The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Gear for sale     Latest  Trending
NAB 2019: New Zoom F6
Old 9th April 2019
  #1
Lives for gear
 
DirkP's Avatar
 

NAB 2019: New Zoom F6

Just detected this on YT:



On edit:

No more gain setting, it is like RAW for audio.

Quote:
Zoom just announced their new F6 field recorder for location sound audio recording. Rather than just tweak the form factor, which they did, they also moved to dual analogue to digital converters which record to 32 bit. So the way you record changes pretty dramatically. When recording to 32 bit, you don't set the gain trim (there is none), you simply adjust the fader for each input channel. If you missed the level, no problem, in post you can cleanly normalize to the levels you need with no degradation in audio quality. They've also switched up a few other things: The screen is a slightly smaller version of the high resolution screen from the F8n, the back of the recorder is a Sony NP-F (L series) battery sled, and you can now power the F6 via its USB-C input. The same quality timecode generator from the F8n is on board, this time with a 3.5mm TRS input/output jack. The F6 has a single SD card slot for recording media and still includes the advanced hybrid limiter when you record in 24 bit. Auto mix is included and the with an adapter, you can use the Zoom Control app for iOS. You can also use the Zoom F-Control control surface. We'll have a full review after the F6 starts shipping, hopefully in June 2019. Pricing still to be announced.
Old 10th April 2019
  #2
Lives for gear
Several DAWs allow 32 bit FP recording, but it's usually dismissed with "why would you want more than 24....that's already more than sufficient for all conceivable real world dynamic ranges recordable" ? However this is a new implementation with 2 AD's (sliding/variable ?) so different to anything we have thus far ?
Old 10th April 2019
  #3
Lives for gear
 
jimjazzdad's Avatar
Quote:
Originally Posted by studer58 View Post
Several DAWs allow 32 bit FP recording, but it's usually dismissed with "why would you want more than 24....that's already more than sufficient for all conceivable real world dynamic ranges recordable" ? However this is a new implementation with 2 AD's (sliding/variable ?) so different to anything we have thus far ?
I believe this is the same AD approach as used by Sonosax and Zaxcom ; it would seem that this technology has now come down to the Zoom price point...

EDIT: Not sure if it is possible to record 32 bit wav files with the SX-R4+? (It was discussed obtusely in this thread)

Last edited by jimjazzdad; 10th April 2019 at 01:52 PM.. Reason: Question added
Old 10th April 2019
  #4
Gear Addict
 

So here's a naive question: A couple of years ago I bought a QSC Touchmix digital mixer for live sound; it works as a recorder too (the Touchmix 8 is about the size of a large laptop) and records multitrack 32-bit floating point WAV files at either 44.1 KHz or 48 KHz; you can choose to record pre-fader/FX or post-fader/FX and you can also just record the stereo mix if you don't want the ISO files.

What I don't understand is whether the fact that the Touchmix records 32-bit floating point WAV files automatically gives me that same degree of dynamic range touted in the new Zoom F6 (i.e., if I'm using it only for recording, not for live sound, can I keep the analog gain low to allow plenty of headroom and not worry about trying to get to a target level like -18dBFs), or whether there's something special going on with the A-D converter(s) used in the Zoom F6 that makes it possible on that unit but not on my Touchmix?

I hadn't seriously considered the Touchmix for recording since it's more of a prosumer machine designed primarily for live sound, but it might be worth a second look.
Old 10th April 2019
  #5
Lives for gear
 
jimjazzdad's Avatar
Quote:
Originally Posted by bradh View Post
...What I don't understand is whether the fact that the Touchmix records 32-bit floating point WAV files automatically gives me that same degree of dynamic range touted in the new Zoom F6 (i.e., if I'm using it only for recording, not for live sound, can I keep the analog gain low to allow plenty of headroom and not worry about trying to get to a target level like -18dBFs), or whether there's something special going on with the A-D converter(s) used in the Zoom F6 that makes it possible on that unit but not on my Touchmix?...
My understanding is that each input on the Zoom F6 uses two preamps feeding two ADs, one 'low gain' preamp writing to bits 1-16 and one 'high gain' preamp writing to bits 17-32. In this way an extremely high dynamic range (of perhaps 135 dB?) is captured in the 32 bit wav (I credit Hugh Robjohns with explaining this - I hope I have understood his explanation). In short, I don't think your Touchmix has this sort of input design.

