Just saw this,
Succesful ABX of 24/96 vs. 16/44.1 - Hydrogenaudio Forums
20/20 ABX score
"I do not think, that I actually hear the presence or absence of stationary HF content. But what I am hearing is louder and somewhat smeared sounding transients in the Redbook version. That is exactly the type of pass-band artifact to be expected from low-pass filtering, but until lately I hadn't considered that to be audible, when done right. And since that step is necessary it can also not be avoided for Redbook delivery"
What is he talking about? I've never heard of transient errors due to downsampling to 44.1 kHz?
Is that part of this TIM theory?
Succesful ABX of 24/96 vs. 16/44.1 - Hydrogenaudio Forums
20/20 ABX score
"I do not think, that I actually hear the presence or absence of stationary HF content. But what I am hearing is louder and somewhat smeared sounding transients in the Redbook version. That is exactly the type of pass-band artifact to be expected from low-pass filtering, but until lately I hadn't considered that to be audible, when done right. And since that step is necessary it can also not be avoided for Redbook delivery"
What is he talking about? I've never heard of transient errors due to downsampling to 44.1 kHz?
Is that part of this TIM theory?
From Cirrus Logic ----- http://www.cirrus.com/en/pubs/whitePaper/DS668WP1.pdf
Figure 2 shows the impulse response of the same
Figure 2 shows the impulse response of the same
filter at a 96 kHz sample rate.
It is apparent that the pre-echo has shorter time duration and less time dispersion than at 48 kHz.
It is this difference that many believe to be a primary source of the audible superiority
of the higher sample rates.
Last edited:
Well, good for you! I don't see anything ironic here.
And, I agree with you, what do we care about audio replay systems 50 years from now anyway!
Jan
Another scientific paper of some interest here I believe
Phys. Rev. Lett. 110, 044301 (2013) - Human Time-Frequency Acuity Beats the Fourier Uncertainty Principle
We study human ability to simultaneously judge the frequency and the timing of a sound.
Our subjects often exceeded the uncertainty limit, sometimes by more than tenfold
Phys. Rev. Lett. 110, 044301 (2013) - Human Time-Frequency Acuity Beats the Fourier Uncertainty Principle
We study human ability to simultaneously judge the frequency and the timing of a sound.
Our subjects often exceeded the uncertainty limit, sometimes by more than tenfold
Indeed. We should be able to discuss the technical merits of technical implementations without feeling that we personally or our personal choices are attacked.
I like vinyl, once in a while, yet I am very aware of the technical limitations.
Doesn't bother me when I listen to great LP music.
Even though my turntable wow & flutter isn't exactly stellar, see attachment...
Jan
I like vinyl, once in a while, yet I am very aware of the technical limitations.
Doesn't bother me when I listen to great LP music.
Even though my turntable wow & flutter isn't exactly stellar, see attachment...
Jan
Attachments
The point exactly, unfortunately not always adhered to.
While I have more or less given up on analog, meaning I'll sooner buy the CD of an LP performance, I am glad John is enjoying his phono section work, and I'm pleased that there are interested parties out there. This comes to me as a sort of late recognition of his earlier work, or, in case these are younger people who were not around way back then, a firm thimbs up for John that he is doing something right enough to be accepted and recognized by the younger generation.
To see your own work appreciated so many years after the original appeared, well, that's really something. Not many can boast that.
"Confounders" as you called them come in many sizes and colours.
It's not often said publically, especially here, but even top flight engineers cannot avoid following fashon up to a point. It simply finds its way to you, no matter how hard you resist, if in not other way than through your commercial products.
An example. The rule here is use very high Ft output devices, because they are faster and hance more resposnive to transients. On the whole, that's true and one cannot argue with it on pure measurement grounds. But life is a whole lot more than just measurement.
It is never said that the faster a device is, the more easily it may oscillate. This not to say it will, or will do it easily,, but that it will do so before slower output devices. It's always a compromise. Motorola/ON Semi's MJ2119x series, for example, is not nearly as fast as today's norm, which is about 30 MHz or more, they do just 6 MHz in the usable range. Briefly, I find them to be more stable than practically all High Ft devices I have come across. "Stable" in this context means they will start to come unstuch much after the others have started oscillating.
The speed rule also doesn't take into account the fact that the Otala/Lohstroh amp used ancient BD 203/204 devices, rated at just 3 MHz, but still had a violtage slew rate of 100 V/uS. This implies a full power bandwidth (off hand) of about 350 kHz at least, which most of the modern devices will not achieve for many reasons, overcompensation to achieve stability being the prime one.
Secondly, such devices often come in TO-3 metal can packaging, which is a pain to build with, but which does enable them to dissipate not the usual 130 or 150W per device, but all of 250W per device. This has the benefit of needing less pairs for a given power output, which in turn means less expensive pairing of devices (though I am not sure many do it these days anyway).
And third, whoever bothers to look over their Data Sheets will quickly discover that the manufacturer's distortion rating for the power devices is unusually good, meaning low distortion. This is especially obvious when you compare THD at 1 and 20 kHz, because the expected rise at 20 kHz is lower with them than with any high speed device I have ever come across.
Yet, very few manufacturers use them any more, and of those who do, say 99% are from the professional sector, and within that, again most are in the sound reinfircement group. This means they are more robust and can take much more punishment than plastic pack devices.
But all this is somehow lost here. It's a strictly high speed, high Ft forum. This is contrary to my personal experience. Every time I changed some fancy output device hitting 10 MHz and over in a TO-3 package for the Motorolas, I was rewarded with a richer, warmer and almost palpable sound.
Similar state of affairs with input stages. Here, you ain't a man unless you use cascoded FET input. I know the theory, but unfortunately, I also know the practice. A well designed as classic as possible differential stage, with a decent bias of say 1-1.5 mA per transistor, with its gain limited to 6-10:1, will sound just as good, and often even better than most of the cascoded FET offerings from the industry at large. Back to the Otala/Lohstroh amp: it uses very simple BJT differential amps, but obviously with a lot of knowledge thrown into the deal, and outshines a lot of fancy shmansy topologoies.
My own H/K 6550 has a simple differntial BJT input, yet still uses only 17 dB of global NFB for excellent sound results.
Makes me wonder what's real, and what's simple and plain fashon. Every time had its trends, why not ours as well?
It's not often said publically, especially here, but even top flight engineers cannot avoid following fashon up to a point. It simply finds its way to you, no matter how hard you resist, if in not other way than through your commercial products.
An example. The rule here is use very high Ft output devices, because they are faster and hance more resposnive to transients. On the whole, that's true and one cannot argue with it on pure measurement grounds. But life is a whole lot more than just measurement.
It is never said that the faster a device is, the more easily it may oscillate. This not to say it will, or will do it easily,, but that it will do so before slower output devices. It's always a compromise. Motorola/ON Semi's MJ2119x series, for example, is not nearly as fast as today's norm, which is about 30 MHz or more, they do just 6 MHz in the usable range. Briefly, I find them to be more stable than practically all High Ft devices I have come across. "Stable" in this context means they will start to come unstuch much after the others have started oscillating.
The speed rule also doesn't take into account the fact that the Otala/Lohstroh amp used ancient BD 203/204 devices, rated at just 3 MHz, but still had a violtage slew rate of 100 V/uS. This implies a full power bandwidth (off hand) of about 350 kHz at least, which most of the modern devices will not achieve for many reasons, overcompensation to achieve stability being the prime one.
Secondly, such devices often come in TO-3 metal can packaging, which is a pain to build with, but which does enable them to dissipate not the usual 130 or 150W per device, but all of 250W per device. This has the benefit of needing less pairs for a given power output, which in turn means less expensive pairing of devices (though I am not sure many do it these days anyway).
And third, whoever bothers to look over their Data Sheets will quickly discover that the manufacturer's distortion rating for the power devices is unusually good, meaning low distortion. This is especially obvious when you compare THD at 1 and 20 kHz, because the expected rise at 20 kHz is lower with them than with any high speed device I have ever come across.
Yet, very few manufacturers use them any more, and of those who do, say 99% are from the professional sector, and within that, again most are in the sound reinfircement group. This means they are more robust and can take much more punishment than plastic pack devices.
But all this is somehow lost here. It's a strictly high speed, high Ft forum. This is contrary to my personal experience. Every time I changed some fancy output device hitting 10 MHz and over in a TO-3 package for the Motorolas, I was rewarded with a richer, warmer and almost palpable sound.
Similar state of affairs with input stages. Here, you ain't a man unless you use cascoded FET input. I know the theory, but unfortunately, I also know the practice. A well designed as classic as possible differential stage, with a decent bias of say 1-1.5 mA per transistor, with its gain limited to 6-10:1, will sound just as good, and often even better than most of the cascoded FET offerings from the industry at large. Back to the Otala/Lohstroh amp: it uses very simple BJT differential amps, but obviously with a lot of knowledge thrown into the deal, and outshines a lot of fancy shmansy topologoies.
My own H/K 6550 has a simple differntial BJT input, yet still uses only 17 dB of global NFB for excellent sound results.
Makes me wonder what's real, and what's simple and plain fashon. Every time had its trends, why not ours as well?
Jan, do you actually HEAR the wow&flutter, as is, or indirectly?
If not, than what do you care?
No I don't hear it. But you know me, I want to know
BTW If you look at the wow frequency in that graph it should be obvious where it comes from....
Jan
It's OK to measure it just so you know, but it would be silly to worry about something you do not hear as such, or implicitly.
I have an old Dual CS 604, direct drive, from 1977 or 1978, I forget which, which is quoted at something like 0.05% or some such. It has never failed me, and it probably never will, and in return, I will refresh it this automn.
It has never sounded great, I'd say very good for its price class, but I must add I am using Ortofon cartridges, which are finnicky about capacitance and need 400-440 pF total to be truly linear. In my day, I measured the whole setup and have written down the capacitance present at the RCA output jacks. This helps me adkist the preamp phono RIAA input capacitance so that the total is what is required.
Although Orotofon says 47k input impedance, by trial and error I have discovered that the best sound comes with a 59k impedance and 400 pF. Hence, once given its optimum, I get better sound than most sonly because I took some time to get it all right.
I suspect you did something like that as well. So, sit back, let the music play ans SMILE!
People speculate as to why reproduction usually doesn't sound quite right, compared to live - and for me the answer has always been that the high end, the treble content, is not being done right, especially at higher volumes. And for me getting that means the circuit needs to be 'fast' - sorry Dejan, ! - the higher harmonics are always under very 'tight control'.
But the car analogy works well here - a vehicle with plenty of capability for speed needs excellent suspension and brakes to make it a complete, well balanced, comfortable package - just dropping a big donk into a conventional runabout may make it exciting for a bit, but it's not something you're happy to live with all the time.
The package has to be in balance, 'fast' electronics will do it beautifully, but it has to be precisely engineered to always behave itself, be totally predictable ...
! - the higher harmonics are always under very 'tight control'.But the car analogy works well here - a vehicle with plenty of capability for speed needs excellent suspension and brakes to make it a complete, well balanced, comfortable package - just dropping a big donk into a conventional runabout may make it exciting for a bit, but it's not something you're happy to live with all the time.
The package has to be in balance, 'fast' electronics will do it beautifully, but it has to be precisely engineered to always behave itself, be totally predictable ...
People speculate as to why reproduction usually doesn't sound quite right, compared to live - and for me the answer has always been that the high end, the treble content, is not being done right, especially at higher volumes. And for me getting that means the circuit needs to be 'fast' - sorry Dejan,
But the car analogy works well here - a vehicle with plenty of capability for speed needs excellent suspension and brakes to make it a complete, well balanced, comfortable package - just dropping a big donk into a conventional runabout may make it exciting for a bit, but it's not something you're happy to live with all the time.
The package has to be in balance, 'fast' electronics will do it beautifully, but it has to be precisely engineered to always behave itself, be totally predictable ...
Why "sorry Dejan"? I'm all for fast electronics, but you have to think a bit where that is best applied. The current stage theoretically has a voltage gain of 1, so its Ft is not really threatened. As the Otala/Lohstroh design amply proved.
Let's put it this way Frank - I prefer to have a very fast amp even if I will never actually need all that speed, than to have a mediam fast amp with which I MIGHT some day need the speed I do not have.
And the car analogy works here. After I had built my own car, I realized that I had actually spent more on balancing it, on road holding and grip, on the suspension and on the brakes than on the engine. Power without control kills, and I don't want to get killed, I want to be able to enjoy the power I have.
I stand corrected, thank you.
To be perfectly honest, I normally don't worry who owns what brand of beer, I just drink it.
No I don't hear it. But you know me, I want to know
BTW If you look at the wow frequency in that graph it should be obvious where it comes from....
Jan
Wow and flutter is easily heard to me on piano recordings. On suspension tables, worn or loose springs will exaggerate this ....
Yes, I'm sure that you could easily hear this on piano, especially long, sustained notes. Is this large variation due to a bad belt or pulley?
- Status
- Not open for further replies.