Umami Audio Perfect Soup MIMO

[Stingber09]

Hi, I decided to make a separate post for my results using Umami Audio Perfect Soup Room Correction Software. I tested it with a 9.8.6 Atmos System.
I never heard my room as cohesive and clear. Bass was precise and tight. Below you will find my before and after measurements.
I plan to add a block diagram of the hardware used.

Umami Audio: https://umamiaudio.com/how-it-works/

System Calibrated:

System Uncalibrated:

System Calibrated with LFE:

System uncalibrated with subs:

Room Layout:

Left Front speaker Calibrated Decay Time:

LFE subs calibrated Decay Time:

Speech Transmission Index (STI) Analysis:

Umami Audio VST Plugin:

 

[PH73]

Hello, chiming in with my own experience of beta testing Perfect Soup with a much simpler 5.1 system. I am seriously impressed with this. It sounds great, the software is very slick and the developer has been really helpful and responsive.
I am somewhat surprised that this worked at all given the quality of the measurements I made: low SNR and potential clock drift issues (I used using flexasio to allow HDMI output to AVR with Microphone In through a 2 ch. interface). I intend to re-run the process with some better measures as I now have a multi ch. interface.
Anyway, for now here are some before and after measurements from REW.

 

[StabMe]

How many measurements per speaker did you guys perform for the calibration? Were you measuring for multiple seats or a single seat? Been intending to make it try it, but the thought of 30-60 measurements per driver makes me hold off

 

[PH73]

I took 30 measurements. The software cycles through each speaker in a couple of minuyes, much like an AVR calibration. Certainly a boring hour but you only need to do it once !

 

[StabMe]

Will devote a day-off for it.

Does the software introduce any latency into the chain or does the chain stay almost latency free and can be used for cases when shifting the video playback is not an option (games, playing of the instrument, live tv with no option to add latency to the video signal)?

 

[PH73]

I have not noticed or measured the latency but it is stated as 10ms in this thread:
https://www.audiosciencereview.com/...r-multi-sub-auto-optimizer.70108/post-2613475
(Sorry must learn how to post links properly)

 

[StabMe]

So my setup is based on an HTPC with a 10 channel interface with Reaper as the host for x-over and VSTs. My mains are 2-way active and i also have 4 subs.
Since Perfect Soup does not handle crossover, i thought i'd create a virtual ASIO device with ReaRoute, make PS calibration wizzard use it and use a USB mic. But calibration wizard requires the use of same ASIO device for playback and recording (mic), so USB mic wont work. I do have an XLR mic that i use with audio-interface but since my intention was to use ReaRoute as some kind of an intermediate virtual device between calibration wizzard and Reaper, i dont think i can make XLR mic connected to the interace available via ReaRoute. Or can it?

What would you suggest i do, dear ASR members? Maybe @umamiaudio can add USB mic support?

P.S. Looks like i can make analog mic available via ReaRoute so that calibration wizzard will see it. Will definitely try it and repoort back. But USB Mic Support would still be lovely

 

[umamiaudio]

I took 30 measurements. The software cycles through each speaker in a couple of minuyes, much like an AVR calibration. Certainly a boring hour but you only need to do it once !

Having a comfy / non-creaky chair way off to the side and something you can do on your phone like reply to some forums / emails or playing a mobile game helps a lot.

 

[umamiaudio]

So my setup is based on an HTPC with a 10 channel interface with Reaper as the host for x-over and VSTs. My mains are 2-way active and i also have 4 subs.
Since Perfect Soup does not handle crossover, i thought i'd create a virtual ASIO device with ReaRoute, make PS calibration wizzard use it and use a USB mic. But calibration wizard requires the use of same ASIO device for playback and recording (mic), so USB mic wont work. I do have an XLR mic that i use with audio-interface but since my intention was to use ReaRoute as some kind of an intermediate virtual device between calibration wizzard and Reaper, i dont think i can make XLR mic connected to the interace available via ReaRoute. Or can it?

What would you suggest i do, dear ASR members? Maybe @umamiaudio can add USB mic support?

P.S. Looks like i can make analog mic available via ReaRoute so that calibration wizzard will see it. Will definitely try it and repoort back. But USB Mic Support would still be lovely

There's no plan for official USB mic support at the moment, reason being that I don't want people to give it a bad rep becuase of some weird clocking issue when it really has nothing to do with my algorithm. At the same time I can't stop people from trying to run it with flexasio or something. I'm happy to see these very nice looking results though!

 

[john61ct]

Yes, my normal converter/interfacii for DSP have no preamps, so for measurements specifically I bought an E-MU 0404 4-Channel interface.

$40 delivered

Sonarworks SoundID Reference mic, includes calibration file - not just on axis 0° but also 90° and 30°

$90 delivered

Will compare to UMIK-2 when using REW with acoustic reference

which apparently is far superior to UMIK-1 wrt measuring all timing issues

 

[Satanica]

There's no plan for official USB mic support at the moment, reason being that I don't want people to give it a bad rep becuase of some weird clocking issue when it really has nothing to do with my algorithm.

Interesting comment.

Are you suggesting that Dirac Live in all of it's forms (DL, BM, BC, ART) has a bad rep?
And if so, could some of that be attributed to USB microphones?
There does seem to be a lot of Dirac Live users using UMIK-1 or UMIK-2.

 

[umamiaudio]

Interesting comment.

Are you suggesting that Dirac Live in all of it's forms (DL, BM, BC, ART) has a bad rep?
And if so, could some of that be attributed to USB microphones?
There does seem to be a lot of Dirac Live users using UMIK-1 or UMIK-2.

I'm not suggesting that at all. I've never used their software and have no idea what their measurement process is like. There are certain things you can do to mitigate and reduce various types of clocking errors, which they may or may not have implemented. It might even be "good enough" out of the box (as seems to be the case here). I just have many other higher priority items on my todo list right now since this was originally designed for the studio market where every user will already have devices with xlr inputs. It might be added in the future, just not right now.

 

[anlin]

Dirac sweeps each speaker in the system individually and then the first speaker again for each measurement position. The second sweep of the first speaker is probably there to compensate for clock drift

 

[john61ct]

And no one is suggesting that Dirac is the be all and end all.

It is targeted at a mass market so obviously compromises are easily made for expediency.