I don't know why Signal Theory only applies to testing - the ref is fine but the talk about analyzer filters misplaced - the tested DACs are simply unacceptable because of the high frequency out of band noise - by their (bad) design
the fact that added filtering is needed for good measurements is due to poor execution in the example DAC design
it is required to filter the images, shaped noise for a quality analog output - whatever DAC technology you are using
it doesn't have to be done at 20 kHz - upsampling, digital filtering lets you use 40-50 kHz analog reconstruction filters even with audio CD source
noise shaping can require higher order filters than you might expect, see on some datasheet example circuits
even V output audio DAC with internal sw-C filter deserve analog filtering after the chip - there is always clock feedtru noise on sw-C filter outputs
the fact that added filtering is needed for good measurements is due to poor execution in the example DAC design
it is required to filter the images, shaped noise for a quality analog output - whatever DAC technology you are using
it doesn't have to be done at 20 kHz - upsampling, digital filtering lets you use 40-50 kHz analog reconstruction filters even with audio CD source
noise shaping can require higher order filters than you might expect, see on some datasheet example circuits
even V output audio DAC with internal sw-C filter deserve analog filtering after the chip - there is always clock feedtru noise on sw-C filter outputs
Last edited:
Hi Demian,
You assert that these "ultrasonics" are "inaudible". Do you suggest that you have evidence that they are inaudible or that some claim they are?
Ciao T
You assert that these "ultrasonics" are "inaudible". Do you suggest that you have evidence that they are inaudible or that some claim they are?
Ciao T
Translation error! "xxx" means claimed by the manufacturer but considered inaccurate by many.
Well we had the live there to compare it to. The ostrich story is a myth BTW.
You dismiss the experience of people who love/know live music (it's Austin after all) without even asking the associated equipment. Please give me licence to do the same (I normally wouldn't use it).
Scott,
Let's talk instrumentation, do you have Bekessy hearing test plots for all of the participants? Did you test for legal and or illegal stimulants? Do you have full TEF plots of the room for each listening position? TEF plots of the actual music before and after? Were all of the listeners blindfolded?
Come on listening is a personal experience. Most folks today are fine with MP3. Yet few seriously argue it is the best method of reproduction. If you tell me it was great, I have no problems believing you. But if you tell me it was perfection, well you know!
Let's talk instrumentation, do you have Bekessy hearing test plots for all of the participants? Did you test for legal and or illegal stimulants? Do you have full TEF plots of the room for each listening position? TEF plots of the actual music before and after? Were all of the listeners blindfolded?
Come on listening is a personal experience. Most folks today are fine with MP3. Yet few seriously argue it is the best method of reproduction. If you tell me it was great, I have no problems believing you. But if you tell me it was perfection, well you know!
Not necessary 🙂, but actually we were pretty tame, and I have been to listening sessions that weren't.
I promised to not post anymore, but there is such misunderstandings and confusions that i would like to explain things for those witch do not rely together things they have read or learned here or here.
So i let you some sort of testament and explanation about your digital questions here:
http://www.esperado.fr/testament.html
Feel free to Pm me if any questions or comments, i will not answer in the thread.
So i let you some sort of testament and explanation about your digital questions here:
http://www.esperado.fr/testament.html
Feel free to Pm me if any questions or comments, i will not answer in the thread.
Last edited:
His graphs show the THD+N at 1kHz and 5kHz, from what I can tell. He said the rise time was 7 microseconds, and I think he is indicating he doesn't have time to take new measurements and post them here.
That's my take on it. I'm sure Pavel will correct me if I'm wrong.
For those who doubt about SACD risetime and slew rate. Measurements of rise time (7us) and 10 kHz square.
Single shot capturing, no averaging. Step-like look of 10kHz square rise and fall is due to digital oscilloscope.
Single shot capturing, no averaging. Step-like look of 10kHz square rise and fall is due to digital oscilloscope.
Attachments
Last edited:
Regarding DSD analysis - DSD stream was converted to 88.2 kHz PCM and then analyzed mathematically - FFT.
I was passing the information from Burr brown now part of TI. The basics on how the systems work seems pretty accurate (no lies on the technology in the app note) and covers many of the issues we have been poking at.
The low pass filter gives good measurements for the data sheet and is consistent with the "accepted wisdom" that information above 20 KHz is inaudible. They even point out that in the real device it would have a different low pass reconstruction filter and that the sharp low pass filter is only for measuring with a conventional THD + NOISE instrument.
Validating the inaudibility of the noise above 20 KHz is not an easy task if done rigorously. There are so many sub issues to be resolved before you can even start with the big one. Simply confirming that your information channel doesn't have IM problems that will make the "ultrasonic" information "sonic" is difficult enough.
Scott,
Scott, you miss my point. A USB to SPDIF Dongle outputs SPDIF Digital signals. If you connect these to a speaker there is no sound, you need to add DAC and Amplification, which are determining the sound.
Hence arguing that "A 29 USD USB to SPDIF Dongle makes DAC's obsolete" is like arguing that due to the high quality of Super Petrol cars are obsolete...
I know. But is was convenient to mention.
Wrong, I dismiss nothing. I do take issue with "the sound" of a USB to SPDIF device being presented as evidence of the sound quality or not of low bit ADC/DAC devices, or as evidence that one does need DAC's anymore...
Scott, if I write completely nonsensical stuff, you have licence to correct me.
Ciao T
Scott, you miss my point. A USB to SPDIF Dongle outputs SPDIF Digital signals. If you connect these to a speaker there is no sound, you need to add DAC and Amplification, which are determining the sound.
Hence arguing that "A 29 USD USB to SPDIF Dongle makes DAC's obsolete" is like arguing that due to the high quality of Super Petrol cars are obsolete...
I know. But is was convenient to mention.
Wrong, I dismiss nothing. I do take issue with "the sound" of a USB to SPDIF device being presented as evidence of the sound quality or not of low bit ADC/DAC devices, or as evidence that one does need DAC's anymore...
Scott, if I write completely nonsensical stuff, you have licence to correct me.
Ciao T
Guys let's keep it respectful, offending post and response removed
You never miss a chance, sorry my fault. OK, we had discussed how much you can pay for a USB to SPDIF converter and I didn't bother to find out what the DAC was and quote a total. It was funny at the time, looking at the setup and seeing a piece of it dangling there.
So the USB to SPDIF doesn't matter?
Just FYI Scot and TL. The USB/SPDIF dongle feeds a slightly modified DCX2496 that acts as the DAC and crossover in SY's system. Nothing super high dollar.
Is that the correct signal path, SY? Or is there something else in the system?
Is that the correct signal path, SY? Or is there something else in the system?
Last edited:
Right. But as SY is deaf, that's probably all he wanted to spend.
He will soon leave Austin to found a hearing aid company.
http://www.nytimes.com/2011/10/18/business/edgar-m-villchur-hi-fi-innovator-dies-at-94.html?_r=1&hpw
- Status
- Not open for further replies.