SSL Meter Plugin Calibration to 0VU

[SonicMethod]

I'm trying to calibrate the SSL meter plugin to my DAW and converters. When I send a 1kHz sine wave at -18dBFS through the SSL meter with the reference level in the plugin set to -18dBFS, I am expecting my meter to read 0VU, but instead it is reading -4db lower. This behavior runs contrary to every VU meter plugin I have ever used.

Why am I not seeing 0VU under this scenario?

Is SSL using a different definition of "reference level" than other manufacturers/developers?

[Hugh Robjohns]

I agree, that seems weird...

...but I think the reason is that there's a second line-up adjustment in the Advanced Settings section which defaults to +4dBu. Try setting that to 0dBu instead.

Or just set the calibration to -14dBFS.

Either way, you are completely correct that your reference tone level should appear as 0VU on the meter.

Calibration of VU meters has been widely misunderstood for at least 30 years... including the manufacturers!

[SonicMethod]

That was my thought as well, but it doesn't make sense because to get 0VU under the testing scenario I outlined the reference level needs to be -22dBFS. With that setting the meter reads 0VU.

In fact, SSL's own video on the UF1 setup, states to get 0VU with the reference set to -18dBFS a -14dBFS tone must be sent. Again, none of this aligns with the normal convention I have seen and just seems wrong.

I just want to understand why.

The adjustment you mention can't be zeroed out only +4dBu and -2dBu can be selected--standard vs. French.

[Hugh Robjohns]

If I were you, I'd just dial in whatever value provides the result you want — apparently -22dBFS with the +4dBu setting — and just get on using it.

If you really want to get to the bottom of the plugin calibration you'll need to talk to SSL... maybe they have a reason in mind, although I highly doubt it.

The inclusion of options for +4 standard / -2 French settings would imply a major misunderstanding by the design team anyway. Not what I expected from SSL.

[RobinC]

I seem to recall that there were some postings about this by SSL in the Facebook group UF1/UF8/UC1/360 Users Group. The upshot was as you say use -22dBFs with the +4dBu setting. I can’t recall the details of the explanation but it’s probably still in there if you wish to search. The postings were from Andy Jackson @ SSL

[SonicMethod]

Thanks Robin that is great info. I know I can just set it and move on, but I'm really trying to understand the logic here. It to me, makes life more difficult for the user, because when I set the reference to -18dBFS in my case, I'm expecting 18db of headroom in my DAW, but I'm really only getting 14db.

I will look for the discussion thread and see if I can find out the rationale.

[SonicMethod]

Well I did find the thread. I'm even more confused:

Here is the response from Andy Jackson explaining how to get to 0VU and then why SSL is doing it this way:

"Our SSL VU meter has an offset so it can be calibrated to the analogue world. By default this is +4 dBu = 0VU.
To put this simply, if you want the SSL VU meter to be consistent to your other plug-in VU meters, just set the reference level to 4 less dBs than what you’re compare it to.
E.g. if your other VU plug-ins are set to “-18 dBFS” then set the SSL Meter to “-22 dBFS”

"The reason is so that it translates to analogue (either the standard VU of +4 dBu = 0 VU or the French standard -2 dBu = 0 VU). You might be sending out of the DAW into some piece of analogue gear where the line-up is critical etc etc. In which case, you could set the Reference to -18 or -24 to match converters etc. Admittedly, might not be a typical use case and maybe my life would have been easier just to add an option where there is no offset at all and what you see is what you get, without having to account for +4 or -2. In any case, it's all a bit arbitrary because the VUs most useful action a lot of the time is to check average signals (old school, are they moving as I expect with this mix), and the reference is really just a tool for moving the meters into the right area to match whatever gain staging you have going on in the DAW, with no significant relation to peaking levels etc."

[Hugh Robjohns]

EDIT: Post crossed with the one above.

I know I can just set it and move on, but I'm really trying to understand the logic here.

I honestly don't think there is any logic. Just confusion. Why would anyone with a functional brain put a selector for analogue operating levels in a digital plug-in?

It smacks of someone looking at a historic meter alignment chart, not understanding it, but copying the options anyway....

I will look for the discussion thread and see if I can find out the rationale.

Please share it if you do.

[Hugh Robjohns]

Well I did find the thread. I'm even more confused:

I'm not surprised... You and Andy, both!

"Our SSL VU meter has an offset so it can be calibrated to the analogue world. By default this is +4 dBu = 0VU.

This is complete loony tunes. ...and makes me rather cross with its idiocy!