Last edited by jimjazzdad; 10th April 2019 at 03:25 PM.. Reason: clarity
Old 10th April 2019
  #6
Gear Maniac
 

Quote:
Originally Posted by jimjazzdad View Post
My understanding is that each input on the Zoom F6 uses two preamps feeding two ADs, one 'low gain' preamp writing to bits 1-16 and one 'high gain' preamp writing to bits 17-32. In this way an extremely high dynamic range (of perhaps 135 dB?) is captured in the 32 bit wav (I credit Hugh Robjohns with explaining this - I hope I have understood his explanation). In short, I don't think your Touchmix has this sort of input design.
Hugh's comments - in the SOS forum - aren't based, as he says himself, on any knowledge of the F6. The interviews at NAB only refer to dual ADCs, not dual pre-amps (indeed, the refs in the interviews to the F6 having the same preamps as the F4 and F8/F8n appear to rather counter that). For another example of dual ADCs (sans dual preamps) in a modestly-priced recorder, there is the Tascam DR100 Mk3, although the F6, of course, promises to be rather better.

But who knows?! The topology of more recent recorders is getting more complex and is not always helped by the absence of published sufficiently detailed block diagrams.

Incidentally the Gotham Sound interview at NAB is the best I have seen so far YouTube, but it will be interesting to see/hear real reviews of the F6 in due course.

Cheers,

Roland
Old 10th April 2019
  #7
Lives for gear
 
DirkP's Avatar
 

I still don't get it, please explain!
The preamps of the Zoom and the gain settings happen before AD, correct?
If I set the gain too low, I have an unwanted ratio between noise and usefull signal (Nutzsignal). If I later normalize the recording or whatever I do with the result, the ratio between noise and wanted signal remains the same.
What is it that I don't understand?
Old 10th April 2019
  #8
Lives for gear
 

The noise floor is 135db down so where you set your signal in the 32bit ‘pail’ you can always lift it up in your mix because there is virtually no noise.
Old 11th April 2019
  #9
Lives for gear
 
DirkP's Avatar
 

Quote:
Originally Posted by emenelton View Post
The noise floor is 135db down so where you set your signal in the 32bit ‘pail’ you can always lift it up in your mix because there is virtually no noise.
Thank you for your reply. Do you talk of the noise floor of the analog part? If so, where is the difference with the 32 bit. If the low noise floor is an attribute of the analog part - the preamps - I could make a recording with a 24 or 16 bit A/D and a gain setting "too low" and still the noise/wanted signal ratio would be relative good.
But how does the 32 bit A/D makes something irrelevant, I do "wrong" in the analog area before the AD?
Old 11th April 2019
  #10
Lives for gear
 

I can’t get into the weeds and defend it. The quality of the new dual ‘stuff’ is characterized by not needing to care about how hot you get your signal. Any nominal signal range, recorded, is no different from the signal recorded at a higher level.

Some here have taken exception to my summarizing so I am trying to be careful

The Sonoax has pre AD, a 20db gain switch only, the fader knobs only adjust the level of the signal after the digitizing. There’s no noise in the pres. Some of the users in video claim the only reason they adjust up the ‘knobs’ is so their post people see a waveform level they are accustomed to.
The Sonosax records a 24bit wave.
Old 22nd April 2019
  #11
Lives for gear
 
surflounge's Avatar
can the gain knobs be cranked all the way up without clipping?
Love to set-and-forget instead of "normalizing" dead signal after in computer.
Old 22nd April 2019
  #12
Gear Addict
 

I have a strong suspicion that the Sound Devices MixPre series (Kashmir preamps) are actually doing something very similar, but they disguise it behind a digital fader that they call preamp gain.

In the earliest release firmware, I discovered a bug where if the MixPre6 was set up to route individual tracks to headphones (rather than LR mix to headphones), the apparent channel gain reaching the headphones would be a fixed value not affected by the preamp gain knob of the channel. The channel preamp could be turned all the way down, not record any signal, and still it would reach the headphones as if the preamp were turned up to 75%.

SD fixed the bug but never answered my question about what that meant as to how the preamps are actually working and what the gainstaging inside the device is doing. Clearly though, the "preamps" are not doing what we normally expect preamps to do (amplify the signal by an adjustable amount before it goes anywhere else).

I also suspect this means that if they wanted to, they might be able to release a firmware update matching this feature of the F6. I don't know if theirs is dual-ADC or not though.

Last edited by dasbin; 22nd April 2019 at 06:00 PM..
Old 22nd April 2019
  #13
Lives for gear
 
DirkP's Avatar
 

Why should they "hide" this feature? Why compressors/limiters in the analog stage when you're using dual-ADC technology that renders limiters superfluous? They would have used 32 bit and dual ADCs etc. to sell their recorders just as Zoom is doing it now.
Old 23rd April 2019
  #14
Quote:
Originally Posted by DirkP View Post
Why should they "hide" this feature? Why compressors/limiters in the analog stage when you're using dual-ADC technology that renders limiters superfluous? They would have used 32 bit and dual ADCs etc. to sell their recorders just as Zoom is doing it now.
Because..... Zaxcom??

Maybe Sound Devices feel like they're spending enough money on their lawyers as it is! ha

(isn't there a legal battle at the moment between Zaxcom and Sound Devices over transmitter bodypacks being recorders too?? btw, Sound Devices own Audio Ltd now as well)
Old 23rd April 2019
  #15
Lives for gear
 

Quote:
Originally Posted by surflounge View Post
can the gain knobs be cranked all the way up without clipping?
Love to set-and-forget instead of "normalizing" dead signal after in computer.

You can definitely run the signal into clipping that way.
Old 4 weeks ago
  #16
Gear Maniac
 
Rtroost's Avatar
Any sign of this yet?
Old 4 weeks ago
  #17
Lives for gear
 

Website says: "Available Fall/Winter 2019"

- so, anywhere between 1 and 6 months. Patience ...
Old 4 weeks ago
  #18
Lives for gear
 
DirkP's Avatar
 

There was a notice on the Zoom page that due to a problem with one component the release is postponed for some month (otherwise it would have been already released).
I like their openess:

Quote:
ZOOM F6 PRODUCTION DELAY

During our initial production for the F6 Field Recorder, we discovered a mechanical component that needed modification to assure consistent, overall durability. As a result, delivery of the F6 Field Recorder has been rescheduled for later in the Fall/Winter of 2019.

At Zoom we have always been committed to providing the highest quality products possible. We apologize for the delay and appreciate your patience and understanding.




© 2019 ZOOM Corporation. All Rights Reserved.
https://zoom.co.jp/news/F6_production-delay
Old 3 weeks ago
  #19
Gear Addict
 
voltronic's Avatar
 

Dirk, thanks for relaying that info. The initial release was supposed to be late May / Early June, and they kept moving it again and again with no explanation until now.

Curious that this news item does not appear on the Zoom North America site.
Old 3 weeks ago
  #20
Gear Maniac
 

I can’t believe, that a small mechanical part can have such an impact. Probably the japanese way of saying, we have a bigger problem

Ronald

Last edited by RFrommann; 3 weeks ago at 12:31 PM.. Reason: Spelling
Old 3 weeks ago
  #21
Lives for gear
 
DirkP's Avatar
 

Better than Behringer and the 808. The german way:-)
Old 3 weeks ago
  #22
Lives for gear
 

It would be nice (but unrealistic) to think they may use the added time to further test and refine the product in the sympathetic hands of careful beta-testers who know how to avoid compromising the "durability" of the product in its existing form.

They probably have tens of thousands of circuit boards etc., sitting in store waiting for the redesign, testing and manufacture of the 'mechanical component', and probably do not want to risk finding another issue that might delay the product.

As speculation is an honoured pastime in this august forum, my instinct says maybe something to do with the battery compartment? Otherwise, the mounting of the main component board and mechanical stress? (If so, this might mean a redesign of the circuit board, so maybe my point in the second para is not relevant.

But, as elsewhere, we learn from our (or other's) mistakes. What does this tell us about the mindset of Zoom? They will hold back the release of a heavily promoted and highly anticipated product, because not to do so might impair their customer's perception of quality of the company's products. (It could also be their bean-counter's advice as to the cost of a subsequent recall if the release proceeded as scheduled? Hmmm... )
Old 3 weeks ago
  #23
Lives for gear
 

Quote:
Originally Posted by panatrope View Post
They will hold back the release of a heavily promoted and highly anticipated product, because not to do so might impair their customer's perception of quality of the company's products. (It could also be their bean-counter's advice as to the cost of a subsequent recall if the release proceeded as scheduled? Hmmm... )
They may have had a chat to Samsung.
Old 3 weeks ago
  #24
Lives for gear
 

Surely 32 bit A/D means a 32 bit fixed point integer A/D chip. One still has to scale the microphone input to the analog input range of the A/D regardless? Is there a free lunch?
Old 3 weeks ago
  #25
Gear Addict
 
voltronic's Avatar
 

Quote:
Originally Posted by David Spearritt View Post
Surely 32 bit A/D means a 32 bit fixed point integer A/D chip. One still has to scale the microphone input to the analog input range of the A/D regardless? Is there a free lunch?
It has dual 32-bit ADCs, and the analog signal gets routed to one or the other depending on the level.

This video has a simplistic explanation at the beginning, but a nice demonstration of 32-bit float recording benefits starting at 5:40:
https://www.youtube.com/watch?v=y4oNd1RgGL0
Old 3 weeks ago
  #26
Gear Addict
 
voltronic's Avatar
 

In this post, Mr. Satz explains how the system used in the F6 would actually work, and why the internal noise of the recorder will be especially important:

http://taperssection.com/index.php?t...803#msg2304803
Old 3 weeks ago
  #27
Lives for gear
 

Harking back to the earliest gain-switching system (say, the 3M DynaTrack) and subsequent gain-switching codecs (u-law/A-law in telephony, or the BBC's Sound-In-Sync or NICAM), noise modulation has always been the bugbear. This can be addressed in the gain-switching mechanism, assuming all other parameters are optimised.

Let's get out the technology speculum. At the pre-amp end, the limitations are pre-amp self noise and pre-amp output clipping (to a first order). The pre-amp supply rail voltage sets the output limitation; the self noise varies with input impedance and gain. At the low end we want to set the high gain pre-amp so that the self noise is as low as possible, below the self nose of the source, and of the quantization noise of the following ADC. At the high end we want unity gain (or maybe less) so that the highest input signal does not clip the pre-amp input. Unfortunately, the self noise of most unity gain pre-amps is a lot higher than the optimised gain version above. So neither pre-amp on its own will give anything like the dynamic range possible of the best convertors, say 125+ dB.

So, if we can combine two chains or pre-amp ADC, one at the lowest noise end, the other at the highest input end, we can achieve a wider dynamic range, providing we have a mechanism for seamlessly transitioning between the two. One way is to avoid a hard switch by having a graduated (level-dependent) and maybe less audible transition (cross-fade) between the two. However, such a system, unless carefully instrumented, can introduce dynamic distortion which may be as annoying as the more coarse switching noise modulation. It is possible to envisage some sort of digital computation process to manage the switching that computes the precise gain offset between the two paths (stable, assuming precise synchronisation of the sampling instant of the two ADCs), computes this with 32-bit precision, and therefore exactly gain matches them so that a cross fade between the two can be undertaken over some level range determined to be least audible. So, a trim-less low noise yet non-overloadable pre-amp/ADC system? Maybe ...

