5D video analysis ...

I had a look at the 5d MKII video sample from dpreview ( http://www.dpreview.co​m …noneos5dmarkII/​page15.asp ) and I noticed that at about 3 sec in the first video, and about 4 sec in the second one, there is a frame (or more, I cannot say) mostly black, white and with pink areas.

has anyone else noticed this or is it my computer playing tricks on me again?

Yeah, I noticed that. It looks to me like the red channel is dropping out temporarily or something. I haven't noticed this in any other samples, so who knows what's up.

I didn't see that, but you should realize that these videos are very demanding on your system -- the 5D videos all tend to hiccup a lot on my "old" (~2 years) computer.

Those of us who will eventually enter these waters will likely need bigger boats!

tonylong wrote in post #6474770
I didn't see that, but you should realize that these videos are very demanding on your system -- the 5D videos all tend to hiccup a lot on my "old" (~2 years) computer.

Those of us who will eventually enter these waters will likely need bigger boats!

Interesting that you didn't see it. I'm not sure if it would be an issue of system power, since you can select/pause on the affected frame(s).
My system seems to play the vids relatively seamlessly. I'm on a 3ghz C2D, 4gig RAM, and a low/midrange vid card...not exactly a behemoth, but decently capable.

I am pretty sure that is a codec issue, not an issue with the 5d's video capturing capabilities.

sethmo wrote in post #6474945
I am pretty sure that is a codec issue, not an issue with the 5d's video capturing capabilities.

I doubt it is a codec or memory issue. the thing occurs on the same frame(s) all the time regardless of the media player used.

I don't know if this is correct but here is a link to the dpreview post (I know ...) on that issue: http://forums.dpreview​.com …rum=1032&messag​e=29647296

One members also says that there are faulty frames in the Laforet videos too. Are those videos samples (the one from dpreview) supposed to be from a final production camera?

petiot wrote in post #6475660
I doubt it is a codec or memory issue. the thing occurs on the same frame(s) all the time regardless of the media player used.

I don't know if this is correct but here is a link to the dpreview post (I know ...) on that issue: http://forums.dpreview​.com …rum=1032&messag​e=29647296

One members also says that there are faulty frames in the Laforet videos too. Are those videos samples (the one from dpreview) supposed to be from a final production camera?

Nope. The only stuff that's been out there have been pre-production models.

bacchanal wrote in post #6474918
Interesting that you didn't see it. I'm not sure if it would be an issue of system power, since you can select/pause on the affected frame(s).
My system seems to play the vids relatively seamlessly. I'm on a 3ghz C2D, 4gig RAM, and a low/midrange vid card...not exactly a behemoth, but decently capable.

Hey, I saw it, on the first one, a bit after three seconds! Wierd! The guy looks like he's getting ready for Halloween!

Edit: I just saw it on the second one in the 4th second -- someone ought to use a good editor to get the frame numbers and forward this to Canon!

It's frame #95 in the first clip and frame #125 in the second.
(If you save the files to your drive and play in QuickTime, you can click the timecode dispay to the left of the timeline to select "frame number" from the pop-up menu)

These are not codec issues ... they're part of the written file.

The Norwegian Broadcasting Corporation has done a very thorough analysis of 5dmkii footage. Here's the link:

http://nrkbeta.no …s-analyzed/#comment-58420