Other than Marantz, Sony Philips etc. did Sharp electronics make anything to contend/compete on the CDP / DAC market,
Anyone know?
Anyone know?
Sharp made a few CD players in the very early days of CD and I remember working on them at the time but couldn't recall now what the DAC was.
DX50 ? 60 ? DX150
DX50 ? 60 ? DX150
Sharp was never big in making chips for audio, except voltage regs. I have definitely never seen a Sharp dac chip in 30+ years in audio biz.
Interesting whats interesting on this one below is the music program indicator - very solid work from SHARP on a whole.
going to browse that list (thanks). I know where to go if I choose to mod something in future 🙂... Classic retro looks
vintage-audio-laser.com/Sharp-DX-3
I don't recognise any of the internal DAC chips on this one, any idea's what's used? Oversampling etc?
going to browse that list (thanks). I know where to go if I choose to mod something in future 🙂... Classic retro looks
vintage-audio-laser.com/Sharp-DX-3
I don't recognise any of the internal DAC chips on this one, any idea's what's used? Oversampling etc?
Last edited:
Managed to find out DAC used, "Sony CX-20012" apparently,
Broken link above in previous post, sorry Vintage Audio Laser and more here.
😎
http://vintage-audio-laser.com/-Sharp-
Broken link above in previous post, sorry Vintage Audio Laser and more here.
😎
http://vintage-audio-laser.com/-Sharp-
The picture shows a CX20017 DAC, same as the original CDP101. These were time shared between channels, in fact if you read the bit in this article called "Inside the Sony CDP101", you will see how I believe Sony made a token gesture at eliminating the audible effect of time sharing the DAC.
Adrian-Kingston.com - Sony CDP-101 CD Player
I say "I believe" because it was me that told this guy about this some years ago... and I see he's added the info to his site with no acknowledgement.
Adrian-Kingston.com - Sony CDP-101 CD Player
I say "I believe" because it was me that told this guy about this some years ago... and I see he's added the info to his site with no acknowledgement.
Fascinating stuff, gonna read that.
Funny, when browsing the vintage audio laser website I came across an ancient technics player, the dac inside: PCM53JP-V.
I research the chip, apparently it's another original NOS implimentation.
So essentially these are the chips used before over/upsampling was introduced.
Correct me if i'm wrong but
As I recall the soul purpose of oversampling was to aid analog filtering after the reconstruction stage but why is filtering so important?
I'm sure most decently built equipment can handle beyond 20khz...
Funny, when browsing the vintage audio laser website I came across an ancient technics player, the dac inside: PCM53JP-V.
I research the chip, apparently it's another original NOS implimentation.
So essentially these are the chips used before over/upsampling was introduced.
Correct me if i'm wrong but
As I recall the soul purpose of oversampling was to aid analog filtering after the reconstruction stage but why is filtering so important?
I'm sure most decently built equipment can handle beyond 20khz...
Last edited:
Yes, oversampling allowed the use of much more "gentle" filters on the audio output. The CD standard ensures that there is no audio present beyond 44.1 /2 KHz (so 22.05kHz).
You need to read up on sampling theory and "aliasing" to understand the whys and wherefores of what happens if you try and record a frequency higher than one half the sample rate. But all that's on the recording side of things...
🙂
On playback, a non oversampled convertor produces a lot of hash that has to be removed. To remove it from a non oversampled convertor needs a filter with really steep roll off... and that can have a detrimental effect on the wanted audio. An oversampled player pushes that noise way up in frequency and so it can be removed much more easily without impinging on the audio band.
Have a search for "CD brickwall filters" and look at the images showing the response.
If the hash and noise were passed to an amplifier it could cause all sorts of distortions and problems (depending on the amp). Speakers may not appreciate a high hf content of noise too and it could cause audible problems as well.
You need to read up on sampling theory and "aliasing" to understand the whys and wherefores of what happens if you try and record a frequency higher than one half the sample rate. But all that's on the recording side of things...
🙂
On playback, a non oversampled convertor produces a lot of hash that has to be removed. To remove it from a non oversampled convertor needs a filter with really steep roll off... and that can have a detrimental effect on the wanted audio. An oversampled player pushes that noise way up in frequency and so it can be removed much more easily without impinging on the audio band.
Have a search for "CD brickwall filters" and look at the images showing the response.
If the hash and noise were passed to an amplifier it could cause all sorts of distortions and problems (depending on the amp). Speakers may not appreciate a high hf content of noise too and it could cause audible problems as well.
Thought so, I understand the recording side more than the playback regarding folding etc.
Alot of lectures on the web don't actually state recording or playback, it's confusing at first and hard to tell which side is which.
But regarding hash on the playback side, are there any examples of the nature of it?
If there's something beyond 22.05, i'm sure any fairly decent amp can manage.
Was it perhaps more to related to the effect on passive components and tweeter combinations?
It would be interesting to test one of these NOS CD players for this "ultrasonic content", that they were keen on filtering out.
Something that really remians unclear, probably a really stupid question but hey ho,
Take a 44.1k CD, is each stereo channel comprised of information up to 44.1khz?
Take a 44.1k CD, is each stereo channel comprised of information up to 44.1khz?
44.1kHz is the rate at which the incoming audio is sampled. 16bits (the other part of it all) sets the dynamic range.
Draw a 1kHz sinewave on a piece of paper and put the timescales in for 1 cycle. Now divide that x axis time scale equally so that there are 44100 points. draw a line up from each point and where it intersects the audio is where a sample (which corresponds to the level) is taken.
Now try it for a 22.05kHz sinewave. Not many samples are there 😀 yet that sinewave can be reconstructed correctly from that data.
So each channel could in theory have info up to 22.05kHz (the magic one half of the sampling frequency). In practice it will be limited to a little lower than that.
The 16 bit part means there are 2 to the power 16 individual levels available (65536).So if you want 2 volts as a max level, then 2 divided by 65536 means each discrete level is "0.03millivolts". 65536 times 0.03 millivolts is 2.
Draw a 1kHz sinewave on a piece of paper and put the timescales in for 1 cycle. Now divide that x axis time scale equally so that there are 44100 points. draw a line up from each point and where it intersects the audio is where a sample (which corresponds to the level) is taken.
Now try it for a 22.05kHz sinewave. Not many samples are there 😀 yet that sinewave can be reconstructed correctly from that data.
So each channel could in theory have info up to 22.05kHz (the magic one half of the sampling frequency). In practice it will be limited to a little lower than that.
The 16 bit part means there are 2 to the power 16 individual levels available (65536).So if you want 2 volts as a max level, then 2 divided by 65536 means each discrete level is "0.03millivolts". 65536 times 0.03 millivolts is 2.
So I'm confusing rate with frequency and there definately isn't any frequecy content beyond 22.0 on a CD and if there is it's not useful.
That's plot explanation really did the trick, thanks.
That's plot explanation really did the trick, thanks.
There is nothing on a CD above around 20kHz (22.05 theoretical limit).
I've just found this that might help explain it better.
CD - The Inside Story - Part 8 - Over-sampling and Filtering
I've just found this that might help explain it better.
CD - The Inside Story - Part 8 - Over-sampling and Filtering
Interesting whats interesting on this one below is the music program indicator - very solid work from SHARP on a whole.
going to browse that list (thanks). I know where to go if I choose to mod something in future 🙂... Classic retro looks
vintage-audio-laser.com/Sharp-DX-3
I don't recognise any of the internal DAC chips on this one, any idea's what's used? Oversampling etc?
The cd player in the picture is a Hitachi 1000 clone, it had a lot of incarnations, I have the Dual one, from ca. 1983.
Thanks for that link moo,
I was keen on gettin a NOS up and running without the analog filter after the TDA but, clearly this woul allow whatever this is in the below picture to get through
What exactly are the additional four bars (I note they repeat infinitely)...
I was keen on gettin a NOS up and running without the analog filter after the TDA but, clearly this woul allow whatever this is in the below picture to get through
An externally hosted image should be here but it was not working when we last tested it.
What exactly are the additional four bars (I note they repeat infinitely)...
The "four bars" represent the noise component from the DAC if the "music" were white noise. Everything from almost DC up to 22.05kHz. OK, that's a paradox, lets say white noise sharply cut off at 22.05kHz 😀
Because the sampling is like a 44.1kHz squarewave, the harmonics of that extend outwards, so 44.1, 88.2 and so on to infinity. In practice the level does fall away pretty quickly the higher up you go because of the limitations of the circuitry in passing such signals. When a signal is sampled, you get an upper and lower sideband centred around the 44.1kHz sampling frequency. Its very important that that lower sideband doesn't "overlap" into the wanted audio below.
I did some more searching and turned this up which looks pretty comprehensive. You'll see its quite a vast subject with a lot of number crunching but some parts of it might help you understand the basics 🙂
The Sampling Theorem
and the next bit,
http://www.dspguide.com/ch3/3.htm
Because the sampling is like a 44.1kHz squarewave, the harmonics of that extend outwards, so 44.1, 88.2 and so on to infinity. In practice the level does fall away pretty quickly the higher up you go because of the limitations of the circuitry in passing such signals. When a signal is sampled, you get an upper and lower sideband centred around the 44.1kHz sampling frequency. Its very important that that lower sideband doesn't "overlap" into the wanted audio below.
I did some more searching and turned this up which looks pretty comprehensive. You'll see its quite a vast subject with a lot of number crunching but some parts of it might help you understand the basics 🙂
The Sampling Theorem
and the next bit,
http://www.dspguide.com/ch3/3.htm
I'm familiar with nyquist and sampling theory, it's just that alot of lectures don't really go in depth explaining "sample pulses" beyond 44 that much, where they are derived from, or why these pulses fold infinitely.
I was after a better explanation on pulses... 🙂
I got the impression they only repeated in os applications, but revisiting the lavry paper shows they exist even in nos. (though it is not clear seeing as it is the paper on "oversampling" as it implies in the title).
So really, just to confirm they definately are present even in nos applications.
Last edited:
- Status
- Not open for further replies.