Rodney Charters, ASC - RED ONE, Canon 7D and 5D MkII side by side

[michael zaletel]

I give up. I should have known that trying to reasonably discuss the point of the piece and prevent it from turning into what it has now obviously become - namely, another rather pointless A/B comparison between one device that costs less than $2K and another that costs more than 10 times that amount - was a bit of a stretch. I'm sorry for trying.

The problem is that it really isn't a comparison between a $2K camera and a $40K camera. The Red One Body is only $17,500 and the 5D2 Body is nearly $3K. The next issue is the cost of all the professional support equipment is nearly the same. Lenses are a major factor and since you can use Canon Lenses on the Red One but can't use all the lenses a Red One supports on a 5D2, that's a wash.

However, the real issue is the cost of talent, lighting, makeup, costuming, writing, props, locations, editing, sound, finishing and so on. Costs and time which CANNOT be easily replaced. It makes little sense to save $14,500 on the body/sensor (and much less if just renting either for a couple weeks) with so much else at stake. Especially considering the tremendous differences in image quality demonstrated by this experiment.

-michael zaletel

 

[M Most]

The problem is that it really isn't a comparison between a $2K camera and a $40K camera. The Red One Body is only $17,500 and the 5D2 Body is nearly $3K. The next issue is the cost of all the professional support equipment is nearly the same. Lenses are a major factor and since you can use Canon Lenses on the Red One but can't use all the lenses a Red One supports on a 5D2, that's a wash.

However, the real issue is the cost of talent, lighting, makeup, costuming, writing, props, locations, editing, sound, finishing and so on. Costs and time which CANNOT be easily replaced. It makes little sense to save $14,500 on the body/sensor (and much less if just renting either for a couple weeks) with so much else at stake. Especially considering the tremendous differences in image quality demonstrated by this experiment.

Once again, you're characterizing this as a one-or-the-other choice. That's not the point. There are many situations in which a device like the Canon cameras, with their very low unit cost but reasonably-good-in-many-circumstances image performance can be extremely useful. Getting an extra body for specialty shots, for instance. How about grabbing available light night exteriors? How about body mounting a camera for unique POV shots? How about rigging multiple cameras to a vehicle in less time than you can rig one or two fully tricked out Reds? How about grabbing some quick shots guerilla style for stock or cutaways without permitting because the police will likely think you're a tourist taking stills? How about the ability to build a multicamera rig - maybe a 9 camera 360 degree rig - for shooting virtual location VFX plates and using the still capabilities to get extremely high quality images for areas of the shots that don't have anything that's moving - and all for less than the cost of a single fully equipped Red One? If one is looking at this as "A" camera vs. "A" camera, one is missing the many things that Canon's inclusion of video capabilities in an extremely high quality still camera is bringing to the table.

And with that, I think I've said all I can say about this subject...

 

[michael zaletel]

Every tool is useful M. Lots of movies have used HDV footage cut in with film and I'm sure the 5D2 will fill that role well. Just don't kid yourself that you're getting a "good enough" image that will match in post or look as good even at 1080p. The difference is night and day and the only way 5D2 footage looks good is when you haven't shot the same scene professionally with a Red One. And to be fair, sometimes Red One footage only looks good because it isn't being compared to the same scene shot professionally on film. Mysterium-X should solve that last bit but for me the Red One is "good enough". The 5D2 isn't even close.

-michael zaletel

 

[Mark Pugh]

...maybe a 9 camera 360 degree rig - for shooting virtual location VFX plates and using the still capabilities to get extremely high quality images for areas of the shots that don't have anything that's moving -
...

But none of what your describing here has anything to do with what the test was designed for - pure image quality comparisons, side-by-side of moving images in tricky, illustrative lighting conditions.

A quality stills camera may find good use on a movie set. No tests required there.

You've just seen a 0 to 60mph performance comparison between a Ferrari and a Hyundai and and you're saying that the Hyundai does a great job, cheaply, of driving to get the groceries from the supermarket. Very true, but not very revealing, and nothing to do with the test criteria.

No offense intended to Hyundai, or Canon.

 

[jimhare]

I think the test is fantastic, and I'm really glad to have seen it.

I agree with everyone. MMOST, the Canon footage is pretty impressive considering where it comes from and the price.

What I think frustrates the Red Owners is that certain members post about the 5D or 7D being "as good" as the Red as opposed to "good for the price." The Red boys will always be fast to point out the shortfalls. Human nature and should be expected when posted on Red's home base.

Kind of funny and more than a little ironic as this is pretty much the same thing that happened when Red came out, and everyone wanted to compare it to a Genesis.

