Zoom F6 Portable Field Recorder Review

[milosz]

For perspective, compare this to the famous Nagra E's specs from => https://www.nagraaudio.com/wp-content/uploads/2019/03/Nagra_E_FR1.pdf

Nagra THD at 0 dB 400 Hz...... -41 dB
Nagra S/N ..................................... 62 dB "A" weighted, likely about 56 dB unweighted

 

[AnalogSteph]

Now you had me going down the rabbit hole... According to measurements even DIN reference level @ 250 nWb/m seems to be manageable with better than -45 dB (or around -50 dB) by a lot of ferric tape, so the THD spec is likely to be tape dependent. Apparently good chrome tape in compact cassette can hit similar SNR, which is remarkable because the Nagra operates at a fixed 19 cm/s (7-1/2 ips), 4 times as fast.

 

[amirm]

Could a ground loop involving the USB cable possibly impact performance?

A real ground loop would show mains frequencies. So it is not that. It could be noise from USB bus though that is not filtered.

 

[DLS79]

Yes, something is wrong there as this device won't even support sampling rates above 48 kHz. I am testing something else but will go back to this a bit later. Thanks for catching it.

Make sure you aren't recording to the SD car as the same time. The f6 can be used as audio interface up to 96 kHz, but you are limited to 48kHz if you record to an sd card at the same time.


As an aside, i think several people that have responded to this thread don't fully understand what field recorders like this are actually used for. Yes they can be used as an audio interface, but that's more of fluff a feature to appease those who don't really know what they want or need.

Videography and the associated disciplines are just a hobby to me, but I am pretty well versed in them. Imo, if you are looking at a field recorder from the Zoom F line or Sound Devices MixPre line then the following things are far more important to you than how it functions as a ADC.

• Does it have timecode support, what type of timecode features does it have, and how accurate is the internal clock.
• How many inputs does it have and how many of them support phantom power. 48v only or 24v and 48v support.
• What are its power options. If it's an internal only devise, what type of battery or batteries does it use. Does it support external power, if so what type of connector does it use. If it supports external power, does it fall back to internal power cleanly so you can hot swap external supplies.
• What bit depths and frequencies does it support.
• what does it use for storage, and what storage related features doe it have. For example recording redundancy is a big deal.
• what type of meta data support does it have, and what other special features like pre-role etc does it have.

The hype from Zoom, SD, and other people in the industry about 32 bit, is about not needing to set and constantly monitor gain. Thats a huge load off the sound guys shoulder. A lot of small productions would happily trade audio quality for that feature. Also keep in mind The F and MixPre lines, are mid to high end consumer, lowend pro grade gear. Mid grade pro gear is going to be something like an SD 833/888, or Zaxcom Nomad. High end gear is probably going to involve a cart.


Edit:
Here is a review by a popular industry reviewer. He reviews a lot of prosumer grade gear.

 

[PeteL]

For perspective, compare this to the famous Nagra E's specs from => https://www.nagraaudio.com/wp-content/uploads/2019/03/Nagra_E_FR1.pdf

Nagra THD at 0 dB 400 Hz...... -41 dB
Nagra S/N ..................................... 62 dB "A" weighted, likely about 56 dB unweighted

Don't get me wrong, we live in an awesome time for audio reproduction, but those direct comparison are not 100% representative. The way of listening to music have changed quite a bit. In those day, a high fidelity system was simpler to summarize. A preamp, a power amp, speakers, ambient noise, noise floor, threhold of loudness, usable dynamic range. Nowadays, we have extremely sensitive IEMs, often many dBs of digital attenuation, gain structures are all over the place with so many types of transducer on the market. In real life situations, for many use case we are closer to the noise floor than we may think, yes it is much better, but it DOES make sense to use all the range available. I know I've heard some background hiss, with actually good electronics, but less than optimal in term attenuation and gain to get the right listening volume. Bottom line, we never know on what someone will listen to your recording.

 

[milosz]

Now you had me going down the rabbit hole... According to measurements even DIN reference level @ 250 nWb/m seems to be manageable with better than -45 dB (or around -50 dB) by a lot of ferric tape, so the THD spec is likely to be tape dependent. Apparently good chrome tape in compact cassette can hit similar SNR, which is remarkable because the Nagra operates at a fixed 19 cm/s (7-1/2 ips), 4 times as fast.

Likely the Nagra 0 dB VU is higher than DIN reference level for cassette tape.

 

[Blumlein 88]

Make sure you aren't recording to the SD car as the same time. The f6 can be used as audio interface up to 96 kHz, but you are limited to 48kHz if you record to an sd card at the same time.


As an aside, i think several people that have responded to this thread don't fully understand what field recorders like this are actually used for. Yes they can be used as an audio interface, but that's more of fluff a feature to appease those who don't really know what they want or need.

Videography and the associated disciplines are just a hobby to me, but I am pretty well versed in them. Imo, if you are looking at a field recorder from the Zoom F line or Sound Devices MixPre line then the following things are far more important to you than how it functions as a ADC.

• Does it have timecode support, what type of timecode features does it have, and how accurate is the internal clock.
• How many inputs does it have and how many of them support phantom power. 48v only or 24v and 48v support.
• What are its power options. If it's an internal only devise, what type of battery or batteries does it use. Does it support external power, if so what type of connector does it use. If it supports external power, does it fall back to internal power cleanly so you can hot swap external supplies.
• What bit depths and frequencies does it support.
• what does it use for storage, and what storage related features doe it have. For example recording redundancy is a big deal.
• what type of meta data support does it have, and what other special features like pre-role etc does it have.

The hype from Zoom, SD, and other people in the industry about 32 bit, is about not needing to set and constantly monitor gain. Thats a huge load off the sound guys shoulder. A lot of small productions would happily trade audio quality for that feature. Also keep in mind The F and MixPre lines, are mid to high end consumer, lowend pro grade gear. Mid grade pro gear is going to be something like an SD 833/888, or Zaxcom Nomad. High end gear is probably going to involve a cart.


Edit:
Here is a review by a popular industry reviewer. He reviews a lot of prosumer grade gear.

I understood all that you say. But the performance is simply so low there is no excuse for it. It shouldn't cost pennies to make it much better.

@amirm
I am bummed out Amir didn't take the time to record and investigate the 32 bit recording to an SD card. That is the big feature which would cause you to choose this device, and it wasn't even addressed. It would have taken some fiddling about, and lots of time. And maybe a few tries to determine which tests told you anything. But this is an unusual feature, and potentially a very, very nice one to have.

 

[DLS79]

I understood all that you say. But the performance is simply so low there is no excuse for it. It shouldn't cost pennies to make it much better.

Those pennies probably went into some of the more useless features imo. For exampe being able to power is via usb.

It also seems obvious they have used cost-cutting measures to reach a certain price point. For example it only supports One SD card.

I would agree that 32-bit feature is the most significant reason to buy the F6. I have an older F4, and in many ways I think it's superior to the F6 when it comes to features related to its actual use.

 

[Tks]

Sensitivity in the mic won't help. Eventually all mics have a self noise defined by the individual air molecules hitting the diaphragm. Larger diaphragms have a lower self noise floor as they average out the impacts more. Similarly the analog stages are limited by the various thermal noise sources they contain. Lots of care can be taken, but there are fundamental limits that can't be got past no matter what.
If you want to combine tracks to reduce ADC errors, just add the two tracks in the DAW. It really is as simple as that. So long as both were recorded with the same source, you will get an extra bit of resolution for every doubling of tracks (in an ideal world, and subject to a few contraints on ADC implementation). But you will simply be able to resolve the thermal noise in the system with perfect precision. There is signal, and useful signal at that, below the noise, but you can't get rid of noise that is welded in to the signal before it gets to the ADC electronics.

Sensitivity on the mic doesnt help? But wouldnt that mean I able to use lower gain and this lower noise?

 

[jerryfreak]

Sensitivity on the mic doesnt help? But wouldnt that mean I able to use lower gain and this lower noise?

in a given class of microphone, more sensitive mics have lower max spl. the best common condenser microphones have a range in mid 120s for dynamic range, but as stated, its difficult to find program material (or playback systems) that can make use of that

 

[Blumlein 88]

Sensitivity on the mic doesnt help? But wouldnt that mean I able to use lower gain and this lower noise?

Let me see if I can go thru an example.

A Shure KSM44a is one of your quieter mics. Also fairly sensitive. Without a pad it is rated at 1% THD at 134 db SPL. That would be about 9 dbV output at that sound level. I happen to have a preamp that has a max input of 9 db V. Self noise is -121 dbV. EIN theoretical max for this microphone which has 50 ohm output impedance is about -138 dbV. The preamp probably has an EIN of -130 dbV at lower gain settings. If you add all that up you'd get -118 dbV. In fact it would be a bit less as noise won't add up being uncorrelated.

Now for a small group miked not too far away or an acoustic instrument and singer I'd use 30 to maybe 40 db. That eats into the noise floor so we'd be looking at a dynamic range of 88 db to78 db above the noise. Ignoring we can hear into the noise. But I didn't mention room noise. Room noise in a good studio or other location would be maybe 30 db SPL A-wtd. That would be about -95 dbV in our signal. And 30 to 40 db gain means we'll only have 55 to 65 db dynamic range left in the end.

If I had a louder source or recorded closer I could lower gain, and get back some of that range to work with. All of which is a long way around to say the range you'll have to work with is mainly going to be limited on the lower end by the ambient noise, and on the upper end by how loud your source is (assuming you aren't over-driving the microphone). As the specs show a 15 dB pad raises max SPL by 15 db it isn't the mechanics of the microphone that are the limit, but the electronics. In this case anyway.

Will the microphone preamp and AD have that much range? Quite a few have 95 to 110 db so they aren't the limiting factor in most cases. The gain is sliding up or down a window of 95-110 db dynamic range. You won't get that at the lowest gains settings, and some preamps lose a bit at the very top if gains are more than 60 db. So in the end this is enough for most of your normal purposes in recording music.

If I had a source that reached peaks of 134 db SPL and a noise floor of 30 db SPL then you have a 104 db window. And I could get by without any gain. The 30 db of room noise is going to swamp self noise of the microphone, and the added preamp noise. When I have to use gain for quieter sources the amplified room noise again swamps noise from other sources. If you had microphones with self noise of 25 db and low sensitivity combined with a preamp of poor EIN you might start to hear that noise with or over the room noise.

 

[Francis Vaughan]

Sensitivity on the mic doesnt help? But wouldnt that mean I able to use lower gain and this lower noise?

Depends where the noise is coming from. In a good system, self noise in the mic is where the limit is. Not the mic preamp. So a sensitive mic just picks up the noise better.
And as above, environmental noise swamps everything in real life.

 

[bennetng]

I understood all that you say. But the performance is simply so low there is no excuse for it. It shouldn't cost pennies to make it much better.

Reminded me of this thread:
https://www.audiosciencereview.com/forum/index.php?threads/interface-mystery.13115/
Sometimes there are unreasonable results like these, while it can be related to test methodology/procedure, it can also be a fixable bug. Hopefully someone else (if not Amir) will figure it out later.

 

[Tks]

Let me see if I can go thru an example.

A Shure KSM44a is one of your quieter mics. Also fairly sensitive. Without a pad it is rated at 1% THD at 134 db SPL. That would be about 9 dbV output at that sound level. I happen to have a preamp that has a max input of 9 db V. Self noise is -121 dbV. EIN theoretical max for this microphone which has 50 ohm output impedance is about -138 dbV. The preamp probably has an EIN of -130 dbV at lower gain settings. If you add all that up you'd get -118 dbV. In fact it would be a bit less as noise won't add up being uncorrelated.

Now for a small group miked not too far away or an acoustic instrument and singer I'd use 30 to maybe 40 db. That eats into the noise floor so we'd be looking at a dynamic range of 88 db to78 db above the noise. Ignoring we can hear into the noise. But I didn't mention room noise. Room noise in a good studio or other location would be maybe 30 db SPL A-wtd. That would be about -95 dbV in our signal. And 30 to 40 db gain means we'll only have 55 to 65 db dynamic range left in the end.

If I had a louder source or recorded closer I could lower gain, and get back some of that range to work with. All of which is a long way around to say the range you'll have to work with is mainly going to be limited on the lower end by the ambient noise, and on the upper end by how loud your source is (assuming you aren't over-driving the microphone). As the specs show a 15 dB pad raises max SPL by 15 db it isn't the mechanics of the microphone that are the limit, but the electronics. In this case anyway.

Will the microphone preamp and AD have that much range? Quite a few have 95 to 110 db so they aren't the limiting factor in most cases. The gain is sliding up or down a window of 95-110 db dynamic range. You won't get that at the lowest gains settings, and some preamps lose a bit at the very top if gains are more than 60 db. So in the end this is enough for most of your normal purposes in recording music.

If I had a source that reached peaks of 134 db SPL and a noise floor of 30 db SPL then you have a 104 db window. And I could get by without any gain. The 30 db of room noise is going to swamp self noise of the microphone, and the added preamp noise. When I have to use gain for quieter sources the amplified room noise again swamps noise from other sources. If you had microphones with self noise of 25 db and low sensitivity combined with a preamp of poor EIN you might start to hear that noise with or over the room noise.

