CamillaDSP - How are you using it?

[oltos]

On 29/11/2023 at 10:59 PM, Keith_W said:

 

I should point out that Camilla can not be used to create convolution filters. Camilla itself is a convolver, i.e. it convolves the convolution filters with music signal for output. You need a separate program to create the filters, and I listed them in my previous post: REW (free), RePhase (free), OpenDRC (free), and the paid options - Acourate, Audiolense, and Dirac. If you are not planning to convert your speakers to an active, then there is also MSO (free). 

 

If you want to control 2 speakers with 4 subwoofers, you will need 6 DAC channels with amplifiers for each speaker/sub (I assume that you already have these amps). There is no way around this, no matter what solution you choose, you will need 6 DAC channels as a minimum. 

 

Alternative convolvers to Camilla: JRiver, Roon, Acourate Convolver, Hang Loose Convolver, EKIO, Audiolense Convolver, and EqualizerAPO (free). 

Room correction software: Acourate, Audiolense, Dirac, RePhase, OpenDRC. Or if using subs only, MSO (Multi-Sub Optimizer). 

 

Acourate is the most powerful and most versatile, and it is also the cheapest of the paid options. But it is difficult to use because everything is manual and requires interpretation. Frankly I think Acourate is overkill for what you want to do and I do not recommend it. 

 

Audiolense is probably the easiest to use and the fastest way to get results because of the high degree of automation. It does not offer the same degree of control or the flexibility of Acourate, but I do not think that you need it. 

 

Dirac, RePhase, and OpenDRC - I haven't used them and I do not have direct experience with any of them. I can't comment on ease of use and flexibility, but I understand that Dirac is easy to use. The downside of Dirac is that it outputs proprietary filters which must be hosted on Dirac's own convolver. I have a thing against proprietary anything, which is why I wrote it off without trying it. The other downside is that it is rather expensive. 

 

MSO only works for integrating multiple subs with speakers and can not do any of the other advanced DSP functions, but that is OK because it looks as if you are not planning to incorporate it. It is free and must be used with REW. You take measurements with REW, then import the measurements into MSO. MSO then outputs a filter which you host on a convolver of your choice (I listed some options above). MSO has an extensive manual, an extensive walkthrough, and a community of people happy to help (like our own @Snoopy8). I think this is probably the best solution for you. 

BTW, I think that you should consider a Merging Hapi IF you are planning to bypass the passive XO's on your speakers and convert them to active. Merging Hapi can be optioned up to 16 DAC channels. 

 

Thanks guys for this great feedback!

 

Small world, Keith!

https://www.audiosciencereview.com/forum/index.php?threads/room-correction-hardware-software-issues-options-please-advise.49900/#post-1793704

 

If I didn't say so there, I don't plan to actively cross my main speakers. Along with my humble Q & A throughout, they are being co-designed by https://josephcrowe.com/ and They will be three or two way using this

or this driver https://www.youtube.com/watch?v=0LMJ-vbxIxA&t=1s

above my Altec 416-8B midwoofers.

https://www.athosaudio.com/2021/01/01/tad-th-4001/ , the speakers will display sound similarly to Pierre's in post #15,266.

 

I'm not competent to explain why but while I've encountered a fair number who do actively cross horn speakers, I've never known of any at Lynn Olson's thread who have, or were satisfied with them after doing it. Again, I hardly have the knowledge base to second guess any of these guys about crossovers for drivers and horns which they everything about. Plus, I'll be paying very serious money on parts and labor for these passive ones.

 

So, you don't have any problems recommending Hapi for doing room correction for my passive speakers and my Rythmik subs?

 

Much earlier, a Camilla user at gearspace.com cautioned to first use any RC software and measurements to acoustically correct the room physically as much as possible, and then follow with creating corrective filters, not that I've yet any experience doing either.  

 

LewinskH01 also has multiple Rythmik subs and uses MSO. He also uses Acourate, but as you said it requires too much knowledge for building filters from scratch; use REW for instead, yes? And six channels of the Hapi?