The Genesis guys would say "ours is still better" to which the Red guys would say "Doesn't matter, yours costs 10X more than ours and ours is good enough..."

Sounds familiar doesn't it? :wink5:

As for me, I own a Red, and would love to have another body to compliment it. The 7D looks very tempting, particularly if Scarlet is still some months away...

 

[Frank Weeks]

Getting an extra body for specialty shots, for instance. How about grabbing available light night exteriors? How about body mounting a camera for unique POV shots? How about rigging multiple cameras to a vehicle in less time than you can rig one or two fully tricked out Reds? How about grabbing some quick shots guerilla style for stock or cutaways without permitting because the police will likely think you're a tourist taking stills?

These are all valid points of interest for me Mike. As someone who usually works alone or with a small crew, having the small camera option is desirable under certain conditions. I have mixed R1 and 5d footage without problems on a couple occasions.

Thanks for your input.

 

[Larry Gebhardt]

Mike...I have to jump in here and say that I totally get what you are saying, as I have been mixing 16mm shots in 35mm shoots forever...this is the same thing here. You use a tool to get a particular shot...if it is small and light, you can get shots that you simply can't get otherwise and your story is visually richer for it. Yes, you can pixel peep (grain peep) the inserted shot and shriek in horror that it doesn't quite match the quality of the RED (the 35mm) shot, but I like to say, the audience will probably not notice the difference if they are ducking for cover...

My two cents.

Thanks for your input.

Larry

 

[Sidney L. Plaut]

MMost - If the test was for the purpose of judging whether it is possible to intercut RED and CANON 5d, I think It is a very clear answer - yes you can intercut the two, and not a single normal person would notice. That is great news!

having a mobile small camera coupled with the RED - you can get shots that would be impossible to get - unless you have a big budget and lots of time.

Some Projects the RED is the best other projects its something else... but hey... dont discuss Religion, politics or Cameras, cuz people will never agree...

 

[joakim_hansson]

I think the whole issue of actually comparing Canons cameras to RED one footage with good lenses is ridiculous, what I do find interesting though is to compare the footage to SI2k for instance. I was seriously considering buying an Ikonoskop DII, but when it was delayed over and over again I took my deposit for it back.
I´ve now ordered a 1DmkIV to use as a second camera for documentaries and short movies alongside with Panasonic P2 based cameras.
I believe it will be a great tool for natural light shots and for situations where you can´t run around with a big shouldercam.

But to compare it with RED or Arri digital footage is kind of comparing a Harley Fat Boy to a Ducati superbike on a racetrack..

 

[Pietro Impagliazzo]

RED is too sharp! It really need some nicely soft lenses for a organic look.

Unless I was shooting for TV, I'm not sure I'd dig this look very often.

 

[Hrvoje Simic]

What I find odd is the type of shots used to determine whether both cameras could be intercut. It is known that DSLR worst drawbacks are aliasing, picture distortion and compression and these shots do not address these issues properly, IMO. It looks more as if this was all shot to check if one can get away with it if one is really careful. With that approach, it is impossible to get the realistic perception of how the footage intercuts in general.

When thinking about mixing footage from these two cameras it’s very important to mention grading challenge for colorists. When a look is decided on and locked it may be very tricky to match the look between Red and D7 footage because D7’s material will fall apart sooner. If going with intercutting these two cameras the look will have to be created on D7 clips and Red footage will have to be graded based on what Canon’s can stand, wasting the advantage of Redcode Raw and limiting grading on the whole scene.

Even on these, in my opinion very modest tests, to me 7D texture looked pastel-ish/gummy and unnatural as on all other prior seen footage. Aliasing is horrible plus I also don’t like the colour accuracy. Both the pixels and colour to me have a weird "fake" look.

The answer whether the footage can be intercut in my opinion is neither firm yes or no, though I’d say more “no” then “yes”. If shot like this, with static shots or very slow movements and with low detailed scenes, and provided that the story is engaging enough - it could be but it’s a far stretch. I still see much more "cons" then "pros".

It's really funny how Red when it came out was under magnifying glass, scrutinized and compared to high end cameras, and standards then were, or seem to be, forgotten with recent constant paroles saying "DSLR footage is exceptionally good for the price".

 

[Hrvoje Simic]

There are many situations in which a device like the Canon cameras, with their very low unit cost but reasonably-good-in-many-circumstances image performance can be extremely useful.

Of course. No one argues that.

Getting an extra body for specialty shots, for instance.

Agreed. In static shot or very slow movement with as low detail in the scene as possible.

How about grabbing available light night exteriors?

Agreed. In static shot or very slow movement with as low detail in the scene as possible.

How about body mounting a camera for unique POV shots?

That would be a horrible disaster, IMO.

How about rigging multiple cameras to a vehicle in less time than you can rig one or two fully tricked out Reds?

That would also be a horrible disaster, IMO.

How about grabbing some quick shots guerilla style for stock or cutaways without permitting because the police will likely think you're a tourist taking stills?

If that was the only option to grab the shots - it could work but has potential to create issues with grading and matching.

How about the ability to build a multicamera rig - maybe a 9 camera 360 degree rig - for shooting virtual location VFX plates and using the still capabilities to get extremely high quality images for areas of the shots that don't have anything that's moving - and all for less than the cost of a single fully equipped Red One?
If one is looking at this as "A" camera vs. "A" camera, one is missing the many things that Canon's inclusion of video capabilities in an extremely high quality still camera is bringing to the table.

That too makes sense but as I interpreted it this test was done to check how both cameras mix.

 

[Tom Lowe]

I notice that it is no longer possible to make comments about direct image-quality comparisons without eliciting 5 pages of lectures about "horses for courses" and "right tool for the job." I thought all of us here already understood horses-for-courses and right-tool-for-the job?

It reminds of when early R1 adopters were subjected to endless lectures about how "a new 4K camera will not improve your script or actors." Yeah, we get it. We got it the first 100 times we heard it.

Some of us want to be able to discuss the minutiae of image quality without being told it is not a proper subject to discuss.

 

[michael zaletel]

I notice that it is no longer possible to make comments about direct image-quality comparisons without eliciting 5 pages of lectures about "horses for courses" and "right tool for the job." I thought all of us here already understood horses-for-courses and right-tool-for-the job?

It reminds of when early R1 adopters were subjected to endless lectures about how "a new 4K camera will not improve your script or actors." Yeah, we get it. We got it the first 100 times we heard it.

Some of us want to be able to discuss the minutiae of image quality without being told it is not a proper subject to discuss.

Amen Brother!

-michael zaletel

 

[Andrew Walker]

I notice that it is no longer possible to make comments about direct image-quality comparisons without eliciting 5 pages of lectures about "horses for courses" and "right tool for the job." I thought all of us here already understood horses-for-courses and right-tool-for-the job?

It reminds of when early R1 adopters were subjected to endless lectures about how "a new 4K camera will not improve your script or actors." Yeah, we get it. We got it the first 100 times we heard it.

Some of us want to be able to discuss the minutiae of image quality without being told it is not a proper subject to discuss.

That's true that the RED One won't improve your script or actors. But it does make your stock footage look like it was shot on 35mm. I mean who wants a soft timelapse shot or a wide shot of some vista. Personally I like it see everything as sharp as it can be. Timelapse shot on the 5D2 in RAW looks amazing and sharp as a tack. But as I've said before an HVX with the 35mm adaptor on it will give you the same results as the video mode on the 5D2.

 

[Bryce O]

This discussion is stupid. Does 7D/5D footage "look" like Red One footage? Yes, if you disagree you are delusional. Is the the Red One sharper and better for grading? Yes, it's 4k RAW vs 2k compressed what did you expect. Aesthetically, they all look the same, 35mm format with decent dynamic range.

 

[Fredrik Callinggard]

Just shot an international car campaign with an ARRI 435, 235 and 3. 5D's. Very happy with all of the footage and some of it wouldn't have been possible if it wasn't for the 5D's.

 

[Andrew Walker]

This discussion is stupid. Does 7D/5D footage "look" like Red One footage? Yes, if you disagree you are delusional. Is the the Red One sharper and better for grading? Yes, it's 4k RAW vs 2k compressed what did you expect. Aesthetically, they all look the same, 35mm format with decent dynamic range.

Yeah and I remember hearing that the DVX looked like 16mm. But is it even close...no. BTW the 5D2 and the 7D are 1920x1080 not 2K.

Just shot an international car campaign with an ARRI 435, 235 and 3. 5D's. Very happy with all of the footage and some of it wouldn't have been possible if it wasn't for the 5D's.

That is true...and here's a picture to prove it. I found it interesting that you were using a 5D2 and an Arri film camera on the same rig. What was the reason you weren't using your RED cameras you have. I never got to ask.

 

[Gary Ploj]

I'm just glad you did a test for all to see and talk.

thanks.

 

[Graeme Nattress]

Yeah and I remember hearing that the DVX looked like 16mm. But is it even close...no. BTW the 5D2 and the 7D are 1920x1080 not 2K.

They may store their video in such a dimensioned video, but they measure significantly less... And the codec itself robs more detail from the image.

Graeme