MixPre II recorders- 32bit float, internal TC generator, lotso dynamic range

Paul Isaacs said:
Started shipping to distributors on Friday just gone.
We use our own patented multiple ADC topology to give the extra wide dynamic range of >142dB. Sorry - can't share any more details than that.
Click to expand...

I see the AKM 5578 https://www.akm.com/akm/en/product/datasheet1/?partno=AK5578EN goes from 121dB to 130dB when using 8 to 1 mode; multiple ADCs = more DR. In cameras, using dual voltage summation to increase DR has been helpful (e.g. ALEV III for the Alexa sensor). Searching the USPTO found this: https://patents.google.com/patent/US9654134B2/en, which also showed up in SD marketing material when searching for the patent number.

Makes sense:
In accordance with this aspect of the invention, windowing functions are repeatedly applied to output frames from two or more ADC's having different analog input gains. A function is then applied to align (or translate) data frames from one ADC output to closely match those from another ADC output prior to conversion of these to window-packets. A selector is used to choose which window-packets supplied to a combiner are subsequently combined to provide a high-dynamic range digital output.
Click to expand...

So that maps well to the high DR afforded by 32-bit float, cool. More ADCs = higher power requirements and less battery time, or was it possible to maintain similar power usage through iterative optimization?

Kudos for being 1st to market with float recorders! (application filed Feb 2016, curious if there's design overlap with Zoom's F6 and any reason for their delayed release).
 
Noiz2 said:
And solid state recording media. I had a Nagra digital recorder crap out on me recording an F-14 starting up because the low end caused a harddrive crash! The fly bys would have benefited from the 32bit float since you didn't actually hear the plane till it was past, it was going kind of fast ;~) It took three passes before I got the levels low enough.
Click to expand...

Wild! For something like that putting the mic and recorder in a dampened box would be helpful!

I grew up in San Diego- we frequently went to the Miramar Air Show, then Top Gun (later MCAS Miramar). Probably have some old 8mm tapes in a box somewhere. While the Navy has arguably the best pilots (night time carrier landings in storms = max skill level), the F16 and F15 demos (especially the one year they flew an F16N, which is ultralight and maneuverable, piloted by the F16 development test pilot) were the best flying jets. The later F22 is/was too heavy (F35 more so). The F16(N) agile movement was incredibly amazing, and the F15 accelerating straight up on full afterburners like a rocket while they played Proud to be an American still evokes powerful emotions! As a kid I'm sure my recordings clipped, but we didn't care- still amazing!
 
I was actually recording at an air show a bit farther north. It was a short lived project to do a series on arial acrobatics (turns out they hate to be called stunt pilots). As part of that we got into a lot of places the audience cant go and get up close and personal with some really interesting planes. I also have some recordings looking in to the back end of a MIG as he taxied away. Basically like a huge hair drier ;~) I think we would have had issues with FAA regulations but we were always on private airfields and military bases. I know one of the cameramen was ex air force and he was amazed they let us on to an active runway. Got buzzed (as in I and the camera man dove for the dirt)
by a small plane that took off and passed us (at the end of the runway) at about 10 feet, upside down. The other pilots fell out of their chairs laughing. They said didn't anyone tell you? Rocky always takes off like that.

Fun times
 
Yep, we recorded some formula 1 races and my partner put on his official "I'm with the press" face and managed to get down to pit row. I didn't but it was LOUD. He also got the short straw and got to ride some practice laps in a formula D car. Not so loud but scary as hell. He got through the first lap and was about to unbuckle when the driver said I wouldn't do that we have another lap to go... Never did any munitions recordings. I almost had a chance once. I was in this town in the middle of nowhere for an ad and at the end of town was a munitions disposal facility. And they apparently set stuff off regularly, but unfortunately not while was in town.

I used some of the clipped out recordings as tests for declipping software. Which is why I bought Izotope RX, it was the only software I found that managed to get those passes to be semi usable. They are still distorted but they came out more like overloaded analog tape rather than digital hash. Quite an accomplishment.

The loudest thing I think I have actually experienced was playback of Beethoven's fifth on a maxed out Meyers theatre system that I did to prove to a director that we could get his show loud enough. We are talking around 10,000 Watts in a 400 seat theatre. I was in the booth and he was in the theatre and it got so loud that people in the Golds Gym across the street came out to find out what was happening. The directors comment was "I guess it will do". We never came close to that level in the show because the audience couldn't stand it.
 
