Audio Note Kits support forum, in co-operation with Audio Note UK
Return to Audio Note Kits Forum
65.19.17.69
In Reply to: RE: Can kit DACs handle 24/96? posted by Paul Folbrecht on May 28, 2009 at 09:12:50
Contrary to what you might read below, the AD1865 is 18 bits, not 16. If you record 16-bits you won’t be using the full resolution of the AN DAC. Recording 24-bits gives you the max resolution you can use today and more for what you might want in the future.
The maximum sample rate is determined by the specs of the digital interface receiver chip. The CS8412 is 48K, the CS8414 is 96K, and the CS8416 is 192K. Upgrading to the CS8414, if it’s not already fitted, is easy...just plug it in. Upgrading to the CS8416 will require some additional logic and a 3.3V power supply.
BTW, I've got the 8414 in my unit. (Heh heh... unit.)
Might as well record 192K and plan on a higher res chip sometime.
And actually I hear Brian/Andy have some top-secret higher-res digital mods possibly coming...
Think my problem might be the USB. That would stink, as the USB transport is truly excellent! And cheap! My used Mac Mini was under $500.
Most humans cannot hear beyond 20KHz. Most microphones and loudspeakers don't work beyond 20KHz. Recording at 192K makes no sense. In theory, all you need to be able to record any and all frequencies up to 20KHz is a 40K sample rate. In practice you need a little more head room to relax the requirements for the anti-alias filter. It is generally agreed that a 60K sample rate meets all the requirements but few audio DACs work at the rate. 88.2K also meets the basic requirements, is handled by most DACs, and creates smaller files than 96K, which is the next best. Recording at 192K provides no more resolution for any signal in the audio range and creates huge files. It also makes more demands on the digital interface and the DAC.
For myself, I'm ripping my CDs at 44.1K/16 and I'll be digitizeing my LPs at 88.2K/24. For playback, the CD files will be resampled to 88.2K/24, on the fly, using a FIR-less interpolator and the LP files will be passed unaltered to DAC, which will be running with a fixed master clock. The PC sound card is synced to the DAC's word clock.
That's the plan, anyway. I'm just now getting started. My silent PC front end is ready with a 2TB NAS. All I have to do is finish the interpolator, build the new DAC, and rip 1000+ LPs and CDs.
jml, you must be aware that there are plenty of people that disagree with your theoretical take, right? I don't feel qualified enough to take a strong view, but there are plenty of knowledgable people who believe 192K sampling gives greater fidelity than 96K. It's certainly *widely* believed that 96K is audibly superior to 44.
Why?
- The Nyquist theory deals with continuous waveforms only - not stops and starts as occurs in real music. This point is critical.
- It is known that reproducing frequencies above 20 Khz has effects in the audio band, via intermodulation and by preserving transient fidelity. You are aware that super tweeters exist, right? (I don't use any myself and can deal with 3 dB down at 17 Khz...)
I will probably also record at 96K - though I'd like to try both.
And furthermore, my AN DAC at 16/44 is thus far the best digital I've ever heard - though I have no experience with hi-res PCM yet!
My point is just really that it is not nearly as cut & dried as you made it sound IMO.
Just because there are plenty of 'knowledgeable people' that disagree with me, I must be wrong. That’s the story of my life.
First of all, it is critical to understand that I was talking about the RECORDING sample rate, not PLAYBACK. The criteria for each in regards to audible fidelity are very different. There is no doubt 44.1K/16 is insufficient for both recording and playback. It is my contention that 88.2K/24 is sufficient for RECORDING LPs. Other than ticks and pops, there is no supersonic content of any consequence on an LP. The typical phono cartridge has enough trouble tracking 18KHz let alone 20KHz and beyond.
Nyquist theory deals with ALL waveforms. It is the practical implementation of the theory that screws things up. At every step, sampling theory demands band limited signals but, in practice, perfect band limiting it very difficult to achieve. In theory, 44.1K is sufficient to record all audio up to 20KHz but the filters necessary for the prefect band-limiting required by the theory have audible side effects. In the lab, with steady-state sine waves, 44.1K measured perfectly but not so with music due to the phase shifts and other anomalies introduced by the brick-wall analog filters.
That realization led to resampling and digital filters. With a digital filter, each output sample point is computed from a weighted average of the nearby input samples. The weighting coefficients determine the characteristics of the filter. In theory, digital filters are perfect and, in the lab with steady-state sine wave, they measure perfectly, as well. But music is not steady state and the averaging effect of digital filters takes its toll, such as with transients and every time there is a change in frequency.
The current state of digital audio means choosing from a spectrum of audible poisons. Audiophiles with a low tolerance for digital filter poison choose NOS and no filters. Others with a low tolerance for NOS poison choose resampling and digital filters. The reason many prefer hi-res recordings is they get the benefits of high sample-rate playback without the poison of resampling and digital filters. The cost is a huge dataset. Pick your poison.
My ideal is to start with lowest practical recording/storage bandwidth, 88.2K/24 for LPs and 44.1K/16 for CDs, and resample to a higher rate for playback without using digital filters. I have been working on this for many years and I now think I have the technology to do it. The first step will be to resample CD files to 88.2K/24 on the PC and feed both LP and resampled CD files to a custom DAC that will resample an additional 1x to 32x using both digital and analog techniques. The second step will be to resample both LP and CD files to 176.4K/24 on the PC and feed the same DAC. I suspect the poisonous effects of digital filters are most noticeable at lower sample rates: That’s why I think I can get away with using traditional digital filter to go from 176.4K to 705.6K and beyond. We shall see.
Mr. JML,
I have no idea who you are but your knowledge of things digital does seem to eclipse my own. If you read what I wrote to you previously I was only pointing out that there are those with completely different conclusions than your own. (It sort of seemed you might not have been aware of that, but obviously you are.)
I have heard a lot of digital setups and I can assure you that my AN DAC fed by the Mac Mini sounds divine - by this I mean not only pleasurable, but realistic in term of timbre, nuance, and dynamics, with no hint of 'digititus' at all. It has been my understanding that jitter is thought to be responsible for that hard, unpleasant digital sound, and since this is lacking in my setup, I have to conclude that it is not high-jitter. (While I may be relying on experience over theory, you do seem to be doing the opposite: have you listened to an AN DAC fed by a USB transport?)
I am open to the idea of using a physical disc transport as I did indicate. In fact, can you answer this for me: Is there some other way I could get 24/96 into my DAC? The chip will truncate to 18 bits but will take a 96 Khz stream, but the Hag USB interface will only take 48 Khz. Until AN Kits has their new USB interface there is no way to do it with USB that I know of. But there is no disc transport that I know of that could, say, take 24/96 AIFF files (such as those produced by a Masterlink with its CD24 burns) and transfer via SPDIF.
Follow-up: I use USB exclusively (Mac Mini) - does USB support 96K? I seem to recall "some" USB devices only supporting 48K...
I don't use USB. It's gotta be the worst possible interface for digital audio. The clock is a VCO phase locked to the 1ms USB heart beat, the is no data integrity, and the specified jitter is +/- ONE SAMPLE!!!
I think again your theory is a little too invasive here - and probably not correct.
Gordon Rankin of Wavelength, whom I believe has a PhD in math, is a huge proponent of USB as a transport interface. Have you read his stuff?
I admittedly have not tried my AN DAC with a SP/DIF transport, but digital with the Mac Mini is soo good I really cannot envision that I am experiencing serious jitter. Admittedly, that's neither rigorous nor conclusive!
I have read Gordon’s rants and I have talked with him. He makes his living selling USB DACs. I’ll say no more.
So, you think a Mac Mini and a USB-fed AN DAC is the ne plus ultra. In most aesthetic pursuits there will be plateaus where, “It doesn’t get any better than this!” and then you experience something better. I think most audiophiles have had that experience at least once in their journey.
I am convinced USB audio was conceived with MP3 quality playback in mind. With USB audio, the PC is in total control: So much so the DAC doesn’t even know what the desired sample rate is. Every millisecond the PC sends the DAC a packet that contains all the samples to be converted during the next millisecond. It’s up to the DAC to adjust the sample clock accordingly. According to the USB audio spec, the PC can change the sample rate at any time. In fact, that happens many times a second when the target sample rate is 44.1K. That sample rate is not evenly divisible in 1ms intervals so some packets will have to have an additional sample to make up for the shortfall in previous packets. That’s where the 1ms granularity comes from. Although the chip makers have tried to minimize the effect with different PLL schemes, it still comes down to a VCO phase locked to the 1ms USB heartbeat using a divisor that changes many times per second. Many people don’t hear it just as many people aren’t bothered by the frame-rate flicker of motion pictures. On top of that, there is no error correction and only rudimentary error detection. If the computed CRC doesn’t match the CRC in the packet header the entire packet is discarded and the DAC is mute for the next millisecond. Given the inherent reliability the digital data transmission, it is a rare event, but it does happen.
I suspect you are experiencing a lot of jitter. Even if you know what jitter is you probably don’t know what is sounds like. You have probably heard from 'knowledgeable people' that jitter is bad, yet most audiophiles, given a choice, prefer more jitter rather than less.
If you’ve spent any time around audio forums where DACs are discussed you have heard about asynchronous reclocking. That’s where the ‘high-jitter’ S/PDIF-derived clock is asynchronously reclocked by a much faster, ‘low-jitter’ clock. The proponents claim significant benefit, which they attribute to reduced jitter. But asynchronous reclocking increases jitter. That’s because the time period of every sample clock is extended to the next edge of the reclocking oscillator.
Some years ago, dCS, the purveyor of ultra high-end digital audio components, introduced the $7K+ Verona Master Clock. Not only was this clock more stable and more accurate than any other, it included a unique feature, dither. That’s right, with the flip of a switch, dither was added to the ultra low jitter clock output and the “golden ears” who reviewed it all preferred the dithered clock. According to dCS, “…it is better to add random noise to the clock, continually forcing [the PLL] out of the dead zone (and allowing the PLL to better do its job) than it is to further improve the stability of the clock…” By “dead zone” they are referring the period when the PLL is locked on frequency and further adjustment of the VCO is not necessary to remain in phase with the master clock.
In the three instances I cited, a PLL is forced to constantly adjust the VCO and hunt for the correct lock. The ‘experts’ say the sample clock should be absolutely stable and without jitter, yet audiophiles seem to prefer, or, at least, not complain, when the sample clock is not stable and contains significant jitter, dither, or other variability. That’s why I put no stock in what ‘experts,’ ‘knowledgeable people,’ and ‘audiophiles’ have to say but only rely on my own knowledge and experience.
One comment: Gordon makes many products besides DACs and it's quite clear hat he chose USB interfaces for his DACs over other methods because he considers it superior. You may disagree, but to insinuate he's not an objective authority on the matter isn't very logical IMO.
(The Wavelength DACs are nice... and my AN Kit 2.1C runs circles around the Cosecant at 60% the price.)
On second thought, the positive qualities you attribute to USB are misapplied because you are not really listening to a pure USB transport. The auxiliary circuit board you use is translating the USB to S/PDIF and that is what is driving the AN digital receiver. In other words, the recovered clock is passing through two PLLs. Perhaps the S/PDIF PLL is ameliorating the damage inflicted by the USB PLL and that’s why you don’t hear what I hear.
It was quite some time ago when USB audio devices appeared and I was intrigued. One of the first USB DACs with audiophile pretensions was the Stereo-Link. I have one. Technically, it does a lot of things right. Separate AC power supply, RCA I/O instead of a mini phone plug, and galvanic separation of the PC and audio grounds. Unfortunately, it sounds like a USB DAC. At about the same time Wavelength introduced their first USB DAC and Gordon was making extraordinary claims about it. I contacted him for clarification and he apparently mistook me for an audiophile. What he told me was total BS. After I followed up with technical challenges to his most outrageous claims he clammed up saying his methods were proprietary.
DACs are my passion. I have been working with computers and audio since 1973: Before PCs, before CDs, and before it was even called digital audio. I think I know a little more about the subject than the average audiophile. As I see it, Gordon saw an opportunity to be get into the immerging, ‘audiophile approved,’ digital audio market and took it. As a method for transporting audio samples from one device to another USB offers no technical benefit. In its current form, as you well know, USB doesn’t go above 48K without proprietary chips and drivers and, as I’ve described, it has poor clock stability and data integrity. USB’s only selling point is plug’n’play simplicity/portability and that apparently trumps all other considerations for members of the iPod generation.
Have you listened to an AN DAC fed by a USB transport?
Why would I want to do that? What exactly would I learn that I don’t already know? I have listened to quite a few USB DACs. In fact, from where I am sitting in my office I can see three, ‘audiophile approved’ USB DACs. I don’t like the way USB puts its mark on the recovered music. In that regard, it’s worse then the S/PDIF. If you don’t hear the effects of USB or it doesn’t bother you then enjoy your USB-based digital audio system. Why do you want my approval?
I have an Audio Note 1.2 DAC kit in my audio system. I like it because it doesn’t offend. Aside from audio shows and dealer showrooms, the only other AN DAC I’ve properly auditioned is the 3.1x. After reading rave reviews and glowing owner testimonials regarding the 3.1x I thought it would be a nice addition to my growing DAC collection. I contacted my local AN dealer to arrange a demo. Eventually the demo was arranged but the AN distributor, who supplied the equipment, insisted I listen to the DAC with the AN CDT and the hideously expensive, AN digital cable. In other words, I had to immerse myself the total Audio Note Digital Audio Experience. What a joke. If your AN DAC sounds anything like the 3.1x, you can have it.
Finally, as long as you cling to USB as the savior of all digital audio then you have to accept its limitations. If you want a plastic disk transport then you have to accept its limitations. If you are willing to use the dread, dare I mention its name, S/PDIF from a PC sound card, then there is no reason you can’t supply your DAC with 192K/24. My AN 1.2 DAC is fitted with a CS8415A that actually plays music with 192K/24 input even though the chip is only rated to 96K. Truncating to 18 bits is no big deal because the linearity of most R2R-type DACs falls off after 18 bits. (Purportedly heard from a Burr-Brown engineer – “Question: What the difference between the PCM1702 and the PCM1704? Answer: Four marketing bits.” In case you don’t get the joke, the PCM1702 is a 20-bit DAC and the PCM1704 is a 24-bit DAC.)
Hi JML. I get the impression that your modified AN DAC1.2 sounds better than the factory AN DAC3. If it may not be too presumptuous, may I ask whether there was there anything in particular you did? modified? aside from the higher grade receiver chip. TIA
It's not that my 1.2 sounded better; it's that the 3.1x I heard exhibited a quality I cannot abide. After playing for only 30 seconds it went back in its shipping box never to heard from again. That was very unfortunate because I was expecting great things from AN in that price range. It was especially unfortunate for the dealer who acquired all the gear and drove 140 miles to my place. We spent the afternoon comparing transports, digital cables, and LP vs CD. I could not hear a difference between the $3K AN digital cable and my $0.15 wire. The same with the CDT. On the other hand, LPs were a clear winner.
BTW, I hear the same defect in other DACs that omit a proper reconstruction filter.
I vaguely recall that the AN DAC1 used a low pass LC from the DAC Chip to the grid of the output tube, while the AN DAC3, cascaded transformers. Could that matter?
As with so many things in audio, there are apparently significant reasons why they should not work well. Examples of apparently 'seriously impaired' technologies include transformer coupling, resonant (ported) speaker enclosures, SET amplifiers, playback via a needle in a plastic groove, 1x oversampling, no digital filtering, corner speaker placement, etc. Yet, properly done, they can and do work remarkably well. Such is the case with the USB interface for digital audio playback. In my music system it works very well indeed.
Bob
While I prefer many of the 'seriously impaired' technologies you mention, I have tried USB for digital audio and found it unsatisfactory. What I find most annoying is its 1ms granularity. If you don't hear it or are not bothered by it, consider yourself lucky.
Sounds like you got pretty bad results using a USB DAC in your system. Not sure what hardware you were using, but there is no sign of granularity (especially as coarse as 1ms)in my system (DAC Kit 2.1 Level C Sig., Mac Mini, 45 SET, Altec 15" 2-way). I have heard grainy sound in other systems, particularly those using solid state amplification (not all though), inexpensive digital or cheap interconnects. I am not sure how you would verify the cause of such granularity as the USB interface or measure it. Perhaps it was due to component interaction or to a coarse sounding dac? In any case, not all digital front ends using a USB interface sound grainy.
Bob
If you don't hear it, you don't hear it. To me, all USB front ends share the same characteristic sound, which I don't like. You, apparently, think USB is the greatest thing since sliced bread. Good for you. Why do audiophiles think everyone should hear what they hear and like what they like? The fact that I don't like USB means my system is not revealing enough, or I use cheap interconnects, or SS amplification, or my hearing is not good enough, or any of a thousand other reasons why my opinion is wrong.
As for verification, the fact I hear the same thing in all USB DACS tells me the problem is with USB and not anything else. I wrote a Winamp DSP plugin that varied a number of things in 1 ms intervals that I thought could be contributing factors and came pretty close to recreating the USB experience with a non-USB DAC.
"To me, all USB front ends share the same characteristic sound, which I don't like."
So you have heard all (or at least many) USB DAC's, each in a variety of systems, and they all exhibited this grainy characteristic in all of those systems?
"You, apparently, think USB is the greatest thing since sliced bread. Good for you."
Ok, so I like the sound of my system with a USB DAC. I have heard better, though not near this price range. Still, others may prefer something else. You clearly prefer the sound of non-usb dacs. I have no problem with that.
"Why do audiophiles think everyone should hear what they hear and like what they like?"
Conversely, why should all audiophiles dislike the same things you do? Actually, if I heard what you describe, I would dislike it.
"The fact that I don't like USB means my system is not revealing enough, or I use cheap interconnects, or SS amplification, or my hearing is not good enough, or any of a thousand other reasons why my opinion is wrong."
No, I said that when I hear a system with grainy sound it is often asasociated with ss amplification, inexpensive digital front ends or cheap interconnects. The fact that the grainy sound is not present in your system when a non-USB dac is used suggests that the cause of the grainy sound is NOT one of these. The fact that you hear this grainy sound suggests that you are a critical listener and your hearing is just fine.
Your observation (grainy sound) is not wrong. No doubt you are hearing this grainy character in the sound of your system when using a USB DAC (and evidently in some other systems as well). IME, people differ substantially in their sensitivity to various colorations. You may actually perceive this grainy sound in every USB dac you try, and I might not hear it at all. Perhaps it is just a matter of differing perceptions (people are not equally sensitive to the same sonic issues).
The conclusion that the USB interface will cause this (grainy sound)to be audible to you in all systems, may be correct. I don't know. I just think it's a big leap to think that all USB dacs will exhibit this grainy characteristic in all systems, AND that all people (or at least audiophiles) will hear this same fatal flaw, as you imply.
The USB interface may be fatally flawed (sonically) to you in all systems, but that does not make it true for everyone. A lot of people are enjoying music reproduced through them. Are we all deaf or listening through systems that lack the resolution necessary to hear what you do? I remain convinced that a good USB interface DAC and a computer/music server/transport currently offer remarkably good preformance and value for many (but not all) people.
There is a lot of equipment to choose from and we are all free to choose what sounds best to us. Enjoy!
Paul,
According to the AudioNoteKits Update post (this forum) dated April 22, 2009:
"Working on a new add on board for the DAC series to support the USB hi res download formats! - Expect something in June time frame"
This would evidently be an upgrade (replacement) to the currently used Hagtech USB board.
Are your Kit 03 speakers running yet? I guess you are now usng a 2A3 SET instead of the Yamamoto 45. I would be very interested in your impression of the performance of the 03's with either of these.
Regards,
Bob
And I thought I had the scoop!
Awaiting new speaker kit. Have the cabs. Really enjoying Supravox field coil drivers on OB right now.
I picked up a pair of Wright 2A3s. The ability to drive into A2 for 9W peaks is really a boon. I felt they were the equal of the Yamamoto in most ways, and more truthful tonally. And they are a steal.
Some day I'll have to try the parallel 2A3 AN monos.
Thanks a bunch. So, if I record 24 bits, the DAC 2.1C will truncate to 18 but that's Ok. It can handle 96K if I have the 8414 (and if this is the current chip I probably do since I bought my kit ~6 months ago).
Now I just have to figure out the best equipment/method to use for A to D.
Post a Followup:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2009, Audio Asylum, all rights reserved ]
[ Go Green ]
[ Take5 ]