Yes, 7.0.3.
The Footage from Leo's shoot...
- Thread starter Ticheli
- Start date
- Status
Yes, 7.0.3.
This is real people. All the more reason to figure out what the deal is. At this stage I CANNOT conclude that the camera is the problem. OS is ruled out. Calibration may play a role...but it CANNOT be the only solution because uncalibrated monitors show AF100 footy with better results...so the hunt continues.
Joe Walker
Well-known member
I'll PM you in the morning Mike, I want to try a little experiment. I'm off to sleep, I've got a long-ass day ahead of me tomorrow.
That's what I have, although it was Pantone Eye-One Display previously. X-Rite bought it from Pantone, but it's still exactly the same. I hear it's not good with glossy displays and not ideal for video (better from photos and printing).
The point was that it's not the pro's opinion in the end that matters. It's the end user or the consumer. If on their "pro" calibrated monitors and such they saw less noise then that's good for them. The point is, is that I'm seeing more noise than my 7D on my monitor at the same time as the AF footage. Calibration is irrelevant when this happens because any calibration or adjustment that would make the noise less apparent in the AF-100 footage would make the 7D footage less noisy as well.
BTW, I'm on a 24" Apple LED Cinema, and yes, it's calibrated for video editing/monitoring.
OK. That's possible. Duly noted. We'll have to grade this after a download (when it's available)...and see the results. I'll try some different workflows and repost once the download is available too. Do you grade? If so...do it and post it, so we can see if that's the issue. Thanks!
It took three hours to download the pro res 422 HQ footage off Joe's ftp site.
We did that to minimize any factors that would affect quality when viewing it.
Both Mac and PC users have access to pro res to view it
How many of those observing of noise have taken that footage and run it out digital (HDMI or DVI, etc) to a plasma and or LCD and still see the noise?
Video quality always suffers from the weakest point in the distribution of it.
That is magnified if there are two or three sequential weak points.
We thought the footage was stunning
We did that to minimize any factors that would affect quality when viewing it.
Both Mac and PC users have access to pro res to view it
How many of those observing of noise have taken that footage and run it out digital (HDMI or DVI, etc) to a plasma and or LCD and still see the noise?
Video quality always suffers from the weakest point in the distribution of it.
That is magnified if there are two or three sequential weak points.
We thought the footage was stunning
I downloaded the mov file and played it back with media player classic ...colors look great ,nice contrast but there's a lot of noise hmmm abit dissapointed BUT
So I waited recalibrated my monitor for the x time .... still the same results.
Then I just played it back with quicktime pro and yes you need to buy this version but geuss what Noise is almost gone , maybe a compression issue, her dress looks clean and so does her hair .
ISO 400 great ISO 800 good.
Now here's the thing I played back 7d files on both players.
And they both looked the same ...slightly more contrast with media player classic but they were of the same quality.
So there's is disconnect !
Now how do we solve this because the end user as in the client/viewer doesnt have the time or maybe not even the knowledge to calibrate theire monitors.
So I waited recalibrated my monitor for the x time .... still the same results.
Then I just played it back with quicktime pro and yes you need to buy this version but geuss what Noise is almost gone , maybe a compression issue, her dress looks clean and so does her hair .
ISO 400 great ISO 800 good.
Now here's the thing I played back 7d files on both players.
And they both looked the same ...slightly more contrast with media player classic but they were of the same quality.
So there's is disconnect !
Now how do we solve this because the end user as in the client/viewer doesnt have the time or maybe not even the knowledge to calibrate theire monitors.
O.K. Mr. Moonay - why don't you go ahead and show us what you've got from that 7D - you spoke of a demo you were putting together - we'd like to see how your stuff stands up to the AF100
pulpfiction007
Veteran
Although this footage could be cleaned up, I still think there's something else wrong, 800 iso should be a lot cleaner to begin with. I think something happened in the chain that must have severely degraded the image. It must be hell for Barry and Joe and Co to KNOW what it is supposed to look like. Sorry about the CAPS by the way!
