Jump to content
IGNORED

Best DSP Hardware


Recommended Posts

I listen from only one position, so I think I will go with DSP to "tune" that listening position rather than room treatments (which I have tried with no success).  My question is, what is the best hardware?  I would put DSP between my preamp and amp.  Roon seems like it has some useful DSP, but that won't work with MQA (I think), so I think I need DSP hardware.  

 

Recommendations to me have been in the pro audio space.

- Mark

 

Synology DS916+ > SoTM dCBL-CAT7 > Netgear switch > SoTM dCBL-CAT7 > dCS Vivaldi Upsampler (Nordost Valhalla 2 power cord) > Nordost Valhalla 2 Dual 110 Ohm AES/EBU > dCS Vivaldi DAC (David Elrod Statement Gold power cord) > Nordost Valhalla 2 xlr > Absolare Passion preamp (Nordost Valhalla 2 power cord) > Nordost Valhalla 2 xlr > VTL MB-450 III (Shunyata King Cobra CX power cords) > Nordost Valhalla 2 speaker > Kaiser Kaewero Classic /JL Audio F110 (Wireworld Platinum power cord).

 

Power Conditioning: Entreq Olympus Tellus grounding (AC, preamp and dac) / Shunyata Hydra Triton + Typhoon (Shunyata Anaconda ZiTron umbilical/Shunyata King Cobra CX power cord) > Furutec GTX D-Rhodium AC outlet.

Link to comment

Don't put your DSP hardware in between your preamp and amp. If you do this, you will be feeding it an analog signal, which will then undergo ADC, then DSP, then through another DAC. In other words, the final DAC you will be listening to will be the cheap DAC in the DSP hardware, as opposed to that beautiful DCS that you own. Not to mention, you will be inserting another layer of ADC-DAC. Some people don't think that is audible, but my own listening tests tell me that it is very easily audible. Even easier to hear than differences between cables. 

 

I suggest you insert your DSP prior to your DCS upsampler. This way, all the processing is done in the digital domain with no conversion to analog. Unfortunately, I do not know whether this will prevent your DCS Vivaldi from being able to "pull" music from your NAS - something you would have to research. 

 

As for WHICH DSP to use, the sky is the limit. My own solution was to build a PC, and run Acourate and HQPlayer. I think this is the ultimate solution. All DSP hardware you care to name lacks the raw grunt of a PC. Lots of grunt means you can have more taps (in simple terms, a tap is a unit of computing power), and thus apply more correction. It also means that you can perform convolution in DSD (a specialty of HQPlayer), without any conversion to PCM whatsoever. There is no other DSP solution that allows you to apply DSP in DSD. 

 

If you want to dip your toes in DSP, then I would recommend MiniDSP as an inexpensive and easy to learn unit. But if you want the ultimate solution, the PC is the way to go. And I can see from your rather tasty equipment list that you are "whatever it takes" kind of guy. 

(edit) I should also point out that if you own any DSD files, the ONLY way to apply DSP to these is to use a computer. None of the hardware DSP solutions will work - these only accept PCM signal.

 

I also realize that I forgot to address your MQA concern. Unfortunately,  PC's are not MQA certified, so you won't be hearing MQA through your system. There is no solution that I am aware of that will allow you to apply DSP to MQA and preserve the "MQA-ness", but I am happy to be corrected if I am wrong. As for me, I am unconvinced of the benefits of MQA so I decided not to pursue it. 

Link to comment
3 hours ago, Keith_W said:

If you want to dip your toes in DSP, then I would recommend MiniDSP as an inexpensive and easy to learn unit. But if you want the ultimate solution, the PC is the way to go.

 

If I understand the miniDSP product range, there is no version that allows you to compare different DSP philosophies.  The OpenDRC version has the convolution engine required to run Acourate (or other DSP programs that generate FIR filters), the Dirac version runs only the Dirac plugin, and most or all the other versions of miniDSP support only IIR parametric EQ filters such as those generated by REW.

 

Is there a version of miniDSP that can do all these things?

 

Since you already use a PC for computer audio, it seems simpler to use the same PC for DSP instead of buying a special-purpose DSP box.

HQPlayer (on 3.8 GHz 8-core i7 iMac 2020) > NAA (on 2012 Mac Mini i7) > RME ADI-2 v2 > Benchmark AHB-2 > Thiel 3.7

Link to comment
7 hours ago, Keith_W said:

 

Hi Tom, as I said in my post, I would put it in between your Synology and the DCS upsampler. 

Hi Keith,

Actually i am not the starter of the topic nor, sadly, do I own a DCS DAC.

I do own a a MiniDSP 2x4 box that works with unbalanced RCA jacks. I can see them in the chain only after the Dac???

Are you aware of mini DSP models that offers digital to digital conversion, as requested from your proposal, i.e. ethernet to usb?

Cheers Tom

Link to comment
1 hour ago, DuckToller said:

Are you aware of mini DSP models that offers digital to digital conversion, as requested from your proposal, i.e. ethernet to usb?

 

The DI suffix means digital in and out.  AES, S/PDIF and Toslink.  USB port is only for programming the filters, not for audio in/out.

 

https://www.minidsp.com/products/opendrc-series/opendrc-di

HQPlayer (on 3.8 GHz 8-core i7 iMac 2020) > NAA (on 2012 Mac Mini i7) > RME ADI-2 v2 > Benchmark AHB-2 > Thiel 3.7

Link to comment
17 hours ago, Keith_W said:

There is no solution that I am aware of that will allow you to apply DSP to MQA and preserve the "MQA-ness", but I am happy to be corrected if I am wrong.

If software decoding is an option (with Audirvana or Tidal), little is lost by doing DSP on the 96 kHz output. You'd only be missing the "render" step, but that's just upsampling with a particularly nasty filter, so you're probably better off without it anyway.

Link to comment
1 hour ago, mansr said:

You'd only be missing the "render" step, but that's just upsampling with a particularly nasty filter, so you're probably better off without it anyway.

Can't the filter be customized by the DAC manufacturer implementing MQA?  For example, Berkeley plans to offer MQA on the Alpha DAC Reference Series 2 in 2Q2017.  It will be an MQA renderer.  I can't believe Berkeley is going to use some awful filter on their $19,500 DAC just so they can tick the MQA box in their specs.

 

I know every implementation needs to approved by MQA.  I can imagine the DAC manufacturers just love that part. 

Pareto Audio AMD 7700 Server --> Berkeley Alpha USB --> Jeff Rowland Aeris --> Jeff Rowland 625 S2 --> Focal Utopia 3 Diablos with 2 x Focal Electra SW 1000 BE subs

 

i7-6700K/Windows 10  --> EVGA Nu Audio Card --> Focal CMS50's 

Link to comment
On 5/17/2017 at 5:44 PM, MarkS said:

I listen from only one position, so I think I will go with DSP to "tune" that listening position rather than room treatments (which I have tried with no success).  My question is, what is the best hardware?  I would put DSP between my preamp and amp.  Roon seems like it has some useful DSP, but that won't work with MQA (I think), so I think I need DSP hardware.  

 

Recommendations to me have been in the pro audio space.

It's not clear from your post what the problem is that you tried to solve with room treatment. Describing that problem might help to get suggestions specifically tailored to your needs. For my needs, I use a DSPeaker box, which is a hardware solution and works very well for what I was trying to accomplish, but may be totally the wrong thing for you.

For my system details, please see my profile. Thank you.

Link to comment

Your right.  I should have been more specific.  

 

I had someone do some measurements, and there are various peaks and dips (some substantial) at certain frequencies.  I am trying to flatten things out, which is why I think DSP will work, especially because I am one person in one chair in one place.

 

I did not want to go with hardware, but I actually think that may be best, and is the only solution that would work with MQA also.

- Mark

 

Synology DS916+ > SoTM dCBL-CAT7 > Netgear switch > SoTM dCBL-CAT7 > dCS Vivaldi Upsampler (Nordost Valhalla 2 power cord) > Nordost Valhalla 2 Dual 110 Ohm AES/EBU > dCS Vivaldi DAC (David Elrod Statement Gold power cord) > Nordost Valhalla 2 xlr > Absolare Passion preamp (Nordost Valhalla 2 power cord) > Nordost Valhalla 2 xlr > VTL MB-450 III (Shunyata King Cobra CX power cords) > Nordost Valhalla 2 speaker > Kaiser Kaewero Classic /JL Audio F110 (Wireworld Platinum power cord).

 

Power Conditioning: Entreq Olympus Tellus grounding (AC, preamp and dac) / Shunyata Hydra Triton + Typhoon (Shunyata Anaconda ZiTron umbilical/Shunyata King Cobra CX power cord) > Furutec GTX D-Rhodium AC outlet.

Link to comment

If the measured peaks and dips are in the low frequency region, the DSPeaker AntiMode Dual Core will help. It has automatic DRC of room modes. It also has a 14 band parametric equalizer, but EQs must be set manually. It will work set between preamp and amp (this is how I use it), though in this configuration it obviously does an ADC-DAC conversion. In my system, this extra conversion does not seem to cause any loss in transparency, as tested through bypassing the device.

For my system details, please see my profile. Thank you.

Link to comment
  • 3 weeks later...

I have found the EQ's as provided in programs like Audirvana+ to be very easy and useful as a first step. In fact I've been doing that for a while now and still haven't taken the deep dive into room correction solutions. My Marantz AV-7005 has an older version of Audyssey that helped in some ways seemed to reduce quality in other ways. I later learned that the implementation in my pre-amp did all it's DSP using a lower sampling rate, perhaps that's what I was hearing. Of course I expect any bit molestation by EQ/DSP is going to cause any downstream MQA decoding to be disabled.

Link to comment
On 6/7/2017 at 9:55 PM, Supperconductor said:

I have found the EQ's as provided in programs like Audirvana+ to be very easy and useful as a first step. In fact I've been doing that for a while now and still haven't taken the deep dive into room correction solutions. My Marantz AV-7005 has an older version of Audyssey that helped in some ways seemed to reduce quality in other ways. I later learned that the implementation in my pre-amp did all it's DSP using a lower sampling rate, perhaps that's what I was hearing. Of course I expect any bit molestation by EQ/DSP is going to cause any downstream MQA decoding to be disabled.

 

The problem with this is that you do not correct for problems in the area of impulse response. Only frequency response. Above that, it is really hard to do it manual and do it good. The right starting point is measuring the room response with a microphone.

Streamer dCS Network Bridge DAC Chord DAVE Amplifier / DRC Lyngdorf TDAI-3400 Speakers Lindemann BL-10 | JL audio E-sub e110 Head-fi and reference Bakoon HPA-21 | Audeze LCD-3 (f) Power and isolation Dedicated power line | Xentek extreme isolation transformer (1KVA, balanced) | Uptone Audio EtherREGEN + Ferrum Hypsos | Sonore OpticalModule + Uptone Audio UltraCap LPS-1.2 | Jensen CI-1RR Cables Jorma Digital XLR (digital), Grimm Audio SQM RCA (analog), Kimber 8TC + WBT (speakers), custom star-quad with Oyaide connectors (AC), Ferrum (DC) and Ghent (ethernet) Software dCS Mosaic | Tidal | Qobuz

Link to comment
  • 1 year later...
On 5/18/2017 at 7:16 AM, Keith_W said:

Don't put your DSP hardware in between your preamp and amp. If you do this, you will be feeding it an analog signal, which will then undergo ADC, then DSP, then through another DAC. In other words, the final DAC you will be listening to will be the cheap DAC in the DSP hardware, as opposed to that beautiful DCS that you own. Not to mention, you will be inserting another layer of ADC-DAC. Some people don't think that is audible, but my own listening tests tell me that it is very easily audible. Even easier to hear than differences between cables. 

 

I suggest you insert your DSP prior to your DCS upsampler. This way, all the processing is done in the digital domain with no conversion to analog.

 

 

 

I have exactly this problem but I am 'forced' to put my DSP after my not-a-million-miles-away-from-dCS Chord DAC because my speakers are active and they come supplied with a bit of modded pro kit as a DSP based crossover. This has now been ditched for a decent DSP/DRC (analogue only out) box (IAP4) that also implements the x-overs

 

So the question is this ... if I do decide to go down the route of moving the DSP/DRC pre-DAC (via a PC or something like DEQX) how do I manage the speaker crossovers, given that any solution must not involve DSP?

Link to comment
  • 1 year later...
On 11/10/2018 at 7:52 AM, User471 said:

 

I have exactly this problem but I am 'forced' to put my DSP after my not-a-million-miles-away-from-dCS Chord DAC because my speakers are active and they come supplied with a bit of modded pro kit as a DSP based crossover. This has now been ditched for a decent DSP/DRC (analogue only out) box (IAP4) that also implements the x-overs

 

So the question is this ... if I do decide to go down the route of moving the DSP/DRC pre-DAC (via a PC or something like DEQX) how do I manage the speaker crossovers, given that any solution must not involve DSP?

 

Sorry for the very late reply. I don't look at this forum very much these days. 

 

The answer to your question: you will need a DAC for every channel that you drive. For example, if you have a pair of speakers which have four drivers each, you will need 8 channels of DAC (i.e. four 2 channel DAC's). Having said that, it is a very bad idea to use four 2 channel DAC's because all their clocks need to be synchronized / slaved to a master clock. There are not many DAC's which accept an external clock, and even if you find them the cost might be prohibitive. It is easier and cheaper to use an 8 channel DAC in the first place. 

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...