Topic: Zoom F6 (32-bit float equipped)

[aaronji]

the contention above concerns the testing methodology used which requires loop-back connection via computer USB interface function.  The suspicion is that the problems are with the USB audio interface rather than affecting the the primary function of recording analog inputs stored locally on the SDcard.    The call above is to re-test using a methodology that better reflects typical recording use rather than use as an interface.

In response to a number of requests, the guy at audiosciencereview performed a second test without USB (straight through the recorder in stand-alone mode to the line-out). He found similar rising low-frequency noise as in USB mode, but that test also had an extra DAC/line-out step. Maybe I am wrong, but the presence of the noise in both tests suggests to me that the noise is coming from the pre/ADC (since it exists without the DAC/output stage and it also exists without the USB). In any event, it peaked at something like -70 or -80 dB at 20 Hz, so probably not audible in many real-world uses...

[jerryfreak]

the contention above concerns the testing methodology used which requires loop-back connection via computer USB interface function.  The suspicion is that the problems are with the USB audio interface rather than affecting the the primary function of recording analog inputs stored locally on the SDcard.    The call above is to re-test using a methodology that better reflects typical recording use rather than use as an interface.

In response to a number of requests, the guy at audiosciencereview performed a second test without USB (straight through the recorder in stand-alone mode to the line-out). He found similar rising low-frequency noise as in USB mode, but that test also had an extra DAC/line-out step. Maybe I am wrong, but the presence of the noise in both tests suggests to me that the noise is coming from the pre/ADC (since it exists without the DAC/output stage and it also exists without the USB). In any event, it peaked at something like -70 or -80 dB at 20 Hz, so probably not audible in many real-world uses...

agreed on all counts

ill do my best to figure it out

and yes we are spoiled, even poorly performing (on paper) modern recorders are well above the gear we were using a decade ago for our purposes

[Ozpeter]

There's always two issues when considering noise - firstly, there's the "does it actually matter" factor and secondly, there's the "never mind whether it matters, am I using the best gear I can afford" factor.

With respect to the former, when it comes to music, especially classical music with its very wide dynamic range, the chances of anyone being troubled by the noise from any reasonable recording system when the replay is at "authentic levels" in a typical listening environment, is close to zero.  By "authentic levels" I mean a level such as the original live recording would reach at the listener's ear.  Of course you can turn up the level way beyond that, at which point you might hear some system noise, but you'd also start to hear ambient noise to a distracting degree - there's not many recording locations where live music is performed that have noise levels below system noise.  (And of course, for pipe organs, you start to hear blower and action sounds pretty soon).  And when it comes to listening environments (ignoring headphones for the moment) I suspect 90% also have background noise levels above system noise.

Now when it comes to nature recordings, if you want to clearly hear a sparrow's tweet 100 metres away, you may want to turn up the level way above what is natural, and system noise might then be more of a factor.

As for using the best gear you can afford - well, you just have to accept that in 6 months' time, you'll need the next new thing.

[rastasean]

Jerryfreak,

Thanks for sharing the link to audiosciencereview and sending Amir your recorder. It'll be interesting to see your results.

Just sonically, I know Curtis Judd and a nature field recordist, George Vlad, have said the F6 and mixpre II series are very close together with the mics they've used. Understandably, they haven't tested either recorder extensively as a USB device.

I understand and appreciate the fact the F6 has a USB interface but to completely dismiss the recorder on those findings, while comparing it to specific devices that are usb input devices, seems a little peculiar. This is the first I've heard of the site so maybe he only focuses on audio devices connected to computers.

In the few times you've talked with Zoom support, have you provided them links to TS or the audioscience page? They could be a little forthcoming with information, so we don't have to use so much conjecture and speculation based on all these tests.

[jerryfreak]

In the few times you've talked with Zoom support, have you provided them links to TS or the audioscience page? They could be a little forthcoming with information, so we don't have to use so much conjecture and speculation based on all these tests.

i sent my contact a link on 8/31. he said give him a few days he wanted to se eif he could get an engineer to weigh in

just bumped him and got an auto reply that hes out from 9/4-9/14. so a few more days probably

[voltronic]

Hello tapers, this is my first post. I did a lot of reading on your site in past years and decided to join the group after reading all the posts about the Zoom F6.
I play pipe organ as a hobby and make recordings of pipe organs using 2 DPA 4090 microphones and an OADE PMD661 super mod recorder or a Denon SD recorder F450R with Lake People C360 preamp.
To allow recording pipe organ and an extra solo instrument I was looking for a recorder with at least one extra channel to use a spot mic.
I tried the MixPre 3, Zoom F6 and Zoom F8N and finally selected the Zoom F6. Just selected by playing/recording (using composed stops on the organ with a lot of harmonics) and careful listening to the result (with AKG K701 head phones).
I did not notice any noise, except very faint from the microphones. And no audible distortion.

Overall I am very satisfied with the recorder, so I was surprised to see a link to a site providing measurements done by Amir with conclusion that he could not recommend this recorder.
Did I buy bad equipment after all? It seems that measurements do (to some extent) not reflect the final result what you can hear.
Maybe someone can give a bit clarification to me.

Welcome, Organfreak!  It's good to have another classical recordist around here.

I think I can put your mind at ease.  Here is a link to some samples of my recordings made with DPA 4061s (same capsules your 4090s) straight into my F6.  The only one in there that isn't made with the F6 is the Wind Quintet file, which used a Tascam DR-70D.

Keep in mind that almost all of the places I record have audible HVAC noise, so they take a trip through iZotope RX Spectral Denoise to remove the worst of it.  I do this at a minimal level, though.  The Festival Sanctus track may not have had RX - I seem to remember that church was very quiet.

[Organfreak]

Thanks all for the remarks and clarifications. There is an awful amount of knowledge and on hand experience available on this forum!
@Voltronic: I will download your samples and pay a listen.
@Gutbucket: It would be nice to discuss about organ recordings. I do it for years, starting with a Philips tape recorder (having this heavy device still at home). Do you also make organ recordings?
@Jerryfreak: Success with your tests. Do you have dedicated test equipment?

I found a test of the Zoom F6 on the internet, sorry that it is in German. For most Dutch people not a problem as the languages have a lot in common. Maybe some of you have German roots?
See graphs on last page. Total distortion is listed as 0,0083%. That seems not much. The test with conclusion is very positive.
I only included the last page of the test report because of size restrictions. If someone is interested I can send to complete test as separate pages.

The preamps in the Zoom F6 should not be the weakest link. I had some correspondence with Doug Oade (who upgraded my PMD661) before I bought the F6; the preamp chip used by Zoom in all its F series recorders is the THAT 1510.

[jerryfreak]

i translated that review the best i could, cant say i agree with the comments

 "The distortion values ​​are outstanding...The F6 takes measurements at full level. Total distortion of about 0.0083%"

"The distortion spectrum shows that despite highly developed 32-bit technology
there is a certain difference in quality compared to expensive studio converters.
However, the performance is very decent for a mobile device."

assuming that number is THD+N, thats not all that great, it is a SINAD of about -81dB, which is more like ultraportable/handheld territory and step below what a balanced input recorder should be

it would be cool if you can link the rest of the pdf so we can see their methodology

the graph in the linked PDF shows the same low-frequency distortion that amir measured

when i try to measure the F6 anywhere near its stated max line input level of +24dBU the distortion blows it out. with a 1kHz sine wave, i can get up close to the max input (well  to +22dBU which is the limit of my DACs output...i could throw a preamp in there to push it to +24 with equally disastrous results)

But cant get anywhere near that with  the low frequencies

heres what the RMAA test file looks like. the two highlighted portions of the waveform are the low frequency portions of the sweep and multitone portion
note that the undistorted level of the original file has all the frequencies at the same level in the sweep and multi-tone portions



here are recorded 32/96 wav files at different input levels. from top to bottom, i tested input levels from +16dBu to+22dBu in 1dB increments. i didnt bother dialing in the gain on the zoom as it really wouldnt matter, so these are slightly off in overall dB but look at those low frequency potions relative to the rest of the waveform.

+16 and +17 dB look good, both look like this



at +18dBu we are starting to see distortion of the low frequencies of the multitone



at +19dBu we are seeing these get worse. the zoom is flashing the 'exceeding input level' on the screen



zoomed in to distorted waveform



at +21dBu the low frequency portion of the sweep is starting to distort as well



this gets even worse at +22dBU



to eliminate any variables of playback system, here is same playback system fed to the AD2K (now 20-year old technology) at +22dBu input. look at the absence of visible low-frequency distortion and how much cleaner the waveform is overall

[jerryfreak]

RMAA tests of zoom, with AD2K plotted as a reference

two sets of measurements, one using benchmark DAC2 for source and the other using mytek brooklyn dac+, to similar results

yeah that low frequency distortion is unshakable at any input level

is -70 to -80 dB distortion below 50Hz a problem? not sure, its pretty low on the audibility curve of human hearing

knowing what i know now, will i be using the HPF? you bet

if you only need 4 channels and dont need phantom, will the tascam dr-2d rival or best it in sound quality? per my tests, seems like it.

at the end of the day it seems like the sound quality of a handheld (at best) in a prosumer body.

[voltronic]

I wonder if Zoom is playing fast and loose with the input level specs.  I see that the F8n is also rated for +24 dB max line level, but only with limiters engaged.

It's doubtful they put a better input stage on the F6, with even less space to work with.

What I'm saying is: maybe the true max input level on the F6 is really nowhere near +24 dB without the limiter inline.  It doesn't say that on the specs page, but still...

Jerryfreak, I wonder what your tests look like with the limiters engaged?


At this point I'm concerned, but still have not run into anything that offends my ears.

[jerryfreak]

i think if i put the limiters on it, it might reduce the LF overload, still we would be left with mediocre performance. even at low signal levels the distortion is a bit out-of-range for gear of its class.

worth noting that if you have an even half decent DAC (most are +4dBu nominal unbalanced) local testing can replicate almost any of those results

[fotoralf.be]

"The distortion values ​​are outstanding...The F6 takes measurements at full level. Total distortion of about 0.0083%"

Almost...  ;-)

"The distortion values aren't quite as outstanding. According to my measurements, the F6 reaches a THD of 0.0083% at maximum level. This is a very good value for a mobile unit running on battery power but most stationary audio interfaces (costing several times as much) nowadays reach far better values."

HTH
Ralf

[Organfreak]

I listened to the samples of Voltronic (Laptop>Audioquest Dragonfly>Rega Brio-R>Totem Arro). It sounded beautiful, no sign of distortion or whatever. The artists were also above average level.

I will use the Zoom F6 only with microphones and will never reach the critical border of ~+17dbU where distortion really starts. I will maybe even not reach the maximum of +4dBU as listed by Zoom for its mic inputs.

Despite the apparently bad specifications I liked the perceived sound quality of the Zoom F6 more than the SD MixPre 3. Using my ears and my microphones... While the MixPre 3 may show much better test results. Would that be the different approach in preamp (Zoom using the THAT 1510 chip <-> SD using their own developed discrete class A preamp)?

[voltronic]

I listened to the samples of Voltronic (Laptop>Audioquest Dragonfly>Rega Brio-R>Totem Arro). It sounded beautiful, no sign of distortion or whatever. The artists were also above average level.

Glad you liked them.  Nice playback chain you have there!

You might also be interested in this folder of samples which has a recording I did recently of an organ selection - Conrad Susa - March for a Joyous Occasion.  This was with my CM3s in a 30 cm / 100 deg array, and were positioned to best capture the choir at this concert.  The organ selection was there mainly to give the choir a vocal rest in lieu of an intermission.  These are not the mics nor the arrangement I would choose for an organ recital, but I think it came out OK, considering.

[jerryfreak]

Despite the apparently bad specifications I liked the perceived sound quality of the Zoom F6 more than the SD MixPre 3. Using my ears and my microphones... While the MixPre 3 may show much better test results. Would that be the different approach in preamp (Zoom using the THAT 1510 chip <-> SD using their own developed discrete class A preamp)?

analog gain stages definitely have more varying sounds than ADCs. most DACs and ADCs have reached the level of 'transparent'

also worth noting that not all 'distortion' is undesirable
thats why transformer preamps, tube amps, and tape saturation are sought after by some people. all impart varying types of distortion which people find pleasing to their ears