Topic: Zoom F3

[datbrad]

Been thinking more about the application of gain in today's recorders and realized there's a distinct reason why 32bit float makes complete sense and leaving out traditional level controls a logical final step in the evolution of audio capture gear.

Most solid state recorders made today have trim knobs, but they are not true gain pots for adjusting preamp and/or analog gain ahead of the A/D. They are essentially for digital gain only. With fixed gain preamps and digitally controlled line levels in menu settings, having gain controls to adjust digital levels in the field isn't necessary.
Back in the day, we wanted to maximize the performance of preamps, since how hard they were driven made a difference in sound, and also prevent analog brick walling in the line level signal before it hits the A/D. With that structure now flipped, the only thing you can do wrong in the field is cause digital overloads. You can't stop analog brick walling, nor optimize the performance of the internal preamp like we did with recorders in the past.
Considering this, it occurs to me that 32bit float can't make you a better or worse taper, it simply removes a part of the equation that only has a negative risk, digital clipping, with no upside to being able to adjust that stage manually. With the evolution of recorders moving away from user control of any of the analog portion of the signal, if 32 bit float should become the standard for A/D, it's almost inevitable and probably should be.....
 

[yug du nord]

^I get what your saying..  I think years ago jerryfreak on board here put together a rig that ran mikes straight into a 24 bit ADC.  I think both for the reason that you said and for lower noise.  I might be wrong on that but I think those were his reasons.

And if I ever said (which I don’t think I did) that 32 bit float makes a person a worse taper, I wasn’t being literal about it.
I was just stirring a ‘devil advocate’s’ attitude about someone stating that it makes person a better taper.
And my wacky rant in my previous post was continuing it.
The only thing that can make a taper a worse taper is not taping.
Apologies to all for derailing the thread.

It’s been quite a week and I was practicing some top notch ‘ass-hattery’. 

If a recorder is in your pocket or gear bag, and the goal is to never look at it.. the F3 achieves that I think.
32 bit float aside, the F3 is exactly (minus the belt loops) what I’ve been waiting for..  a pocket sized recorder with full XLR’s, P48, and good enough battery life to run P48 for a full show.
Pretty amazing!!

[checht]

Esteemed Mr. North,

Your provocative points challenged many unquestioned assumptions. Thanks for that!

I own that I am a bit of a worrier, and that that can be distracting at a show. The combination of 9 hours of battery, 32 hours of SD card, and 32 bit float has been a game-changer for my experience. I imagine it's not so impactful for others.

One of the things that keeps me recording is that it continually presents opportunities to learn. Currently embarking on OMT4. Last time I ran more than 2 mics was Greeks '85, so really appreciating GutBucket's wisdom as a guide.

Don't see the need for any regrets. Your post was on topic and respectful as always.

Regards, Chris

[edited to fix typo]

[goodcooker]

Currently embarking on OMT4. Last time I ran more than 2 mics was Greeks '85, so really appreciating GutBucket's wisdom as a guide.

Not to veer even more off topic (or to minimize Gutbucket's contributions to our little corner of the world which are considerable and much appreciated) but the stereophonic zoom paper was written by Michael Williams and published by the AES in 1984.

[gaijin]

3rd outing:

The F3 is continuing to be a winner.  The gaff tape, while not ideal, is proving to be effective in avoiding erroneous stops.  Battery life continues to be great with the Eneloop Pros - I am never below 50% by the end of a ~ 1.5 hour set @ 32/48 w/ 48v phantom on.  Bag checks don't even see the Hermitshell case I have the F3 in opened up - it is small enough that it is entirely disregarded by security so far - obviously that will vary by venue.  I'm going to do a Herbie show soonish - he tends to have much tighter screening from experience, so it will likely get its real test then.

In terms of ease of stealth operation, it really doesn't get any simpler.  I am eyeballing levels based on house music before hitting record and dumping it till the end of show - not checking anything but an occasional glance at the red recording light (out of paranoia from the past lost recording on the first outing).  Results have been great so far and the 32-bit would make even significantly botched levels a non-issue.

This time there was far too much bass so I've done some minor post to the samples as it was not listenable as it was. Still too much cymbal/hh "crashiness" going on for my taste so I did minor subtractive EQ on that as well.

All things considered, after 3 outings with different types of venues / different mic setups / etc I am giving this one a big thumbs up.  As mentioned previously I am inclined to use the MixPre3-ii whenever there is no concern for security rummaging through the bag, based on my subjective assessment that it just sounds "better" raw from device, but when stealth is a factor I no longer feel like I am sacrificing as much as I did with the stealth oriented Sony A10, D50, D1, M10 / Sony XLR-1 /  Battery Box setups I was using previously.

Recorded directly behind soundboard using a slightly narrow (practical stealth issues) NOS on stealth "object" peeking over the rail ~ 6ft high (very low "balcony") @ DFC, FOB with Line Audio CM4 > F3. 

Artist: Keiko Matsui
Venue: Bethesda Blues and Jazz Club, Bethesda, MD.

Sample 1:
https://we.tl/t-Nc0HP9SPXl

Sample 2:
https://we.tl/t-kxPahRPcVZ

Follow on from the above - made it in to the Herbie show with a minor deflection.  *Looks at device* ... "Is that recording equipment?" ... "No" ... "Thanks, keep your mask on during the show please!".   Here's sample from the show with subtractive EQ to remove excessive bass and some added gain. 

https://we.tl/t-pMfQN6tla4

Line Audio CM4 > Zoom F3 (P48 on, 32 / 48) > r8brain dither to 24, 48 > FLAC. 

Final verdict - I'm entirely sold on this thing.  It has everything you could ask for (minus M/S decode) and nothing you wouldn't (e.g. sub-par built-in mics).  As a bonus, it's discrete enough to pass through persnickety venue checks and can even dodge a minor security interrogation due to its size / appearance.

[Gutbucket]

Currently embarking on OMT4. Last time I ran more than 2 mics was Greeks '85, so really appreciating GutBucket's wisdom as a guide.

Not to veer even more off topic (or to minimize Gutbucket's contributions to our little corner of the world which are considerable and much appreciated) but the stereophonic zoom paper was written by Michael Williams and published by the AES in 1984.

Yes! I make no claims to the ownership of good ideas, but I will champion and downright proselytize for what has served to empirically illuminate and expedite my own pursuit of greater taper enlightenment.  William's Stereo Zoom was and continues to be a foundational cornerstone in much of that.  I believe the most tangibly useful single thing I've contributed to the taper community here is the improved PAS table - which at it's core is 100% William's Stereo Zoom.  All I did was carefully think it through and repackage it in such a way as to make it directly applicable to tapers and easy to use.

However, I do think the OMT multichannel approach is different in an important way, being less derivative and representing something of a different paradigm entirely, even-though the basic Stereo Zoom relationship remains fundamental within it.  The combination of multiple channels of a purposefully-arranged array of microphones set up so as to optimize the interactions between them represents something quite different, requiring the reassessment of what was widely known to work well for a pair of microphones alone.  The basic acoustics and signal aspect underpinnings remain the same, but to effect something greater than what might otherwise come from the simple sum of stereo pairs requires the reconsideration of the ways in which a single stereo pair is best optimized.  The parsing of what works for a single pair and also applies to the interactions of a purposefully designed multiple microphone arrangement verses what doesn't is something of a brave new world.  My thanks to this forum as the place to discuss exploration of working this stuff out!

[aaronji]

So will the F3 squash a Mytek or AD2K or SD or any other 24bit AD??
Or does it just make it easier??
That’s my question.

I think it just makes it easier. 24-bit ADs are more than sufficient for any music recording (even 16-bit, for that matter).

And is anybody running a preamp in front of the F3 or does that defeat the purpose?
Are all 32 bit float dual ADC’s the same?

If you have a preamp with a flavor you like, or that pairs well with your mics, I think it still makes sense to run one. Also, as I understand it, there aren't really 32-bit float ADCs; these units use multiple 24-bit ADCs, with differing gains, and combine the results post-conversion.

With respect to Williams and the Stereophonic Zoom, I agree with Gutbucket. It's illuminating and time hasn't really dimmed it much. I recall reading that it was based on rather limited data, though, so maybe it could be tweaked a little. A couple of degrees or cm here or there, perhaps, but the take-homes probably wouldn't change much...

[Gutbucket]

The Williams curves are close enough for our use.  There are a few alternate data sets that differ somewhat, I think partly based on the signal used as stimulus for the testing: impulsive clicks, sinewaves, noise, speech, music or whatever, but can't recall specifics as it's been years since I looked that far into it. [confirmed, see edit note below..]

For deriving all the values in the Improved PAS Table I used Eberhard Sengpiel's data, in the form of the online sengpielaudio visualizer rather than William's published curves. 

Here are two links to PDF's hosted at www.sengpielaudio.com showing the differences in the underlying trade-off curves derived by various researchers:
http://www.sengpielaudio.com/InterchannelLevelDifferenceTimeDifference1.pdf
http://www.sengpielaudio.com/InterchannelLevelDifferenceTimeDifference2.pdf

Differences sure, but leveraging the general trend of the underlying relationship is what is important for informing microphone setup.


Edit-
Note found on the bottom of the page of the first PDF linked above reads:
Red curve: Master's Thesis by Gert Simonsen at the Technical University of Denmark Lyngby (1984), also
used by Michael Williams. Test signals: maracas and claves clicks. http://www.sengpielaudio.com/Claves.htm
Blue curve: Eberhard Sengpiel and tonmeister students (1992) used as test signals monaural human speech
and a Beethoven string quartet in a usual control room with Bowers & Wilkens 801 matrix 3 loudspeakers.

[aaronji]

^ Yes, that is the article to which I was referring. One thing I can say: I am very glad that Dr. Sengpiel's website lives on despite his passing. It might not be beautiful, but there is a TON of useful information there...

[eman]

FYI, the F3 looks like it is recording when it isn't. Make sure the red light is on. I thought I had a lot of time and told the GF how to start it, and apparently she hit record twice which turns it back off. https://archive.org/details/deadco2022-06-14.m60.flac16 Second recording with my new Telefunken M60s, they are flat out amazing.

[datbrad]

Hope you weren't too hard on her, sounds like an easy mistake to make during daylight hours. I know it burns memory, but I usually start the deck if I need to step away within 30 min of the expected start time. If I get back in time, I just stop the recording and restart it to create a new file, then later I delete the junk file.

[morst]

back to off topic, Neumann has a great app called Neumann Recording Tools, for android and iOS, which calculates recording angle for nearly any pair.
review here: https://www.soundonsound.com/reviews/neumann-recording-tools
https://apps.apple.com/us/app/recording-tools/id576702914 iOS
https://play.google.com/store/apps/details?id=air.com.neumann.recordingtools&hl=en_US&gl=US Google Play Store

[eman]

No, I wouldn't blame a recording error on anyone but myself. If anything I considered not saying anything but was honest about it and just shared the info to show what we are up against getting good recordings. So many stupid things to go wrong, and I'm not the most fastidious taper around. I would have started the recording but the last one I screwed up was due to batteries running out in the F3 and I'm not 100% confident there. I ended up with 2 bars left on the battery meter after recording just the second set using my Powerex AAs.

[Gutbucket]

FYI, the F3 looks like it is recording when it isn't. Make sure the red light is on.

Is this behavior common to all Zoom recorders?  It is to the F8 (the only Zoom I have much on-hands experience with), which remains in essentially the equivalent of record pause mode whenever it is turned on and not recording or playing, with phantom power on and meters bouncing if mics are attached.  Difference may be that on the F8 it's pretty obvious when actually recording as the record button becomes illuminated solid red, in addition to the indication of  recording on the display. 

I notice this all the time because I rarely disconnect my pre-rigged multichannel microphone array from the recorder.  Both occupy the same bag, hooked up at all times.  So when playing back files directly from the recorder through the car stereo, head-phones or whatever other playback system I may plug into, as soon as playback ends or is stopped the mics immediately become active.

[voltronic]

FYI, the F3 looks like it is recording when it isn't. Make sure the red light is on.

Is this behavior common to all Zoom recorders?  It is to the F8 (the only Zoom I have much on-hands experience with), which remains in essentially the equivalent of record pause mode whenever it is turned on and not recording or playing, with phantom power on and meters bouncing if mics are attached.  Difference may be that on the F8 it's pretty obvious when actually recording as the record button becomes illuminated solid red, in addition to the indication of  recording on the display. 

I notice this all the time because I rarely disconnect my pre-rigged multichannel microphone array from the recorder.  Both occupy the same bag, hooked up at all times.  So when playing back files directly from the recorder through the car stereo, head-phones or whatever other playback system I may plug into, as soon as playback ends or is stopped the mics immediately become active.

The F6 illuminates the button solid red in the same way. I still can't wrap my head around the physical interface choices on the F3.