The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Custom DAC - Good Design?
Old 8th December 2016
  #31
Lives for gear
 

Yes, 192k in, out and trough, selected by end-user. Irrespective of ''internal'' rate, as in some converters. Personally I think sound quality is best at 192k but i also have a suspicion that the src still kills the signal by a small degree (at 192k). Am curious to hear your expert assessment.
Old 11th December 2016
  #32
Lives for gear
 
Casey's Avatar
 

Yes the extra filter that the SRC imposes is audible, but here is my take. For studio recordings you won't hear it.

In the case that you are recording jazz or orchestral performances in a good acoustic space, I would hear a similar filter on PCM quite easily compared to a raw DSD recording.

So for example if I take a raw DSD recording of an orchestral performance and then convert that to 384khz for editing in Pyramix, then convert the 384 back to DSD, the two files are very different. Now the only thing that would cause this difference is the 384khz filter used in the conversion to PCM.

The difference between the files is obvious. The differences are all related to the sense of space and soundstage. The raw DSD represents the acoustic space that is captured in the recording accurately. The sense of accurate space and soundstage is diminished significantly in the file that was converted to PCM and then back again into DSD.

Again this difference is obvious to me, but it is related to the sense of space. This never exists in a studio recording in the first place so the difference would not be heard in a studio recording.

It is a can of worms but somehow the very slight smearing of a 384kHz filter is enough to alter the timing of the sound in a way that to me is quite damaging.

So yes a 192khz filter is audible, but not in a studio recording.

Sorry to run on, and this is all IMO of course, but you did ask.



-Casey
Old 12th December 2016
  #33
Quote:
Originally Posted by Casey View Post
It is a can of worms but somehow the very slight smearing of a 384kHz filter is enough to alter the timing of the sound in a way that to me is quite damaging.
-Casey
Others have found similar effects. Seems the transient signals above the human hearing range offer some location cues and depth perception that is not fully explored yet.

The January 2017 issue of The Absolute Sound has an interesting article tiltled: "Could This be Audio's Final Frontier"? The concept of "infinite bandwidth" is explored with discussions of Haas and "leading edge" effects.

A practical example using the EnigmAcoustics Sopranino super tweeter is explained. This $4995 horn loaded electrostatic super tweeter goes up to 80k hz. When run in parallel with ordinary speakers it adds a more accurate "leading edge" to the sound because the arrival time of musical information is faster. This adds to the sensory perceived "depth" and "location" of the sounds.

The Haas Precedent effect says if two similar sounds arrive at our ears at the same short period of time, they will be perceived as a single sound and the one that arrives first will be dominant in determining the perception of location to the sound's source.

Using wider bandwidth audio gear can help in preserving these location cues. That is one reason why I try to maintain a 2~200k hz bandwidth in analog gear, to preserve the location cues while avoiding phase shift in the audio band.
Old 12th December 2016
  #34
Lives for gear
 

Quote:
Originally Posted by Casey View Post
Yes the extra filter that the SRC imposes is audible, but here is my take. For studio recordings you won't hear it.

In the case that you are recording jazz or orchestral performances in a good acoustic space, I would hear a similar filter on PCM quite easily compared to a raw DSD recording.

So for example if I take a raw DSD recording of an orchestral performance and then convert that to 384khz for editing in Pyramix, then convert the 384 back to DSD, the two files are very different. Now the only thing that would cause this difference is the 384khz filter used in the conversion to PCM.

The difference between the files is obvious. The differences are all related to the sense of space and soundstage. The raw DSD represents the acoustic space that is captured in the recording accurately. The sense of accurate space and soundstage is diminished significantly in the file that was converted to PCM and then back again into DSD.

Again this difference is obvious to me, but it is related to the sense of space. This never exists in a studio recording in the first place so the difference would not be heard in a studio recording.

It is a can of worms but somehow the very slight smearing of a 384kHz filter is enough to alter the timing of the sound in a way that to me is quite damaging.

So yes a 192khz filter is audible, but not in a studio recording.

Sorry to run on, and this is all IMO of course, but you did ask.



-Casey
thanks for that, much appreciated. i actually compartmentalize audio into PCM and 'real' audio, so when i posed my question to you in my previous post i was thinking only within the context of pcm audio, the effect of SRC. pcm to me is an ADSR mangler that cannot be switched off, a timing funny mirror.

so any chance of a Bricasti 8 channel DSD da? I've been thinking about stacking several of the existing units but alas i can't afford it.
Old 12th December 2016
  #35
Lives for gear
 

Quote:
Originally Posted by Jim Williams View Post
Others have found similar effects. Seems the transient signals above the human hearing range offer some location cues and depth perception that is not fully explored yet.

The January 2017 issue of The Absolute Sound has an interesting article tiltled: "Could This be Audio's Final Frontier"? The concept of "infinite bandwidth" is explored with discussions of Haas and "leading edge" effects.

A practical example using the EnigmAcoustics Sopranino super tweeter is explained. This $4995 horn loaded electrostatic super tweeter goes up to 80k hz. When run in parallel with ordinary speakers it adds a more accurate "leading edge" to the sound because the arrival time of musical information is faster. This adds to the sensory perceived "depth" and "location" of the sounds.

The Haas Precedent effect says if two similar sounds arrive at our ears at the same short period of time, they will be perceived as a single sound and the one that arrives first will be dominant in determining the perception of location to the sound's source.

Using wider bandwidth audio gear can help in preserving these location cues. That is one reason why I try to maintain a 2~200k hz bandwidth in analog gear, to preserve the location cues while avoiding phase shift in the audio band.
but if one looks at pcm as a filter, the sound is very different in pcm as compared to just an analogue filter, running short cables with high integrity signals?
Old 7th February 2018
  #36
Here for the gear
 
Alessa's Avatar
 

Quote:
Originally Posted by David Rick View Post
Meanwhile, the ASRC camp has pushed back, because ASRC chips have gotten better. Crane Song's Dave Hill is the latest convert to this approach. His latest-generation Avocet controller uses ASRC to a fixed 207kHz sample rate.
So the Crane Song Solaris uses systematic oversampling to 207khz, could it be the main reason of it's sonic excellence?

And would systematically oversampling at 768Khz produce even better results in DA conversion?

Maybe the RME ADI-2 PRO makes it possible maybe to oversample everything to 768Khz...
Old 7th February 2018
  #37
Motown legend
 
Bob Olhsson's Avatar
 

My understanding is that filters are the issue and not bandwidth. ASRC introduces latency that is a problem for people not using a recording console.
Old 7th February 2018
  #38
Quote:
Originally Posted by Alessa View Post
So the Crane Song Solaris uses systematic oversampling to 207khz, could it be the main reason of it's sonic excellence?

And would systematically oversampling at 768Khz produce even better results in DA conversion?

Maybe the RME ADI-2 PRO makes it possible maybe to oversample everything to 768Khz...
Historically, oversampling had benefit because it reduced the need for high-quality analog anti-aliasing. That's a CD-era story though. More recently, there have been arguments that higher sample rates yield less pre- and post-ringing, but not everyone agrees the difference is audible.

The main advantage of using ASRC like Crane Song (and Benchmark before them) is that it allows the DAC to run from a fixed-frequency clock of the highest possible stability. No PLL can ever match the phase noise performance of a great non-pullable oven-controlled frequency standard. The ASRC processes itself is a potential sonic detriment, although ASRC chips have gotten a bit better over the years. Still, you're doing real-time interpolation of filter coefficients based on clock timing that can't be known precisely. But overall, the sonic balance sheet may come out on the positive side. Certainly Dave Hill and John Siau both think so, as do many of their customers.

The fact that RME can work at 768 kHz means that they are using a different DAC chip than the previous two products. It could be either better or worse, independent of the sample rate at which you run it.

Trust your ears, not the internet!

David L. Rick
Seventh String Recording
Old 10th February 2018
  #39
Lives for gear
 
Adebar's Avatar
Quote:
Originally Posted by David Rick View Post
But overall, the sonic balance sheet may come out on the positive side.
asynchronous SRC is less easy to make than synchronous and it has to be very good that in the end the balance is positive.

As always it is all about implementation.

The other approach ist to avoid ASRC and achieve a very stabil clocking via good PLL. Keeping the clocking lines clean and damping distortions also helps.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump