Roon DSP - Individual Procedural faders [Ticket In]

My question:

I run an active 3way speaker setup.
Roon does all speaker EQ and crossover.

Signal path is:
RPI USB > Ropieee > Roon > Motu Mk5 lite > 3x stereo amplifiers.

I have noticed that applying negative or positive gain on any procedural fader alters all routed/mixed channels by the same amount (Even if you have discretely routed signal to particular channels). So effectively each procedural fader acts as headroom master gain, rather than individual channel gain which is what I would expect.

This fader:

For instance if I apply -10db of signal on channel 5/6 assigned fader, this also removes 10db on the remaining channels in the mix, I.e. 1/2, 3/4. This seems very odd indeed, (especially if you are dealing in negative gain) as I have no ‘easy’ way to level compensate, other than applying EQ in ‘positive gain’ and using headroom management to compensate, or of course use a mix of amplifier gain and reverse application of EQ to level control channels.

It seems counter intuitive for procedural faders to act as master faders, instead of individual faders for the channel they are assigned, especially in the context of speaker crossovers/management.

My thought is, once Procedural EQ has been applied to a particular channel say 1/2, its applicable fader should be discrete to those channels, and not impact remaining channels in the mix. Headroom management can then be used to prevent clipping. This is more inline with pro interfaces or DSP units, albeit most modern DSP’s use floating point to manage digital gain.

Channel 1/2

Channel 3/4

Channel 5/6. You can see a reduction of 12db is applied to the channel.
This is my de250 compression driver.

This reduction appears as a level reduction of 12db across channels 1-4, visible on my Motu signal levels. This is not right. Procedural fader control should be applied to the routed/mix channels only.

3 Likes

Moved to support area

Hello @grizaudio ,

Thank you for the report here! I wanted to touch base with some good news, which is that we’ve been able to reproduce this issue and we have put in a ticket for our developers to look into this further. As per policy, I cannot comment on timelines regarding a resolution time, but we are aware of and tracking this, thanks again for the report!

4 Likes

That’s good news.
Definitely easy to replicate. :grinning:

If you need anyone to beta test, let me know.

It seems the Procedural engine was/had been initially created with an understanding PEQ/DSP would be applied to a stereo signal only.

Obviously users like me, and those with active setups apply DSP across many channels. So discrete gain control per procedural EQ is important.

The difficulty is how do you correct this issue, without potentially impacting users who have configured multiway setups configured around this limitation.

Thanks.
Steve

2 Likes

This topic was automatically closed 45 days after the last reply. New replies are no longer allowed.

Any information on this rectification for the procedural implementation?

Hi @grizaudio ,

Thanks for checking in! I see that we have a development ticket in the queue regarding this, but work has not begun yet. I’ve sent the team a reminder on the ticket, but generally speaking we’re unable to comment on when a ticket will complete the dev lifecycle. Thanks for your patience while this ticket makes its way through the queue!

1 Like

Any update one year later?