 

Your August 12th post here is extremely informative. My pc

https://ark.intel.com/content/www/us/en/ark/products/212267/intel-xeon-w1390-processor-16m-cache-up-to-5-20-ghz.html

32GB ECC memory

https://www.gigabyte.com/Motherboard/W480M-VISION-W-rev-10/support#support-dl-driver

https://www.tomshardware.com/reviews/evga-nvidia-geforce-gtx_1660_super-sc-ultra

 

https://www.silverstonetek.com/en/product/info/computer-chassis/GD05/

With https://www.argusmonitor.com/ and 120mm fans; built by https://www.steigerdynamics.com/ So, it will likely run cool and quietly with most RC software.  

 

Btw, you mentioned at the ASR thread that you have the NADAC; almost twice above my MCH DAC budget. While the Hapi is clearly the way for me, since you said that I don't need to install any ADC card, it's too bad that you can't install two

of these 8 channel DAC cards https://www.merging.com/products/interfaces/specifications#d-a8-d-a8-p-option-card where the box could parallel them for better sound quality, something like what can be done with the NADAC.  

[Keith_W]

Yep that's me on ASR. 

 

If you have the ability to make an active digital crossover for your speakers, that really is the ultimate way to go. The downsides are extra cost for the amplifiers, DAC channels, software, and the learning curve. There are at least a dozen advantages for choosing to go down the digital active route. I strongly encourage you to consider this, but bear in mind that until you get the hang of it the result is likely to be worse than a well designed passive XO. If your speakers haven't been constructed yet, you could ask the manufacturer for the crossover configuration, so to start off with you could simply copy the configuration into your digital XO. This will more or less replicate the sound of the passive XO with all the usual advantages of active speakers. The rest of the optimization (individual driver correction, time alignment, room correction) can come later. Of course I am more than happy to help you along your journey. 

 

I myself have the system that I described to you on ASR and here. I have horns, which have been converted to active, with a digital XO and DSP. Link to my system in my signature. 

 

I have no problem recommending the Hapi. It sounds as if you may eventually need more than 8 channels (if you choose to go active), so the Hapi is the most versatile solution. Don't do what I did and buy a NADAC. NADAC's advantages is that it looks nicer, is easier to use than Merging's pro equipment (because most of the pro features have been removed), and it comes with standard XLR and RCA sockets. Even though it was designed 10 years ago, the SINAD is comparable to 2023 equipment. But then you get nearly all the features with a Hapi at less than half the cost, and the Hapi can do more than the NADAC. I have to warn you though, that a Hapi is pro audio equipment. It has a tonne of features that will likely confuse you. It does not "simply work" when you plug it in like any consumer DAC. There is an extensive list of configuration you need to make before you even get sound, not to mention you have to order a custom D25 cable for it with the correct connectors. 

 

That Camilla user in gearspace gave you good advice re: room treatment, but most people are limited with how much room treatment they can implement. My recommendation is do as much as you can within the limits of your situation, but DO NOT OVERDO IT. You will be surprised by how little room treatment you can get away with. 

 

If you want the easiest software to use, you should consider Audiolense. It will give you great results very quickly. If you want the most flexible, consider Acourate. If you don't want to pay any money, use REW + MSO. You can decide how much time you want to invest learning about audio and acoustics, how fanatical you are, how much money you want to pay, how much you value ease and convenience, etc. Nobody can make that decision for you, which is why there is a different solution for everybody. 

 

All this is off topic for this thread. I suggest you either start a new thread, or get a mod to move all these replies to a new thread, or we can continue our discussion in ASR. 

[Wirrunna]

Found this thread this morning.

I have been using CamillaDSP on rPi as per the previously mentioned https://www.audiosciencereview.com/forum/index.php?threads/rpi4-camilladsp-tutorial.29656/

and https://www.audiosciencereview.com/forum/index.php?threads/rpi4-camilladsp-tutorial.29656/page-63#post-1579025

 

Posted building a config for modified K-Horn tri-amped with phase flattening using REW and rePhase - https://www.diyaudio.com/community/threads/camilladsp-cross-platform-iir-and-fir-engine-for-crossovers-room-correction-etc.349818/page-189#post-7213813

which may be of interest if you are starting out.

[Keith_W]

That sounds like a fantastic project @Wirrunna! Would love to see you post a showcase thread with details of your system and DSP. 

