Adding a 10-band PEQ to a DAC - What is holding manufacturers back?

[Music1969]

Is it too difficult?

Nope - Merging Anubis supports 12 bands PEQ per channel. 84 bands total.

Its been out for a few years now.

 

[Geert]

Can we agree on a deadline for the implementation of eq in DAC's, and from that moment on rate all DAC's without this feature as 'poor' ? It's 2023..., a Yamaha SPX90 effects processor could do parametric eq in 1985! It had less processing power than a modern door bell.

 

[BeerBear]

And here's what a SoundBlaster from the 1990s can do, with the onboard DSP chip and a custom driver:


AFAIK the KX Project driver was almost entirely coded by some Russian guy in his spare time, with some other guys who wrote some FX plugins. And it still works in newer Windows versions (here's a video). You can probably find one of those compatible soundcards for free at your local landfill, but they need a PCI slot to work.

This is just to say that you don't need expensive high tech DSP hardware or a multi-million dollar developer team for this stuff.

 

[Jeromeof]

A Raspberry PI Zero has easily enough processing power for a full DSP engine and costs about $10 - to manufacturers a similar ARM CPU would be < $1 - I hear the excuse that its the software that is the bottleneck and hardware companies are not good at software which is probably true but I think that is getting better (look at Wiim / LinkPlay) and some DSP engines are now open source and run on ARM already - though some of the best ones like camilladsp have GNU licences which would definitely put off some manufacturers.

 

[Tangband]

Sony could do a 3-band one in 1989.

What is it that seems to make a proper stereo DAC with just 7 more bands too much to ask for in 2022?

No software interface required, a simple screen and a couple of front panel controls would do. Even be preferred actually.

Is it too difficult? Too niche? Both? Something else?

Probably because the market ( customers ) are more interested in bluetooth and the latest flavour of the month dac chips.

A good eq that dont worsening the sound is not doable in the low pricerange ( WiiM pro is an example ) .

With that said - a dac should have a preamp with good volumeregulation, a 10 band eq with variable peq and an active dsp crossover .
Minidsp flex is almost there.

 

[pkane]

I had two of these in series using digital-only mode, connected by Toslink, for a total of a 20 point PEQ ten years ago:

https://www.behringer.com/product.html?modelCode=P0146
https://www.audiosciencereview.com/forum/index.php?threads/quick-measurements-of-behringer-ultracurve-deq2496.11382

Both units purchased used for a total of $200

 

[Rednaxela]

I had two of these in series using digital-only mode, connected by Toslink, for a total of a 20 point PEQ ten years ago:

https://www.behringer.com/product.html?modelCode=P0146
https://www.audiosciencereview.com/forum/index.php?threads/quick-measurements-of-behringer-ultracurve-deq2496.11382

Both units purchased used for a total of $200

They're brilliant. I'm using one myself too at the moment, digital in digital out. They're a great example of how to effectively offer a 10 band PEQ in a fully self-contained unit. And it's been around for 20 years, and cheap as chips.

Now if that could be added to the modern DACs we see so many of these days. With multiple inputs, remote control, volume control, bit nicer looking, and 2023 D/A conversion performance.





(Which is not entirely what RME offers, although it does get close.)

 

[mglobe]

Probably because the market ( customers ) are more interested in bluetooth and the latest flavour of the month dac chips.

A good eq that dont worsening the sound is not doable in the low pricerange ( WiiM pro is an example ) .

With that said - a dac should have a preamp with good volumeregulation, a 10 band eq with variable peq and an active dsp crossover .
Minidsp flex is almost there.

What do you feel the flex is missing? In terms of switching between multiple digital, and one analog input, it can act as a pre. There’s volume regulation that works well. I don’t remember if it is a full 10 bands for eq, but I think it is. You can do crossovers to integrate subs, and there is a Dirac option. Am I missing something?

 

[Rednaxela]

Nope - Merging Anubis supports 12 bands PEQ per channel. 84 bands total.

Its been out for a few years now.

Thanks. Though for £2,280.00, at least something must have been difficult about it.

Is it entirely self-contained?

 

[Music1969]

Thanks. Though for £2,280.00, at least something must have been difficult about it.

Is it entirely self-contained?

Check out the user manual. It is a Swiss Army Knife

https://www.merging.com/uploads/assets/Installers/Firmware/MERGING+ANUBIS%20User%20Manual.pdf

Ravenna network audio input too. Advanced stuff.

