Klanghelm VUMT 2 and VUMT deluxe released

[Compyfox]

I appreciate your input. With your apparently deep knowledge about VU meters, do you agree with the VUMT manual that all three VUMT modes are "within VU specs"?

Yes, they are well within specs.

The VU spec that you've studied (I haven't) allows for VUMT1's less linear ramp/fall behavior?

In fact, old hardware VU's had a way higher tolerance (see Hermetech Mastering's post on "no VU was 100% the same). So no matter which behavior, it doesn't matter.

It struck me that VUMT has a mode named "ideal", but VUMT defaults to a different VU mode. Later I may appreciate the options, but for now, I just want VU meter that emulates my old, real VU meters; and my old VU meters did not require me to select modes. If I want it to work like VU meters from back in the days of mode-less VU meters, should I set the mode to mode 1, mode 2, or mode ideal?

Again... don't get hung up on the modes. It's basically just giving you a different visual. There was a lot of time put into VUMT2, and I'm sure it was fine-tuned to get as close to the "source" (the device that the developer used to compare, though please don't ask me which exactly) as possible.

If you're used to old hardware VU meters, just leave the setting at default (which is VUMT2) and set up your reference point (0 VU), then forget the rest. You probably won't see any big difference at all.

[MusicProgrammer]

Again... don't get hung up on the modes.

Asking to Tony to finish documenting his software isn't what I'd call "getting hung up". I'd read the VU spec and code my own before getting "hung up" on anyone else's, but since I did buy this one, I thought I'd first try requesting documentation here to supplement the missing documentation in the manual.

It's basically just giving you a different visual.

RMS is basically just a different visual too. Peak meter is basically just a different visual too. But I would not want to interpret the output from a peak meter as RMS. A Warhol is basically just giving you a different visual from a Rembrandt.

There was a lot of time put into VUMT2, and I'm sure it was fine-tuned to get as close to the "source" (the device that the developer used to compare, though please don't ask me which exactly) as possible.

I don't do faith in developers, even when they come highly recommended by their beta testers. I just want straight info about math, no speculation about "the math must be good because I know Tony". That's not how math or science is done.

Anyway, thank you for fielding my question.

[MusicProgrammer]

The original spec for VU meters from 1940 is here:

http://www.aes.org/aeshc/pdf/chinn_a-new-svi.pdf

It allows for certain tolerances, so yes, all the modes are within spec, (much like there are tolerances for measuring ISPs, and therefore different ways of implementing them, and therefore certain meters give slightly different results).

Every real VU is gonna be slightly different too. I've been using large calibrated Sifam VUs fore about a decade now, and no plugin version comes close, they all look jerky compared to the real thing, but I do think VUMT is the best attempt so far.

Thank you. That makes sense to me.

[poonna]

The original spec for VU meters from 1940 is here:

http://www.aes.org/aeshc/pdf/chinn_a-new-svi.pdf

It allows for certain tolerances, so yes, all the modes are within spec, (much like there are tolerances for measuring ISPs, and therefore different ways of implementing them, and therefore certain meters give slightly different results).

Every real VU is gonna be slightly different too. I've been using large calibrated Sifam VUs fore about a decade now, and no plugin version comes close, they all look jerky compared to the real thing, but I do think VUMT is the best attempt so far.

Have you tried Waves VU? For some reason, I think that one is very smooth. Not sure about the accuracy though.

[scalawag]

The original spec for VU meters from 1940 is here:

http://www.aes.org/aeshc/pdf/chinn_a-new-svi.pdf

It allows for certain tolerances, so yes, all the modes are within spec, (much like there are tolerances for measuring ISPs, and therefore different ways of implementing them, and therefore certain meters give slightly different results).

Every real VU is gonna be slightly different too. I've been using large calibrated Sifam VUs fore about a decade now, and no plugin version comes close, they all look jerky compared to the real thing, but I do think VUMT is the best attempt so far.

Have you tried Waves VU? For some reason, I think that one is very smooth. Not sure about the accuracy though.

Would love to know this too. Maybe Compyfox can give his input on this...

[Hermetech Mastering]

Sorry, never used a Waves plugin so wouldn't know! Of the ones I have tried, VUMT is best, I once had it set up with the real VUs right below the monitor, and lined them up exactly at a similar size, and the ballistics were very close, although the actual needle movement was a pale comparison. Not sure why but imagine due to screen/graphic refresh rate? I think you'd need it way high to be anywhere as close as smooth as the hardware.

[Compyfox]

Take your meds.

And this is the very reason why I barely post on KVR Audio anymore... And it looks like this isn't your first rodeo regarding giving/receiving feedback. So you just got yourself a report towards the mods. Nice work - hope it was worth it.

Have you tried Waves VU? For some reason, I think that one is very smooth. Not sure about the accuracy though.

Would love to know this too. Maybe Compyfox can give his input on this...

I have the Waves VU in my possession indeed. But not in constant use.

Last time I used/extensively tested it, it was fairly accurate (+-0,3VU to +-0,5VU being well within tolerances). I think they've updated it even with a "graph delay" feature (it is not changing the needle inertia on top of the ballistics however, it is for manual PDC). It is a fixed 300ms rise/fall VU though - so no custom settings like PSP TripleMeter or zplane's PPMulator XL.

Waves is giving this thing away for free ever so often. You just have to live with the WUP shenanigans.


My go to VU is still Klanghelm VUMT2 because of it's capabilities for customization (also - UI resize!), although I do use Hornet Plugins VUMeter mk3 for "quick project gain staging setups" if I'm short on time (since it has an auto-gain feature that first checks for max peak, then average signal strength -- haven't upgraded to v4 yet).

[poonna]

Since we are talking about meters, may I ask one more off-topic question? There’s also the Dorrough meter in Waves catalog. Are the VU and the Dorrough meters used for the same purpose? Do they give the same information?

[Compyfox]

I've wrote about this on page 4 in 2016 already:
viewtopic.php?p=6627464#p6627464
viewtopic.php?p=6627668#p6627668


The TL;DR:
The Dorrough (patent) meter actually runs "twice as slow as a VU", so roughly 600ms. The meter is therefore offset by a couple of dB if you send in the exact "program material" (like a kick - I've ran into offsets of about -1,5dB to -3dB -- hence the +-1 "hotspot" around the 0 point). Compared to a pure VU, the Dorrough meter is also a "bargraph" and offers a PPM readout on top of the realtime average signal strength measurement.

It is possible to use for individual signals (and this was the main appeal why some companies later created "combi-meters" on consoles), however personally I'd not use it for that purpose, but only for a sum of various signals (as in: stereo mix, or an instrument section).

The Dorrough 40A meter was also the main backbone for the K-System v1 meter concept. Only with a shifted reference level.



VUMT2's RMS mode is based upon the Dorrough patent

[Russell Grand]

Wow. The VUMT Deluxe GUI is sexy as hell. Nice! Instant buy IMO. 

[poonna]

I've wrote about this on page 4 in 2016 already:
viewtopic.php?p=6627464#p6627464
viewtopic.php?p=6627668#p6627668


The TL;DR:
The Dorrough (patent) meter actually runs "twice as slow as a VU", so roughly 600ms. The meter is therefore offset by a couple of dB if you send in the exact "program material" (like a kick - I've ran into offsets of about -1,5dB to -3dB -- hence the +-1 "hotspot" around the 0 point). Compared to a pure VU, the Dorrough meter is also a "bargraph" and offers a PPM readout on top of the realtime average signal strength measurement.

It is possible to use for individual signals (and this was the main appeal why some companies later created "combi-meters" on consoles), however personally I'd not use it for that purpose, but only for a sum of various signals (as in: stereo mix, or an instrument section).

The Dorrough 40A meter was also the main backbone for the K-System v1 meter concept. Only with a shifted reference level.



VUMT2's RMS mode is based upon the Dorrough patent

Thanks for your detailed response! So, with a Dorrough meter, would you still recommend aiming at -18 dB average on most signals and -9 to -6 dB peak on transients for gain staging individual channels? I happen to like how the Dorrough meter visualizes loudness better. It’s also easier to see from a distant (e.g., when I’m trying to set gain for drum recording and I’m sitting at the drum set).

[Compyfox]

I understand the visual appeal of the Dorrough Meter with the "red hotspot". It is a bit easier to read from far away. Although VUMT2 can be set up with a "hotspot range" as well.

Sure, in theory you can use the Waves Dorrough meter for individual signals. But as mentioned, the ballistics are "twice that of a VU", so 600ms (aka slower than a VU). So if you use a -18dBFS reference level and your average signal strength (lowend heavy kick/toms, bass) is in this "hotspot zone" (+-1 from the reference), you actually have somewhat "undershot" compared to a VU at 300ms. So you should aim for about +1 to +2 on the bargraph, or light up 1 yellow LED to the right/the top of the hotspot ever so often as absolute max (for average signal strength).

If you measure peak, the Dorrough meter is actually a PPM with around 2-5ms, I would never(!) go higher -6 regarding maximum signal peak. Always cross check with a digital meter (the meter bargraph from your host). With instant readouts, that should never exceed -3dBFS digital max with rogue peaks.


I hope that helps for the time being.

[poonna]

I understand the visual appeal of the Dorrough Meter with the "red hotspot". It is a bit easier to read from far away. Although VUMT2 can be set up with a "hotspot range" as well.

Sure, in theory you can use the Waves Dorrough meter for individual signals. But as mentioned, the ballistics are "twice that of a VU", so 600ms (aka slower than a VU). So if you use a -18dBFS reference level and your average signal strength (lowend heavy kick/toms, bass) is in this "hotspot zone" (+-1 from the reference), you actually have somewhat "undershot" compared to a VU at 300ms. So you should aim for about +1 to +2 on the bargraph, or light up 1 yellow LED to the right/the top of the hotspot ever so often as absolute max (for average signal strength).

If you measure peak, the Dorrough meter is actually a PPM with around 2-5ms, I would never(!) go higher -6 regarding maximum signal peak. Always cross check with a digital meter (the meter bargraph from your host). With instant readouts, that should never exceed -3dBFS digital max with rogue peaks.


I hope that helps for the time being.

Thanks a lot for the tip!