Got a much better grasp now. Well explained Blum!

@Francis Vaughan

You mentioned something about combining audio in post to get a cleaner result. How does that work btw?

 

[Francis Vaughan]

You mentioned something about combining audio in post to get a cleaner result. How does that work btw?

It is really only a theoretical point. There was mention made of ADC implementations that internally parallel ADCs. My point was simply that if you really wanted to get the same effect, so long as the ADCs were correctly implemented, you could perform the same trick yourself. Split the feed to multiple channels, each with its own ADC, and sum the result in the DAW. Correct implementation pertains to corrent dithering on the input, and doing so in a manner which is not correlated between the ADCs. For every doubling of ADCs ganged up you get an extra bit of depth in the AD process. It doesn't help with noise in the source.
I'm not suggesting that anyone should do this. It is more a matter of theoretical amusement.

 

[PeteL]

It is really only a theoretical point. There was mention made of ADC implementations that internally parallel ADCs. My point was simply that if you really wanted to get the same effect, so long as the ADCs were correctly implemented, you could perform the same trick yourself. Split the feed to multiple channels, each with its own ADC, and sum the result in the DAW. Correct implementation pertains to corrent dithering on the input, and doing so in a manner which is not correlated between the ADCs. For every doubling of ADCs ganged up you get an extra bit of depth in the AD process. It doesn't help with noise in the source.
I'm not suggesting that anyone should do this. It is more a matter of theoretical amusement.

Could it be the same idea where many stereo or multichanels DAC chips, when use in a dual mono-2 chips implementation would gain a few dBs of dynamic range with the same specified word length, or is it a completely different concept?

 

[jerryfreak]

For every doubling of ADCs ganged up you get an extra bit of depth in the AD process..

that seems erroneous

from ADC specs ive seen, they gain 1-2 dB when combining channels (and then again from 4-8)

for accurate mic manufacturers, ive heard +3dB in S/N when doubling up

i havenot seen any that give a full 'bit of depth' (6 dB) on combining

just reporting what ive seen, cant say i truly understand the physics behind it

 

[bennetng]

that seems erroneous

from ADC specs ive seen, they gain 1-2 dB when combining channels (and then again from 4-8)

for accurate mic manufacturers, ive heard +3dB in S/N when doubling up

i havenot seen any that give a full 'bit of depth' (6 dB) on combining

just reporting what ive seen, cant say i truly understand the physics behind it

Seems to agree with RME's measurements.

BTW, I assume you use the mono trick at the input to get 3 dB more SNR? Are you aware you can do the same at the output? You can use a simple split cable, two XLR female to one XLR male. Let the generator send out the same signal on both channels. Internal resistors in the ADI cause an unchanged output level and THD, but 2.5 dB less noise. I use that quite often.

 

[Francis Vaughan]

The above is really just doing noise averaging, which is more brutal in its cost benefit. I was referring to a theoretical way of using the dithered LSB. But that assumes it is the limiting noise. So indeed, you won't see 6db gains doing it yourself because things are already noise limited in other ways. Just ignore me.

 

[Tks]

It is really only a theoretical point. There was mention made of ADC implementations that internally parallel ADCs. My point was simply that if you really wanted to get the same effect, so long as the ADCs were correctly implemented, you could perform the same trick yourself. Split the feed to multiple channels, each with its own ADC, and sum the result in the DAW. Correct implementation pertains to corrent dithering on the input, and doing so in a manner which is not correlated between the ADCs. For every doubling of ADCs ganged up you get an extra bit of depth in the AD process. It doesn't help with noise in the source.
I'm not suggesting that anyone should do this. It is more a matter of theoretical amusement.

Maybe I'm just a brutal moron. But when you say split the feed. Like what does that mean, how would I do that? Like, do I need to have two channels recording the same thing with multiple mics, or one mic that feeds into multiple channels with a splitter? And then somehow take that recording and.. do what exactly in software? Like, I'm wondering what does it mean literally from a directions point of view (what's actually happening when you're doing this goes over my head, I'm wondering specifically what would I be doing with options in a DAW that would be doing this sort of "combining" process that yields better results).

Also, this adds bit-depth as you say, but not reduction in noise. What use is such if noise hits before your theoretical bit-depth? Or lets say your noise is super low (amir posted a video once of some company recording insects eating and a caterpillar walking), could you then do this technique you speak about to achieve something that company was able to with their mics?