TBProAudio releases mvMeter - Multivariable Meter including RMS, EBUR128, VU and PPM measurement for

VST, AU, AAX, CLAP, etc. Plugin Virtual Effects Discussion
Post Reply New Topic
RELATED
PRODUCTS
mvMeter

Post

Compyfox wrote: VU: yeah, smoothed but depending on the incoming signal, it's still offset by +/- 1dB to 1,5dB. I consider this within "acceptable parameters"
It would be helpfull to tell the auditorium which test sequence and environment, thank you.
Compyfox wrote: R-128 meter: due to the +3LU "relative" scale, it's still nonsense to use this type of meter. The official R-128 test sequence goes +5LU and beyond. This makes this meter completely unusable for measurements
Needle goes up to +3LU, right; numerical values go correctly a far beyond that.
Compyfox wrote: - all numeric values and the max-hold needle is reset, if you stop/rewind your project in Cubase 7.5 x64 (no infinite hold) - this is a known problem with your engine
This not a bug but a feature: you can change needle function by selecting "M" on the right middle button.
Compyfox wrote: - sporadic "jump" of volume knob on copied instance still happening
Does not happen here, needs more investigation
Compyfox wrote: - no "numeric insertion" of gain values
Click with right mouse button on knob, online manual is updated
Last edited by TB-ProAudio on Sun Jan 22, 2017 4:23 pm, edited 1 time in total.

Post

provoc wrote:
TB-ProAudio wrote:
provoc wrote:Found a nasty bug in reaper where when you click on the gain to drag but it jumps straight away to +24dB!
Latest release v1.0.2 should fix this.
Thanks, tested and no problems with 1.0.2 now!
Thank you!

Post

Compyfox, I appreciate your efforts in testing and pointing out all the technical details regarding these meters. Some valuable information there and glad to see things got improved along the way. However, your tone has been pretty aggressive with it. TBPro handled that issue best possible way, keeping on working and simply not engaging in useless arguing. I think you would have to agree with that because by your own statement, there were no proper arguments against anyway. I'm interested in seeing if proper communication and discussions are possible going ahead.

Where I have to disagree somewhat with you are the copycat blames. From the distance all needle VU meters will by their very nature look more or less similar, they actually have to if going by any regular standards. I do agree that most of the nine examples look like from a close family, both companies offering three skin options and one of them being whitish/creme. But if we really look at the details, they are all different. Exactly the details you pointed out and more, they are simply not the same. I don't think just the idea of having three skins is patended, but maybe TBPro should go for four, just because.

I don't see the point in your channel strip example. You could've chosen some other companies if you really wanted to show features looking visually closer to DDMF. Yes, even some of the controls arranged similar way like in the DDMF strip. Yet none of them would've looked exactly alike, and neither does TBPro. If we put non-similar looks aside, all that is left are very common controls found in too many products to even count. That many of them follow similar ways to arrange their very same controls points to something else than crime. But trying to keep this non-copying concept, we surely only see one SSL, API, Neve etc. style board anywhere?

I'd like to think these things are not any coincidence but deliberate choices. Choices made to offer a specific feature set amongst dozens of other companies doing the same. You are definitely not going to make it without overlap. Especially when the buying crowd keeps looking for the same features time and time again, with whatever little can be enhanced here and there.

Again, I appreciate your work in testing the actual features. But can't help seeing the irony of having to try getting the behavior and specs exactly like they are in other similar products (for a good reason too regarding the meters), then turning around and going on about copying others. Funny old world. :shrug:

Post

Oh, so now we're having a conversation?
Interesting...

Can I now please get an answer how long your metering tool was in development?


