It's been while since I've used AES, but I'm encountered an odd issue (again). When I connect my Mytek Brooklyn 2 channel ADC to Burl B80 via AES+BAES4 card, the signal that reaches the DAW (and RME Totalmix) is 6db higher than what the Mytek peak meters show. (I am also clocking Burl from Mytek via BNC)

This does not happen when I use the Mytek USB output to a separate computer. When I do that, the input level is exactly the same as the meters, and roughly the same as the BAD4 ADC card.

I can't find a digital input trim in TotalMix or the DAW that could be doing this.

I've also experienced this in a different scenario: Home audio. Using an ethernet bridge to AES, the input of my Mytek Manhattan II DAC would always be 6db hotter.

I have all my converters calibrated to -18. What is going on? Is there something about the way the format handles digital level, or how Mytek/Burl build AES send/receivers?

Hmm the Sunday challenge...

I would guess that on the separate units 0dBfs is not referenced to the same dBu value.

Additionally I remember a few broadcast bits of kit that had a +6dBu boost option on AES, I'm guessing for super long cable runs.

Mystery.

This:

Not this:

Cable signaling levels have nothing at all to do with the audio playback levels at which the received digital data are heard.

Pretty weird. Not something I would be comfortable with not to know what caused it. What happens if you go in the red on the TotalMix meters?

The only thing that comes to mind is a signal doubling somewhere. If the same signal is sent to the same channel twice you get 6 dB gain. Maybe a monitoring signal mixed in? I can get that on my MOTU 828 if I'm monitoring through the DAW and through hardware at the same time.

Additionally maybe check the routing in TotalMix. There are ways to route and mix signals that are not immediately obvious by looking at the UI alone.

Maybe check on parallel signal in TotalMix Matrix? I fell for this once.

I'll have to double check this. It seems very straightforward, but TotalMix isn't always.

you can also disable totalmix...

You can? I didn't know that. How would you do that?

I know you can exit the application but as far as I know the routing persists. Maybe I should test that.

in the daw settings (at least when using nuendo) - i have totalmix permanently disabled (or then on a simplistic 1:1 patch) as i do all my patching via my digital desk; it would be far too complicated to troubleshoot if there would be an issue with my two madiface xt's...


oh, and regarding the inital post: i'm not familiar with any of the gear but couldn't it be that the reference level is different on one piece of gear, meaning levels read higher but do not record higher?

Ah, you mean digital audio workstation mode. I did know about the operational modes. Yes, you're right. That disables anything other than default routing.

It is possible. But I think AES/EBU prescribes -18 dBFS. That would mean that the Mytek uses -24 dBFS. That is not very common. So if I were the OP I would like to be sure it's not something else.

There is no actual standard for analogue level vs. digital level. Or if you prefer, there are a number of competing standards. The converter manual should tell you and some converters like my old Prisms will have internal jumpers or switches so you can choose.
--scott

Hugh Robjohns claims the EBU standard is -18 dBFS in this article

https://www.soundonsound.com/sound-a...-audio-systems

He certainly does know a lot more about these matters than I do.

The first thing I do when I get a new converter is figuring out what its reference level is. It varies wildly, I agree. Sometimes inputs and outputs are even aligned differently. I think that was the case with the Steinberg MR816x if I remember correctly.

Yes, that is an EBU standard for European broadcasters. I think it is part of EBU standard R-128.

It is not part of the AES3-1985 standard which we call AES/EBU, nor is it part of IEC-458 which is what has replaced the AES/EBU standard.
--scott

Why can't these folks just make up their minds already?

I still can't get over the whole pin 2 hot vs. pin 3 hot standard mess.
--scott

I was struggling to understand this til I looked at the article - because you’ve missed off half the quote.

He said 0VU = -18dBFS is the EBU standard; but that only applies at an AD or DA converter.

A digital output should be a straight 1:1 transfer, since there’s no conversion and no gain being applied.

Out of one devices via AES and into another should mean the exact same digital signal arriving, assuming the 2 boxes are clocked correctly (but jitter etc is irrelevant until playback in the analogue domain).

So yes you’re right, but it doesn’t appear relevant to the OP. Gain seems like it’s being applied somewhere for them!

I thought it was obvious that a synchronized digital link should be a bit-for-bit transfer -- anything else is broken!

Not sure what your point is. The Mytek the OP is talking about IS an AD converter. I thought he was talking about the meters on the Mytek which I assume are showing the analog levels. Which is then converted into digital using some reference level. Once that is done the level should not change, I agree with that. But apparently the receiving end uses a different reference level or there is something else going on like I suggested.

Could you possibly somehow be combining Stereo L/R to mono? That would add the 6 dB.

It’s a bit confusing, but he’s talking about the input meters showing one thing (post conversion), but the AES signal received at the computer showing a different thing.

That difference doesn’t happen when connected via USB.

So it’s nothing to do with the conversion aspect (and hence my confusion with you talking about reference levels for conversion).

It’s confusing for sure!

Very true!

That would only be true if the left and right signal are 100% identical. Usually not the case if it is stereo.

Yes, it is confusing. In fact I made an error myself when I said that the Mytek would have to have -24 dBFS as its reference level. That should be -12 dBFS.

Maybe a practical example can clarify. So let's assume the Mytek is indeed -12 dBFS. If you feed it with a 0 dBu signal its meter will show 0 dB. Then it gets converted and transferred digitally to another device that is calibrated to -18 dBFS. The analog output of that converter will then be +6 dBu. A gain of 6 dB, purely caused by the fact that both converters have different reference levels.

Yes, but if they're broadly similar, it'd be about 6dB, near enough. If it's exactly a 3dB rise and not a mono signal, then yes - this isn't the issue. But it's worth asking the question.

Um - that's not very well expressed - you're missing the units on "it's meter will show 0dB". dB what? The usual lineup config is "xx dBFS = 0VU" not dBU - that's where your +4dBU comes in. In this case, your desk should be showing 0VU, your actual level is +4dBU and your converter shows -12dBFS.

(so in fact, if you feed "0dBU" to a converter lined up to 0vU = -12dBFS, you'll actually see about -8dBFS on the display, not 0dBFS. That's where the headroom comes in. 0VU = -12dBFS doesn't actually leave much headroom when your desk can put out +24dBU without distortion, as many can!).

Yes - if you have your AD/DA chain lined up so the inputs are different, then you'll get a level difference when converting back to the analogue domain. That's expected, is understandable and simply user error (if you wanted unity gain).

That's not relevant here though - @ IanBSC is seeing a digital gain boost of +6dB between converter output and Totalmix input. Which is odd, but also seems like something fixed is adding the gain.

It's something I've run into quite a bit at my day job at a big radio/media facility. Folks have software set to record stereo tracks by default but are only recording one mic. Playback pans are centered, so therefore... Also some gear has a dual mono AES option, particularly more broadcast oriented gear.

You're probably right but I've never seen VU used in a digital context. When I think of a VU meter I think of it is as showing RMS levels, not the peak meters that most digital devices use (if they have any metering at all).

I think what you describe comes close to my own understanding. When I read the Hugh Robjohns article I was surprised that he says the EBU standard is -18 dBFS equals 0 dBu. Maybe he made a mistake there and meant +4 dBu. I don't know.
I always link it to +4 dBu so then EBU would actually be -14 dBFS.

I'm always using -18 dBFS as my reference when working in the DAW. I even created my own meter plugin to have this visualized in a way that I like and that my DAW doesn't provide.


I know, but -12 dBFS is surprisingly common. It should be illegal.
Well, if you're dealing with consumer levels it is plenty and that's probably the reason.


I don't know. I'm not so sure it is not relevant here. I think the OP's description of his issue was not very precise either. But we seem to have lost him.

When you're adding two signals together that are not identical, anything can happen. Depends totally on what kind of signal, music, speech, noise. If it is true stereo and doesn't have a strong mono component (like loud bass) the gain increase will be minimal. There are even circumstances conceivable where there will be a gain decrease.

When 2 sources in the left and right are 100% coherent (ie the same), there'll be a 6dB increase when summed to mono.

When the two sources are 100% out of phase, there'll be a cancellation to silence.

Anywhere in between, will go anywhere in between (as you said).

What I said was that if a source is strongly coherent, even if it's not completely mono (eg a mono source with some slight stereo ambience), the sum will be close to, but not exactly a 6dB gain boost.

That's why I asked the question - if it were exactly a 6dB gain boost but not a mono signal, summing to mono would be unlikely. If it were a close to 6dB gain boost with a signal with little stereo information, that is a stronger indication something like this is happening.

None of what you say is *wrong* but it doesn't contradict what I'm pointing out.

An AD meter goes analogue-digital. The VU meter shows the analogue end of the chain, the dBFS represents the digital (of course).

The VU meter is on the desk or other analogue device, the dBFS meter on the converter.

That's how digital gear is lined up. Send a tone at 1kHZ at 0VU out of the console (being tone, VU is steady of course). Line up digital side to read whatever you want.

Hmm. I was about to say dBFS references ing ALWAYS done to VU when lining up in music - because in my experience it always has been - but the EBU standard is very clearly to 0dBu!

Regardless - you have to STATE these things clearly, always including the units, so you're being clear.

In pro studios in the UK (and I think I worked in enough to say it's "the standard"), you reference to VU. So -18dBFS = 0VU, at a nominal +4dBU. Which (if I've got my sums correct) gives you an additional 4dBish of headroom so you can hit the desk output harder and not clip converters.

Well, you have to reference it TO SOMETHING - that's the whole point. If you're just doing it within the plugin chain, then you've just got a pseudo-VU reference to where the plugin operates "best" I guess.

I do a similar thing - I have virtual VU meters calibrated to 0VU = -18dBFS at the end of my chain. Though sometimes I cheat and if it's running a bit hot but sounding good, I'll nudge them up to -16...it doesn't really matter. It's just an indication for the levels the plugins expect to "see" and means you don't end up pinning compressors (and your mixbuss chain operates at similar levels consistently).

We used to line up DAT to -14, but never went that low!


I think OP has been around long enough to know all the above hopefully he'll report back what it was when he finds out!