Become a Site Supporter and Never see Ads again!

Author Topic: Zoom F6 (32-bit float equipped)  (Read 32093 times)

0 Members and 1 Guest are viewing this topic.

Offline carpa

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 163
Re: Zoom F6 (32-bit float equipped)
« Reply #315 on: June 04, 2020, 03:32:13 AM »
@Jerryfreak, thanks for all the work you've done! From what's been answered by Zoom  I get that the digital trim also in 24 can only be adjusted after the analog input stage.
It seems, the,  that the padded line-in input may produce a bit more noise as a result;  I still wonder if in real world use the unpadded  +4 db limit may actually cause overload problems/distorted file even in 24 bit mode. Given the F6 design, maybe something more than +4 would have be a safer option?
If I'm right, recording two parallel tracks in 32 and 24 will result in having  a 24 bit files probably ok "out of the box"  if you've trimmed levels ok, and in a 32 bit files which free from setting inaccuracy; both files, though, are equally subjected to the same risk of input overload - if rare, I argue.
I just wonder if the "traditional" F4/F8 of recording a "more generous" file and a safety track 12 db lower could lead to a safer result and less fiddling at the end of the day.

Apart from measurements, which are interesting indeed, I'd like to figure out what performance F6 will give in an average scenario - i.e. recording a  classical or jazz concert or a choir or whatever - with a pair or average sensitivity mics. How does it compare in sound quality to an F4/8 or Mixpre or even a cheaper recorder like Zoom H6?  Reading here and there it seems that, using condenser mics, the limit in noise floor is most likely stated by the mic self noise more than preamps.


Offline jerryfreak

  • Trade Count: (30)
  • Needs to get out more...
  • *****
  • Posts: 4171
  • high desert for life
Re: Zoom F6 (32-bit float equipped)
« Reply #316 on: June 04, 2020, 03:39:21 AM »
@Jerryfreak, thanks for all the work you've done! From what's been answered by Zoom  I get that the digital trim also in 24 can only be adjusted after the analog input stage.
It seems, the,  that the padded line-in input may produce a bit more noise as a result;  I still wonder if in real world use the unpadded  +4 db limit may actually cause overload problems/distorted file even in 24 bit mode.


it would be identical. if you overload a 24 bit file by going over +4/+24, it would be the same in 32 bit

i can ask them if the limiter helps a hot signal but im thinking not based on the block diagram it lives in the digital portion. however, limiter is disabled in float vs integer so maybe theres an advantage

Apart from measurements, which are interesting indeed, I'd like to figure out what performance F6 will give in an average scenario - i.e. recording a  classical or jazz concert or a choir or whatever - with a pair or average sensitivity mics. How does it compare in sound quality to an F4/8 or Mixpre or even a cheaper recorder like Zoom H6?  Reading here and there it seems that, using condenser mics, the limit in noise floor is most likely stated by the mic self noise more than preamps.

i think its great. when used correctly it should be basically indistinguishable from F4/F8/mixpre. I cant personally speak to H6, it is known as a 'lower grade' of gear relative to the others, but that probably has to do with how much gain youre using. almost any modern recorder does a good job when fed a decent line signal and not leaning on the gain stage
« Last Edit: June 04, 2020, 03:41:47 AM by jerryfreak »
my facemask is a windscreen

People whose posts I wont see:
capnhook, daspyknows, Melanie, morst, Rob D., Scooter123, Sloan Simpson, Walstib62

Online Gutbucket

  • record > listen > revise technique
  • Trade Count: (15)
  • Needs to get out more...
  • *****
  • Posts: 13782
  • Gender: Male
  • Gunther Theile nailed it!
Re: Zoom F6 (32-bit float equipped)
« Reply #317 on: June 04, 2020, 10:49:27 AM »
4) Right channel signal is hotter (equating to lower noise after equal level adjustment of signal) but is less clean. I prefer the Left channel signal.

yes they were different mics (one dpa and one schoeps). the fact that you can hear differences in noise character between them indicates the mic self noise is at least as apparent as the input noise floor.

To clarify, I didn't notice a difference in noise character between the two channels in each sample, but rather between different samples in the same channel. In listening for difference in noise character I was trying to determine if I was hearing the same source noise in each sample, or the noise of different things. Any difference in character I heard was relatively subtle and I could be imagining it, influenced by noise level differences and the slightly different profile at the start of each file as it settles.

The character and level of the noise floor between left and right channels appeared about the same as I recall, but with a slightly hotter signal in the right channel, which I presume is due to differences in mic sensitivity.  The character of the signal in each channel was different (expected from the different mics), with the left channel signal sounding cleaner to my ear, be that from a subtle difference in frequency response, resonances, or some other form of subtle distortion. 