TB-ProAudio wrote:
Compyfox wrote: VU: yeah, smoothed but depending on the incoming signal, it's still offset by +/- 1dB to 1,5dB. I consider this within "acceptable parameters"
It would be helpfull to tell the auditorium which test sequence and environment, thank you.
It would be helpful for both the "auditorium" and the developer to not just read the TL;DR version (or just the list view). Because I pretty much documented which work environments I'm in (Cubase 7.5 and Wavelab 9), and what I did. :roll:


However, we're beyond static and burst test signals at this point. Now we're talking about actual audio signals like kicks, snares, vocals, bass, strings, etc. Content that has a different bass range for the VU to respond to. And this particular test (in direct comparison to other metering tools) shows if the metering tool is linear in terms of "analysis" or not.

I do consider an offset of mere +-0,5dB/VU more than excellent. Considering a needle overshot of 1-2%, that is well within parameters in the wild with actual hardware VU's as well. +-1dB can be considered "fine" still, but it's starting to push it.

If I however have a readout difference to other established/highly accurate VU meters of anything higher +-1,5dB - then I consider the meter inaccurate and unusable and the problems need to be addressed.




TB-ProAudio wrote:
Compyfox wrote: R-128 meter: due to the +3LU "relative" scale, it's still nonsense to use this type of meter. The official R-128 test sequence goes +5LU and beyond. This makes this meter completely unusable for measurements
Needle goes up to +3LU, right numerical values go a far beyond that.
That is not the point.

The point is, that your meter has a fixed reference point of -23LUFS (like a EBU R-128 meter, which is slowly starting to get outdated, btw - see AES TD1004.1.15-10) and a "red zone" of +3LU max. This is however insufficient for actual measurements.

The official EBU R-128 test sequence (seq-3341-2011-8_seq-3342-6-24bit-v02.wav) does have a +-1LU ILk offset from -23LUFS, but the signal itself reaches up to -18LUFS SLk and -14,8LUFS MLk - on purpose even. The long-term measurement (Integrated / ILk) still shows +-1 after the test sequence because of the math happening behind the scenes. Something that this particular meter does not offer.

See the image you've probably ignored with my last post:

Image


Looking into both the EBU R-128 "recommendation" paper and the ITU-R BS.1770-x white paper, there are "headrooms" of up to +9LU mentioned for burst signals and loudness jumps, which are within acceptable parameters as long as the ILk readout is +-1LU for broadcast streams, which in case of EBU R-128 hover around +-1LU of -23LUFS.

And what do you know, 99,5% of all ITU-R type meters on the market have a digital bargraph (not a needle, unless we talk Grimm Audio LevelView and consider "arch view" as "needle meter"), with two types of relative scales: +9LU and +15LU. Just for this particular purpose.

We can talk +3LU or even +4LU for SLK material if we use the meter for Loudness Range limited music content (see my K-System v2 concept). But even then, the meter scale doesn't offer an accurate visual representation.

You could fix that in two ways:
1) change the relative scale and offer a "red zone" up to +9LU bar minimum
2) change the scale to absolute




Honestly... the more tools you create, and the more you communicate (or actually not), the more I get the impression that you just code plugins for the sake of "programming" - to proof that you're capable of doing that. But you give the impression that you have absolutely no idea as to why XYZ (metering) tool does what it does.

We had this type conversation already with dpMeter and the AES-17/+3dB offset, remember? And now you once more confirm this particular impression. So why should I trust you as developer, and further support you financially (with our other tools) in this case?

Also considering what I wrote regarding your... er... highly "inspired from" recent creations...




TB-ProAudio wrote:
Compyfox wrote: - all numeric values and the max-hold needle is reset, if you stop/rewind your project in Cubase 7.5 x64 (no infinite hold) - this is a known problem with your engine
This not a bug but a feature: you can change needle function by selecting "M" on the right middle button.
It is not a "feature," it's an annoyance.

If you use 12+ instances of mvMeter (one per individual channel, have it setup to M, then run your complete project through the VU's in order to setup your project properly / gain staging), it is essential that the max readout does stay where it is.

However (and this has been a criticism from me for dpMeter v1 already), once you hit stop and or happen to accidentally hit "rewind" or jump to a marker in your project for better visibility as to what you're doing, or simply just wait too long while doing edits in your projects (gain settings, opening/closing an UI) - the max readouts of all (other) instances are reset. Your work was therefore in vain.

If your tool wasted my time due to this, why should I even use it?

So no, it's not a "feature".




TB-ProAudio wrote:
Compyfox wrote: - sporadic "jump" of volume knob on copied instance still happening
Does not happen here, needs more investigation
Then investigate... could be one of the many iPlug issues...




TB-ProAudio wrote:
Compyfox wrote: - no "numeric insertion" of gain values
Click with right mouse button on knob, online manual is updated
Was not written in the PDF manual (up until this point). I therefore consider it non-existent. Can't find your so called "online manual" on the actual product page either.

Furthermore, I did a right mouse click on the gain knob of v1.02 and nothing popped up in Cubase 7.5, it does however in Wavelab 8/9.





SUMMARY (for this particular topic)
You're still in a very bad position IMO, and the more you write (or lack therefore), the more I get the confirmation that you're just a "free loader". Especially with this particular plugin creation.

Collectively, I wasted about 2 (unpaid) workdays on all releases so far (including this post), sniffing out all of your bugs/issues - not to mention properly documenting them (which resulted in insults and "you're ripping on the reputation of the developer", rather than objective conversations and counter-arguments/tests). And still find more problems with each update. Something that should be the job of your Beta Testers - not me as unpaid "Joe Normalguy" (though Metering Tools are my forte, and I find accurate tools and proper education in this area important - I just try to be one of the good guys).

I find that a real pity, because some of your creations have an interesting aspect to them, your Channel Strip does look interesting. So your company "has" potential in the independent programmer realm.

But this whole... "thing"... just leaves a very bad aftertaste IMO






--------------------------------------------------






EDIT-BEFORE-POSTING: Just saw this - I'll answer within the same post, to not spam the thread with double posts.
subterfuge wrote:Compyfox, I appreciate your efforts in testing and pointing out all the technical details regarding these meters. Some valuable information there and glad to see things got improved along the way. However, your tone has been pretty aggressive with it.
So we only need to allow "everything is fine, let's hold hands and dance our names" situations these days? Last time I checked, I wasn't the one starting to throw around unnecessary insults. :shrug:


subterfuge wrote:TBPro handled that issue best possible way, keeping on working and simply not engaging in useless arguing. I think you would have to agree with that because by your own statement, there were no proper arguments against anyway.
Not quite right. (IMO and all that)

The developer ignored questions and after I took a closer look a couple of hours later myself, and pointed out obvious mistakes not to mention "copying", he suddenly went into Radio Silence mode.

Make up your own mind as to what this could mean.


subterfuge wrote:I'm interested in seeing if proper communication and discussions are possible going ahead.
I find it more interesting that you first have to call out a developer, to even start an objective conversation (after asking legit questions that were being ignored). And even then, you still see the developer having a defensive position.

I invited all other thread participants to prove me wrong. And you know what, if I would have been proven wrong (and I could confirm that), I wouldn't have been too shy to say "I messed up, you're right, sorry! Let's try a different approach". Something that you didn't see from the developer so far. Instead I got insulted and barked at from... er... "fanboys" (though I am surprised that this didn't go further, like it usually does). And after 2 maintenance updates, turns out that I wasn't so wrong with my findings after all (example: the silent inertia fix for the PPM).

I find that way more problematic to be honest... blind agreement, defense for a developer that "he didn't do anything wrong", etc. This is what other people call "snake oil" or "drinking the cool aid" - there is no natural skepticism in this world anymore. Why? Because there are just too many "bedroom engineers" these days, and those never learned how to properly put a tool to it's paces before actually using it.

So if there is something new, shiny and affordable - pick it up, don't question it. If it costs just 10bucks too much, shred it to pieces!

Now you can argue that "KVR is not a professional audio forum" (which seems to be the main argument as of late), but I disagree. This particular audio realm is professional (always has been) - technical discussions like these are considered professional - the communities just opened more and more doors for "normal" people (which is good, because I still experienced very awful elitism - which completely changed my viewpoint and made me more "outgoing" to help others).

And things like that have a way bigger impact as you'd imagine. First there is Joe Normalguy, not knowing anything about metering tools and not having any money, so he picks up this tool (since it's free), thinks "this is great" and uses it without question (since he doesn't know it better). Then there are the off-forums and magazines that praise this release to no end, while ignoring the bugs/issues - maybe even add their own spin of "how to use things the right way". These places were also to blame for the "Loudness War" to begin with - mix everything loud, use this and that meter, "the K-System meter is usable for mixing and/or setting up a project" (spoilers: it's not!).

Just... no. I do have a problem with that. Because what we (outgoing) engineers tried to accomplish with sharing our knowledge, is then instantly destroyed. It's like fighting against windmills.


subterfuge wrote:Where I have to disagree somewhat with you are the copycat blames. From the distance all needle VU meters will by their very nature look more or less similar, they actually have to if going by any regular standards. I do agree that most of the nine examples look like from a close family, both companies offering three skin options and one of them being whitish/creme. But if we really look at the details, they are all different. Exactly the details you pointed out and more, they are simply not the same. I don't think just the idea of having three skins is patended, but maybe TBPro should go for four, just because.
That is actually completely ignoring the issue.

What I criticize is the obvious copied concept (almost 1:1!), the way too similar UI's (down to needle blips!), and the nearly 1:1 design copy of the manual. This is not cool towards Klanghelm IMO. In fact, it's not the first time we see this among companies. Though at least they are creative to change the UI, have an own manual concept, etc.

But this is more than blatant. Hence my question "what was the development time?" (again, I seem to repeat myself)

Speaking of "skin amount", Klanghelm's VUMT2 Deluxe has 8 skins, plus "flip contrast" (summed up: 16, Standard has 8 skins, VUMT1 had 4). So another skin for mvMeter wouldn't... er... cut it.


subterfuge wrote:I don't see the point in your channel strip example. You could've chosen some other companies if you really wanted to show features looking visually closer to DDMF. Yes, even some of the controls arranged similar way like in the DDMF strip. Yet none of them would've looked exactly alike, and neither does TBPro. If we put non-similar looks aside, all that is left are very common controls found in too many products to even count. That many of them follow similar ways to arrange their very same controls points to something else than crime. But trying to keep this non-copying concept, we surely only see one SSL, API, Neve etc. style board anywhere?
No, what I see is that a long-known company (DDMF) created a Channel Strip that can also be found on various magazines at this point. Then I see another company that is not "as known yet" to happen to have the "same channel strip concept", with a similar UI, and more than similar setup - trying to sell for the same amount of money. Why? To make yourself "known"?

This is also my main criticism with this metering tool, the design and the usability.
If you copy somebody, at least don't make it as obvious please.



Okay, now the next argument could be (as already hinted at) "but look at zplane's PPMulator, PSP Audio Triple Meter and Klanghelm VUMT2 - aren't they all the same?". To which I can answer "no they're not".

The basic gist of it is, but the rest is in the fine print.

For example PSP Triple Meter vs Klanghelm VUMT2. Both have several metering modes, but the setup is completely different (usability/comfort), the scales are different, we talk different screen sizes and UI placements, etc. In fact... if we look at the history of both plugins: first there was PSP VU Meter, then Klanghelm VUMT, then PSP Triple Meter, and in parallel VUMT2 was in the making. You could say that both inspired each other, yet both developer "teams" went their own route. Which is as far as it could go from each other.

"But PPMulator and VUMT2 must be similar, no?" - far from that... PPMulator actually uses bargraphs (no Needle Meter possible) for DIN and Nordic, there is no "RMS" mode, the "VU" mode needs to use one of the available DIN/NORDIC/EBU/BBC scales, also the EBU R-128 meter is a bargraph. VUMT2 is (only) a needle meter, highly customization at that as well (no fixed dropdown settings), has independent modes, the k-weighting filter and the possibility to set it up to an MLk meter in fact a proof of concept (and for educational purposes). And that's just scratching the surface.

Similar "roots", but completely different concept, usability and design. They are(!) distinguishable. And they can absolutely complement each other.


Now compare VUMT2 with mvMeter...
And suddenly things are a bit too similar.


subterfuge wrote:I'd like to think these things are not any coincidence but deliberate choices. Choices made to offer a specific feature set amongst dozens of other companies doing the same. You are definitely not going to make it without overlap. Especially when the buying crowd keeps looking for the same features time and time again, with whatever little can be enhanced here and there.
There is no argument about that. People will always look for alternatives because "reasons".

But I find it more than coincidence that this particular meter, that looks nearly 1:1 like Klanghelm's VUMT2 (and works similar as well) surfaced merely 1,5 months after Klanghelm's release, and is praised until infinity while ignoring bugs and the blatant "copying".

That is what I've been addressing.


subterfuge wrote:Again, I appreciate your work in testing the actual features. But can't help seeing the irony of having to try getting the behavior and specs exactly like they are in other similar products (for a good reason too regarding the meters), then turning around and going on about copying others. Funny old world. :shrug:
It's not about getting towards the same behavior as with the competition - in case of metering tools, they are standardized. If they're not accurate, they're useless. End of story.

The UI and manual copying is a complete different issue and more than food for thought.



You seem to have started to think about it, question it, it started a conversation... so I did my job.
I can't ask for more - and can now walk away with a good conscious.
Last edited by Compyfox on Mon Jan 23, 2017 4:04 pm, edited 1 time in total.
[ Mix Challenge ] | [ Studio Page / Twitter ] | [ KVRmarks (see: metering tools) ]

Post

Phew, that's a lot of reading (and I won't pretend I understand all of it)! Bottom line for me is, I did actually quite like mvMeter but then when Compyfox recommended VUMT, for the €14 it costs it was silly to not buy it. I love Klanghelm stuff and VUMT is no exception - really well coded and works perfectly. And the great thing is, when funds allow I can upgrade to Deluxe and get a whole bunch more features.
A bit fried in the higher freqs

Post

Yeah, Compyfox knows his stuff! But IMO, most people don't even need these types of plugins. If you're not sure if you need it or don't really understand how they work you probably don't need them.

Post

*Feels kinda sorry for TBProAudio getting Compyfoxed*

Post

xx JPRacer xx wrote:If you're not sure if you need it or don't really understand how they work you probably don't need them.
Personally, I would rephrase that as "If you're not sure if you need it or don't really understand how they work you really do need them!" probably because your gain-staging is crap (and I'm speaking from personal experience).
A bit fried in the higher freqs

Post

xx JPRacer xx wrote:Yeah, Compyfox knows his stuff! But IMO, most people don't even need these types of plugins. If you're not sure if you need it or don't really understand how they work you probably don't need them.
Then I ask myself why EBU R-128 meters are so praised by so many users (unskilled/entry level especially)? As if this is some sort of holy grail. Not to mention that it "has to be" bundled with xyz metering tool these days as well?


Which brings me to the next quote:
cprompt wrote:Personally, I would rephrase that as "If you're not sure if you need it or don't really understand how they work you really do need them!" probably because your gain-staging is crap (and I'm speaking from personal experience).
I think you're still phrasing it wrong. The main questions should be:
"Am I willing to learn how a specific metering tool works? What will this offer me? And what tools are appropriate for that task?"

