Holst and a Headache
Albert Ross
that the BBC can still do technical excellence when it wants to.
Did anyone else watch a recent programme on BBC4 about Gustav Holst?
It was very interesting, I didn't know much about him, unlike Vaughan
Williams and Cecil Sharp, and only really knew The Planets.
The programme was made up of current interviews and concert footage,
interspersed with vintage clips.
All the clips were in 4:3 and inserted into the programme in
stretchyvision.
At first I thought either my TV or the Toppy had fallen over and kept
flicking through the various aspect ratios before I twigged the
programme was actually MADE like this.
Long before the end this was making my brain hurt.
Is it now too expensive to scan-and-pan or whatever they call it in
these digital days?
Dave Plowman (News)
Albert Ross <spam@dev_null.com.invalid> wrote:
> Is it now too expensive to scan-and-pan or whatever they call it in
> these digital days?
Doubt it. But the program makers simply don't care.
--
*Forget about World Peace...Visualize using your turn signal.
Dave Plowman da...@davenoise.co.uk London SW
To e-mail, change noise into sound.
Roderick Stewart
Regrettably it's the same on the published DVD, and also on the
otherwise excellent programme about Ralph Vaughan Williams by the same
production team. It's annoying because it's such a simple thing to get
right but I guess either they don't care or they're employing people
who don't know how. Best just to listen to the music.
Rod.
--
Virtual Access V6.3 free usenet/email software from
http://sourceforge.net/projects/virtual-access/
Max Demian
news:rfias6hq92ebtutk0...@4ax.com...
Usually they just chop the top and bottom off to turn 4:3 into 16:9. Then if
you watch it on a 4:3 set you can chop the left and right off and end up
with a little postage stamp area of the original image.
"Pan and scan" is a way to show a widescreen cinema film on a non-widescreen
TV, not the other way round. There are a lot of older films around that are
still being shown from this stock as they haven't got round to putting the
original films through a modern telecine machine.
--
Max Demian
J G Miller
> I have to admit the Royal Wedding was pretty excellently done, proof
> that the BBC can still do technical excellence when it wants to.
Only with the excellence of the technical assistance and facilities of SIS though.
J G Miller
> I have to admit the Royal Wedding was pretty excellently done, proof
> that the BBC can still do technical excellence when it wants to.
Only with the excellence of the technical assistance and facilities of SIS though.
Jim Lesurf
> Did anyone else watch a recent programme on BBC4 about Gustav Holst?
Yes.
> It was very interesting, I didn't know much about him, unlike Vaughan
> Williams and Cecil Sharp, and only really knew The Planets.
I can recommend the old 'Lyrita' Boult recordings of items by Holst. The
newer ones by Hickox on Chandos are also probably good. Alas, BBC R3 have
tended to overlook Holst even more than they have overlooked VW over the
years. Still, at least they notice The Planets. Alas others like Rubbra
almost never get a hearing. Holst's Choral Symphony was performed at a
recent Prom, though. But was ignored by BBC4 so not on TV.
> The programme was made up of current interviews and concert footage,
> interspersed with vintage clips.
> All the clips were in 4:3 and inserted into the programme in
> stretchyvision.
[snip]
> Is it now too expensive to scan-and-pan or whatever they call it in
> these digital days?
IIRC other documentiaries produced by the same film-miker also
streatchtvisioned inserts. This looked really daft to me. They also had the
same kind of images of things like the orchestra playing with wildly
emphasised 'light and shadow'. As a result I found the content interesting,
the music enjoyable, but some of the images annoying. I suspect this is the
maker just being 'arty', though. The aim may be to keep reminding you "this
is a film". Priority of format or presenter over content.
I just tend to regard it as an audio documentary with occasional bits of
video that are worth seeing. Probably not the reaction the film-maker
wishes... :-)
Slainte,
Jim
--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scots_Guide/intro/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html
Paul Ratcliffe
> Only with the excellence of the technical assistance and facilities of SIS though.
Can you stop your double posting (approx. 30s apart) please.
It's rather tiresome.
J G Miller
> Can you stop your double posting (approx. 30s apart) please. It's rather
> tiresome.
I appreciate that you are seriously tolerance-challenged and offer my
most humble apologies for causing you unecessary mental trauma, but
please rest assured that it is not intentional.
Albert Ross
wrote:
>In article <rfias6hq92ebtutk0...@4ax.com>, Albert Ross
><spam@dev_null.com.invalid> wrote:
>
>> Did anyone else watch a recent programme on BBC4 about Gustav Holst?
>
>Yes.
>
>> It was very interesting, I didn't know much about him, unlike Vaughan
>> Williams and Cecil Sharp, and only really knew The Planets.
>
>I can recommend the old 'Lyrita' Boult recordings of items by Holst. The
>newer ones by Hickox on Chandos are also probably good. Alas, BBC R3 have
>tended to overlook Holst even more than they have overlooked VW over the
>years. Still, at least they notice The Planets. Alas others like Rubbra
>almost never get a hearing. Holst's Choral Symphony was performed at a
>recent Prom, though. But was ignored by BBC4 so not on TV.
Oh I loved Boult doing Vaughan Williams. He was Da Man.
Yes over the years when I listened to Radio Three I came across a
whole bunch of British composers whose names have now faded back into
the mists, some of them were wildly underrestimated.
>> The programme was made up of current interviews and concert footage,
>> interspersed with vintage clips.
>
>> All the clips were in 4:3 and inserted into the programme in
>> stretchyvision.
>[snip]
>
>> Is it now too expensive to scan-and-pan or whatever they call it in
>> these digital days?
>
>IIRC other documentiaries produced by the same film-miker also
>streatchtvisioned inserts. This looked really daft to me. They also had the
>same kind of images of things like the orchestra playing with wildly
>emphasised 'light and shadow'. As a result I found the content interesting,
>the music enjoyable, but some of the images annoying. I suspect this is the
>maker just being 'arty', though. The aim may be to keep reminding you "this
>is a film". Priority of format or presenter over content.
>
>I just tend to regard it as an audio documentary with occasional bits of
>video that are worth seeing. Probably not the reaction the film-maker
>wishes... :-)
Ah, I couldn't recall seeing anything by the same guy. Obviously one
to look out for - and ignore.
Jim Lesurf
<spam@dev_null.com.invalid> wrote:
> On Sat, 07 May 2011 16:09:29 +0100, Jim Lesurf <no...@audiomisc.co.uk>
> wrote:
> >
> >> It was very interesting, I didn't know much about him, unlike Vaughan
> >> Williams and Cecil Sharp, and only really knew The Planets.
> >
> >I can recommend the old 'Lyrita' Boult recordings of items by Holst.
> Oh I loved Boult doing Vaughan Williams. He was Da Man.
I've tended to prefer Barbirolli for items like the 3rd symphony or Tallis.
And his versions of the Sinfonia Antartica and Tuba/Oboe Concertos seem
great to me. [Hint. Try the Barbirolli Society CDs. :-) ] But I also like
Boult's versions.
[snip]
> >
> >I just tend to regard it as an audio documentary with occasional bits
> >of video that are worth seeing. Probably not the reaction the
> >film-maker wishes... :-)
> Ah, I couldn't recall seeing anything by the same guy. Obviously one to
> look out for - and ignore.
I like the content of his films. But dislike the visuals of some of the
more recent examples. But then I feel the same way about a lot of modern TV
documentaries. Cue the standard examples, wobblycam, saturateovision, etc
Roderick Stewart
> > >I can recommend the old 'Lyrita' Boult recordings of items by Holst.
>
> > Oh I loved Boult doing Vaughan Williams. He was Da Man.
>
> I've tended to prefer Barbirolli for items like the 3rd symphony or Tallis.
> And his versions of the Sinfonia Antartica and Tuba/Oboe Concertos seem
> great to me. [Hint. Try the Barbirolli Society CDs. :-) ] But I also like
> Boult's versions.
Try Bernard Haitink for the 8th and 9th.
Alan White
<rj...@escapetime.removethisbit.myzen.co.uk> wrote:
>Try Bernard Haitink for the 8th and 9th.
Andrew Davis.
Bernard Haitink for 'A Sea'
--
Alan White
Mozilla Firefox and Forte Agent.
By Loch Long, twenty-eight miles NW of Glasgow, Scotland.
Webcam and weather:- http://windycroft.co.uk/weather
Jim Lesurf
Roderick Stewart <rj...@escapetime.removethisbit.myzen.co.uk> wrote:
> In article <51d354b...@audiomisc.co.uk>, Jim Lesurf wrote:
> > > >I can recommend the old 'Lyrita' Boult recordings of items by Holst.
> >
> > > Oh I loved Boult doing Vaughan Williams. He was Da Man.
> >
> > I've tended to prefer Barbirolli for items like the 3rd symphony or
> > Tallis. And his versions of the Sinfonia Antartica and Tuba/Oboe
> > Concertos seem great to me. [Hint. Try the Barbirolli Society CDs. :-)
> > ] But I also like Boult's versions.
