Stephen Mick
Totally Usable
I didn't clock the transcoding, but on my i7 MacBook Pro, it absolutely flew.
AF100 noise
- Thread starter Steve Kahn
- Start date
Jan_Crittenden
Veteran
There is also a difference between the two shoots if for no other reason than they used different gamma curves. Leo used HD Norm and I believe Timurs group used Cine-Like D.
You can see a difference between these on a monitor.
Best,
Jan
You can see a difference between these on a monitor.
Best,
Jan
Barry_Green
Moderator
No need, it works as it's supposed to. The white balance button is also used for black balance, you just have to hold it down longer.
i just didint know you had to hold it down for longer than a second. the HVX was very quick to activate, i stupidly assume it was the same
Rick Burnett
Section Moderator
Given I am #2 on that same list, I too have been following the different sample videos as well. I think there is a lot of truth to the curve you are using playing a significant role in what happens to the noise in the blacks. I graded some of the noisy clips I could get and was easily able to tame the noise down, and the results looked awesome. I too use the same monitor that I edit ALL my footage on and I did in fact calibrate it at one time, so I am very familiar with what to expect on it.
lawrenceingram
Active member
Just as a demonstration, I picked up my HVX200 and shot a test clip before black balance and one after. Here are the results:
lawrenceingram
Active member
Oh, and as I understand it, the changes are exacerbated by extreme temperature changes (as is the case right now where I am and would explain the big difference after I black balanced).
Lawrence
Lawrence
Barry_Green
Moderator
I've got two laptops with different black levels. On one, there's visible noise in the blacks on some AF100 footage. On the other laptop, the exact same footage file plays clean as a whistle -- which is also how it looks on my TV, on my production monitor, on Stephen's television and on his laptop, on Matt's laptop, on Joe Walker's editing suite, and on nearly every other screen it's been shown on.
It's a problem with the laptop screen. It's showing things it shouldn't be showing. It's like if you went to someone's immaculate house and pried up the floorboards and found roaches down there -- yes, if you go looking for them, you'll find them. But if you just look at their house, it looks immaculate. Same thing with AF100 footage. There's a little noise in the darks, like there is on pretty much every video camera ever made. But some monitors do a serious black stretch on their gamma, and just exaggerate and firehose the appearance of noise, and that's just not right. It's viewing the footage wrong. What noise there is, blends nicely into near invisibility on a properly-set-up, proper-gamma monitor. If your monitor isn't using video gamma and is instead stretching the blacks out, you'll see a crawling mess of noise.
I know some people aren't going to like this answer, but it's the actual truth, so please consider it carefully before rejecting it.
On the laptop that can show the noise, I just did a little experiment -- I put up color bars in Premiere. And yeah, the 11.5 IRE bar is glowing like a lighthouse, it's so much brighter than the black that it's not even funny. And there's one particular AF100 clip I've got that, when I show it like that, ooh -- it's noisy noisy noisy. But if I tilt my screen so that the color bars look correct and the 11.5 IRE slight gray bar is just barely visible against the black (like it's supposed to be) then guess what -- all the noise disappears, and the shot looks like a million bucks.
I just ran the Windows 7 "display calibration" tool, and guess what -- the gamma was way off on this monitor. Windows has a tool for setting the gamma using some circles with dots in them, and when you tune that so that the central dots disappear (i.e., blend in with what's around them) then everything gets so much better -- now the Premiere Color Bars show up properly, and now the AF100 footage looks brilliant. If I maladjust the monitor by raising the gamma so that I can clearly see the dots in the circles, then the color bar 11.5 IRE band starts to glow, and yes there becomes lots of noise in the blacks on some of the AF100 footage.
But it's noise you should never see because on a proper-gamma properly-calibrated display, it will never, ever be visible.
And now that I've properly calibrated this monitor, the footage looks immaculate on it too.
So for those of you wondering, look at this picture of color bars. Look in the lower right. Do you see the gray stripe? If you can just barely detect it, to where it's just barely slightly visible against the black on both sides of it, then your monitor's gamma is probably about right. If you can clearly and easily discern the gray stripe from the black around it, then your monitor's gamma is set wrong for video, so don't be surprised if you see things in the video that aren't right.
In summary: I can make the noise show up if I force the monitor to display it. But if I set my monitor up correctly, the noise disappears.
It's a problem with the laptop screen. It's showing things it shouldn't be showing. It's like if you went to someone's immaculate house and pried up the floorboards and found roaches down there -- yes, if you go looking for them, you'll find them. But if you just look at their house, it looks immaculate. Same thing with AF100 footage. There's a little noise in the darks, like there is on pretty much every video camera ever made. But some monitors do a serious black stretch on their gamma, and just exaggerate and firehose the appearance of noise, and that's just not right. It's viewing the footage wrong. What noise there is, blends nicely into near invisibility on a properly-set-up, proper-gamma monitor. If your monitor isn't using video gamma and is instead stretching the blacks out, you'll see a crawling mess of noise.
I know some people aren't going to like this answer, but it's the actual truth, so please consider it carefully before rejecting it.
On the laptop that can show the noise, I just did a little experiment -- I put up color bars in Premiere. And yeah, the 11.5 IRE bar is glowing like a lighthouse, it's so much brighter than the black that it's not even funny. And there's one particular AF100 clip I've got that, when I show it like that, ooh -- it's noisy noisy noisy. But if I tilt my screen so that the color bars look correct and the 11.5 IRE slight gray bar is just barely visible against the black (like it's supposed to be) then guess what -- all the noise disappears, and the shot looks like a million bucks.
I just ran the Windows 7 "display calibration" tool, and guess what -- the gamma was way off on this monitor. Windows has a tool for setting the gamma using some circles with dots in them, and when you tune that so that the central dots disappear (i.e., blend in with what's around them) then everything gets so much better -- now the Premiere Color Bars show up properly, and now the AF100 footage looks brilliant. If I maladjust the monitor by raising the gamma so that I can clearly see the dots in the circles, then the color bar 11.5 IRE band starts to glow, and yes there becomes lots of noise in the blacks on some of the AF100 footage.
But it's noise you should never see because on a proper-gamma properly-calibrated display, it will never, ever be visible.
And now that I've properly calibrated this monitor, the footage looks immaculate on it too.
So for those of you wondering, look at this picture of color bars. Look in the lower right. Do you see the gray stripe? If you can just barely detect it, to where it's just barely slightly visible against the black on both sides of it, then your monitor's gamma is probably about right. If you can clearly and easily discern the gray stripe from the black around it, then your monitor's gamma is set wrong for video, so don't be surprised if you see things in the video that aren't right.
In summary: I can make the noise show up if I force the monitor to display it. But if I set my monitor up correctly, the noise disappears.
Last edited:
Rick Burnett
Section Moderator
I keep hearing this excuse of the monitors not being set up properly, but it's not like people are just watching video for the first time on their screens. And it's not like EVERY PERSON in the world who looks at your clip that you post places, like trailers and such that end up being watched online many times are going to care for ONE SECOND what excuse you come up with why the noise in your clips are dancing all over the place.
The FACT is that many people are comparing these posted movies to what they already know and to other cameras that they may own. I've compared the footage to what cameras I use and the grading that was done is VERY noisy on the same monitors that i use over and over again to look at things. Sure, on my calibrated monitors the noise is almost non-existent and looks great. But, on other monitors, that are not calibrated, that are at the same factory settings that came on the monitor, the noise is VERY noticeable. Compared to other clips shot say on my 7D with similar settings, the noise is VERY MUCH lower.
HOWEVER, I attribute this like I said to the curves being used in camera as I took the footage and made it look MUCH better on the LCD with the wrong gamma, and it looked just as good on the properly calibrated screen.
Someone who color corrects footage has got to look at ALL the situations that your movie is going to be watched, including home theaters that vary incredibly in their curves as well. This is just plain common sense to me. It's no different in the audio world. I mix on VERY expensive audio speakers, but I also have a pair of speakers that simulate television speakers in plastic housing (Triple-P monitors) so that I can hear what my audio sounds like on the crappiest of circumstances, because it needs to sound good in both, and properly mixed audio will.
I think what the original poster is asking, is given the noise he sees, how can he reduce it so it looks good in all circumstances. I think, however, that it's a function of what curves you are using in camera as well, which makes it a bit more complex of a question.
The FACT is that many people are comparing these posted movies to what they already know and to other cameras that they may own. I've compared the footage to what cameras I use and the grading that was done is VERY noisy on the same monitors that i use over and over again to look at things. Sure, on my calibrated monitors the noise is almost non-existent and looks great. But, on other monitors, that are not calibrated, that are at the same factory settings that came on the monitor, the noise is VERY noticeable. Compared to other clips shot say on my 7D with similar settings, the noise is VERY MUCH lower.
HOWEVER, I attribute this like I said to the curves being used in camera as I took the footage and made it look MUCH better on the LCD with the wrong gamma, and it looked just as good on the properly calibrated screen.
Someone who color corrects footage has got to look at ALL the situations that your movie is going to be watched, including home theaters that vary incredibly in their curves as well. This is just plain common sense to me. It's no different in the audio world. I mix on VERY expensive audio speakers, but I also have a pair of speakers that simulate television speakers in plastic housing (Triple-P monitors) so that I can hear what my audio sounds like on the crappiest of circumstances, because it needs to sound good in both, and properly mixed audio will.
I think what the original poster is asking, is given the noise he sees, how can he reduce it so it looks good in all circumstances. I think, however, that it's a function of what curves you are using in camera as well, which makes it a bit more complex of a question.
Barry_Green
Moderator
You can call it an "excuse" if you want, but ... look. I have two monitors here, one that shows the noise and one that doesn't, on the exact same footage. And I tested the calibration of the monitor that shows the noise, and it was off. And I calibrated it using the Windows calibration tool, and the problem disappeared. The difference is massive.
I would call that a pretty definitive examination of the problem, identification of the actual root cause of the problem, and a fix that eliminates it. I don't call that "an excuse".
If the advice was "just crush the blacks on your monitor to hide the noise" then I'd be right there with you, saying that's totally inappropriate. But that's not what I'm saying. I'm saying to set up your monitor so that it PROPERLY displays the video signal the way it was meant to be seen. And when you do that, you have a case of the camera working the way it was designed, delivering footage to a monitor that's displaying it the way it was intended to be displayed. That is a solution, not an "excuse."
People can use whatever they want, or whatever they've got, but if they're using a monitor that is artificially stretching the blacks, nothing is going to look right. That's just an absolutely inescapable fact. And if you want to dumb down your footage to go chasing after every conceivable misalignment that someone is possibly capable of inflicting on a display device, then ... well, there is no way to do that and preserve the footage's integrity.
If you want to run Neat Video's noise reduction on the blacks that are under 7.5 IRE, I guess you could do that; nobody should see anything under 7.5 anyway (as it's all supposed to be black, and on a properly calibrated monitor it will show as black, and any noise that's in there will be invisible). If you want to show your footage on a monitor that stretches out and exaggerates the range under 10 IRE, then yes you're going to see noise there.
So I guess you have two potential solutions -- either calibrate your monitor properly, or run noise reduction software so that your footage will look noiseless when it's displayed on a monitor that's showing it how it was never intended to be displayed.
I would call that a pretty definitive examination of the problem, identification of the actual root cause of the problem, and a fix that eliminates it. I don't call that "an excuse".
If the advice was "just crush the blacks on your monitor to hide the noise" then I'd be right there with you, saying that's totally inappropriate. But that's not what I'm saying. I'm saying to set up your monitor so that it PROPERLY displays the video signal the way it was meant to be seen. And when you do that, you have a case of the camera working the way it was designed, delivering footage to a monitor that's displaying it the way it was intended to be displayed. That is a solution, not an "excuse."
People can use whatever they want, or whatever they've got, but if they're using a monitor that is artificially stretching the blacks, nothing is going to look right. That's just an absolutely inescapable fact. And if you want to dumb down your footage to go chasing after every conceivable misalignment that someone is possibly capable of inflicting on a display device, then ... well, there is no way to do that and preserve the footage's integrity.
If you want to run Neat Video's noise reduction on the blacks that are under 7.5 IRE, I guess you could do that; nobody should see anything under 7.5 anyway (as it's all supposed to be black, and on a properly calibrated monitor it will show as black, and any noise that's in there will be invisible). If you want to show your footage on a monitor that stretches out and exaggerates the range under 10 IRE, then yes you're going to see noise there.
