Become a Site Supporter and Never see Ads again!

Author Topic: Zoom F6 (32-bit float equipped)  (Read 139933 times)

0 Members and 1 Guest are viewing this topic.

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: Zoom F6 (32-bit float equipped)
« Reply #210 on: January 06, 2020, 08:46:59 PM »
I did some "SD Performance tests" , of some cards i had on hand, with the handy menu feature. The "full test" test was thorough, took 1-2 hours per card

ill link the info on the cards i tested that weve aggregated data on in the mixpre media thread

lower "access rate" is better, 100=fail

128GB Integral UltimaProX (is now one of the few cards on the Sound Devices approved media list, and the only affordable one over 32GB)

https://taperssection.com/index.php?topic=191314.msg2306038#msg2306038



64GB Transcend UHSII (fast card, esp on UHS-II, but failed on mixpre after 16 hours of testing)
https://taperssection.com/index.php?topic=191314.msg2312164#msg2312164



I havent run the integral in any recording tests, but ive put about 7 hrs of 6ch 32/192, 5 hrs of 8ch 32/96  and 20 hrs of 2ch 24/96 on the transcend without a single issue

« Last Edit: January 06, 2020, 08:53:00 PM by jerryfreak »
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline justme

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 231
  • Gender: Male
Re: Zoom F6 (32-bit float equipped)
« Reply #211 on: January 08, 2020, 12:57:08 AM »
Thank you JF for testing.
Zoom really delivers.

Internal card tests is far from a new function but it would be a great feature for SD to implement as well.
They have some slack to pick up in some areas.

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: Zoom F6 (32-bit float equipped)
« Reply #212 on: January 08, 2020, 09:59:23 AM »
ok i got it to connect with iphone via usb

this didnt work
https://www.amazon.com/gp/product/B07WLMFMRH

but i had this OTG adapter from a while back which works with a standard usb-a to c cable
https://www.amazon.com/gp/product/B07TWTXQ51

i can get the iphone to see the zoom and it works in metarecorder but it makes mixlr freak out for some reason and send out weird chunks of intermittent sped up samples interspersed with silence. which is weird because mixlr works fine with both built in iphone mic as well as d:vice on lightning input

zoom>iphone>metarecorder works flawlessly to record a backup, but you cant use 32-bit float, and are limited to 44.1 or 48K
 
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline EmRR

  • Trade Count: (5)
  • Taperssection Member
  • ***
  • Posts: 779
    • ElectroMagnetic Radiation Recorders
Re: Zoom F6 (32-bit float equipped)
« Reply #213 on: January 08, 2020, 12:15:59 PM »
Sounds like the differential IC is an Integrated Circuit with a Class-B or similar on board.

Almost nothing runs pure class B anymore, it's usually associated with broadcast transmitters.   I think many IC's run class AB, and can be biased to run pure class A if there's adequate power supply to do so.   In AB they run A on low level signals and move into B as they grow larger.   

In layman's terms, AB and other higher classes use variable amounts of power depending on need, with class A you get a constant for both power draw and circuit power artifacts imprinted on audio. 
Mics: DPA 4060 w/MPS 6030 PSU/DAD6001/DAD4099, Neumann KM 131, Oktava MK 012, Sennheiser MKH 105, MKH 20, MKH 30, MKH 40, MKH 800 TWIN
Recorders: Zoom F8n, Sony MZ-R50

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: Zoom F6 (32-bit float equipped)
« Reply #214 on: January 13, 2020, 12:51:55 PM »
OK, I realized that I had a major DUH moment here regarding the Reference / REC level for the channel knobs in 32-bit float mode:

As the block diagrams in the manual clearly show, all level adjustments in 32-bit float mode are post-ADC.  So WiFiJeff was correct: Zoom is putting a lot of faith into 32-bit float mode, because you really can't adjust it where it's going to prevent damage to your recording.  All that you can adjust is post-ADC level that is being written to the 32-bit float container, and it has already been shown that you can amplify or attenuate that until the cows come home with nothing lost.

can it be argued that the zoom really doesnt have "preamps" in the traditional sense? that it relies on a relatively sensitive modern ADC where the clipping point is matched to a +4dBU max input (a "purer" signal path, as per the block diagram, "line" mode is just the same circuit with a 20 dB pad inserted)

the way i used to run my CMRs into the AD2K was similar to this.... no additional gain stage. premise being despite the fact that the signal only peaked at -30dbFS, i still had over 80 dB of dynamic range to work with which put mic noise floor right around converter noise floor (both well below crowd noise floor). the CMR does not put out a relatively hot signal, and the AD2K isnt looking for anything but hot balanced line level signals, but i would contend that if you can get a conventional AD converter to see a mic output without an additional preamp, and get your levels peaking *anywhere close to -20dB*, you would have results worthy of any of the 32-bit float recorders. 20+ dB of headroom at which point your mics would be distorting (or youd be clipping the input of your 32bit recorder), and enough down-bit resolution to cover the dynamic range of almost any live performance

something with a little higher output like a schoeps at 60V>baby nbox or DPA MMP-G directly into any modern 24-bit converter, levels be damned
« Last Edit: January 13, 2020, 01:23:37 PM by jerryfreak »
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline voltronic

  • Trade Count: (40)
  • Needs to get out more...
  • *****
  • Posts: 4104
Re: Zoom F6 (32-bit float equipped)
« Reply #215 on: January 13, 2020, 08:28:36 PM »

can it be argued that the zoom really doesnt have "preamps" in the traditional sense? that it relies on a relatively sensitive modern ADC where the clipping point is matched to a +4dBU max input (a "purer" signal path, as per the block diagram, "line" mode is just the same circuit with a 20 dB pad inserted)


Maybe when set to 32-bit float mode, but not when in 24-bit mode where it appears to have a somewhat traditional preamp stage.
I am hitting my head against the walls, but the walls are giving way.
- Gustav Mahler

Acoustic Recording Techniques
Team Classical
Team Line Audio
Team DPA

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: Zoom F6 (32-bit float equipped)
« Reply #216 on: January 14, 2020, 12:01:27 AM »
its not clear (to me) from the block diagram where the ADC stage is

also its technically possible that 'trim' is attenuation, not gain

Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: Zoom F6 (32-bit float equipped)
« Reply #217 on: January 14, 2020, 01:38:31 PM »
got a disappointing 2:15 of run time first try with 2200 mAH dracast (6ch phantom). 2:12 on second test

does not compute as i get over 3 hours from 2500 mAH NiMH at 5V (confirmed since with different batteries)

notice that Voltronic's 6600 mAh test was 5 times the capacity of the 2200 mAh test. Assuming the battery isnt defective it would have to be premature voltage cutoff by the zoom

apparently under high load the voltage sag on the 2-cell battery trips the relatively high voltage cutoff of the zoom (around 7.0V) much earlier than the 6-cell 6600 mAh (where we would expect something slightly over 3X capacity vs the 2200 mAh battery, perhaps 3.5X, but not 5X!  best guess on the NP-F discharge rate, based on a proper 3.0V cutoff, would be close to 0.3C. the 6 cell 6600 mAh would be well under 0.1C, probably closer to 0.05C with a proper 3.0V cutoff. look at the graph below how much juice is left in the battery below 3.5V/cell, close to 40%! Most other manufacturers spec 3.0V as cutoff in their devices, which is still high enough to not damage the battery. a deep discharge would be something like 2.7V which some devices do. remember that cutting off at 3.0V *under load* is still leaving a safe amount of juice in the cell as not to damage it.

submitted this to support today:

-----------------------------------------------------------------------------------------------------------
i did come across some bugs in 1.20 firmware, and some suggestions for future firmware

bug: sometimes tracks dont arm despite the knobs being clicked on... a reboot corrects it

bug:display power saving acts erratically sometimes. where normally it goes dark after 30 seconds, sometimes it will just stay on. ive also seen it randomly click on and off by itself while recording. last night when it was stuck on, it flashed off for a second then back on (and stayed on). it was right when displayed NPF battery voltage was flipping from 7.2 to 7.1V, not sure if that is related in any way

suggestion: make an option for screen timeout in menu. something like 30 sec/60 sec/5 min/never
suggestion: make order of power sources used in menu selectable if possible (even if its just selecting between NPF and AA))
suggestion: allow user adjustable cut off-voltage when using NPF battery, as of now it cuts off at about 7.0V (3.5V/cell), which leaves upwards of 40% of the battery unused with smaller batteries like 2200 mAh used in high-drain situations like 6-channel w/phantom power. (between 0.3 and 0.5C, indicated by blue oval on graph below).  Having this cutoff adjustable down to 6V (3V per cell) would offer extended runtime and is more on par with other devices. These batteries are cheap, lost takes are not!




-----------------------------------------------------

its difficult to estimate state of charge of a resting li-ion cell by simple voltage since the curve is so flat. after cutting off, resting voltage after 10 mins or so is 7.6V (3.8V per cell... which seems like theres a lot of juice left in there...)

in fact i can take the 'dead' battery (7.0V cutoff from 6-ch phantom), and boot up the zoom in lowest power case i can imagine (1 channel no phantom), it displays 7.4V under load and runs for 40 more minutes before cutoff (woulda been probably an 1.5hr+ to 3.0V/cell cutoff.. but anyway)

i would venture to guess i could do the same with "dead" AAs. drop down from 6 to 2 channels and eek out some more runtime... good in a pinch

depending on cutoff voltage of this device, it might be able to be used to squeak the rest of the power out of the NPF battery and into the USB-C port
https://www.amazon.com/Power-Junkie-NP-F-Blind-Spot/dp/B07NF14XW4

in any case, for now, absent a firmware update, smaller NPF cells like 2200 mAh are a poor match for high-drain recording such as multichannel w/phantom. 4400, 6600 or larger would get well over 2-3X the usable capacity

im in the process of doing a runtime test with just 2 ch phantom, which is a better fit for this battery

Time 0 = 7.9V
20 mins -> drop to 7.8V
1.5 hrs -> 7.7V
2.25 hrs -> 7.6V
2.9 hrs -> 7.5V
3.9 hrs -> 7.4V
4.1 hrs -> 7.3V
4.3 hrs -> 7.2V
4.5 hrs -> 7.1V
4.7ish hrs ->switched over to internal AAs

idle battery voltage after this test was right around the 7.0V cutoff, so there was much less voltage sag at this <0.2C test

___________________________________________
TLDR: i orderd a 4400 mAh battery. About the same price and should give about 3X the runtime of the 2200 mAh due to high cutoff voltage
« Last Edit: January 16, 2020, 07:01:40 PM by jerryfreak »
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: Zoom F6 (32-bit float equipped)
« Reply #218 on: January 15, 2020, 03:57:42 PM »
re:


can it be argued that the zoom really doesnt have "preamps" in the traditional sense? that it relies on a relatively sensitive modern ADC where the clipping point is matched to a +4dBU max input (a "purer" signal path, as per the block diagram, "line" mode is just the same circuit with a 20 dB pad inserted)


Maybe when set to 32-bit float mode, but not when in 24-bit mode where it appears to have a somewhat traditional preamp stage.

I asked support:

Quote from: jerryfreak
I'm trying to understand the gain topology of the F6

it seems trim is active in linear mode but not float.

if i understand correctly, in float mode, basically the device acts as 'wide open' and has the ADC stage matched to the maximum +4dBU input (which is really the same in mic-in or if the 20 dB line-in pad is on)

in linear mode, is the 'trim' early in the block diagram an active analog gain stage or just an attenuator? does it operate in the analog or digital range?

is there a 'unity gain' ideal input level for the device which is free of gain stages and/or offers the highest dynamic range?

response:

Quote from: zoom support
The Trim on the F6 acts as a digital stage. 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.

so it would appear there is no analog gain or attenuation whatsoever, and that 'unity gain' would be the level at which the zoom operates in 'reference mode' where trim is disabled (probably same as trim at zero in all other modes but not confirmed). I requested clarification

the claim that the dynamic range of the input exceeds the 144 dB range of a 24-bit word seems like a bit of a stretch though.
« Last Edit: January 15, 2020, 04:02:22 PM by jerryfreak »
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (16)
  • Needs to get out more...
  • *****
  • Posts: 15700
  • Gender: Male
  • "Better to love music than respect it" ~Stravinsky
Re: Zoom F6 (32-bit float equipped)
« Reply #219 on: January 15, 2020, 04:23:44 PM »
the claim that the dynamic range of the input exceeds the 144 dB range of a 24-bit word seems like a bit of a stretch though.

