The DSP Rolling & How-To Thread
post-14275651
Post #136 of 155

castleofargh

Sound Science Forum Moderator
Joined
Jul 2, 2011
Messages
8,968
Reaction score
4,298
Joined
Jul 2, 2011
Posts
8,968
Likes
4,298
To Whomever Would Like to Assist:

In advance, I would like to extend my sincere appreciation to those who are willing to assist, and I apologize for my loquaciousness...
...and I certainly hope this is the most appropriate forum in which to post this inquiry...

An OLD audiophile here, (periodically seeking ever improving sound reproduction for 30+ years...
...), so, though my BASIC understanding of DSP is "OK", the plethora of available MiniDSP products here are a bit daunting if not confusing...

Essentially, I possess one connected audio system, with two output legs/chains, or, imagine a wishbone, or a river, in which a single river source splits into two rivers, (Both the signal sources and the D/A converter being the single river before it splits into two rivers, the digital signals of which are converted by a D/A converter, the analog signals then continuing their journeys downstream on the two separate rivers... :) ...

Signal Sources: (1) "Garden variety" OPPO digital disc player, output via standard digital SPDIF, terminated with BNC connector at D/A end,
(2) Computer USB to to Schiit Audio EITR, outputing to D/A connector via standard digital SPDIF interconnect...

D/A Converter: Schiit Audio Yggdrasil...which outputs the converted/analogue signal via balanced interconnects to my Primary/System (1),
as well as outputs the converted/analogue signal via unbalanced/single-ended interconnects to my secondary/System (2)

System (1):
Pre-amplifier: Schiit Audio Freya, (fully balanced and receiving the balanced signal output from the D/A converter--->balanced output to Marchand Electronics XM44-2 crossover, (Linkwitz-Riley, 24 dB/Oct. @ 60 Hz), which drives an Aragon 4004 amp., via single-ended outputs/inputs, which drives a pair of Lipinski Sound L-707 monitors. the active crossover also driving a pair of active, SVS subwoofers via single-ended outputs/inputs.

System (2):
Integrated amp.: Schiit Audio Ragnarok:, (receiving single-ended output from the D/A converter)---> the amplifier section driving a pair of Audience Audio Clairaudient 1+1 V2+ monitors, and via single-ended interconnects, feeding a pair of JL Audio Dominion D-108 subwoofers (possessing Linkwitz-Riley 24/dB/Oct. low-pass filters only, crossover set at approx. 60 Hz.).

I am very seriously considering upgrading my system(s) with a DSP device, and it appears that a MiniDSP box might be my best solution...and that the DDRC-22D might be the best of those solutions for me...Unless one can a case for the new MiniDSP SHD or SHD "Studio" units.

1. I VERY much like the sonic qualities of my Schiit Audio Yggdrasil D/A converter, and wish to do NOTHING that would adversely affect its qualities. THEREFORE might I assume that I must implement a MiniDSP unit upstream of my D/A converter...which then would suggest the DDRC-22D...(UNLESS I can be convinced that a MiniDSP unit installed downstream of my D/A converter would NOT adversely affect the sound of the system...I understand that the input/output design of the "22D" would allow me to channel both the digital disc player digital output AS WELL AS my computer digital output into/through the "22D"...

2. IDEALLY, I would love to possess a single DSP unit which could correct/compensate for frequency level and timing output issues in both my primary AND secondary stereo systems simultaneously, or at least possess the ability measure/store compensation algorithms for both systems and then simply select/enact the appropriate compensation algorithm for the stereo system to which I am listening at a given time.

However, if this is NOT possible, and I must select only one stereo system for which the DSP could measure/correct/compensate, is there a distinct advantage to selecting a stereo system with an active crossover, or would the MiniDSP unit do an equally good job compensating for both frequency and timing issues with either stereo system? (the secondary system currently lacks a dedicated/active x-over between the mini-monitors and the subwoofers, such that there is more overlap between the upward frequency output of the subs and downward freq. output of the monitors...Subwoofer drivers and forward monitor drivers are on the same plane relative to the listener, but subwoofers possess inherent15 millisecond group delay, and monitors are crossover-less design)...Can the "22D" compensate for the inherent 15 millisecond group delay in the subwoofers?...

2. Does MiniDSP manufacture an active crossover that is, (in terms of sonic qualities), at least the equivalent of the Marchand active x-over, which I could set at 60Hz./24 dB./Oct. ...?...

3. Are there any alternative options utilizing the MiniDSP units that would meet my needs, that would DO NO HARM to my audio chain/reproduction...(the 1+1 mini monitors are a crossover-less design, such that they generate well nigh NO phase issues within their output range, and I do not want to compromise THOSE inherent qualities..., nor do I want to compromise the sonic qualities possessed by the Yggdrasil D/A converter...)

I genuinely appreciate any and all input/suggestions/recommendations, and avidly look forward to your responses!

THANK YOU ALL.

Sincerely,
T.A. Kogstrom
the DDRC-22D seems like a typical design so I doubt it handles 2 different outputs. usually people have one or more sources to feed the same speakers, so that's the situation usually discussed. a mail to miniDSP might be more helpful for that question.
about where to put it, well it's before the DAC in your case. you input a digital signal and output still a digital signal that has been processed by convolution. some gears can do more, but the basic concept is a digital process.
about your concerns of altering the sound in ways you don't want, as bigshot said, the role of such a device is to alter the signal. it's based on the assumption that speakers in a specific room will not give a flat response at the listening position. and that's almost always true. you're concerned about small details while it will most likely correct much more significant stuff. how you'll subjectively feel about the change is another story.
there exist many auto calibration solutions, from getting a mic(or even using a cellphone...), and using the REW and some free convolution option on a PC, to very advanced and expensive auto-calibrating stuff. it's always hard to know what we need/want.
 
     Share This Post       
post-14315699
Post #137 of 155

ironmine

500+ Head-Fier
Joined
Jun 13, 2008
Messages
636
Reaction score
55
Location
Far East of Russia
Joined
Jun 13, 2008
Location
Far East of Russia
Posts
636
Likes
55
Signal Sources: (1) "Garden variety" OPPO digital disc player, output via standard digital SPDIF, terminated with BNC connector at D/A end,
(2) Computer USB to to Schiit Audio EITR, outputing to D/A connector via standard digital SPDIF interconnect...
Why do you need 2 digital sources? Get rid of your OPPO disc player and keep the DAC.

As for your existing CDs, you can either copy them to your hard drive in the computer, or (it's even easier) just find on the Internet and download the images of your CDs.

Are there any alternative options utilizing the MiniDSP units that would meet my needs
There are lots of such solutions. I recommend the MathAudio Room EQ VST. It can be used either as a Foobar2000 component or as a VST plugin. It's free, simple to use and it offers the same or even better quality processing as more expensive solutions.
 
Last edited:
     Share This Post       
post-14338740
Post #138 of 155

prescient

Head-Fier
Joined
Oct 31, 2006
Messages
73
Reaction score
41
Joined
Oct 31, 2006
Posts
73
Likes
41
A few thoughts on the miniDSP boxes as people seem to have interesting views on these. I have the ddrc-88a it would be difficult to have a software solution that works as well and as simply as this box if you have a large number of input devices. Many people are using these boxes in their home theaters or listening rooms and in those use cases I am of the opinion that it is invaluable.

Typical home theater or listening room use case is to put the box between your processor and amp so that when you change inputs your signal gets passed through and automatically corrected for your room/speakers. The only way to do this on a pc would be if you had some sort of 8/8 or 16/16 audio card with a pc running 24/7. The other nice thing about these boxes is that you have multiple presets + bass management allowing you to set up profiles for different scenarios. Finally, if you have a serious home theater setup (e.g. 7.2.4), which I don't, you can run multiple minidsp units together. If you check AVS forums people have done comparisons and it seems to out perform just about everything out there maybe excluding some very high end home theater solutions (Trinnov altitude 16 will run you about ~$16K or more).

Being able to store multiple profiles for movies, late night (reduced bass), music, etc is great. This way if I decide i want to watch a movie or play video games at 2AM I don't wake up the house with explosions. I could use a dedicated PC, which I have, but once you factor in the setup costs and complexity versus the minidsp solution it doesn't seem to make much sense and Dirac works quite well. The other nice thing about it is that you can save your curves in a spreadsheet and just feed them into the Dirac software. This allows you to say move a speaker, retake the measurements, recalculate the filters and away you go. It is fairly trivial and is important as my wife and/or housekeepers like to move stuff when cleaning.

For $1K I'm not sure what else you would buy with better functionality and ease of use. For headphones it probably isn't the best solution, but it is great when you use it for what it is built for. If you are listening from a Mac I'd second audio hijack as a polished user friendly solution. Minidsp also has a headphone amp, dac, dsp combo. I'm not sure how well it works but it looks intriguing.
 
Last edited:
     Share This Post       
post-14339286
Post #139 of 155

ironmine

500+ Head-Fier
Joined
Jun 13, 2008
Messages
636
Reaction score
55
Location
Far East of Russia
Joined
Jun 13, 2008
Location
Far East of Russia
Posts
636
Likes
55
For those people who don't like digitally clipped & dynamically compressed music, I recommend the Thimeo Stereotool. I use it as a VST plugin and I am able to achieve great results with its DeClipper and Natural Dynamics functions.
 
Last edited:
     Share This Post       
post-14581397
Post #140 of 155

Synergist969

100+ Head-Fier
Joined
Jun 17, 2009
Messages
274
Reaction score
33
Location
Central Florida
Joined
Jun 17, 2009
Location
Central Florida
Posts
274
Likes
33
Dear bigshot:

Did not realize you had responded so quickly...my thread response alert function must not have been working...

Yes, I understand that by its very definition that a DSP will indeed alter the sound of the system... :) ...and YES, you are correct with regard to my system being a dedicated Stereo, (2 channel only system)...and I DO wish to alter the frequency curve/output of my system, as well as timing-wise, match the impulse response of my subwoofers to that of my mini-monitors...(the mini-monitors are a full-range driver, crossover-less design, such that they have NO inherent group delay for all intents and purposes, and my subwoofers have an approx. 18 millisecond group delay...so that I would like to delay my mini-monitors' impulse response by at least 18 milliseconds, (if I physically align the drivers of my subwoofers and mini-montitors in the same physical plane)...and possibly delay the mini-monitors even more if I must place the subwoofers further behind that physical plane...(I realize that I can fully/infinitely adjust the phase of the output of my subwoofers via the subwoofer controls, such that the subs and monitors can be properly "in phase", HOWEVER, that does not correct for the group delay issue)...

Ultimately, I wish to continue to enjoy the clean, low distortion sound of my stereo system, HOWEVER, I ALSO want to create a system whereby the subwoofers are perfectly time-aligned with my mini-monitors, and that the properly integrated subwoofer-mini-monitor system would also exhibit a very well balanced frequency curve, from the low end of the subwoofer output...(30 Hz.,), to the upper end of the mini-monitor output...(20KHz....)...

I have read via several other reviews that the Mini DSP units are NOT the last word in noise and distortion levels, (though I have read that the SHD series are an improvement over the earlier models/lines), and I simply do not want to give up the relative high fidelity I now enjoy by utilizing in my stereo system a device that might compromise those qualities...

Peace,
T. A. Kogstrom
 
     Share This Post       
post-14581419
Post #141 of 155

bigshot

Headphoneus Supremus
Joined
Nov 16, 2004
Messages
21,252
Reaction score
3,397
Location
Hollywood USA
Joined
Nov 16, 2004
Location
Hollywood USA
Posts
21,252
Likes
3,397
Website
www.facebook.com
Ultimately, I wish to continue to enjoy the clean, low distortion sound of my stereo system, HOWEVER, I ALSO want to create a system whereby the subwoofers are perfectly time-aligned with my mini-monitors, and that the properly integrated subwoofer-mini-monitor system would also exhibit a very well balanced frequency curve, from the low end of the subwoofer output...(30 Hz.,), to the upper end of the mini-monitor output...(20KHz....).
Maybe I don't understand this particular application. It's interesting. You're using bookshelf speakers (probably something like my KEF speakers) with a sub. I'm assuming they are near to each other and the room isn't affecting your mains different than the sub, and I assume you use bass management. I can totally see how an equalizer would help make your bass crossover balanced. That would make a clean handoff at 80Hz or 100Hz or wherever you have it set. If the crossover is working correctly, there shouldn't be any overlap, so there shouldn't be any phase problem. There would just be a tiny sliver of mismatch right at 80Hz. I can't see how that would be audible. Setting the phase dial on the back of the sub will fine tune that kind of thing. I don't see how the noise in the Mini DSP would be audible. The mini DSP is supposed to have an excellent equalizer, so it would be great for that. When you have speakers spread out all around a room at different distances, I can see how time alignment helps. But my guess is that the time alignment wouldn't add up to anything beyond just a theoretical improvement with your setup. It's not my bailiwick, so I might be wrong though. Perhaps you have some sort of room considerations that make it impossible for you to put the sub near to the mains. I can see how you might run into problems if that is the case.
 
Last edited:
     Share This Post       
post-14581487
Post #142 of 155

Synergist969

100+ Head-Fier
Joined
Jun 17, 2009
Messages
274
Reaction score
33
Location
Central Florida
Joined
Jun 17, 2009
Location
Central Florida
Posts
274
Likes
33
Dear bigshot:

Thank you for your prompt response!...

Here is my current situation...and set-up/signal path...

Signal Sources: Streaming audio output via laptop PC...,
Dedicated CD transport, (new)...(Cambridge Audio CXC)...

BOTH feeding a Schiit Audio Yggdrasil D/A converter...

Which feeds a Schiit Audio Freya PRE-AMP...

Which feeds a Marchand active stereo crossover, (set at 60 Hz./24 dB. /octave) (recommended/suggested crossover point for the mini-monitors via the manufacturer)...

The Marchand crossover outputs both the left and right channel bass frequencies (sub 60 Hz. frequencies) to a pair of JL Audio Dominion D-108 subwoofers,,.(and the subwoofers exhibit an 18 millisecond group delay due to the amplifier/electronics between the signal entering the sub and the driver receiving that signal),...the 60 Hz. and above frequencies outpul to an Aragon 4004 amplifier...

The amplifier feeds the signal to a pair of "Audience" Clairaudient 1+ 1 v2 mini-monitors...(crossover-less design, bipolar active front and rear drivers, with bipolar side-firing passive drivers), the active drivers being full-range...so that the mini-monitors exhibit NO inherent group delay due the absence of any passive x-over...)

(Currently, this is a near-field listening arrangement in a relatively small room...however, that in all likelihood will eventually change to a larger space where room effects might come into greater play...)

Though I am reasonably happy with this result, in a PERFECT world, I would like to improve on the system response/output...

Qualities I would ideally like to improve:

1. The frequency response of the mini-monitor-subwoofer stereo system...(a reasonably flat frequency response curve from bottom to top...)
2. The impulse response of both the subwoofers relative to the mini-monitors...(such that the subwoofers AND the mini-monitors respond to the input signal at exactly the same time, NOT just simply be in phase...(18 millisecond group delay of the subwoofers must be "cancelled" by delaying the output of the mini-monitors...)
3.In addition, if there are any additional timing errors exhibited in the system, be it stereo system induced OR room induced, IDEALLY, I would like those to be corrected as well...

I imagine that the only way to do ALL of this, (including the delay of the mini-monitors), as such:

REPLACE both my current Marchand active crossover AND my Schiit Audio Freya pre-amplifier with the Mini DSP SHD Studio... placing the Mini DSP SHD Studio unit between my signal sources and my Schiit D/A converter, essentially utilizing the Mini-DSP SHD Studio unit as my dual purpose, X-over AND pre-amplifier...sending the 60Hz and above signal to my amplifier to my mini-monitors, and the sub-60 Hz. signal to a secondary D/A converter...(a more modest Schiit D/A converter?), and send THAT signal to my subwoofers...???...

Have you any alternative ideas on how I might accomplish my above goals...?...

What I do NOT wish to do is "toss out the baby with the bathwater",...i.e., sacrifice the reasonably good sounding system qualities I now enjoy by replacing my current X-over and pre-amplifier with an audio component that might introduce noise and or other distortion artifacts which ultimately worsen the overall sound of my system...

Peace,
T. A. Kogstrom

P.S. I welcome all input from others regarding my question(s)...
 
     Share This Post       
post-14581656
Post #143 of 155

bigshot

Headphoneus Supremus
Joined
Nov 16, 2004
Messages
21,252
Reaction score
3,397
Location
Hollywood USA
Joined
Nov 16, 2004
Location
Hollywood USA
Posts
21,252
Likes
3,397
Website
www.facebook.com
My gut feeling is that balancing your response will make the biggest improvement. I'm not sure you would notice any difference with timing fixes. But if you're getting the mini DSP for the equalizer, you can certainly play with the timing and see if it helps any. I'd also suggest trying an 80Hz crossover to the sub rather than 60. That's low enough that directionality isn't an issue, and high enough that it will take some of the load off the mains and make it easier to flatten out to suit your room. I imagine you have done a bit of room treatment. If not, I would say that would give a lot of margin for improvement too.
 
     Share This Post       
post-15170358
Post #144 of 155

jgazal

500+ Head-Fier
Joined
Nov 28, 2006
Messages
728
Reaction score
133
Joined
Nov 28, 2006
Posts
728
Likes
133
Website
www.head-fi.org
Guys, I am trying to use the Equalizer APO to separately equalize each channel of my living room stereo system, but every filter I create in the config tab affects both channels.
If I create new tabs for the left and right speakers with separate filters, still only the config tab affects the output device.
Does anyone could provide instructions on how to do it?
 
     Share This Post       
post-15170402
Post #145 of 155

jgazal

500+ Head-Fier
Joined
Nov 28, 2006
Messages
728
Reaction score
133
Joined
Nov 28, 2006
Posts
728
Likes
133
Website
www.head-fi.org
Apparently I got it. It basically edit config files, but it always uses the base config.txt file and then you can insert other config files inside the config.txt. I just can’t effectively implement a digital balance. I have tried to insert a preamp control in the left channel txt file, but I cannot hear the balance shift. Only the main preamp inside the config.txt works for both channels. Maybe am I creating the preamp in the wrong order?
 
Last edited:
     Share This Post       
post-15170456
Post #147 of 155

castleofargh

Sound Science Forum Moderator
Joined
Jul 2, 2011
Messages
8,968
Reaction score
4,298
Joined
Jul 2, 2011
Posts
8,968
Likes
4,298
you had a problem and now you don't, we're just that good!
you're welcome.
 
     Share This Post       
  • Like
Reactions: jgazal
post-15178795
Post #148 of 155

jgazal

500+ Head-Fier
Joined
Nov 28, 2006
Messages
728
Reaction score
133
Joined
Nov 28, 2006
Posts
728
Likes
133
Website
www.head-fi.org
I have just implemented a FIR filter for my room using REW, Rephase and Equalizer APO and the result is a clear improvement. Thank you all for the tips!
 
     Share This Post       
  • Like
Reactions: RRod
post-15180079
Post #149 of 155

ironmine

500+ Head-Fier
Joined
Jun 13, 2008
Messages
636
Reaction score
55
Location
Far East of Russia
Joined
Jun 13, 2008
Location
Far East of Russia
Posts
636
Likes
55
I have just implemented a FIR filter for my room using REW, Rephase and Equalizer APO and the result is a clear improvement. Thank you all for the tips!
Two months ago I also discovered rePhase software. It took me lots of time, trials and errors to master it, with the help of some knowledgeable people from different web-forums. Learning how to eliminate the excess phase of speakers and bring it to the minimum phase has taken my enjoyment of music to a yet new higher level. Before, I only knew how to fix the room resonances only with REW, now I can also make the speakers to be minimum-phase. I never thought it was possible to improve the quality of audio so much without any extra investments (you do need to buy a good mic such as UMIK-1). Fixing the phase makes the room to "disappear".

By the way, I found this great free, simple and stable VST convolver: MConvolutionEZ.
(just use "custom path" to show to the plugin where you keep your correction impulses)




Unfortunately, you cannot measure the headphones with UMIK-1 mic as easily as speakers.

But I found this way (it corrects the frequency response of headphones, but not their phase): you can take a correction impulse filter (it's a .wav file) from website AutoEq (it has lots of headphone and earphone models, even those not supported by TB Morphit or Sonarworks) and load it into a convolver plugin of your choice. You can play with Dry/Wet setting to make the sound to your liking (some correction impulse files, when they are applied 100%, make the resulting frequency response 100% flat which most people will not enjoy).

However, what do you do when you want to have the full control over frequency correction process and its result but you are stuck with some earphones or headphones which are not supported by TB Morphit, Sonarworks, etc.? You need to have the frequency measurement of your phones, but you cannot measure it, you cannot find the measurement online. The only thing you find (e.g., at AutoEq) is a correction impulse filter.

Well, in case the target curve of a correction impulse is the perfectly flat line (which is not always so), you can invert it and basically get the frequency measurement of your phones. Inversion can be done in REW, by dividing the perfect impulse by the correction filter you've found. (Future versions of REW will have the additional feature: arithmetic operation 1/A, where 1 = perfect impulse.)

Once you get the measurement of your phones, you can assume full control over it and correct it any way you want it: in REW or in rePhase. You can set as a goal any target curve you prefer, you can correct using minimum phase or linear phase, etc...
 
Last edited:
     Share This Post       
  • Like
Reactions: jgazal
post-15194091
Post #150 of 155

ironmine

500+ Head-Fier
Joined
Jun 13, 2008
Messages
636
Reaction score
55
Location
Far East of Russia
Joined
Jun 13, 2008
Location
Far East of Russia
Posts
636
Likes
55
I've found one VST plugin which emulates the tube sound, but, unlike other tube-emulating plugins, it has a unique feature: it keeps the original dynamics of audio:


MeldaProductin - MPhatik

Try these settings:
Dry/Wet - 100%
Dynamics - 100%
Mode: Matching
AMP - enabled, Vintage 1, drive - 25%, character - 25%
Compressor - disabled
Convolution - disabled
All other settings (drive, range, speed, attack, etc.) - by default (don't touch them)
Side chain, eq, phychoacoustic prefiltering are all off.

I tried activating this plugin on several tracks already for testing. It does make the sound more three-dimensional, the bass sounds fuller and rounder, the stage depth is increased. Sounds as if the vocalist takes a few steps back. There are also other changes in the perceived sound which are hard to describe in words. Music starts flowing in a more relaxed manner.

If you tried other tube-emulating plugins and were disappointed with them, give this MPhatik a chance. I was surprised how carefully MPhatik preserves the transparency of the audio (compared to other tube emulation VST plugins).

Its side-chain capability looks also interesting (you can emulate the tube sound with other plugins, and use MPhatik to restore the compressed dynamics back to original). But I haven't figured out yet how to use this feature properly.
 
Last edited:
     Share This Post       

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top