Cam Audio Recording Request: 32-Bit Float

Further to my post just above...

On the surface, this thread is a request for Blackmagic to add 32-bit float. There’s no reason to stop there. Why not noise reduction, which would obviously be far more useful and probably cheaper to implement. And ...

I do use Trew Audio and trust their recommendations.

I do use 32bit recording on the MixPre-6 II and appreciate being able to forget about it once I set my audio levels; but I also record 24bit XLR on my URSA Mini 4.6K successfully… no nearby thunderclaps to test, just the occasional gunshot!

I do use Trew Audio and trust their recommendations.

I do use 32bit recording on the MixPre-6 II and appreciate being able to forget about it once I set my audio levels; but I also record 24bit XLR on my URSA Mini 4.6K successfully… no nearby thunderclaps to test, just the occasional gunshot!

And Trew sells the Deity PR-2. Chris should be able to pick up the international version at Trew Vancouver, although being based in Los Angeles it might well decline to ship one to the U.S. As mentioned above, Deity also says that five Vancouver camera stores sell its products.

Sound effect recording of gunshots doesn’t require 32-bit float, at least if the recordist knows what he’s doing. It’s not like recording of gunshots only started five years ago. Thunder claps are a little less predictable The New York City area is known for its thunderstorms, but thunderclaps that powerful are not common - as I said above, maybe a couple of times a year.

A few years ago, there was a thread calling for Blackmagic to offer a full sound recording function. I think that this thread ultimately leads to that kind of request. I have no idea why someone would request 32-bit float over, for example, noise reduction. Even the basic MixPre features include things that would be more useful for camera sound recording than 32-bit float.

Thanks for the info you two!

I had performed a simple google search on stores in Vancouver and if I had seen Trew, I would have looked them up because I've been to their store in Los Angeles. Now that I know they're in Vancouver, I'll see if I can get a ride up with my friend and go pick those up. It would be nice to get the International version.

And I agree, noise reduction seems like a very useful tool. I saw a video where a Sound Devices guy was showing that off. Initially I confused it with the MixAssist and was going to buy it thinking it was $99. Sadly, it's actually a $300 single mic plug-in. Ouch.

I don't typically record in 32-bit float, but maybe for my upcoming interviews. Rob, thanks for the thunder recording info. I do a lot of work in Texas/New Mexico so that might come in handy for me. When I've gotten to ye olde freeway overpass since often I've been dodging hail, tornadoes, etc.

Now that I know they're in Vancouver, I'll see if I can get a ride up with my friend and go pick those up. It would be nice to get the International version.

...

thanks for the thunder recording info. I do a lot of work in Texas/New Mexico so that might come in handy for me. When I've gotten to ye olde freeway overpass since often I've been dodging hail, tornadoes, etc.

If the Deity PR-2 preamp and lavs are decent, you might have a look at 3:25 of the video below. Deity’s Andrew sets up a PR-2 to make an AB Spaced stereo recording of a waterfall. I sometimes use a couple of DPA 4060 lavs AB Spaced to record stereo. It can be quite effective, and you don’t need a “proper” stereo bar to do it.

Deity PR-2 User Guide | Features & Specs

Also shown at 2:35 of this video:

I'll record an example video for everyone with the PR-2. Can't share the video and audio from last week's client shoot. But that boom audio sounded fantastic straight into the URSA. Even if I saw the levels starting to reach the red zone. PR-2 also sounded great for that shoot.

I'll record something for you guys soon.

I'm just going to say the quick test I did was "ew" because I left my AC on. And, unless I do voice isolation to remove that AC then you're literally not getting an accurate portrayal. And, the AC just sounds... well, it's an AC unit.

I had the Sennheiser 416 via XLR going into the P6KPro, and the Deity PR-2. Both sound exactly the same quality via my headphones. Especially with 20% Voice Isolation to remove the AC Unit.

I did both 32-Bit Float Mono and 24-Bit Stereo on the PR-2 for that test. Plain and simple my levels were good and no issues. But again, the AC Unit makes it all so "ew."

Tomorrow I'll do a better test if I can for you guys.

However, one thing I'd like to test out is using the Sound Devices MixPre with the Sennheiser 416 on Boom going into that, and then have the Deity PR-2 on the subject. See how that overall compares. That I need to organize with my friend who has the MixPre to meet up for that test.

Timecode wise it all is working fantastic.

If the Deity PR-2 preamp and lavs are decent, you might have a look at 3:25 of the video below. Deity’s Andrew sets up a PR-2 to make an AB Spaced stereo recording of a waterfall. I sometimes use a couple of DPA 4060 lavs AB Spaced to record stereo. It can be quite effective, and you don’t need a “proper” stereo bar to do it.

Woo hoo! I just tossed my honking huge Mixpre3 into the trash! Those Pr-2's are tiny in comparison.

Curtis Judd now says these are his favorite. Most reviews I've seen have said the lavs are quite good.

If the Deity PR-2 preamp and lavs are decent, you might have a look at 3:25 of the video below. Deity’s Andrew sets up a PR-2 to make an AB Spaced stereo recording of a waterfall. I sometimes use a couple of DPA 4060 lavs AB Spaced to record stereo. It can be quite effective, and you don’t need a “proper” stereo bar to do it.

Woo hoo! I just tossed my honking huge Mixpre3 into the trash! Those Pr-2's are tiny in comparison.

Curtis Judd now says these are his favorite. Most reviews I've seen have said the lavs are quite good.

Yes, the Lav for the PR-2 is very good. I was pleasantly surprised how good it was considering the price.

I don't think you need to trash your MixPre-3. The MixPre is still a perfect option for when doing boom audio. And, no matter what I will always prefer a boom to a lav. And, you can't deny how nice using a MixPre is. The PR-2 is just another tool in your kit.

Woo hoo! I just tossed my honking huge Mixpre3 into the trash! …

Don’t be alarmed if you see an old guy with glasses and a Panama hat rummaging through your trash. it’ll be me!

Okay, straight from camera and audio recorder. Brought into Resolve, synced, trimmed the fat, and exported. BRAW is set to Extended Video.

https://vimeo.com/999748420/86cf350e75

Should give you an idea on the audio quality unfiltered. Plus, may be the first time you guys are seeing me. Putting a face to my name and handle. Plus hearing my voice.

Should I appear on camera more in a YouTuber sort of way to talk tech?

Røde created this entire market. It’s genuinely interesting to research how it did it. Have a look at the early videos and written marketing material. Røde knew from Day 1 that its biggest challenge was to get filmmakers to accept, as normal, aesthetics that were considered completely unacceptable. Røde succeeded in this campaign, so much so that DJI and others started copying it.

Having created the market, Røde and its copyists have now moved to stage two - they're increasing the cost of entry.

100% agree with this. You see lots of news teams that used to mic reasonably subtly with a standard personal/lavalier mic now clipping on those large and ugly Røde combined mic and transmitters. They may be quicker and easier to use - but they are ugly as...

Marketers love float32!

Take a crappy old pre-amp, a cheap AD converter, and encode the result using float32, and voila you got a float32 recorder!

I am a great fan of float32 encoding (and not just for audio but also for video - we are not quite there yet!), but what I am not a fan of is marketing nonsense that implies that if a device can encode with float32 it must be automatically very high quality.

By the way, the effective bit-depth of float32 is 24-bit. Sure, there’s more resolution for audio near the noise floor, which might be advantageous—especially for someone who can’t figure out how to set levels properly while recording the subtle sounds of a zen garden at night in a remote mountain temple.

You still don't understand the main advantage...

Sent from my iPhone using Tapatalk

You still don't understand the main advantage...

I am sure you will provide me with some marketing materials to explain it to me.

At least 32-bit float has become easier to work with. Four years ago, when I decided to try it out for high dynamic range sound effects, there were problems processing it. Neither Logic Pro nor iZotope RX would work. I wound up installing Reaper. It could process the files, but having to install and use another DAW was a nuisance.

I still prefer to get my levels right in the first place. When it comes to recording the human voice, apart from on some dramatic or documentary film sets, almost everybody using 32-bit float is demonstrating lack of confidence in their skills, if not plain incompetence. It’s become part of the “fix it in post” mentality.

almost everybody using 32-bit float is demonstrating lack of confidence in their skills, if not plain incompetence. It’s become part of the “fix it in post” mentality.

"Almost everybody"? You've surveyed the field from coast to coast, evaluated the recordists and are ready to damn them for incompetence or insecurity? Were they allowed to make any defense in the matter? Or was it a summary judgment to which there's no appeal?

Who knew that embracing a format which removes common errors from the recording chain was another milestone in the decline of civilization....

"Almost everybody"?

John, I can’t be bothered with your never-ending search for a fight

If you want to know where 32-bit functionality has become widespread, see my earlier post about the explosion in use of 2.4GHz wireless.

If you want to think that clipping is a “common recording chain error” for people who understand the basic function of setting recording levels for the human voice, you go right ahead.

Rob, it might be useful to vet your own pronouncements for provocation (and insult, for that matter) before assuming the "me, pick a fight? Never!" pose.

That aside, we can probably leave the choice of format to the individual recordist, and no urgency to praise or condemn the decision. Each to his own, after all. Who knows? The guy might even have reasons of which *you* would approve.

And of course your editorial comments are not responsive to the feature request or the technical questions at issue.

your editorial comments are not responsive to the feature request or the technical questions at issue.

Thanks for the invitation to reiterate some of my earlier comments:

A lot of people are now using 32-bit float as a matter of course. That’s cool, but it doesn’t interest me. I know how to set levels, which I’ve found is not rocket science, and I’m not keen on make-work.

I also don’t care whether Blackmagic offers 32-bit float. However, I have no intention of paying for it. For me, that would be a reason to buy another camera make.

On the surface, this thread is a request for Blackmagic to add 32-bit float. There’s no reason to stop there. Why not noise reduction, which would obviously be far more useful and probably cheaper to implement.

A few years ago, there was a thread calling for Blackmagic to offer a full sound recording function. I think that this thread ultimately leads to that kind of request. I have no idea why someone would request 32-bit float over, for example, noise reduction. Even the basic MixPre features include things that would be more useful for camera sound recording than 32-bit float.

I am also the only person in this thread who has discussed an actual use case for 32-bit float: viewtopic.php?f=2&t=206405#p1072760

In addition, I wrote an entire post discussing the subject of 32-bit float in the context of sound recording gear marketing: viewtopic.php?f=2&t=206405#p1072782 Two other people then commented on the marketing phenomenon.

I'm not sure what you hoped to achieve by quoting your opinions at such length, but you proved my point. The substance of the thread -- what 32 bit float is, how it works, its applications, its usefulness, etc. -- will be found elsewhere.

Yes, the Lav for the PR-2 is very good. I was pleasantly surprised how good it was considering the price.

I don't think you need to trash your MixPre-3. The MixPre is still a perfect option for when doing boom audio. And, no matter what I will always prefer a boom to a lav. And, you can't deny how nice using a MixPre is. The PR-2 is just another tool in your kit.

Damn, you're right, the huge XLR input from my Sennheiser won't plug in to this stupid Deity! Better retrieve my Mixpre from the trash. Oh, wait, there's an old guy in a hat running down the street with it! Give that back!

Seriously, for my interviews, I plan to boom to the Mixpre and record to the lavs so I can double my pleasure and double my fun. One of the people is 92 so I need to make sure I get this recorded since it might be my one shot.

You’re all good, Chris. No worries.

https://youtu.be/0Wk_VPEi8Z8?si=N7edhsnnzJ4bQcid


Sent from my iPhone using Tapatalk

I think this video with a conversation with someone from Sound Devices on the subject of 32-Bit Float really is the most informative. And, it does illustrate why there is an aversion to it from some.

Damn, you're right, the huge XLR input from my Sennheiser won't plug in to this stupid Deity! Better retrieve my Mixpre from the trash. Oh, wait, there's an old guy in a hat running down the street with it! Give that back!

Seriously, for my interviews, I plan to boom to the Mixpre and record to the lavs so I can double my pleasure and double my fun. One of the people is 92 so I need to make sure I get this recorded since it might be my one shot.

Yeah, I'd boom always for interviews and then have lavs on the subjects. That's my preferred method. The one thing I wish in the USA was that that patent didn't prevent simultaneously recording and transmitting with the Deity Theos Lavs. As I do like to at least monitor audio for anything that may cause you to want another take. Too often have I heard garbage trucks collecting trash only to look and see some old guy in a hat sticking out of the garbage can suspended in midair. Then into the back of the garbage truck he goes... still clinging on to that MixPre.

I'm no professional by any means, but I know a few pro audio people, and every one of them agree that 32-bit recording is overrated and ultimately unnecessary. I've heard some audio post houses refuse to even accept 32-bit source files, requiring you to re-render them before sending.

On a professional recorder? Maybe. On a camera, though, seems an unnecessary expense...

Hmmm.... it's quite simple really.

Most audio people concentrate on audio alone. Often in very well controlled environments, where surprises are seldom, if not frowned upon, when coming from the talent.

Most run and gun outfits have no way to control, what happens In a parade, demonstration or carnival procession - even some religious processions get fairly loud. Especially within short distances, when situated in a crowd full of excited people. Trying to control gain here is hopeless, if you also have to concentrate on camera, focus and whatnot else.

We're all interested in good sound, I assume.

Now with 24-bit (or worse 16 bit) inputs of whatever quality, if I exceed maximum level, it's fairly problematic. Whatever the quality of the preamp and microphone, when clipping sets in.

I've used 32-bit float (Zoom F6 and F3 and F2-BT for lavaliers) for years, and the freedom given to me has been educating. The F6 and F3 preamps are very low-noise very high level capable critters (F3 and F6 have +4 dBu standard, but "Line - with Phantom" is really just a +24 dBu microphone input, when things get crazy out of hand.

It happens, but I can quickly "flip a switch", if I suspect extreme levels, and still be confident, that noise is controllable.

I regard 32-bit (with good preamps and good microphones) as similar to travel insurance.

A lot of people are convinced, that they do not need travel insurance, and unless independently wealthy, they're in deep dodo, if experiences the hard end of a hit and run driver.

I always have full travel insurance (world wide). If I can't afford that, I really can't afford travel either, and certainly not an accident on my personal bank account when overseas.

I have had ample experience with blown 16 or 24-bit recordings. Not because the microphone(s) clipped, but because levels turned out to be periodically out of wack. Either far too loud or mere whispers, that I couldn't attend to in the midst of scrum of partying people, agitated demonstrators (loud screaming) or sudden "silence" in a spanish town, during a solemn Semana Santa session after dark.

You still want the best possible sound, but as a one man band - travel, reportage, parties, weddings - you name it - you simply cannot control everything, unless you have a dedicated sound man, and - ideally - everything happens in a studio, well controlled environment.

I've never had that.

Since I started using 32-bit float on good gear with good microphones (for the particular purpose), I have never lost a recording to "clipping" or drowned in "noise". I only have to make sure, that the pre-digital analogue frontend doesn't clip, and the microphone capsule survives the SPL.

That's simple. My loudest microphone (a real hottie) delivers 13 dBu with less than 1% distortion. That leaves a lot of headroom on the receiving end. My most "timid" ribbon microphone still delivers really low noise results without any "cloud lifter". In both cases handled equally well by my gear - in real life, without me wasting precious resources on riding the levels.

It's an insurance kind of approach. It works for me. 24-bit or far worse 16-bit sound does NOT deliver the "leeway" I need in real life.

A sound engineer, that does not have to control available light scenarios, aiming focus, stability and whatnot of course will be perfectly fine in a 24-bit controlled recording setting. Maybe only a bit stressed in "lively" 16-bit scenarios. Almost all the time ("life" still happens

Regards

https://youtu.be/0Wk_VPEi8Z8?si=N7edhsnnzJ4bQcid


Sent from my iPhone using Tapatalk

I think this video with a conversation with someone from Sound Devices on the subject of 32-Bit Float really is the most informative. And, it does illustrate why there is an aversion to it from some.

Thanks, I just listened to that interview. The Sound Devices employee makes precisely the point that I made 10 posts up about the “fix it in post” mentality:

YouTube presenter:
One of the comments that I got in my last video is that it’s still important to set gain even if you’re recording in 32-bit float. It sounds like that’s only important insofar as you want to save work in post normalizing things. Is that correct?

Sound Devices employee:
Yes, that’s exactly what it is, saving yourself the headache later by doing it now. It’s kind of like the fix it in post mentality. You can fix it in post in this case, but if you set it correctly to start you have less work to do in the end.

Elsewhere in that interview, he says that his general preference is to record in 24-bit. Why? Because he understands basic level setting and generally doesn’t need 32-bit float, or the additional work that it creates.

The one thing I wish in the USA was that that patent didn't prevent simultaneously recording and transmitting with the Deity Theos Lavs. As I do like to at least monitor audio for anything that may cause you to want another take.

Just want to note that I’ve seen a suggestion on YouTube that Deity’s app for its Theos products, when used on an American phone, will defeat the benefits of international Theos versions. If I recall, I saw this on the Sound Speeds YouTube channel. I have not seen it confirmed.

I don’t know whether this is true of the PR-2, which in any event can fully function (see Curtis Judd’s review) without using Deity’s app.

Since microphone clipping has been mentioned several times it's probably important to note that mics don't really clip outside of super high SPL level events like a gunshot. Shure lists the maximum human voice at 135 db 1 inch from the mouth. Even a pedestrian 416 (130db max SPL) will never clip 2 feet from the loudest human on earth. The clipping associated with microphones is more likely a function of an overloaded preamp, A to D, or recorded bit depth.

Good Luck

The one thing I wish in the USA was that that patent didn't prevent simultaneously recording and transmitting with the Deity Theos Lavs. As I do like to at least monitor audio for anything that may cause you to want another take.

Just want to note that I’ve seen a suggestion on YouTube that Deity’s app for its Theos products, when used on an American phone, will defeat the benefits of international Theos versions. If I recall, I saw this on the Sound Speeds YouTube channel. I have not seen it confirmed.

I don’t know whether this is true of the PR-2, which in any event can fully function (see Curtis Judd’s review) without using Deity’s app.

The main thing with the PR-2 is you can't record and monitor simultaneously because of the patent. With the Deity App you can still see the audio levels, but you can't listen to it in terms of monitoring. The only time you can monitor is when it is not recording. So, that's the main issue there.

As for the Theos, I am uncertain if it's specifically a USA phone issue. I think it's got to do with it recognizing that you're in the US and thus are restricted by the Patent. But as of now I'm not fully sure as I haven't tested anything regarding this. I haven't bought the Theos yet. It's on my wishlist.

I do know the patent is specifically covering wireless transmitters for body packs. The DRTX for the boom mic is outside the patent and is allowed to wirelessly transmit and record internally simultaneously. This is because that is not a body pack. So the first Deity Theos I would buy is that because that matters more than the wireless transmitters at the moment since I do have Sennheiser G4 that works in the meantime.

As for the Theos, I am uncertain if it's specifically a USA phone issue. I think it's got to do with it recognizing that you're in the US and thus are restricted by the Patent. But as of now I'm not fully sure as I haven't tested anything regarding this. I haven't bought the Theos yet. It's on my wishlist.

Yes, I don’t know whether the Canadian and U.S. Deity apps are different, or whether the app is the same but recognizes where it’s being used. As I said above, I also haven’t seen confirmation that this is an issue with Theos, let alone with the PR-2, and Curtis Judd says that one doesn’t need the app to operate the PR-2.

Assuming, we're talking about the same video, I saw a Sound Speeds review on the Theos and what he said was that if you have the US version and you were in a foreign country, you'd still be locked out from monitoring. It would only give you the options to change the transmit power and frequency ranges in that country. He talks about it at 30 minutes in:

Interesting for me is that my plan is to buy the global version of the PR-2 in Canada so I can do interviews in Mexico. Perhaps I'll install the app on a Mexican phone.

And yes, I will set my gain accordingly and record in 32-bit float.

Since microphone clipping has been mentioned several times it's probably important to note that mics don't really clip outside of super high SPL level events like a gunshot. Shure lists the maximum human voice at 135 db 1 inch from the mouth. Even a pedestrian 416 (130db max SPL) will never clip 2 feet from the loudest human on earth. The clipping associated with microphones is more likely a function of an overloaded preamp, A to D, or recorded bit depth.

Good Luck

SPL specs are only one side of the equation in the capsule end.

The second part is what comes OUT of the connection end. My most potent microphones deliver 50 and 35 mV at 94 dB SPL (1 Pa) at 1m distance. The 35mV critter has no built in attenuation. My least sensitive microphone delivers 0,9 mV in the same situation (94 dB SPL at one meter).

That’s a difference of 38.9 times in level at the same SPL or - rounded up - 40 times or a 32 dB for the identical SPL in the capsule end.

My most potent condenser microphones are able to deliver 13.5 dBu (140 dB SPL) and 8 dBu (132 dB SPL), but my ribbon microphone handles higher SPL gracefully, hardly overloading any mixer/recorder input at roughly -24 dBu compared to the 132 dB SPL on one of the microphones mentioned. The microphone easily handles proximity levels around 150 dB SPL - drums and trumpets come to mind - but still.

You have four main elements in the recording chain.

1. Front end SPL (what can the capsule handle).

2. Maximum microphone output (after any attenuation present)

3. Maximum analogue input (mixer/recorder microphone preamp end).

4. Maximum recording level (digital transmission/recording/clipping level).

The last step is ALSO critical.

