My personal opinions about the current state of Digital Room Correction.

About 6 years ago, I earnestly invested quite a bit of time, in the promise of DRC - Digital Room Correction, bought a Dayton Audio microphone, which had a calibration file, which I downloaded from their web site, and went through as much of the documentation of the DRC tool, the DRCDesigner app, and eventually discovered REW (Room Equalisation Wizard).

I had also tried out the demo of Sonarworks - whatever name their product was called at the time, cos this seems to be evolving, I dismissed Sonarworks cos I did not like the end result, sounded a bit too Hi-Fi ish. Came to the conclusion that there was some custom curve that no matter what settings I used, cos there is a fair amount of customisation you can make to the automated correction within Sonarworks' speaker/room correction product, it seemed to have a slant, some smiley kind of curve, with a it of mid range emphasis, which I did not enjoy. In other words - Sonarworks never eq'd flat, in my opinion, so I was not prepared to invest further in it.

The best way to describe my concern with all kinds of digital room correction, is the lack of convergence. It reminds me of a time before science became well understood and we knew what elements were, like hydrogen, and oxygen, and nitrogen and silver, and helium, etc. Prior to proper science there was just speculation and hype and ignorance, and I can imagine that each alchemist had a personal interest in convincing you to ingest their own potions, and because there was so much secrecy, the details of their concoctions were not divulged. These were the days before intellectual property and copyright, so secrecy was the only way to hide your commercial goldmine of chemical superiority.

Then science was invented, and finally we had convergence, based on especially measurement, not personal opinions. Two people anywhere in the world could conduct the same chemical reaction and obtain the same result, and eventually we understood why elements behaved in a certain way, and things have improved. Above all so much of what is sometimes secret, for a season, not wanting to publish it too early, eventually becomes public knowledge when presented scientifically and verified by others.

I'll continue in the next post to avoid this becoming too long, in one post

My 1st attempts with DRCDesigner/REW(for the measurement) were not enjoyable to listen to and over time I realised I was applying too much correction, almost like micro correction, but as soon as I relaxed the strength of correction, the results were more listenable.

Then I moved on to REW and instead of using EQ correction computed by REW, I used REW to generate the correction filter as a wave file which I could import into a convolution tool such as the wonderful in built ReaVerb in Reaper, my DAW. Reaper has versions for Windows, Mac and Linux, for those who need to know. Relatively pleased with the results, but I also lost interest in audio/music for a while, cos the initial aim of listening to music, became futile cos so much music that is commercial is based on subjects, that are superficial. So I had done all this work, to hear music properly, but was pissed off with the mundane subjects that a lot of music revolved around. How many times does one need to hear about sex in popular music, before the topic becomes ear grating - oh not another one. So I lost interest for a while.

Just before I switched off, I had become acquainted with MitchCo the author of the book on how to use Acourate. I had hoped to release a similar book, hopefully inexpensive one, that was easily affordable, but never got round to it, cos one bother with the tools I was using was a lack of convergence - exactly how many measurements?, from what positions?, where exactly should you point the measurement microphone at and from where - Up, Forward, towards the center of the speakers, or towards the speakers, at the tweeter or at the point between the tweeter and the woofer, and if you have a sub, how do you integrate this. The DiY approach, which I had intended to write about using REW, had too many variables.

Continued below.

The 1st thing that put me on to the issue of convergence was - each time I tried out the Sonarworks product, every measurement was not the same. i.e whatever method Sonarworks was using is not consistent, i.e because of human intervention, it was impossible to repeat the measurement in exactly the same way twice.

Considering that the speaker had not moved position, why should a measurement system give me perceptively and objectively different results each time I measure. Hmmmm.. No convergence in this method.

So this caused me to pause any further efforts I had, in developing a deterministic method using openly and freely available DRC, because I had not arrived at convergence, so there was no point in publishing any method where I had not put in the effort to achieve convergence - Convergence being - hey - do exactly this and achieve a fantastic result, and I call that method OK1's DRC Method version 1. Simple things like, exactly what level of smoothness should be applied to the determination of the correction, what level of SPL in the room should I be aiming at, for this size of room, too many variables that I had not fully resolved.

