Dolby C vs. dbx
Steve Hannoford
works, but my understanding (explained by more than one audio
engineer) is that Dolby is a reduction system which requires
the same implementation during playback that was used during
recording, where-as dbx will improve the quailty of the tape
even if you do not have dbx at playback (in fact I think that
dbx is only effective during recording).
What does that mean?
Dolby has two classes of NR: Type B and the newer Type C.
Manufacturers can buy Dolby NR at several price levels, so
no all Dolby C are the same. The type refers to how it reduces
noise, the level determines the quality of circuitry used, etc.
Dolby labs researched tape noise and isolated the frequencies
where the noise typically occurrs. When you make a recording
with Dolby, the signal is boosted in that range, so that the
signal to noise ratio is increased. (The signal is boosted so
that it 'covers up' the noise.) This results in a top heavy
(increased treble range) recording. During playback, the
signal is reduced by the same amount, restoring the proper
treble levels, but the noise isn't there (well not as much
of it).
If you record with Dolby and play back without, the tape will
sound brighter - obviously, because the treble was boosted.
It's like turing the treble all the way up and the bass all the
way down on the pre-amp... Some people 'prefer' this sound
(the brightness of dolby record without dolby playback), but
the signal is not like the original signal - the sound is altered.
My understanding with dbx is that its noise reduction is during
record only, and is not required during playback... Not only
is it a better reduction system (results in less noise than
either Dolby B/C), but it is also preferential because it
works without the requirement that the implementation at
playback match the implementation used to record....
--
+-------------------------+--------------------------------------+
| ste...@pilot.dmg.ml.com | My employer has no opinions and has |
| Steveh Hannaford | nothing to do with anything I say. |
William Nau
>works, but my understanding (explained by more than one audio
>engineer) is that Dolby is a reduction system which requires
>the same implementation during playback that was used during
>recording, where-as dbx will improve the quailty of the tape
>even if you do not have dbx at playback (in fact I think that
>dbx is only effective during recording).
I believe you are getting dbx mixed up with hx-pro. HX-pro is
effective only during recording. DBX compresses the signal
and must be uncompressed during playback or it really sounds
terrible.
Lon Stowell
>I can tell you how Dolby works, but I can't really say how dbx
>works, but my understanding (explained by more than one audio
>engineer) is that Dolby is a reduction system which requires
>the same implementation during playback that was used during
>recording, where-as dbx will improve the quailty of the tape
>even if you do not have dbx at playback (in fact I think that
>dbx is only effective during recording).
>
record bias based on the amount of treble signal present...to
avoid treble losses due to saturation.
HX-Pro is a record-only technique.
DBX is a compression scheme on record and a expansion on
playback. There are several different DBX units, but the type
included in deck with built-in DBX is a non-frequency dependent
compression and expansion. It sounds like heck if not played
back on a decoding deck.
Dolby B just sounds a bit bright if played back on a non-Dolby
deck. Also Dolby C sounds a bit bright if played back on a
Dolby B deck. Dolby C on a non-Dolby deck gets rather annoying
due to excessive treble....but even that is more listenable
than undecoded DBX. (you can always just turn down the treble a
bit...)
>
>Dolby has two classes of NR: Type B and the newer Type C.
>Manufacturers can buy Dolby NR at several price levels, so
>no all Dolby C are the same. The type refers to how it reduces
>noise, the level determines the quality of circuitry used, etc.
Although this might have been true in the very early days of
Dolby C, I'd have to take issue with the statement...other than
for individual quality differences in decks NOT related to
Dolby.
And the level of Dolby doesn't determine the "quality of the
circuit". There are superb Dolby B decks and horse-apple Dolby
C ones.
j...@cmkrnl.com
> I can tell you how Dolby works, but I can't really say how dbx
Several replies have already noted the flaws in this person's statements about
dbx. As for Dolby B and C...
> Dolby labs researched tape noise and isolated the frequencies
> where the noise typically occurrs.
Yes, but...
> When you make a recording
> with Dolby, the signal is boosted in that range, so that the
> signal to noise ratio is increased. (The signal is boosted so
> that it 'covers up' the noise.) This results in a top heavy
> (increased treble range) recording. During playback, the
> signal is reduced by the same amount, restoring the proper
> treble levels, but the noise isn't there (well not as much
> of it).
This isn't quite correct. It isn't just a matter of boosting the treble on
record and cutting it on playback -- that would be merely a different EQ curve,
and all tape decks already apply EQ on recording and apply the reverse EQ on
playback. Simply boosting the treble even more would cause tape saturation.
Rather, Dolby B and C apply *compression* to the high frequencies on record,
and expansion on playback. Parts of the program that already have a lot of
high-freq. content won't be much affected by Dolby. But in parts in which the
highs are of low level, highs are boosted on record and cut on playback.
During the "cut on playback", the tape hiss, which is also mostly high freqs,
gets cut too. Voila. (When there is a lot of high-freq content in the
program material, you don't notice the hiss nearly so much, as you're paying
attention to the program. This is a psychoacoustic affect known as "masking".)