  • Like
Reactions: jcs
Anyone know if I can sync my Blackmagic pocket 4k camera which takes timecocode from its 3.5mm jack and sync it with the Mixpre 3 ll or do you still need to use something like a tentacle sync to keep sync between the two? I read the manual and different recorders have different options for timecode in and out and I'm not sure I understood those details. I'm curious specifically about the mixpre 3 ll. Thanks
 
mico said:
Anyone know if I can sync my Blackmagic pocket 4k camera which takes timecocode from its 3.5mm jack and sync it with the Mixpre 3 ll or do you still need to use something like a tentacle sync to keep sync between the two? I read the manual and different recorders have different options for timecode in and out and I'm not sure I understood those details. I'm curious specifically about the mixpre 3 ll. Thanks
Click to expand...

For the MixPre-3 II, it takes TC in only. It does not send it out. Options for TC in are the mini HDMI in and the 3.5mm aux in.

Your best bet is to use Tentacle Sync boxes.
 

As noted earlier, here's Shure's info on the SM58 and high SPL: https://www.shure.com/en-US/support...rophone-handle-really-loud-sounds-maximum-spl

Looks like SD can handle this at the analog stage without needing an attenuator- cool. So basically after attaching the mic the only thing you need to worry about are the limits of the microphone: self noise and clipping. Curious how the SM7B and ribbon mics sound (mics needing a lot of gain).

Paul- were the ambisonic and headphone level issues addressed for this release?


MixPre II 3 & 10 in stock: https://www.bhphotovideo.com/c/search?Ntt=mix pre 2&N=0&InitialSearch=yes&sts=ma&Top+Nav-Search=
 
Alex H. said:
For the MixPre-3 II, it takes TC in only. It does not send it out. Options for TC in are the mini HDMI in and the 3.5mm aux in.

Your best bet is to use Tentacle Sync boxes.
Click to expand...

That is not correct. All MixPre-IIs can output timecode including the MixPre-3 II. You switch the stereo out from audio output to LTC output in the Outputs menu.

Paul
 
Paul Isaacs said:
That is not correct. All MixPre-IIs can output timecode including the MixPre-3 II. You switch the stereo out from audio output to LTC output in the Outputs menu.

Paul
Click to expand...

Thanks for the correction, Paul! I had read through the user manual and somehow missed that function! Though, that does sacrifice the audio output, so no cam hop or Comtek in that case. So, it’s still better to use external TC boxes.

I was, though, wondering why the 3 II and 6 II, with the addition of full TC support and the USB drive transfer, were not also given internal sound reports. Yeah, I can always pop the thumb drive over to my laptop and generate a report with Wave Agent, but it’s nice just to hand the thumb drive off to the DIT (or the client) straight from the recorder.

About to upgrade my 6 to the 6 II. It’s primarily my traveling sound design recorder for grabbing ambient tracks and SFX wherever I go, but on very small shoots (and especially low-budget) it’s really nice to have something so compact that can handle the workload with zero gear-carrying fatigue.
 
Alex H. said:
Thanks for the correction, Paul! I had read through the user manual and somehow missed that function! Though, that does sacrifice the audio output, so no cam hop or Comtek in that case. So, it’s still better to use external TC boxes.

I was, though, wondering why the 3 II and 6 II, with the addition of full TC support and the USB drive transfer, were not also given internal sound reports. Yeah, I can always pop the thumb drive over to my laptop and generate a report with Wave Agent, but it’s nice just to hand the thumb drive off to the DIT (or the client) straight from the recorder.

About to upgrade my 6 to the 6 II. It’s primarily my traveling sound design recorder for grabbing ambient tracks and SFX wherever I go, but on very small shoots (and especially low-budget) it’s really nice to have something so compact that can handle the workload with zero gear-carrying fatigue.
Click to expand...

Your request is noted Alex!

thx

Paul
 
If 32bit float is as amazing as it sounds, I wonder why that option wasn't incorporated into the new Scorpio or 833 that was just announced?
 
Not being sarcastic, but perhaps three quick guess:

1) Perhaps SD thinks users of Scorpios and 833s understand how to control levels and are fairly happy with the limiters in those devices (not that they expect to use them much), and don't want to kick the can down to post...

2) Perhaps the advantages of 32-bit float are more important to people who get levels wrong and/or want to record music or just about anything in uncontrolled environments, rather than the more (maybe) dialog-focused uses of Scorpios, 833s, etc. So OMBs rolling camera & sound and have way too much to do, Jam-band fans who are, um, distracted, sound designers...

3) Perhaps some 32-bit stuff is in Scorpio & 833 but hasn't been activated yet and SD doesn't want to announce such a thing until it's closer to real. Says both have really robust and extensible processing engines. We may need to wait at least until SD staffers are back from IBC to find out.


In the meantime, these two articles, which have probably already been posted here, make for interesting reading (and listening/tweaking):


32-bit Float Files Explained
Posted on August 29, 2019
The MixPre II models introduce the ability to record 32-bit floating point WAV files. For ultra-high-dynamic-range recording, 32-bit float is an ideal recording format. The primary benefit of these files is their ability to record signals exceeding 0 dBFS. There is in fact so much headroom that from a fidelity standpoint, it doesn’t matter where gains are set while recording. Audio levels in the 32-bit float WAV file can be adjusted up or down after recording with most major DAW software with no added noise or distortion. To understand the nuts and bolts of 32-bit files, keep reading. This paper discusses the differences between 16-bit fixed point, 24-bit fixed point, and 32-bit floating point files.
https://www.sounddevices.com/32-bit-float-files-explained/


Sample 32-bit float and 24-bit fixed WAV files
Posted on August 29, 2019
These [downloadable] samples were recorded on a MixPre-3 II to demonstrate the advantage of using 32-bit float WAV files for recording. We split a signal from a MKH40 mic into two MixPre-3 IIs, one recording 24-bit fixed WAVs, and the other 32-bit float WAVs. We then applied too much gain to loud dialogue. Low-cut was set to 80 Hz, and no limiters were active. Both the original 32-bit float and 24-bit fixed file are heavily distorted and unusable.
https://www.sounddevices.com/sample-32-bit-float-and-24-bit-fixed-wav-files/
 
It just seems that the capability of capturing such a wide dynamic range of undistorted sound without limiters (which you can hear when active) would be of value to everybody. Insurance.



Jim Feeley said:
Not being sarcastic, but perhaps three quick guess:

1) Perhaps SD thinks users of Scorpios and 833s understand how to control levels and are fairly happy with the limiters in those devices (not that they expect to use them much), and don't want to kick the can down to post...

2) Perhaps the advantages of 32-bit float are more important to people who get levels wrong and/or want to record music or just about anything in uncontrolled environments, rather than the more (maybe) dialog-focused uses of Scorpios, 833s, etc. So OMBs rolling camera & sound and have way too much to do, Jam-band fans who are, um, distracted, sound designers...

3) Perhaps some 32-bit stuff is in Scorpio & 833 but hasn't been activated yet and SD doesn't want to announce such a thing until it's closer to real. Says both have really robust and extensible processing engines. We may need to wait at least until SD staffers are back from IBC to find out.


In the meantime, these two articles, which have probably already been posted here, make for interesting reading (and listening/tweaking):


32-bit Float Files Explained
Posted on August 29, 2019
The MixPre II models introduce the ability to record 32-bit floating point WAV files. For ultra-high-dynamic-range recording, 32-bit float is an ideal recording format. The primary benefit of these files is their ability to record signals exceeding 0 dBFS. There is in fact so much headroom that from a fidelity standpoint, it doesn’t matter where gains are set while recording. Audio levels in the 32-bit float WAV file can be adjusted up or down after recording with most major DAW software with no added noise or distortion. To understand the nuts and bolts of 32-bit files, keep reading. This paper discusses the differences between 16-bit fixed point, 24-bit fixed point, and 32-bit floating point files.
https://www.sounddevices.com/32-bit-float-files-explained/


Sample 32-bit float and 24-bit fixed WAV files
Posted on August 29, 2019
These [downloadable] samples were recorded on a MixPre-3 II to demonstrate the advantage of using 32-bit float WAV files for recording. We split a signal from a MKH40 mic into two MixPre-3 IIs, one recording 24-bit fixed WAVs, and the other 32-bit float WAVs. We then applied too much gain to loud dialogue. Low-cut was set to 80 Hz, and no limiters were active. Both the original 32-bit float and 24-bit fixed file are heavily distorted and unusable.
https://www.sounddevices.com/sample-32-bit-float-and-24-bit-fixed-wav-files/
Click to expand...
 
  • Like
Reactions: jcs
DNN said:
It just seems that the capability of capturing such a wide dynamic range of undistorted sound without limiters (which you can hear when active) would be of value to everybody. Insurance.
Click to expand...

I bet that production pros are a lot more conservative when it comes to change, so investing a lot of money into something that they are not familiar with might not be a good fit.
 
Reading the patent it appears they are using multiple ADCs (more AKM chips- looks like they’re already using e.g. 4 to 2 or 8 to 1 modes to extend dynamic range).

More chips = more power and complexity. Perhaps 10 channels in that form factor is the practical limit right now.

32-bit float’s only drawback is 1/3 more disk space.
32-bit float will be standard in the near future (no clipping or limiter artifacts, ever (unless u manage to clip the mic)).
 
Last edited:
TheDingo said:
I bet that production pros are a lot more conservative when it comes to change, so investing a lot of money into something that they are not familiar with might not be a good fit.
Click to expand...

Perhaps it's helpful to view 32-bit float as analogous to auto-focus: A useful too that's not needed or even beneficial to plenty of productions.

Yes, before too long 32-bit float will be more common and more accepted; it's not a bad thing, but if the 833 doesn't have the ability to implement 32-bit float, that won't affect my decision to buy one or not (and I am considering buying one).

Jim "does a lot of location audio work" Feeley
 
jcs said:
Reading the patent it appears they are using multiple ADCs (more AKM chips- looks like they’re already using e.g. 4 to 2 or 8 to 1 modes to extend dynamic range).

More chips = more power and complexity. Perhaps 10 channels in that form factor is the practical limit right now.

32-bit float’s only drawback is 1/3 more disk space.
32-bit float will be standard in the near future (no clipping or limiter artifacts, ever (unless u manage to clip the mic)).
Click to expand...

Is 33% really that big when it comes to audio? If it was video, I would unstand, but audio is way smaller.
 
32-bit float’s only drawback is 1/3 more disk space.
32-bit float will be standard in the near future (no clipping or limiter artifacts, ever (unless u manage to clip the mic)).
Click to expand...

Maybe. In my experience film sound is about ten years behind music production and about five behind theatre. That gap is probably shrinking but? Big film is VERY conservative, there is a lot of $ involved if you make a mistake. If music moves over and all the DAW's come with 32 bit float built in then maybe film will start to budge. In 2005 or so I was at a post studio in LA and they were still using Mac's with SCSI drives and some very vintage version of PT.

So it might be common in some fields in the near future but I wouldn't count on it as standard in production film work all that soon. Now the SFX folks are going to jump all over it, they will just convert files or post will have assistants converting files. They are not that likely to switch over in post either for a couple of reasons. One not enough equipment supports it right now. If I dropped a few million on a digital mixer I'm not looking at allowing 32bit in till that mixer gets an update. And post places HATE to have a library of mixed file types. So if I'm a post house and we decide to go 32bit someone is going to get paid to convert the libraries.

In post for everything but micro budget no crew films it is all about compatibility. If I can't hand off my sessions to the mix with out having someone do conversions it's not happening.

So I do believe it will happen but at least for films with a budget it's going to take awhile.

As an example Soundtrack Pro and FCP 6 used 32 bit float in 2006, and we are talking about it like it is new 13 years later.
 
Noiz2 said:
Maybe. In my experience film sound is about ten years behind music production and about five behind theatre. That gap is probably shrinking but? Big film is VERY conservative, there is a lot of $ involved if you make a mistake. If music moves over and all the DAW's come with 32 bit float built in then maybe film will start to budge. In 2005 or so I was at a post studio in LA and they were still using Mac's with SCSI drives and some very vintage version of PT.

So it might be common in some fields in the near future but I wouldn't count on it as standard in production film work all that soon. Now the SFX folks are going to jump all over it, they will just convert files or post will have assistants converting files. They are not that likely to switch over in post either for a couple of reasons. One not enough equipment supports it right now. If I dropped a few million on a digital mixer I'm not looking at allowing 32bit in till that mixer gets an update. And post places HATE to have a library of mixed file types. So if I'm a post house and we decide to go 32bit someone is going to get paid to convert the libraries.

In post for everything but micro budget no crew films it is all about compatibility. If I can't hand off my sessions to the mix with out having someone do conversions it's not happening.

So I do believe it will happen but at least for films with a budget it's going to take awhile.

As an example Soundtrack Pro and FCP 6 used 32 bit float in 2006, and we are talking about it like it is new 13 years later.
Click to expand...

Yeah Scott float support goes, way way, back, I think circa 1998-2000 32-bit float (singles) became faster than integer for doing the math (DSP / digital signal processing). I remember doing tests with assembly language 32/64-bit fixed point vs. simple float 32 and was thrilled float was faster! And no more worrying about range issues. That's another reason float is so fast: we don't need to check range before/after each operation to prevent over/underflow for typical audio DSP (and even physics simulations (there are checks after integration finishes, for example, but not on every operation)).

Pro Tools is still the gold standard for pro audio work- it and every audio tool I'm aware of supports 32-bit float. Even Audacity (free) supports 32-bit float. Reaper does DSP in 64-bit float (doubles): https://www.reaper.fm/ . Hardware has gotten so powerful, even phones can do 64-bit float with ease. Why process with 64-bit? Float-32 is 23 bits of fraction plus the sign bit and an 8-bit exponent. Float-64 (doubles) provide 52 bits of fraction. For some detailed audio work, this can make a difference apparently (I haven't personally A/B'd). Pro Tools also performs DSP in 64-bit float (doubles).

Converting 32-bit float to 24-bit integer without editing it first could result in clipping or loss of information (levels recorded too high or too low).

The only drawback is disk space- if 1/3 more disk cost affects a tight budget, that could nix recording in 32-bit float (or limit it for uncontrolled situations).
 
You must log in or register to reply here.
Back
Top