Canon C300 Mark II - $15,999 4K camera

[Guest Ebrahim Saadawi]

It does actually look like an extremely clean sensor, plus the DR does look 12 stops. It's probably the highest image quality sensor at 1" size, all it lacks is the shallow DOF capability. 

If you look at the image, it's not rx10/FZ1000 league of 10 Stops of DR, terrible noise above 800 ISO, and video rev 709 gammas. Just because two cameras have the same 1" sensor size doesn't mean they're equal in image quality, this is the mistake we made looking at the rx10 & fz1000 as being equals for half the price and that's the reason we dismissed it momentarily, they are not, at least not on IQ side. 

When you add in the colour science and way superior codec, colour sampling, broadcast work approval, better ergonomics, modular design, removable grip, matching C-Log image, smaller weight and size for drones & pov (it really can be broken down to tiny gopro like device not a DSLR at all after removing the grip, the lens hood, the EVF attachment and any hotshoe accessory) the XC10 starts to make a lot of sense for its price. It's a first for it's kind as a 4K, modular, POV camera, that can be built up for use as a handheld doc/broadcast ready form, with C300 like colour, DR, Log mode, 422 300mbps codec, very versatile optical range, IS and AF that actually works. 

I hate to sound like a Canon defender but everytime they release a new product I find my self bashing it, then realize they knew better than me in how it will serve me as a filmmaker/videographer. 

It's sometimes hard to step out of the camera-geek suit where one is lusting for the highest possible numbers, widest lenses, without caring about versatilitt or real world shooting ease of use. The same reason why I dismissed the C300 for the Scarlet the day they were announced. But in actual reality, the entire package is all that matters not just one head grabbing number and a letdown overall package. 

[Jimmy]

Agree completely. At first, it made no sense to me, as it seemed about GH4 size... But it is quite a bit smaller, when stripped down for drone use. Add 12 stops and 4:2:2 at 4k and it starts to make sense.

The IQ looks really nice, the one thing that paper specs can never measure.

[mtheory]

This is a perfect broadcast camera, but its not a filmmaking camera.

[Jimmy]

It is not an A cam, but it will be great as a drone/action cam within a film.

Can't work out why people are not seeing this? 12 stops, 4:2:2, 4K in a body not much bigger than a gopro.... If someone makes a gimbal for this, it'll be quality,

[Guest Ebrahim Saadawi]

It also looks a clear step up in build from smaller mirrorless/dslrs/hybrids 

Here it is on a POV shot



If you notice the 5 shots taken with xc10 in the film, it blends perfectly with C300 mkii. In the car exterior shot it especially shows how the image is almost indistinguishable from the C300 in sharpness, identical colour rendition of the desert & sky colour, and a pretty wide DR. Just seems like a very similar image to Canons C100/C300. 

[kedbear]

​No you are not because there are no straight comparison shots with Alexa right there. You are using a "mental image" of the perfect image of Alexa as a benchmark, not actual reality. 

I don't need direct comparison with the Alexa. I have shot with the Alexa multiple times and that forms the basis of my comparison, not an imagined benchmark. The idea that you need to see direct comparison from one camera to another in order to form an opinion is one that can only come from somebody not familiar with these cameras. The highlight roll off on the c300 mkII sucks, the over sharpened image feels like video. If you like that look fine. 

[jcs]

A direct comparison is helpful- here the Sony F55 matches (or exceeds) the Alexa:

And the F65 exceeds the Alexa (don't need to take my word for it- the guys doing the comparison did their own tests: Luc Besson & his DP): http://www.eoshd.com/comments/topic/7757-lucy-amazing-color-shot-on-sony-f65/

 C300 vs. Alexa vs. FS7:

It really is helpful to do side by side tests under the same conditions to evaluate camera systems. Canon can match the Alexa too. Crazy spaces/formatting courtesy forum software bugs.

[hmcindie]

I don't need direct comparison with the Alexa. I have shot with the Alexa multiple times and that forms the basis of my comparison, not an imagined benchmark. 

But you are comparing them like you have shot both. You haven't. You are comparing an imagined benchmark (if you have indeed shot with the Alexa, you have not shot it in the same environment, with the same color grader). In this benchmark the Alexa is super smooth and retains all highlight details. Then you compare it to a graded C300mkII video that was shot with another crew who may have added sharpening and even burned highlights in the grade. 

You may not need it, but I need straight on benchmarks. I read a lot of computer articles (benchmarks with different GPU's etc) and I never assume that one CPU/GPU is better than another if someone plays a completely different game with different settings on and doesn't do a straight up comparisons. Science trumps feeling.

Imagine how ridiculous it would be that when Nvidia releases a new graphics card, I would just say "I've used the newest Radeon and it has better anisotropic filtering because of this clip I saw from youtube". That would just be insane.

[AaronChicago]

 I haven't seen this brought up yet, but it BLEW my mind. Check out this manual focus feature that uses auto face detect to guide focus: https://youtu.be/ElZTQIP_BaQ?t=5m11s

 

[kedbear]

But you are comparing them like you have shot both. You haven't. You are comparing an imagined benchmark (if you have indeed shot with the Alexa, you have not shot it in the same environment, with the same color grader). In this benchmark the Alexa is super smooth and retains all highlight details. Then you compare it to a graded C300mkII video that was shot with another crew who may have added sharpening and even burned highlights in the grade. 

You may not need it, but I need straight on benchmarks. I read a lot of computer articles (benchmarks with different GPU's etc) and I never assume that one CPU/GPU is better than another if someone plays a completely different game with different settings on and doesn't do a straight up comparisons. Science trumps feeling.

Imagine how ridiculous it would be that when Nvidia releases a new graphics card, I would just say "I've used the newest Radeon and it has better anisotropic filtering because of this clip I saw from youtube". That would just be insane.

So basically the suggestion is that a person with experience, even a top class DoP, can't possibly form any opinion of the camera from watching something shot with it in a real world scenario. The only way is to shoot the exact same shot with two cameras and then grade them the same? 

Why do you presume the way that you assess a camera is the only way it can be done? Do you have experience shootng with high end cameras? There are people out there with enough experience to be able to assess the basic quality of a camera from watching it's footage. I personally would wager that the c300mkii has the hallmarks of v1. Poor highlight roll off and overly sharp with a videoy feel.

 

[kedbear]

A direct comparison is helpful- here the Sony F55 matches (or exceeds) the Alexa:

 

The above example is not designed to push the cameras. Having shot both of these cameras I can tell

you the highlight roll off of the f55 is poor. The LUT helps match the colours (though the colours are still visibly different in the above example) but the two images still feel different, even moreso when watched on a big screen without vimeo compression.

 

[Jimmy]

There are people out there with enough experience to be able to assess the basic quality of a camera from watching it's footage. I personally would wager that the c300mkii has the hallmarks of v1. Poor highlight roll off and overly sharp with a videoy feel.

​You have access to the 4K files? Nice, the rest of us only saw a 720p version, compressed to YouTube... What about the behind the scenes video, where they clearly demonstrate insane levels of DR in the highlights? What about the part of the video where they said they graded it to fit the heist genre?

Put simply, when you have one 720p video that has been heavily graded to suit one colourist's style, you would be crazy to use that as anything close to a definitive idea of what that camera can do.

I do wonder if Canon/Blackmagic had accidentally uploaded the wrong video to each of their accounts... Would some here have claimed the opposite video was better, due to some pre-existing idea of what they think will be wrong with Canon/BM footage

[hmcindie]

So basically the suggestion is that a person with experience, even a top class DoP, can't possibly form any opinion of the camera from watching something shot with it in a real world scenario. The only way is to shoot the exact same shot with two cameras and then grade them the same? 

​Yes, that is my suggestion. There are loads of deluded DoP's out there with very weird ideas about what constitutes image quality in a camera. They can still make a great looking image though. Claiming that you can see from a graded image what camera is which is just bonkers. Did you ever see the Zacuto shootout? It was quite difficult to tell certain cameras apart especially when no one told you before hand.

