GH6 with ProRes

[ahalpert]

Absolutely incorrect.

But I'll try to be understanding because it's obvious you never used it because the example you provided is foolish. Explosions? Who is filming explosions besides the 1%? [And these can be created in post.]
___

Working by yourself in so many environments and having the option of not worrying about clipping during interviews, multiple speaker panels (low and high talkers), events, sports games (and sports celebrations), etc. is one of the best inventions [and comforts] this industry has ever seen.

Almost every sound guy/gal I know in New York City has a 32-bit kit. Doesn't mean all do and doesn't mean all will use it and doesn't mean they are now better because of it, but 32-bit float is still extremely useful for them and even more for solo camera operators.

I also have a 32-bit float recorder. Never used that mode, and never clipped with it. plenty of solo camera operation where I'm not even mixing levels during the recording

we had this discussion dvxuser recently and actual sound guys on this forum basically said it's a feature that non-sound people obsess over and isn't as useful as they think. explosions = actual use case for that much DR

bottom line: gain staging > everything else

 

[NorBro]

Yeah, let's see if they say that if they have to operate a camera in a hectic environment at the same time and only get to have one audio track and no back-up safety tracks.

Every one of them would crumble to their knees to have 32-bit float when it's time to deliver the audio and get paid.

I understand people being fine with what they have in their own circumstances, but saying we don't need more than 24-bit is ignorant when such beautiful technology exists.

 

[Tom Roper]

“Absolutely incorrect.”

No soup for you.

 

[ahalpert]

Yeah, let's see if they say that if they have to operate a camera in a hectic environment at the same time and only get to have one audio track and no back-up safety tracks.

Every one of them would crumble to their knees to have 32-bit float when it's time to deliver the audio and get paid.

I understand people being fine with what they have in their own circumstances, but saying we don't need more than 24-bit is ignorant when such beautiful technology exists.

yeah that's why no camera operator was ever able to deliver flawless audio before 32-bit float

I'm not saying it's useless. just that it's not necessary

 

[NorBro]

I don't understand how you can see such a definitive purpose for AF but you can't comprehend 32-bit float.

AF isn't necessary either, especially since you think you're a good focus puller (your words), but you use it because it makes your life easier (mainly on a gimbal) and your work better.

I'm this close [ ] to shaking my head, mister.

___

Yes, it's not necessary but I want it. I don't want to worry about audio like I don't want to worry about focus, you dig?

 

[ahalpert]

I missed shots with AF, buzzed focus, acquired focus later than I do now, etc. In contrast, I have 32-bit float available to use right now, and I never feel compelled to use it because I never screw up my audio, and I don't even break a sweat. This includes situations where I'm not monitoring/mixing during interviews/events. The fact of the matter is that you already have a lot of headroom with 24-bit

 

[NorBro]

There is going to be a time when the wedding crowd yells so loud in celebration for a couple and you will blow your audio and think about this thread. :engel017:

 

[ahalpert]

Honestly, 32-bit has created some laziness. It’s useful for recording sources with really high and unpredictable dynamic range, especially sound effects that include sudden peaks like explosions. But for dialog, nothing beats good, old-fashioned, proper gain staging. I know a couple of mixers who have 32-bit recorders now and don’t even pay attention to proper setup and record levels, swearing that it’s 32-bit so they don’t even have to worry about it. This is nonsense.

The pre-amps in the MixPre series are excellent and the limiters are quite effective and transparent...

 

[NorBro]

Non sound guys who know a decent bit about sound definitely care about 32-bit, it is game changing for a single operator. For the simple reason that it’s as close to error proof as you can get.

Docu: multiple recorders to multiple people with zero need to monitor levels. I’m one person. Shooting an ambient scene, you can keep audio levels optimal, and then never worry about that large truck that passes or the loud thump or the yell from the crowd. Shooting an interview with a soft spoken subject with an emotional topic as they speak quietly, but then a moment later start a new thought with a raised voice. The take is one of a kind, one take only emotional perfection, except it clipped - except it didn’t, because it’s 32 but.

as a solo operator I choose 32 bit over AF. I look forward to future cameras where 32 bit is standard, it’s the RAW of audio.

Maybe some dedicated sound guys don’t care about 32 bit because a big part of their job is to monitor levels full time for living.

 

[ahalpert]

I think Alex's point is that if you set yourself up right to begin with, you can tolerate just as much volume variation with 24-bit. But thank you for quoting the post someone else just made in this thread without adding comment. That wasn't obnoxious at all

 

[NorBro]

So you can do it without providing any additional thoughts, context but I can't? My response is fitting.

And with all due respect to Alex, he is not a camera operator; he is a professional audio individual whose primary focus is to handle audio.

 

[ahalpert]

the difference is I'm bringing in his point from a separate conversation that people here didn't just read 2 posts up the page. feel free to follow the link to his post if you want to see the broader discussion

I think the implication people are making here is that sound mixers don't like 32-bit float because it renders them redundant. but I think the truth is closer to something like sound mixers think that camera operators don't know how to properly set up sound and see it as a magic bullet for getting around it

if filmguy123 has use cases where it makes all the difference, then go for it

the main point about 24-bit is that if you have proper gain staging and good limiters, you can record to a low level without fear of clipping when something loud happens

camera op perspective from the same convo:

One thing that has improved my sound over the last couple of years is putting levels lower. Without really thinking about it I had come to believe that recording audio 12dB lower to then raise it in post was analogous to underexposing the image a stop. So... probably ok but there is a cost to pay in increased noise.

This might well be true for analogue levels (I'm not a soundie) but for digital levels at the recorder stage, it is not the case. 24-bit audio is FAR more luxurious in terms of headroom and data allocation than a 10-bit log image. So for interviews I now set levels to peak at around -18dB with a -6dB limiter to catch laughs or sudden explosions. If an interviewee is peaking as low as -24dB I usually leave it.

https://www.massivemastering.com/blo...ing_Levels.php

 

[NorBro]

Yes, it makes all the difference for those that use it, that's the point of using the device.

 

[DLD]

I shoot a lot of 16:9 cropped to 2.4:1 in post. frame guides on the external monitor

But wouldn't you prefer it done in camera the way all other available aspect ratios are?

PS. Side note - I just cropped a friend's 3:2 vertical still into a wide screen. It has "enough" pixels for a social net photo but I see why 16K would be the next target for pro video.

 

[ahalpert]

But wouldn't you prefer it done in camera the way all other available aspect ratios are?

PS. Side note - I just cropped a friend's 3:2 vertical still into a wide screen. It has "enough" pixels for a social net photo but I see why 16K would be the next target for pro video.

why - for cropping wide screen video into a vertical still? why would you need 16K?

would I prefer what done in camera? the frame guides? or the aspect ratio to be recorded as it will be shown?

having frame guides available would be a nice option, although I wouldn't use them since I basically never monitor off the mirrorless camera built-in monitors. (there are frame guides available on the camera, I just can't remember if it has 2.4:1...)

I don't mind recording the "look-around" on top and bottom. occasionally it helps you adjust headroom in post. it probably wouldn't generate enough data savings to be worth it to lose that.

if I could record at a higher frame rate by recording in 2.4:1 only (as on blackmagic and myabe red and z-cam, etc), then that might be worth it

 

[ahalpert]

Yes, it makes all the difference for those that use it, that's the point of using the device.

I don't think it's universally needed, certainly not to the point that it needs to become a standard offering. I have it and don't use it.

and if you're comparing 32-bit float to recording RAW in video, well we all know that you can **** up recording RAW video and it's never going to be good

another cameraman from that convo who says it's useful but, in his case, only for getting usable audio rather than good audio:

32-bit float recording on the MixPre II is pretty amazing in what it can help you recover. I've used it once with a completely inexperienced boom operator who had no idea how to set the gain, and again with mics that were just planted because we had nobody to operate the mixer. In both cases the audio was usable, which I think it wouldn't have been with 24-bit tracks. I wouldn't say that the audio was actually good, but I think the poor mic placement had far more to do with that than the lack of a sound person operating the mixer.

 

[NorBro]

I don't think it's universally needed, certainly not to the point that it needs to become a standard offering. I have it and don't use it.

and if you're comparing 32-bit float to recording RAW in video, well we all know that you can **** up recording RAW video and it's never going to be good

another cameraman from that convo who says it's useful but, in his case, only for getting usable audio rather than good audio:

filmguy compared it in his post - but, no, you can't mess it up unless the microphone is too far or you don't press record.

Why you choose to have these conversations is beyond me.

Such amazing technology was created and you're sitting here rationalizing...okay, now I'm shaking my head.

Please, for the love of God, try 32 on your recorder.

Yell "I love 24-bit!" at the top of your lungs into your microphone, import the file into FCP, and drag down the dB and watch the magic happen.

 

[ahalpert]

I always debate when I read something I disagree with. I shoot chaotic live events too and I don't miss anything and I use 24-bit...

But the issue at hand is not what 32-bit does and how it can be used. The OP was what is its relative value and importance (ie should it be a standard feature on every camera and is it more useful than AF)

I feel the same way about RAW. That's nice. I'd like to use it (depending on trade-offs for data storage, etc). Do I need it? Not really. Like Doug said in another recent conversation, it's most useful when you screw things up. I could probably benefit from it on 5 shots per year in that regard. And having it wouldn't mean that I could afford not to control my exposure

 

[NorBro]

The original thought was suggesting we collectively would like 24-bit in cameras, and I think if we're asking for better audio, we want 32-bit float instead.

Just like how you/anyone asks for video features...more DR, no-crops, more resolution, higher-framerates.

If someone says I want 4K/60p, I don't say, "Well, I'm good with 4K/24p and only use 4K/60p 5x a year so I don't think it's really necessary, or universal."

No offense, but you always debate selfishly and it's always about what's best for you.

If Thomas is saying to ask companies for higher-quality audio in cameras/24-bit, I'm basically saying, "We can do better than that, we can ask for more."

 

[ahalpert]

that's actually a great point. Thomas started the topic by using 24-bit as an example of high-DR audio with overhead equivalent to 14+ stops of exposure DR or 8K for 4K.

Shooting 8k is more about having overhead much like why we want 14+ stops of DR or 24 bit audio.

people do indeed say they don't need things they won't use. I don't need an option to record uncompressed video. it may be "better" (the best, in fact), but I know for a fact that I would never use it

I'm debating about what is generally required in cameras. I acknowledge that some users don't need or want AF. In fact, if you're using the best lenses, you won't be able to make use of sensor-based AF anyway. I'll always qualify the features I'm discussing with the use cases they address.