Another Mk3 or the C100?

[Michael Kraus]

A bit surprised that the 6D didn't get a vote in this discussion. I imagine the extra $1000 would come in handy at some point.

[Adrian Tan]

To be honest, it just wasn't on my radar; I suppose because the camera news blogs I follow don't really talk about it.

In hindsight... You're right! it does look like a better video buy to me than a Mk3.

I have problems with SD cards, but I think that's just peculiarities of my setup.

Absence of headphone jack really doesn't fuss me; I never use the jack on the Mk3. If I really care about sound, it gets recorded separately.

[Nigel Barker]

Originally Posted by Michael Kraus

A bit surprised that the 6D didn't get a vote in this discussion. I imagine the extra $1000 would come in handy at some point.

Originally Posted by Adrian Tan

To be honest, it just wasn't on my radar; I suppose because the camera news blogs I follow don't really talk about it.

In hindsight... You're right! it does look like a better video buy to me than a Mk3.

I have problems with SD cards, but I think that's just peculiarities of my setup.

Absence of headphone jack really doesn't fuss me; I never use the jack on the Mk3. If I really care about sound, it gets recorded separately.

The 6D is Canon's new bargain priced full frame camera. Nobody has used one as they aren't shipping yet (CVP expect them 10th December). I suspect that Canon were forced to announce earlier than intended because Nikon announced & are even shipping their own new bargain priced full frame camera the D600.

The most important feature that the 6D lacks is the super new improved AF of the 5D3. Hopefully it will have the same amazing low light performance & moire free video of the 5D3 although it doesn't appear to use the same sensor so who knows? Here is a moderately informative review albeit without any sample images Canon EOS 6D Hands-on Preview: Digital Photography Review

[Adrian Tan]

Revisiting this old thread (in which I think I've made several inaccurate statements, and, as usual, several quite silly statements, but anyway)...

I'm jealous of Art and Tim's C100 footage. I definitely wish Santa would put a C100 in my stocking. I not infrequently type "canon eos c100" into eBay or keep a watch on the marketplace (I think they've sold for $5,000 and $5,500 there).

But whenever I've seriously considered the matter, the conclusion I keep coming to is that, for me, in my particular situation, it doesn't make money sense to invest in a C100 (I currently shoot mainly weddings, with two Mk3s and two Mk2s and two camera operators, shooting mainly on zoom lenses). If I really need another camera (I sort of do, I sort of don't), it would be another Mk3.

Here's the thinking.

C100 cost savings:
-- data storage -- more efficient than a Mk3. From memory (I might well be wrong), C100 fits about 6 hours onto two 32gb cards; Mk3 fits four hours; Mk2 fits three hours.
-- built in wide dynamic range profile could mean less time/money spent grading

Cost increases from a C100:

-- more different types of batteries and battery chargers (and more junk to carry around)
-- possible extra cost in trying to match footage
-- the elephant -- the actual cost of purchase.

And, really, that's all that the financial argument, in my case, amounts to. A lot of expense for no appreciable savings. The money could more wisely be spent on lenses for image quality improvement, or an extra computer to speed up post production (BluRay rendering times are murder), or on advertising.

The difference in image quality from a Mk3 (eg sharpness, dynamic range, built in NDs to avoid higher shutter look, better lowlight) would not translate to a difference in income from weddings, partly because what sells your work is what people view online, and it's not straightforward to tell the difference. If I really wanted to up my image quality, switching from zooms to primes would make a bigger difference to me than a C100 sensor.

The difference in functionality (long recording time, ability to check focus while recording, etc) are things that elbow grease can compensate for. No real financial impact.

If I need improved quality for a particular project, then it would be Mk3 + external recorder, or Mk3 raw hack, or rent a camera...

Of course, everyone who reads this post will be in a different situation. Your mileage will definitely vary.

Edit: 6D I no longer think about, incidentally. Seems to be some compromise on image quality vs Mk3. For the difference in price, I'd much prefer a Mk3.

[Ger Griffin]

how do you find the mk3 intercuts with the mk2 Adrian?

[Eric Coughlin]

Originally Posted by Adrian Tan

If I really wanted to up my image quality, switching from zooms to primes would make a bigger difference to me than a C100 sensor.

Are you sure about that? I did a test below comparing the 5D Mark III with a Canon 100mm f/2.8 and Canon 135mm f/2.8 (one of Canon's sharpest lenses), and compared it with the C100 with various lenses (primes and zooms). I found the dullest lens I put on the C100, which was the Canon 70-200 f/2.8 version I (not II), was still sharper than the 5D Mark III with the Canon 135mm. That's speaking only in terms of sharpness.

Canon C100, 5D Mark III, & 60D Tests w/various Canon & Nikon Lenses on Vimeo

[Adrian Tan]

Eric, I'm not sure about my claim! I find it hard to compare sharpness in your test because of the way it's cut, because you're shooting different objects, and because you dared to use real-world situations instead of concentric circles. But I should say that 135mm on a MkIII in your video does appear sharper to me than the 70-200 on the C100, or at least close enough that I can't easily tell (just looking at the priest and the girl's eyes). And also that the 70-200 MkI is supposed to be photography's second sharpest zoom (the sharpest being the MkII).

