VB-Audio's Forums

Hi,

Since I saw a request for feedback on Release Candidates, I thought it would be okay to start a new thread on 3.0.2.5 (Potato).

Noting the 'Correction/new functions' lists following 3.0.2.2 I tried to test the items that were relevant to my usage (sorry for not testing it all).

I absolutely like 3.0.2.5 and see NO major issues.

I've never seen crashes or anything, but I did run into some useability issues and nice to haves that I'd like to mention. Nothing dramatic.


Compressor and Gate details

Very cool to have this level of control!

Useability

-At first I found it strange, that a compressor threshold of say -30dB, with a high ratio, did not reduce the channel output level even though I did see a big reduction on the red meter. Then, I found this was due to the 'Auto MakeUp' which is on by default. Perhaps turn it off by default?

-Sometimes, clicking the AutoMakeUp button discards all settings. Not so nice

-When in the Compressor/Gate details window, clicking the Compressor/Gate view still sets the underlying effect (modulation/voice/postion). Odd

-Setting exact values is a bit difficult using the sliders. Perhaps an idea to make small adjustments possible when pressing shift or alt or something?

Nice to haves

-The maximum release time of 500ms is a bit low, could do with up to 5 seconds in cases where I want to level music from various sources to the same level (using it as AGC more than as a limiter). Perhaps a non linear scale would help to allow this range of settings?

-The ratio is limited to 8. Having a hard limit would be nice (infinite ratio ie flatline, brickwall limiting)


C5-Limiter (Multiband Compressor)

Awesome again, but with some useabilty and nice to have remarks.

Useability

-Ability to save/load settings for just the C5

-ABS link type is tricky as it will reset any finetuning of other bands when a value is changed, and there is no 'undo' (and no save/load)

-The limiters do distort more than what is indicated, especially when driving them hard (clipping)

-I do think from what I've heard that Voicemeeter has some headroom internally, so when driving the multiband to get high density I don't hear distortion when reducing output gain, but a final compressor limiter on the output is missing

Nice to haves

-Again, ratio on the channel limiters could do with an indefinite setting, even though the channel limiter does take care of that....

-Coupling compressors between bands (to some degree) would be nice

-Being able to set the slope of the crossovers


VBAN audio streaming

Useability

-When receiving a stream, the stream name has to be an exact match. Sometimes this causes confusion when working with others and getting stream details right. Then it is visible that there is an incoming stream, but the streamname isn't shown


Macro Buttons

Awesome to now have the Button.Color control. Very nice indeed! Works like a charm.

Nice to have

Being able to change the button name or button sub name as well, programmatically


Again, these are all minor items. I hope this feedback helps in some way.

I Just wanted to add my "+1" to all of the above, and say thanks stoepie for posting this.

My thoughts....I'll put the important parts in bold so busy people can just read those as a 'TL;DR'

I especially agree about compressor ratios, ability to set specific values, and crossover slopes

I'm already loving the button colour macros.

I think 'auto makeup' being on by default might be smart - For audio pros like you stoepie, it will be easy to find the parameter causing the unexpected results... but if you are a newbie, the compressor might seem a bit dumb without 'auto makeup' being on, and you wouldn't know that is the parameter which you need to change.

I tested out the new RC after your post about latency, but I wasn't able to replicate the fault. Glad you got that sorted out.

I did test the new denoiser and found it to be very effective, but the latency increase was too high. In my existing VST denoiser plugin, I chose to sacrifice some denoiser effect, in exchange for lower latency, by selecting a shorter FFT window. Because VM's denoiser introduced more latency, I'm not able to use it. I think an option to use a shorter window/buffer/whatever the magic is here in VM, would be fantastic, if that's possible? That, combined with C5 would allow me to completely remove the VST chain and just use VM's onboard effects

I noticed the registry keys for advanced users were reset during the upgrade/installation, but I don't think that's a problem. They're for advanced users. I guess this is intentional, but I mention it just in case

There is an old bug which still exists - Virtual input device bit depth and sample rate are still strange. The output devices default to 24/48 as per a few versions ago, but the inputs' settings are still strange. One is 44.1, one is 48, and one is 96k, and they all seem to default to 16 bit. This means that the default configuration forces sample rate and bit depth conversions on everything passing through the cables, which is directly opposed to the advice given on the correct configuration.

Thanks, also for posting good stuff.

You're right about about the 'auto makeup'. For me it would have been RTFM, but there is no 'Fine Manual' yet for these functions, is there? Makes sense for a release candidate. No issue.


Regarding the macro button colors, I am having issues using the API, but I already started a thread about it. Will add some Python test stuff there as well (as it's more accessible than writing console apps in C++)

Tried VBAN Receptor Lite today, on my Android phone. Now that's a cool little application!

