[Stephan Mire]

So I was thinking, if people spend so much money buying DACs and amps trying to modify the sound to achieve a sound that matches their preference, why not buy a PEQ or a graphic equaliser to achieve that function?

Wouldn't a MiniDSP solution offer a lot more flexibility in achieving a specific sound? Like if someone wanted a warmer sound, or a more detailed sound, just use the MiniDSP to mold the sound to your liking? Can the MiniDSP do this, and in real time?

It sounds very interesting to me.

 

[RayDunzl]

So I was thinking, if people spend so much money buying DACs and amps trying to modify the sound to achieve a sound that matches their preference, why not buy a PEQ or a graphic equaliser to achieve that function?

Wouldn't a MiniDSP solution offer a lot more flexibility in achieving a specific sound? Like if someone wanted a warmer sound, or a more detailed sound, just use the MiniDSP to mold the sound to your liking? Can the MiniDSP do this, and in real time?

It sounds very interesting to me.

Some people do.

DSP invokes a delay, so strictly speaking, it isn't real-time.

Take the the number of taps in use and the sampling frequency and find the offset.

The OpenDRC-DI (stereo, digital in and out) reclocks to 48khz, and has 6144 taps. If the filter is centered then there is a 3077 sample delay through the box, 64ms.

 

[PrimeTime]

Most modern AVRs that have Room EQ also have graphic/parametric equalization. These are useful to test and explore your own personal tastes and fr profiles. And less expensive than a targeting equalizer like, say, a Dirac.

Once your curiosity about such tools/toys is piqued, there's probably no way to scratch that itch other than to succumb and get one. That itch comprises a large part of what AVS is all about.

Having spent a goodly amount of my career in audio labs, I am now, thankfully, immune.

 

[jaddie]

Most modern AVRs that have Room EQ also have graphic/parametric equalization. These are useful to test and explore your own personal tastes and fr profiles. And less expensive than a targeting equalizer like, say, a Dirac.

While true, the room EQ and graphic EQ are not the same thing, and in most AVRs it's either one or the other. The graphic EQ in an AVR is something like a 9-band, which is a nice "tone control" but useless for room EQ. Parametric in an AVR would be rare. Room EQ is not tone control at all. I've always been confused as to why they are not permitted simultaneously, but it's probably DSP resources.

You can learn the effects of EQ of all kinds using something like Audacity, loading up your favorite tune, then go nuts with any EQ you like. All for free.

 

[PrimeTime]

Graphic EQs abound in media players. Good ol' Winamp has a 10-band EQ.

My wife's thirty-year-old Sharp combo (vinyl disc, Phillips cassette and FM radio/amp) that I am using in the basement has a five-band EQ. It's surprisingly useful. I find that when I'm watching TV late at night through it with the Volume turned way down, I can hear dialogue much better with the 4 kHz slider cranked up. All of that extra HF is absorbed by the room treatment and is inaudible elsewhere.

All AVRs have Bass and Treble shelving. Treble shelving can pretty much implement the Harman/B&K House Curve.

Once you get above entry level, things improve. Denon and Sony then offer 7-band graphic EQ.

The Yamaha RX-V677, one of the more popular AVRs under $400, offers manual 7-band PEQ as an alternative to their YPAO room EQ. That's a pretty powerful tool for just about anything.

 

[4DHD]

My experience with autoEQ in AVRs just suck out the life of the audio..
What passes for good movie soundtracks is no where good enough for music. imo.
The Dirac room correction, on the other hand, allows you to taylor the FR to your liking, w/o sucking out the life of the music.

 

[FAH]

So I was thinking, if people spend so much money buying DACs and amps trying to modify the sound to achieve a sound that matches their preference, why not buy a PEQ or a graphic equaliser to achieve that function?

Wouldn't a MiniDSP solution offer a lot more flexibility in achieving a specific sound? Like if someone wanted a warmer sound, or a more detailed sound, just use the MiniDSP to mold the sound to your liking? Can the MiniDSP do this, and in real time?

It sounds very interesting to me.

The idea is not to color the audio. Their looking for acoustic transparency. DSP and such all color sound, post conversion.

 

[4DHD]

The idea is not to color the audio. Their looking for acoustic transparency. DSP and such all color sound, post conversion.

Dirac colors nothing! It gives you a flat response, or a house curve if one is needed to tame an overly live room.

 

[FAH]

Dirac colors nothing! It gives you a flat response, or a house curve if one is needed to tame an overly live room.

Three things:


1. No such thing as a flat response, from a multi driver speaker system
2. So-called flat 1/3 measurements don't equal better sound.
3. You cannot eq reflections (live reverberant rooms)


All post filter implementations augment (change the original signals)...

 

[jaddie]

Three things:


1. No such thing as a flat response, from a multi driver speaker system

Beg to differ, but it also depends on what you call "flat". In the old analog tape days, "flat" was +/- 2dB. What do you call flat today? Many multi-driver speakers can be equalized flat, or more correctly, flat to a house curve. I can get flat within +/- .5dB of my target curve over most of the spectrum in most rooms. Is that "flat" enough?

2. So-called flat 1/3 measurements don't equal better sound.

Yup, and that's why we don't do that any more.

3. You cannot eq reflections (live reverberant rooms)

But you can equalize the result, and at least some auto EQ systems that work in the time domain actually do. So the reflection is there, but some if its effect is compensated for to varying degrees.

All post filter implementations augment (change the original signals)...

Sure, or they wouldn't be equalizers, they'd be flat amplifiers, and what would be the point of that?

 

[jaddie]

The idea is not to color the audio. Their looking for acoustic transparency.

Well, sort of. Speakers and rooms "color" audio quite a bit, so to achieve transparency you'd need to compensate for what they're doing. That's the definition of EQ. Some may wish to deliberately soften highs, or boost lows in response to a particular recordings failings.

DSP and such all color sound, post conversion.

A DSP-based equalizer alters the frequency response as a deliberate act to compensate for some deficiency in frequency response elsewhere in the system, usually speakers. The result, if done correctly, would be less "colored" sound.

All digital processing occurs post analog to digital conversion, and pre digital to analog conversion, in other words, in the digital domain...that being the D in DSP. In every audio signal chain there must be both A/D an D/A somewhere, even if the A/D is back in the original recording studio.

 

[4DHD]

Three things:


1. No such thing as a flat response, from a multi driver speaker system
2. So-called flat 1/3 measurements don't equal better sound.
3. You cannot eq reflections (live reverberant rooms)


All post filter implementations augment (change the original signals)...

Keep talking newbe...you might actually convince yourself...
but not anyone else.

If one reduces any peaks at the output, then those reflections of the same frequency are also reduced, thus become much less noticeable.

Much like lowering the volume in a live room. But if one turns the volume too high, in a live room, the reflections become unbearable.

 

[PrimeTime]

Much like lowering the volume in a live room. But if one turns the volume too high, in a live room, the reflections become unbearable.

And lo! -- the House Curve was born.

 

[JHAz]

Beg to differ, but it also depends on what you call "flat". In the old analog tape days, "flat" was +/- 2dB. What do you call flat today? Many multi-driver speakers can be equalized flat, or more correctly, flat to a house curve. I can get flat within +/- .5dB of my target curve over most of the spectrum in most rooms. Is that "flat" enough?
Yup, and that's why we don't do that any more.
But you can equalize the result, and at least some auto EQ systems that work in the time domain actually do. So the reflection is there, but some if its effect is compensated for to varying degrees.


Sure, or they wouldn't be equalizers, they'd be flat amplifiers, and what would be the point of that?

depends,sadly,on what you're measuring if you are measuring in-room FR at a specific location. If you don't time window your measurements consistent with at least average human hearing perception, what's flat to a measurement microphone may not be flat in your ear/brain system because much of the reflections that flatten it out arrive too late to be part of the main sound, and become, in our brains, simply part of the ambiance. It's why many speaker manufacturers look for a smooth rolloff in the highs as you get off axis horizontally . . . the reflections of the off axis output then become less confounding for real hearing experience, because they're quieter. While I'd always prefer to ring a room out with a good measurement mic, in the end what a person experiences in a specific listening location is simply different from what the mic "hears." And different from what the same person will hear two or ten feet away from the initial microphone location, less of a problem in live sound settings simply because all the reflections are way late for most everybody in the room., distances from the walls being significantly bigger . . . .

 

[PrimeTime]

. . . the reflections of the off axis output then become less confounding for real hearing experience, because they're quieter. . .

Taken to its logical conclusion, this implies that speakers with significant off-axis response, no matter how "smooth," will sound more "confounding" than speakers that beam (are more directional).

I for one believe this to be true in most cases.