> Try Bernard Haitink for the 8th and 9th.
Afraid I found Haitink somewhat 'clinical' to my ears when it comes to
British composers. However this may be because of the fashion we went
though for 'by the book' performances that pleased musicological academics.
Or maybe just my habituation to the older versions by then. I still like
some of the old MFP/CFP recordings of other things because they were all I
could afford at the time.
I tend to prefer an older style like that of Barbirolli or Boult. Or even
Mr Preview and his Band for the 7th. The downside is that some of the older
recordings can have poor sound. Frustratingly, in some cases because PRT
[1] didn't take any care with the master tapes so they deteriorated by the
time people wanted to do re-issues. Sadly ironic considering that they
employed the 'Mercury' team for some of the recordings.
To pile even more irony on that, some of the very early EMI stereo
recordings that have appeared on CFP are of remarkably good sound quality.
Slainte,
Jim
[1] Best regarded as 'Purveyors of Retreads', perhaps. :-)
Roderick Stewart
> > > > >I can recommend the old 'Lyrita' Boult recordings of items by Holst.
> > >
> > > > Oh I loved Boult doing Vaughan Williams. He was Da Man.
> > >
> > > I've tended to prefer Barbirolli for items like the 3rd symphony or
> > > Tallis. And his versions of the Sinfonia Antartica and Tuba/Oboe
> > > Concertos seem great to me. [Hint. Try the Barbirolli Society CDs. :-)
> > > ] But I also like Boult's versions.
>
> > Try Bernard Haitink for the 8th and 9th.
>
> Afraid I found Haitink somewhat 'clinical' to my ears when it comes to
> British composers. However this may be because of the fashion we went
> though for 'by the book' performances that pleased musicological academics.
> Or maybe just my habituation to the older versions by then. I still like
> some of the old MFP/CFP recordings of other things because they were all I
> could afford at the time.
Agreed Barbirolli's recordings are superb, despite the sound quality, but
being accustomed to RVW2 & 8 by him, and Boult's 9th, made Haitink's
recording of the 8th and 9th sound almost like new works. The same sort of
thing happened with Beethoven's 5th, which I thought I knew (as everybody
does) until I heard the Ben Zander recording. Just hearing a different
interpretation of a familiar work can often enable you to notice things you
never realised were there.
Albert Ross
wrote:
>In article <ed1ts6tqh30e5k4ou...@4ax.com>, Albert Ross
><spam@dev_null.com.invalid> wrote:
>> On Sat, 07 May 2011 16:09:29 +0100, Jim Lesurf <no...@audiomisc.co.uk>
>> wrote:
>
>> >
>> >> It was very interesting, I didn't know much about him, unlike Vaughan
>> >> Williams and Cecil Sharp, and only really knew The Planets.
>> >
>> >I can recommend the old 'Lyrita' Boult recordings of items by Holst.
>
>> Oh I loved Boult doing Vaughan Williams. He was Da Man.
>
>I've tended to prefer Barbirolli for items like the 3rd symphony or Tallis.
>And his versions of the Sinfonia Antartica and Tuba/Oboe Concertos seem
>great to me. [Hint. Try the Barbirolli Society CDs. :-) ] But I also like
>Boult's versions.
Boult was delightfully cool and understated. Yes I've been surprised
by some of the people who've "done" VW in significantly different
ways, surprisingly including Bernstein.
>I like the content of his films. But dislike the visuals of some of the
>more recent examples. But then I feel the same way about a lot of modern TV
>documentaries. Cue the standard examples, wobblycam, saturateovision, etc
You got that right!
I think digital TV is a good analogy for many things that are wrong
with the modern world. They remove as much of the signal as they can
and then emulate its existence with gadgetry.
Albert Ross
White<alan....@windycroft.co.uk> wrote:
>On Sun, 15 May 2011 10:17:26 +0100, Roderick Stewart
><rj...@escapetime.removethisbit.myzen.co.uk> wrote:
>
>>Try Bernard Haitink for the 8th and 9th.
>
>Andrew Davis.
Oh yes, I like his approach to many composers.
I've recorded some proms in recent years including a couple of
Russians whose names I can't currently recall. One of them made
Stravinsky's Rite Of Spring sound an order of magnitude better than I
ever heard it before. Like some of the best sound engineers they bring
out stuff that was always there but you never noticed before.
Analytical but passionate.
Jim Lesurf
<spam@dev_null.com.invalid> wrote:
> On Sun, 15 May 2011 11:42:41 +0100, Alan
> White<alan....@windycroft.co.uk> wrote:
> >On Sun, 15 May 2011 10:17:26 +0100, Roderick Stewart
> ><rj...@escapetime.removethisbit.myzen.co.uk> wrote:
> >
> >>Try Bernard Haitink for the 8th and 9th.
> >
> >Andrew Davis.
> Oh yes, I like his approach to many composers.
I quite like his older Sibelius recordings. But my favourite Sibelius
concert was actually a Rattle/CBSO performance given back in the 1980s.
> I've recorded some proms in recent years including a couple of Russians
> whose names I can't currently recall. One of them made Stravinsky's Rite
> Of Spring sound an order of magnitude better than I ever heard it
> before.
I'll Guess Gergiev. Must admit I tend to have the feeling that he gets the
'electric attention' effect by simply not really rehearsing with the
orchestra and then not giving them clear signals on timing. The result
seems to me to simply dump onto the orchestra the load of making the music
and keeping together, then he turns around and takes the credit at the end!
:-)
Probably not like Solti or Reiner or Toscanini, though! I doubt a modern
orchestra like the LSO or BBCx would put up with that.
Alan White
wrote:
>I quite like his older Sibelius recordings.
Weren't they Colin Davis?
Phil Cook
> On Mon, 16 May 2011 13:28:34 +0100, Jim Lesurf<no...@audiomisc.co.uk>
> wrote:
>
>> I quite like his older Sibelius recordings.
>
> Weren't they Colin Davis?
>
Both have recorded Sibelius, Andrew on CBS Colin on RCA. Colin is
reckoned to be a great Sibelian amongst other things.
--
Phil Cook
Jim Lesurf
White<alan....@windycroft.co.uk> wrote:
> On Mon, 16 May 2011 13:28:34 +0100, Jim Lesurf <no...@audiomisc.co.uk>
> wrote:
> >I quite like his older Sibelius recordings.
> Weren't they Colin Davis?
Oops! Yes, sorry. Misread what had been written.
Yes, I very much like Andrew Davis for many composers, particularly UK ones
like Elgar. Really enjoyed his Proms in recent years. And although I am
usually wary of 'finished' or 'constructed' works I found his performances
of the Elgar 3rd spellbinding.
FWIW I also really like Hickox's performances. Great shame that he died so
soon. Which brings me back to VW again. For example his VW 'London'
Symphony on Chandos is a stunning performance and recording.
bugbear
> Is it now too expensive to scan-and-pan or whatever they call it in
> these digital days?
While we're griping, I really HATE
when rostrum camera work is done digitally
with low resolution JPEG images - so that
when you're zoomed in you can see both pixels
and massive quantisation artefacts.
Guys - when you're doing digital rostrum work,
use high res, non lossy images. Or just get
Ken Morse in.
Thank you.
BugBear
Java Jive
end of major series like 'Civilisation', and, I think, 'The Ascent Of
Man', 'The Age Of Uncertainty', etc. That was long before I knew what
a rostrum camera was. I think I only noticed the name because I'd
been struggling with Morse Code a year or two previously.
On Tue, 17 May 2011 11:34:16 +0100, bugbear
<bugbear@trim_papermule.co.uk_trim> wrote:
>
> Guys - when you're doing digital rostrum work,
> use high res, non lossy images. Or just get
> Ken Morse in.
--
=========================================================
Please always reply to ng as the email in this post's
header does not exist. Or use a contact address at:
http://www.macfh.co.uk/JavaJive/JavaJive.html
http://www.macfh.co.uk/Macfarlane/Macfarlane.html
Jim Lesurf
bugbear
<
> Guys - when you're doing digital rostrum work, use high res, non lossy
> images. Or just get Ken Morse in.
Some years ago my wife and I noticed how often Ken Morse did the rostrum
work for a programme and became quite impressed by the low-key way he did
such a neat job. Is he still alive and well? I can't recall seeing his name
on a new programme for some time. Presumably production companies now
adopt the "we have a computer so someone here can do it" attitude... bit
like some of the messes that pass as 'graphic design' because someone has a
Mac and think this makes them able to layout documents well. :-/
IIRC They once had a programme that featured KM and his work.
Jim
> In article<j7qdnTYyVJK10k_Q...@brightview.co.uk>,
> bugbear
> <
>> Guys - when you're doing digital rostrum work, use high res, non lossy
>> images. Or just get Ken Morse in.
