Welcome to our community

Be a part of something great, join today!

  • 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 :)

Canon 5Dm2 "RAW" files... FAIL

Why would you have two sets of incompatible, yet expensive, lenses? One for stills and one for motion picture; two sets of filters; two sets of recording media, etc...what a waste!

That's certainly my thinking. I want the same lenses for both. And when a guy from a local ad agency who's shot a number of spots on RED with Cine lenses told me his last spot was shot with Nikons and then says the Nikon footage seemed sharper it ends the discussion for me. He didn't care what the name on the lens was, he looked at the footage. Like you should.
 
I went through and extracted some still images, cropped the same as the original post to 100%.

Working with shallow depth of field and dealing with slow shutter speeds, manual focus, and vibrations from both the focusing/fiilming and the environment, I think you need to be a little more selective with which stills really represent the capabilities of the camera.

I'm sure people can find plenty to fault with these frame grabs as well, but to me they are a better representation.

MVI_0795.jpg


MVI_0492.jpg
good one snipeR.. be welcome to our club.. you rock as non-affiliated open-minded poster..
 
mr jukic.. we're all waiting..
 
good one snipeR.. be welcome to our club.. you rock as non-affiliated open-minded poster..
I'm sorry, I thought providing an alternate perspective was the whole concept behind being open minded?

Red is great, I'm a big fan of Mr Jannard's and his products, but I won't be biased in my view of others products--even if it is an 'evil empire' (Canon).

It is true I just registered and my first post is (potentially) a positive post towards Canon, but that is because I felt the OP was some-what misleading.

If this were a conversation about Red One on a Canon forum and someone had posted screen grabs from some sample footage and I found they were not the best representation I'd register there and post my opinion (potentially in favor of Red) as well.

So I'm sorry if you question my intentions, but what can I do?
 
Anyone see the ginormous macroblocks in the water? Or the blocks in the shadows? Lovely... Has anyone plopped them on the FCP timeline, and even on a 8 core, how sluggish it is - no RT - render everything. Tried correcting out the tungsten balance?

Graeme

Yes, indeed. I just downloded the clips and I can see considerable compression artefacts. Look at the nose and forehead of the dude in the heli, it is all clipped to a single colour, no gradation, no shoulder for highlights. Also, no headroom in CC. Try to drop the clips on a time line and do any CC, then look at the histogram and see what you got left - there is no lattitude, none, zero - the colour information is not there to play with.

Good for webcam, tho; Expensive webcam. I'm sure BBC and Discovery will use it :bleh: LOL
 
hey snipeR.. seems to me you're completely out of any light on my mind.. open-minded first time poster.. save up your sorries by now.. learn to know my posting.. how should you know it?.. i'm just a filmmaker.. then we'll speak it out again.. i'll be here to listen your sorries.. in the meantime.. keep going your posting.. don't refrain yourself.. me likes..
 
I went through and extracted some still images, cropped the same as the original post to 100%.

Working with shallow depth of field and dealing with slow shutter speeds, manual focus, and vibrations from both the focusing/fiilming and the environment, I think you need to be a little more selective with which stills really represent the capabilities of the camera.

I'm sure people can find plenty to fault with these frame grabs as well, but to me they are a better representation.

MVI_0795.jpg


MVI_0492.jpg

Unfortunately with this type of compression, the worst frames are sort of the "weakest links." If you have 10 nice frames then 3 jacked-up frames then 6 nice frames, etc etc... you have to judge the footage by the weak links in the chain.

Keep in mind that I have probably been one of THE most enthusiastic people here about the 5Dm2. Just for me, the footage does not hold up, pixel for pixel - which is important... to me. Other people might not care about that. To each his own!

I'm waiting to see a test off the HDMI port. :gun:
 
Keep in mind that I have probably been one of THE most enthusiastic people here about the 5Dm2.
:) ehehehehehe you're applying to the sj-amazing-award this civil year?..
 
sorry, but bashing a simple bonus feature, enabling live-view recording on a prosumer still camera tells one thing only... the redowners are scared shitless
 
My good fellows, we cannot forget these files -- or at least a part of those when my memory doesn't serve for everything, they are ISO 3200 or 6400 (granted in accordance with the shooter he would have shot at 1/60th). So...
 
Woof. Those aren't looking good at all. The Nikon D90 stuff looks fantastic even on auto, how could this technically "superior" movie mode look so bad? Well, unless they add 24p, it won't matter much to me but if they do, I can't wait to see the tests and see if there really is a problem on the production models. Remember, this was a preproduction unit after all.
 
...I can't wait to see the tests and see if there really is a problem on the production models. Remember, this was a preproduction unit after all.

With Canoon, it is generally that the producton units are the biggest concern :)
 
I'm sorry, but I have to step in. I have discovered a problem with quicktime incorrectly handling the native colorspace these clips have. Apparently, during conversion into RGB color, it simply clips the YUV data below 7.5 IRE and 100 IRE. I imported the same clips into a YUV native editor (NewTek SpeedEDIT) and brought the contrast down to correctly map ALL the data into RGB color. It also shows much less macroblocking, although there is some left. This is the result...
 
I dunno what happened. Ok, so this is a bit more compressed than I'd like, but all the artifacts you see are jpeg.
 
Unfortunately with this type of compression, the worst frames are sort of the "weakest links." If you have 10 nice frames then 3 jacked-up frames then 6 nice frames, etc etc... you have to judge the footage by the weak links in the chain.

Keep in mind that I have probably been one of THE most enthusiastic people here about the 5Dm2. Just for me, the footage does not hold up, pixel for pixel - which is important... to me. Other people might not care about that. To each his own!

I'm waiting to see a test off the HDMI port. :gun:

I completely understand your position, I was just trying to say that some of the clips short-comings may have been the film maker's (lack of) experience working with video. Meaning perhaps in the hands of a more seasoned pro that chain may have had fewer (if any) weak links.

Really everything is just speculative right now, once it is in the hands of the consumers we'll have a better idea of this camera's potential.
 
Back
Top