Page 2 of 2 FirstFirst 12
Results 16 to 29 of 29

Thread: Is Good DAC Still Necessary With DSP Sound Processor?

  1. #16
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,942
    Robs last post is good advice

  2. #17
    Dang. Amateur speakerdave's Avatar
    Join Date
    May 2003
    Location
    Northern California
    Posts
    3,734
    For me, and possibly for anyone else not using DSP, the idea of getting an outboard DAC is more about questionable implementation of analog audio output sections in consumer CD players. When my Philips SA1000 died and I went to a Denon it was no doubt a step backward sonically. A Bryston DAC brought the life back, except, of course, for SACD's.
    "Audio is filled with dangerous amateurs." --- Tim de Paravicini

  3. #18
    Administrator Mr. Widget's Avatar
    Join Date
    Apr 2003
    Location
    San Francisco
    Posts
    9,720
    A few thoughts on subjects covered in this thread:

    1. In my opinion the most significant audible differences in DACs have more to do with clocking and the analog output than the chip sets and digital topologies used.

    2. In my experience sample rate conversions are almost always a bad idea and in many DACs upsampling isn't a great idea either.

    3. If you are using a DSP you should also be concerned with your A to D too unless everything in your system is already digital.


    Widget

  4. #19
    Senior Member
    Join Date
    Mar 2008
    Location
    UK
    Posts
    235
    Speakers with their electro mechanical interface must be the weakest link in the chain and in the analogue domain also record player interface.

    In the digital domain things seem a bit more blurred the method of outputting the digital signal (computer, streamer, any EQ, DSP etc.) and then the quality of the DAC.

    As usual everything we put in the signal path makes a difference and measuring and modifying the signal to get a straight line output does not always give the desired result.

    My own experience is that the DAC is fundamental to the sound of a system as this is the component that changes the digital signal to an analogue output. The good thing for us is there are people who devote their lives to getting the best possible sound, in fact many are obsessed.
    For someone with limited knowledge and limited budget it’s a nightmare trying to sort out the utter rubbish from the actual facts.
    I have ended up with a Level 4 Lampizator DAC none DSD and I’m sure there are many such units out there.

    My dilemma is it possible to obtain an all-digital system, including the crossover and still use the DAC of ones choice without having to purchase multiple units? Well out of my budget.

    It’s easy to forget that what sounded good last year still sounds good now.

    Dave

  5. #20
    Senior Member
    Join Date
    Feb 2008
    Location
    Ashland, MA
    Posts
    908

    Dead thread but wanted to comment here

    Quote Originally Posted by Ducatista47 View Post
    A voice of sanity. I involuntarily laugh or facepalm, depending on my mood, whenever someone tries to improve on the theoretical work of Harry Nyquist. Those who consider themselves Golden Ears get pretty defensive when faced with engineering facts. At that point what follows reminds me of fairy tales.

    As for digital volume controls in DACs throwing signal away, the one I use will if I attenuate more than, I think it is, 55dBs. Who turns down their system 55dBs?
    Reviving a Zombie Thread since this is some of the most lucid thinking I've seen on the subjects. Great stuff guys.

    I am reintegrating some DSP, specifically a DBX Venu360, into my rig. The above conversation has been en-pointe for the most part, but I want to address this above 55dB issue-

    I regularly turn down my system 55dB or more, and have spent a lot of time with attenuation in various forms trying to maintain resolution and low noise. I've made transformer volume controls of multiple types, as well as buffers, traditional preamps and resistive attenuators. For the DSP, my input levels will be exceedingly low if I don't work around that issue. The problem is that high efficiency speakers, coupled with powerful and sensitive home amplifiers (without any input trimmers) will operate in home spaces at very low output relative to their maximum. I have some 200-500 "real" watts per driver available and most of my drivers are generally 97dB or more, in a modest space. I prefer the sound of big efficient systems (as we tend to around here), but it means that typical gain structures are way too much.

    The solution is well known to many in the pro/venue space, where it's generally a mixer or the like feeding the DSP. Input trimmers on amplifier inputs allow you to structure your gain differently, so that you can avoid pushing against either the noise/resolution floor or potential clipping. The former is usually able to be fixed easily by adjusting the trimpots on amplifiers downward and the mixer master volume upward, leading to a higher voltage throughput on the DSP, and better noise and quantization performance. If you're clipping the mixer to get that voltage, you done gone too far, and you need to back off the mixer volume and up the trims on the amps.

    So for me, I've spent the last day deciding around some of what I'll do to correct the issue- I've been waffling between two options:


    1. Simple DIY resistor networks, either as part of cables, or wired on the input XLRs of the amps
    2. Having someone else do it for me- https://naiant.com/custom_audio_repr...nline-devices/, either as an adaptor attenuator or built into some cable.


    There are cheap XLR attenuators on market but they do not seem to be of sufficient quality for my taste, with many designed around 600 ohm operation (too low) or not properly balanced (an L-pad on just one leg of a balanced connection will lower the volume of a differential input, but isn't ideal). Naiant doesn't seem to suffer from any of those issues, and the adjustable units are very, very cool.

    One can also change the gain structure as another way to get down that path, but the point is, where your gain takes place in a system with ADC is very important, and 55dB isn't an unheard of level of attenuation.

  6. #21
    Senior Member
    Join Date
    Mar 2008
    Location
    UK
    Posts
    235
    Quote Originally Posted by badman View Post
    Reviving a Zombie Thread since this is some of the most lucid thinking I've seen on the subjects. Great stuff guys.

    I am reintegrating some DSP, specifically a DBX Venu360, into my rig. The above conversation has been en-pointe for the most part, but I want to address this above 55dB issue-

    I regularly turn down my system 55dB or more, and have spent a lot of time with attenuation in various forms trying to maintain resolution and low noise. I've made transformer volume controls of multiple types, as well as buffers, traditional preamps and resistive attenuators. For the DSP, my input levels will be exceedingly low if I don't work around that issue. The problem is that high efficiency speakers, coupled with powerful and sensitive home amplifiers (without any input trimmers) will operate in home spaces at very low output relative to their maximum. I have some 200-500 "real" watts per driver available and most of my drivers are generally 97dB or more, in a modest space. I prefer the sound of big efficient systems (as we tend to around here), but it means that typical gain structures are way too much.

    The solution is well known to many in the pro/venue space, where it's generally a mixer or the like feeding the DSP. Input trimmers on amplifier inputs allow you to structure your gain differently, so that you can avoid pushing against either the noise/resolution floor or potential clipping. The former is usually able to be fixed easily by adjusting the trimpots on amplifiers downward and the mixer master volume upward, leading to a higher voltage throughput on the DSP, and better noise and quantization performance. If you're clipping the mixer to get that voltage, you done gone too far, and you need to back off the mixer volume and up the trims on the amps.

    So for me, I've spent the last day deciding around some of what I'll do to correct the issue- I've been waffling between two options:


    1. Simple DIY resistor networks, either as part of cables, or wired on the input XLRs of the amps
    2. Having someone else do it for me- https://naiant.com/custom_audio_repr...nline-devices/, either as an adaptor attenuator or built into some cable.


    There are cheap XLR attenuators on market but they do not seem to be of sufficient quality for my taste, with many designed around 600 ohm operation (too low) or not properly balanced (an L-pad on just one leg of a balanced connection will lower the volume of a differential input, but isn't ideal). Naiant doesn't seem to suffer from any of those issues, and the adjustable units are very, very cool.

    One can also change the gain structure as another way to get down that path, but the point is, where your gain takes place in a system with ADC is very important, and 55dB isn't an unheard of level of attenuation.
    I now use a lightspeed attenuator as my main control and noticed an immediate increase in bass control, I guess a better impedance match. Now I am going to contradict myself, my Lampizator DAC went faulty so in desperation I hooked the system up direct to the computer and am amazed at the sound its producing. It is a high end Laptop but its not supposed to sound that good!!!!!! Another problem one of the amplifiers also decided it had enough and started smoking. Again in desperation to have something to listen to I connected my old Denon RCD M37DAB driving the compression drivers. Another revelation, I had never heard such clarity from the top end. My old Yamaha M60 amp is going up for sale but as its switchable Class A its going to be difficult to replace maybe try a Chinese offering?

  7. #22
    Senior Member
    Join Date
    Feb 2008
    Location
    Ashland, MA
    Posts
    908
    Quote Originally Posted by David Ketley View Post
    I now use a lightspeed attenuator as my main control and noticed an immediate increase in bass control, I guess a better impedance match. Now I am going to contradict myself, my Lampizator DAC went faulty so in desperation I hooked the system up direct to the computer and am amazed at the sound its producing. It is a high end Laptop but its not supposed to sound that good!!!!!! Another problem one of the amplifiers also decided it had enough and started smoking. Again in desperation to have something to listen to I connected my old Denon RCD M37DAB driving the compression drivers. Another revelation, I had never heard such clarity from the top end. My old Yamaha M60 amp is going up for sale but as its switchable Class A its going to be difficult to replace maybe try a Chinese offering?
    You may be doing that very human thing of "new"="improved". It's very common in this hobby to hear a change as an improvement by default, which is part of why people seem to always upgrade over and over and over, even though at each point they tell themselves "I'm done, this is amazing".

    Regarding amp, I don't know your budget, but I am a serious advocate of hypex NCore amplification. Nord Acoustics is an excellent and fairly priced source for pre-built, in any configuration you can imagine. It's not cheap stuff but it's world class amplification.

  8. #23
    Member sebackman's Avatar
    Join Date
    Feb 2004
    Location
    Europe
    Posts
    675
    Hi,

    Not knowing how your signal chain looks like it is in general a bad idea to introduce "resistors" in the signal chain unless the source is very low impedance and the "reciver" is high impedance. If the source is "pro" ie <600ohms output it is usually fine within limits. I concur with previous posts that you should keep the level into the DSP (if analogue) as high as possible to maintain decent S/N and that may post a problem later in the chain if there are no input trimmers as stated.

    Is this 2-channel or multi, active or passive speakers? Are you running balanced cabling?

    There are several attenuator alternatives in the "pro" market, both passive and active.

    I'm on a fully digital chain with separate multi Burr&Brown VCA's after the DAC's but the newer BSS units do provide a good attenuator function that can be controlled from any iPhone or iPad. They use 41 bit internally which provides for retained resolution even when reducing volume in the digital domain. Most other DSP's does not. In fact, many of them are direct poor in volume control as they do it in or before the DAC chip (usually 16 or 24 bit) and not in the DSP chip (often up to 41 bits FP). Some use analogue VCA on the output which is muck better if correct implemented. Or just a pot for gain control. BSS solved this by implementing attenuation as a DSP function and it is just a nice feature that you can control it via (AA) via an app (iPhone/iPad) or indeed even by a single analogue hard wired pot attached to the rear of the BSS.

    Even if that is not what you want to hear I would sell the DBX and get a BSS BLU160 (or BLU100-103 which is fixed IO), feed it digitally via an digital input card or use the BSS BLU-USB which is a very good asynchronous sound card with USB in and digital BSS BLU-LINK out (I have both). The beauty with BLU-Link is that the BSS sound card will use the BSS BLU-XXX DSP clock via BLU-LINK and as it is asynchronous the potential limitation from the computer OS i limited. This would give you up to 256 channels out @48kHz. I would then use the BSS digital volume control controlled via iPhone or iPad and feed analogue signal to your power amps.

    There is currently no other simple solution to a fully digital DSP/active crossover chain with working volume control in the market today, at reasonable price. There may be separate DAC's in the market with good volume controls implemented (digital or analogue) but the problem will be to calibrate levels and to control all of them at the same time if more than 2 channels. I've seen some attempts but not really any successful solutions.

    The balanced Burr&Brown VCA's (2 x 8 channels) I'm using cost an arm and half a leg at the time and was introduced before I went with BSS. Today I would use the BSS volume control to control volume to my 8+8+2 power amps (3-way active 5.2). That is also what JBL is using in the Mark Levinson M2 package (the JBL DSP SDSC's is a re-badged BSS)

    Kind regards
    //Rob
    The solution to the problem changes the problem.
    -And always remember that all of your equipment was made by the lowest bidder

  9. #24
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,942
    Hi Rob,

    Are those and BSS BLU160’s (or BLU100-103) the ones that make so much fan noise they become an annoying?

    I recall this discussed in a thread sometime back.

  10. #25
    Member sebackman's Avatar
    Join Date
    Feb 2004
    Location
    Europe
    Posts
    675
    Hi Ian,

    Sure is. However you can mount silent Noctua fans or a slow 120mm fan and be done. Even in a quiet living room.

    Some use them without cover and the fans disconnected. Seems to work fine.

    Alternatively use a BLU50 without any fan and add BLU BOB’s BLU BIB’s to get desired number of output and input channels. No fans at all.

    Kind regards
    //Rob
    The solution to the problem changes the problem.
    -And always remember that all of your equipment was made by the lowest bidder

  11. #26
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,942
    There must be an easier way surely.

  12. #27
    Member sebackman's Avatar
    Join Date
    Feb 2004
    Location
    Europe
    Posts
    675
    I agree Ian,

    The truth is that the HW is not important, the algoritms are. From having had the fortune to having sample many DSP’s I would say that BSS is superior to most from a sonic perspective. You get what you pay for....

    And if fan noise is a concern buy the fan less BLU50.

    kind regards
    //Rob
    The solution to the problem changes the problem.
    -And always remember that all of your equipment was made by the lowest bidder

  13. #28
    Senior Member
    Join Date
    Apr 2003
    Location
    Australia
    Posts
    7,942
    Quote Originally Posted by sebackman View Post
    I agree Ian,

    The truth is that the HW is not important, the algoritms are. From having had the fortune to having sample many DSP’s I would say that BSS is superior to most from a sonic perspective. You get what you pay for....

    And if fan noise is a concern buy the fan less BLU50.

    kind regards
    //Rob
    From my own research the out come is related to how the dsp is integrated into the loudspeaker design and hardware.

    The “how” is the expertise and this is the single biggest barrier to success in any diy audio project or journey.

    The diy user might have skills in a few areas but it’s unlikely they will have have expertise and practical application in all the key deliverables or what they are? Hence the question of this thread and the rabbit hole that follows. The ultimate unknowns are will the project get finished and will it deliver the outcomes? That lack of certainty is where help is most needed.

    An analogy is go build a modern car on your own or an aircraft. The dude decides to buy the most elaborate mag wheels because his ego tells him that is what’s important.

    How many people can seriously say l can do that and how many would say are you crazy 😜?

  14. #29
    Senior Member
    Join Date
    Oct 2009
    Location
    Denmark
    Posts
    256
    Quote Originally Posted by sebackman View Post
    Dear David,

    I look to be a fine piece of equipment and if it sounds good to you it does.
    Using different gear may produce a different sound which by it selves does not imply that your or the alternative gear is better or worse. All gear alters the signal in way or the other, the trick is to find an “alteration” that suits your ears. -And that may or may not be the combination of gear with the overall lowest sound alteration/coloration. J The important piece is that you like the sound your combination of gear produces.

    If you are open to alternative information, first of all I would check that your USB sound card (DAC) is true asynchronous, it is not possible to read from the web page. Info on the implication can be found here.
    http://www.audioresearch.com/ContentsFiles/DAC8_white_paper.pdf
    http://www.hifi-advice.com/USB-synchronous-asynchronous-info.html

    I my experience when using a reasonable computer (not esoteric high-tech dedicated) the asynchronous protocol does make a difference. -More than any of my DAC’s. I personally cannot attribute noticeable sound deterioration to any of the DAC’s I have if the rest of the chain is correct. Not by listening or measuring. I do happen to have digital out from some of my DSP’s, both SPDIF, DSD and BLU-LINK and I cannot hear a difference with by just changing the DAC chip. Sorry. (Digitalout and separate DAC). However, I’m not saying that others cannot.

    My recipeis as Einstein said; “keep it as simple as possible, but not simpler” . J

    Get a good asynchronous sound card with XO clock and a good digital out. Set the computer to output native format. Feed your DSP digital signal (or analogue) and use the analog outs in the DSP. If your DSP does have clock input or output do connect the sound card and DSP to share the clock and turn of the SRC.

    My favorite right now, as I’m using only BSS DSP, is new nice little device from BSS called BLU-USB. It is an asynchronous sound card that feed the BSS DSP with a proprietary digital signal (BLU-LINK) but the neat thing is that there are now SRC’s anywhere and the BLU-USB uses the clock in the DSP so they are always paced. Signal path is short and sweet. But that is for a different post.

    Kindregards
    //Rob
    This is an old thread, but hoping it can be made active

    You mention that with the BLU-USB you can disable SRC anywhere... But on my BLU160 you are forced to select either 48khz or 96khz in the configuration. Isn't it a requirement for BLU-link that all devices are set to the same sampling rate? So what happens if you play a 96khz file on your computer? And then right after a 48khz file? (Or a 44.1 for that matter). Won't it be re-sampled to whatever sampling rate you have chosen in the BSS? The clock will be shared though which is a good thing, but I can't see how you can bypass SRC?

    EDIT: I will hopefully soon be able to use Dante and the question above is very relevant to me.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. SK2-1000 did not sound good - is it me?
    By Jonas_h in forum Lansing Product General Information
    Replies: 5
    Last Post: 01-21-2013, 11:44 AM
  2. How good will this speaker sound?
    By dkalsi in forum Lansing Product General Information
    Replies: 24
    Last Post: 05-03-2007, 04:26 PM
  3. Have-Good-Sound-with-your-beer-project...
    By Ralf in forum Lansing Product General Information
    Replies: 45
    Last Post: 04-09-2006, 06:06 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •