7000 image had better perceived DR (You can see more in the shadows of the faces and the highlights are not blowing as on the faces in the AF100 image). It is true that a flatter image does not necessarily show better DR if it does not have texture/detail in those highlights and shadows. The image above does say anything in the sense that by the direction of the light, it is very early morning in at least Autumn and even if the sun is out, its power can still be very low during this time.

A test of the Alexa log, Sony f3 slog and the Canon 60D. How many people would take the ungraded Alexa footage, washed-out etc ( if you did not tell them that it was the Alexa) but the amount of info is just phenomenal with so much DR. The Sony F3 is also very very nice. It is the interior shot that really show where DR is so much important. Where the Alexa and F3 still can expose the whole scene correctly, the 60D is completely blown for the outside shots.


I've seen some amazing stuff from the Alexa without using Slog .. so it's really a choice you make. In the case of these shots (except for
the dark ones), there's really no need to use Slog because the latitude of the shot is well within the latitude of the camera.

What's curious to me is that the Alexa with an Slog curve should have more latitude (after post) than the 7D does without compression.
Yet, when these were graded, the 7D appears to have more. That suggests to me that the colorist didn't know what
they were doing.

jeff
 
That suggests to me that the colorist didn't know what
they were doing.
Well you've just hit a big nail on the head. The trouble with powerful tools these days is that they often end up in the hands of rank neophytes that wanna be rock stars without having to learn any guitar chords at all- let's call them the Guitar Heroes of HD Video. Camera manufacturers have worked hard to 'can' profiles so that much of the learning curve in finesse is removed altogether but your point is clear. You can't substitute years of color grading, audio engineering, directing, lighting etc. with 'canned' presets.
I daily prove that experience trumps technological superiority and am constantly dismayed by the money that pours into the camera maker's wallets, even as film schools struggle to stay alive.
It must be our consumer mentality.
 
Hi Guys

I bought an AF101 in September and shot my first short with it in December and I have to say that I wasn't pleased with the results! The fault lies entirely with myself as I used used an uncalibrated monitor and subsequently found the white balance was way off. Couple that with the rustiness of not being behind a camera in over 3 years.

I also got a bit carried away in a couple of scenes with the shallow DOF from a 70mm lens so much so that the background is mush which my DSLR mates love but I hate.

I've run subsequent tests with the camera hooked up to a 50" tv and am more than impressed with the results and I'm using "old" Canon FD lenses.

I've been more than blown away with the quality acheived so far and am glad to say that I'm planning another short in March.

Regards

T
 
The trouble with powerful tools these days is that they often end up in the hands of rank neophytes that wanna be rock stars without having to learn any guitar chords at all- let's call them the Guitar Heroes of HD Video.

Totally stealing this analogy.

I daily prove that experience trumps technological superiority and am constantly dismayed by the money that pours into the camera maker's wallets.

Technology is changing so fast, and cameras are continually evolving. It's not like the money's involuntarily leaving your pocket to go into that of the camera makers.
 
There are a couple of different worlds at play here. There is the world where footage will be graded after it is cut and the world where the editor will take what they are given and cut with an eye towards minimal color correction.

Slog will almost always win in the former world as it does have the greater dynamic range, but it REQUIRES a grade. Scenes shot with a look baked in are can win in the latter world if shot properly.

You have to know the project, the budget and the client needs in each case. Handing over Slog footage to some one not prepared for it might be a technical win for image quality ultimately but be the completely wrong thing to do for the project and you client and your relationship with your client.
 
True, but redundant since our man doesn't seem to want to engage in the application of scene files.


Wait, when did the OP say he was opposed to using scene files? All he said was that he was depressed that the D7000 screen shot looked better to him in this one test. I think we've helped him & should give him the benefit of the doubt.
 
Gary. steal that analogy man!
The way that some people talk ( no proof here ) you'd think that these forums were like a book of the month clubs.
Some day I'd like to hear the story about that person that's run their 5 year old camera into the ground- winning awards and and packing the bank with cash along the way.
Well, I'm still trying to kill my crappy gear so I'm about 2 years away from that point and sadly... no awards yet :)

Totally stealing this analogy.



Technology is changing so fast, and cameras are continually evolving. It's not like the money's involuntarily leaving your pocket to go into that of the camera makers.
 
Excellently put. Fundamentally there are three scenarios:
1) breaking news (for this you need 'baked in' presets- no grading possible)
2) Features with requisite lead time and an experienced editor/grader waiting (Slog)
3) Everything else including Indie productions. (take your pick)

There are a couple of different worlds at play here. There is the world where footage will be graded after it is cut and the world where the editor will take what they are given and cut with an eye towards minimal color correction.

Slog will almost always win in the former world as it does have the greater dynamic range, but it REQUIRES a grade. Scenes shot with a look baked in are can win in the latter world if shot properly.

You have to know the project, the budget and the client needs in each case. Handing over Slog footage to some one not prepared for it might be a technical win for image quality ultimately but be the completely wrong thing to do for the project and you client and your relationship with your client.
 
Wait, when did the OP say he was opposed to using scene files? All he said was that he was depressed that the D7000 screen shot looked better to him in this one test. I think we've helped him & should give him the benefit of the doubt.
Valid point. I stand corrected. You're right. The OP didn't specifically say he was opposed, nor did I say he was opposed. I said he "doesn't SEEM to want to engage in the application of scene files" - hence the depression. Seems like he wanted the footage to be superlative without fiddling. That's what I inferred from the "depression." My take? It's nothing to be depressed about. Paint with the scene files and grade and it'll mimic or blow most things away in its price point. If you don't want to do either, I just wouldn't recommend the AF100. From what we've seen from some talented individuals on this forum, the AF100 can mimic quite a few looks out there. That's my humble opinion and you don't have to agree with it. Like Barry said, it's an example of shooting with factory settings. There's very little you can't fix with the scene files or grading or good lighting. Scene files or grade or both. I do both. You want to talk about depression - see what the 44-176mbs hack GH2s are putting out, see Bloom's Winter shoot-out, there's no scene file work-around for that. Nonetheless, most of us are still content with our purchase. So have we helped, Turbodog? I hope so. We're all saying the same thing with different styles. I probably just wasn't clear enough. No harm intended on my part.
 
Back
Top