Big name in Pro Audio recording but no reason why can't be used at home. Good headphone amp built in

Look at what is top left of the desk

 

[Music1969]

Though for £2,280.00, at

And you have have Googled the wrong model

 

[Keerati]

Because the expertise required to build a DAC is very very different from the expertise to build a software capable of manipulating audio.

There are only three companies I'm aware of that make DACs with digital processing. That's RME, Qudelix and MiniDSP. It baffles me that Qudelix is a one person business, just imagine how much effort went into making the Qudelix 5K from scratch. RME is a multi-million dollar companty and MiniDSP was founded on that niche (DACs with EQ) so they make sense.

I would like to add Fiio to this list. Their K7 & K9 DAC/headphone amp series have parametric eq function (through Fiio mobile app). Fiio K7 bt cost ~$250.

I just recheck. Fiio PEQ works only on Bluetooth input.

 

[Keerati]

By the way, if you look at car amplifiers, DSP is quite common feature. DSP is a necessity for a proper car audio installation. For car audio, you don't usually have a proper speaker, you normally have woofers & tweeters that you need to tune by yourself.

I have this unit in my car that cost ~$180.

Nakamichi NDS4835AU

PRODUCT TECHNICAL DATA Product Data Dynamic Range(RCA Input) ≥100dB S/N(RCA Input) ≥90dB THD ≤0.05% Frequency Response 20Hz~20 KHz Input Impedance Low level:20KΩ;High level :51Ω Low Level Output Impedance 51Ω Signal Input Range RCA :5.5Vpp;High level:26Vpp Power DC 9V~16V REM Input High level...

nakamichicaraudio.com

It is a DAC/Amp with Bluetooth and DSP (8 Channels low level,4 CH×50W power). Yes, their DSP support PEQ function.

 

[AudioX3]

So very timely thread for me .... Several options I know of and love anyone's experience or adding here to share with others:

1) HW PEQ: Behringer Feedback Destroyer or the Ultra Curve. Not clear on their SNR as to if there is anything more harm then good, but they seem to fit the bill for a HW Digital PEQ (think 10 or 20 per channel) at reasonable price. I have an old version that was supposedly lower quality but the newer is supposed to be much better and think Amir reviewed if I recall. The Destroyer allows you to say like 20 or more settings so you can apply your favorites (1 - automatic, 2 - tuned to your room, 3 - EQed to taste (Jazzy, Tubie, Rock, Metal, Base heavy, whatever....thru choice N
2) Monolith Headphone DAC also recommended by Amir. It has some quirks and not state of art in all areas but if not a 10, then a 9 which is quite good enough for most. Has DIRAC, PEQ, Shelf EQs, high end DAC....an RMI for 1/3 the price. I have this as well and it fits the bill, though have thoughts of replacing it with the newer Begringer Destroyer for ultimate HW PEQ control independent of DAC.
3) Every AVR has Room Correction PEQ that can be user modified. Why I not just going that road not sure why other than fear that somehow AVRs are not high fidelity enough .. may be just fear or snobishness (really mostly FUD).
4) Then there is SW based PEQ / DSP. That I don't know much about and look to this year experiment with and learn. I am behind the times not switched yet to streaming or ripping (still a CD guy mainly with 1000s of them and digitizing to cloud or disk seems fabulous but the time to do so seems insurmountable or at least intimidating LOL) but setting up a HTPC with SW for DSP/PEQ/Room Correction seems to be the way to go. Any advise on that journey from folks at ASR in this thread or point me to dedicated ASR threads or Internet links on SW based PEQ/DSP/Room Correction, would love any advice!

 

[kemmler3D]

The technology is super basic and almost every consumer audio device already has the capability, a typical Bluetooth speaker has 10 or 20 bands of PEQ ready to go if you have the tools to configure it.

I've said it elsewhere but the reason that user-controlled PEQ is not common is manufacturers don't believe that enough users are interested and capable of using it properly. The juice is not worth the squeeze for them.

I say this with experience, I put a graphic EQ on one of our products. When units would come back for repair or return, often all the bands were maxed out, or something equally harmful / stupid like that. In fact, we think some people damaged the woofers by abusing the EQ.

