Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Red MX kicks 5D's ass all the way to the moon

Cant wait to get my Epic, and my RED MX'd! Thanks for posting!
 
Try putting both in separate underwater housings and mount them as a helmet cam then go night surfing on a full moon and tell me then what camera kicks the others arse. Good chance that one of them is near on impossible, right?
Haven't we had enough of these comparisons? I bet Canon is loving this attention, being compared to a Red one.
 
very difficult to gather much from this test because of the horrible youtube compression. It's evident that there's a bit more range and resolution on the MX but it's so terribly garbled with all sorts of nasty that it's essentially useless. Try ProRes or H.264
 
Michael Totten said:
very difficult to gather much from this test because of the horrible youtube compression. It's evident that there's a bit more range and resolution on the MX but it's so terribly garbled with all sorts of nasty that it's essentially useless.
Click to expand...

Personally, there's a notable difference even at 360p. Currently downloading the 1080p stream which is reasonably good quality.
 
To be honest unless your audience obsesses over subtle dynamic range cues then the 5D is all the more impressive for the price really.

MX is fantastic, but the 5D isn't bad, not at all.
 
andrewhake said:
To be honest unless your audience obsesses over subtle dynamic range cues then the 5D is all the more impressive for the price really.

MX is fantastic, but the 5D isn't bad, not at all.
Click to expand...

The 5d isn't bad, untill you start doing bad stuff with it. Like mounting it on a car, running with it or feeding it too many critical lines.

Then i goes crap all over.
 
andrewhake said:
To be honest unless your audience obsesses over subtle dynamic range cues then the 5D is all the more impressive for the price really.

MX is fantastic, but the 5D isn't bad, not at all.
Click to expand...


Did you see the 1080 version? I wouldn't say the differences are subtle.
 
I did some ISO testing with M-X and 1d mk III (in stills mode, as it doesn't do video) - lighting my bookshelf as dimly as I could so I could go all the way to ISO6400 on both cameras. It's obvious I wouldn't be posting this unless the M-X looked better, both in terms of noise and overall tonality.

The main problem with the Canon (and it's not just Canons that have this issue) is that they rely on analogue gain for ISO, which means if you're in pure low light you're ok enough, but if there's any kind of extended dynamic range, you're going to clip severely on the top end and eat into the images tonality, ruining skin tones and showing a low dynamic range image. With the M-X in such a situation the DR remains high and constant, giving a fantastic range of values for any colorist to deal with and a very smooth highlight rolloff as-is.

I've seen nighttime 5D2 tests where the skin on actors turned to plastic because it was (necessarily) shot at high ISO, but as soon as the actor got too bright compared to the background, the skin turned to plastic.

And no, often the audience doesn't care. But we as professionals should care.

Graeme
 
I don't think the 5D looks terrible until you see the comparison. Then there is quite a difference....and it's not that subtle. The ECU of his face at 3200 ISO is AMAZING on the Red MX.
 
Graeme Nattress said:
I've seen nighttime 5D2 tests where the skin on actors turned to plastic because it was (necessarily) shot at high ISO, but as soon as the actor got too bright compared to the background, the skin turned to plastic.

And no, often the audience doesn't care. But we as professionals should care.

Graeme
Click to expand...
I'v found plastic skin is caused by a combination of improper white balance and over exposure. In particular, over exposing the red channel will render unacceptable skin tones - unacceptable that is, unless you like the alien complexion look:)

Another problem for the 5D/7D shooters is using variable NDs which leak blue and also contribute to awful skin tones.
 
Food for thought

Food for thought

I agree with Graeme, we as professionals should care about what looks better and why. I think it's fair to say that if anyone was at red at nab or have seen this
http://carlosprio.com/blog/wp-content/uploads/2010/04/Resolution-Chart1.jpg

Then it's clear who leading the game.

It's funny that you say canon should be proud because they are being compared with red, but isn't red the one being compared with Arri.

All cameras have their place whether I'm using a hero pro strapped underneath my Harley, or my iphone to take a quick picture. But I think the main purpose here is to see the EV range for dramatic work, not strapping your epic to your head while surfing at night.

But for dollars/quality no one is even coming close to what Red has accomplished in such a sort time.
 
Graeme Nattress said:
I did some ISO testing with M-X and 1d mk III (in stills mode, as it doesn't do video) - lighting my bookshelf as dimly as I could so I could go all the way to ISO6400 on both cameras. It's obvious I wouldn't be posting this unless the M-X looked better, both in terms of noise and overall tonality.

The main problem with the Canon (and it's not just Canons that have this issue) is that they rely on analogue gain for ISO, which means if you're in pure low light you're ok enough, but if there's any kind of extended dynamic range, you're going to clip severely on the top end and eat into the images tonality, ruining skin tones and showing a low dynamic range image. With the M-X in such a situation the DR remains high and constant, giving a fantastic range of values for any colorist to deal with and a very smooth highlight rolloff as-is.

I've seen nighttime 5D2 tests where the skin on actors turned to plastic because it was (necessarily) shot at high ISO, but as soon as the actor got too bright compared to the background, the skin turned to plastic.

And no, often the audience doesn't care. But we as professionals should care.

Graeme
Click to expand...

I think even drunk 14 year old girls will say that something "looks cheap" "looks crappy" even though they will not be able to say why. This is the reason many European and Canadian documentary-ish-looking indy films do poorly. They just don;t look very pretty, and it feels depressing. You don't need any tech expertise to feel it.
 
Perhaps most revealing was the facial hair at the end. Even on 360p there's a massive difference in detail. In 1080p it is almost embarrassing.

But of course, the audience wouldn't mind were it not intercut with M-X. The 5D still looks decent. Next to M-X, however, it falls apart. I can't imagine how disastrous these DSLR footage will look once we move into 4K video standards and beyond.
 
Why is the Canon so blurry? Line skipping looks aliased, not blurry. Is the compression so bad?
 
Yeah, a lot of compression artifacts in the Canon. No question, to me, the MX looked better. I suspect the question for a lot of users will be whether it's $15,000 better.

I would've preferred to see footage with more horizontal motion, to try to judge how the rolling shutter issues look on each device. BTW, as long as the images aren't clipped, I think good color correction could match the overall looks pretty well (particularly in terms of skintone). But there's no easy way to overcome data-compression problems.
 
I think the MX works better in making a rich detailed and clean image, but I'm worried about color rendition. In the 5D footage the shirt is clearly an apricot - cream tone and the skin is more like brown sugar, while the colder lighting of some backgrounds are blue. In MX material I would say the shirt and the face are almost the same color, and the global image has a little sepia tone. Sure there are parts where the skin tones of the 5D are terrible and the MX are better, but I think the 5D pictures shows more variety of colors. I think the newer color science has been an improvement in skin tones, but I still need to give some saturation in RCX to help define the different colors of the scene. The bad side is that warmer greens get over saturated easily, and there's (still) no saturation curves in RCX. My opinion is that RED images has a wood-like sensation in skin tones, and I know there's work in progress about this issue (FLUT color), but tests like this are useful to improve things.
 
This was a quick test, but it it gives you 1000 times more information than the biased Zacuto static bubble girl tests. We are motion picture filmmakers, henceforth all comparison tests should involve some kind of moving action. Even all charts should be moving. This especially goes to Jim and Jarred, we need Epic moving images not still pictures of Lizards, turtles and prototype cameras. I love y'all, but shoot some action for crying out loud. If you can't shoot send me an Epic-X, I'll shoot some real stuff. LOL. Please don't ban me from Red studios!!!
 
You must log in or register to reply here.
Back
Top