[Wirrunna]

Thank you for the quick reply Keith, there are pics of the amp etc in this link

https://www.audiosciencereview.com/forum/index.php?threads/rpi4-camilladsp-tutorial.29656/page-63#post-1579025

 

and the DIY Audio link in my earlier post has the procedure I used, lots of screen grabs of REW measurements and how to flatten SPL and make the filters for CamillaDSP, screen grabs of using rePhase to flatten phase and get the resulting FIR filters into CamillaDSP.

 

The K-Horns are dinosaur speakers from the previous century with B&C DCX464 compression drivers and an Eliptrac horn which is a kit made from machined MDF and was popular in the Klipsch world until they stopped being made. The B&C required a 5msec delay to time align with the bass horn, easy using CamillaDSP, try that using a passive !

I used .pdf files rather than screen grabs in the post, the .pdf allows a reader to zoom in to read file print in the screen grabs.

Edited February 14, 2024 by Wirrunna
Added bit about delay.

[Keith_W]

@wirrunna I was hoping to see a pic of your horns I am not the only person on SNA who does DSP, and I am not even the only person on SNA with horns and DSP. But I AM the only person who regularly posts measurements, and talk about my struggles / stupid ideas / experiments (all found in my system thread). At times it feels as if I am talking to myself so it would be nice to have some company

[Grizaudio]

On 01/12/2023 at 8:35 PM, oltos said:

8 channel DAC cards https://www.merging.com/products/interfaces/specifications#d-a8-d-a8-p-option-card where the box could parallel them for better sound quality, something like what can be done with the NADAC.  

 

Alternative. 

https://rme-audio.de/m-32-pro.html

 

[oltos]

20 minutes ago, Grizaudio said:

 

Alternative. 

https://rme-audio.de/m-32-pro.html

 

How might that be better than the Hapi for sound quality and/or functionality?

[Keith_W]

2 hours ago, oltos said:

How might that be better than the Hapi for sound quality and/or functionality?

 

There will be bugger all difference in sound quality. Both are superb DAC's and the SINAD is vanishingly low on both. Whether there is a difference depends on whether you believe there is a difference. In this case, Merging are well known for their focus on DSD whereas RME are not. My RME is PCM only. 

 

As for functionality, there are a few differences: 

 

- The Hapi can be configured any way you like via drop in cards. The RME has a fixed hardware configuration. Not only is the Merging solution more flexible, it is also more future proof. 

- Hapi has Ravenna (vs. AVB for the RME). I have read that it is theoretically possible to convert all the AES67 protocols from one to another (i.e. Ravenna, AVB, and Dante) but you might lose some functionality. I don't want to deal with converting AES67 protocols, its a complication I would rather not live with. If you decide to live within the Ravenna or AVB ecosystem, you will find there are relatively few devices with AVB support vs. numerous devices for Ravenna. Note that not all of RME's own products support AVB, as opposed to Merging where 100% of their products support Ravenna. 

- Merging uses Pyramix, RME uses Totalmix. Both are a complete pain in the arse for those of us who are not audio professionals because the operation is so unintuitive. In addition, Merging has ANEMAN, a utility that connects all networked audio devices together. I have not used it myself, but I have read multiple reports on how difficult it is to get working. Even the pros find it difficult. 

 

Then there is the price. The RME is half the price of the Merging once fully configured. 

[Grizaudio]

2 hours ago, oltos said:

How might that be better than the Hapi for sound quality and/or functionality?

how many channels do you need?

[Grizaudio]

41 minutes ago, Keith_W said:

Ravenna (vs. AVB

Non issue for HiFi, unless your signal chain is AoIP.  

 

[oltos]

On 04/05/2024 at 10:12 PM, Keith_W said:

 

There will be bugger all difference in sound quality. Both are superb DAC's and the SINAD is vanishingly low on both. Whether there is a difference depends on whether you believe there is a difference. In this case, Merging are well known for their focus on DSD whereas RME are not. My RME is PCM only. 

 

As for functionality, there are a few differences: 

 

- The Hapi can be configured any way you like via drop in cards. The RME has a fixed hardware configuration. Not only is the Merging solution more flexible, it is also more future proof. 

