So in essence, both sides are correct or incorrect - the vinyl side "hear" that vinyl sounds better - the digital side "theorise" that digital is better. The real measure is in the sonics!
Many people use the term digital and RBCD interchangeably. RBCD is only one digital standard, there are many others. In fact even digital audio tape is recorded to a different standard. Digital resolution can be increased to any arbitrarily defined level, analog cannot. In fact the most important signals in the world are converted and transmitted digitally. For example those from the Hubbell Space Telescope and from deep space probes that travel to the outer reaches of the solar system. If we had to rely exclusively on analog signals, these systems would not work.
we are talking about CD as per the O/P as well you know!
OK so let's arbitrarily choose 23 bits. What timescale to design a 23bit ADC do you think? Over breakfast? 😎
<edit> oh and one more thing - the signals returning from space probes most assuredly are analog signals.
Last edited:
You still need to use 40 khz minimum. The level must change at least twice every 20,000th of a second. That's the time for any arbitrary change between two amplitude levels at the highest audible frequency. The number of bits per word and the required dynamic range will define how many levels there can be. If music has a dynamic range of 100 db and you have 2 to the 14th levels (16 bit word) in that range you have 16,000 available levels. If they are evenly spaced logarithmically that's a resolution of 1/160th db between any two levels or 16 times finer than the resolution JA claims he can hear and 80 times what I think I can hear at best. With 2 to the 21st levels in a 23 bit word that would be 32 times greater resolution. It is obviously unnecessary and overkill. 44 khz scanning rate leaves a 10% reserve margin in the time base. The system exceeds all possible combinations of audible change in the information channel a human can hear and no further increase will subjectively improve the results. The only way that that real world equipment for this system can be flawed is if it does not function as intended. A non linear A/D or D/A would be such a flaw but it would show up as severe non linear distortion in conventional analog tests. These may have been relatively common flaws at certain amplitude levels in early recorders and players. Generally like all digital systems they either function perfectly or they don't work at all. Audiophile tweaks invariably result in performance degradation whether someone enjoys it more or not.
Even that amazing device leaves me wanting, and I have spent enough time with it to have come to that conclusion as a choice. Just to twit SY, I still prefer op amps with ground plane ballast ala electron pools / Ground Control. I don't think the digital information format is the only culprit here, I think poorly handled grounds for op amp based buffers are a large part of the problem. Once dealt with I have no problem with Red Book audio or op amps, as you know and have heard in my system, except with the information thing.
Bud
You have, good sir, nailed it! I fact I have never lived at one address 5 years in my life - not ever. The average is a good bit lower. Lugging around all that vinyl is nigh impossible, but I have left a good trail of it across the globe.
Moving often keeps the clutter down a bit.
Interesting - largely the same as my own conclusions. Got any fun links about this to share? 🙂
<edit>
Pano said:
Yep, me too. I think Bud's on the right path, but then I would say that...😛
Last edited:
Interesting ideas Bud, as always.
I can't swear that too much information bothers me, tho it might. The better my system has gotten, the more emotionally taxing it is. I can only take so much good music per day. Reminds me of visiting the Louvre. All that beauty and splendor will wear you out!
The edginess tho does bother me. It's a common complaint against digital. I don't believe that most of that edge is the fault of the format, but the fault of its implementation.
I can't swear that too much information bothers me, tho it might. The better my system has gotten, the more emotionally taxing it is. I can only take so much good music per day. Reminds me of visiting the Louvre. All that beauty and splendor will wear you out!
The edginess tho does bother me. It's a common complaint against digital. I don't believe that most of that edge is the fault of the format, but the fault of its implementation.
Hi Bud,
It seems that grounding has become a fairly frequent topic in regard to digital lately, although you may have been the first to directly express it.
Best Regards,
Terry
The Berkeley DAC looks to be one of the few designs that's really taken grounding design seriously. Can't help thinking that's a major contributor to its excellent reputation. Incidentally, any news yet about the mods undertaken to the Weiss which you were going to look into? 😉
Lin and Larry Pardley were able to do that by living on a 23 foot sailboat for 10 (?) years, as they sailed around the world. Too much stuff and you'll end up swimming.
😀
Best Regards,
Terry
The DAC implementation is only part of the problem - what is fed to the DAC also has to be pristine or it's back to the digital heebie-jeebies!
The whole implementation chain on the digital side warrants much more attention than it currently seems to receive & it suffers as a consequence.
The whole implementation chain on the digital side warrants much more attention than it currently seems to receive & it suffers as a consequence.
The Berkeley Dac certainly had some first rate people that worked on it. 😀
I'll have to apologize, I was over at Genesis talking to Gary a little over a week ago, and I only remembered about the Weiss Dac after I left and was 25 miles down the road. I'm interested in finding out myself.
Best Regards,
TerryO
Indeed ! In fact the edginess can be removed from a CDP by carefully planning.
I did it in my CD53 🙂
One of my best resulting mods consisted in placing a dedicated clock on the servo chip and upgrading the driver´s PSU so the machine reads better and uses less the error correction system. Detail and bass are in order here.
It is interesting to know that a better mech works wonders in digital as well as a better TT works good on Analog.
Yes, that's what I'm talking about - to get things right in digital, implementation needs to be more extreme. It is the equivalent, as you say, of the turntable tweaking needed for best analogue sound. I'm convinced this is the lowering of jitter, others feel it is better RF treatment or grounding but whatever the ultimate reason, it would seem to be needed. Then there are others who will say this is all rubbish , "bits is bits" & on a "properly implemented" system digital is perfect. What they are unable to cite is a properly implemented system - QED 🙂
Last edited:
Well... IMO it is much more difficult to implement a good digital solution than an analog one. But a top notch analog system is much more expensive. (Just look at the prices of a good MC cart 🙂)
Agreed. It is much cheaper to get to great digital sound than to get great analogue sound i.e the digital parts are cheap compared to the precision engineering needed for analogue parts but the know-how seems to be lacking in digital design (flame repellent jacket now being donned 🙂)
A transformer output stage on the DAC does it for me. Seems to cure a lot of the nasties. I've even measured it - typical opamp stages tend to have a lot of high harmonic noise. Transformers don't.
Certainly there are other ways, but transfos do it for me.
Certainly there are other ways, but transfos do it for me.
- Status
- Not open for further replies.