Jump to content

Zoom F6 (a 32bit recorder!)


IronFilm

Recommended Posts

2 hours ago, Vincent R. said:

Oh wow. Why did i miss this critical piece of info. That's quite a major detail. Where you got this intel from?

 The zoom guy in the videos clearly states that the pres are the same as in the F8 and that the mic/ line switch needs to be in the correct position. The idea that there is "no analog circuitry before the converter" is not really believable IMO. In order to connect a microphone you need some balanced analog circuitry at the front end of the AD converter and the mic needs to see a bridging impedance for optimum signal transfer. 

Link to comment
Share on other sites

3 minutes ago, Werner Althaus said:

 The zoom guy in the videos clearly states that the pres are the same as in the F8 and that the mic/ line switch needs to be in the correct position.

 

So yeah i remember that quote. Indeed 180 degree from what he told @Patrick Farrell

 

@ZoomOfficial care to chime in on the subject?

Link to comment
Share on other sites

2 hours ago, Vincent R. said:
 

So yeah i remember that quote. Indeed 180 degree from what he told @Patrick Farrell

 

@ZoomOfficial care to chime in on the subject?

 

Yeah, that would be helpful. I'm not an EE and am going way outside of my comfort zone but I am under the impression that even AES3 digital inputs (except TOSLINK) , balanced or unbalanced require analog circuitry at the input. As long as copper is involved you are plugging into analog circuitry, the difference is whether the signal itself is analog (as in microphone signals) or digital, if it's analog then the circuit / op amp(s) determine the gain, signal to noise ratio and audio bandwidth. Please correct me if I am wrong but that is my layman's understanding. 

Link to comment
Share on other sites

5 hours ago, Bouke said:

Please, someone, record the same souce twice in 24 bits, once normal, once 30 dB lower, and amp the second one by 30 dB in post, and do a phase reverse test. Then tell me I'm stupid.

 

Parenthetically, this is what Wolfe Seeberg recommended when 24-bit recorders first became available alternatives to 16-bit DAT. He thought that recordists, accustomed to the -8db lineup tone settings of the Nagra, tended to set levels too hot. Recording at lower levels - he urged normal peaks at -30 rather than -20db - allowed ample headroom and softer portions could easily be raised without harm. 

 

Of course, there are career liabilities that may come of swimming against the usual practice. 

 

David

Link to comment
Share on other sites

2 minutes ago, Werner Althaus said:

I am under the impression that even AES3 digital inputs, balanced or unbalanced require analog circuitry at the input.

No, AES3 is already PCM. I think this "Zoom design" actually comes close to a kinda "pseudo AES42" design, or even "pseudo MEMS". 

 

Link to comment
Share on other sites

2 hours ago, Vincent R. said:

No, AES3 is already PCM. I think this "Zoom design" actually comes close to a kinda "pseudo AES42" design, or even "pseudo MEMS". 

 

 

I know that AES 3 is PCM but the electrical (as opposed to optical) signal travels over copper and has an analog signal strength, it is subject to limitations imposed by cable length, capacitance, EMI/ RFI etc. It needs to be plugged into either a balanced input (110 Ohm) or unbalanced input (75 Ohm). The fact that it is a PCM stream doesn't change the fact that it needs analog circuitry to be distributed. My point is that the claim that there is no analog circuitry involved at the input doesn't seem plausible to me.

Again, if someone with more knowledge could clear up some of these issues that would be useful.

Link to comment
Share on other sites

2 hours ago, Vincent R. said:

AES doesn't need any amplification on the receiving end if that is what you mean? Also, it can be carried over optical fiber;
https://tech.ebu.ch/docs/tech/tech3250.pdf

 

I don't mean it needs amplification on the receiving end, it just needs the correct signal strength into a balanced input with the right impedance. Sounds pretty analog to me.

I am under the impression that connecting 2 pieces of audio gear via 110 or 75 Ohm AES3 involves analog circuitry despite the fact that the signal is digital.  So we can say that even these digital inputs are subject to analog circuits before getting encoded.

To then claim that there are no analog circuits involved when connecting an analog microphone to an ADC seems utterly implausible.

 

Link to comment
Share on other sites

1 minute ago, Werner Althaus said:

To then claim that there are no analog circuits involved when connecting an analog microphone to an ADC seems utterly implausible.

 

That is not the claim, the claim is that there is no preamp involved (making up mic level to line level). 

4 minutes ago, Werner Althaus said:

I am under the impression that connecting 2 pieces of audio gear via 110 or 75 Ohm AES3 involves analog circuitry despite the fact that the signal is digital.  So we can say that even these digital inputs are subject to analog circuits before getting encoded.

You confuse me now. AES3/42, hence 67, are digital distribution protocols. Indeed it needs to be stabilised/controlled/whatever (and because of the high freq nature of a digital signal, a high impedance cable is preferred, the 110 ohms for example, to resist interference/reflections/return loss)  to get properly into the "receiver", but besides the initial A to D conversion at the "transmitting" end, there is no D/A/A/D conversion involved anymore with the aforementioned protocols. 

Link to comment
Share on other sites

2 hours ago, Vincent R. said:

That is not the claim, the claim is that there is no preamp involved (making up mic level to line level). 

 

 

It is what Patrick was supposedly told by Zoom

 

"While this is all true for traditionally designed recorders, it doesn't really apply to the F6. There are no preamps or analog circuitry before the converters which would saturate into nonlinearity. The mic signal directly feeds the converters."

 

This info contradicts what was said in the video that states that the mic pres are the same as in the F8. Even if these pres are set to minimum gain (10 dB IIRC) or no gain at all they will exhibit non linearities and limited dynamic range that do not exceed 24 bit. If not I'd call that a miracle.

Link to comment
Share on other sites

8 minutes ago, Werner Althaus said:

 

It is what Patrick was supposedly told by Zoom

 

"While this is all true for traditionally designed recorders, it doesn't really apply to the F6. There are no preamps or analog circuitry before the converters which would saturate into nonlinearity. The mic signal directly feeds the converters."

 

This info contradicts what was said in the video that states that the mic pres are the same as in the F8. Even if these pres are set to minimum gain (10 dB IIRC) or no gain at all they will exhibit non linearities and limited dynamic range that do not exceed 24 bit. If not I'd call that a miracle.

Although it is contradicting (Sam/zoom statement), it is not unknown in the audio/music bizz, to bypass a preamp. Lots of people use "hotter" microphones straight into compressors without preampfification, they do it for an effect, usually to really PUMP the sound, but it is not unheard of.

 

SO, since we are speculating here ANYWAYS, it is an interesting theory, if the AtoD (32 bit?) converter indeed can handle lower signals, thus do not need the amplification a preamp gives (all in all, a preamp is there just to add gain), It is all up to the microphone what the actual noise/spl specs are. 

 

Link to comment
Share on other sites

1 hour ago, Werner Althaus said:

 

Yeah, that would be helpful. I'm not an EE and am going way outside of my comfort zone but I am under the impression that even AES3 digital inputs (except TOSLINK) , balanced or unbalanced require analog circuitry at the input. As long as copper is involved you are plugging into analog circuitry, the difference is whether the signal itself is analog (as in microphone signals) or digital, if it's analog then the circuit / op amp(s) determine the gain, signal to noise ratio and audio bandwidth. Please correct me if I am wrong but that is my layman's understanding. 

If you want to call copper wire analog circuitry then sure, there necessarily has to be some. But copper wire doesn't care what voltage it is carrying (within our normal operating range of mic or line) and thus won't introduce distortion. I haven't seen schematics so I don't know for sure but I asked specifically and he said the mic signal goes straight into the converters. Also if feeding it line level, he said the mic/line switch shifts the input range of the converters.

Link to comment
Share on other sites

3 hours ago, Vincent R. said:

 

You confuse me now. AES3/42, hence 67, are digital distribution protocols. Indeed it needs to be stabilised/controlled/whatever (and because of the high freq nature of a digital signal, a high impedance cable is preferred, the 110 ohms for example, to resist interference/reflections/return loss)  to get properly into the "receiver", but besides the initial A to D conversion at the "transmitting" end, there is no D/A/A/D conversion involved anymore with the aforementioned protocols. 

 

 

Yes, I can see that I am not getting my point across. I never said that there was any subsequent conversion, my point is that just because it's PCM audio doesn't mean it doesn't involve analog circuitry. Every AES3 input utilizes some form of analog components, transformers, resistors, etc. that's all I am saying. For me this matters in the context of the claim that there's no analog circuitry involved at the input of the Zoom F6. The point being even if there were digital inputs on the F6 it would still involve analog circuitry and with analog inputs there certainly are. As such the analog circuitry dealing with mic signals certainly does affect the dynamic range and linearity, regardless of whether it's a mic pre or just a line amp. I hope I'm making sense now.

2 hours ago, Vincent R. said:

 

 

SO, since we are speculating here ANYWAYS, it is an interesting theory, if the AtoD (32 bit?) converter indeed can handle lower signals, thus do not need the amplification a preamp gives (all in all, a preamp is there just to add gain), It is all up to the microphone what the actual noise/spl specs are. 

 

 

 

I get the concept, it's been around since the early 2000s with Stagetec's "true match" converters. But in this context it's made to sound like mic preamps only add noise, therefor eliminating them will reduce noise and give you the microphones true dynamic range. The reality is a bit more complicated and you can already do this anyway with any 24 bit encoder to get 144 dB dynamic range, ah if it were only this easy.

Link to comment
Share on other sites

16 minutes ago, Werner Althaus said:

doesn't mean it doesn't involve analog circuitry. Every AES3 input utilizes some form of analog components, transformers, resistors, etc. that's all I am saying.

Sure, but in your line of thought, a whole computer is analog then, since they have transformers/resistors/caps etc... 😁
 

 

16 minutes ago, Werner Althaus said:

As such the analog circuitry dealing with mic signals certainly does affect the dynamic range and linearity, regardless of whether it's a mic pre or just a line amp. 

And that depends; if indeed the claim being made by Patrick/Sam is true, and there is no amplification in there (yes, maybe some circuitry to get it all proper, but NOT gain), it is getting interesting. Hence my previous mentioned "Pseudo MESS" microphone (these MESS mics, usually in phones and tablets, also take the direct electricity from the mic into the digital domain, kinda).  

 

Link to comment
Share on other sites

2 hours ago, Patrick Farrell said:

If you want to call copper wire analog circuitry then sure, there necessarily has to be some. But copper wire doesn't care what voltage it is carrying (within our normal operating range of mic or line) and thus won't introduce distortion. I haven't seen schematics so I don't know for sure but I asked specifically and he said the mic signal goes straight into the converters. Also if feeding it line level, he said the mic/line switch shifts the input range of the converters.

 

I'm not calling copper wire analog circuitry, I'm calling transformers, resistors, op amps , etc. analog circuitry. 

And that argument was made by me with regards to digital signals (PCM) being carried over copper, with the Zoom F6 it's analog signals that will pass through analog circuitry and it will affect the bandwidth linearity and dynamic range whereas with digital signals the analog circuitry will not affect the analog audio signal per se, just the quality of transmission, error rate, jitter, etc.

Link to comment
Share on other sites

I think the main point here is that Zoom seem to imply that they have implemented something like StageTec‘s TrueMatch converter. While this also doesn’t work without analog circuitry, it does work without a preamp. And it claims to convert the entire mic‘s dynamic range, and it boasts a 153dB dynamic range - packed into 28-bits. And its own noise is lower than the thermal noise of the microphone. That is StageTec‘s claim anyway. 

If Zoom has come up with something like that, and at such a price point, it would be amazing. If done well...

There are converter chips around that are capable of 145+ sampling rate, so it’s getting there. Just not sure about the whole implementation 

Link to comment
Share on other sites

2 hours ago, Constantin said:

I think the main point here is that Zoom seem to imply that they have implemented something like StageTec‘s TrueMatch converter. While this also doesn’t work without analog circuitry, it does work without a preamp. And it claims to convert the entire mic‘s dynamic range, and it boasts a 153dB dynamic range - packed into 28-bits. And its own noise is lower than the thermal noise of the microphone. That is StageTec‘s claim anyway. 

If Zoom has come up with something like that, and at such a price point, it would be amazing. If done well...

There are converter chips around that are capable of 145+ sampling rate, so it’s getting there. Just not sure about the whole implementation 

 

Yes, even though true match mic inputs provide up to 70 dB of gain per Stagetec's website.

 

"

Gain Up to 70 dB (clickfree digital adjustment in 1-dB steps)

 

 

Anyway, my guess is that the analog circuit inside the XMIC+ is a big, if not the most important part of the equation. It also should be noted that digital gain, while in theory "free" is never able to increase the dynamic range at capture and that depends on the analog circuitry used.

Link to comment
Share on other sites

2 hours ago, Werner Althaus said:

 

It is what Patrick was supposedly told by Zoom

 

"While this is all true for traditionally designed recorders, it doesn't really apply to the F6. There are no preamps or analog circuitry before the converters which would saturate into nonlinearity. The mic signal directly feeds the converters."

 

This info contradicts what was said in the video that states that the mic pres are the same as in the F8. Even if these pres are set to minimum gain (10 dB IIRC) or no gain at all they will exhibit non linearities and limited dynamic range that do not exceed 24 bit. If not I'd call that a miracle.


Maybe the Pre Amps are only used in 24 bit mode and not 32 bit mode?
 

Link to comment
Share on other sites

9 hours ago, Vincent R. said:

Sure, but in your line of thought, a whole computer is analog then, since they have transformers/resistors/caps etc... 😁
  

 

 

I dunno if I'd go along with that, instead i'd offer this quote from Barry Henderson of IZ technologies (RADAR) " All digital signals are analog signals and have to be treated as analog signals".

Link to comment
Share on other sites

7 hours ago, IronFilm said:


Maybe the Pre Amps are only used in 24 bit mode and not 32 bit mode?
 

 

Maybe, who knows? we can speculate all day but the video from NAB gives me a few clues about what's really going on here. The Zoom person claims something along the lines of "24 bit audio converters do really well recording at higher levels but not so well at lower levels" or something to that effect. This statement is misleading if not false. What exactly are these lower levels, what makes them "record bad" and what does the word length have to do with it?

I don't have experience with current super high end converters but If I record with my Euphonix AM703  into the MADI input of my RADAR Studio I can record "low levels" just as good as high levels, no quality difference as a result of low recording levels whatsoever, at least to my ears. Same is true for really high levels, no problem, high and low level content will be recorded without negative side effects as long as I stay away from the maximum analog input level that corresponds to 0 dBFS. 

If I'm recording with a cheap interface or recorder the same is not true even though the cheap interface also has a 24bit converter but low levels sound anemic and noisy while levels close to 0dBFS sound distorted and edgy. There's a smallish range where the audio sounds reasonably good but both hot and low levels tend to be less than good sounding. What this tells me is that the difference is in the analog realm, the power supply, the op amps, the clock, all part of the analog topology surrounding the ADC. If low levels fed into a 24bit ADC sound "bad" it's probably a function of those components rather than the word length used to encode the audio since it is already (theoretically) capable of a staggering dynamic range of 144 dB

Link to comment
Share on other sites

2 hours ago, Werner Althaus said:

If I'm recording with a cheap interface or recorder the same is not true even though the cheap interface also has a 24bit converter

2 hours ago, Werner Althaus said:

at least to my ears.

 

Yeah for shit and giggles I read this threat again a while ago at Gearslutz about that...😁
Spoiler alert; most people preferred the cheap as shit Behringer over the Aurora...

https://www.gearslutz.com/board/gear-shoot-outs-sound-file-comparisons-audio-tests/335267-lynx-aurora-16-vs-behringer-ada8000.html

 

Link to comment
Share on other sites

6 hours ago, Werner Althaus said:

 

I dunno if I'd go along with that, instead i'd offer this quote from Barry Henderson of IZ technologies (RADAR) " All digital signals are analog signals and have to be treated as analog signals".

 

Oh well, the signal integrity requirements are certainly different. Digital signals tolerate certain amounts of noise that would make a transmission channel unusable for quality analog transmission. But of course at the end of the day signals are signals and common transmission line problems can ruin your day. 


Exhibit one, a silly example. Unterminated RS485 serial bus (note the overshots). It worked but before terminating resistors were installed it tended to fail sometimes. I guess when someone went near the unshielded twisted pair cable with a mobile phone. The second image shows a better terminated bus. 

 

Now, modern networks at high speeds enter the realm of microwaves, where circuit design needs to take Maxwell's equations into account rather than simple Kirchoff's Laws :) Try to assemble a digital circuit on a simple breadboard. It will work at 1 MHz. Now try 100 MHz ;)

 

 

owa3x.png

DS1Z_QuickPrint2.png

16 hours ago, Werner Althaus said:

 The zoom guy in the videos clearly states that the pres are the same as in the F8 and that the mic/ line switch needs to be in the correct position. The idea that there is "no analog circuitry before the converter" is not really believable IMO. In order to connect a microphone you need some balanced analog circuitry at the front end of the AD converter and the mic needs to see a bridging impedance for optimum signal transfer. 

 

I understand that he means "no analog gain stages" or even "no active components in the analog signal path". 

Link to comment
Share on other sites

  • 5 weeks later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...