 

- Hapi has Ravenna (vs. AVB for the RME). I have read that it is theoretically possible to convert all the AES67 protocols from one to another (i.e. Ravenna, AVB, and Dante) but you might lose some functionality. I don't want to deal with converting AES67 protocols, its a complication I would rather not live with. If you decide to live within the Ravenna or AVB ecosystem, you will find there are relatively few devices with AVB support vs. numerous devices for Ravenna. Note that not all of RME's own products support AVB, as opposed to Merging where 100% of their products support Ravenna. 

- Merging uses Pyramix, RME uses Totalmix. Both are a complete pain in the arse for those of us who are not audio professionals because the operation is so unintuitive. In addition, Merging has ANEMAN, a utility that connects all networked audio devices together. I have not used it myself, but I have read multiple reports on how difficult it is to get working. Even the pros find it difficult. 

 

Then there is the price. The RME is half the price of the Merging once fully configured. 

 

I just wish there were more MCH DAC choices with the features I need, minus the ones I'd surely rather not pay for. I'm not doing live recording or digitizing multiple analog sources, so I don't need ADCs-maybe just one ADC to switch between 6 or 8 DAC channels for recording live test signals from my speakers and subs for setting up room correction software.  

 

I also hate paying for that Ravenna and the rest of those arcane protocols and platforms. And while the cards in the Hapi might be upgradeable, if Merging offers no trade in policy good luck selling your old cards.

 

But the Hapi does have the highest output voltage in case of room correcting software induced gain loss, which is what always worries RC software neophytes like me about MCH DACs. https://www.sowter.co.uk/decibels.php

 

 

 

On 04/05/2024 at 10:35 PM, Grizaudio said:

how many channels do you need?

 

Edited May 12, 2024 by oltos

[rthorntn]

Hi,

 

I keep coming back to this thread and so I thought I would post.

 

Perpetual beginner here, I'm about to finally scratch the DIY speaker itch but first... I have a spare set of wee Diamond 9.0 bookshelf speakers that I removed the crossover from... to safely tinker with CamillaDSP before moving to a DIY speaker.

 

I'm an IT nerd, happy with RPi's and Linux.  Roon is my main source.  My only experience with DSP is with my MiniDSP SHD and UMIK-1.

 

I already have a RPi5 and the Hifiberry DAC8X arrived yesterday.  I have various amplifiers including some AliExpress PCBs: wee TDA8932 "blue" boards (arriving soon) and scary powerful (at least for tweeters) SAMP-200 dual TPA3255 boards.  I have quite a few PSUs kicking around, probably run the TPAs at 24V with an ALPS 20K volume pot initially at least.

 

I was going to start with a 2.2KHz crossover.  I ordered some 80UF film capacitor "coke cans" from Digikey to protect my tweeters.

 

So what's the path of least resistance here, I think CDSP on Pi5 with "USB Gadget" mode connected to my Roon endpoint is the way to go?

 

I'm unclear on how to replicate a "DIRAC-esk" experience, UMIK-1 into something free or low cost that guides me through a sound test and spits out something that CDSP can work with?

 

Thanks.

Richard

[Guest]

31 minutes ago, rthorntn said:

So what's the path of least resistance here, I think CDSP on Pi5 with "USB Gadget" mode connected to my Roon endpoint is the way to go?

Use Roon DSP engine and save the trouble of learning.

 

The easiest way with Pi5 is load moOde.

[rthorntn]

Thanks, I'm not interested in using the Roon DSP because Roon is my main source but not my only source and in the future I'll share those (SHD) inputs to CDSP over USB.

 

I should add that Audiolense and Acourate are out of my price range.  Does that only leave REW?

Edited September 17, 2024 by rthorntn

[Guest]

9 minutes ago, rthorntn said:

Thanks, I'm not interested in using the Roon DSP because Roon is my main source but not my only source and in the future I'll share those (SHD) inputs to CDSP over USB.

Note CamillaDSP is a DSP engine, similar to miniDSP software in the SHD,  its GUI not as friendly as miniDSP.  moOde adds some templates to make things easier.  But there is no equivalent of Dirac room correction.