What's insanely missing pretty much all of the time, is the learning factor. People just assume "this is great", and then comes the sad reality "uhm... how do I use this? *hits Youtube* AWESOME - I'm now using a mastering-task-type meter xyz for mixing - because the tutorial told me to do so".


Which brings us back to the criticism with the lack of debunking...
Probably because of the lack of knowledge to begin with.



MogwaiBoy wrote:*Feels kinda sorry for TBProAudio getting Compyfoxed*
Not sure if...
Image

...I should take this as a compliment, or this was a fair warning to other devs.
[ Mix Challenge ] | [ Studio Page / Twitter ] | [ KVRmarks (see: metering tools) ]

Post

Compyfox wrote:Not sure if... ...I should take this as a compliment, or this was a fair warning to other devs.
The fact that they are now on a 1.0.2 release addressing some of the issues you have raised implies both! :D
A bit fried in the higher freqs

Post

TB-ProAudio wrote:
Chris-S wrote:What is the difference between "stereo" and mid channel?
@provoc: Can't reproduce your bug in Reaper.
"stereo" is either summed (EBU) or averaged loudness of the 2 stereo channels.

"Mid" is just the mid signal only of the stereo signal, fed to both internal channels, in opposite to "Side" signal.
Sorry, don't yet understand what "averaged loudness of the 2 stereo channels" means.
In the "Mid" mode the meter behaves like other mono-meters which I checked.
In stereo mode mvMeter is showing higher levels for stereo signals.

Post

cprompt wrote:The fact that they...
I still don't think that we talk about a developer "group". I think we only talk about one developer and one graphic designer - else we'd see a different form of quality and communication. Yet constantly writing "we" on forums, the manual (where only "the team" is written) and the page implies something "bigger".

It's a psychological thing. :shrug:


But then again... food for thought.
[ Mix Challenge ] | [ Studio Page / Twitter ] | [ KVRmarks (see: metering tools) ]

Post

Chris-S wrote:
TB-ProAudio wrote:
Chris-S wrote:What is the difference between "stereo" and mid channel?
@provoc: Can't reproduce your bug in Reaper.
"stereo" is either summed (EBU) or averaged loudness of the 2 stereo channels.

"Mid" is just the mid signal only of the stereo signal, fed to both internal channels, in opposite to "Side" signal.
Sorry, don't yet understand what "averaged loudness of the 2 stereo channels" means.
In the "Mid" mode the meter behaves like other mono-meters which I checked.
In stereo mode mvMeter is showing higher levels for stereo signals.
In stereo mode, the average level of both channels is calculated: (l+r)/2.0.

Post

TB-ProAudio wrote:In stereo mode, the average level of both channels is calculated: (l+r)/2.0.
This would clear the offset I've been mentioning on material that is "not" a test signal.

Slightly stereo offset material would actually read "lower" if the math is (L+R)/2 if you use this math for showing a mono needle (which is also the "mid" needle - see formula here: DE / EN). Rather than picking up the higher readout from either of the available channels and show that on the "mono needle view".


In fact, VUMT is doing this for the "Mono Needle View" with stereo signals as well (meaning: display M, unless you force the view into L/R mode). But not if you send in a mono signal.

This particular offset did however happen with mvMeter - which tells me this VST Plugin does have an issue with mono signals. And on top of that, the "stereo mode" doesn't take the highest peak of either available channel as readout.



Looks like I found another one...

*sigh* Does this not get old at this point?
Last edited by Compyfox on Mon Jan 23, 2017 8:00 pm, edited 1 time in total.
[ Mix Challenge ] | [ Studio Page / Twitter ] | [ KVRmarks (see: metering tools) ]

Post

TB-ProAudio wrote:In stereo mode, the average level of both channels is calculated: (l+r)/2.0.
And this is different from Mid?

My understanding is that Mid=Mono=(L+R)/2

EDIT: After some testing I think that in Stereo mode you calculate the max(L,R) and in Mid Mode the average (L+R)/2.

Post Reply

Return to “Effects”