Tools like Acourate have many claims, including endorsement by Bob Katz, no disrespect to him, but in the music industry, endorsements are not worth that much, take them with a pinch of salt, cos you are never informed when Bob Katz has moved on from a tool he previously endorsed, to a tool you never ever hear about cos that is a tool that gives him competitive advantage and he is never going to share all his secrets with you! Then there is Dirac, ARC by IK Multimedia, and more. There really was no way to do a proper demo of Acourate, which is a bit of a pity. Can't recall there is a demo of ARC, or any other commercial DRC tools, except Sonarworks.

TBC in next post

So there is a bit of a trend here, commercial tools which would have cost me about a thousand dollars if I wanted to try them out, and no refund. So I never bothered. I like to see some proof before investing in something, with money that cannot be refunded.

Then I heard of Trinnov, and that's expensive, way outside my budget. What I like about Trinnov, is you do not hear one or two testimonials from a few industry leaders, but this solution is implemented in at least a few hundred high end commercial venues, such as cinemas, and high end studios, so it gives me some confidence that there must be some truth to their method. The implementation is not black magic, measurement is completed within a few minutes, and within an hour or two, the whole process of measurement, correction and testing, can be completed. And it is scalable from a stereo pair of speakers, to Dolby Atmos setups with over 10 speakers. Costly yes. Out of my budget yes. If I had the money, that is what I would get and forget about it. Anything that Trinnov cannot achieve is probably not worth bothering about, cos like most things in life, we have the really important stuff, and anything else is minutiae that is not really worth the effort or expense, more like sweating the small stuff. So we have Trinnov as the gold standard.

Audio engineering is a terrible place to be passionate about, cos it has ample examples of snake oil. Its so easy to pull wool over many many people - my opinion of Sonarworks, from the issues I had with its method, posted above, or in other prior posts, who are not in a position to know any better. But I caveat that Sonarworks for headphones, holds more promise. In my opinion, cos it cuts out some of the variability in measurement. So I'm not dismissing Sonarworks outright, just qualifying my opinion of it. Placing it in the appropriate context.

Then in another thread, I am led to discover that the MitchCo, has gone from a book on DRC, to providing a service on DRC and now also has a plugin Convolver for DRC. A nice tidy business, and then the light bulb goes off. Nothing against a prospering business, but I wonder, who is verifying anything that MitchCo claims.

Acourate the tool he uses is so complex, a book had to be written to explain it.

And then another light bulb goes off - the best way to be an alchemist in today's world, is to provide a service, or write a book about topics, that have not been properly proven to work, only claimed, cos the method is secret or so complex, most people who use it now need an expert to be paid - to read their book, or paid to help you use Acourate (or REW) properly, and now here goes a nice money spinner HLConvolver, over 100 dollars.

I did an experiment a few years ago which is 100% repeatable, which proved that when properly setup, the mathematics of convolution is 100% exact between two accurately implemented convolution plugins. Easy - they completely cancel out, below audibility. It's easy to prove this from mathematics, and also test it, in real life on a computer, so in conclusion, all the different convolution tools that add mojo to convolution are actually distorting the signal, so I am now wondering, why do we need another convolution tool when we have perfectly good ones, for free. Everyone is entitled to their money spinner.

Software is the new alchemy - the place to hide secrets, that cannot be proven, so no one bothers to verify exactly what they do - cos we have this belief in software that its right. A few years ago, the almighty Dan Worrall demonstrated that Reaper's in-built compression plugin, had an oversampling option that simply did not work, and actually did nothing. People like me had been using this oversampling option, in good faith for many years, believing that it worked. It was such an embarrassment, to discover that I and thousands of other Reaper users had been deliberately deceived, cos I am 100% sure the Reaper developers knew exactly what they had done. And this was from Reaper developers who have a most honorable presentation of the laudable causes for why they develop Reaper, which has become the democratic DAW, for the people, yet they in full knowledge hoodwinked us deliberately - like some internal joke.

The lack of understanding of room acoustics and speaker engineering, by most lay folk has opened up this void by which everyone who has a spin, is welcome to rip our pockets a bit more, with absolutely no real proof, only reputation, especially when there is no demo version to try before you buy. The perfect sale. No refund. This applies to the products costing between about $200 and $500.

This is a bit of chicken change to many aspiring musicians and studio owners, it will not kill them to part with such monies, but its unconscionable to think of the businesses profiting on the snake oil of unproven promises, with zero refund. A perfect victim (or sorry - customer!). Capitalism. its legal. The product does not have to work, cos it's promises are not actually measurable !!.

At some point in medicine and pharmacy, the need for proper testing of drugs came into the picture, otherwise anyone would sell you anything, including cocaine in Coca Cola, and it was legitimate, or mercury or lead infused concoctions, and these were legal.

We need a similar level of scrutiny in our audio world, but we will never have it cos these products are not killing anyone, only nipping their pockets a bit - a few hundred quid here and there, of legitimate placebo based tools.

At the lower end of the DRC market, is the potential for legitimate snake oil. I would only add the obvious - buyer beware.

Most of this was about DRC for speakers and rooms.

But there is a similar battle in DRC for headphone, with products like Waves NX, Realphones, Dear VR, and the whole Binaural movement, leading on from the Ambisonics based immersive sound. The implementation of Ambisonics with real speakers is 100% valid, once you know the math, no issues there, but each binaural implementation for headphones, simply adds their own voodoo. No convergence, who is right? Absolutely no idea.

So for room simulation (which typically includes speaker correction or rather more correctly headphone frequency correction) another huge opportunity for snakes in oil skins. No proof, only claims. Ah, even Acustica is in on the bandwagon, everyone here to collect their share of the booty, on claims that cannot be validated, in any way aka Sienna.

No convergence.

But there is a ray of hope, and I have the utmost respect for Toneboosters.com

A proper researcher in acoustics, with the programming chops and mathematics know how, a real professional.

I highly recommend his Morphit headphone correction product. I have used it on my AKG 702's, and they really impressed me, further to some customisations in the tool, which I'd rather keep to myself. Oh come on spill the beans.

Well its not that difficult, Morphit, in addition to correcting frequency anomalies on your headphone model, furthermore, allows you to emulate another headphone, so play around with it and emulate as many of the headphone options it provides, until you find the one you think is most balanced, for your preferred personal definition of "flat".

Then phase 3, in Morphit, you can still customise this stock emulation, a bit further. In the low cost department, it is the only tool that I have found to make a significant improvement in audio results, results worth far more than its current price. Hope the surge in sales from this unsolicited recommendation does not jack up the price of Morphit which is currently only $39..

Some truth, finally at low cost, and you can demo it. But I'd go so far as to say, do not bother - just buy the damn thing. Some good lunches cost more than $39 per person. Pretty sure many good bottles of wine or something stronger, cost more than $39. Do I have proof that Morphit is the real deal - no. But so far its the only low cost converged solution, that goes straight to the point, with not too many variables to tweak, and it just works - one decent well developed and effective product, in this sea of snakes in dubious oils. A tool everyone can demo and make up their own minds about, without having to call an expert or read an experts book, or hire further consulting services, or pay for another tool to make it work properly.

Link below.

https://www.toneboosters.com/tb_morphit_v1.html

And while you are at it, find the crossfeed plugin tool - which uses the Meier algorithm, and place this before the audio is passed to Morphit.

I think you could place it after, ?Morphit, and it should not matter, cos both of these are linear, kind of like 2+3 = 3+2, does not matter which comes 1st.

I place Meier crossfeed - 64 bit version for Windows, developed by Case, before Morphit.

Find this tool - Meier crossfeed either 32 bit or 64 bit version - same result. If I posted a link here, it would be too easy, enjoy the thrill of the hunt for it.

Just feel the need to debunk the unfortunate profiteering in the lower cost end of the DRC market place, where so many invest in tools that achieve very little, in proven, verifiable results, and call out the disgrace that this end of the market has turned into. Everyone trying to make a quick buck from ignorant victims - or sorry customers. Well intended I can imagine, but you really wonder how honest the manufacturers are?

Time for the resurgence of honesty in our digital audio tools. Tools with proofs, should be the norm.

Can we see a photo of your room?

:>)
I was about to post.. perhaps along similar lines? But would like to go a lot farther. To get this out of the way.. I'm interested in the subject -for the 'tech aspects, generally learning. I've no interest in using DRC at this late point in my very modest studio/audio years, and intend to tread lightly, questions with respect. For the OP, didn't see anything in the post, or profile to mention if you are primarily in home audio, studio -both?
How well acoustically treated?
I have a general understanding -impression/belief re what DSP may or may not be able to do or fix. Tend to follow the notion some may expect more from it than the realities of it, and is best suited as 'final touch after jobs one- excellent speakers, and room treatment.
Thanks
Wayne

Well that is it, isn't it...

I just find it a 7 post diatribe with little to no substance at all. Anyone can play the expert, and toss around words like 'science', 'snake oil' and 'convergence' like they mean something but unless they can coherently bring it all together in a way that highlights a clear understanding of the purpose of these tools with appropriate measurements, the whole thing just ends up sounding like a poor, naive chap got upset because DRC didn't perform magic.

The science is pretty clear. There is a reason why we still have highly skilled and qualified acousticians. There is a reason why they take measurement after measurement after measurement, and use tools like spatial averaging rather than relying on single sweeps. There is a reason why you can hand 30 of them the same problem, and end up with 30 different solutions. While the science is clear, it is also very complex.

The people who often have the most problems with DRC are those that treat it like it is the solution to every problem. Obviously there is nothing scientific about that. There is nothing even slightly informed about that. DRC are tools. Nothing more, nothing less. Some are more feature rich than others. Some can solve more problems better than others. But that is also why they cost a lot more, or are intrinsically more time consuming to implement.

The final room measurements are all that matter. The antithesis to a 'lack of data' is not vitriol. In acoustics, a measurement indicating no positive change is often just as productive as a measurement showing a slam dunk. Let's start simple with a photo of OK1's room... because that may well be enough to put this all to bed.

Thank you for posting.

Basically, do your research when it comes to DRC.

Brüjo

Yes, agreed, do your research when it comes to DRC.

Unfortunately, DRC is a widely misunderstood tech, even misunderstood by the author of this post.

There are DRC tools with proofs, including the ability to independently verify the results.

To help people understand the state of the art of DRC, and what it can and can't do, along with verifying the results, I put this video together:

https://youtu.be/yfGAUvyvdNU

I'm running DL in my studio and find it a massive improvement over what I previously thought was decent sound. The image is so much better and every time I bypass the plugin, it just sounds like a phasey mess. Quite happy with it.

I run the plugin on a bus in Pro Tools just downstream from my print bus. My audio interface is a UA x8 and I monitor my live inputs during tracking via the Console software. Now that my ears have acclimated to the sound with Dirac on, I'm starting to wish I could enjoy the corrected sound while monitoring my inputs in a recording situation. I'd be able to dial in my eq and compression more effectively.

I obviously can't run the plugin in Console as its not UA, but I seem to remember someone mentioning some kind of 3rd party non DAW based software that could host it. But I can't seem to find this discussion anymore.

Can anyone think of a way I could have DL effect the sound BEFORE console, or am I asking the impossible?

You're the only one using the word "DL" in this thread. Yet you expect everyone to know which product it is?

Good point, brain fart. Dirac Live.

Thanks! I looked at Dirac Live yesterday actually. Have been aware of it for some time and I do think it looks interesting.

I also think the miniDSP’s DDRC-24 is very interesting. I need to make room and monitor decisions first though as I'm currently making do with Head Phones since I moved the studio.