For reference, here are a couple of samples of Af100 from Japan, that exhibit pretty low noise compared to the studio test. Again, I have viewed all via download played back through PS3 to large plasma.
This one is really some of the most filmic footage I've seen from the Af100. Very low noise in darker or black clothing, mid tones, etc. (I also see no evidence of edge hallowing, or disagreeable sharpening) This clip looks really good even standing right in front of the 50". Graded only for saturation and primary density. Camera setitngs:
Camera settings: Cine-Like D
Details -7
Saturation -3
http://vimeo.com/16737321
This one, though not the best shooting technique in my opinion, has night footage near the end which shows very little noise in the blacks. I believe Cine-D was used for both these examples, but I am not certain. I somewhat doubt this one was graded.
http://vimeo.com/16798262
This one is really some of the most filmic footage I've seen from the Af100. Very low noise in darker or black clothing, mid tones, etc. (I also see no evidence of edge hallowing, or disagreeable sharpening) This clip looks really good even standing right in front of the 50". Graded only for saturation and primary density. Camera setitngs:
Camera settings: Cine-Like D
Details -7
Saturation -3
http://vimeo.com/16737321
This one, though not the best shooting technique in my opinion, has night footage near the end which shows very little noise in the blacks. I believe Cine-D was used for both these examples, but I am not certain. I somewhat doubt this one was graded.
http://vimeo.com/16798262
Compressing it right now...
Anyways while I was editing my dumb little test the answer hit me... and it's so obvious. The AF-100 sensor only has a 3-4MP, if I remember correctly, which is great to reduce aliasing because not much down scaling is being done. However, the 7D and GH2 both have a HUGE mega-pixel count in comparison, 16-18MP. That means that the noise is physically being applied to smaller pixels in the first place... PLUS, it's being down-sampled further, which creates the possibility of aliasing, but decreases noise due to more pixel averaging. Therefore I actually believe it's impossible for this camera to perform BETTER than the GH2 and 7D in high ISO areas. The AF-100 image might be brighter at ISO 800 than a 7D because of bigger pixel size, but there will always be more noise because it's starting with less pixels.
VERY COOL and validated point. You can CLEARLY see the difference in your screen capture. Thanks for sharing!
...back to the Vimeo issue...we know that we are viewing the footage using the SAME player, because there isn't an option to use another player. (That doesn't mean that I will not personally note the differences between Qtime Pro and Qtime classic. So I think we're going to find that this is only the tip of the ice-berg of discoveries as to how we can optimize our "post to web" workflow(s).
Your observation here has to do with black levels, RGB vs 601/709 mismatch can often reveal this. If the blacks get crushed enough, either in a camera or in post or on a monitor, it will hide the noise that's dancing in there. When you bring up the blacks, you also bring up and reveal the noise. The less noise you start out with, the less there is to be revealed. I've dealt with this quite often with the HVX. It can look super clean in a low light situation if the blacks are held down.
Paul Vicente
Well-known member
Barry? Barry? Where are you?
I think you guys have upset this man.
I think you guys have upset this man.
Hey mrmoonay...thanks for sharing. (Honestly) I think what you're going to find...(I could be wrong)...is that these results are NOT indicative of the AF100 performance AT ALL. I think you're going to find that this is a workflow issue and that the AF100 is NOT inferior to the 7d in low-light performance. Phillip Bloom, Barry Green and other professionals have taken their time to evaluate the AF100 performance and they have all used the 7d. So....no offense...thanks for sharing...but I think that you'll need to have a little more faith in these professionals than to think that they ~just forgot to notice~ how the camera performs in low light. THIS VIDEO IS NOT AN INDICATOR OF AF100 LOWLIGHT PERFORMANCE. I may eat those words in 3 months....but I seriously doubt it. Barry knows what he's talking about. So does Phillip.
Steve Kahn
Well-known member
- Status