Canon C300 MKII Announced

A couple of things thus far in this thread. Let's not get hysterical over anything first and foremost. It's another tool to be discussed as such.

Anyone is welcome to discuss any other camera as a comparator. People also should have the ability to show their interest in this camera with out person A coming and saying numerous times, well but Camera Y has this over the C300MKII. We get it, and Ill be warning if I see it anymore. Please self moderate as always, use the did you find this post helpful buttons (in the bottom right corner), three down and it will become hidden.

Some thoughts from me on this - Ive read in this thread numerous times the C300 can't do 12 stops. Ive seen numerous DR tests that say 12.5, after three years of use I'd say a solid 12 any day of the week if you're experienced shooting with it or you've read up on how to shoot with it from experienced shooters (ah the beauty of the internet is usually someone else has done it). I know theres some DR tests out that that say differently, I am saying in my experience I disagree.

I also read that Shane Hurlbut jumped ship from Canon to Red. Shane's an experienced cinematographer - he's shooting Red Epic Dragon for his next production and testing the hell out of the Dragon before he does. He's not on any ship - just shoots with what works for what he wants to achieve.

Just to clarify some hyperbole there.

The C300MKII for me, pretty much looks like what we've been promised and they seem to have listened to what we weren't so happy about on the C300. It's early days, but like any camera I'm interested in putting it through its paces. 15 Stops of DR in my experience usually equates to what the individual finds acceptable as far as noise goes. The camera, Im sure can do 15, but again Im interested in how it actually plays. Ive always found as a rule, its about one stop under for me as to whats in the spec, however the C300 seemed to give me all of its 12. Might be pleasantly surprised?

As far as the shoulder mount - its not designed out of the box to be one - third party like Zacuto fill the void there.

DLD said:
But, I think, as roughly what harddrive posted on the Varicam thread, dual readout is about the only way a modern camera can deliver an image similar to a photo Raw.
Click to expand...

Except thats not the entire story - ah if only camera manufacturers would give up their tech openly.

Pana are going to give me some charts on that FWIW. Waiting. (this doesn't belong here - but just wanted to point it out.
 
Sttratos said:
I've been saying this from page 1. Everybody behaving like the press release is the gospel. For several reasons I will be very surprised if it has real 15 stops.
Click to expand...

If anything, Canon has under-spec'd their cameras in the past...and knowing the people at Canon, they would not put anything in their spec sheets that was inaccurate.......

Jim Martin
 
steve phillipps said:
Well that's the key point isn't it - with that pricing they obviously see it in a different class to FS7 etc....but based on what? On the shoulders of the success of the original? Maybe - and maybe they've got it right - again, I just don't get it - but if I'm in the minority I'm happy with that.
Steve
Click to expand...

I do understand why it looks a certain way to many, but to perhaps suggest why there is a disconnect:
I believe a huge part of the C series love comes from day to day ownership and lots of diverse shooting. So, while it may be true that on paper, the C series look like cameras that are spec deficient and perhaps over-priced compared to others, the real world cameras present a much different picture for countless owners and shooters.

It's all good though, the Sony is a tremendous value, perhaps the best around right now.
 
DLD said:
Oh, and, unlike XC10, C300 MK II doesn't seem to be crippled in any way, shape or form. Its competitors, even including the FS7 and GH4 among then, have a few strikes going against them.
Click to expand...

4K 30p internally only is not being crippled?

No real viewfinder, LCD in weird position, bad ergonomics. How does the FS7 have more strikes against it?
 
Sttratos said:
4K 30p internally only is not being crippled?

No real viewfinder, LCD in weird position, bad ergonomics. How does the FS7 have more strikes against it?
Click to expand...

Postulating is against our forum rules. No more warnings. I said as much just a few posts back. Specs are for comparison only.

Guys try to have an open discussion.
 
Eric Coughlin said:
Don't know if anybody knows which I asked a page back, but are the Atomos CFast cards compatible with the C300 Mark II?


The bit rates are listed as, "Intra 410/225/220/110Mbps, LongGOP 50Mbps, Long GOP 35 or 24Mbps (for proxy recording on SD cards)".

Can anyone point me to info on how those data rates correspond to resolution (4k vs 1080), color space (4:4:4, 4:2:2, and 4:2:0), and bits (12 bit, 10 bit, 8 bit)? Can 4k be recorded at a data rate lower than 410 Mbps? On the FS7 4K at 30p can be recorded at 220 Mbps with 10 bit 4:2:2 as well as 100 Mbps with 8 bit 4:2:0 which is nice.

At 410 Mbps on a 128 GB CFast card, you'd be getting around 42 minutes of recording if my math is close. 1 hr 24 minutes on a 256 GB card which are currently only from Lexar at about $1k each. I'd figure four cards would be sufficient, so an extra $4k. With Atomos' 128 GB cards, you could get eight for $2k. Maybe 2 256 GB Lexars and 2 Atomos 128 GB cards for $2440 would be a good bet if the Atomos' work. I don't see the point in paying $800 for a 128 GB SandDisk when the Lexars are just $200 more for 256 GB.
Click to expand...
I believe that CFast 2 cards would be required. The Atoms are CFast 1.
 
