Question for the DSP Experts - how often do you DSP?

[klettermann]

If you're just coming to DSP as anything less than an audio engineer, mathematician or physicist, the learning curve seems pretty steep. To be clear, I'm not talking about plug-and-play solutions, like Audessy. I refer to REW, MSO, phase alignment, multiple sub setup, yada yada. Beyond grasping the theory there's a helluva lot of experimentation and experience that comes into play. There's software to learn, and it's complicated. REW and MSO aren't what I'd call simple or intuitive.

So, on to the question: How often do you some kind of DSPing on your system(s)? I myself have been pretty well sucked in, finding myself doing more DSP fiddling than actually listening. No bueno. What's your experience? Thanks and cheers,

 

[Rednaxela]

DSP as a replacement term for lovemaking. Try it.

Sorry.

 

[Kiko1974]

In a Word: NEVER. And this is regardless if I'm playing an LP record, a CD or High Rez digital files.

 

[Astoneroad]

Totally agree, unfortunately. I understand what a powerful tool that DSP is... as I do with the Webb Space Telescope... however with both... I'm awed by the results and find the process/learning curve a deterrent to being able to enjoy music (or cosmology), without considering the gravitational effects of my room.

Before and after DSP

 

[Rednaxela]

In all seriousness - I DSP with a DEQ2496. Started anew a couple of times. New measurements, new analysis, new approaches, etc. After a while I felt I had tried everything and found myself arriving at the same point more and more. This made the urge to constantly tweak go away.

Next time I'll DSP again will be when I have new speakers.

 

[Pareto Pragmatic]

I am no expert, but...

I measure and re-EQ when something in my room changes.

However, in the past I have measured a lot at times, like when trying to find the best location for new speakers. That generally involved setting it up in a spot, eq, and then later on realizing there was a problem that I could not eq sufficiently, then moving a bit and doing it all again. A bit closer to the wall, maybe too much bass, but a little further away and not enough, and eventually just right. So how often I measure varies from time to time.

Here's how I avoid infinite fiddling. I have a goal, I want such and such level of distortion, a certain target curve, and a frequency range of +/-3db at 1/6 smoothing. When I hit that, I am done. Why done? Because I can't really get better than that in my room (limited to 10 PEQs at the moment). Guess how I know that I like that, and that I can stop there?

Lots and lots of experiments.

It does not take long to learn to use these things at a basic level, but which I mean "Sweep, eq, check". To be honest, I rarely go beyond that, though I do look at the waterfall once for any new set up. If you are feeling overwhelmed, just learn to sweep and then use EQ wizard to develop and check the results.

If you want to go beyond that, I suggest learning one measure at a time. That's a lot easier than trying to "learn REW/MSO". if you learn one measurement per day, or even week, you'll learn a huge amount before you realize it.

 

[ppataki]

I do it whenever I have a new speaker (these days literally every other week if you follow my DIY posts here...)
But other than that I do that whenever there is a new version of Dirac
Or just for fun if I need a change

 

[LGD_]

If you're just coming to DSP as anything less than an audio engineer, mathematician or physicist, the learning curve seems pretty steep. To be clear, I'm not talking about plug-and-play solutions, like Audessy. I refer to REW, MSO, phase alignment, multiple sub setup, yada yada. Beyond grasping the theory there's a helluva lot of experimentation and experience that comes into play. There's software to learn, and it's complicated. REW and MSO aren't what I'd call simple or intuitive.

So, on to the question: How often do you some kind of DSPing on your system(s)? I myself have been pretty well sucked in, finding myself doing more DSP fiddling than actually listening. No bueno. What's your experience? Thanks and cheers,

I have measured my 2ch system twice, using REW and a calibrated microphone. From that, made FIR convolution files that I use in my player (foobar2000).
I use those correction files everytime I use foobar, which is my one-and-only player.

Results are fantasic - my speakers sound better than I ever thought possible.
In fact, I would recommend anyone contemplating new speakers to do room correction first - you will be amazed.
And it's not hard, if you can follow directions there are videos and tutorials available to walk you through it.

 

[ZolaIII]

I do my speakers once or twice a year. After spring cleaning and renovation and sometime on late autumn. When I change something from equipment and so on. I do it for friends and random strangers hire. When I am in a good mood I experiment and learn. I do it far more often for headphones.

 

[pablolie]

I just used in-room measurements to position speakers and especially the sub (I tried two but went back to a single one). I just don't want to introduce more complication than absolutely necessary and am more than happy enough with the results. Personally, I have found the effect is most notable in bass management, other improvements are far more subtle.

 

[Kiko1974]

I have several problems with DSP:
1- If I'm playing an LP record, I want a full Analogue chain. I don't see the point of playing an analogue medium and digitize It for playing, unless for a certain album, the LP record IS the best sounding available version.
2- DSD can't be processed natively on the digital domain, It needs to be converted to PCM first. I want my SACD's and DSD files to be played natively, with no intermediate PCM convertion.
3- Many A/V Receivers that feature DSP/Room Correction, are unable to play High Rez PCM, either at 96/24 or 192/24 , when any kind of DSP/Room Correction is on.
Somewhere well hidden in the User's Manual states that if you want to play High Res PCM at its full native resolution, change to the "Direct" or "Puré Direct" (depending on the brand the name changes) mode.

 

[gberchin]

Every single day. I work the math, create custom algorithms, program the computer to implement them, measure the response, and actually listen to the results.

 

[ZolaIII]

I several problems with DSP:
1- If I'm playing an LP record, I want a full Analogue chain. I don't see the point of playing an analogue medium and digitize It for playing, unless for a certain album, the LP record IS the best sounding available version.
2- DSD can't be processed natively on the digital domain, It needs to be converted to PCM first. I want my SACD's and DSD files to be played natively, with no intermediate PCM convertion.
3- Many A/V Receivers that feature DSD/Room Correction, are unable to play High Rez PCM, either at 96/24 or 192/24 , when any kind of DSP/Room Correction is on.
Somewhere well hidden in the User's Manual states that if you want to play High Res PCM at its full native resolution, change to the "Direct" or "Puré Direct" (depending on the brand the name changes) mode.

It's not DSP, it's you!
1 chain is analog and broadband at best, depending on how old and rare recording is and depending of state of vinil you will want to restore and digitalise it.
2 DSD is single bit stucked in 44100 or 48000 X64\16. Can be bit perfect converted to 16 bit PCM.
3 many of them are expensive ruins and even ones that are not are still rather limited regarding processing power. 24 bit 96000 is plenty for FIR and such, there's really no need for more regarding playing at least. Internally DSP processing moved to FP and noise is a no issue whatsoever (as SINAD is beyond).
If you like crackling and other noise that's fine with me. The main purpose of DSP is to correct drivers response and room impacts to a listener not to correct or impact recording, choose your sources carefully.

 

[Kiko1974]

It's not DSP, it's you!
1 chain is analog and broadband at best, depending on how old and rare recording is and depending of state of vinil you will want to restore and digitalise it.
2 DSD is single bit stucked in 44100 or 48000 X64\16. Can be bit perfect converted to 16 bit PCM.
3 many of them are expensive ruins and even ones that are not are still rather limited regarding processing power. 24 bit 96000 is plenty for FIR and such, there's really no need for more regarding playing at least. Internally DSP processing moved to FP and noise is a no issue whatsoever (as SINAD is beyond).
If you like crackling and other noise that's fine with me. The main purpose of DSP is to correct drivers response and room impacts to a listener not to correct or impact recording, choose your sources carefully.

I do choose my sources carefully within my limited budget.
I don't agree that DSD is 44.1-48/16 resolution. It's more like a bit more than 44.1 kHz sampling frequency, DSD64 can transparently reproduce Up to 30/35000 Hz, and when well mastered, DSD64 has a resolution of 20 bits or more.

 

[ZolaIII]

I do choose my sources carefully within my limited budget.
I don't agree that DSD is 44.1-48/16 resolution. It's more like a bit more than 44.1 kHz sampling frequency, DSD64 can transparently reproduce Up to 30/35000 Hz, and when well mastered, DSD64 has a resolution of 20 bits or more.

No it can't, DSD128 can reach 19 bits and all beyond what you hear (20 KHz) is used for noise shaping and with a bandwidth limit set to 45 to 60 KHz. Come on buddy use the basic math. You can get as much SINAD from 16 bit PCM with noise dithering.
We are talking about FP double precision 64 bit processing precision and then gluing it back to 48~96 KHz 24 bit integer and that's DSP to DAC.
PDM is used when you can't afford bus and synchronisation and will continue to be dominant in limited applications as microphones for all kind of mobile devices (that's far more than all other combined).

 

[Keith_W]

I DSP often. Once you have purchased all the hardware and software, the only possible improvements come from:

1. Purchasing a new house or renovating your living room ... $1-2 million or so.
2. Changing speakers. $10-30k. Even if you DIY it's still $10k ... DIY'ers always forget to factor in purchase cost of carpentry tools!
3. DSP

Of the three, DSP costs me nothing. I can improve my DSP by upgrading how I take measurements and working out new DSP procedures. That's why I love it.

 

[Waxx]

I set my dsp and forget, untill i changee something on the setup or in the room (major change of layout, like moving or adding furniture). In reality that is not that often, maybe 1-3 times a year.

But i'm in the process of changing my diy speakers (i used them almost 10 years in various iterations, first passive crossover, now dsp for some years), so i probally will mess with the dsp a lot somewhere in the future when the new setup is ready. I'm still in the designing phase altough. i'm keeping the old speakers for another space (probally with a passive crossover again) But i use diy speakers mostly in my house, that involves a lot more thinkering to get everything right than predesigned commercial speakers.

I started working with dsp almost 30 years ago, as sound engineer for high power dj systems. So I know the danger of the endless tweaking, which does often make your sound worse (bcause you loose perspective). Measuring helps you to avoid that a bit, but only to a certain level. Sometimes you just need to leave it like it is for a while untill you find out what is wrong, or that nothing is actually wrong. These days i can see on measurements that it will sound good to my taste, but that requires some experience. But it makes the calibration of dsp a lot easier if you can connect a graph to a sound in your mind, without hearing it. Off course the final judgement stays listening, but for someone who's trained in this, a graph (or at least a series) can tell you 99% of the equation, and it makes the job easier.

 

[Purité Audio]

I have several problems with DSP:
1- If I'm playing an LP record, I want a full Analogue chain. I don't see the point of playing an analogue medium and digitize It for playing, unless for a certain album, the LP record IS the best sounding available version.
2- DSD can't be processed natively on the digital domain, It needs to be converted to PCM first. I want my SACD's and DSD files to be played natively, with no intermediate PCM convertion.
3- Many A/V Receivers that feature DSP/Room Correction, are unable to play High Rez PCM, either at 96/24 or 192/24 , when any kind of DSP/Room Correction is on.
Somewhere well hidden in the User's Manual states that if you want to play High Res PCM at its full native resolution, change to the "Direct" or "Puré Direct" (depending on the brand the name changes) mode.

LP is the ‘best sounding’,is it certainly not in my experience.
Regarding EQ spend some time adjusting for lower bass room issues and then forget all about it.

Keith

 

[Sokel]

Set and forget until the next big change.
I have gone down the rabbit hole for a long time so to know that improvements are nice but do not bend the laws of physics.

So, next itch - next speakers, that's the way it goes for me, after a treated room it's the single most changing game element in the chain.

 

[Kiko1974]

No it can't, DSD128 can reach 19 bits and all beyond what you hear (20 KHz) is used for noise shaping and with a bandwidth limit set to 45 to 60 KHz. Come on buddy use the basic math. You can get as much SINAD from 16 bit PCM with noise dithering.
We are talking about FP double precision 64 bit processing precision and then gluing it back to 48~96 KHz 24 bit integer and that's DSP to DAC.
PDM is used when you can't afford bus and synchronisation and will continue to be dominant in limited applications as microphones for all kind of mobile devices (that's far more than all other combined).

I like how well mastered DSD64/SACD's sound.