You've made me go and do some research into lens sharpness, which has just left me even more confused -- resolution depending on aperture and distance from centre of sensor and focal length, perceived sharpness being a combination of resolution and contrast, etc. End of the day: I'm still inclined to believe that lens matters more than sensor, but you've raised a doubt in my mind. Need more tests!

Ger, this is just my opinion -- Mk2 intercuts perfectly with Mk3 except under when it starts to get dark, where Mk3 just looks so much better. Less noisy, more saturated. But even then, they cut fine as long as you have enough light on your main subject.

Here's a recent 5-minute clip (did zero attempt at trying to match colours). Password: "password". The very last shot and very first shots were Mk2s, and all the steadicam shots through the whole clip were Mk3 with a 16-35 lens. I think you can spot the Mk2 shots in some of the general dance footage, because of noise and underexposure, but that the difference otherwise is not so straightforward.

I think the difference between lenses is more obvious than the difference between cameras. For instance, when the family are congratulating each other outside the church, that's a Mk2 with a 70-200, and I think it's so much sharper when compared with the 16-35 steadicam shots.

Private Video on Vimeo

[Eric Coughlin]

Originally Posted by Adrian Tan

For instance, when the family are congratulating each other outside the church, that's a Mk2 with a 70-200, and I think it's so much sharper when compared with the 16-35 steadicam shots.

Particularly for wide shots (not to be confused with close shots with a wide angle lens), I would argue for the camera over the lens. On close face shots, like in the samples in my test video, the difference in sharpness is not as noticeable on a 5D vs a C100. The C100 holds detail in wide shots much better than the 5D, where the 5D image really starts to fall apart and look soft, no matter the lens. I think the 5D is incapable of creating wide shots with a lot of detail (unless you use the RAW hack), while it is capable of providing sharp close shots with good lenses and good lighting.

I find the girl with the 70-200 to be slightly sharper than the priest, mainly looking at the eye brows which have more detail in them. The priests eyes may appear sharper mainly because they are better lit, where the girl's eyes have a good deal of shadows in them which makes them harder to decipher. If you compare the closer comparison, the shot before the priest, of the girl in the same light with the 5D III vs the girl in the same light with the C100, I think the different is more noticeable. It's easier to tell the difference if you open two tabs with the video and switch between them. There's a lot more detail in her hair as well, even in the C100 shot at f/2.8. And of course, here we're talking about the difference between the dullest lens on the C100 vs the sharpest lens on the 5D III; put the 135 on the C100 and then the difference between cameras would be much more noticeable.

I've seen a lot of test videos online comparing the 5D to the C100 with wide shots, and it always looks like the 5D is out of focus, but it's really just that the 5D never handled wide shots well. I think it is mainly because the codec compresses the detail too much, which is why wide shots in RAW hold a lot of detail. This is also why a lot of people who own a single C100 or C300 and a 5D will in a two camera interview use the 5D for the tight shot of the face and the C100 as the wider shot of the interview.

Consider these videos...

This one uses the Zeiss 15mm, a $3000 wide angle lens, with the 5D Mark II.

Zeiss Distagon T* 2.8/15mm meets "Luftbrückendenkmal Frankfurt am Main" on Vimeo

And here, the C100 with a $450 Rokinon Cine Lens.

Testing Rokinon 14mm Cine Lens on C100 on Vimeo

Certainly not a great way to compare cameras since we don't know the compression settings used in those videos and they're shot by different people in different locations, etc., but my point is that as far as wide shots go the C100 can get wide shots with lots of detail while the 5D cannot (as far as I know) without a RAW hack.


Another thing to consider with the C100 is, as you mentioned reading, that better apertures will increase sharpness. All but one of the tests in the C100 video I did were at f/5.6 (one test was at f/2.8), shooting at 850 ISO which is the native ISO in Canon Log on the C100. The native ISO of the 5D III is ISO 160. So for an interview setup, with the lighting that I own (I could always buy higher powered lights, but that costs considerable amounts of money, and blinds the interviewee more), I'm forced to shoot at f/2.0-2.8 on a 5D if I want to stay at the native ISO (which I generally do). On the C100, with the native ISO at 850, I can shoot at f/5.6 and get better results from the lenses. Add to that, generally cheaper lenses (and good zooms) struggle more when wide open, but are more comparable to expensive primes when stopped down to a more ideal aperture.


I'd also take note that some people say that you can't notice the difference in sharpness between the cameras when it gets compressed online. I think this depends on how you compress the videos. If you don't upload and upgrade your video on Vimeo to 1080p (from 720p), and if you don't upload at a high data rate (say 40,000-60,000 Mbps instead of 5,000), then the difference between cameras will be much less noticeable.

[Dan McGuckin]

I just had the tossup between a 5dm3 and C100 to replace my HMC40 and to go with my 550D. The C100 won and just purchased it today. For it it was all about the XLR and ability to shoot all day. 5d may be in the future as a B Cam, but just not today.

[Ger Griffin]

Thanks for that mk2 vs mk3 in a nutshell Adrian.
For me with weddings its dslrs all the way.
for the simple reason that people act so much more natural around a stills camera than a video camera.
But this is nothing new to any of us at this stage.