Let's all remember that a choice of camera is not an us v. them thing. It's just a choice. If you own a different model camera you should not feel the need to justify your possession by criticizing another item.

On paper a $20,000 car can have specs better than a $40,000 car. But that's not the whole story and some people prefer the more expensive one. Fair enough.

Forgetting all other aspects including technical specs and price points, some people simply prefer the way a Canon camera's image looks. Others prefer Sony.

If there were no reasons for all this, there would only be one brand of camera makers out there and all the cameras would be the same. That is not the case and it never will be. The same is true for just about every other industry on the planet.
 
Noel Evans said:
... Except thats not the entire story - ah if only camera manufacturers would give up their tech openly.

Pana are going to give me some charts on that FWIW. Waiting. (this doesn't belong here - but just wanted to point it out.
Click to expand...
Noel, you know what post I was referring to, right? It wasn't about the Varicam per se but about why the dual-read makes sense at this moment in time. (as in, basically, one processor handles the highlights, the other shadows and they all meet in the middle ... a grossly simplified explanation, obviously, but that's the gist).

Sttratos said:
4K 30p internally only is not being crippled? ...
Click to expand...
No, I don't think it's being crippled. They're neither trying to protect another higher end camera nor sell you a piece of kit that they produce themselves.

And one can still get 4K 60P with an external piece from ... well, how about Mitch ...

Mitch Gross said:
No, just a typo. I'm just so used to answering questions about the C500. I've corrected it.
Click to expand...
I figured that. Mine was just a leeetle joke.
 
Personally I'm very grateful for all the great cameras we have available at the moment. Feels like some kind of golden age - may not always be true as cell phones continue to eat the lower end of the market taking out profit centers for all the camera manufacturers.
Looking forward to trying the 12 bit 4:4:4 2K internal - if the 15 stops of DR are real and it gets close to the Amira this could be a great documentary camera.
 
Mitch Gross said:
If you own a different model camera you should not feel the need to justify your possession by criticizing another item.
Click to expand...

Or worse if you have a vested interest in the success of one. That's why real names are always a very good idea.
I haven't owned any camera for over 5 years but have used enjoyed gear from Sony, Canon, Panasonic, Vision Research and RED. I'm as impartial as you can get!
Steve
 
If Canon has indeed made a 15 stop camera then the C300 Mk2 will be a very successful and widely loved camera, despite the lack of 60p. 15 stops, even if it's only 13 'usable stops' (many people rate the Alexa at 400, meaning they sacrifice a stop for cleaner shadows) puts this sensor into the 4K top class along with RED and Panasonic (Sony outclasses, Alexa lacks resolution). The potential highlight rolloff will be fantastic. Put that together with the nice workflow and great lens environment, I predict this camera doing very well. It seems like it will serve owner operators for the price (relative to it's expected demand), name recognition and dependability. Adding a 7Q isn't that expensive if you want to do some 60p 4K, though having to do it RAW is a bit of a pain.

All the included color science is also quite exciting. Obviously designed before the Alexa Mini announcement, but most likely a superior camera, being able to match it to LogC and the inclusion of a variety of color spaces and Log options can make the camera super versatile. I predict that many more features will be shot with the C300 II alone.

All this news makes me excited for a C500 II, which has been rumoured as a more Alexa style body camera. This sensor could do big things if unleashed F55 style. I hope this means Canon releases some new cinema primes as well. It's great how the 4K environment is providing a ton of camera options, unique colormetrics, each with their own plusses and minuses, but all of them good enough to shoot a feature with.
 
Regarding DR, specs and measurements:

Long, long time ago, I measured the 5D2 at "close to 13 stops" with my most extreme Flaat profile. The method was as follows: point at a uniform surface, bring exposure down until there's no more signal in the waveform. That's what I call "absolute DR": if there's a signal, it counts. But then I realized that on real footage anything below a certain level in the waveform was just noise. Noise that gets darker when you close the iris, but noise anyway.

So, I measured again (now with the 5D3), this time looking at some actual target image (here), and with a rule that was "the shadows only count if I can discern the black box from the dark brown table, and it's not so noisy that it can't be used". With this definition of "usable DR", my measurement was "10 stops".

And to make things even more complicated: when testing the RAW mode, I had to decide whether completely colorless highlights that were nevertheless not completely white would be useful or not. If grey-but-not-white highlights are good for you, it has 11.3 stops of DR, but if you want your highlights to keep their colors, it has just 10.3 stops.

My conclusion: comparing numbers makes no sense, unless they come from the same test, ran by the same people on the same day. I'd rather look at a side-by-side comparison with a wide exposure ramp. Like this one, where, from 10:00 onwards, you can see that the F3 had nearly two stops better DR than the C300 (nearly half a stop on the highlights, about a stop and a half on the shadows).

If the C300II has a stop more DR than the FS7, it's a beast. But show me proof, not claims.
 
Samuel H said:
Regarding DR, specs and measurements:

Long, long time ago, I measured the 5D2 at "close to 13 stops" with my most extreme Flaat profile. The method was as follows: point at a uniform surface, bring exposure down until there's no more signal in the waveform. That's what I call "absolute DR": if there's a signal, it counts. But then I realized that on real footage anything below a certain level in the waveform was just noise. Noise that gets darker when you close the iris, but noise anyway.

So, I measured again (now with the 5D3), this time looking at some actual target image (here), and with a rule that was "the shadows only count if I can discern the black box from the dark brown table, and it's not so noisy that it can't be used". With this definition of "usable DR", my measurement was "10 stops".

And to make things even more complicated: when testing the RAW mode, I had to decide whether completely colorless highlights that were nevertheless not completely white would be useful or not. If grey-but-not-white highlights are good for you, it has 11.3 stops of DR, but if you want your highlights to keep their colors, it has just 10.3 stops.

My conclusion: comparing numbers makes no sense, unless they come from the same test, ran by the same people on the same day. I'd rather look at a side-by-side comparison with a wide exposure ramp. Like this one, where, from 10:00 onwards, you can see that the F3 had nearly two stops better DR than the C300 (nearly half a stop on the highlights, about a stop and a half on the shadows).

If the C300II has a stop more DR than the FS7, it's a beast. But show me proof, not claims.
Click to expand...

We'll get that proof soon enough. Whoever tests these cameras first will get a massive amount of traffic as everyone who holds a camera will be watching. Hell of an incentive.
 
I'm afraid the first reviewers may do a sharpness and low light test, and maybe a DR test where they point the camera at a sunny street and say "it looks fine".

BTW I just read the eoshd piece on the C300II and wow, he truly has no idea how sensors work down at the low level.
 
Last edited:
Whoah - so much argument about C300 II vs FS7 vs X. The reality for me and pretty much most of the people I know is that the most important factors in order of important to getting work are 1) Previous working relationship with client. 2) Networking, networking, networking. 3) Your reel and past completed works 4) Your gear. That's it folks! Oh, some people do want 4K over HD as a deliverable requirement now.
 
DLD said:
Andrew Reid at EOSHD says the dual readout is an "in-camera trickery".
Click to expand...
What Canon are doing with the C300 I have no idea, but what the Alexa does may be "trickery", but it's very good as such. It's a way of getting out all the sensor itself is capable of, and not being limited by the read mechanism.

DLD said:
But, I think, as roughly what harddrive posted on the Varicam thread, dual readout is about the only way a modern camera can deliver an image similar to a photo Raw. In an audio world, it's sort of like comparing a full-range speaker with a 2-way, a separate tweeter for high frequencies and midrange/woofer for the lower frequencies. A single driver can only cover so much.
Click to expand...
I had to go back to see what I did put ( :) ) and I think what you're referring to is something a much more knowledgeable friend sent me, ( http://www.dvxuser.com/V6/showthrea...ricing&p=1986529793&viewfull=1#post1986529793 ) especially the last paragraph where he talks about the Alexa read.

It’s worth mentioning the principle behind the Alexa, as that does indeed use dual analogue circuitry, and individually such circuitry is not capable in isolation of doing justice to the sensor. The Alexa sensor is sub-4K resolution, which means individual photosites are large compared to 4k s35 sensors. Which gives it an inherent native DR advantage compared to the 4K versions. But the 14 bit A-D convertors can’t take full advantage of that (they become the limiting factor, not the sensor). To get over this they employ dual simultaneous read, one low gain A-D dealing with highlights and the high gain path dealing with the dark areas. Combine the results and you have what they call meaningful luminance information in all 16 bits. See http://www.arri.com/camera/alexa/tec...alexas_sensor/ .
Click to expand...

The real point about the Alexa is that dual-read works there because the individual A-D convertors are the weakest link in the DR chain - not the sensor itself. And that is perhaps more because the sensor is so good than the A-Ds are poor! :) It's largely because the Alexa isn't 4K that it wins in this way - it enables the individual photosites to be larger for a s35 chip size.

So when we talk about other cameras - where now 4k is the norm - the issue becomes less. It becomes easier to design the analogue circuitry to roughly match the sensor, so the need for dual read tends to go away. Your loudspeaker analogy is pretty good in principle.

In his note, my friend starts off by saying "I’m afraid what follows is complicated enough, but the practical reality is even more complicated. For example, I have ignored noise, but hopefully it may be of use in helping to understand basic principles." and whilst the fundamentals of DR may be straightforward enough, then real world measurement is as much an art as a science, as I think someone else has already said. [EDIT - Samuel H, a few posts back!] To make meaningful comparisons, it's essential the same methodology is employed unit to unit - and ideally all done by the same person. In practice that doesn't happen, and hence why widely differing figures get seen for the same camera. Hardly surprisingly, the marketing departments of the manufacturers tend to prefer the highest figures....... :)
 
Last edited:
You must log in or register to reply here.
Back
Top