Ted is there any kinds of distortion that some are pleased with hearing. I always wondered how much of what we like is the imperfections and making things better may introduce a sterile sound. In No way did I ever feel the ds sounds this way ever just curious.
People have argued about 2nd harmonic (or even harmonic) distortion being pleasing vs. odd harmonic being displeasing. That doesn’t seem entirely correct in my experience, but it’s a possible example for you. Torreys A is considered a little richer by some. Lately I’ve determined that it has a little more harmonic distortion than other releases so that might be a minor example for you.
In general inharmonic distortions (aliasing, FM modulation, noise modulation, …) are more audible and objectionable than harmonic distortions. Harmonic distortions can indeed more benignly add richness to sound.
Perhaps even more simply, reverb is often used to add richness to an otherwise perhaps sterile recording. But conversely I’ve found that lowering the noise floor allows us to hear the natural reverb that’s in a recording which can also add more reality to a recording.
Paul McGowan saidMust be a tough choice between the two. Can't you just flip a coin?Can’t promise the weekend but can tell you MQA is on for the full decode within the Bridge.
Arnie has found 'the one" and is falling over himself with excitement calling this Ted’s finest masterpiece - beyond anything yet done - I will hopefully have a chance to go through the two versions he’s narrowed it to today but that depends on a lot - I am just getting back, have one day at work to catch up, then off to the LA Audio show to be on the Legends panel.
Fingers crossed but breath not held.
Just read the newest Paul’s Post… so excited, and so very impressed that Ted and co managed to keep some of the specifics of this update a secret for so long. Can’t wait for the weekend!
[Here is the link to Paul’s Post. Elk]
Sweeter sound. 3 decibels (+18 ultrasonic) less noise. Tidal / MQA compatibility.
I wish I had purchased the Dad as opposed to the Jr., if anything to be able to get Huron tomorrow not next week! Ha.
Well done guys this sounds really promising.
dvorak saidYes, this was an awesome readJust read the newest Paul’s Post… so excited, and so very impressed that Ted and co managed to keep some of the specifics of this update a secret for so long. Can’t wait for the weekend!
Question to Paul;
Since there were 2 alternative versions of Huron shortlisted, did the non-preferred version excel on any areas compared with the preferred one?
I know this is ‘water under the bridge’ but it denotes that there still exist a potential for even further improvements.
Just read Paul’s daily post, ‘incredible’ !! A tour de force from Ted. Unfortunately, I wont be able to hear the marvel that is Huron until close to the end of the month as I am overseas, but I can at least read all the forum reviews… Well done Ted.
The part that surprised me the most in the announcement is the part about going from ten times DSD to twenty times DSD. I’m surprised the FPGA has the processing headroom. Very anxious to hear the improvement.
amsco15 saidThe part that surprised me the most in the announcement is the part about going from ten times DSD to twenty times DSD. I’m surprised the FPGA has the processing headroom. Very anxious to hear the improvement.
That was a surprise to me as well! I’m ordering an SD card with Huron from PS Audio as soon as it’s available!
Frode saiddvorak saidJust read the newest Paul’s Post… so excited, and so very impressed that Ted and co managed to keep some of the specifics of this update a secret for so long. Can’t wait for the weekend!
Yes, this was an awesome read
Question to Paul;
Since there were 2 alternative versions of Huron shortlisted, did the non-preferred version excel on any areas compared with the preferred one?
I know this is ‘water under the bridge’ but it denotes that there still exist a potential for even further improvements.
Each version Arnie tested had its benefits and detractors. Some had extraordinary bass, others midrange, still others top end. Only a few had all three and that’s what Huron wound up being. Lots of hard work on the part of Ted and Arnie to suss this out.
What size SD card do I need to use ? Am I still only able to use a 2 Gb ?
Thank You
I was very excited to read Paul’s post this morning! It sounds like the hits keep coming for the Directstream. Lower noise floor? Better bass? Tidal with MQA? Upsampling to DSD x 20?!! Dayum. Great job, Ted! (as always - what an asset for all of us) I can’t wait to hear it! Unfortunately, that will have to wait until Sunday since I’m going out of town on Friday! 
wonderful news… I can’t wait… The thing I absolutely love about this is that the ‘audio jones’ we all have … well this helps feed the beast!! 
I too - as all of you - was surprised at the specifics… Ted did mention something about altering things a bit in regards to algorithms… the 20X DSD… will be really curious about that… but it certainly appears as if the changes seem fairly dramatic… I like drama!!! haha…
Nothing has changed with the requirements of the SD card. If we change things so you can use a bigger card at some point you’ll still need to use the current sized/formatted card to get that update first
Like the stealth release of the DS Jr I don’t like to talk about things before they are in the can. A bit of background about Huron:
Richard Murison of BitPerfect had sent me some email and we were discussing possible sigma delta modulator improvements and it dawned on me that I had the code to simulate some of our questions/speculations and when I did I realized that I could eliminate the most problematic part of the upsampling chain since it was redundant with the type of sigma delta modulator I was using. Since I’d always gotten better results when I relaxed that final stage of the upsampler I knew that eliminating it entirely would be good in more than one way.
I had done a lot of the work to use fewer resources in Torreys: after eliminating the final part of the upsampler chain there was enough room to double the rate of all of the downstream processing in Huron. I knew that that would also be good for noise and indeed I got exactly the theoretically expected improvements in noise…
The work I’ll be doing in the next release also springs from further brainstorming with Richard.
If you haven’t been reading his column in Copper I recommend it highly. I also enjoyed his reviews of the DS enough that I read all of his other blogs which included that review. They cover a lot of interesting topics, among them upsampling, noise shaping, DSD, etc. but they are much broader than that: for example he has great music recommendations.
In case you missed it here’s the first of his four part review of the DS:
http://bitperfectsound.blogspot.com/2014/05/directstream-i.html
Here’s one of his recommendations that I enjoyed much more than I thought I might:
http://bitperfectsound.blogspot.com/2016/03/master-bates.html
Ted,
Thank you so much for allowing us to gain insight into your design and engineering thoughts and approach in creating Huron. The improvemdnt of Torreys over Yale was significant, and I can’t wait to listen to Huron. Your willingness to share your thoughts on its creation really makes owning my DSJ a positive experience.
Ted Smith saidRichard Murison of BitPerfect had sent me some email …
Wow, a shout-out from Ted Smith! An honour indeed
Ted Smith said Nothing has changed with the requirements of the SD card.This is a bit confusing. Longtime PSA owners, like me, got it firmly in their heads that 2GB was the maximum size card. I thought that this was true for the DS DAC when it first came out and that later it was upgraded to use larger cards. Reading the FAQ about SD cards seems to indicate that the DS has always used 8GB cards, and that is in line with Ted's comments. Is that right? But I still have my stash of 2GB cards and don't plan on giving them up.
I meant to say that the requirements on the SD cards in the DS hasn’t changed from the DS’s first release. A new bootstrap routine was required to upgrade from the PWD to the DS and that is the kind of thing that allows changes in the requirements of the SD card (or USB stick in the case of the DS Jr.) I’m sorry for any confusion I may have caused on that point. I don’t pretend to know the exact requirements on the SD cards and I defer to Dennis, etc. on the details.
I’m pretty sure I’ve used some very unusal SD cards on the DS. The past couple times I updated it, I was using 8GB and I think 32GB Micro-SD (from my Android smart phones) in a Micro-to-SD adapter. I don’t think I’ve ever used 2GB SDs. So if you have some SD cards, it can’t hurt anything to try them; you’ll only waste a little time if they don’t read.
Ted, question for you about this jump to 20x internal processing:
None of this effects the DS being able to accept more than DSD2X on the USB, correct? I ask, because I’m finding (as are many others in the various digital audio forums) that DSD2X seems to be the tipping point (or sweet spot) where the digital music really starts sounding more analog. I’m seeing trends of really good upconverting software supporting DSD4X and wondering if the DirectStream could ever get there? I know we touched on this a long while ago when the DS was still new, and it sounded like it might be possible, but we didn’t see why. Current trends have me thinking about it (and maybe trying out another DAC like the TEAC NT-503 in my second system).
I’ve not been around at all - it’s good to see lots of familiar names and excited talk on the latest DS OS. I’m really excited about this, Ted and Paul!
~Eric~
There are a lot of things that need to work together to support quad rate. The problem is that it’s way past what S/PDIF and AES/EBU (and certainly TOSLink) were designed to do. It’s also past what most I2S source devices were designed to do. Newer dac chips support higher rate I2S but most sources don’t do I2S well at higher rates. Ignoring the random details getting quad rate DoP over I2S is a real problem for many sources, quad rate Native DSD over I2S works on more devices (if they support Native DSD at all.) On the USB side the maximum of 1024 bytes / packet and 8000 packets / second put limits on the number of channels allowable and most computers sort of like 32 bits / sample (64 bits / stereo sample), an even number of channels and both record and playback channels, which all taken together -> ACK : something has to give for quad rate DSD, but what decisions each piece of software makes as to what has to give is pretty random. Once again native DSD takes about 1/2 the bandwidth of DoP so it’s more possible for systems that already do double rate native DSD.
So - lining all of the ducks up for quad rate involves new USB drivers, new USB code in the XMOS USB processor in the DS, new controller code in the DS, and cooperation of the playing software (what ever the user is using.) But the FPGA is much closer with Huron.