Highlight Clipping Seems to be the Main Concern

[Justyn]

jannard said before April on the scarlet. i believe he would not have said that, knowing his own past history on dates, if that wasn't a feasible date barring any unforeseen issues. 2012 no way. This is coming soon.

I have some swamp land for you down here in FL too. I'll just say that you and anyone should wait cause otherwise you'll be highly disapointed. Of course he's saying this.. he doesn't want everyone to buy an AF100.. even if the scarlet rock the house, people might have bought something else. I've reserved the right to NOT believe what they say till it's delivering.

 

[Paul Vicente]

is means "presently", no?

 

[Paul Vicente]

present 3d singular of be

 

[Duke M.]

I am supposed to have access to another prototype after they're done with some upcoming shows and such. Interbee in Japan is taking up just about all of the existing models. Once all these shows are over, they are going to be able to send me one with, hopefully, all the latest upgrades and latest software, and then I can do some controlled testing and figure out what the bottom line is.

While others will be interested in some gamma testing, I consider that an education issue not a camera issue. If you could throw in a real resolution test (which will also tell us about various moire frequencies) I'll know where I stand.

Thanks.

 

[Rick Burnett]

It is amazing how much an LCD can influence the footage. On our two 7Ds we known that if it looks right in the LCD on the back, the darken it a bit because it will be lighter on our calibrated monitors. And then of course if there is any sunlight hitting near the back and I don't have an VF on it, then it gets worse.

I had an external LCD hooked up and the contrast was SUPER apparent in some crane shots we were doing. Offloaded onto a computer, once again, scene was not as contrasty as we thought, not even close. Point being made, it takes time to learn the relationship to different viewing mechanisms and the final results.

 

[Justyn]

good points Grime.... I often get caught when I don't do all my routine setup procedures. Sometimes It's a simple thing like not white balancing correctly and bingo...

 

[PaPa]

I am supposed to have access to another prototype after they're done with some upcoming shows and such. Interbee in Japan is taking up just about all of the existing models. Once all these shows are over, they are going to be able to send me one with, hopefully, all the latest upgrades and latest software, and then I can do some controlled testing and figure out what the bottom line is.

Can i come too?

 

[Duke M.]

I thought the clips by Emmanuel Pampuri showed nice skin tones and transitions with little clipping with high contrast in harsh direct sunlight.

 

[Dino]

It is amazing how much an LCD can influence the footage. On our two 7Ds we known that if it looks right in the LCD on the back, the darken it a bit because it will be lighter on our calibrated monitors. And then of course if there is any sunlight hitting near the back and I don't have an VF on it, then it gets worse.

I had an external LCD hooked up and the contrast was SUPER apparent in some crane shots we were doing. Offloaded onto a computer, once again, scene was not as contrasty as we thought, not even close. Point being made, it takes time to learn the relationship to different viewing mechanisms and the final results.

That's why a waveform monitor is so useful, or at least zebras. The waveform never lies. A REC-709 calibrated on-set monitor is also very useful.

 

[groveChuck]

Barry, I saw your post on LCD vs Zebras. I've been meaning to ask- did you, Philip or Filippo use the waveform?
If so, what did you see? Thanks.

 

[Rick Burnett]

That's why a waveform monitor is so useful, or at least zebras. The waveform never lies. A REC-709 calibrated on-set monitor is also very useful.

I absolutely agree. One of the selling points for me to upgrade to the AF100 was those things exactly. On the 7D, I don't have them. On a shoot a few weeks ago, smallHD allowed me to demo their DP6 and it has a false color mode that I was playing with. I can say the DP6 contrast and brightness were much closer to what we see in post, especially compared to the DP1X which I own and use. Significant improvement.

The AF100 cannot get here soon enough. It will be nice to take the 'guessing game' out of the equation.

I can feel for Barry with response to seeing the zebras and seeing the results on the LCD if in fact the backlight was set too low. Even if the zebras are right, as a creative professional, it's hard to just rewire your brain in an instant to try and get it exactly right. You fight yourself because you WANT it to look good in the LCD. It's natural.

If this is the new LCD, I'd imagine all the DPs were also trying to get into using the new LCD and also fighting with their previous experience of using the prototype LCD which probably was a lot different. Just switching between a 7D and a t2i was interesting to me on a shoot. I try to not shoot on the t2i because it just looks so different, I'm not always 100% sure how that will translate.

I do have a curious question. If the AF100 has around 10.5 stops of DR, that would seem to me you could certainly use a lot more highlight protection if you wanted given you have more starting range to work with. I'd imagine there is a lot of tuning and testing that goes on. Does Panasonic ask for your clips as well and your feedback Barry? Just curious because engineers can do all they want, but at the end of the day, a professional DP with good discerning eyes are going to provide the most useful and amazing feedback to them.

 

[PaPa]

im wondering if an increase in latiude can be done in such a way that we only get more detail in one end of the spectrum ( i.e. we get an increase in detail in the darks and no gain in the hi's ) ? Is this possible, or if one goes from a camera that previously had 8.5 stops (7D) to a camera that has 10.5 stops (af100) is it assumed that we gain at both the top and bottom ends of the luma? Just hard to believe with what we've seen so far that there really is any increase in latitude. But again, always feels slightly useless posting queries to a product that isn't yet finished.

 

[Barry_Green]

That's why a waveform monitor is so useful, or at least zebras. The waveform never lies. A REC-709 calibrated on-set monitor is also very useful.

Yep. And the LCD does. Which is why I always advise never to rely on the LCD for exposure. And that's a rule I broke sometimes with the AF100, because... it just looked too dark, and hey, it's a new camera, I thought I would explore the boundaries, but ... the immutable law is the immutable law. If you're seeing zebras at 105, you're clipped and you have to stop down. I set the zebras down to 100 and let them show up more than I would on, say, an HVX or HPX170, and in retrospect I'll say that didn't work. You still have to guard your highlights carefully. That's a standing rule with all digital video -- you get far more latitude in underexposure, than you get in overexposure, and overexposure is always a big no-no. I let too much overexpose, and I think most of the guys did. I can say the LCD lulled me into a false sense of thinking that the footage must be darker than it really was, and that I was way too generous with the zebras, something that I will obviously correct for the next time I shoot with it.

 

[Barry_Green]

Barry, I saw your post on LCD vs Zebras. I've been meaning to ask- did you, Philip or Filippo use the waveform?
If so, what did you see? Thanks.

I did, I have no idea if they did or not. We were rarely next to each other, especially at the Temple setting, there was, I don't know, 10 acres or more of space there, and we were all over the place, so I didn't really get to observe them.

I would occasionally pop it on, and when I did, I got great shots. The overexposed shots I got were when I was relying on the LCD or the "Iris Level" meters, and ignoring the hottest zebras (to my peril). But hey, you have to learn how far you can go, and the only way you learn is by pushing too far and seeing when it breaks, and what I learned is -- the zebras are not fooling around, they're serious, when they show up at 100 you'd better be extremely careful about what you let show up with 100.

 

[Rick Burnett]

im wondering if an increase in latiude can be done in such a way that we only get more detail in one end of the spectrum ( i.e. we get an increase in detail in the darks and no gain in the hi's ) ? Is this possible, or if one goes from a camera that previously had 8.5 stops (7D) to a camera that has 10.5 stops (af100) is it assumed that we gain at both the top and bottom ends of the luma? Just hard to believe with what we've seen so far that there really is any increase in latitude. But again, always feels slightly useless posting queries to a product that isn't yet finished.

That is a good question. There are two different mechanisms at play with DR. One is noise. If you reduce your noise floor, you increase your DR. This is always the subjective end of the DR discussions. At some point the noise is so great that you cannot tell a difference, but clearly the acceptable point is far above that where there is noise, but you can live with it (and it is different for everyone I am sure). The other mechanism is the storage capability of the pixel. The larger that storage ability, the more highs you can capture (and of course the depth of the AD converter). Clearly most designs differ in both these. When someone tells you a camera has a DR of a certain amount, it still leaves a lot to question. I typically try to work with DR and what some refer to as Native ISO of the sensor. I think this is a good starting point to getting an idea, but of course, ISO on digital cameras is not standard either.

 

[Barry_Green]

I can feel for Barry with response to seeing the zebras and seeing the results on the LCD if in fact the backlight was set too low. Even if the zebras are right, as a creative professional, it's hard to just rewire your brain in an instant to try and get it exactly right. You fight yourself because you WANT it to look good in the LCD. It's natural.

This is exactly the case, and while I offer no excuse for myself (I mean, I should have known that the zebras weren't just gentle guidelines, they were hard-and-fast rules) I can well understand if the other shooters were lulled by it. Especially those who shot primarily DSLR before, because the DSLR doesn't have any sort of valid hard exposure tool, so if they're used to judging off the LCD, I can see how they may have been led to overexpose. That said, the LCD offers seven different backlight levels, so I am certain we can get it to look more accurate, and there are enough valid hard-and-fast exposure tools that yes, all the guesswork should be able to be completely eliminated.

If this is the new LCD, I'd imagine all the DPs were also trying to get into using the new LCD and also fighting with their previous experience of using the prototype LCD which probably was a lot different. Just switching between a 7D and a t2i was interesting to me on a shoot. I try to not shoot on the t2i because it just looks so different, I'm not always 100% sure how that will translate.

It was the new LCD, and it certainly seemed different to me. I don't accept that as an excuse, I still say I blew it because I should have known to trust the hard meters more, but ... I'm just saying that I definitely found it tempting to open up more to make it brighter on the LCD, and I fell victim to that temptation. I am not speaking for the other guys as we didn't discuss it, but I would not blame them at all if they fell victim to the same temptation.

I do have a curious question. If the AF100 has around 10.5 stops of DR, that would seem to me you could certainly use a lot more highlight protection if you wanted given you have more starting range to work with.

Agreed. Which makes it doubly important to never allow clipping, expose for the highlights, and let the DR pick up the rest.

I'd imagine there is a lot of tuning and testing that goes on. Does Panasonic ask for your clips as well and your feedback Barry?

Yes, they have all the clips.

 

[PaPa]

"Agreed. Which makes it doubly important to never allow clipping, expose for the highlights, and let the DR pick up the rest."

Good point. Range is range, and clippin is clipping. More dynamic range has nothing to do with allowing clipping to go further... Silly me.

 

[Rick Burnett]

"Agreed. Which makes it doubly important to never allow clipping, expose for the highlights, and let the DR pick up the rest."

Good point. Range is range, and clippin is clipping. More dynamic range has nothing to do with allowing clipping to go further... Silly me.

Or as I like to say "Once it's gone, it's gone!" The editor I work with spent time ingraining this into my shooting method. If you have to err, err on the side of under expose in most of the situations because at least some noise can be removed if needed, but once you blow out the detail, you cannot get that back.

The other problem with most LCD screens is they depend on the environment around you. A shot we did yesterday also had a lot of sun behind us. Since we were shooting on a crane, the LCD setup wasn't fully ideal. I am thinking of using a C stand near the crane in the future when the sunlight is directly behind us and using some sort of throw over fabric. We had to shoot this way as we were shooting across a lake and needed the light to hit the trees right. It was a bunch of 'point and pray' that's for sure. We could use the LCD for framing, but for exposure it was tough.

This is why the meters will be so important. However, I do believe I read that the overlays cannot be sent out the HDMI or HD-SDI, so in this instance, up on the crane, we'd still be in the same boat.

I will repeat, I cannot wait to get those zebras and waveforms! While I know my eyes will still play a huge role, I consider this "having more than one opinion" about my chosen settings Barry just forgot to use the best 2 of 3 principle

 

[dillont]

I've never trusted an LCD to give me a proper exposure. It's brightness depends on the viewing angle. I always use the EVF to set exposure.

I tell my students "You know what LCD stands for? LOW CUALITY DISPLAY."

(Yes, quality is misspelled).

Edited to add: Of course, I only tell my amateur students this. This doesn't apply to professionals who have been shooting for decades.

 

[Alpha Beta Omega]

Shooting DSLR style video and setting exposure by the LCD screen alone will at the beginning result in many out of focus and overexposed images. Once you start to look out for when an image starts to clip then you can control what you shoot. But the LCD displays does trick you into turning up exposure since it looks so darn good at the display. I always feel safe when using false colours and will feel safe in having other proper tools like the ones in the AF101.

I shoot alot of stills and using RAW format I can save many shoots in correcting for wrong colours or exposure but with digital video (except RED to a certain degree) you gotta be on target or slightly underexposed for it to look great. Same thing with areas that are way underexposed. It's better to have the whites lowered and the black raised and then in post stretch both to your likening. That's just the nature of video as I see it.

Same thing with audio - to high=distorted - to low=noisy. But with some headroom you're safe and get a good result.