So I guess you have two potential solutions -- either calibrate your monitor properly, or run noise reduction software so that your footage will look noiseless when it's displayed on a monitor that's showing it how it was never intended to be displayed.
I have had a few DP's call me and ask me "what is wrong" with that sample AF100 footage.
I edit for many producers and DP's who have shot with VariCam and HVX and wanted my opinion on the AF100.
I refer them to the versed trailer and the footage Timur shot in NYC.
They say they can see a visual difference. These are 20+ year DP's who are saying this.
These are guys caught up in the great DSLR debate and don't know which way to go.
They all like the look of some DSLR footage but hate the clunkeness....
I know an HVX shooter that gets hired as a wide camera & audio for
multi-cam DSLR shoots.
I tell them that the AF100 looks like a great value but I want to see more footage
before I recommend them going for it.
I edit for many producers and DP's who have shot with VariCam and HVX and wanted my opinion on the AF100.
I refer them to the versed trailer and the footage Timur shot in NYC.
They say they can see a visual difference. These are 20+ year DP's who are saying this.
These are guys caught up in the great DSLR debate and don't know which way to go.
They all like the look of some DSLR footage but hate the clunkeness....
I know an HVX shooter that gets hired as a wide camera & audio for
multi-cam DSLR shoots.
I tell them that the AF100 looks like a great value but I want to see more footage
before I recommend them going for it.
Rick Burnett
Section Moderator
What I am saying is are you going to go and expect every person who watches your material online that you have to give them a disclaimer that they need to properly calibrate their monitor before they watch what you give them? I mean, it's not like people are bringing up these questions without any frame of reference. I can go back and watch footage from a variety of cameras on these same LCD screens that are not calibrated and they don't show the same level of noise at the approximate ISO/gain levels as what some of the posts have shown with the AF100. That's why people are a little disconcerted that they see this much noise in a circumstance where they've watched PLENTY of other material and haven't seen anywhere near that much dancing in the dark.
I watch tons of trailers and videos on the uncalibrated screen and I've yet to see ANY noise that comes close to some of the clips I have seen with the AF100 that people have released. Which is why I downloaded them to play with them so I could understand what was going on. That's how I could see that the grading could be improved and really push out that noise, without having to use neat video.
Do you really think that every theatre, every home television is going to be ANYWHERE near calibrated? Just go to a Best Buy and look at the LCD wall and look at the MASSIVE differences between any different screens and the variety of settings that they use. It's all over the place, which I understand, is why it's hard to hit a moving target. But like I said, this is no different than music. I can create audio with massive awesome bass below 80Hz that sound amazing in my studio. I play that out of a cheap pair of speakers and the bass is so strong, the music distorts and sounds awful. And my studio monitors are VERY flat response. Does this mean that I should just tell people, sorry about your crappy speakers, go buy good ones that have a good response. No, the onus is on me to create a good mix that handles well in a variety of circumstances. That is part of being a good mastering engineer (which I hardly would call myself anyways
)
The same is true with video. I have seen AMAZING noise free footage from the AF100 that I felt was graded very well. I tested that footage on both my calibrated and uncalibrated screens and it held up well in both circumstances. I believe that some experimentation is in order to find out what i can do to get the best of both worlds. I don't want to reduce the quality of the image in the calibrated monitor, but I do want to enhance the viewing experience in displays that enhance that range, which obviously, is going to be a lot of circumstances since uncalibrated LCDs seem to show this.
Don't misunderstand my feelings, I am not seeing a problem with the AF100. I think the footage is just different than what people have been playing with and so people like the OP are wanting to find a good workflow to get the noise characteristic seen in Timurs piece (or other pieces). And I would NEVER hold it past FCP to do something to the footage as I've had problems with it before.
I watch tons of trailers and videos on the uncalibrated screen and I've yet to see ANY noise that comes close to some of the clips I have seen with the AF100 that people have released. Which is why I downloaded them to play with them so I could understand what was going on. That's how I could see that the grading could be improved and really push out that noise, without having to use neat video.
Do you really think that every theatre, every home television is going to be ANYWHERE near calibrated? Just go to a Best Buy and look at the LCD wall and look at the MASSIVE differences between any different screens and the variety of settings that they use. It's all over the place, which I understand, is why it's hard to hit a moving target. But like I said, this is no different than music. I can create audio with massive awesome bass below 80Hz that sound amazing in my studio. I play that out of a cheap pair of speakers and the bass is so strong, the music distorts and sounds awful. And my studio monitors are VERY flat response. Does this mean that I should just tell people, sorry about your crappy speakers, go buy good ones that have a good response. No, the onus is on me to create a good mix that handles well in a variety of circumstances. That is part of being a good mastering engineer (which I hardly would call myself anyways
)The same is true with video. I have seen AMAZING noise free footage from the AF100 that I felt was graded very well. I tested that footage on both my calibrated and uncalibrated screens and it held up well in both circumstances. I believe that some experimentation is in order to find out what i can do to get the best of both worlds. I don't want to reduce the quality of the image in the calibrated monitor, but I do want to enhance the viewing experience in displays that enhance that range, which obviously, is going to be a lot of circumstances since uncalibrated LCDs seem to show this.
Don't misunderstand my feelings, I am not seeing a problem with the AF100. I think the footage is just different than what people have been playing with and so people like the OP are wanting to find a good workflow to get the noise characteristic seen in Timurs piece (or other pieces). And I would NEVER hold it past FCP to do something to the footage as I've had problems with it before.
There's an inherent problem with a discussion like this happening online, instead of in a place where we can look at the same monitor. There are differences in people's inherent abilities to discern these things, combined with different display devices, and importanly, subjective bias based on what someone may "like" to see.
I have purchased and will receive by month's end, my Af100. But I also saw the noise in parts of that footage, which I watched through PS3 to a Panny studio monitor via HDMI, and the footage was the same full-res file uploaded to vimeo, just re-wrapped to mp4, not trancoded. I've worked in this business for 21 years, and I saw excessive noise in the same shots I've seen others mention. The reason it did not worry me badly enough to cancel the order, is that it seemed an anomaly, for which I felt there must have been a rational explanation. I disagree with characterizing an observation like that as a "complaint". It simply means, for me, I need to know why I'm seeing it, and whether it represents an exception, or an unavoidable norm.
Again, what Timur described as his experience after black-balancing the AF100 as something of the clouds opening and rays of light beaming down from the heavens with harp music, seemed to confirm that there could be a situation with that camera that could lead to excessive noise, which can be eliminated. So with that and other very clean samples from the AF100, I have continued my order for the AF100.
I have purchased and will receive by month's end, my Af100. But I also saw the noise in parts of that footage, which I watched through PS3 to a Panny studio monitor via HDMI, and the footage was the same full-res file uploaded to vimeo, just re-wrapped to mp4, not trancoded. I've worked in this business for 21 years, and I saw excessive noise in the same shots I've seen others mention. The reason it did not worry me badly enough to cancel the order, is that it seemed an anomaly, for which I felt there must have been a rational explanation. I disagree with characterizing an observation like that as a "complaint". It simply means, for me, I need to know why I'm seeing it, and whether it represents an exception, or an unavoidable norm.
Again, what Timur described as his experience after black-balancing the AF100 as something of the clouds opening and rays of light beaming down from the heavens with harp music, seemed to confirm that there could be a situation with that camera that could lead to excessive noise, which can be eliminated. So with that and other very clean samples from the AF100, I have continued my order for the AF100.
It was specifically the shuffle of angels feet.... no harps.
Steve Kahn
Well-known member
Exactly. If this turns out to be a complex question let's figure it out.
This doesn't explain why some video has noise and other video does not - on the SAME monitor.
Timur seems to believe it is a black balance issue. You seem to not agree.
Exactly again. The point is: HOW do we optimize footage for many viewing situations.
+1. Nailed it again. We're not pointing fingers here. We're still in. But this is becoming our camera now. We need to understand what's going on.
Would be very interested in your opinion. You said you saw the noise that we saw at ISO 400 which made you worry. Is this just a black balance issue? Or, do other factors play in as well?
Melvin Harris
Black Flag Member
(sigh)
heavyG
Underwater Giraffe Member
Seems to me that extensive testing of various settings (in camera as well as in post) will be needed once the cameras arrive - and then a thread or two where those of us who get the camera early (fingers crossed) can share footage and settings. I assume that this cam will have scene files that can be uploaded similar to the HVX et-al?
heavyG
Underwater Giraffe Member
I'm feeling you on that…
Shooter
Veteran
@grimepoche
Sorry man ..I disagree with you so much when you say things such as "Do you really think that every theatre, every home television is going to be ANYWHERE near calibrated? "
It is absolutely impossible for any Colorist (or a Cinematographer) to compensate to "the lowest common denominator" . eg grandmas 1982 color telly
Images should always be CC / graded , using known specs on a properly calibrated reference monitor.
Its fundamental to the workflow and its not rocket science.
Black is black...white is white and the grey scale is everything in between. A properly calibrated monitor will display these precisely.
Sorry man ..I disagree with you so much when you say things such as "Do you really think that every theatre, every home television is going to be ANYWHERE near calibrated? "
It is absolutely impossible for any Colorist (or a Cinematographer) to compensate to "the lowest common denominator" . eg grandmas 1982 color telly
Images should always be CC / graded , using known specs on a properly calibrated reference monitor.
Its fundamental to the workflow and its not rocket science.
Black is black...white is white and the grey scale is everything in between. A properly calibrated monitor will display these precisely.
Barry_Green
Moderator
Okay, I'll try again -- Steve, Grim, etc... I'm talking about the exact same clip. Timur's clip. With the magical black balance, and Timur's magical settings.
When shown on an uncalibrated display, looks noisy.
When I calibrated the display, it looks magical.
I'm not comparing footage from one shoot to footage from another. I'm not comparing vimeo to youtube. I'm not comparing CC'd to raw. I'm playing a raw clip, in windows media player, on this screen. And proper calibration made all the difference in the world.
Do with that what you will. Make of it what you will. I can't go chasing down every miscalibrated monitor in the world. I'm not going to color-correct the Hulk to magenta, so he'll look green on someone's horribly-miscalibrated monitor with the tint control all the way out of whack. I'm gonna make him green, so that he looks green. If someone's monitor shows him as magenta, that's the monitor's fault, not mine and not the footage's.
Does that mean the noise is there? Yes it's there. But it's invisible. Should Panasonic just crush everything in the below-7.5 IRE range to solid black? Maybe. Maybe that's what folks want? If you want noiseless footage, then consider crushing 0-8 ire to black. Maybe that's the post workflow people should employ then, if they're worried about people seeing noise on non-calibrated gamma-stretched displays. Because 0 to 7.5 is all supposed to be black anyway, according to the TV engineers.
Maybe that's a concession people will have to make in this new age when video footage designed for video monitors, is instead going to be displayed on computer monitors... I don't know. It's up to you and how you want to show it.
When shown on an uncalibrated display, looks noisy.
When I calibrated the display, it looks magical.
I'm not comparing footage from one shoot to footage from another. I'm not comparing vimeo to youtube. I'm not comparing CC'd to raw. I'm playing a raw clip, in windows media player, on this screen. And proper calibration made all the difference in the world.
Do with that what you will. Make of it what you will. I can't go chasing down every miscalibrated monitor in the world. I'm not going to color-correct the Hulk to magenta, so he'll look green on someone's horribly-miscalibrated monitor with the tint control all the way out of whack. I'm gonna make him green, so that he looks green. If someone's monitor shows him as magenta, that's the monitor's fault, not mine and not the footage's.
Exactly. And on a properly-calibrated display, this noise people are talking about, simply disappears. If you were here I could show it to you.
Does that mean the noise is there? Yes it's there. But it's invisible. Should Panasonic just crush everything in the below-7.5 IRE range to solid black? Maybe. Maybe that's what folks want? If you want noiseless footage, then consider crushing 0-8 ire to black. Maybe that's the post workflow people should employ then, if they're worried about people seeing noise on non-calibrated gamma-stretched displays. Because 0 to 7.5 is all supposed to be black anyway, according to the TV engineers.
Maybe that's a concession people will have to make in this new age when video footage designed for video monitors, is instead going to be displayed on computer monitors... I don't know. It's up to you and how you want to show it.
Last edited: