I used to get upset about this. You see a person being so deluded and ignorant and you know it is relatively simple to explain how it really works.
Except that the vast majority IS NOT interested in how it really works. They are happy and cozy in their self-generated reality, and because they no longer see the difference with the real world they present their nonsense as if it is a fact, as you noted.
I find it still regrettable and still, after 12+ years at diyaudio, find it sometimes difficult to just let it go. It won't kill the guy, after all. Generally.
Jan
When i worked out the differences between producing an analogy( such as running along), sampling and binning, thou were silent, Jan. As thou did not praise me then, thou should not call me ignorant now.
No, 56K was designed by Motorola and i think never produced by TI.gnobuddy said:
Some of them have ears and brains. Some of them despise the over-technicised world. Some of them just want to give you the kicks.
I can't keep an eye on everyone here 😱 !
Nothing personal, but you showed ignorance on the issues that were discussed.
Quickly rectified of course, if you really want.
Jan
I can only assume that this was a poor attempt at humour, combined with a poor attempt at punctuation. A smiley at the end would have confirmed this.gabdx said:
Conscience is only an abstract, a resumé, a Zusammenfassung of one's brain activity. If one takes only facts into account and leaves no slack for feel, then one is gonna loose one's brain, as we know it, in the long run.
Oh good grief. Vinyl isn't that bad. In fact, it can be amazingly good. Just like digital.
Do we have to resort to hyperbole to make a point?
Do we have to resort to hyperbole to make a point?
56K was a great chip for audio, 24/56 bit DSP, great instruction set. We worked closely with MOT Israel on the device, as an early volume user - somewhere I still have the letter "thanking" me and my partner for discovering the pipeline in the first silicon didn't work properly... We wrote the first ever single processor V32 modem on the 56K. Tx, Rx, near and far echo canceller, trellis and AT commands all on the one DSP... All in assembler!
I got to the point at 4:19 and I STOPPED THE VIDEO to post here - the display software is displaying a sine wave made up of stairsteps!
I just checked, Audacity doesn't do that. Its display has each data point as a point (as in the Oppenheim/Schafer textbooks), and it draws a straight line between consecutive points. This is a good approximation for a sine wave, but doing it with a square wave shows that (it really is just a line between points, and) this method doesn't emulate a reconstruction filter. Cool Edit 2000 (now Adobe Audition) DOES apply a reconstruction filter to generate the line it displays connecting the points, and correctly shows square waves and other transients as having the Gibbs phenomenon.
But in the textbooks it does NOT show stairsteps, it shows an impulse for each point, a line going up or down from the axis to each data point. Mathematically and engineering-wise this doesn't matter, as either a staircase or an impulse train will be filtered to the correct wave by the reconstruction filter, but ... oh, geez. Peeps point to that and say "See, it's a stairstep! It's DIGITAL!"
And they say "It's DIGITAL" with great distain. Ironically, when CDs first came out, those two words mean the opposite, that it's the best, and even more, it's perfect. "Perfect Sound Forever."
Woops, I just ranted... did he say what software he was using? I suppose I should watch the rest of the video...
ETA: Okay, he saves himself [not quite the word I want to use] at 6:05 by saying "stairsteps are wrong to begin with." Back to the video...
Last edited:
Perhaps not, but (in my best whiny schoolboy voice) "they started it!" 😀
I have, indeed, heard vinyl sound very good. I have also frequently heard it sound very bad, the most irritating aspect for me usually being enormous amounts of very harsh treble distortion, particularly near the centre of the disc.
That's where recorded wavelengths are shortest, due to the low linear speed of the groove. Because of the short wavelengths, tracing distortion, is at its worst here.
Similarly, peak stylus acceleration is inversely proportional to the square of the wavelength of the wiggles in the groove, so the centre of the disc is also where the most demands are placed on the stylus having very low tip mass. If it's not low enough, the stylus is literally too heavy for the vinyl walls to force it to follow the wiggles in the groove.
Instead, the vinyl undergoes plastic deformation where the tip of the stylus contacts it, and the stylus carves its own erratic line through the unfortunate vinyl. (The cartridge still tracking in the groove, but the tip of the stylus not following the intended path - remember the stylus only contacts a narrow patch on each wall of the V-shaped groove, and this can be plowed into a furrow by excessive tip mass.)
A lot of this was revealed by a collection of fascinating electron-microscope photos of the walls of the microgroove in a vinyl record. They were in a slim little book called, I think, "Pickups: The Key to Hi-Fi". I still remember one photograph, which showed treble content in a vinyl groove that had been utterly destroyed by one single playback with a cartridge that had too much tip mass; you could clearly see that the tip of the stylus had ploughed its own furrow through the vinyl, ignoring the original curvature of the groove walls.
I have read that the era of quadraphonic records in the 1970s forced many improvements in playback cartridge design. In particular, one of the quadraphonic formats used a 30 kHz carrier wave recorded into each wall of the vinyl disc, with the additional two channels encoded onto it.
To play these back, and recover the encoded rear channels, the playback cartridge had to be able to track frequencies up to 45 kHz. 😱
At a time when most cartridges struggled to reproduce 15 kHz well, this must have been a tremendous engineering challenge. To triple the upper frequency limit, without exceeding the limited force available from the vinyl groove walls, stylus effective tip mass would need to be lowered by a factor of nine!
I don't think that tremendous engineering challenge was actually ever fully met, as those records supposedly had a reputation for rapidly losing their "quad" sound after a few plays. The 30 kHz carriers simply got smeared out by the stylus being too heavy to follow it.
But I'm betting that manufacturers did manage to lower the tip mass substantially, even if not by the full factor of nine times. The "rock on a stick" did become lighter, and that would have benefitted ordinary stereo record playback, too.
-Gnobuddy
Okay, saw the whole video (and can no longer edit my post). He goes into dither (a rather general topic for digitizing, or more specifically, bit reduction) and noise shaping (specifically for audio, though I've seen where a filtered noise spectrum can be added outside the bandwidth of interest to improve digitizing in RF systems). This National Semiconductor app note may be of interest, and perhaps worth passing on to the same people you'd want to see the video:
http://www.ti.com.cn/cn/lit/an/snoa232/snoa232.pdf
http://www.ti.com.cn/cn/lit/an/snoa232/snoa232.pdf
For years I've used Goldwave as my audio editor. I like it because I know it, and it has some cool tools. Goldwave displays the waveform as stair steps. That's great because it lets me get in and edit sample by sample if need be. The stair step makes a good visual representation of the actual samples. That stepped graphic on the screen makes it easy to grab a single sample and change it.
Audacity, to its credit, shows the samples as dots with a line between them.
It does not mean that the analog waveform coming out of the DAC is stepped. It's just a handy way of representing the samples at a high zoom setting.
Audacity, to its credit, shows the samples as dots with a line between them.
It does not mean that the analog waveform coming out of the DAC is stepped. It's just a handy way of representing the samples at a high zoom setting.
https://linearaudio.net/downloads a list of tip mass figures. Note
0.055mg Technics EPC-(P)100CMK4 (MM, boron pipe cant.)
Real shame they won't bring that back as all accounts were it was a stunning cartridge, but not flooby enough for the audio press...
Now, if we could only get that message out to all the digitally-paranoid people in the world...
-Gnobuddy
JP and I are both huge fans. I have one left that hasn't gone into cantilever collapse. IMO, best cartridge ever made.
Got into a discussion on this topic with a customer a few years ago, standing out in a parking lot after finishing a service call. He was trying to explain to me about stairsteps coming out of the DAC, illustrating with his finger in the dirt on the back window of my service van. I replied, "Sorry, but that's a myth, that's not what is coming out of your CD player," and drew a few pictures of my own. We parted ways without violence. I forgot about the pictures.
Next day, my boss comes back from lunch and says, "Would you please wash the PCM tutorial off the back of your truck?! Jesus..."
-- Jim
Next day, my boss comes back from lunch and says, "Would you please wash the PCM tutorial off the back of your truck?! Jesus..."
-- Jim
When I tried Goldwave the redraw time when you transversed a huge file was excruciating. The CoolEdit developers considering the time were very clever their virtual memory model even impressed our SPICE developers. I could expand a full 10 hour mp3 into 32/96 and zoom in/out anywhere in the file with no redraw latency at all.
- Status
- Not open for further replies.