To get an idea of how compression and expansion works, imagine that you're
recording a series of numbers in the range 0 through 1. But your recording
scheme somehow has a problem with the lower numbers -- anything much below 0.9
will be corrupted by noise on playback. So you first divide the numbers by 10
and then add 0.9. Now you are recording numbers in the range 0.9 to 1.0. On
playback you reverse the procedure and recover the original numbers.
(This by the way would be a ten-to-one compression ratio. Most companders used
in audio work are far less aggressive than that, 2:1 or so. The compander that
is part of VHS HiFi is, I believe, about 2.5:1. THe main difference between
Dolby B and C, btw, is in the compression ratio used -- I don't know the
exact numbers but C has about twice the compression ratio of B.)
Now, anyone who works with floating point numbers in programming will spot a
flaw: In digital computers you only have so many bits of precision per number.
Since you can't record with infinite precision you will end up with some
rounding errors on playback. Suppose, for instance, you can only record two
decimal places after the decimal point. In the "companding" scheme, inputs
0.91 and 0.92 will both be recorded as 0.99. So on playback you don't know
quite what the original number was, and you have to settle for something close,
but you regard this as an ok tradeoff in exchange for avoiding the problems you
used to have with recording small-valued numbers.
The same thing happens in analog recording -- we aren't getting something for
nothing here; we are merely shifting the errors to a domain where they're less
objectionable. In companding schemes (Dolby B, C, S, and A, and dbx, are all
companding schemes) we are getting rid of tape-hiss "errors" but we might not
be getting all of the pieces of the original program played back at exactly the
right levels with respect to one another. This is called "tracking error" and
it is usually an acceptable tradeoff, as the hiss is far more bothersome. Most
Dolby C implementations seem to be an exception: Dolby C has proven to be
extremely difficult to get right as far as level-tracking is concerned. Even
slight head misalignments between the recording and playback decks are enough
to cause bad tracking errors. (In the "digital" example above, the effect of
head misalignment would be to shift the entire range of recorded numbers
downward -- say from 0.9 to 1.0 down to 0.7 through 0.8. Since the playback
decoder is being presented with bad data it can't possibly recover the original
numbers.) In fact, the cassette mechanism itself is sloppy enough so that
record and playback on the same deck sometimes causes problems.
This is why Dolby is trying to push Dolby S -- it is just as effective as Dolby
C, maybe more so, but is far less subject to tracking errors due to head
misalignment. It also sounds quite acceptable when played through a Dolby B
deck. If the Digital Compact Cassette (DCC) turns out to be a dud, you may see
the prerecorded cassette industry switch to Dolby S.)
In some companding schemes, notably that used by VHS HiFi to get its wide
dynamic range, other errors are common. These frequently show up as noise
"tails" at the beginning of quiet sections following crescendos, and in an
audible modulation of the background noise level (known as "breathing" and
"pumping").
--- Jamie Hanrahan, Kernel Mode Consulting, San Diego CA
Internet: j...@cmkrnl.com, hanr...@eisner.decus.org, or j...@crash.cts.com
Uucp: ...{crash,eisner,uunet}!cmkrnl!jeh
Odd Arild Urtveit
Dolby B boosts the treble, but not in a linear way. The weaker the signals are the more they are boosted. Strong signals are not increased at all. This results in a 'compression' of the treble. During playback the inverse process takes place: The Dolby system 'uncomresses' the treble and 'hide' the tape-hiss away. The improvement in S/N-ratio can be up to 10dB. Because of this non-linear boost, this is not equivalent to increaseing the treble.
Dolby C works in a similar way as Dolby C by boosting the treble. However, Dolby C also works in the mid-range (>500Hz). The boost is even more nonlinear for Dolby C than for Dolby B resulting in a bigger improvement in S/N-ratio. Up to 20dB (or 15dB depending on the frequency that are measured). Dolby C also increases the saturation-level of the tape by a few dB. (= higher recording-levels).
dbx works by 'compressing' the entire signal by boosting the weak signal more than the strong ones. Put into other words: The dynamic range of the signal is squeezed together so that it would fit into the much narrower dynamic range of the tape. This results in a recording that would sound very unpleasent without 'uncompression'. Therefore, dbx should also be used during playback. The result is an improvement in the S/N-ratio of aprox. 30dB.
Because of the way these NR systems works properly, it is vital that the tapedeck is correctly adjusted. (Bias & tape sensitivity level). The heavier the alterations to the signal the NR system performs, the more important is this matter. dbx is the most effective, followed by Dolby C and Dolby B, but because ov the reason mentioned above dbx is also most critical to the tapedeck adjustment and it's therefore easier to ruin a dbx recording than a Dolby B/C recording.
Dolby HX Pro is not a NR-system, but a system to obtain constant bias. Without Dolby HX Pro the treble will work together with the bias, resulting in a varying bias. The more treble the more bias, and more bias results in a reduction in the tapes treble saturation level. By keeping the bias constant, Dolby HX Pro tries to elliminate this effect resulting in a higher treble saturation level and therefore better treble. (<=> better frequency range)
O.A.Urtveit
re...@mi.uib.no
Tom Mathes
of track. I was asking the subjective opinions out there. So far, I've found
that most folks that object to dbx hate the noise pumping artifacts. If noise
pumping doesn't bother the listener, he/she prefers dbx over Dolby C.
Noise pumping artifacts bother me, so I'll skip on dbx. Thanks to all who
responded to my query by e-mail. All of you were very helpful.
In article <1992Oct4....@cmkrnl.com> j...@cmkrnl.com writes:
>Rather, Dolby B and C apply *compression* to the high frequencies on record,
>and expansion on playback. Parts of the program that already have a lot of
>high-freq. content won't be much affected by Dolby. But in parts in which the
>highs are of low level, highs are boosted on record and cut on playback.
>During the "cut on playback", the tape hiss, which is also mostly high freqs,
>gets cut too. Voila. (When there is a lot of high-freq content in the
>program material, you don't notice the hiss nearly so much, as you're paying
>attention to the program. This is a psychoacoustic affect known as "masking".)
You've got your theory right, but terminology wrong. Dolby amplifies, or
expands the high frequencies during recording, then compresses them during
playback. That's one reason to critically check the recording levels during
recording with all Dolby NR systems. Too high a level will saturate the tape
and distort like crazy. That's where Dolby HX helps; it can help reduce the
tape saturation effects of high frequencies by attenuating the recording bias
during high-level high-frequency passages.
>Dolby C implementations seem to be an exception: Dolby C has proven to be
>extremely difficult to get right as far as level-tracking is concerned. Even
You're right: the biggest problem with Dolby C is tape sensitivity mismatches.
If it's easy to set this on your deck, along with bias level, Dolby C works great.
Problem is that most decks don't allow easy adjustments for tape sensitivity.
My deck has these external pots, so it's easy to tweek the deck.
>This is why Dolby is trying to push Dolby S -- it is just as effective as Dolby
>C, maybe more so, but is far less subject to tracking errors due to head
>misalignment. It also sounds quite acceptable when played through a Dolby B
I've read in "Audio" that Dolby Laboratories is being very picky as to what
tape decks will be allowed to use the Dolby S circuit. The manufacturer has
to make the prospective deck with very tight tolerances of head alignment, and
it also has to have wide frequency repsonse. Also, the deck has to have some
type of provision for tape sensitivity adjustment, either automatic or manual.
Dolby Labs learned from the problems of Dolby C to do this; I'm glad they are
doing it. Unfortunately, only high-end decks ($900+) contain Dolby S right now.
Subjective tests in "Audio" and "Stereo Review" of Dolby S decks rank them almost
as good as DAT. On an oscilloscope it was easy to tell the analog deck, but to
the ear, the reviewers said it sounded practically as good as DAT.
Tom
Bill Vermillion
>In article <1992Oct1.1...@pilot.dmg.ml.com>, ste...@pilot.dmg.ml.com (Steve Hannoford) writes:
here)
>dbx works by 'compressing' the entire signal by boosting the weak
signal more than the strong ones. Put into other words:
The dynamic range of the signal is squeezed together so that it
would fit into the much narrower dynamic range of the tape.
This results in a recording that would sound very unpleasent
without 'uncompression'. Therefore, dbx should also be used
during playback. The result is an improvement in the S/N-ratio
of aprox. 30dB.
dbx MUST be used in playback. Besides using a straight 2:1
compression on the entire tape, it also does some equalization.
Because of the compression it can put more high end boost in
place - SNs of over 100db are possible in pro machines.
>Because of the way these NR systems works properly, it is vital
that the tapedeck is correctly adjusted. (Bias & tape sensitivity
level). The heavier the alterations to the signal the NR
system performs, the more important is this matter.
dbx is the most effective, followed by Dolby C and Dolby B,
but because ov the reason mentioned above dbx is also most
critical to the tapedeck adjustment and it's therefore easier
to ruin a dbx recording than a Dolby B/C recording.
Wrong on that point. The Dolby units are most critical in
record and playback adjustment. There is a 'dolby' level that
is set in record, and MUST be matched on playback for proper
decode. Only signals below a given level are compressed and if
the playback level is set wrong the levels used to decode are
also wrong. dbx is relatively insensitive to level changes.
The worst problems I ever had with dbx were caused by 'head
bumps' in low frequency playback causing a disparate increase
in LF response at the bump center. To compensate we typicaly
notched on record to give linear playback. We didn't sweat
compatibility too much as there were fewer than 6 (by our
estimates) 32-track 2" analog machines in the states at that
time. (Between the 2 rooms we had 48 tracks of dbx. (Only
time that was a problem was if we needed 32 in A, and 24 in B
- but we worked our schedules to preclude that).
>Dolby HX Pro is not a NR-system, but a system to obtain constant
bias. Without Dolby HX Pro the treble will work together
with the bias, resulting in a varying bias. The more treble
the more bias, and more bias results in a reduction in the tapes
treble saturation level. By keeping the bias constant,
Dolby HX Pro tries to elliminate this effect resulting
in a higher treble saturation level and therefore better treble.
(<=> better frequency range)
To be more accurate - the system varies bias current because
excessive HF content cause the signal to be self-biasing. The
bias current is reduced to compensate so the the 'effective
bias' is constant .
Bias can be optimized for several parameters. The typical bias
points are set just beyond maximum output. As the bias
increases past MOL the output decreases. However the
distortion decreases. Overbias gives less distortion
sacrificing the high-frequency output. HX-Pros use is not so
much to give a 'higher treble saturation' but to give a linear
(or perhaps to state it more accurately) a consitant high
frequency response curve. Without the HX-Pro, the high
frequency response curve would vary depending on the signal
content.
(I spent most of my life around high-speed reel-reel so I
didn't have to sweat the slow stuff :-) )
--
Bill Vermillion - bi...@bilver.oau.org bill.ve...@oau.org
- bi...@bilver.uucp
- ..!{peora|ge-dab|tous|tarpit}!bilver!bill
Tom Mathes
>
>In article <1992Oct4....@cmkrnl.com> j...@cmkrnl.com writes:
>>Rather, Dolby B and C apply *compression* to the high frequencies on record,
>>and expansion on playback. Parts of the program that already have a lot of
>>high-freq. content won't be much affected by Dolby. But in parts in which the
>>highs are of low level, highs are boosted on record and cut on playback.
>>During the "cut on playback", the tape hiss, which is also mostly high freqs,
>>gets cut too. Voila. (When there is a lot of high-freq content in the
>>program material, you don't notice the hiss nearly so much, as you're paying
>>attention to the program. This is a psychoacoustic affect known as "masking".)
>
>You've got your theory right, but terminology wrong. Dolby amplifies, or
>expands the high frequencies during recording, then compresses them during
>playback. That's one reason to critically check the recording levels during
I'm eating crow on this one. I realized soon after I posted this that Jamie
was right and I was wrong. Dolby does amplify the high-frequencies, and does
indeed compress the dynamic range of those signals. I guess my brain was in
neutral when I wrote this.
Sorry about the screw-up part folks.
Tom
Norm Strong
}record only, and is not required during playback... Not only
}is it a better reduction system (results in less noise than
}either Dolby B/C), but it is also preferential because it
}works without the requirement that the implementation at
}playback match the implementation used to record....
Not true. All noise reduction schemes require complementary playback
circuits--ESPECIALLY dbx. It's HX-Pro that does not require any
playback compensation.
--
Norm Strong (str...@tc.fluke.com)
2528 31st S. Seattle WA 98144 USA
j...@cmkrnl.com
The source of confusion here may be that "compression" can mean either boosting
the signal or cutting it, or some of each, depending on the original signal
level and on compressor's parameters!
For example, "compressing" the dynamic range of a signal by 10:1 might mean
turning a 0-to-1.0 range into 0.0-0.10 (reducing the level), or into 0.9-1.0
(raising; this is what Dolby NR does to the high freqs on record, though not by
anything like a 10:1 ratio), or 0.45 to 0.55 (some inputs are boosted, others
are cut), or to 0.2 to 0.3 (most signals are cut, but the very lowest-level
ones are boosted)... The term "compression", as in "make smaller", applies not
to the signal level but to the dynamic range.
Anyway, you brought up a very valid point, which ought not to get lost in the
shuffle: use of Dolby NR requires that you pay close attention to record
level. When you're recording with Dolby (and who doesn't on cassette decks,
unless you have and prefer dbx?), it is important not to push the meters into
the red. Your deck's record level meters probably show the signal *before*
Dolby boosts the highs, and they may not respond well to the high frequencies
anyway! (A few cassette decks have "meter weighting" switches that cause the
meters to respond more to the highs, so that the "red" range corresponds better
to the hf saturation point of the tape.)
Bill Vermillion
I have had a CD player in my car for a little over 5 years. It
skips on some bumps in the road, but I have fairly stiff
suspension and tires.
I noticed that on one particular road I travel, sometimes it
skips, sometimes it doesn't.
It appears the discs that skip are usually the long discs, and
the ones that don't are much shorter in length.
Just a data point if anyone cares.
Bill Vermillion
>In article <1992Oct6.1...@oakhill.sps.mot.com>, rvk...@aus18a25.sps.mot.com (Tom Mathes) writes:
>> Sorry about the screw-up part folks.
>The source of confusion here may be that "compression" can mean either boosting
>the signal or cutting it, or some of each, depending on the original signal
>level and on compressor's parameters!
Jamie - in most audio circles compression means to limit the dynamic
range - make it smaller than the original. The reverse is usually
called expanison.