OK, my conjecture as to how such a system might be implemented. In the end, there is no such thing as perfection in the world of audio, but this may be a system that advances real-world sonic and operational benefits in the field. However, seems we may not find out as late as (northern) spring. Until then ... jus' keep speculatin'! (Sometimes I ask myself why I bother!)
Old 2 weeks ago
  #28
Lives for gear
 

As speculation appears to be in season (Apple, do I hear your name?), I will venture into an interesting area here, in relation to the just announced SD MixPre II series. Apparently stemming from a hardware upgrade (mainly processor and ADCs?), many new features have emerged for the new MixPre series II, not the least of which is 32-bit float recording and (claimed) 142dB dynamic range. Now where have I heard that before? Oh, yes, in relation to the subject of this thread .... release of said subject suddenly announced as delayed by months!

Now, my neurons in associative mode fired up in recalling similar circumstances in my (non-audio) professional life. I have noted in the SD manuals that the ADC technique in the SD MP II is protected by patents in USA and Canada and pending in other places. Could it be that Mr Z's 'trimless' input technique (of which they have provided no details, and no indication of patent) may have something in common with that of SD? To the extent that if they released the F6, they may find themselves owing licence fees to Mr SD? (And the urge to avoid paying licence fees to a competitor is one stronger even than sex ...)

So (the speculation continues) did they pull the F6 while they do a patent dodging re-engineering exercise, to avoid the above scenario? I have seen it in other fields. Possibly this one? To me, this makes more sense than "mechanical redesign of a critical mechanical component" as a reason to delay the product release. As always, the truth may be stranger than fiction (but maybe not as amusing ...)

I stress that I actually have no knowledge of the facts. Just my idle conjecture for which I may be castigated by William of Occam's Razor (the unkindest cut of all). Just remember, though, in choosing between conspiracy and stuff-up, always bet on the latter. Truth is the daughter of time ...

Offered in ignorance for your amusement ...
Old 2 weeks ago
  #29
Gear Head
Here's one of the patents.
https://pimg-fpiw.uspto.gov/fdd/34/541/096/0.pdf


Quote:
Originally Posted by panatrope View Post
As speculation appears to be in season (Apple, do I hear your name?), I will venture into an interesting area here, in relation to the just announced SD MixPre II series. Apparently stemming from a hardware upgrade (mainly processor and ADCs?), many new features have emerged for the new MixPre series II, not the least of which is 32-bit float recording and (claimed) 142dB dynamic range. Now where have I heard that before? Oh, yes, in relation to the subject of this thread .... release of said subject suddenly announced as delayed by months!

Now, my neurons in associative mode fired up in recalling similar circumstances in my (non-audio) professional life. I have noted in the SD manuals that the ADC technique in the SD MP II is protected by patents in USA and Canada and pending in other places. Could it be that Mr Z's 'trimless' input technique (of which they have provided no details, and no indication of patent) may have something in common with that of SD? To the extent that if they released the F6, they may find themselves owing licence fees to Mr SD? (And the urge to avoid paying licence fees to a competitor is one stronger even than sex ...)

So (the speculation continues) did they pull the F6 while they do a patent dodging re-engineering exercise, to avoid the above scenario? I have seen it in other fields. Possibly this one? To me, this makes more sense than "mechanical redesign of a critical mechanical component" as a reason to delay the product release. As always, the truth may be stranger than fiction (but maybe not as amusing ...)

I stress that I actually have no knowledge of the facts. Just my idle conjecture for which I may be castigated by William of Occam's Razor (the unkindest cut of all). Just remember, though, in choosing between conspiracy and stuff-up, always bet on the latter. Truth is the daughter of time ...

Offered in ignorance for your amusement ...
Old 2 weeks ago
  #30
Lives for gear
 

Thank you for that reference. It will take a while to absorb, but interesting how in broad brush terms it reflects my conjecture about six days ago. It may need more than the occasional glance while watching the telly ...

One - note the lodgement date (2016).

Two - while the patent describes a particular embodiment of that technique, I have encountered situations where the holder of the patent claims overarching control of that general configuration (sometimes incorrectly in my view). The patent system is not perfect and can sometimes be mis-used to inhibit innovation, not just reward the originator of the idea.

Three - after reading what processes they put that poor old audio waveform through, no audio purist would have a bar of it - they would be driven back to analogue forever. :o)))
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump