Some footage...please!

He's talking about the lowish quality B-frames where the zooms take place. There are a bunch of them, where the bitrate isn't sufficient to capture the motion. The 1st B-frame after an I or a P deteriorates more than the 2nd. The deterioration is quite evident around the jacket pattern (which looks remeniscent of GH1 mud) , mouth , glasses

I uploaded a couple frames to demonstrate, where the P-frame looks fine, but the lower bitrate B-frame deteriorates (they are marked in the top left corner), recovers somewhat the 2nd B-frame, and cycle repeats... (if this were solely a focus issue, you wouldn't expect the 2nd B-frame to "pop" back into higher quality)

I realize this is just a random handheld shoot, and normally you might control it better, and this is a not 100% finished retail product etc, etc...; but I suspect these are the types of situations where an external recorder might be useful

http://www.mediafire.com/?z2bkl8l9nzopz7j
 
Last edited:
The highlights in the foliage have the typical HVX200/200A/HPX170/500 look to them, that is to say too much chroma, cartoonish/video like. I guess this is because the processing is similar to the other low cost Panasonic cameras. It is very different from an HDX900/HPX2700/3700 highlight look(using film gammas). Not surprising, but not the filmic look of a properly setup higher end Panasonic camera. I wonder what DRS would have looked like on that scene?
 
K first of all, Barry uploaded a very high contrast scene. open shade with lots of hot spots all over the leaves behind the subject.
I just transcoded this using cineform and other than the scene that it was shot under, I think this footage is EXCELLENT!!!!

Just for a second, imagine if this scene was shot with ANY Dslr out there? That jacket and the pants would have never held the way it did here. The glasses would have aliased all day with the HDSLR. In fact, most of this shot would have fallen apart from the HDSLR.

The AF100 held up REALLY nice from what I've seen in FCP. I mean come on, with a HDSLR you would have asked that guy to remove that jacket as it would be a nightmare to deal with.

I even graded this shot using some filters I usually give to my work and all I can say is WOW! I pushed the grade pretty good and I saw VERY LITTLE loss of gradation in his skin, where on my 5D it would have fell apart. I was shocked!

Now, this is still the pre-production? COme on guys. I'm viewing this on my 100 in. veiwing wall, super graded, and all I can say is I WANT THIS CAMERA! Mud? That tiny little bit at 14:01? Really? 99.9999999% of NORMAL viewers would say you're nuts! But they would surely be turned off by that jacket if filmed on an HDSLR.

Why all the HDSLR comparison. Too me, that what this camera addresses. All the faults of a large chip recorder with the price below $5000. THis camera cost the same as a Canon 1Dmk4. Think about what this camera will do with an external recorder? Will it improve in the final production? I'm sure it will. But this clip, in my NLE, passes my test for wanting one real bad. To me, this looks like it will be my HVX200 with a nice DOF and no grain. Easy transition....
 
I think it looks great.

If you are going to blow up the image 500% and start pixel peeping, then you are better off buying a RED.
 
mcgeedigital said:
I think it looks great.

If you are going to blow up the image 500% and start pixel peeping, then you are better off buying a RED.
Click to expand...

+1,000,000 for real. I'm really liking what I'm seeing here and it's not really a great shot (sorry Barry). But it proves that it's handling difficult texture! I mean, even that darn little rope along his hat looks great, something the 5d would have destroyed!
 
If that shot had been something for a real job it would have been ruined due to the artefact. It's brief but never the less unacceptable to any of my producers.
 
editman said:
If that shot had been something for a real job it would have been ruined due to the artefact. It's brief but never the less unacceptable to any of my producers.
Click to expand...
If your producers are that picky, why would you be shooting with a $5,000 video camera?

Edit:
and I don't believe that what you are referring to in the clip is mud.
It looks like something caused by the shaky zoom or an autofocus freakout.
 
Last edited:
to the couple people that siad they are impressed by the dynamic range shown in this clip of the man drawing in the park. can you please tell me how so? To me all the leaves in the background look blown up just as they would on any other DSLR. what is so impressive about it?
 
holyzoo said:
"Unsupported Video File: The video contains an unsupported audio codec." Not sure what that is all about. Software on the AF100?
Click to expand...
It's because the AF100 uses uncompressed audio, but ... wait -- I think this was a slightly overcranked shot, which means there shouldn't be any audio at all? Either way, it might explain it, Voltaic might need to be updated because both those aspects are new for AVCCAM.

That issue aside, the resolution looks plenty to me, I couldn't see any aliasing, and considering this is a fairly high contrast shot, the latitude looks to be pretty empowered.
Click to expand...
Resolution is excellent, and I toned down the detail to -4, which to me appears to be the point at which you can't tell any noticeable edge enhancement, so it has pretty minimal edge enhancement. -4 on the scale of +7 to -7. Latitude is good, but I can't measure it here to see how it compares to the 10.5 stops I got on the first unit.
 
This problem is blown out of propprtions

Whats the settings barry? Its a bit too saturated for me. Im guessing this isnt cine d?
 
Jesse Brauning said:
I honestly see zero mud in those images.
Maybe some motion blur, but not mud.
Click to expand...

It's not solely due to motion blur or focus, or you wouldn't get that pattern of suddenly poping out/in of focus or suddenly no motion blur on the good frames - the motion is fairly constant in that sequence. Step through the frames and you will see.

This is primarily a problem with inadequate compression for this shot.





You don't have to blow anything up to see the B-frame issue . You can see it at 1:1. I attached a 1:1 crop

Overall I think it looks ok, but people have to make their own choices about whether it meets their needs. Better than ignoring the truth.

We don't even know if it's a problem. This is an unfinished unit, correct?



Uploaded with ImageShack.us
 
barry any chance of a/b shots with something like GH1 (or gh2) or ANY other cam just so we can see side by side the same footage and see how dynamic range and such compares on af100?
 
I feel sorry for the Japanese guy being put under a microscope, how weird would that be for him to find this thread :)
 
PDR said:
This is primarily a problem with inadequate compression for this shot.


You don't have to blow anything up to see the B-frame issue . You can see it at 1:1. I attached a 1:1 crop
Click to expand...

I wish I couldn't see it but I do. It almost looks like the lens jitters while zooming. Maybe that's what it is? But it is there :huh:

We don't even know if it's a problem. This is an unfinished unit, correct?
Click to expand...
I think they said it's 95% done. That sounds pretty finished to me...
 
You must log in or register to reply here.
Back
Top