 

So, you can add a cross over, apply a PEQ etc.  However, you have to use some other software to do the calculations, e.g. values for the PEQ,  convolution correction etc.

[rthorntn]

Thanks, Audiolense and Acourate are out of my price range.  Does that only leave REW?

 

With moOde, I'm not sure how that will fit in with Roon, does it support acting as a Roon endpoint, does it support "USB Gadget" mode, trying to figure out how to do the calculations with REW as I assume I'd have to have a mic input to REW and REW outputting audio through CDSP and out my DAC to my speaker drivers, or do I install REW on the RPi5? 

 

I'm not sure what my REW output audio device is, I was assuming it has to be the same DAC and channel arrangement as my "final" setup, I'm not even sure if I setup my crossover frequencies in CDSP prior to REW or if REW helps you figure out the optimal crossover frequencies?

 

I also see mention of REW + MSO is that just for multiple subs?

[Guest]

6 minutes ago, rthorntn said:

Thanks, Audiolense and Acourate are out of my price range.  Does that only leave REW?

I use Audiolense, expensive but produces excellent convolution room correction. 

 

REW can do only frequency room correction. There is also rePhase, but that is going down a deep rabbit hole (was there previously!).

 

9 minutes ago, rthorntn said:

With moOde, I'm not sure how that will fit in with Roon, does it support acting as a Roon endpoint, does it support "USB Gadget" mode, 

moOde has Roon end point which you can apply DSP.

 

11 minutes ago, rthorntn said:

trying to figure out how to do the calculations with REW as I assume I'd have to have a mic input to REW and REW outputting audio through CDSP and out my DAC to my speaker drivers, or do I install REW on the RPi5? 

 

I'm not sure what my REW output audio device is, I was assuming it has to be the same DAC and channel arrangement as my "final" setup, I'm not even sure if I setup my crossover frequencies in CDSP prior to REW or if REW helps you figure out the optimal crossover frequencies?

You appear to be new to REW, so it will be a steep learning curve.  Lots of help available but you have to spend a lot of time and effort. 

 

SHD, with Dirac, is much, much, much easier and may be better to stay there?

 

18 minutes ago, rthorntn said:

I also see mention of REW + MSO is that just for multiple subs?

MSO is for subs integration only.

----

So, what are you missing in your setup that makes you consider using Camilladsp?  

[Keith_W]

1 hour ago, rthorntn said:

Thanks, I'm not interested in using the Roon DSP because Roon is my main source but not my only source and in the future I'll share those (SHD) inputs to CDSP over USB.

 

I should add that Audiolense and Acourate are out of my price range.  Does that only leave REW?

 

There is also DRC-FIR which is free. Focus Fidelity is the new kid on the block, USD$249 so it is a little bit cheaper than Acourate or Audiolense. 

 

In general there are two styles of DSP software products - toolbox style and automated. Dirac, and to a lesser extent Audiolense, are more automated. The advantage is that you are less likely to screw it up, it is much faster, and is much easier to learn. The disadvantage is that you have less control, particularly with Dirac. In fact, Dirac is famous for screwing things up so badly  that there are multiple threads in other forums complaining about Dirac issues. IMO Audiolense should be package of choice for beginners. It works well, it is less opaque than Dirac, and you can fiddle with the controls as much as you like. It is ultimately less flexible than toolbox style DSP products but that is not an issue for most people. 

 

REW + Rephase and DRC-FIR is for advanced users. I guarantee a 100% chance that you will screw it up in your first 20 attempts. If you don't know what the terms "frequency dependent windowing", "minimum phase" or "excess phase" means, don't bother trying until you have a firm grasp of these concepts. DRC-FIR is particularly unfriendly. You write a script file telling DRC-FIR what to do, then run it. This is a product for engineers who are well versed in signal control theory, and not for casual hobbyists. 

 

There are no good free DSP tools for beginners, period. You have to acquire the knowledge, and then you hit another barrier - there are also no good DSP resources for beginners, either paid or free. 

[rthorntn]

Thanks, I might be getting a bit carried away.

 

I'll concentrate on getting CDSP setup as a crossover first.  So if moOde has CDSP built in and works with Roon (is there a webpage confirming moOde is Roon Ready as I cant find it?) maybe I start there?  I would like whatever I choose to support analog/digital inputs over USB.  I also can't figure out if moOde supports "USB gadget" as I would like CDSP to show up as a "USB sound card" for one of my computers, I'm quickly coming to the conclusion that moOde doesn't support USB gadget, I know CDSP does...

 

Any tools/tutorials I can use to help me with setting good "levels", as I understand it the same 4CH amplifier at the same volume on all channels connected to my stereo woofer and tweeter drivers is going to have wildly different levels, is there something I can use with my UMIK to figure out the optimal attenuation on the tweeters?

 

I'll maybe grab Audiolense next month (I assume I need the XO version?).  If I do need XO but Focus Fidelity will be just as good then should I just get that, I want the simplicity and quality results?

Edited September 17, 2024 by rthorntn

[Guest]

37 minutes ago, rthorntn said:

So if moOde has CDSP built in and works with Roon (is there a webpage confirming moOde is Roon Ready as I cant find it?) 

You can manually install a Roon Bridge on moOde.  No such thing as Roon Ready certification on any of the Linux distributions. If that is mandatory, then you need to stick with commercial offerings. 

 

39 minutes ago, rthorntn said:

I would like whatever I choose to support analog/digital inputs over USB.  I also can't figure out if moOde supports "USB gadget" as I would like CDSP to show up as a "USB sound card" for one of my computers, I'm quickly coming to the conclusion that moOde doesn't support USB gadget, I know CDSP does...

What is analog/digital inputs over USB???  USB is always digital !!

 

And sharing a USB sound card from Linux to other computers on network ?  Suspect this is for a technical person to solve  - can you?   As for CDSP supporting it ?

 

51 minutes ago, rthorntn said:

Any tools/tutorials I can use to help me with setting good "levels", as I understand it the same 4CH amplifier at the same volume on all channels connected to my stereo woofer and tweeter drivers is going to have wildly different levels, is there something I can use with my UMIK to figure out the optimal attenuation on the tweeters?

Given you do not know REW, suggest starting there.  This is one of many available

 

 

[Keith_W]

14 hours ago, rthorntn said:

Any tools/tutorials I can use to help me with setting good "levels", as I understand it the same 4CH amplifier at the same volume on all channels connected to my stereo woofer and tweeter drivers is going to have wildly different levels, is there something I can use with my UMIK to figure out the optimal attenuation on the tweeters?

 

This is how I set levels (note, I am using Acourate). 

 

1. Do nearfield driver measurements and linearise all the drivers. Convolve with crossover filters. 

2. Place the mic at the main listening position and time align all the drivers. 

3. Level match all the drivers by doing repeated sweeps and adjusting each amplifier until the corner frequencies line up.

4. Do room correction. 

[rthorntn]

Just tried to google "CamillaDSP vs HQPlayer" (ultimately it took me back to this thread) both are convolvers but what's HQP "providing" when you choose filters like sinc-M, AHM7EC5L, SDM?

 

Does HQP just have a bunch of included filters, that mess with the sound in ways that some people find pleasing?

 

Can CDSP replicate the HQP "try one after the other and pick one that sounds good" experience, perhaps by importing a bunch of community filters?

[Guest]

38 minutes ago, rthorntn said:

Just tried to google "CamillaDSP vs HQPlayer" (ultimately it took me back to this thread) both are convolvers but what's HQP "providing" when you choose filters like sinc-M, AHM7EC5L, SDM?

 

Does HQP just have a bunch of included filters, that mess with the sound in ways that some people find pleasing?

 

Can CDSP replicate the HQP "try one after the other and pick one that sounds good" experience, perhaps by importing a bunch of community filters?

HQP is an up sampler software package, with filters, convolver that requires a server with grunt. 

 

CDSP is a DSP engine, that you need to configure, with no templates. There is no such thing as a community filter because every room and setup is different.  CDSP requires knowledge and experience to use...

[Keith_W]

16 hours ago, rthorntn said:

Does HQP just have a bunch of included filters, that mess with the sound in ways that some people find pleasing?

 

You understand that you have to design the filters yourself, right?