Okay, somewhat unrelated to 3.0.2.5. but wanted to mention it. Give it a try!

thanks all for your feedback

1- AUTO MAKE UP is made to compensate the compression and finally boost the voice. then you can decrease the output gain by the Slider if you feel it too loud. RATIONALS: Since our slider is limited to +12 dB we considered it was not possible to make the auto make up with it (because +12 dB is not enough for big compression) - so i decided to activate the AUTO MAKE UP per default.

2- Now this AUTO MAKE UP option can be disabled by the detailed view in Voicemeeter Potato. But I did not see any way to discard other settings by switching this AUTO MAKE UP Option.

3- Don't understand your remarke : "clicking the Compressor/Gate view still sets the underlying effect"

4- you can edit value by CTRL + Right Click

5- OK for limiting the Release Time to 5000 ms.

6- ratio max stay on 8 (because we have also a limiter if needed)

7- You can Save / Load C5 preset by the System menu (click on icon on top-left of the C5 Dialog box) - it's the same for all FX in Dialog box (EQ, Reverb, Delay).

8- the C5 Limiters distortion parameter control is to manage the limiter reaction rate (distortion rate is pending on many other paremeters before), this is just an indication to make you understand the moving this knob will have consequence on loudness right, but distortion amount.

9- Voicemeeter implement brickwall limiter on all BUS anyway.

10- the VBAN Stream name is part of the stream identification. This is the only way to allow streaming several streams from a same IP to another.
(basically, a stream is identified by an IP and a stream name. then this couple must be unique).

11- I note your remark about button name/sub name.

12- denoiser add latency by nature. but we will see if we can reduce it at term...

13- for Virtual Input configuration you may check the latest DriverChecker app , provding a page to possibly change the audio format of all our virtual audio devices...

Hi Vincent,

Thanks for going through all items in such detail!

Here's my feedback/notes. One in bold I think you maybe missed?

1 -> the 'output gain' slider in the gate goes up to +24dB which I think is enough, but you have a valid point with the AUTO MAKE UP. It works, it just was a bit unexpected to me, that's all.

2 -> I tried but could not reproduce. Odd but okay, must have been drunk

3 -> See the image below, hope this explains it. 4 -> Nice one, cool. Thanks!

5 -> Thanks!

6 -> Okay good point.

7 -> Wasn't aware, having never used any FX before may explain why....

8-> Understood about the limiters. What about a final compressor like the band channels, for the final mix in C5 (output limiter?)

9 -> On all bus but I'd like to have a 'legal' level from C5 to begin with, not having to put limiters everywhere (and forget...)

10 -> I understand (I do a thing or two with streaming media), and the uniqueness isn't what I was talking about. Not being able to see the stream name on the VBAN receiver is what I meant to address. Then, you see an incoming stream but until sniffing with Wireshark there is no way to know the streamname. As a remark: I would be quite happy to have IP::PORT uniqueness, would be nicer from a router/firewall perspective anyway. But that's a whole new discussion

11 -> Thanks!

12 -> Great!

13 -> Wasn't even aware of this but I'll go have a look. Live and learn!

I'm using Voicemeeter more and more, thank you for creating and supporting it. Awesome

-David

Installed the 3.0.2.7 version... and really laughed from joy when seeing the "Precision Tri Slider". Now that's pure art!

A bit over the top lol but it works beautifully. Kudos!

I expected that pressing shift would just put the mouse on a handbrake so setting the right value would be easier. Would have been good too. LOL.

Wow, duly impressed.

The 5 second Release Time in the Compressor is very nice too, thank you!

Installed 3.0.2.7 out of interest for Noise Reduction and Advanced Compression, been running for 8 hours and no issues, in fact I believe that the ASIO Bridge implementation is better, previously I had sync issues with Cantible causing robotic voice, clicks and pops. No longer have those problems. I still have to understand the compressor and tune noise reduction but it is stable and just works even with ASIO and a VST chain of 8 plugins!

Waiting eagerly for updated manuals etc.

Great update!

Thank you to all the developers!

thanks all for your feedback and follow up.
the tri slider is an old control from my early plug-in series (2000-2005).

I have also improved Virtual ASIO driver and virtual audio callback mechanism, to stay stable in different cases and mainly remove the problem referenced on this topic (timing callback error):
viewtopic.php?f=11&t=506

For the capabilities to connect any DAW or ASIO APP to Voicemeeter, it was crucial to provide more stable virtual ASIO driver even if based on unstable audio stream. And today, most of audio stream provided by audio driver (Windows or ASIO) are often not very stable regarding real time and timing audio callback.