I'll try and give another listen this weekend to actual downloaded files on my higher quality Senn phones.
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline jerryfreak

  • Trade Count: (30)
  • Needs to get out more...
  • *****
  • Posts: 4171
  • high desert for life
Re: Zoom F6 (32-bit float equipped)
« Reply #318 on: June 04, 2020, 10:58:13 AM »
by the nature of the setup they werent on same vertical axis so there may be some difference in sound ,mic'ing a 2-way speaker from slightly different spots at a reasonably close distance

more from zoom

Quote from: me
does the limiter help with input signals over +4/+24dB? or is it too far into the digital chain to effectively prevent overload?

Quote from: zoom
In 32-bit float the limiter will have no beneficial effect. In 24-bit it will reduce analog gain and add it back after conversion to buy some extra headroom. There is also a look-ahead mode which works as effectively as a analog limiter.

Quote from: me
i saw the F4 and f8 were spec'd at +14dBU input with limiter on, is this in the same range?

Quote from: zoom
Yes. Limiter is the same as the F8n in 24-bit mode.


seems worthy of some experimentation.
if all of what he is saying is correct, namely:
- noise floor is fixed and identical in both float and integer modes
- + 4dBU will clip input on 32bit and limiter is not available
- +14dBU is available with limiter in 24-bit mode

this could create a scenario where 24-bit would actually have more usable input headroom than 32-bit float (albeit compressed on the top end, better than clipping)
« Last Edit: June 04, 2020, 11:05:54 AM by jerryfreak »
my facemask is a windscreen

People whose posts I wont see:
capnhook, daspyknows, Melanie, morst, Rob D., Scooter123, Sloan Simpson, Walstib62

Online Gutbucket

  • record > listen > revise technique
  • Trade Count: (15)
  • Needs to get out more...
  • *****
  • Posts: 13782
  • Gender: Male
  • Gunther Theile nailed it!
Re: Zoom F6 (32-bit float equipped)
« Reply #319 on: June 04, 2020, 11:25:31 AM »
by the nature of the setup they werent on same vertical axis so there may be some difference in sound ,mic'ing a 2-way speaker from slightly different spots at a reasonably close distance

Noted, thanks.  That could easily account for much of the difference I was hearing.
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline voltronic

  • Trade Count: (33)
  • Needs to get out more...
  • *****
  • Posts: 3096
Re: Zoom F6 (32-bit float equipped)
« Reply #320 on: June 04, 2020, 11:43:37 AM »
Apart from measurements, which are interesting indeed, I'd like to figure out what performance F6 will give in an average scenario - i.e. recording a  classical or jazz concert or a choir or whatever - with a pair or average sensitivity mics. How does it compare in sound quality to an F4/8 or Mixpre or even a cheaper recorder like Zoom H6?  Reading here and there it seems that, using condenser mics, the limit in noise floor is most likely stated by the mic self noise more than preamps.

I don't have an F4 or F6 to compare with, but here some samples of concerts I recorded with DPA4061 > F6.
https://taperssection.com/index.php?topic=190175.msg2324851#msg2324851

Keep in mind that I have to use iZtotope Spectral Denoise on almost every recording due to noisy HVAC in the places I am recording, especially with the 4061s which pick up everything in the room.  I believe that a couple recent chamber choir concerts I recorded with my CM3s did not need any denoising because the HVAC was off (huzzah)!  I'll post those when I find them.
DPA 4061 | Line Audio CM3 | Naiant X-Q | AT 853
Naiant PFAs | Shure FP24
Zoom F6 | Sony PCM-M10

Team Line Audio
Quote
I am hitting my head against the walls, but the walls are giving way.    ///    If a composer could say what he had to say in words he would not bother trying to say it in music.
- Gustav Mahler

Online Gutbucket

  • record > listen > revise technique
  • Trade Count: (15)
  • Needs to get out more...
  • *****
  • Posts: 13782
  • Gender: Male
  • Gunther Theile nailed it!
Re: Zoom F6 (32-bit float equipped)
« Reply #321 on: June 04, 2020, 11:45:09 AM »
some questions zoom has answered for me in last week or so

Quote from: zoom
The dual ADC is used when recording in 32 bit float and when recording in 24 bit. The Trim allows you to set the level that will be recorded to the 24 bit file. This makes it so that if the signal clips in 24 bit mode, it is due to the file clipping from the limitations of the 24 bit integer format, not from clipping at the converter. Since linear formats cannot accommodate the wide dynamic range of the dual ADC on the F6, a software Trim is necessary to set the level of the recorded file.

^Good. Establishes that in 24bit mode, the ADC no longer represents a bottleneck.


Quote from: jerryfreak
i've done some tests recording relatively quiet source material , both line in, and mic in, and i really can't hear a difference in the integer vs float files when normalized. it seems like the EIN is the limiting factor.

Quote from: zoom
As you point out here, the theoretical implications of 32-bit float are limited by hardware. With a quiet signal, I wouldn't expect you to hear a difference in the integer vs float files.

^This! [bolding is my emphasis]


Quote from: jerryfreak
considering that in either integer or float the max input level is +4dbu, and the EIN is fixed, i don't quite understand how the float files increase dynamic range between these two endpoints. i understand how the two ADCs can have a wider range of 200 db or more, but i still can't wrap my head around how that increases the 120ish db dynamic range of the input

Quote from: zoom
The preamp and ADC are going to act the same whether in float or integer. Floating point allows the final recorded file to have an extended dynamic range. So if recording in integer, the bit depth itself will limit the dynamic range meaning that somewhere in the full dynamic range of the preamps and ADC's it will have to cut it, meaning there is a ceiling and a floor which will either clip or not record. When recording float files, you can capture the entire dynamic range possible.

^ To my way of thinking, this is the critical question. 24 bit integer format can fully accommodate a 120ish dB dynamic range limit imposed by the hardware (the range between EIN and +4dBu) .


Quote from: jerryfreak
one more question: since the trim is acting as an attenuator, when recording in 24-bit mode would the best performance be achieved with trim set to 0dB?

Quote from: zoom
Not necessarily. Because both AD converters are working all the time, where the trim is set doesn't have as much effect on the performance as what the AD converters decide to do 

I read this as confirming that we can set trim to minimum (thus minimizing the possibility of overloading the input) and not incur an increased noise-floor from the recorder. So far it looks like there is no disadvantage to recording in 24 bit mode with minimum trim, assuming one is fine with adjusting levels as necessary in post, which one would need to do if recording in 32FP mode anyway.



more from zoom

Quote from: jerryfreak
does the limiter help with input signals over +4/+24dB? or is it too far into the digital chain to effectively prevent overload?

Quote from: zoom
In 32-bit float the limiter will have no beneficial effect. In 24-bit it will reduce analog gain and add it back after conversion to buy some extra headroom. There is also a look-ahead mode which works as effectively as a analog limiter.

Increases headroom for hot inputs but potentially reduces overall dynamic range by attenuating the input signal.  Does input attenuation raise EIN < it should not. But because input signal level is reduced by the pad while EIN remains the same, the result is an increase in apparent noise floor of the recorder in relation to signal. [several edits made here as I've re-thought this through, with distractions]

^ [edit: I have to think this part through more, and maybe listen to the low level samples again]


Quote from: jerryfreak
i saw the F4 and f8 were spec'd at +14dBU input with limiter on, is this in the same range?

Quote from: zoom
Yes. Limiter is the same as the F8n in 24-bit mode.

Good to get this cleared up.



Quote from: jerryfreak
seems worthy of some experimentation.
if all of what he is saying is correct, namely:
- noise floor is fixed and identical in both float and integer modes
- + 4dBU will clip input on 32bit and limiter is not available
- +14dBU is available with limiter in 24-bit mode

this could create a scenario where 24-bit would actually have more usable input headroom than 32-bit float (albeit compressed on the top end, better than clipping)

Yes. Would be ironic, no? Except dynamic range may be reduced by the same amount.

[note: initial editing to streamline the questions/answer quoting above, no content changed]
[multiple re-edits to my comments only]
« Last Edit: June 04, 2020, 05:24:37 PM by Gutbucket »
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline carpa

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 163
Re: Zoom F6 (32-bit float equipped)
« Reply #322 on: June 04, 2020, 01:38:28 PM »
Apart from measurements, which are interesting indeed, I'd like to figure out what performance F6 will give in an average scenario - i.e. recording a  classical or jazz concert or a choir or whatever - with a pair or average sensitivity mics. How does it compare in sound quality to an F4/8 or Mixpre or even a cheaper recorder like Zoom H6?  Reading here and there it seems that, using condenser mics, the limit in noise floor is most likely stated by the mic self noise more than preamps.

I don't have an F4 or F6 to compare with, but here some samples of concerts I recorded with DPA4061 > F6.
https://taperssection.com/index.php?topic=190175.msg2324851#msg2324851

Keep in mind that I have to use iZtotope Spectral Denoise on almost every recording due to noisy HVAC in the places I am recording, especially with the 4061s which pick up everything in the room.  I believe that a couple recent chamber choir concerts I recorded with my CM3s did not need any denoising because the HVAC was off (huzzah)!  I'll post those when I find them.

Very nice recordings! Thank you for the link. Actually I ended up buying an H6 which is on it's way now (which isn't the same league as F6, of course) 'cause I had the chance of a good discount on one of the last "old" grey model vs the "new", which is only just painted black with no other differences than coming without the m-s capsule ( sound ironical, but true) and bag. So at less price I still have the two capsules. I don't expect much quality over my old R09HR , but I have XLR and I'll take it as a chance to make some practice with more than 2 tracks and can still make some experiments/comparisons with my littlebox as an external preamp. I think I'll consider to be covered with mic options on the cheap side and save cash for a "serious" pair of condensers, which in any case I think will make a difference.   In the worst scenario it will serve as interface for Skype calls. 
« Last Edit: June 04, 2020, 01:40:13 PM by carpa »

Offline voltronic

  • Trade Count: (33)
  • Needs to get out more...
  • *****
  • Posts: 3096
Re: Zoom F6 (32-bit float equipped)
« Reply #323 on: June 04, 2020, 03:38:36 PM »
The H6 is a decent recorder and will serve you well.  I don't own one, but have used one several times.  The preamps are not the quietest, if that matters to you.  My main gripe is that the level knobs are very fiddly.  It is otherwise very simple to use.
DPA 4061 | Line Audio CM3 | Naiant X-Q | AT 853
Naiant PFAs | Shure FP24
Zoom F6 | Sony PCM-M10

Team Line Audio
Quote
I am hitting my head against the walls, but the walls are giving way.    ///    If a composer could say what he had to say in words he would not bother trying to say it in music.
- Gustav Mahler

Offline jerryfreak

  • Trade Count: (30)
  • Needs to get out more...
  • *****
  • Posts: 4171
  • high desert for life
Re: Zoom F6 (32-bit float equipped)
« Reply #324 on: June 04, 2020, 10:42:10 PM »
Does input attenuation raise EIN < it should not. But because input signal level is reduced by the pad while EIN remains the same, the result is an increase in apparent noise floor of the recorder in relation to signal.

thats my take.
my facemask is a windscreen

People whose posts I wont see:
capnhook, daspyknows, Melanie, morst, Rob D., Scooter123, Sloan Simpson, Walstib62

Offline voltronic

  • Trade Count: (33)
  • Needs to get out more...
  • *****
  • Posts: 3096
Re: Zoom F6 (32-bit float equipped)
« Reply #325 on: June 09, 2020, 10:28:11 AM »
I'm not sure if this has been discussed before, but seeing as I always record in 32-bit float mode I never discovered this odd behavior.  It seems that in 16 or 24-bit fixed mode, you cannot use the knobs to change your ISO track level; just the stereo downmix.  In order to change your ISO levels in that mode, you have to go through the menus to each track's PFL menu.

Here's a video explaining the issue.  I agree with him that a firmware update is called for, and it would be great if there was a quick way to toggle between the knobs controlling ISO levels and the downmix as you can on the F4 and F8.
https://www.youtube.com/watch?v=ro5Op9d-LKI
DPA 4061 | Line Audio CM3 | Naiant X-Q | AT 853
Naiant PFAs | Shure FP24
Zoom F6 | Sony PCM-M10

Team Line Audio
Quote
I am hitting my head against the walls, but the walls are giving way.    ///    If a composer could say what he had to say in words he would not bother trying to say it in music.
- Gustav Mahler

Offline jerryfreak

  • Trade Count: (30)
  • Needs to get out more...
  • *****
  • Posts: 4171
  • high desert for life
Re: Zoom F6 (32-bit float equipped)
« Reply #326 on: June 09, 2020, 10:35:10 AM »
correct, see test 1 and test 5 in my last series. faders at opposite extreme, isos came out the same

agreed the firmware should allow control of trim

i just requested. Response was

Quote from: zoom
Currently not possible, but it is a comment we have alerted our engineers to.
« Last Edit: June 09, 2020, 02:15:31 PM by jerryfreak »
my facemask is a windscreen

People whose posts I wont see:
capnhook, daspyknows, Melanie, morst, Rob D., Scooter123, Sloan Simpson, Walstib62

 

RSS | Mobile
Page created in 0.053 seconds with 53 queries.
© 2002-2020 Taperssection.com
Powered by SMF