Different interfaces have different analogue /digital alignments. The digital plugin meter can make no assumption of what that is — that's why it needs an understandable user-calibration function (without hidden and pointless offsets!).

The "default analogue VU alignment" is completely and utterly irrelevant to the DAW/ plugin calibration.

"To put this simply, if you want the SSL VU meter to be consistent to your other plug-in VU meters, just set the reference level to 4 less dBs than what you’re compare it to."

What an embarrassing admission!

"The reason is so that it translates to analogue (either the standard VU of +4 dBu = 0 VU or the French standard -2 dBu = 0 VU).

That is hilariously muddled — or perhaps just depressingly stupid — thinking... I'm genuinely astonished!

"You might be sending out of the DAW into some piece of analogue gear where the line-up is critical etc etc. In which case, you could set the Reference to -18 or -24 to match converters etc

...and what if your converters or interface have different headroom and alignment levels... as so many do?

"...maybe my life would have been easier just to add an option where there is no offset at all and what you see is what you get

YES!!!!

"...without having to account for +4 or -2

No one in the real world has to account for that — but if they did they might also want an option for the Mackie and euro-broadcast 'standard' of 0dBu too.

In reality, you simply need to be able to align 0VU in the plugin to your chosen digital operating/ reference level, whatever that is, without some nonsensical offset.

And then adjust the interface gain structure as necessary for the external analogue alignment (only) if that's relevant to the workflow.

"... the reference is really just a tool for moving the meters into the right area to match whatever gain staging you have going on in the DAW, with no significant relation to peaking levels etc."

It has an absolute relation to the desired DIGITAL operating level... so the SSL plugin's 4dB (or -2dB) offset is utterly meaningless and nonsensical.

Just incredible.

[SonicMethod]

Thanks Hugh for confirming that I have not lost my mind.

I had just gotten to the point of feeling like I had a very solid grasp on all this digital to analog gain staging stuff and then I try and setup this plugin and because it was created by SSL, I thought I must be doing something wrong, or, that I had not grasped things as clearly as I had thought.

You hit the nail on the head here:

"In reality, you simply need to be able to align 0VU in the plugin to your chosen digital operating/ reference level, whatever that is, without some nonsensical offset."

This was my very initial instinctive reaction as well.

Again, in practice, I can deal with it, but good grief!

[Hugh Robjohns]

[Hugh Robjohns]

This saga has been troubling me overnight, and I've been thinking about why SSL configured this plugin as they have.

Eventually, I have come up with what I figure is their reasoning. It's still a very muddled and nonsensical approach, IMHO, but there is a kind of warped logic to it... I think.

Essentially, the 'reference level' setting in the plugin doesn't relate to the chosen DAW operating level at all. It actually relates to the converters/interface being used with the DAW (but only if you're working with professional converters /interface!!!).

So, if you're using pro converters that clip at +24dBu SSL want you to select the -24dB reference level option.

When you do that, a +4dBu analogue signal input to the converter would appear in the DAW at -20dBFS, of course, and the VU meter would show 0VU because of the +4dB default offset applied to the selected reference level.

This unusual approach essentially defines and fixes the DAW's operating level according to the converter/interface design, rather than a user preference. I guess SSL argue that this is a practical solution when a DAW is always used with a professional analogue console and professional converters...

However, they also claim this works for converters with different sensitivities... but it doesn't! As an alternative scenario, if you're using converters that clip at +18dBu for an EBU environment, say -- like most older RMEs* -- the SSL advice is to set the reference level to -18dB.

A 0dBu analogue signal (the normal reference level for the EBU environment) would then appear in the DAW at -18dBFS, but the meter would show -4VU and would only reach 0VU at -14dBFS. And so if you're a European broadcaster that would have you tearing your hair out in confusion!

Basically, I think SSL have tried to make life simpler for users working with professional +24dBu converters /interfaces and professional consoles... but they actually made it massively more confusing and complicated for everyone else.

Bottom line — the default 4dB offset built into the VU plugin is unnecessary and highlights a major misunderstanding in the way the VU meter was historically designed to be used.

I find this rather surprising, because it was a (long since retired) SSL engineer who re-educated me as to the ways and wherefores of the VU meter!

Personally, I think it would have been so much simpler just to have a calibration option to set 0VU at any desired DAW level... like everyone else does. So you'd set -20dBFS for a pro SMPTE setup, -18dBFS for an EBU setup, and -14dBFS (or whatever) if you're mixing /mastering for streaming etc.

*O do know that most older RME interfaces actually clip at +19dBu, not +18dBu, but the intention behind that was to leave a gnats of converter safety headroom when adopting the EBU alignment. ie. Operating reference level, 0dBu = -18dBFS with 18dB headroom — ie, clip at +18dBu.

This differs from the SMPTE alignment which states Operating reference level +4dB =-20dBFS with 20dB headroom — ie, clip at +24dBu.

[sonics]

Marvellous, Hugh. Thanks for that. Why not just let us set the calibration level we need, indeed?

It's wonderful to know that you have sleepless nights, so that we don't have to!

[Hugh Robjohns]

[SonicMethod]

Hugh, I agree, this is a most vexxing situation!

At the very least they need to include a 0dBu option in the software as the default setting. This would solve the problem--as inelegant as it is!

-2dBu
0dBu
+4dBu

Then, if someone wanted to engage the basterdized version of the VU meter they can if their heart so desires. For the rest of us, we can go on living our best life!

I know SSL is following these threads, so I'm asking you guys to do the right thing here and fix this, if not for our sanity, do it for your industry and brand reputation!

I don't want to have to account for some nonsensical offset when setting my digital reference levels.

I love my SSL gear and am rooting for you all to step up and fix this in the next software update. I could care less about any acknowledgement of the error in your thinking, just make it right!

[Hugh Robjohns]

You're quite right — if SSL are wedded to their wacky offset concept then adding a 0dB option would work for EBU analogue installations, as well as make the reference level function work properly for everyone, regardless of their choice of interface/converter alignment.

[BWC]

I know SSL is following these threads, so I'm asking you guys to do the right thing here and fix this, if not for our sanity, do it for your industry and brand reputation!

...I could care less about any acknowledgement of the error in your thinking, just make it right!

You know it's the right thing to do!

[SonicMethod]

Yes, along with anyone with a "functioning brain".

[BWC]

My comment was aimed at (hopefully) someone from SSL following this thread, just to be clear.

[SonicMethod]

Yes, I got it BWC, thank you. I'm thinking about bringing this up in the SSL Facebook forum to see if anything can be done. I have been in contact with SSL via the help desk, but they are somewhat clueless and don't understand what I'm talking about.

Hugh mentioned earlier in this thread that this implementation seemed like "It smacks of someone looking at a historic meter alignment chart, not understanding it, but copying the options anyway...."

And that is exactly what SSL did!

How do I know this?

Well, in my back and forth with them, they sent me a PDF copy of the World Audio Level Reference Chart, explaining that that was the source document they used for the "alignment" levels used in the meter during development!

[BWC]

Yes, I got it BWC, thank you. I'm thinking about bringing this up in the SSL Facebook forum to see if anything can be done.

...in my back and forth with them, they sent me a PDF copy of the World Audio Level Reference Chart, explaining that that was the source document they used for the "alignment" levels used in the meter during development!

Sad, but reputations must be maintained. Hopefully, they will. We all make mistakes, and get confused, sometimes. It's not too late for them to make this just an embarrassing isolated incident, quickly corrected.

[Hugh Robjohns]

Hugh mentioned earlier in this thread that this implementation seemed like "It smacks of someone looking at a historic meter alignment chart, not understanding it, but copying the options anyway...."

And that is exactly what SSL did!

How do I know this?

Well, in my back and forth with them, they sent me a PDF copy of the World Audio Level Reference Chart, explaining that that was the source document they used for the "alignment" levels used in the meter during development!

Thats hilarious! ...and also very disappointing, at the same time.

If SSL want to get in touch privately I'd be very happy to help them with their understanding.

[Jay257]

Is there a definitive modern guide somewhere documenting these different level standards?
If not, it would be a good subject for an SOS article...

[Sam Spoons]

https://www.soundonsound.com/sound-advi ... y-hardware

[Hugh Robjohns]

Is there a definitive modern guide somewhere documenting these different level standards?

Inevitably, there are a few 'special situations' with odd 'standard levels', but in most professional situations there are only two common operating level standards:

SMPTE = +4dBu analogue / -20dBFS digital
EBU = 0dBu analogue / -18dBFS digital

These standards mean the SMPTE standard enjoys 20dB of headroom, while the EBU standard has 18dB — both in the digital domain. Analogue gear used in either settings could potentially offer more headroom.

...and given the context of this thread, 0VU is adjusted to be the same as whatever the chosen analogue/digital operating level is.

In the semipro world, the normal operating level is -10dBV which is (near as makes no difference) -8dBu.

If you're seeking a chart comparing different analogue and digital meter scales, try this: