DecayCore — Free FIR room correction with temporal decay control, automatic optimization and measurement workflow

[VilhoValittu]

Hi everyone,

I wanted to introduce DecayCore , formerly known as CamillaFIR.

DecayCore is a FIR room-correction and filter-generation tool for CamillaDSP and other FIR-capable DSP engines. The main goal is not just to draw a target curve and force the response to match it, but to generate correction filters that remain physically sensible, band-limited, and useful in real listening rooms.

The software can generate filters from REW exports, WAV/IR captures, or from its own integrated measurement workflow. It has Automatic, Basic, and Advanced modes, and can export ready-to-use FIR filters as WAV files, with optional CamillaDSP-related configuration assets.

The current filter types are:

• Asymmetric — recommended default for many real-world systems
• Linear Phase — for cases where linear-phase behaviour is preferred and latency is acceptable
• Minimum Phase — for lower-latency causal correction
• Mixed Phase — for users who want a balance between magnitude correction, phase handling, and practical time-domain behaviour

One of the main design ideas behind DecayCore is temporal decay control. Instead of only optimizing the steady-state frequency response, the tool also tries to avoid corrections that make room-mode decay or time-domain behaviour worse. The automatic mode includes target selection, safety limits, stereo-aware ranking, and filter scoring intended to produce filters that are usable without requiring the user to manually tune every parameter.

A few key features:

• FIR filter generation for CamillaDSP and other FIR-capable systems
• Automatic target and correction optimization
• Controlled low-frequency correction
• Temporal decay control
• Phase-aware correction
• Multiple filter strategies: Asymmetric, Linear, Minimum, and Mixed Phase
• Browser-based graphical UI
• Export of WAV FIR filters and summary reports
• Multi-rate export support up to 192 kHz
• Max samplerate : 384000 hz with 1048576 taps

Important note about the measurement function:
The integrated measurement function is available only in the packaged versions published under the GitHub Releases section. It is not included in the public source tree. The source repository contains the filter-generation side, while the measurement/acquisition workflow remains available through the released builds.

GitHub repository:
DecayCore

Latest builds are available from the Releases section.

Feedback, measurements, comparisons, and criticism are very welcome. I am especially interested in real-world results, difficult rooms, and comparisons against other FIR/room-correction workflows.

Old thread

 

[longts]

Hi there! I’m using a MacMini 2018 with Intel processors, Motu M4, Plus Kef R3 speakers, and a subwoofer SVS PB1000.
I’m really interested in learning how to measure and integrate the subwoofer with my main speakers. I read your manual, and I think it mentioned that the Decay Core can automatically measure and set the best crossover and delay for both the main speakers and the subwoofer. However, it also stated that the subwoofer measurement only runs in Windows.
I would greatly appreciate your advice on this matter.

 

[longts]

Maybe an error here @VilhoValittu

 

[VilhoValittu]

Maybe an error here @VilhoValittu View attachment 532299

Does macOS ARM-version work with Intel processor? Bass integration is still on development. Not available yet. It was my mistake to leave that info in manual.

 

[longts]

Does macOS ARM-version work with Intel processor? Bass integration is still on development. Not available yet. It was my mistake to leave that info in manual.

I have two Macs: 1 Mac mini Intel and 1 Mac Air M2. This photo from Mac Air M2.

 

[VilhoValittu]

I have two Macs: 1 Mac mini Intel and 1 Mac Air M2. This photo from Mac Air M2.

Ok, I will look why GitHub actions not include sounddevice at package.

 

[VilhoValittu]

Maybe an error here @VilhoValittu View attachment 532299

Please try version 1.0.3 and report how that works on Mac. v1.0.3

 

[longts]

Please try version 1.0.3 and report how that works on Mac. v1.0.3

OK, this one is okay. I can choose the sound device on Mac M2.

I still saw the subwoofer measurement here, but the FILE Tab only shows the measurement for a two-channel system. So at this time, this tool can only do the FIR for a 2.0 system only?

Thanks for your prompt response!

 

[VilhoValittu]

OK, this one is okay. I can choose the sound device on Mac M2.

I still saw the subwoofer measurement here, but the FILE Tab only shows the measurement for a two-channel system. So at this time, this tool can only do the FIR for a 2.0 system only?

Thanks for your prompt response!

You can still make own filters for subs. Just place files File fields and use automatic mode with subwoofer selection or adjust smart scan range 20-200hz with advanced mode.

 

[VilhoValittu]

Hi,

Version v1.0.5 out :

Automatic mode is significantly faster now with same results than before. Speed up on Linux machines almost 80%, Windows about 20%. No data for Mac speed up.

Minor UI change : added own scrollbar to Auto mode details.

Bypass filters are also now produced to filter package.

 

[rp86]

Thanks for your work on this great software. I'm just firing it up for the first time before trying some proper measurements and had some suggestions / requests for future improvements:

• Is it possible to have the analysis display the latency of the suggested filters? (Would need a note stating is in addition to the latency of the device used to run the filters)
• This is crossing into MSO territory, but I have 4 subs and would love the ability to use more than 2 in DecayCore

 

[Curvature]

Can you please post measurements when you get a chance?

 

[VilhoValittu]

Thanks for your work on this great software. I'm just firing it up for the first time before trying some proper measurements and had some suggestions / requests for future improvements:

• Is it possible to have the analysis display the latency of the suggested filters? (Would need a note stating is in addition to the latency of the device used to run the filters)
• This is crossing into MSO territory, but I have 4 subs and would love the ability to use more than 2 in DecayCore

Hi,

Latency is displayed at final plot, right top corner, "filter delay removed". With linear filter, info is always available at right top corner of the program.
(edit) Asymmetric : ~85ms , mixed ~85ms, minimum 1-2ms , linear 749ms, this numbers are with 65536 taps @ 44,1khz


About subs : Bass integration is still at beta. When I get enough feedback about it and get it working properly, 4 subs wouldn't be a problem.

You can use 4 or any number subs with DecayCore by measuring your speakers with subs. Program gives excelent results with that method.
But your subs timing (aka distance) must be correct. I presume that everyone know's how to do that, specially multisubs owners.

 

[VilhoValittu]

Can you please post measurements when you get a chance?

Before/after pics?

 

[VilhoValittu]

Can you please post measurements when you get a chance?

My own living room. No acoustic treatment, just couple sofa's and thick carpet.
(edit) Automatic mode with adaptive target.

Link :
before/after doc.

Source :
Pics

 

[Redacted]

I thought "DecayCore" was a really bad thrash-metal band!

It's probably just me, but a product with "Decay" in the name seems rotten
I'm sure it's excellent though, and 'free' is a price I like.

 

[john61ct]

Handling the timing, both phase issues and time to LP is really my primary need for DSP in the first place, especially FIR

But it's great there are different types if tools that each have their speciality.

It's just hard to keep track which does what

 

[VilhoValittu]

Handling the timing, both phase issues and time to LP is really my primary need for DSP in the first place, especially FIR

But it's great there are different types if tools that each have their speciality.

It's just hard to keep track which does what

That's actually one of the reasons I started developing DecayCore.

A lot of DSP tools are very good at frequency-response correction, but when you start looking at timing, phase behavior, excess phase, decay control, crossover integration, and subwoofer alignment, it becomes difficult to understand which tool is solving which problem.

My goal with DecayCore is to handle frequency response and timing behavior as a single optimization problem rather than treating them as completely separate tasks.

For bass especially, timing often matters just as much as the magnitude response. Two systems can measure similarly in frequency response but sound quite different if the phase relationship, group delay, or subwoofer integration is not well aligned.

Different tools still have different strengths, of course. REW is excellent for analysis, rePhase is extremely flexible for manual FIR design, and DecayCore focuses more on automated optimization of both frequency and time-domain behavior with minimal user intervention.

DSP can become a rabbit hole very quickly. The challenge is often not finding a tool that can do something, but understanding which tool is actually solving the specific problem you're hearing.

One area where I still think DecayCore needs significant work is subwoofer integration.

I'll be the first to admit that it is not nearly where I want it to be yet. In some rooms it produces excellent results, while in others the outcome can be far from optimal.

The biggest challenge is simply the huge variation between rooms, subwoofer placements, crossover implementations, and system topologies. At the moment I don't have enough real-world measurement data from a wide enough range of systems to reliably optimize for every scenario.

That's why subwoofer integration is currently one of the main development areas. Every new measurement set helps reveal cases where the assumptions are wrong and where the algorithms need to become more robust.

 

[VilhoValittu]

I thought "DecayCore" was a really bad thrash-metal band!

It's probably just me, but a product with "Decay" in the name seems rotten
I'm sure it's excellent though, and 'free' is a price I like.

You're not the first person to say that.

When I renamed the project, I was aiming for the acoustic meaning of decay rather than biological decay. The goal is actually to reduce unwanted low-frequency decay, not add more of it.

That said, "DecayCore" absolutely sounds like a thrash-metal band that released one album in 1992, disappeared for 30 years, and then reunited to play at a small festival in Finland.

At least the software is free, so unlike most audio upgrades, trying it won't damage your wallet even if it damages your ears.

 

[john61ct]

The biggest challenge is simply the huge variation between rooms, subwoofer placements, crossover implementations, and system topologies. At the moment I don't have enough real-world measurement data from a wide enough range of systems to reliably optimize for every scenario.

I am personally working on, planning for a multi-sub system that will be "somewhat portable", sometimes used outdoors.

So room correction needs to be kept a separate "layer" from the crossovers "speaker building" / bass management side.

Even maybe doing the timing / spatial tuning, smoothing for a wider LP area separately from magnitude, otherwise the driver layout, relative distance differences must be precisely maintained setting up every time.

but if truly automated, that is A Good Thing

then maybe doing it all together is OK if the workflow is sped up.