Camera Labs

Hi folks,

Thanks to 4/3 Rumors for flagging the Zacuto Great Camera Shootout 2010. The video compares a couple of film cameras with the Canon 5D MarkII, 7D, 1D MarkIV, the Nikon D3s and the Panasonic GH1 and, fortunately, it isn't advertorial for Zacuto products though I'm sure they'd like us to buy some.

There's rather a lot of "chatting" at the start but I found the tests themselves to be very instructive and well worth 35 minutes of my time. Episode 2 is yet to be published and it will include the recent Canon 550D/T2i as well. No download link that I could find, though.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

I checked episode 1 and liked it very much. Clearly, as far as dynamic range is concerned, film is the clear winner, though the 5DMKII performs incredibly well!

It also shows that Nikon is far behind Canon on HD videos on DSLR's (no offense to Nikonites here), though I'm waiting for the other episodes to see if Canon retains the lead on the other aspects tested.

_________________
Canon 5D Mark II
Canon 24-105mm f/4.0 L

I'm rather surprised at how well digital held up against film. I'm familiar with the claims that film is superior, better latitude, etc. But to see film and digital compared side by side like that, is a real eye opener.

There are times when digital doesn't retain details in the highlights (see the bricks, the light bulb) but that's really splitting hairs. You need to know what to look for and even then it wouldn't have been obvious unless you had film to compare against.

Now if only they'd added other cameras to the list. I'm curious to see how the Pentax DSLRs do.

_________________
PhilipGoh.com

Film in terms of dynamic Range is nothing short of impressive. I've read an interview with the DP of Sunshine (2007), and at one point he claimed that they had to overexpose the scene by NINE (!) stops in order to get blown highlights...

_________________
I take pictures so quickly, my highschool was "Continuous High".

.
I'd settle for a DSLR sensor with an extra two or three stops of dynamic range. Don't need any more megapixels than my current 21 but I'd love for exposure bracketing (HDR) to be consigned to the trash, though to do so completely might need that nine stops! Am I right in assuming that it's the non-linear nature of film's response curve which allows so much detail to be recovered from the highlight areas in those videos shot with film?

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Fascinanting video,impressive results on both sides.

_________________
Gear: Nikon D80, 18-105mm, 35mm F/1.8, 85mm F/1.8, SB-600
Visit my work here

I'm pretty sure digital is also non-linear, considering you can pull out much more shadow detail than highlight detail in 8 bits?

_________________
I take pictures so quickly, my highschool was "Continuous High".

The digital imaging sensor's response to light is linear. When the RAW data is converted (either in camera or on computer), we apply a nonlinear sigmoid (i.e. S shaped) tone curve to transform the data from 12+ bits to 8 bits.

That's enough geekiness from me on a Sunday evening.

_________________
PhilipGoh.com

Don't complain I'm still feeling last night

It makes sense though, the compression we apply is non-linear. film isn't compressed.

_________________
I take pictures so quickly, my highschool was "Continuous High".

.
By non-linear I was struggling to remember stuff I've long since forgotten. Fortunately as memory fades Internet search engines improve so here's the shape I was after, courtesy of this article on Answers.com about sensitometry:

There's also a detailed explanation in this Kodak paper (PDF file). If I'm interpreting it correctly the shoulder area in the graph above allows a greater dynamic range to be captured than would otherwise be the case. As I understand it the sensors we are familiar with in our cameras are essentially linear in their response until (my interpretation) that sudden cut-off when the well is full, though the shape of the output is modified by a gamma correction. Non-linear sensors have been produced but apparently suffer from problems such as fixed pattern noise (source) so don't really feature in consumer cameras.

But essentially all the above falls into the category of too much information for our discussion as the benefits of film are very visible in the sections of the shoot-out showing the light bulb. But maybe DSLRs shine in low light situations - sorry, bad pun. I'm looking forward to Episode 2 of the shoot-out.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Ah, well noted Bob!
As far as low-light goes: I'm pretty sure DSLRs would (or digital in general) would win, IF you go HD.

For instance, when I watch a general movie shot on film in HD, I tend to see a clean image (relatively clean, anyhow, there's always grain).

However, if I look at a film where higher film speed is used (saving private ryan for instance, to get a higher shutter speed/choppy look), I can see more grain....

Hmm....

_________________
I take pictures so quickly, my highschool was "Continuous High".

Just wanted to say I LOVED the video at the end, some great videography there!

Sony, hurry up! I wanna shoot DSLR-videos!

- Bjorn -

_________________
Street and documentary photographer | Google+ | Twitter | Blog | Tumblr

Leica M9-P (my article on Camera Labs) | Leica D-Lux 5 | 50mm Summilux

Check out the episode two, where they do low-light tests:

http://www.zacuto.com/shootout

Particularly, check the ultra-high ISO test they do between the 1DMKIV and the Nikon D3s. Nikon's results are impressive. It beats the crap out of Canon by a huge margin, and it beats film too.

This D3s has a wicked low-light performance, simply unbelievable, good job Nikon!

_________________
Canon 5D Mark II
Canon 24-105mm f/4.0 L

It's very very impressive. I wasn't so keen on the Nikon video at low ISO images, but at high ISOs it looks great!

_________________
PhilipGoh.com

Hi folks,

I finally caught up with Episode 2. Pretty amazing but I was left wondering what the various DSLRs would have been capable of if a video equivalent of RAW was available. The extra detail in the light-bulb when shot with film is self-evident but the image stored on film is sort of like RAW recording with the data massaged down into a digital format externally and, presumably, with much more control than is available with in-camera processing.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.