DSP 01: Real, Legit Audiophile Goodness

About six months ago, we saw [tshen2]’s work on the DSP 01, a 2-input, 6-output DSP and crossover for extreme audiophiles, and we’re not talking about oxygen free rooms here. The DSP 01 turns a USB audio output into six outputs that will give you perfectly flat eq across bass, mids, and highs, integrates with a 6x100W amplifier, and compensates for room noise. There was a huge update to the project recently and [tshen] is more than happy to share the details

Getting to this phase of the project hasn’t been without its problems. To get the DSP communicating to a computer through a USB port, [tshen2] found a potential solution in the CP2114 USB to I2S Bridge. This device should function as a USB audio sink, translating digital audio into something the DSP understands. This chip did not work in [tshen]’s design. The CP2114 simply does I2S wrong; the I2S spec says the clock must be continuous. This chip implements I2S with a SPI, firmware, and a few other things, making it incompatible with to-spec I2S.

While there was some problems with getting audio in to the device, the core of the device has remained unchanged. [tshen2] is still using the Analog Devices DSP, with the interesting SigmaStudio being used to compensate for the frequency response of the room. This real, legit, science-based audiophile territory here, and an impressive development for a field that – sometimes understandably – doesn’t get the respect it deserves.

22 thoughts on “DSP 01: Real, Legit Audiophile Goodness

  1. For one of our shows (I do some sound and light tech for a community theatre near me), someone borrowed a “magic” box from their school that basically does exactly this + removes feedback loops.

    Gotta admit, I wish we had one of those on a more permanent basis. The thing made my life real easy.

    1. a decade ago, you could get a DSP core, six 24-bit data converters, a flexible self-booting digital interface, an auxiliary ADC and an awesome software package for $10? not really!
      and then this chip came out in 2007 :)

      1. The software package is awesome if all you want to do is within it’s very severe limitations.

        You’re a capable programmer … on something like a cortex M4 you could do stuff like 256+ point filters with FFT overlap add/save, It would be complete overkill, but you could get one silky smooth bode plot if you optimized for a single position with that.

  2. “Tuning” a room isn’t just an audiophile thing. This is a for-real makes-it-sound-better technique that’s used in professional environments every day around the world.

    Back when I was a sound engineer we used Smaart (http://en.wikipedia.org/wiki/Smaart) to look at transfer curves, and various different gizmos (depending on the year and budget) to get a flat response curve– both outdoors and indoors.
    My rig is still around here somewhere…

    If you’re not familiar with it, you’d be stunned to hear the before vs. after most of the time.

  3. I had to suffer the whips and chains of a crack sound man once helping set up for a major jam band at our brew hall.
    We are into “ringing out” the monitors.
    He’s onstage, I am in house.
    Turn one up he barks, weeee. Turn down 512 Hz 4 and a half Db he shouts… silence (noticing that it’s light was lit on the EQ/display). Turn it up more… thru 3 iterations, he nailed the correct slider and amount each time on a 31 band EQ. The gain was now up many Db from when we started.
    Three more monitor channels to do.
    Done in a minute. He ran house.The sellout show was awesome. I could hear everything including the vocals.
    When it comes to the room you have to be able to sense the average and not EQ for one place unless you are the only listener. Weather at home or in the house.

    1. hi echodelta!
      you are correct about ‘don’t EQ for one place’.
      any particular EQ is only correct for one (extremely specific) location. i get around that by calculating a weighted average of four listening positions. i then correct for this averaged curve, which produces a decent result at all four positions.
      this problem with this approach, as i have just realized, is that EQ-ing from the listening position only makes sense below a few hundred Hz. above that, your brain hears a complex mix of direct sound, early-reflections and steady-state reverb. so you have to blend the listening-position measurement with speaker-only measurements (preferably anechoic) to figure out a really good EQ. that’ll be the next update :)

  4. Utter nonsense. First of all an completely flat frequency response is impossible it will always no matter what be +/- couple thousands of a dB on a best case scenario. Second all all all these audiophiles do not even know about the importance of room acoustics and acoustic treatment. No matter how perfect your room is it will have a bigger impact on the sound then anything. Purpose build and treated million dollar studios get within+/- 1db or so and even cheap a/d converters acheve much much more linear responses. Anything that does psychoacoustic compensenstion for room”noise” is nonsense and it will only further damage the audio. In the end the most important thing is how well the record was produced and how good the room/speakers are. Not silly overpriced dac,s

    1. Now leave your theoretical super-studio and try to set up some sound in a real room, or in the … bar for example. I’ve heard too many awful installations in public places that convinced me – some people need tools to get it right.

      And seriously, this project should go on Kickstarter. Much more useful than Pono.

    2. So, utter nonsense is suggesting the closest you can get to flat response is “thousands of ” dB off, “best case”. Also, apparently you’re the only one here who understands acoustics, but “psychoacoustics” is nonsense too (according to you). Here’s a good book:

  5. Serious audiophiles, as opposed to ordinary consumers of audio gear, pay a lot of attention to adjusting for the influence of the room on sound. I worked for a B&O retailer for a while. Their top end speaker the BEOLAB 5 had a built in mic and the capability to test the room and apply EQ. B&O’s top speakers were all self powered with internal DSP for the best response. This was a wildly successful arrangement, though at a very high price. There are a number of other benefits such as allowing high SPL outputs without overdriving the least sensitive drivers by limiting peak power levels.

  6. Can the DSP board be used on its own to correct for room characteristics? I have a soundbar-5.1 setup where the subwoofer is wireless and drives the satellites (because I’m finicky about wires), but it would be cool to stick this in between my S/PDIF receiver ( I’m using the same wireless NTSC trick) to make things sound great from my couch.

    1. hello Trevor!
      yes, you can definitely use just the DSP board as a 2-in, 2-out to perform EQ. but let me ask, how many channels are being received by your S/PDIF receiver?

      at present, my S/PDIF receiver only works for stereo S/PDIF – it does not work for Dolby multi-channel sound. this is because Dolby patents and binary-blobs everything into inscrutability, and if you aren’t paying licensing fees, you won’t have S/PDIF 5-channel capability. my design should work just fine for stereo, if that’s what you’re doing!

      i’m glad to see more people are doing the wireless NTSC thing :)


      1. Oh! My mistake, I assumed that’s what the 6x amplifier was for. Too bad there’s not an easier way to extract the channels – this setup would be even more amazing with surround sound.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.