Magic Lantern enables 14-bit RAW CinemaDNG on 5D

Um, no. Not seeing any aliasing on the raw Mk3 vids. Just the FPN on a couple out of the hundreds that now exist, notably that of James a couple of days ago. But again, he's using a camera with the OLPF removed, possibly affecting things.

Watch the guy's pants/kimono after 1:38.
https://vimeo.com/66170436
The fact that it is far less objectionable than the BMC doesn't mean it is not there.

There is also noticeable aliasing on straight contrasty lines.


edit: And turns out Kholi already mentioned this exact video.
 
Last edited:
* On the aliasing debate: I think the 5D3-RAW has about the same aliasing/moire issues as the D5200, which, in practical terms, for me, means "none". I've seen aliasing in a D5200 test clip shot by macgregor, and horrible FPN in the shadows of his comparison 5D3-H264 test clip. But that was on my very nice 52" Sony TV. When I move to the 32" Samsung TV that I have on my editing system, I can't see the aliasing in the D5200 clip, or the FPN on the 5D3-H264 clip; everything is slightly muddled, and these issues are so small that they disappear. My guess is that Kholi has a much better screen than what I'm using, and that my average viewer has a much worse screen than the worst one I've ever had (a screen in which the rainbow moire of the 5D2 will be painful to watch, but in which the 5D3-RAW aliasing will go completely unnoticed). I'm not even sure your average movie theatre will show these issues.

I'd argue that the vast majority of projects shot on DSLRs end up being viewed on the web on computer displays (or tablets). These are much cleaner than TVs and any aliasing artifacts are much easier to spot.
 
As there is no footage monitoring I was wondering how easy it is to jump back and forth between RAW and all-I or AVCHD recorded on the SD card to review footage. The chief reasons I like to review are to see if my overcrank is working in the given scene or to check for flickering if I'm using practicals. Otherwise, I don't review all that much.

Also, is 48p available and at what dimensions? I find that more useful than 60p
 
Last edited:
The fact that it is far less objectionable than the BMC may end up meaning a lot to many. Having lived with the 5d2 for many projects of the past few years, it may just look like practically nothing to my abused eyes!

Watch the guy's pants/kimono after 1:38.
https://vimeo.com/66170436
The fact that it is far less objectionable than the BMC doesn't mean it is not there.

There is also noticeable aliasing on straight contrasty lines.


edit: And turns out Kholi already mentioned this exact video.
 
I think some of the excitement is boiling to hyperbole with this camera. Is it a really nice image? Yep. Does it look miles beyond other good camera footage. Nope. The fact that it looks so much better than regular 5DMK3 footage is causing delusions of grandeur.

An email hit my inbox this morning, friend on the east coast linking me to the latest EOSHD blog post asking about the 5D Hack... "He said it looks as good as F35 and better than C300. Should I buy an MK3?"

Well... I don't even know what to say. But, it immediately reminded me of this post.
 
The main cameras being worked on seem to be the 5D Mark III, the 5D Mark II, the 600d and I think the Canon 6D also.
The SD card controller appears to be the bottleneck on the 600D/T3i, unfortunately greatly limiting the data rates, and therefore the resolutions, possible. Most unfortunate.
 
Weren't you people all geeking out about removing the OLPF from the 5D Mark III last year? This is why it's there in the first place.
 
Picture is something but the camera rig to shot properlly with the BMC has more contraints than with a 5d III isn'it?

It would have been nice if both shots had the same FOV considering wider shots tend to Moire more than close up ones in my experience. The BMC must have been twice as far back to get a wider FOV than the 5D.
 
Picture is something but the camera rig to shot properlly with the BMC has more contraints than with a 5d III isn'it?

Judging from the pictures of both cameras, they looked to be about the same size. They both could have been framed similarly, they were not. Or, they chose examples that were not similar.

That entire test is flawed from the onset, anyway. But, once it's out there you can't really undo it, so.
 
Judging from the pictures of both cameras, they looked to be about the same size. They both could have been framed similarly, they were not. Or, they chose examples that were not similar.

That entire test is flawed from the onset, anyway. But, once it's out there you can't really undo it, so.

These were my 1st thoughts also. How hard is it to use same framing and same lenses for anything called a 'Test'.

This example shows nothing for me .
 
For the love of God, +1

I don't know if it has been asked or not, but can we merge these 2 threads?
For what? For 90 pages, no thanks : )

Open a sub-forum instead, that's a new camera series!
The first and only HDSLRs to run @24/25fps RAW :P as someone said... ;) a dream of ages!

For less, an upcoming release ;-) last year Landmine Media opened a whole forum dedicated to!
 
Last edited:
Watch the guy's pants/kimono after 1:38.
https://vimeo.com/66170436
The fact that it is far less objectionable than the BMC doesn't mean it is not there.

There is also noticeable aliasing on straight contrasty lines.


edit: And turns out Kholi already mentioned this exact video.
OK, I'll give a fair example... Imagine you have two girls:

the kisses of one come with bad breathe but she is smoker, the other one too but she doesn't smoke...

This is what happens to BMCC user, instead the 5D team supporter's girls are... one of them loves to be around with other males, the other one only for work ;-)
 
Weren't you people all geeking out about removing the OLPF from the 5D Mark III last year? This is why it's there in the first place.


Who can remember that far back?

And as the starter of both threads, my apologies. I didn't think both would blow up simultaneously.
 
Back
Top