And oddly Mr. Nirvana has what appears to be a lovely Onken cabinet, or Ultraflex as his avatar. The sort of speaker that vinyl lovers love. Strange.
What is ridiculous about tubes, exactly?
Also,could someone enlighten me about why a digital piano sounds so utterly unconvincing no matter what the price?
There are some pretty good electronic pianos out there, but you are correct, they don't really sound real. Is it the samples, the speakers, the amps?
I suspect the speakers. It might be interesting to hear what a good sampled piano sounds like with decent room reverb played on your home hi-fi. How close would it be?
I suspect the speakers. It might be interesting to hear what a good sampled piano sounds like with decent room reverb played on your home hi-fi. How close would it be?
I think its all of the above. I used to have a Clavinova but upgraded it to an acoustic piano. Have not had any desire to return to a digital piano. In theory I think its possible to make a digital piano sound lifelike. The limitations of the Clavinova I had were in dynamic range - the note's die-away became grungy - and polyphony. 16 note polyphony was not enough - when using the sustainiing pedal, I heard older notes disappear when I played new ones. I also think the modelling of how a piano actually operates was too simplistic - simply sampling individual notes gives no interaction between them. Notes certainly interact on a real piano. The way that the touch of a key was turned into the note's amplitude was also not like a real piano - I'm guessing simply sensing note velocity doesn't do justice to how a real piano works.
I suspect that sampling of individual notes may be done well but that the chief glory of a piano is the complex interactions in the soundboard when the sustaining pedal is used and this may be quite a daunting thing to even approximate digitally in the sense that the variables are endless. I am sure you are both right however and if the samples were played through the kind of nutty equiptment we use it would be a whole lot better. I recently heard a recording of a digitally sampled organ; simpler basic waveforms on each note and no soundboard or sustaining pedal-it was really quite convincing!
Ha! you beat me to it abraxalito.
Ha! you beat me to it abraxalito.
I agree with you piano3 - the challenges of digitally sampled organs are quite different. I have played one and I found it much more convincing than my digitally sampled piano. The limitations with an organ are definitely in the speakers and amps not in the signal processing and sampling.
Yes, I agree, I considered replacing 'sound better' with 'sound more like the master tape'. However, I did go on in my post to make an exception for when the “audiophile” is not a true audiophile and actually prefers higher levels of noise and distortion. My understanding is that in unsighted tests music listeners generally prefer lower noise and distortion and flatter frequency response, unless their hearing is damaged. That is why I used the phrase 'sounds better': even though the word 'better' is subjective, it has been tested and the shock finding is it is NOT just a matter of taste, in fact more accurate reproduction is preferred. So I put my heart in my mouth and left the word 'better'.
Werner, you are a hero for patiently helping to educate people who don't want to know the truth. I don't think I could do it! I suppose you realise that if they accept your entirely factual teaching they will have to change their opinions? What chance? 😱
Like I said, an amazing thread, which by rights could have ended with the succinct post #23 by krachkiste: http://www.diyaudio.com/forums/anal...better-turntable-compared-cd.html#post2304847
Please think about that statement, and then ponder it again. Really well put, that is exactly at the heart of the sonic mystery (in my humble, possibly ill-informed opinion).
The variables are indeed almost without end. Now go sample them all, and model those interactions that you cannot sample... 😱
--
Complete agreement here! We've found everything stated above to be true as well.
Best Regards,
TerryO
If by 'sonic mystery' here you're meaning the original topic of the thread, I'd say that's another issue entirely.
Here the limitations are not in being digital, they're in having inadequate and incomplete models for how pianos generate the sounds they do. When the models and therefore by extension the algorithms aren't complex enough the sounds won't be similar enough to a real piano. Same argument applies to lossy compression (mp3s and the like).
I find I get the best results when simultaneously running a program which calculates the prime numbers. It does something to the jitter you could describe as noise shaping, it inserts more randomness into the jitter.
I just hope my program never runs out of prime numbers, because I don't quite believe the proof that there are infinitely many.
Kenneth
😀 Excellent - reminds me of when Captain Kirk had to give the ship's computer the full expansion of PI to distract it from alien influence.
So you reckon that skimping on a proper SPDIF interface feeding your DAC doesn't matter ? Or the generic SMPS in the el cheapos doesn't matter either ?
I doubt that you really believe that, because you like to look at waveforms.
I would be very surprised if you haven't compared the waveforms from different level players, and seen the nasty influence of the SMPS residual too,
compared with a player using a linear PSU. However you aren't going to get a new player using a linear PSU for $15 to $25 , are you.
BUt, bits are bits, aren't they, and an el cheapo CD player sounds the same into a decent DAC. as a $4000 player.
An externally hosted image should be here but it was not working when we last tested it.
😀
O.K. Richard, enough of my mischief, have a good day !
Alex
I'm designing a DAC and would like the signal source to be fairly poor. I don't want to design a DAC that only works well with high-end transports, rather to me a good DAC needs to be relatively transport immune.
So to answer your question - with a competently designed DAC, that's the ideal yes. I don't know how closely I can approach that ideal with my own design.🙂 I'm of the view that all transport problems can be fixed in the DAC (assuming that the data isn't corrupt). My design is a way to try to falsify my view on this.
Sure it does matter with most DACs, including the performance I'm getting currently with mine. But its not finished yet... Incidentally, I was intending to mention the PSU on my blog, I'm a bit overdue for a posting there. Its got quite a nice new chip in it.
You might be surprised to learn I found out about RF entering my DAC by listening, not by looking at its output waveform.😀 When I use my scope, I really don't know what's relevant on the waveform for getting the sound I want, which is freedom from glare, and some...
Right. I'm a modder - I'd have no mods to do if the kit I bought was already perfect. So that's one of the things I've been thinking of experimenting with, changing the PSU.
Is this parody of the 'objectivist' position? I'm a bits-is-bits person for sure, but bits ain't the whole story, not by far. If someone's paid $4000 to get the same data as my $15 player, that's an awful lot of money for improvements in RFI. So I'd like to do that a bit cheaper.
Oh, but I appreciate your 'mischief', all good questions you're asking. Its a sign you're by no means a has-been Alex 😀
All good answers too Richard. I hope you achieve the results you are looking for with your design. I would like to see more of what you come up with!🙂
Regards
Alex
Regards
Alex
Then go ahead. I've done my share of things. You may want to spend
a moment or two thinking about the ramifications of band-limiting that
wobbling signal close to Fs.
True enough, and even so for SACD.
And all too often the upsampling is even done in a most incompetent way. Sinatra At The Sands '192kHz' is a prime example of this. It must have been an enormous effort, botching it the way they did.
The future of disc-based high-resolution audio got severely compromised first by the stupidity and arrogance of the relevant DVD working group and then by the Solips SACD run-away, (oxy)moronically adopting DSD as a production format.
It was then killed-off by the music companies not delivering as promised, perferring to defraud their customers with fake recordings.
Generally speaking. In the classical world things were a bit better.
Correct. The high-speed DS modulation with AA-filtered decimation is just one way of building low-speed long-word ADCs. The cheapest way, for a given level of performance.
Right. But one addition that still could have been made was that when all is said and done, a well-done 50s recording replayed on a good record player remains a total marvel.
But then, so is 1981-era optical disc technology 🙂
You can add turntable sound to the list. A turntable is such a complex amalgam of groove-triggered structural resonances that it too defies modelling.
- Status
- Not open for further replies.