You've been saying two specifics things. 1) Over sharpened. You can turn sharpening down on a C300, you can also disable it completely (of course, that means you have to dig into the menus). The image is also 720p on vimeo and the sharpening can come from grading and rescaling (Premiere sharpens with default settings for example) 2) Highlight roll of. You say this several times. Lets take a look at the first shot of the film. It looks badly graded. It looks like someone took some curves adjustments and whomped them up and called it a day. Like every post you've made has been about "highlight roll-of". Yes Alexa has good highlight roll-of but come on now. Is that the only thing your looking at because it does sound like it?

Did you even look at the behind the scenes where you can clearly see that those highlights are not burned in, but they did it in grading?

[kedbear]

​You have access to the 4K files? Nice, the rest of us only saw a 720p version, compressed to YouTube... What about the behind the scenes video, where they clearly demonstrate insane levels of DR in the highlights? What about the part of the video where they said they graded it to fit the heist genre?

Put simply, when you have one 720p video that has been heavily graded to suit one colourist's style, you would be crazy to use that as anything close to a definitive idea of what that camera can do.

I do wonder if Canon/Blackmagic had accidentally uploaded the wrong video to each of their accounts... Would some here have claimed the opposite video was better, due to some pre-existing idea of what they think will be wrong with Canon/BM footage

​Ok watched the behind the scenes. The BTS video does demonstrate good DR. However, i can't for the life of me understand why on earth it would be graded so that there's a nasty highlight roll off, it just looks bad. The idea that it fits a heist genre doesn't hold water for me at all. You don't see anything like that in The Town for example amongst many many others. I can't help but suspect something fishy/odd is going on. Anyway, taking DR out of it and the colouring style, i still find the footage videoey. Of course, it's way better than many cameras, but amongst the Epic and Alexa i don't think it has the 'feel' to the image. Lets see when the camera arrives i guess... 

[Guest Ebrahim Saadawi]

I do agree that from the BTS DR the camera clearly has a mind-blowing range and highlight rendering, and also agree it was a bad grading choice to burn them so abruptly and the grade overal feels ''video''. If it was a standalone film, I respect the look they wanted, but to showcase a new DR camera, bad choice. Good thing the uploaded  the BTS because I was initially disappointed in the DR. All they left there is 11-12 stops, c300 mk i kind of level, that plus uploading in 1080p eliminates the entire step up from the original to 4K resolution and 15 stops DR.

[jcs]

The above example is not designed to push the cameras. Having shot both of these cameras I can tell

you the highlight roll off of the f55 is poor. The LUT helps match the colours (though the colours are still visibly different in the above example) but the two images still feel different, even moreso when watched on a big screen without vimeo compression.

 

​Hey kedbear- I get your point regarding the Alexa and its ability to create a great image more easily than the F55 and many other cameras- that's why it's so popular. In the clip which matches the F55 to the Alexa, can you select frames which show the highlight roll off being poor on the F55 and not on the Alexa, describing which area/pixels are having the issue?

[Raafi Rivero]

The specs on this camera are just ridiculous, and that autofocus feature looks amazing. But the price is high; there's no way around that. The question is deciding whether there's actual value for that price. For instance, based on these specs and the one film we've seen so far, I'd choose to buy this camera over the lesser Reds -- you'd probably be saving a few grand in that case, and gaining a better post workflow. There's a ton of value compared to the FS7, even though you're paying twice as much. But compared to FS7 we get better low-light capability, weather-proofing, color science, lens selection, number of NDs, multiple card/recording options, superior monitoring, and the aforementioned focus options. It pretty much has everything we say we want. The catch is that it comes with a price tag many people will find too high.

The Kinefinity cameras are very interesting as well. The footage coming from them is cinematic-looking in a way that the one C300mkII film isn't. But the Kine cameras pose their own set of issues: support and workflow to name two. Most people who've reviewed also say that the Kine system's menus are cumbersome to navigate which is a big strike against if you ever plan to be in a run-and-gun or documentary type situation.

Although I've had projects shot on Alexa, S-16mm film, Red, C500, C300, F5, BMPC 4k, etc. on down the line, the only camera I've ever owned is the 5DmkII. Both the C300 MKII and the Ursa Mini (half the cost, once kitted out), have me seriously considering buying one. But the (presumed) low-light deficiencies of the Ursa-mini are a great worry. I rented the BMPC 4k once and the grain on that thing at 800 ISO was a HUGE disappointment -- they have to present some serious improvement there for it to be a consideration.

As an independent filmmaker, the simplicity of the system can pay as many dividends as the quality of the image. Add to that the fact that Canon is a reputable company and a worldwide operation. Things like that give me confidence in spending a little more. And keep resale values higher. Kinefinity could close their doors tomorrow and not a single person here would be surprised. Does that make me happy at the idea of spending $16k for something the Kine folks are selling for ~10? Absolutely not! But those are the big-boy decisions we have to make.

I once bought one of those Rokinon Cine lenses. And the image quality was an instant boost over my vintage still lenses, comparable to the Canon L series. I considered buying the rest of the set. But then one of the glass elements inside of the lens came loose and the lens is almost never in focus on the outer edges anymore. Well, it's a $350 lens, what did I expect? The next time I spent money on lenses I bought vintage Lecia R series and am adapting them for cine. The metal housings are rock solid in my hands. I can feel where that extra money went. AND it shows up on the screen. The point is that there's a lot of quality packed into places beyond the spec sheet, and we as creators have to decide how much value we place in them.

[kedbear]

​Hey kedbear- I get your point regarding the Alexa and its ability to create a great image more easily than the F55 and many other cameras- that's why it's so popular. In the clip which matches the F55 to the Alexa, can you select frames which show the highlight roll off being poor on the F55 and not on the Alexa, describing which area/pixels are having the issue?

Hi jcs. There is no issue in your example, because the lighting has been controlled, favouring the DR of the f55. Start increasing the power of the backlight and it would be a different story. My point is a controlled test such as the one you posted does not show much apart from that the f55 has a Lut that gets it close colour wise to the alexa's rec709.

[kedbear]

​Yes, that is my suggestion. There are loads of deluded DoP's out there with very weird ideas about what constitutes image quality in a camera. They can still make a great looking image though. Claiming that you can see from a graded image what camera is which is just bonkers. Did you ever see the Zacuto shootout? It was quite difficult to tell certain cameras apart especially when no one told you before hand.

You've been saying two specifics things. 1) Over sharpened. You can turn sharpening down on a C300, you can also disable it completely (of course, that means you have to dig into the menus). The image is also 720p on vimeo and the sharpening can come from grading and rescaling (Premiere sharpens with default settings for example) 2) Highlight roll of. You say this several times. Lets take a look at the first shot of the film. It looks badly graded. It looks like someone took some curves adjustments and whomped them up and called it a day. Like every post you've made has been about "highlight roll-of". Yes Alexa has good highlight roll-of but come on now. Is that the only thing your looking at because it does sound like it?

Did you even look at the behind the scenes where you can clearly see that those highlights are not burned in, but they did it in grading?

Tell that to Roger Deakins.

I did see the zacuto shoot out, I thought the Alexa stuck out. A lot of the mid range cameras looked very alike. I'm sorry but if you think the c300 mkII looks great and comparable to an Alexa that's fine, however for me I can tell immediately that it isn't. The sharpness, the roll off, the motion, the noise, the colours, the overall organic 'feel' of the image simply isn't as attractive to my eyes as the Alexa, and still bears the c300 mkI family of 'look'. There is a reason the Alexa is the most used digital camera in the industry. Can you get great images out of lower end cameras? Of course, I think 'Blue Ruin' showed that. Can the c300 look like the 'collider' music video for Jon Hopkins. No.

[Guest Ebrahim Saadawi]

We still have NO comparison between alexa and C300 mkii under the same environment. Until then, any conclusions one is superior/inferior to the other are merely baseless predictions.