Devices which do both are often called 'companders'.
Tom Bruhns
>It appears the discs that skip are usually the long discs, and
>the ones that don't are much shorter in length.
>
>Just a data point if anyone cares.
Another data point: I have a player that skips more on the early
part of a disc than the later part. It's a D9 "Discman". It also
skips much more if it's horizontal than if it's at an angle, or
even vertical. Works fine till you get back to horizontal
upside down.
just a guy made of dots and lines
After overdosing on Halcion, bi...@bilver.uucp (Bill Vermillion) typed:
[ regarding skipping in car CD player ]
>It appears the discs that skip are usually the long discs, and
>the ones that don't are much shorter in length.
>
>Just a data point if anyone cares.
I had a car CD player for about two years (until yesterday -- it
was ripped off, sigh) and I'd noticed a similar correlation between
disc length and likelihood of skipping. In recent months, however,
it had started skipping on some discs even in the absence of road bumps;
these skips seemed less correlated to length.
-- Stewart
--
Is there a dry unoccupied aquarium?
That may be refilled with fish and tranquility.
-- Ernest Noyes Brookings
/* uunet!sco!stewarte -or- stew...@sco.COM -or- Stewart Evans */
William A Leeper
The data is encoded from the inside out. If you open the CD player and
insert a disk you will see that as the disk spins on the playhead the uneven
parts of the disk go up and down. This requires the read head to adjust very
rapidly to the motion of the disk. If the read head has dust or is slightly
damaged it will not be able to keep up with the motion. The only recomendation
for CD players with this problem is to have them cleaned or to clean it yourself
This newsgroup is abundant with cleaning suggestions.
--
: "To many good things happen to : Bill Leeper
: good people to be attributed to : wle...@en.ecn.purdue.edu
: anything but a carring God. " :
Carl A Haverl
Older CD players that start skipping can usually be rejuvenated with
one or two drops of teflon containing oil on the rails that the optical
assembly slides on. I use Tri-Flow, which is available in bike shops.
The aeresol is OK too, just don't hit the optics.
--
Carl Haverl
hav...@cats.ucsc.edu
305 Chace St.
Santa Cruz, CA 95060
(408) 458-3259
Bill Vermillion
>Older CD players that start skipping can usually be rejuvenated with
>one or two drops of teflon containing oil on the rails that the optical
>assembly slides on. I use Tri-Flow, which is available in bike shops.
>The aeresol is OK too, just don't hit the optics.
If you go back and read what I said when I started this new
title, you will see that I reported it skipped only on the
longer discs when I hit a bump. (This is a car player, a
Sanyo, about 5.5 years old - only one on the market I could
find then.)
I guess I expected that the readers would understand from what
I said.
My intuitive guess is that since the long discs achieve their
long lengths by using the minimum spacing tolerance in the
specifications, that these discs, because of the narrower
spacing of the groove, cause the mis-tracking by vibration,
whereas the discs cut with normal groove spacing are tolerant
of the vibrations.
Lon Stowell
>
>Older CD players that start skipping can usually be rejuvenated with
>one or two drops of teflon containing oil on the rails that the optical
>assembly slides on. I use Tri-Flow, which is available in bike shops.
>The aeresol is OK too, just don't hit the optics.
>
Anecdotally it seems to work.
M Eric Carr
>
>I have had a CD player in my car for a little over 5 years. It
>skips on some bumps in the road, but I have fairly stiff
>suspension and tires.
>
>I noticed that on one particular road I travel, sometimes it
>skips, sometimes it doesn't.
>
>It appears the discs that skip are usually the long discs, and
>the ones that don't are much shorter in length.
>
Perhaps it has something to do with the RPM of the CD -- the longer
CD's
would spin at slower RPM's at the end of the CD -- the shorter ones,
as the data is closer to the center, spin faster... Perhaps a
gyroscopic effect keeps the shorter CD's balanced better? Try
playing a long CD -- on the first few tracks (~10 minutes or less
into the CD). See if that helps.
P.S. I bought a Radio Shack CD mount ($20ish). People may like to
cut down Radio Shack (I know Realistic ain't Pioneer, Kenwood, Alpine,
etc.), but they're not all **that** bad. The CD mount (I have a Fisher
portable CD) works like a charm -- almost completely eliminates
skipping.
I think I've finally managed to replace the tape deck with a working
CD setup -- I can't STAND skipping: I even prefer tapes over that!
--Eric <ca...@xanth.cs.odu.edu>
--
Eric Carr
-- The Cyber Linguist --
j...@cmkrnl.com
ca...@chrysanthemum.cs.odu.edu (M Eric Carr) writes:
> In article <1992Oct13.1...@bilver.uucp> bi...@bilver.uucp (Bill Vermillion) writes:
>>It appears the discs that skip are usually the long discs, and
>>the ones that don't are much shorter in length.
>>
> Perhaps it has something to do with the RPM of the CD -- the longer
> CD's
> would spin at slower RPM's at the end of the CD -- the shorter ones,
> as the data is closer to the center, spin faster...
uh-uh. CDs are played from the outside in, but the unused portion is always
at the *outer* rim of the CD. So all CDs spin at the same speed at the end,
but they have different starting speeds -- longer ones spin slower at the
beginning (since the start is closer to the outer edge).
My guess -- and it's only that -- as to the mechanism that's at work here is
that any wobble in the CD's rotation will resort in more up-and-down movement
the closer you get to the outer edge. This means that, even in the absence of
bumps and jolts from the outside, the tracking mechanism has to work harder
at the outer edge, so it will have less "margin" when an outside jolt comes
along.
If you believe that adding mass will help, you can get something like a CD
Micro-Mat -- a semi-pliable, nonresonant disc with a tacky (as in
sort-of-adhesive, not cheap-looking!) underside. This goes on top of the CD
you're playing. Or you can simply try playing two discs at once, using the top
one as a weight. I'm a "bits is bits" person myself, so I don't think much of
such things (nor of green pens, etc.) as far as improving the sound quality
goes, but there's no denying that *if* you have a situation where making the
disc more massy will help, these things *may* help. At $40 it's a lot cheaper
than CD rings, too, since you only buy one.
On the other hand, making the disc more massy also means that the motor that
spins the CD will have to work a little harder to bring it up to speed and to
make the speed changes that are required to maintain CLV. I have no idea how
much "margin" CD drive motors have in this area.
Andrew Koenig
> uh-uh. CDs are played from the outside in, but the unused portion is always
> at the *outer* rim of the CD.
Nope; they're played from the inside out.
--
--Andrew Koenig
a...@europa.att.com
M Eric Carr
> ca...@chrysanthemum.cs.odu.edu (M Eric Carr) writes:
>> In article <1992Oct13.1...@bilver.uucp> bi...@bilver.uucp (Bill Vermillion) writes:
>>>[...]
>>>It appears the discs that skip are usually the long discs, and
>>>the ones that don't are much shorter in length.
>>>
>> Perhaps it has something to do with the RPM of the CD -- the longer
>> CD's
>> would spin at slower RPM's at the end of the CD -- the shorter ones,
>> as the data is closer to the center, spin faster...
>
>uh-uh. CDs are played from the outside in, but the unused portion is always
>at the *outer* rim of the CD. So all CDs spin at the same speed at the end,
>but they have different starting speeds -- longer ones spin slower at the
>beginning (since the start is closer to the outer edge).
>
re-adjust it. Maybe this one particular CD player and this one particular CD
were different, but it started playing the CD from the **inside**, and
as the track number was advanced from 1 to 10 (The cd had 10 tracks),
the lens moved **outward**. Which, due to this very "skipping"
problem, is how I would have designed the beasties myself. Good
explanation of why the outside part (whatever order it's played in)
skips... That's what I meant in my explanation, but I wasn't as clear.
Once again, tho, I think the "car cd player mounts" are WELL worth the
$20 or so they cost (Unless, of course, you went and bought a Sony Car
Discman 8-) )
j...@cmkrnl.com
How right you are. the unused portion is still on the outside. Sigh. I have
been making far more than my share of net.gaffes lately...
Bill Vermillion
>> Perhaps it has something to do with the RPM of the CD -- the longer
>> CD's
>> would spin at slower RPM's at the end of the CD -- the shorter ones,
>> as the data is closer to the center, spin faster...
>
>uh-uh. CDs are played from the outside in, but the unused portion is always
>at the *outer* rim of the CD. So all CDs spin at the same speed at the end,
>but they have different starting speeds -- longer ones spin slower at the
>beginning (since the start is closer to the outer edge).
Jamie - you have that one backwards. All CD's - as a matter of fact
all home laser products, CDs & LDs, read from the inside out. All have
the same starting speed, but the ending speed is different.
>My guess -- and it's only that -- as to the mechanism that's at work here is
>that any wobble in the CD's rotation will resort in more up-and-down movement
>the closer you get to the outer edge. This means that, even in the absence of
>bumps and jolts from the outside, the tracking mechanism has to work harder
>at the outer edge, so it will have less "margin" when an outside jolt comes
>along.
I have realized that my problem in communcation comes from the fact
that I don't give enough details. But they seemed obvious to me.
When I said it was the longer CD's that skipped, I meant that it was
the CD's that are longer than the orignal 74 minute specification, the
ones that cut to the low end tolerance and put more than 74 minutes on
the disk.
These will skip at different place, - eg a long cd can skip in the
middle - not just near the long end. That is why I suspect that the
player is more susceptible to jolts when the low end tolerance are at
the limit. It perhaps becomes easier to lose the tracking when cut at
the finer pitch. That is just a suposition. However it could be
just a fluke that that happens. It could be some other differences
and that I have only noticed them on the longer disks - and have not
had a shorter disk that exhibits the same phenomenom.