It also requires a lot more than just enabling the feature. To make it worthwhile for the manufacturer you must:

• Create firmware that accepts user input for DSP configuration
• Build an app that can write to your hardware DSP
  • Which might require licensing or other interaction with the DSP module maker
  • Which might cost money
• Maintain the app indefinitely at indefinite cost
  • You can't just ship it once, because iOS / Android / Windows / Mac updates will eventually obsolete the app even if it works perfectly.
• Document the PEQ feature in the manual
  • Requires someone to successfully explain how PEQ works to a complete noob, in about 100 words or less. Any volunteers?
• Deal with the inevitable support inquiries and returns where people abuse the PEQ and conclude the speaker / DAC is broken
  • Yes, there will be some, even if the feature is buried in the "advanced" menu.
• Market the feature as a value-add, to justify all of the above
  • Even though the initial assumption is virtually nobody understands the feature to begin with. It ends up being a distraction, you take real estate away from things most people DO understand, to explain something most people DON'T understand. It's tough to market the feature except as an afterthought, which will make you wonder why you built it in the first place.

This is basically how manufacturers think about PEQ. It's a long detour from their normal product development process, and as far as they know that detour leads nowhere.

If WiiM starts eating everyone's lunch, we might see a lot more PEQs out there. Until they see demand for it, they will avoid this whole mess entirely.

 

[formdissolve]

The classic "smiley-face" EQ in winamp always made me chuckle when I saw it. Bass and treble maxed, with almost ALL midrange shelved to -10dB or lower. Sounded so bad, but everyone I knew did it.

Schiit has some nice hardware solutions, but some of them seem to affect the sound in worse ways. My buddy runs his entire balanced system through this $1500 ARS DJ EQ (has a nice bypass switch though): https://www.ars-direct.jp/product/204

I'd like to send to Amir for testing.. but don't have a spare 210K yen

 

[kemmler3D]

I'm going to write a counterpoint to my previous post, now that my butt is firmly planted in the consumer's chair, here's how I WISH audio manufacturers would look at it:

• If you're going to have an app anyway, might as well build in PEQ controls.
  • The market is going that way anyway.
  • Your incremental hardware cost is zero. So you have to update an app once a year. Suck it up.
• Before every device under the sun has user-facing PEQ, better to establish yourself as one of the brands that's good about it.
  • Don't be reactive, lead the market!
• Support inquiries and returns will be negligible if you hide the feature and put warnings on it.
  • Have some faith in your designer and copywriter. If you don't have those, hire at least one.
• The most knowledgeable consumers are the ones that ultimately influence most consumer audio purchases.
  • If you don't know this already, you probably don't do any customer research. So this one's on the house.
  • The most knowledgeable consumers are currently clamoring for PEQ.
  • If you want everyone's nerdy brother-in-law to recommend your products next holiday season, put PEQ in the app now.
• If your engineer complains they're not a UI designer, tell them to quit whining and hire one on upwork.
  • PEQ interfaces are old news. Nothing needs inventing here. Google it.
• ASR in particular is SEO gold. Don't worry about marketing the PEQ. Just get Amir or one of the regulars to review it and the rest will take care of itself.
• Bonus: Getting more users to install your app means you get a shiny new marketing channel that you own, plus you can collect everyone's info and sell it. Are you in business or are you a charity?

 

[Rednaxela]

Love the recent contributions, thank you all.

I understand the app creation and maintenance hurdles. Which is why I wrote this in the opening post:

No software interface required, a simple screen and a couple of front panel controls would do. Even be preferred actually.

RME shows how this can be done. Their app is 100% optional and has in fact been non-existent for years. Behringer is another example of PEQ functionality that does not require an app.

In short I would like to get out of the app-first mindset here. I strongly believe that going back to a firmware-first approach is the way to crack this one. A bit hard to get user friendly but once you have something workable and bug free, you might have a feature/USP that you can choose to add to any model in your lineup and then call it a day. To me it seems to be the most sustainable way for both the manufacturer and the consumer. Perhaps at a relative usability penalty, but speaking for myself I’d rather learn to operate a Behringer than worry if they’ll still update my app five years down the line.

 

[CedarX]

I strongly believe that going back to a firmware-first approach is the way to crack this one.

Is a web server / web app an intermediate option? No Android/iOS app to maintain and (a lot) more flexible than firmware + display…. AutoEQ is a great example of what can be done with a web interface.

 

[Rednaxela]

Is a web server / web app an intermediate option?

Cool suggestion!