Yes, and if a somewhat larger "dynamic range container" is necessary, I've wondered why not simply go to 32 bit linear and not have to rework everything else around floating point along with people's thinking about how it all works and what is going on?  Would effectively solve the problem just as well in a way that is much easier to understand.
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<< (note: This is a 1st draft, now several years old and in need of revision!  Stay tuned)

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: Zoom F6 (32-bit float equipped)
« Reply #220 on: January 15, 2020, 04:32:34 PM »
exactly, see this post above

https://taperssection.com/index.php?topic=190161.msg2320079#msg2320079

people claim that 32bit float is easier and more precise to manipulate, but as far as volume adjustment which is the #1 feature we crave, couldnt you just losslessly 'shift bits' 6dB at a time with any linear format?
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: Zoom F6 (32-bit float equipped)
« Reply #221 on: January 16, 2020, 02:09:03 AM »
interesting. if your input is too hot, it flashes this message for 10 seconds to let you know



i guess thats in the manual but i missed it. happened when i accidetnally hot-unplugged a mic but can make it overload by screaming right into a 4011

NOTE
• When Float (32bit) is selected, if a signal is input that exceeds the maximum input level for the input source (+4 dBu when Mic or +24 dBu when
Line), an “Exceeding maximum input level” message will appear. If this
message appears, adjust the output levels of the devices connected to
the input jacks.




« Last Edit: January 16, 2020, 02:18:00 AM by jerryfreak »
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline voltronic

  • Trade Count: (40)
  • Needs to get out more...
  • *****
  • Posts: 4104
Re: Zoom F6 (32-bit float equipped)
« Reply #222 on: January 16, 2020, 05:53:42 PM »
I'm a bit disappointed to hear you were able to overload the input using a 4011.  That's not a highly sensitive mic.

Mind trying the same thing with the input set to Line+Phantom?  That should keep you out of the red.
I am hitting my head against the walls, but the walls are giving way.
- Gustav Mahler

Acoustic Recording Techniques
Team Classical
Team Line Audio
Team DPA

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: Zoom F6 (32-bit float equipped)
« Reply #223 on: January 16, 2020, 06:03:09 PM »
I'm a bit disappointed to hear you were able to overload the input using a 4011.

i literally had to SCREAM into it. multiple times because first wasnt loud enough. 4011C spec says it can top 4V (+14 dBU)

at the 10 mV/Pa that scream shoulda been in the upper 130s

doubt it would overload in normal use. maybe with a 40mV/Pa cap like a 4006 or 2006 on an MMP-A body without the pad
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline DSatz

  • Site Supporter
  • Trade Count: (35)
  • Needs to get out more...
  • *
  • Posts: 3347
  • Gender: Male
Re: Zoom F6 (32-bit float equipped)
« Reply #224 on: January 16, 2020, 07:47:41 PM »
When someone literally screams into a mike within fractions-of-an-inch distances, 130 dB SPL, 140 dB SPL and even higher can be generated for short periods of time. This is why some head-worn microphones, such as DPA's, are designed with such high maximum SPL limits. With even a foot of distance between the mouth and the mike, most people can't produce 130 dB SPL at the mike with the voice alone.

But there's an art to screaming, and I've recorded a couple of trained operatic sopranos whose limits I'm not sure that I know, but who have produced 130+ dB SPLs in a microphone two feet or so away. That got me to be more serious about testing preamps for their input overload limits.

I also found that some microphone manufacturers are more conservative than others in their published maximum SPL ratings, i.e. a microphone may be able to put out considerably more signal than you would project from the sensitivity and maximum SPL specifications alone.

--best regards
music > microphones > a recorder of some sort

 

RSS | Mobile
Page created in 0.094 seconds with 40 queries.
© 2002-2024 Taperssection.com
Powered by SMF