>If you believe that adding mass will help, you can get something like a CD
>Micro-Mat -- a semi-pliable, nonresonant disc with a tacky (as in
>sort-of-adhesive, not cheap-looking!) underside. This goes on top of the CD
>you're playing. Or you can simply try playing two discs at once, using the top
>one as a weight.
I don't believe in these, but in this case it would not work. THis is
an in-dash car unit - the one that skips on severe bumps only on long
disks - and the slot would not accept a disk with a ring on it - even
if I decided to do that.
A. J. Dean
: Jamie - you have that one backwards. All CD's - as a matter of fact
: all home laser products, CDs & LDs, read from the inside out. All have
: the same starting speed, but the ending speed is different.
will be less disc warp there, and thus it is a more reliable place to get
the initial focus position. After that the lens will stay pretty much in
focus. (from a sony book)
: These will skip at different place, - eg a long cd can skip in the
: middle - not just near the long end. That is why I suspect that the
: player is more susceptible to jolts when the low end tolerance are at
: the limit. It perhaps becomes easier to lose the tracking when cut at
: the finer pitch. That is just a suposition. However it could be
Apparently (again...) the track spacing (standard 1.6 micro m I think) was
determined because it places adjacent tracks close to a zero in the optical
system's response, minimising crosstalk from adjacent tracks. The 'zero'
occurs because of the diffraction-limited focussing, ie the spot focusses to
concentric rings rather than a blurred dot. I think the crosstalk can affect
the performance of the tracking servos before it upsets the data stream,
which would explain the skipping.
Antony. (dea...@elec.canterbury.ac.nz)
Bil Gonzalez
> Bill Vermillion (bi...@bilver.uucp) wrote:
> : Jamie - you have that one backwards. All CD's - as a matter of fact
> : all home laser products, CDs & LDs, read from the inside out. All have
> : the same starting speed, but the ending speed is different.
> True. Apparently the reason they start at the inside is/was because there
> will be less disc warp there, and thus it is a more reliable place to get
> the initial focus position. After that the lens will stay pretty much in
> focus. (from a sony book)
While it is true that they read from the inside out, it is not true
that the speed always changes. While CAV laser disks speed up from
beginning to end (~600? - 1800 rps), CLV (constant linear velocity)
maintains a 1800 rps speed during playback. I always thought that
this was true of audio CD also, which is the reason why CDV was in CLV
format. Does anyone know anything more about this?
Bil Gonzalez
bgo...@math.rutgers.edu
Farul A Ghazali
>
>While it is true that they read from the inside out, it is not true
>that the speed always changes. While CAV laser disks speed up from
>beginning to end (~600? - 1800 rps), CLV (constant linear velocity)
>maintains a 1800 rps speed during playback. I always thought that
Wait a minute. Correct me if I'm wrong but I'm pretty sure that CAV
discs maintain 1800rpm during playback while CLV discs vary from
600-1800 rpm. Audio CDs are spun at a varying rate, since they're CLV.
---> farul ghazali.
columbia university in the city of new york.
Bill Vermillion
>Bill Vermillion (bi...@bilver.uucp) wrote:
>: Jamie - you have that one backwards. All CD's - as a matter of fact
>: all home laser products, CDs & LDs, read from the inside out. All have
>: the same starting speed, but the ending speed is different.
>True. Apparently the reason they start at the inside is/was because there
>will be less disc warp there, and thus it is a more reliable place to get
>the initial focus position. After that the lens will stay pretty much in
>focus. (from a sony book)
But there is one other good reason, which was pioneered (no pun
intended) by LD industry in laser read information medium. Besides
having a consistant starting point, you can vary the size of the media
as the stopping point is encoded in the data stream.
This is why we can have 8" and 12" LDs and CD singles (3") and standard
= 12 cm.
Records for commecial use started from the outside, and you had to set
the size because they had to go to an ecentric groove for auto-stop (on
the very early units) to auto-eject - as occured later on.
When the LP's came out the mechanism for dropping from a center spindle
was perfected (after being show the way by the RCA 45) and a trip level
sensed the LP size on it's decent to the platter. Previously they
were pushed from the bottom of the stack.
Discs recorded for industrial (broadcast) use often were cut from the
inside out - partially for cutting convenience, some for musical
quality, and some sets were cut both ways to minimize audible
differnces when selections crossed from one disk to another.
Inside out makes a lot of sense.
CJ Herman
> While it is true that they read from the inside out, it is not true
> that the speed always changes. While CAV laser disks speed up from
> beginning to end (~600? - 1800 rps), CLV (constant linear velocity)
> maintains a 1800 rps speed during playback. I always thought that
> this was true of audio CD also, which is the reason why CDV was in CLV
> format. Does anyone know anything more about this?
You have these backwards. CAV (Constant Angular Velocity) implies a
constant angle is swept out over a specific period of time. For this
to occur the disk must spin at a constant RPM.
CLV operation requires that the optical sensor trace out a constant
linear distance per unit time. Since the circumference of the circle
traced by the head increases as the radial distance out from the
center increases, the disk must necessarily spin slower.
If you have any doubt that audio CD's operate this way, just find a
player with a window in it (most portables), step from the first to
last tracks and observe how fast the disk spins.
--
-------------------------------------------------------------------------------
|"In general, thinking about hi-fi equipment |I don't think my | C.J. Herman |
| before listening can be dangerous." |employer cares | c...@kentek.com|
| -- Anonymous cable manufacturer |about disclaimers| Boulder, CO |
Bil Gonzalez
> You have these backwards. CAV (Constant Angular Velocity) implies a
> If you have any doubt that audio CD's operate this way, just find a
> player with a window in it (most portables), step from the first to
> last tracks and observe how fast the disk spins.
Yes, You are all right, I did mix this up. Sorry, I'm feeling pretty
stupid. I was half asleep when I responded. I have a portable cd
played, designed for 3" cds, and have watched the disc rotation slow
down as it plays....I guess I just wasn't thinking clearly, but I knew
that one of the formats had a constant rotational speed.
I stand corrected.
Bil Gonzalez
Peter Wood
Bil> While it is true that they read from the inside out, it is not true that
Bil> the speed always changes. While CAV laser disks speed up from beginning
Bil> to end (~600? - 1800 rps), CLV (constant linear velocity) maintains a
Bil> 1800 rps speed during playback. I always thought that this was true of
Bil> audio CD also, which is the reason why CDV was in CLV format. Does
Bil> anyone know anything more about this?
I don't know if anyone knows anything more, but we certainly know something!
CAV means Constant Angular Velocity. That means that the disk rotates at the
same speed, so its angular velocity, strangely enough, is constant.
CLV means Constant Linear Velocity. That means that whatever part of the
disk is being read is passing under (or strictly speaking over in the case of
most CD players) the head at a constant speed. That implies that as the head
moves from the inside to the outside of the disk, Angular Velocity must
decrease (ie it slows down!).
Figures we have seen bandied about on this net suggest CD speeds of the order
of 600-250 rpm or thereabouts. Not 1800rps, which might be a bit quick!
Pete
--
-----------------------------------------------------------------------------
- Peter Wood, Nokia Telecommunications, Transmission Product Development -
- Cambridge, UK. Tel: +44 223 423 123 -
- External Email: wo...@uk.tele.nokia.fi Internal Email: NTL02::WOOD -
-----------------------------------------------------------------------------
A. J. Dean
: In article <BwI7x...@cantua.canterbury.ac.nz>, dea...@elec.canterbury.ac.nz (A. J. Dean) writes:
: > True. Apparently the reason they start at the inside is/was because there
: > will be less disc warp there, and thus it is a more reliable place to get
: > the initial focus position. After that the lens will stay pretty much in
: > focus. (from a sony book)
:
: While it is true that they read from the inside out, it is not true
: that the speed always changes. While CAV laser disks speed up from
: beginning to end (~600? - 1800 rps), CLV (constant linear velocity)
: maintains a 1800 rps speed during playback. I always thought that
: this was true of audio CD also, which is the reason why CDV was in CLV
: format. Does anyone know anything more about this?
[whoops, bit late] What are laser discs? Nah, I do know, but they are
extremely rare in this country - even now they are only really used in video
games, karioke, videophile's homes, etc. One of the most popular CD shops in
town here has about 4 of them total!
BTW (many others may have corrected you by now...) CLV discs (like CD)
change speed, CAV discs rotate at constant Angular speed.
Antony. (dea...@elec.canterbury.ac.nz
Cary D Renzema
the speed of CDs. This is what I have.
Scanning Velocity (CLV): 1.2 - 1.4 m/s
Starting Radius: 25 mm
Maximum Ending Radius: 58 mm
Given the above and with a little geometry and math the following table
has the maximum and minimum RPMs for the scanning velocity extremes.
S.V. Starting RPM Ending RPM
---- ------------ ----------
1.2 458 198
1.4 535 231
Another intersting point is that given the two radiuses, the minimum
scanning velocity and a minimum track pitch of 1.5 um (1.6 - .1 um) you get
a maximum play time of under 80 minutes. Now I know I've heard of CDs with
longer play times then that, so somebody's pushing the specs to get it!
Cary D. Renzema
ca...@tekig6.PEN.TEK.COM
Tom Moellering
>Another intersting point is that given the two radiuses, the minimum
>scanning velocity and a minimum track pitch of 1.5 um (1.6 - .1 um) you get
>a maximum play time of under 80 minutes. Now I know I've heard of CDs with
>longer play times then that, so somebody's pushing the specs to get it!
The way you get 80 minutes has to do with the built-in slop in the track
pitch spec. It was put there over 10 years ago, but manufacturers have
gotten more and more talented and now they can actually keep the track
pitch close enough to its minimum value from the inside all the way to
the edge, and over the whole disc you do gain a couple of seconds.
Tom (moe...@convex.com)