Maker Pro
Maker Pro

Data write/read and peak detector animation

T

Tom Del Rosso

http://www.grc.com/animation.htm

Comments?

I know, the read head should be magneto-resistive instead of inductive.

But should it be Differential Manchester? The square wave transitions are
not encoded like this:

http://en.wikipedia.org/wiki/Differential_Manchester_encoding


For some people the most interesting thing is that the animation is coded in
javascript, with no flash or similar, and it's neatly hand-coded and
commented, so you might want to save the source for future reference.
 
J

Jasen Betts


Very nice!
I know, the read head should be magneto-resistive instead of inductive.

Inductive is oldschool. but not wrong.
But should it be Differential Manchester? The square wave transitions are
not encoded like this:

I think It should be some sort of RLL, differential manchester is
extremely oldschool.
http://en.wikipedia.org/wiki/Differential_Manchester_encoding


For some people the most interesting thing is that the animation is coded in
javascript, with no flash or similar, and it's neatly hand-coded and
commented, so you might want to save the source for future reference.

No crosspost to comp.lang.javascript :)

Your code?

The author has left the mouse coordinates display turned on
in the lower right corner. (when I put debug code in javascript I
use a hidden activation method eg: enable it with a key value in
location.hash or location.hostname)

The doctype is wrong, the document is not HTML4.01 Transitional.
 
R

Rod Speed

Tom Del Rosso said:
Comments?

That’s a very superficial/dumbed down view of what actually happens.
I know, the read head should be magneto-resistive instead of inductive.
But should it be Differential Manchester? The square wave transitions are
not encoded like this:

For some people the most interesting thing is that the animation is coded
in javascript, with no flash or similar, and it's neatly hand-coded and
commented, so you might want to save the source for future reference.

Yeah, its quite elegantly done.
 
T

Tom Del Rosso

Jasen said:
Very nice!


Inductive is oldschool. but not wrong.


I think It should be some sort of RLL, differential manchester is
extremely oldschool.

I was thinking in terms of a balanced signal, then realized the encoding
didn't have to be balanced.
 
A

Arno

In comp.sys.ibm.pc.hardware.storage Tom Del Rosso said:
http://www.grc.com/animation.htm

I know, the read head should be magneto-resistive instead of inductive.

Inductive is very historic.
But should it be Differential Manchester? The square wave transitions are
not encoded like this:

Manchester is also historic. All modern drives use some PRML
decoding, i.e. the signal is read analog and the decoding
and error correction coding takes that into account. The
encoding itself is also a lot more sophisticated than in
the old RLL or MFM days.
For some people the most interesting thing is that the animation is coded in
javascript, with no flash or similar, and it's neatly hand-coded and
commented, so you might want to save the source for future reference.

Not really. Unless you are trying to fix a very old tape data
recorder, this animation is irrelevant today.

Arno
 
T

Tom Del Rosso

Arno said:
Manchester is also historic. All modern drives use some PRML
decoding, i.e. the signal is read analog and the decoding
and error correction coding takes that into account. The
encoding itself is also a lot more sophisticated than in
the old RLL or MFM days.

Manchester produces a balanced signal. That's what I was thinking of.
Balance isn't needed for this apparently.

Not really. Unless you are trying to fix a very old tape data
recorder, this animation is irrelevant today.

I meant for those who write javascript. It's ability to do this was a
revelation to me.

The author's purpose was to explain something about Spinrite, the old disk
tool. He claims it does more thorough testing than a long self test. I
doubt it, but he is a good programmer. I told him recently he should put
that talent to use on a new product, but people still buy Spinrite. It has
a following, like fans.
 
A

Arno

Manchester produces a balanced signal. That's what I was thinking of.
Balance isn't needed for this apparently.
I meant for those who write javascript. It's ability to do this was a
revelation to me.

Oh, I see. That is actually interesting. Agreed.
The author's purpose was to explain something about Spinrite, the old disk
tool. He claims it does more thorough testing than a long self test. I
doubt it, but he is a good programmer. I told him recently he should put
that talent to use on a new product, but people still buy Spinrite. It has
a following, like fans.

And it is completely bogus today, basically a scam. Hence
my negative reaction to it. As modern disks and controllers
do not offer the interface MFM and RLL drives offered (you
could do a raw digital reading of the data). The equivalent is
not even possible today without uploading specialized firmware
to the drive or using secret vendor functionality. There
certainly is no standardized interface for asking the drive
to give you the analog signal. And the on-disk encoding is
also not standardized. SpinRite would need both to do any
better than a long selftest.

With MFM/RLL that was all different, and SpinRite did indeed
do better. These days are over, but the SpinRite fans
do not understand that. They just want their "god" tool
and feel superiour.

Arno
 
Top