The “virtual” clipping level on one of my IEEE internal recording/transmission units (when input “clips” electrically) is around +17 to +18 dB and the most quiet lavalier has a noise level around -84 dB RMS (actually very good) recording in the same quiet environment. Actually the same lavalier microphone.

Note, that actual SPL from unknown source distance cannot be deduced from that.

The 32-bit IEEE Float recorder decides maximum input level (if microphone doesn’t) and the microphone - typically - decides maximum noise level (above recorder “internal noise”).

Since none of my lavaliers have built in attenuators, this setup will allow a dynamic range of 101-102 dB, where the highest levels far exceeds “normal 0 dB” without input gain control. That’s “narjs”!

Minimum noise level is actually decided by movement in air molecules “bombarding” the capsule, typically (in simplified popular terms) “4dB” and lower in cold weather (-30C) than in hot weather (+50C). Extremely few condenser microphones have a low noise level, limited by “air noise”).

Regards

Addendum. Simple facts:

Loudest human voice: Jill Drake,129 dBA (https://www.guinnessworldrecords.com/wo ... individual)

Sound level to distance is a square function.

A “naked” (no amplifiers) open air symphony orchestra easily produces a general level of 90 dB or so at 1 m. A trained opera singer (no amplifier) is easily heard by over the orchestra. Let’s say 96dB at 1 m.

Now, take a microphone, put it at 1m (40 inches) and you get 96 dB from the singer. At 50cm/20 inches, you get 102 dB, at 25 cm/10 inches you have 108 dB SPL if I’m correct (see https://www.engineeringtoolbox.com/inve ... d_890.html). There are other effects at play, but the gist of it all, is that most trained singers can deliver extreme sound levels to a microphone.

Continue at own pace and understand why DPA microphones for “performance use” can handle extreme sound levels. Jet fighters a mile away are easily outperformed by trained singers with “enthusiasm”, when microphones are held 1 to 2 inches from the lips.

Smile!

One of the problems with 32bit float recording is the headphone monitoring, if your input is 30db low, it will record fine but you will need an extra 30db of headphone gain to hear it.

"The loudest known sound in history was the eruption of the Krakatoa volcano on August 27, 1883, which was estimated to be 310 decibels (dB)."


There are many sounds that can exceed the human voice; there are many sounds that will defeat the max SPL, of the best mics; even if these occurrences might be extremely rare; but a jet engine might be another example, more often encountered; at least IME. Dynamic mics, like Shure's take more SPL than condensers; though I feel there is little pedestrian about the trusty 416 ; mine have taken a lot of abuse over the years; and still deliver on spec.

I am also the only person in this thread who has discussed an actual use case for 32-bit float: viewtopic.php?f=2&t=206405#p1072760

Professionally 32 bit float is often used for special FX; an example might be rain in a thunderstorm with lightning; where the huge DR range from light patter to crack could both be accommodated, usefully.

Interesting about the firmware update at SD for 32bit float, for the Scorpio et al, I missed that, also Rob

You might not have to worry about levels with 32bit (though you still have to set them for the pre-amps); but the window into that 1500+ db still depends on the quality of those pre-amps, ADCs and mics too. Limiters and 24bit still have their place and relevance at times, also.

Granted that the mic clip thing is probably more theoretical than real, but if it's not stated upfront as a qualifier somebody will use the omission to demonstrate how dumb or incompetent you are for using 32 bit float. The same with preamps -- not an issue with SD equipment, and a marginal one for cheaper gear (the zoom $99 special is unobjectionable compared to the competition and quieter than sound gear used for years on low-budget film production), given the advantages of the format, but if you don't say so.....

This resistance to 32 bit float, and in some cases contempt for those who choose to use it, is puzzling.

Next we'll hear that color grading is only for incompetent shooters (get it right on set, doofus!). Or that log (and raw!) formats are for the hapless and insecure. Only wusses want to dial-in WB after the fact!

As someone who works solo (and unpaid) I've found the Rode Wireless Pro to be a game-changer. I no longer need to rely on a backup recorder, nor two level recording. I can quickly mic my victim up with a W/Pro TX (and lav if appropriate), plug the RX into the 3.5mm socket and not have to worry about levels. I can set the camera to a mid approximation that will record a usable track that, will often be just what I need. However, if that track needs much work on highs and lows, I can pull the 32 bit recording off the TX and sync it in.

Of course, I'm just an amateur hobbyist; I don't need broadcast quality. Good audio should be unnoticeable to the listener, so my measure of success is nobody commenting on it.

This resistance to 32 bit float, and in some cases contempt for those who choose to use it, is puzzling.

Often pure ignorance. Many people still think it's just more dynamic range. How often do you hear "24bit is more than enough if you know what you're doing".

They neither know that two separate ADCs record at different gain stages before anything "32 bit float" even enters the pipeline nor where the true advantages lie compared to regular recording.

This resistance to 32 bit float, and in some cases contempt for those who choose to use it, is puzzling.

Often pure ignorance. Many people still think it's just more dynamic range. How often do you hear "24bit is more than enough if you know what you're doing".

They neither know that two separate ADCs record at different gain stages before anything "32 bit float" even enters the pipeline nor where the true advantages lie compared to regular recording.

So you think float32 implies multiple ADCs?

(spoiler: it doesn't).

this guy...

This resistance to 32 bit float, and in some cases contempt for those who choose to use it, is puzzling.

John, I certainly have no resistance to it; it's a major technological advance. But I hardly ever use it; that is not to say it doesn't have it's place. My answers were in relation to Tim's original request for it in camera, when no camera has it from any make, as far as I am aware - I would first like to see limiters and ALC. For ENG particularly they have been next to essential over the years; in single cam op.

Even when I was a location recordist on Nagras; I often had Charlies scream down the mic, since they thought it was a laugh - I always monitored low, to protect my lugs. This is not to mention loud transients in war and riot neither. That at least was easier to deal with than being spat at and guns pulled on me.

Not all pre-amps have multiple ADCs neither. SD's Kashmir ones are pretty new in this regard. And that has nothing to do with 32 bit float neither - but it sure does squeeze more out of it.

There's a lot of assumption here we can't read and we don't know what 32 bit float might be about. Yes it's not just DR - that's was already more than adequate in the digital age and a lot better than the 60db or so SNR than we had to play with in the tape days. Sound recording is about so much more; and mic craft still comes top, in any serious recording.

There's a lot of assumption here we can't read and we don't know what 32 bit float might be about.

Hey Steve, the tech and pipeline is laid out quite clearly by manufacturers of 32bit float recorders. Maybe I misunderstand what you mean?


The hard-/software pipeline difference basically is this:

Screenshot 2024-08-19 at 18.07.38.jpg (50.24 KiB) Viewed 5224 times

Screenshot 2024-08-19 at 18.07.jpeg (60.48 KiB) Viewed 5224 times

Thanks Michel, yes it would stand to reason but there are pre-amps and there are others; and think we can both agree the SD's Kashmir ones are pretty good, that is all.

I might add apropos of earlier comments and nothing to do with 32 bit float; I've got both mix assist and Noise Assist plug-ins on my MixPre - the NR is very good and very useful; not far behind the Cedar. I use it a lot in VO work in less than dry conditions; but it is quite remarkable in high ambient conditions too - next to a roadway, for example. Yes it costs 300 bucks but it is well worth the extra. I've only got the one instance and that's really enough, mostly; but I'll get a 2nd one when the pennies allow, too.

Assuming, we're talking about the same video, I saw a Sound Speeds review on the Theos and what he said was that if you have the US version and you were in a foreign country, you'd still be locked out from monitoring. It would only give you the options to change the transmit power and frequency ranges in that country. He talks about it at 30 minutes in:

Interesting for me is that my plan is to buy the global version of the PR-2 in Canada so I can do interviews in Mexico. Perhaps I'll install the app on a Mexican phone.

The point he makes at 30:30 about using a Theos device, purchased in the U.S., in another country, is a different point from what I recall hearing or reading. He’s basically saying that a U.S. and Canadian Theos device are configured differently. That much seems to be clear. I recall a suggestion that the Deity app, when used in the U.S., might enforce the U.S. patent on a Theos device purchased in Canada. As I said, I have not seen any confirmation of this.

I might add apropos of earlier comments and nothing to do with 32 bit float; I've got both mix assist and Noise Assist plug-ins on my MixPre - the NR is very good and very useful; not far behind the Cedar. I use it a lot in VO work in less than dry conditions; but it is quite remarkable in high ambient conditions too - next to a roadway, for example. Yes it costs 300 bucks but it is well worth the extra. I've only got the one instance and that's really enough, mostly; but I'll get a 2nd one when the pennies allow, too.

If Blackmagic wants to start adding sound recording features to its cameras, noise reduction is where the real value-added is. That’s why Sound Devices can demand a significant add-on premium for it. Leaving aside your outdoor roadway example, noise reduction addresses such common issues as an air conditioner system running in the background. In the U.S. and Canada, this time of year it’s a constant problem for indoor recording.

The quality of noise reduction that Røde, DJI, etc. are offering in 2.4GHz wireless units leaves a lot to be desired. Shure’s new MoveMic is a major step up in quality. This is not professional noise reduction, but competing 2.4GHz wireless systems pale in comparison:

MoveMic vs Mobile Phone Built-in Mic Sound Test Comparison | Shure

I might add apropos of earlier comments and nothing to do with 32 bit float; I've got both mix assist and Noise Assist plug-ins on my MixPre - the NR is very good and very useful; not far behind the Cedar. I use it a lot in VO work in less than dry conditions; but it is quite remarkable in high ambient conditions too - next to a roadway, for example. Yes it costs 300 bucks but it is well worth the extra. I've only got the one instance and that's really enough, mostly; but I'll get a 2nd one when the pennies allow, too.

If Blackmagic wants to start adding sound recording features to its cameras, noise reduction is where the real value-added is. That’s why Sound Devices can demand a significant add-on premium for it. Leaving aside your outdoor roadway example, noise reduction addresses such common issues as an air conditioner system running in the background. In the U.S. and Canada, this time of year it’s a constant problem for indoor recording.

The quality of noise reduction that Røde, DJI, etc. are offering in 2.4GHz wireless units leaves a lot to be desired. Shure’s new MoveMic is a major step up in quality. This is not professional noise reduction, but competing 2.4GHz wireless systems pale in comparison:

MoveMic vs Mobile Phone Built-in Mic Sound Test Comparison | Shure

I'll be honest, 20% Voice Isolation in Resolve easily gets rid of any air conditioner units. To such a degree that I don't even worry about them anymore knowing it's that easy to get rid of them. Still, if I can turn it off I will for the cleanest possible audio. Yet they're not really a worry these days.

I'll be honest, 20% Voice Isolation in Resolve easily gets rid of any air conditioner units. To such a degree that I don't even worry about them anymore knowing it's that easy to get rid of them. Still, if I can turn it off I will for the cleanest possible audio. Yet they're not really a worry these days.

Sound Devices can demand a premium for noise reduction because some people prefer to get sound right at capture.

The number one thing that I’ve learned about iZotope RX, which I’ve used since version 3, is to get my sound right so that I don’t have to use it.

I'll be honest, 20% Voice Isolation in Resolve easily gets rid of any air conditioner units. To such a degree that I don't even worry about them anymore knowing it's that easy to get rid of them.

You'd think it would go without saying that a correction you can easily handle in post, with way more processing power, monitoring and control than will ever be available in any camera, might be less urgent than a simple change of recording format which prevents defects which can never be fixed or at best with a kludge -- and which doesn't introduce artifacts.

But you'd be wrong, if we go by this thread. Alas.

You'd think it would go without saying that a correction you can easily handle in post, with way more processing power, monitoring and control than will ever be available in any camera, might be less urgent than a simple change of recording format which prevents defects which can never be fixed or at best with a kludge -- and which doesn't introduce artifacts.

But you'd be wrong, if we go by this thread. Alas.

John “fix it in post” Paine now characterizes the addition of 32-bit float to video cameras as “urgent”

Not that a single camera offers this feature, or that he’s shown a single example of him using 32-bit float.

In fact, so far I’m the only person in this thread who’s discussed an actual case of using 32-bit float and why I used it, and shown the related waveform.

I’m going to repeat the following, because I think that it’s useful advice about some very expensive software that I use on a quite powerful computer:

The number one thing that I’ve learned about iZotope RX, which I’ve used since version 3, is to get my sound right so that I don’t have to use it.

John “fix it in post” Paine now characterizes the addition of 32-bit float to video cameras as “urgent”

Not that a single camera offers this feature, or that he’s shown a single example of him using 32-bit float.

In fact, so far I’m the only person in this thread who’s discussed an actual case of using 32-bit float and why I used it, and shown the related waveform.

Depends on how you define “camera”.

Any iPhone 15 Pro (Max) is able to record up to four channels of IEEE 32-bit float as part of standard ProRES 422 at several qualities and frame rates - one example is Blackmagic Camara (since the first version).

The new Panasonic GH7 is also capable of recording IEEE 32-bit float internally (I assume a ProRES format is required), so…

Limited knowledge or outdated beliefs may hamper the argument.

Regards