«5D Mark III outperforms Scarlet»

[Philipe Ratton]

The only reason for blowing up the 1080p-2k downscale to 400% is to show that the deliverables (Scarlet 2k downscaled to 1080p vs. 5Dmk3 at "1080p") look about the same. Which makes sense, since anything more than 1080p is NOT in the average household and not realistically available via streaming (yet). Granted, 4k downscaled to 1080p would presumably look much better (which is why it wasn't included)...

I get REDuser being defensive when things like this come up, but you have to look at it for what it is: trying to encourage 5D users by saying they get somewhat comparable results (to varying degrees) with the only camera they can afford (in this case, the 5Dmk3.) There's no point getting bent out of shape over it. Obligatory car analogy: If someone claims on the road their civic and your ferrari go the same max speed, it's technically true (going past the speed limit is illegal, right?) You wouldn't get pissed off, you'd smile/laugh at the irony (but still love the fact that you have a ferrari.) Currently the speed limit is 1080p, and more often than not, compressed 1080p, so the results can be comparable... of course you'd have to ignore 13+ stops, HDRx, and r3d RAW, but ignorance is bliss.

The real reason people should be upset is why this is even posted here... You wouldn't go to an Alexa forum/page to look up RED information, right?

I understand what you mean Mike and this is not an attack to you but 1080p is not the limit when you can finish and deliver in 4k and 2k. Digital cinema camera used to mean something around here.

[Arnold F.]

I wonder if he can test the reason I bought a RED camera: how much work it gets me. It's not worth my time looking at that test because I know the Scarlet will move me faster. Period.

As long as every producer wants to shoot with RED (even if only a fraction of them can say why), it was the right choice for me even if the 5D is 10 times better. I look forward to the time when I have the luxury to choose the camera. Until then it's this one.

[Trevor Meeks]

Real nice test Sebastian/Cinema5D.

What a load of shit.

[Nick Wernham]

Why are people mad?

Transparent play for page views is transparent.

[David Mullen ASC]

Finishing a test to 1080P is fine if that's your intended market, but I don't understand the legitimacy of the test if they are starting out by shooting 2K cropped mode on the Scarlet before conversion to 1080P. Particularly compared to a full-frame sensor image like from the Canon 5D, the depth of field is going to be radically different compared to the 2K area of the Scarlet sensor. To be fair, he should have also cropped the Canon image in half and then converted it to 1080P.

[Gavin Greenwalt]

The 2k test is fatally flawed by suffering from some obscene Chromatic Aberration. A significant amount of the RED resolution is being lost to that.

That being said, I think the 4k -> 1080p -> 4k test vs 1080p -> 4k test isn't a bad one since it actually demonstrates well in that shot how much better the RED is... even as a 4k camera. The 5D MkIII is softer in the first test--showing off why you want to supersample.

In a slightly unrelated comment... it would appear that the 5D has a stronger UV filter.

[Sergio Perez]

This test is ridiculous. No one is going to shoot a Scarlet a 2k if not to get a longer telephoto shot with a lens or to get slow motion at 60fps. If the intent is to shoot slow motion, the comparison should be Scarlet at 2k 60fps vs 5D at 720 60p. To compare the 1080p output of these cameras the test should be Scarlet at its full resolution the. Downscale to 1080p vs the 5D, which is using its full sensor downscaled to 1080p.

Really disappointed with cinema 5d.

[Andy White]

I agree with Sergio... and following this train of thought (full 5DMkIII sensor vs Scarlet full sensor) - wouldn't the HDMI/SDI output via a Pix (or similar) have been a better test?

It's so weird to have these tests - all to crank up hits on the interweb... and really only serving to confuse people.

[Gavin Greenwalt]

EDIT: Ok my apologies to Sebastion I missed the other mp4

[sergio arguello]

this guy sebastien from 5D is plain and simple trying to self promote at any cost...test is benign and he gets moron status for not thinking so