>
> Some years ago my wife and I noticed how often Ken Morse did the rostrum
> work for a programme and became quite impressed by the low-key way he did
> such a neat job. Is he still alive and well? I can't recall seeing his name
> on a new programme for some time. Presumably production companies now
> adopt the "we have a computer so someone here can do it" attitude... bit
> like some of the messes that pass as 'graphic design' because someone has a
> Mac and think this makes them able to layout documents well. :-/
>
> IIRC They once had a programme that featured KM and his work.
>
> Slainte,
>
> Jim
>
IMDb lists his last credits as 2009.
Dave Plowman (News)
Jim Lesurf <no...@audiomisc.co.uk> wrote:
> Some years ago my wife and I noticed how often Ken Morse did the rostrum
> work for a programme and became quite impressed by the low-key way he
> did such a neat job. Is he still alive and well? I can't recall seeing
> his name on a new programme for some time.
I'd hope he made enough money to be enjoying a happy retirement now - as
I'd guess he's way beyond working age.
--
*Hang in there, retirement is only thirty years away! *
bugbear
> In article<51d4c4b...@audiomisc.co.uk>,
> Jim Lesurf<no...@audiomisc.co.uk> wrote:
>> Some years ago my wife and I noticed how often Ken Morse did the rostrum
>> work for a programme and became quite impressed by the low-key way he
>> did such a neat job. Is he still alive and well? I can't recall seeing
>> his name on a new programme for some time.
>
> I'd hope he made enough money to be enjoying a happy retirement now - as
> I'd guess he's way beyond working age.
born 1944, according to wikipedia.
Also found:
http://www.guardian.co.uk/notesandqueries/query/0,5753,-2039,00.html
BugBear
Jim Lesurf
<da...@davenoise.co.uk> wrote:
> In article <51d4c4b...@audiomisc.co.uk>, Jim Lesurf
> <no...@audiomisc.co.uk> wrote:
> > Some years ago my wife and I noticed how often Ken Morse did the
> > rostrum work for a programme and became quite impressed by the low-key
> > way he did such a neat job. Is he still alive and well? I can't recall
> > seeing his name on a new programme for some time.
> I'd hope he made enough money to be enjoying a happy retirement now - as
> I'd guess he's way beyond working age.
I've been beyond working age since I was about 14. :-)
Albert Ross
wrote:
>In article <fe02t6d7dsl2t0ibn...@4ax.com>, Albert Ross
><spam@dev_null.com.invalid> wrote:
>> On Sun, 15 May 2011 11:42:41 +0100, Alan
>> White<alan....@windycroft.co.uk> wrote:
>
>> >On Sun, 15 May 2011 10:17:26 +0100, Roderick Stewart
>> ><rj...@escapetime.removethisbit.myzen.co.uk> wrote:
>> >
>> >>Try Bernard Haitink for the 8th and 9th.
>> >
>> >Andrew Davis.
>
>> Oh yes, I like his approach to many composers.
>
>I quite like his older Sibelius recordings. But my favourite Sibelius
>concert was actually a Rattle/CBSO performance given back in the 1980s.
Oh yes he's another one who can bring a new perspective onto various
composers.
>> I've recorded some proms in recent years including a couple of Russians
>> whose names I can't currently recall. One of them made Stravinsky's Rite
>> Of Spring sound an order of magnitude better than I ever heard it
>> before.
>
>I'll Guess Gergiev. Must admit I tend to have the feeling that he gets the
>'electric attention' effect by simply not really rehearsing with the
>orchestra and then not giving them clear signals on timing. The result
>seems to me to simply dump onto the orchestra the load of making the music
>and keeping together, then he turns around and takes the credit at the end!
>:-)
<G> you may be right, but that fits with my physiology, I love jazz
and improvisation in general.
Quote from Thelonious Monk, when asked how often his quartet rehearsed
"Rehearse? You mean CHEAT???"
I have a fetish for recordings of live performances for a similar
reason, and off the wall stuff. Somewhere I have a record of Sonny
Rollins playing a bunch of jazz classics, including Swing Low Sweet
Chariot featuring very bluesy African bagpipes, which I didn't
previously know existed.
>Probably not like Solti or Reiner or Toscanini, though! I doubt a modern
>orchestra like the LSO or BBCx would put up with that.
Ah, Solti was another of my favorites.
Be interesting to hear who's on this year's Proms.
Albert Ross
wrote:
>In article <8jm2t6d0md7sj2cav...@4ax.com>, Alan
>White<alan....@windycroft.co.uk> wrote:
>> On Mon, 16 May 2011 13:28:34 +0100, Jim Lesurf <no...@audiomisc.co.uk>
>> wrote:
>
>> >I quite like his older Sibelius recordings.
>
>> Weren't they Colin Davis?
>
>
>Oops! Yes, sorry. Misread what had been written.
Ah yes, he had IMO a narrower range but was good at what he was good
at.
>Yes, I very much like Andrew Davis for many composers, particularly UK ones
>like Elgar. Really enjoyed his Proms in recent years. And although I am
>usually wary of 'finished' or 'constructed' works I found his performances
>of the Elgar 3rd spellbinding.
Yes, Andrew goes out of his comfort zone more often and to great
effect.
>FWIW I also really like Hickox's performances. Great shame that he died so
>soon. Which brings me back to VW again. For example his VW 'London'
>Symphony on Chandos is a stunning performance and recording.
Ooh, don't think I've heard that one. It's too easy to make early VW
and other Brit composers sound like film music. Neville
Mariner/Academy of St Martin in the Fields did an excellent bunch of
his minor works including the Tallis Fantasia and Lark Ascending which
were stand-out performances.
Albert Ross
<bugbear@trim_papermule.co.uk_trim> wrote:
Good point, that name takes me back.
Phil Cook
http://www.bbc.co.uk/proms/whats-on/2011/performers
http://www.bbc.co.uk/proms/whats-on/2011/composers
<takes a shuftie>
Bax...
Hmmm, I wonder which tone poem is getting an airing this year....
<click><click>
<falls off chair>
SYMPHONY No.2!!
Tuesday 16 August
7.00pm – c. 10.10pm
Royal Albert Hall
Copland
Fanfare for the Common Man (3 mins)
Bax
Symphony No. 2 (39 mins)
INTERVAL
Barber
Adagio for strings (8 mins)
Bartók
Piano Concerto No. 2 (28 mins)
INTERVAL
Prokofiev
Symphony No. 4 in C major (revised version,1947) (37 mins)
Yuja Wang piano
Royal Philharmonic Orchestra
Andrew Litton conductor
--
Phil Cook
Jim Lesurf
<spam@dev_null.com.invalid> wrote:
> On Tue, 17 May 2011 09:45:05 +0100, Jim Lesurf <no...@audiomisc.co.uk>
> wrote:
> >FWIW I also really like Hickox's performances. Great shame that he died
> >so soon. Which brings me back to VW again. For example his VW 'London'
> >Symphony on Chandos is a stunning performance and recording.
> Ooh, don't think I've heard that one.
Recommended. Not only is it a very good performance and recording. It is
also the first modern recording of the orginal version of the London
Symphony. Rather longer than the version people have become accustomed to.
It can be a bit of a surprise if you are used to the established version.
But since you like improvisation the 'swerves and addition' (actually
replaced parts that were removed, etc) can be really interesting.
Great shame that Hickox died before he could do the entire VW set. The
Chandos discs also have a number of 'first recordings' of other VW items.
> It's too easy to make early VW and other Brit composers sound like film
> music.
Alas an entire generation of UK composers went though a time of being
performed in the 'cow looking over a gate' style. View inflicted on us by
an establishment period when 'experimental' was all the rage and had to be
obviously so.
Quite ironic when you realise what the VW pastoral symphony is *really*
about. Think Owen and "Bugles calling for them from sad shires". Britten's
War Requiem by more subtle means.
> Neville Mariner/Academy of St Martin in the Fields did an excellent
> bunch of his minor works including the Tallis Fantasia and Lark
> Ascending which were stand-out performances.
Agreed. I also like the old 'Argo' recordings he did. But in the end prefer
Barbirolli for works like the Tallis.
Albert Ross
wrote:
>In article <1v5ft6popcnvaq88v...@4ax.com>, Albert Ross
><spam@dev_null.com.invalid> wrote:
>> On Tue, 17 May 2011 09:45:05 +0100, Jim Lesurf <no...@audiomisc.co.uk>
>> wrote:
>
>
>> >FWIW I also really like Hickox's performances. Great shame that he died
>> >so soon. Which brings me back to VW again. For example his VW 'London'
>> >Symphony on Chandos is a stunning performance and recording.
>
>> Ooh, don't think I've heard that one.
>
>Recommended. Not only is it a very good performance and recording. It is
>also the first modern recording of the orginal version of the London
>Symphony. Rather longer than the version people have become accustomed to.
>It can be a bit of a surprise if you are used to the established version.
>But since you like improvisation the 'swerves and addition' (actually
>replaced parts that were removed, etc) can be really interesting.
>
>Great shame that Hickox died before he could do the entire VW set. The
>Chandos discs also have a number of 'first recordings' of other VW items.
When I find some more circular tuits I might look that out.
>> It's too easy to make early VW and other Brit composers sound like film
>> music.
>
>Alas an entire generation of UK composers went though a time of being
>performed in the 'cow looking over a gate' style. View inflicted on us by
>an establishment period when 'experimental' was all the rage and had to be
>obviously so.
>
>Quite ironic when you realise what the VW pastoral symphony is *really*
>about. Think Owen and "Bugles calling for them from sad shires". Britten's
>War Requiem by more subtle means.
If you dig into the music it's actually pretty "modern" but the
stunning discords etc. don't sound as striking in the context of the
rest of the music as they would if someone else used them (true also
of Holst), there's something very disturbing under the surface beauty.
One of my all time favourite pieces.
>> Neville Mariner/Academy of St Martin in the Fields did an excellent
>> bunch of his minor works including the Tallis Fantasia and Lark
>> Ascending which were stand-out performances.
>
>Agreed. I also like the old 'Argo' recordings he did. But in the end prefer
>Barbirolli for works like the Tallis.
I just remembered, somewhere I have a recording of VW himself
conducting the Fourth. Apart from the generally lousy sound he seems
to play it much more "straight" than some who have come since. True
also of Rachmaninov playing his own stuff (piano rolls), without the
usual hadwringing emotions it actually sounds better. Bartok was a
bugger though, he timed each part of the music to the microsecond but
when he played or conducted himself he really let loose.
Albert Ross
That sounds like a nice collection.
Ah, they're doing the first three of Bartok's three piano concertos
<G> they don't play the First as often as I'd like.
I'll look through the rest of the lists later, I can smell cooking . .
.
Jim Lesurf
<spam@dev_null.com.invalid> wrote:
> On Sat, 21 May 2011 13:52:46 +0100, Jim Lesurf <no...@audiomisc.co.uk>
> wrote:
> >Great shame that Hickox died before he could do the entire VW set. The
> >Chandos discs also have a number of 'first recordings' of other VW
> >items.
> When I find some more circular tuits I might look that out.
FWIW I've just got a copy of the London Symphony as 96k/24bit flac files.
I keep trying to decide if they sound any better than the CD. But when I
try I just end up becoming lost in the music. Ho hum, investigating audio
can be such a strain at times... 8-]
Albert Ross
wrote:
>In article <481qu618t30ftuk2k...@4ax.com>, Albert Ross
><spam@dev_null.com.invalid> wrote:
>> On Sat, 21 May 2011 13:52:46 +0100, Jim Lesurf <no...@audiomisc.co.uk>
>> wrote:
>
>> >Great shame that Hickox died before he could do the entire VW set. The
>> >Chandos discs also have a number of 'first recordings' of other VW
>> >items.
>
>> When I find some more circular tuits I might look that out.
>
>FWIW I've just got a copy of the London Symphony as 96k/24bit flac files.
>I keep trying to decide if they sound any better than the CD. But when I
>try I just end up becoming lost in the music. Ho hum, investigating audio
>can be such a strain at times... 8-]
By a strange coincidence it was used as part of the backing track to
something I watched the other day, Coast or similar. Now I have one of
his tunes stuck in my head.
Java Jive
Lossless Audio Codec after all?
On Mon, 06 Jun 2011 18:16:50 +0100, Jim Lesurf <no...@audiomisc.co.uk>
wrote:
>
> FWIW I've just got a copy of the London Symphony as 96k/24bit flac files.
> I keep trying to decide if they sound any better than the CD.
Jim Lesurf
<spam@dev_null.com.invalid> wrote:
> On Mon, 06 Jun 2011 18:16:50 +0100, Jim Lesurf <no...@audiomisc.co.uk>
> wrote:
> >In article <481qu618t30ftuk2k...@4ax.com>, Albert Ross
> ><spam@dev_null.com.invalid> wrote:
> >> On Sat, 21 May 2011 13:52:46 +0100, Jim Lesurf
> >> <no...@audiomisc.co.uk> wrote:
> >
> >> >Great shame that Hickox died before he could do the entire VW set.
> >> >The Chandos discs also have a number of 'first recordings' of other
> >> >VW items.
> >
> >> When I find some more circular tuits I might look that out.
> >
> >FWIW I've just got a copy of the London Symphony as 96k/24bit flac
> >files. I keep trying to decide if they sound any better than the CD.
> >But when I try I just end up becoming lost in the music. Ho hum,
> >investigating audio can be such a strain at times... 8-]
> By a strange coincidence it was used as part of the backing track to
> something I watched the other day, Coast or similar. Now I have one of
> his tunes stuck in my head.
Bit showing the old docks? I now can't recall if it was the VW or the
London Overture by Ireland! :-)
I was fascinated by two aspects of the programme.
One was the image of the ship at the end of the road, seen between the
houses. I remember seeing that sort of thing back in the days when I could
also hear the ship's hooters from The Pool at New Year.
The other was Doctor Feelgood being featured in the section on Canvey.
Remarkable to see that juxtaposed with Einstein on the beach (Belgium)...
(Musicalogal joke. ;-> )
Jim Lesurf
<ja...@evij.com.invalid> wrote:
> There *shouldn't* be any difference, should there? FLAC is Free
> Lossless Audio Codec after all?
I'll break into the top posting here: :-)
The snag is the usual one. In theory, theory and practice agree, but in
practice... ?
The requirement isn't just to output the correct sequence of values. It is
to ensure that all the values are output and get to the DAC with uniform
timing. No pauses, breaks, duplicates, rate variations, etc.
The difficulty is that the 'computer' has to do more counting on its
fingers to 'decode' a format like flac than is needed to simply shove a
stream of LPCM values. The whole process has to be timed and bufferred and
controlled adequately to let the DAC conversions tick along without any
problems.
So I'll need to do an 'IQ test'[1] comparing playing flac and lpcm before I
decide on this. Probably using more than one player on more than one setup.
[1] See an example on
http://www.audiomisc.co.uk/Linux/Sound3/TimeForChange.html
if interested in the kinds of differences I'm referring to. Note that just
used lpcm data, which minimises the amount of work the computer process has
to do (particularly as I used the ALSA 'play' command to avoid having a GUI
app doing the job, or go though other layers).
Plus, the people who write real-world software often make assumptions that
turn out not always to be true. <ahem> You have probably encountered this
yourself. :-)
So for example, I have one audio player program that refuses to even *try*
to play files with sample rates above 48k. The author assumed that any such
value must be an error. Another says it is playing 24bit, but the output is
16bit values in 24bit slots.
FWIW I've also encountered different mp3 decoders that give markedly
different output from the same source mp3 file. Yet the idea of methods
like mp3 is that the 'clever bit' is the encoding and packing the chosen
info into the defined format. The decoder should then just apply a defined
algorithm to recover what the data in the mp3 file/stream defines. The
'loss' is supposed to be in the coding end. The decoder isn't supposed to
know or care how the encoding was done. Just how to interpret the supplied
file/stream.
So IME assuming "all is well" is risky. I've already measured and reported
on a number of unobvious 'problems' in the past. So I don't take for
granted that a system 'works perfectly' because 'theory says it should'.
Need to compare, measure, analyse, etc.
Above said, I compared a RO port of the flac tool with SoX on Linux this
morning. They gave identical results for the 96k/24bit lpcm payloads I
tried. But gave oddly different wave headers when used to generate the
audio in a wave file. Alas, creating a file doesn't have to be done in real
time with perfectly uniform timing. That bit has its own requirements.
The listening is the easy part. :-)
Slainte,
Jim
> On Mon, 06 Jun 2011 18:16:50 +0100, Jim Lesurf <no...@audiomisc.co.uk>
> wrote:
> >
> > FWIW I've just got a copy of the London Symphony as 96k/24bit flac
> > files. I keep trying to decide if they sound any better than the CD.
--
bolta...@boltar.world
Jim Lesurf <no...@audiomisc.co.uk> wrote:
>The difficulty is that the 'computer' has to do more counting on its
>fingers to 'decode' a format like flac than is needed to simply shove a
>stream of LPCM values. The whole process has to be timed and bufferred and
>controlled adequately to let the DAC conversions tick along without any
>problems.
With the increase of storage density and broadband speeds I suspect the
days of lossy audio compression systems are numbered except in certain limited
applications such as over the air broadcasting where bandwidth can't be
increased.
B2003
Andy Champ
> There *shouldn't* be any difference, should there? FLAC is Free
> Lossless Audio Codec after all?
>
> On Mon, 06 Jun 2011 18:16:50 +0100, Jim Lesurf<no...@audiomisc.co.uk>
> wrote:
>>
>> FWIW I've just got a copy of the London Symphony as 96k/24bit flac files.
>> I keep trying to decide if they sound any better than the CD.
errmmm... CD is 44.1KHz. 96KHz FLAC should sound a _lot_ better.
_If_ your ears and speakers are good enough that is.
Andy
Phil Cook
Only *if* it really is 96kHz/24bit. More than likely it is the same
thing in a bigger box.
--
Phil Cook
Java Jive
On Tue, 07 Jun 2011 19:42:32 +0100, Andy Champ <no....@nospam.invalid>
wrote:
>
> 96k/24bit flac
Paul Ratcliffe
>>> FWIW I've just got a copy of the London Symphony as 96k/24bit flac files.
>>> I keep trying to decide if they sound any better than the CD.
>
> errmmm... CD is 44.1KHz. 96KHz FLAC should sound a _lot_ better.
No it shouldn't. You are well down the law of diminishing returns curve.
Jim Lesurf
One of the problems here is to have versions which only differ in a
carefully controlled way. i.e. may have a different sample rate or sample
depth, but *not* be different in other ways. It is routine in the music biz
to 'master' <sic> different versions on different assumptions about the
'market'. So a CD may be level compressed and clipped and eq'd whilst the
same source material on another type of delivery is not.
The industry has a widespread presumption that for CD 'louder is better',
and that clipping doesn't matter. But that those paying more for 96k/24
expect wider and more natural dynamics, crest factors, etc.
For me, one of the points of having some 96k/24 material is I can know for
myself how I then generate 'CD' or other versions. e.g. choice of
downsampling filter shape, dithering, noise-shaping, etc. This means I can
know what changes have *not* been applied.
However moving back to my earlier point, one of my interests is to see if
flac decoding-and-playing rather than lpcm playing has any effect on
timing, etc, so far as the DAC is concerned. In theory it shouldn't. But my
experience is that practice often departs from the assumptions people jump
to. :-)
Jim Lesurf
Champ
Yes, I often hear/read such general assertions. They are bound to return
TRUE once you have noticed the careful inclusion of the "should" qualifier
and the self-referential defenition of the "if". ;-> That may not tell
you much about the reality, though...
In practice the assessable evidence from carefully run listening tests is
often less clear.
Four points to ponder.
1) That many recordings may simply not have anything audible above about
20kHz in the first place due to the nature of what was recorded and/or the
choice of microphones and/or the mixing equipment used.
2) That a controlled test published in JAES showed no ability to
distinguish under domestic listening conditions.
3) That some of the 'high rez' releases have been found *by measurement* to
contain the same info as the CD version - simply 'upsampled' to a higher
rate. Yet no-one seemed to notice this before measurement.
4) That some of the 'dual version' releases where people have said "I can
hear an obvious difference" have measurably been processed in different
ways. e.g. SACD/CD dual-layer discs where the CD version is level
compressed and eq'd in ways clearly different to the version on the SACD
layer.
So it is easy for people to presume this is a forgone conclusion, but the
reality may often not conform to the belief.
Andy Champ
> In article<slrniutgds...@news.pr.network>, Paul Ratcliffe
> <ab...@orac12.clara34.co56.uk78> wrote:
>
>> No it shouldn't. You are well down the law of diminishing returns curve.
With my ears you're off the end :(
<snip>
> For me, one of the points of having some 96k/24 material is I can know for
> myself how I then generate 'CD' or other versions. e.g. choice of
> downsampling filter shape, dithering, noise-shaping, etc. This means I can
> know what changes have *not* been applied.
>
I've pulled some of my old vinyl in to the computer. I record at
88.2khz, 24 bit, then post-process and resample down to 44.1khz, 16 bit.
Just to make sure the no-more-than-adequate CD standard doesn't get in
the way.
(Don't misunderstand me; it is adequate, just not by much.)
> However moving back to my earlier point, one of my interests is to see if
> flac decoding-and-playing rather than lpcm playing has any effect on
> timing, etc, so far as the DAC is concerned. In theory it shouldn't. But my
> experience is that practice often departs from the assumptions people jump
> to. :-)
>
You should be able to convert LPCM to FLAC and back again and compare
the results - and see it bit-perfect.
Andy
Jim Lesurf
Champ
<no....@nospam.invalid> wrote:
> On 08/06/2011 09:22, Jim Lesurf wrote:
>
> I've pulled some of my old vinyl in to the computer. I record at
> 88.2khz, 24 bit, then post-process and resample down to 44.1khz, 16 bit.
> Just to make sure the no-more-than-adequate CD standard doesn't get in
> the way.
> (Don't misunderstand me; it is adequate, just not by much.)
The real problems with 'CD' are down to the people making the CDs. Given
how often they do things like clip and excessively level-compress, how much
confidence can we have that they have any clue about how to do even sample
rate conversions and associated filtering well?
If you ask yourself why so many EMI CDs issued in the past have a 'grainy'
or 'sandpaper' effect on strings the answer may be more in the letters
"EMI" than "CD". :-)
> > However moving back to my earlier point, one of my interests is to see
> > if flac decoding-and-playing rather than lpcm playing has any effect
> > on timing, etc, so far as the DAC is concerned. In theory it
> > shouldn't. But my experience is that practice often departs from the
> > assumptions people jump to. :-)
> >
> You should be able to convert LPCM to FLAC and back again and compare
> the results - and see it bit-perfect.
Actually, I've already found differences can arise in such conversions if
the person doing them doesn't know what they are doing. However that is
largely a 'nut behind the wheel' problem.
My concern is in the "decoding-and-playing", though.
i.e what happens when you *play* a flac file. That requires the conversion
to be done in a way that does not alter the timing of the *output stream to
the DAC*. Not enough for all the values to be correct. They also have to
arrive with suitable timing.
If you look at the measurements I've already done you can see we have no
guarantee that all computer systems+software+DACs will do this well.
The problem here is that many in the 'computer' world have adopted the
assumption that it "must work" and confused "I can hear something" with "it
works perfectly". As often in domestic audio, when you actually bend down
and turn over the rock, you may just find something 'unexpected'... :-)
Andy Champ
> Not enough for all the values to be correct. They also have to
> arrive with suitable timing.
Surely that part of the system - the DAC - will be running whether
you're playing FLAC, PCM or even MP3?
Andy
Jim Lesurf
Champ
You'd have to define more carefully what you mean before I could comment.
Have you read the 'IQ test' pages I gave a URL for a while ago? That shows
the kinds of timing problems that can and do arise in practice.
Measurements, not theory.
Andy Champ
> In article<CLWdnfqMYK5tgWzQ...@eclipse.net.uk>, Andy
> Champ
> <no....@nospam.invalid> wrote:
>> Surely that part of the system - the DAC - will be running whether
>> you're playing FLAC, PCM or even MP3?
>
> You'd have to define more carefully what you mean before I could comment.
>
> Have you read the 'IQ test' pages I gave a URL for a while ago? That shows
> the kinds of timing problems that can and do arise in practice.
> Measurements, not theory.
>
The FLAC / MP3 -> PCM conversion will be running entirely in the digital
domain, and the odd bit of jitter in the bit timing really doesn't
matter - _so_ _long_ as they arrive in the DAC and get clocked through
at the correct time. Which isn't going to be closely related to the
decode time, I'm sure there will be buffers of several kB at least in
that pipe.
Or to put it another way - my FLAC/MP3 decoder can run way above real
speed. It'll decode perhaps 100mS worth of data in under 10mS. Then
stick that data into a buffer and queue it for the DAC. It doesn't
matter when it does this as long as the DAC never actually runs out.
Several seconds worth of buffer are no problem, and can be a real advantage.
On the other hand the rate at which the DAC turn the PCM into a voltage
must be very tightly timed. Even a 1% error will result in severe
distortion, and that's an error of about 2uS (if my mental arithmetic is
working tonight).
Don't suppose you have that URL handy?
Andy
Jim Lesurf
Champ
<no....@nospam.invalid> wrote:
> On 10/06/2011 09:17, Jim Lesurf wrote:
> > In article<CLWdnfqMYK5tgWzQ...@eclipse.net.uk>, Andy
> > Champ <no....@nospam.invalid> wrote:
> >> Surely that part of the system - the DAC - will be running whether
> >> you're playing FLAC, PCM or even MP3?
> >
> > You'd have to define more carefully what you mean before I could
> > comment.
> >
> > Have you read the 'IQ test' pages I gave a URL for a while ago? That
> > shows the kinds of timing problems that can and do arise in practice.
> > Measurements, not theory.
> >
> The FLAC / MP3 -> PCM conversion will be running entirely in the digital
> domain, and the odd bit of jitter in the bit timing really doesn't
> matter - _so_ _long_ as they arrive in the DAC and get clocked through
> at the correct time.
Agreed. Your proviso is critical. That is my point.
> Which isn't going to be closely related to the
> decode time, I'm sure there will be buffers of several kB at least in
> that pipe.
Yes, such systems normally have buffers. But timing variations still occur.
So again, what happens in practice may not be as assumed if you think
having a buffer will cure any problems. Buffering can sometimes alter the
details of the resulting timing problems rather than eliminate them.
> Or to put it another way - my FLAC/MP3 decoder can run way above real
> speed. It'll decode perhaps 100mS worth of data in under 10mS. Then
> stick that data into a buffer and queue it for the DAC. It doesn't
> matter when it does this as long as the DAC never actually runs out.
> Several seconds worth of buffer are no problem, and can be a real
> advantage.
Yet timing variations occur in reality even with LPCM source material.
In part the problem is that the 'computer' end is multitasking with
hardware interrupts, etc. Hence despite any individual process being
'quick' and 'buffered' can still end up with timing problems for a task
like sound where timing is critical. Now add in a decoding process whose
intensity varies unpredictably from second to second in accord with the
details of the compressed input at that section of the data.
And in part - unless you use a modern asynch DAC or have all the clocks
locked together - there will also be problems with clock differences though
the system.
> On the other hand the rate at which the DAC turn the PCM into a voltage
> must be very tightly timed. Even a 1% error will result in severe
> distortion, and that's an error of about 2uS (if my mental arithmetic is
> working tonight).
Note also that modern recordings may well be 96k/24. Some labels also now
offer 192k/24 as 'master' files. And to match this DACs and transfer
methods have to cope with higher data rates and clock accuracy.
> Don't suppose you have that URL handy?
Have two. :-) One for measurements on a practical example, the other for
a simplified explanation of the method
http://www.audiomisc.co.uk/Linux/Sound3/TimeForChange.html
http://www.audiomisc.co.uk/Linux/Sound3/TheIQTest.html
FWIW I've now almost finished the public versions of the programs I wrote
for generating and analysing IQ waveforms. Hope to have them on my website
in the next couple of days. I'll mention here when they are available in
case anyone wants to try it themself, or write their own (better) code.
Albert Ross
wrote:
>In article <0bsru6l22n9hoac7a...@4ax.com>, Albert Ross
><spam@dev_null.com.invalid> wrote:
>> By a strange coincidence it was used as part of the backing track to
>> something I watched the other day, Coast or similar. Now I have one of
>> his tunes stuck in my head.
>
>Bit showing the old docks? I now can't recall if it was the VW or the
>London Overture by Ireland! :-)
Ah now I'm confused, but the VW was on *something" recently, for some
unaccountable reason there have been a few recent programmes of above
average quality, and not all on BBC4. Makes me feel quite nostalgic.
Good informational content allied to proper production without the
trendy crap.
>I was fascinated by two aspects of the programme.
>
>One was the image of the ship at the end of the road, seen between the
>houses. I remember seeing that sort of thing back in the days when I could
>also hear the ship's hooters from The Pool at New Year.
back in the day I worked in some of those old warehouses in the docks,
they had hydraulic lifts and cranes working off a high pressure water
system. The lifts had a rope going through holes in roof and floor
which you pulled to start and stop (with a bounce).
Most of the other docks I know (Bristol, Southampton) have likewise
become Executive Housing.
>The other was Doctor Feelgood being featured in the section on Canvey.
>Remarkable to see that juxtaposed with Einstein on the beach (Belgium)...
>(Musicalogal joke. ;-> )
Yes they obviously let someone with a brain escape for a while.
Jim Lesurf
> FWIW I've now almost finished the public versions of the programs I
> wrote for generating and analysing IQ waveforms. Hope to have them on my
> website in the next couple of days. I'll mention here when they are
> available in case anyone wants to try it themself, or write their own
> (better) code.
Finished that quicker than I thought! The programs are now available from
http://www.audiomisc.co.uk/software/index.html
(Scroll down to near the bottom of the page.)
The complied apps are for RO and RO Linux. However the source code is
included and the Linux version is in GCC 'C' so should be easy to adapt to
other platforms if someone wishes. The Linux versions can also easily be
run on machines that don't use ROX Filer.
Terry Casey
11:01:43 +0100
Albert Ross <spam@dev_null.com.invalid> wrote:
>
> back in the day I worked in some of those old warehouses in the docks,
> they had hydraulic lifts and cranes working off a high pressure water
> system. The lifts had a rope going through holes in roof and floor
> which you pulled to start and stop (with a bounce).
>
I remember those lifts in many buildings I used to visit in the City of London
- and the glove the operator wore to grip the rope!
In the city, the lifts were driven by Artesian wells, rather than the hydraulic
accumulators used in the docks to provide power to the cranes.
--
Terry
Andy Champ
find two things spring to mind:
On 11/06/2011 09:46, Jim Lesurf wrote:
> And in part - unless you use a modern asynch DAC or have all the clocks
> locked together - there will also be problems with clock differences though
> the system.
What's the clock source for the DACMagic? Not USB I hope? I also find
myself wondering if there is some kind of beat between the recorder and
the DAC. Could the periodic errors be due to dropped samples?
>
>> > On the other hand the rate at which the DAC turn the PCM into a voltage
>> > must be very tightly timed. Even a 1% error will result in severe
>> > distortion, and that's an error of about 2uS (if my mental arithmetic is
>> > working tonight).
> Note also that modern recordings may well be 96k/24. Some labels also now
> offer 192k/24 as 'master' files. And to match this DACs and transfer
> methods have to cope with higher data rates and clock accuracy.
Not on (ordinary) CD it won't be :)
Andy
Jim Lesurf
Champ
<no....@nospam.invalid> wrote:
> Jim, I've read through the two items (and will again - it's late!) but I
> find two things spring to mind:
> On 11/06/2011 09:46, Jim Lesurf wrote:
> > And in part - unless you use a modern asynch DAC or have all the
> > clocks locked together - there will also be problems with clock
> > differences though the system.
> What's the clock source for the DACMagic? Not USB I hope?
Yes and no, m'lud. :-)
AIUI The DACMagic (without using the Halide Bridge) uses the - now old
fashioned - method of trying to follow the rate at which data is sent, but
then 'upsampling' these values into the DACMagic's internally clocked 192k
series for conversion. The DACMagic has a lock loop that tries to ensure that
in the long term the clock follows the USB rate. Theloop seems to have a
locked bandwidth of the order of a Hz or two. So *short term* variations in
timing should be suppressed by the loop and buffering. The problem is that
sometimes a change persists...
The 'rate jumps' shown in the graphs are the result. Note they last for
many seconds. At present can only speculate wrt the internal details beyond
that as various causes are possible. cf below.
> I also find myself wondering if there is some kind of beat between the
> recorder and the DAC. Could the periodic errors be due to dropped
> samples?
The recorder was making an *analogue* recording of the output of the DAC.
So has no idea what internal sample rates the DAC (or USB) uses. It just
sees how well the result emerges as an analogue IQ waveform. If there are
any dropped or duplicated samples in the DAC or USB transfer that then
shows up as phase rate (i.e. frequency) modulation.
The analysis also works on the basis of determining the change of phase
during each successive short period (as defined by the clock in the
recorder). So a dropped or repeated or pause sample in the DAC/PC would
produce a brief 'spike' in the measured result, not a steady change.
Note that the output I obtain is the effective *frequency* of the recorded
analogue. So a dropped sample would just show up as a very brief change in
frequency before returning to the earlier frequency. Whereas you can see
that for the DACMagic there are quite long-term changes in frequency. So
represent a change in the effective sample clock rate as many many samples
are covered during each stretch of a given frequency.
The DACMagic does try, I think, to lock to the incoming frequency. So I
suspect the problem here is that the *computer* keeps changing its mind
about what rate to use, maybe based on how full an internal buffer is as
determined by its own clocks - which may also vary.
Note that the problem vanishes when I use the Halide Bridge. This has its
own internal clock and takes over responsibility for USB transfer. It then
tells the computer when more data is required. The result is that both
computer and DACMagic have to follow what the Halide Bridge decides. The
DACMagic then seems to lock to the spdif output of the bridge and be quite
happy with that. The computer just feeds out the data at the rate, and in
blocks, as demanded by the Bridge.
Only fair to also point out the effects are small. parts per million or
billion are appropriate units. Somewhat lower 'wow and flutter' than even
the best LP turntables or analogue tape decks! :-)
Also worth pointing out that the results may vary if you change the source
'PC' or alter the processes or hardware it is juggling.
I can't be sure. But to me it looks like a clear sign that the internal
clocking of a domestic computer may simply not be uniform and reliable. Not
surprising given all the multitasking, hardware interrupts, clock-rate
changing, interference, etc, inside a typical 'PC'. But this is why I am
also wary of assuming that, for example, 'flac will work exactly as lpcm'.
Maybe, maybe not. :-)
My prediction is to expect async USB DACs and transfer methods to replace
the older transfer methods. But no doubt that will throw up its own
problems... :-)
Paul Ratcliffe
> I can't be sure. But to me it looks like a clear sign that the internal
> clocking of a domestic computer may simply not be uniform and reliable. Not
> surprising given all the multitasking, hardware interrupts, clock-rate
> changing, interference, etc, inside a typical 'PC'. But this is why I am
> also wary of assuming that, for example, 'flac will work exactly as lpcm'.
Be all of that as it may, but it is just clouding the issue. FLAC is
turned into LPCM in software and the buffers are delivered to the OS,
sound driver and sound hardware in exactly the same way as a WAV which
is already LPCM.
I really don't understand what you are on about... but never mind.
Andy Champ
> I can't be sure. But to me it looks like a clear sign that the internal
> clocking of a domestic computer may simply not be uniform and reliable. Not
> surprising given all the multitasking, hardware interrupts, clock-rate
> changing, interference, etc, inside a typical 'PC'. But this is why I am
> also wary of assuming that, for example, 'flac will work exactly as lpcm'.
> Maybe, maybe not.:-)
The clock on the PC ought to be a quartz-quality clock. And it'll
probably be running at several GHz, so what you see should be pretty
steady. BUT it'll also be full of digital noise. And it's not unknown
for digital systems to do something like dividing down, and deciding
that a 24MHz clock is close enough to 25 for their purposes... or even,
as has happened in the past, that 8 1/3 is close enough to 8.
The problem you're seeing just should not be happening (TM). The
problem I expected you to show me is due to clock jitter moving the
samples slightly sideways - which then gives effective amplitude errors.
However, as you've shown, it is happening. I suspect the halide bridge
just has a better clock generator.
>
> My prediction is to expect async USB DACs and transfer methods to replace
> the older transfer methods. But no doubt that will throw up its own
> problems...:-)
I have a fairly decent USB sound device at work, bought some years ago
when I was working on an audio project. I've had to stop using it now,
Vista has it as a legacy device, and insists on running it at 48KHz.
Which means it resamples all my 44.1KHz stuff on the fly. It's bad
enough to be annoying, even when all I'm doing is using the headphones
to cut out office noise.
Andy
Java Jive
basic problems which I'd like to put forward for discussion ...
Long ago I digitised my Audio-Cassette and my VHS collections, but I
still have some outdated audio recordings in the form of:
ACs: 'The Hitchhikers Guide To The Galaxy' first 2 radio series
About 60 MiniDisks.
About 125 LPs
About 5 x 45 rpms
At one time there were over 400 LPs and over 100 45s. Most of these
have been either replaced by CDs, already digitised, a few wanted
tracks found elsewhere, or just become superfluous to my artistic well
being!
Then of course, there is the attendant hardware:
Denon HiFi (seperates) downstairs in my lounge
Sanyo Midi Tower (seperates) upstairs in my bedroom/office.
MD deck in the downstairs hifi (but I think it may be faulty)
MD deck in the upstairs hifi
Dual turntable, aged, some rumble, no preamp.
Phono inputs on downstairs hifi.
NAD Phono Preamp.
Project turntable with inbuilt pre-amp.
Vinyl washing machine
2 x Desktop P4s W2k (still), each with ...
SB Live with Digital IO dongle (SPDIF Coax and Optical In & Out)
Dell Latitude 610 laptop + docking station
USB Terratec Aureon MKII (SPDIF Optical In & Out)
In order to reduce the chances of picking up hum and stray e-m, I
always prefer to use optical digital connections whereever possible. I
can do this upstairs by connecting optically between the MD output and
an optical input on the PC - for AC or any other analog source, I
put a blank MD in the player, pressing MD Record but also Pause, which
allows the analogue source to be routed out via the MD's optical to be
recorded by the PC; I assume, perhaps incorrectly, that this enables
me to convert the analogue to digital without incurring the
compression lossiness of recording it onto MD and then playing it
back, but in truth this is an informed guess, I don't actually know
this for certain, and don't recall actually testing this belief, which
I suppose I ought to have. Bit late now though.
Unfortunately, there is no such solution for the vinyls.
I record everything to LPCM wave files.
For some time, I've been thinking that it would be good to:
:-) Guard against irreplacable but obsolescent equipment going
down
:-) Digitise these remaining vinyls before further deterioration
:-) Be able to play the vinyls without the associated hassle
:-) Lose this significant pile of junk, but keep the recordings.
Since Christmas, I've digitised H2G2 and all the MDs, and have nearly
finished the first pass through the vinyls. However, and hence this
post, I've had some totally unexpected problems ...
H2G2 (ACs) ...
I first recorded this via the Terratec, using the optical arrangement
described above. However, on playback, the recordings sounded
slightly but definitely, noticeably speeded up, vaguely suggestive of
Mini Mouse on 'speed' and helium. When I simply moved the optical
input to the SB Live, everything was fine.
Why?!
MDs ...
Thus forewarned, I recorded these via the SB Live without mishap.
Interestingly, I was able to compare some vinyls previously recorded
onto MDs, for greater convenience of playing, with the originals.
There is a noticeable difference between the two. Even my aging ears
can hear the drop in sound quality from MD lossy compression. It's
not terrible, I still would far rather have been able to make all my
old folk club recordings on MD than AC, but it's there.
Vinyls ...
Here the big problems are
Finding a stable surface for the deck(s)
Dust and gunge on the vinyls
Mains Hum
I decided that it would be best to put the decks on the living-room
floor, as it's concrete. This would make it less likely to be
affected as I walk about, open drawers in my filing cabinet, etc.
Also, that's where the best HiFi is, and, there being so much
electrical equipment upstairs, the e-m noise level must be lower
downstairs. However, this left me with a problem - the PCs are
upstairs, and I can't simply move one downstairs, because I have only
one screen, keyboard, and mouse. Accordingly, I decided to buy the
laptop and docking station second-hand on eBay.
I've described how I deal with the dust/gunge problems here ...
http://www.macfh.co.uk/JavaJive/AudioVisualTV/Vinyls/VinylRestoration.html
... my plan being to record each vinyl, or at least each wanted track
thereof, twice:
1st Pass unwashed, via the Dual
2nd Pass washed, via the Project
... and ultimately keep only the best.
That page also mentions mains hum in passing, but a fuller discussion
follows ...
I found that however I connected up the Dual, I got a big hum. I
tried all the following combinations severally and together, all
without making much of a difference:
Earthing the metal of the deck via the mains lead, and not
Earthing via the earth terminal of the amp, and not
Using the seperate preamp, and the phono inputs on the amp
Finally, in desperation, I took the turntable out of its box, and
discovered that the grounds of the cartridge connections were
connected to the metal of the deck. I cut these connections and
rewired them so that the metal of the deck is earthed via the mains
lead, and seperately the grounds of the cartridge, like the signals,
come straight out of the back to the outputs. This is a HUGE
improvement: there is no build-up of static, and it's effectively
hum-free until I start the motor. With the motor running, there is
some residual hum, which I'd like to get rid of, but at least the
result is listenable.
When I tried the new arrangement with the NAD pre-amp, there was just
a little more hum, so I've been using the HiFi phono inputs.
When I tried it additionally connecting the metalwork to the earth
connection on the amp either there was either no difference or it was
worse, I can't now remember which, only that no benefit was obtained.
My conclusion from all of this is that a vinyl record-deck should be
wired as follows, but I'd be interested to see if others agree ...
If the deck is driven from mains voltages, then its metalwork should
be earthed via the earth in a three-core mains lead connected to the
earth in a 3-pin plug. In this case, the metalwork of the arm should
be insulated from the deck and a seperate earth point provided to
connect the arm to the amp earth.
Whether or not the deck is driven by mains voltages, the cartridge
should be connectioned to the outputs via screened cable over the
entire distance. The problem is that screened cables are stiffer than
the tiny wiring commonly used in an arm, and their stiffness might
affect the tracking if the job's not carefully done.
I did once wire up an old Garrard deck like this, as an experiment. My
recollection is that it was hum-free. I got the shielded cabling from
RS or Maplin's - can't remember which, but I do remember that I used
the same cabling in my TR when I swapped out some worn heads. It was
two cores, about the same thickness as you'd find in a pick-up arm,
with a common braiding outside, then a thin outer insulation. It was
quite flexible, I do not recall ever noticing tracking problems with
it.
Thoughts on this reasoning?
Next came recording via the laptop.
Having had more success with the SB Live than the Terratec, I thought
I'd better stick with that. Although the docking station was really
bought so that I could connect the laptop instantly to my KVM and
office network upstairs, it has a single PCI slot, so I bad farewell
to convenience upstairs, I put one of the SBs in it, connected up to
the HiFi, and started recording. However, when I checked the
playback, there was a problem with distortion - some albums in
particular were ruined by it.
WTF? Why? And why some albums consistently so much more than others?
The card has been working perfectly well for years in a desktop PC
running Windows v5.0, why not in the docking station of a laptop
running v5.1? Thinking, however that it might be something to do with
driver versions under XP rather than 2000, I checked for the latest
driver downloads, but those I've been using all along for 2000 are
also the correct and latest for XP.
So I tried the Terratec. Success at last!
"Great!", I thought, "I don't need the docking station downstairs
anymore!". So I got the normal PSU for the laptop, and removed the
docking station. Big, big hum! Replaced the docking station. Back
to normal. Removed it again. Big, big hum! Replaced it again. Back
to normal.
I conclude that the DS has a decent PSU, while that supplied for the
laptop is crap. Or is it more subtle than that? Is there a mains
loop being formed that I can do something about? I really would like
to have the docking station back up here. Any thoughts?
On Sat, 11 Jun 2011 09:46:37 +0100, Jim Lesurf <no...@audiomisc.co.uk>
wrote:
>
> > Don't suppose you have that URL handy?
>
> Have two. :-) One for measurements on a practical example, the other for
> a simplified explanation of the method
>
> http://www.audiomisc.co.uk/Linux/Sound3/TimeForChange.html
>
> http://www.audiomisc.co.uk/Linux/Sound3/TheIQTest.html
>
> FWIW I've now almost finished the public versions of the programs I wrote
> for generating and analysing IQ waveforms. Hope to have them on my website
> in the next couple of days. I'll mention here when they are available in
> case anyone wants to try it themself, or write their own (better) code.
Jim Lesurf
<ab...@orac12.clara34.co56.uk78> wrote:
> On Sun, 12 Jun 2011 10:08:28 +0100, Jim Lesurf <no...@audiomisc.co.uk>
> wrote:
> > I can't be sure. But to me it looks like a clear sign that the
> > internal clocking of a domestic computer may simply not be uniform and
> > reliable. Not surprising given all the multitasking, hardware
> > interrupts, clock-rate changing, interference, etc, inside a typical
> > 'PC'. But this is why I am also wary of assuming that, for example,
> > 'flac will work exactly as lpcm'.
> Be all of that as it may, but it is just clouding the issue. FLAC is
> turned into LPCM in software
Not quite. It is done by *hardware* that is executing software - and with a
lot of other processes also being run 'at the same time' <sic> on that
hardware. Plus hardware interrupts, hardware items having their own
clocking requirements, etc.
> and the buffers are delivered to the OS, sound driver and sound hardware
> in exactly the same way as a WAV which is already LPCM.
Are they? But what if the later transfer processes have their timing
affected by the CPU having the added requirement to decode the flac to
lpcm? To name just one obvious possible change that *might* have an impact.
> I really don't understand what you are on about...
I appreciate that. At its simplest it is what I've said. That practical
reality does have a habit of *not* working as neatly as simple theories and
assumptions may lead people to expect. A 'PC'+USB+DAC chain is a fairly
complex one. If the 'PC' is a general purpose device running a full OS and
a set of other processes, expect complications. Some of which may duly
affect the behaviour.
At present I have no idea at all if flac decoding will make any difference
at the 'output end of the pipe'. May be no difference. But my bias is
towards using experiment and observation rather than theory to tell me.
And my experience of finding out things by measurement/observation makes me
think it quite possible that there may well be a difference.
The maxim of the Royal Society (removing the fancy latin) is "Take no-one's
word for it". Good advice in my experience.
FWIW A number of times during my professional career I was told words to
the effect of "Why investigate that, there isn't anything there to be
found!"... and then discovered something interesting that helped move some
part of science or engineering along a bit.
My first work as a postgrad was building new detectors for mm-wave
astronomy (then called 'far infrared'). At the time no-one had seen
anything at these wavelengths other than the Sun or Moon. Many physicists
said, "There's nothing to observe at these wavelengths. Why bother?"
Then we started finding all the dust and gas clouds, and other sources that
no-one expected to show up. And to start finding out things about them the
visible and other bands hadn't told us. For a few mm-wave groups we then
had a 'boom' decade or two where some of the people I worked with couldn't
write the papers reporting their findings fast enough. At one time we used
to have 'advance block bookings' on some of the telescopes in Hawai'i.
Every time we used the telescope we found new sources no-one had detected
or expected. :-)
Much the same happened in later decades when I and others looked at other
fields - e.g. observations of volcanoes, looking thought the dust clouds.
Lots of established people said "No point in this" or even "This can't be
done" at first. Then all rushed to join in once the results started to
appear. ;->
So finding out for myself is a habit I've formed that often seems to turn
up something interesting or unexpected. Fun, too. :-)
Jim Lesurf
Champ
<no....@nospam.invalid> wrote:
> On 12/06/2011 10:08, Jim Lesurf wrote:
> > I can't be sure. But to me it looks like a clear sign that the
> > internal clocking of a domestic computer may simply not be uniform and
> > reliable. Not surprising given all the multitasking, hardware
> > interrupts, clock-rate changing, interference, etc, inside a typical
> > 'PC'. But this is why I am also wary of assuming that, for example,
> > 'flac will work exactly as lpcm'. Maybe, maybe not.:-)
> The clock on the PC ought to be a quartz-quality clock.
However the 'clock' seen by different parts of the system may well be
something else, and clock rates will vary. [As your comment below
identifies.] And various processes will be distracted by hardware
interrupts, pre-emptive multi-tasking, etc.
> And it'll probably be running at several GHz, so what you see should be
> pretty steady. BUT it'll also be full of digital noise. And it's not
> unknown for digital systems to do something like dividing down, and
> deciding that a 24MHz clock is close enough to 25 for their purposes...
> or even, as has happened in the past, that 8 1/3 is close enough to 8.
And some part of the system may keep changing its mind about what divided
down ratio of some master clock to use.
I'm not an expert on modern CPUs, but IIUC they also do things like switch
up and down a 'comb' of clock rates depending on things like loading and
temperature. Trying to synch all that with other devices like USB
controllers or soundcards can clearly become complicated once you add in
all the things that are happening.
> The problem you're seeing just should not be happening (TM). The
> problem I expected you to show me is due to clock jitter moving the
> samples slightly sideways - which then gives effective amplitude errors.
Hence my earlier comments about practice not always meekly following the
theory of what 'should happen'. :-)
Note that the changes in rate last for many seconds - i.e. a slight
alteration in clock rate spread uniformly over tens or hundreds of
thousands of samples. That looks more to me like the clock rate being used
to output the USB is being switched up and down every now and then by some
monitoring process deciding "I'm running a bit fast" or "I'm running a bit
slow" and trying to get the right rate - on long term average. Typical of
two processes running with different clock sources with a connecting
buffer, with one altering the rate at which it fills or empties the buffer
judged on an 'too empty' and 'too full' set of levels.
FWIW I've seen similar behaviour in other comms systems. It ensures all the
data is transferred with no lost or misplaced values. Fine for systems
where the output doesn't immediately have to be delivered with uniform
real-time behaviour. But leads to problems like this when audio is required
to be heard rather than just dumped onto store.
Note also the regularity of the durations and levels of the rates giving a
mark-space rectangle wave effect to the *rate*.
> However, as you've shown, it is happening. I suspect the halide bridge
> just has a better clock generator.
TBH I doubt that. The LF and HF noise doesn't look that much different to
me.
> >
> > My prediction is to expect async USB DACs and transfer methods to
> > replace the older transfer methods. But no doubt that will throw up
> > its own problems...:-)
> I have a fairly decent USB sound device at work, bought some years ago
> when I was working on an audio project. I've had to stop using it now,
> Vista has it as a legacy device, and insists on running it at 48KHz.
> Which means it resamples all my 44.1KHz stuff on the fly. It's bad
> enough to be annoying, even when all I'm doing is using the headphones
> to cut out office noise.
I had a similar problem with the internal cards in my machines at first.
But I solved the problem by using ALSA and a suitable .asoundrc file.
One advantage of the newer USB devices like the DACMagic and Halide bridge
is that they don't require a specific driver to be installed. So you may
well find they work fine with Vista and other versions of Windows. Can't
comment directly, though, as I don't use Windows at all these days.
FWIW I'd personally recommend the DABMagic with or without the Halide
Bridge. Sounds very good either way. And quite nice that it shows the
sample rate it is getting so you can immediately see if that is switching
with the source material as it should.
But at present I suspect we will find that many other good similar devices
will appear. Examples like the arcam rDAC, etc, have already become
available and seem to work nicely. (Although I've not produced any results
with them as yet. So only saying that on the basis of general listening.)
Norman Wells
> As always, Jim's work is very interesting, but I have some much more
> basic problems which I'd like to put forward for discussion ...
>
> Long ago I digitised my Audio-Cassette and my VHS collections, but I
> still have some outdated audio recordings in the form of:
<six years snipped>
Please .... tell us you can't get them already digitised from Amazon or
iTunes.
How much is your time worth?
Andy Champ
>
> Please .... tell us you can't get them already digitised from Amazon or
> iTunes.
>
>
my old folk club recordings on MD than AC, but it's there."
Andy
UnsteadyKen
Java Jive wrote...
> I
> put a blank MD in the player, pressing MD Record but also Pause, which
> allows the analogue source to be routed out via the MD's optical to be
> recorded by the PC;
On Sony MD decks you can skip the insert disc business, pressing Record
with no disc inserted will work as above and if a digital input is
selected instead of analogue then the deck will perform as an external
DAC. Thus you can't record over a wanted disc by mistook and when using
this method the laser is not activated, the laser life of MD decks seems
to be a weak point
> I assume, perhaps incorrectly, that this enables
> me to convert the analogue to digital without incurring the
> compression lossiness of recording it onto MD
I think this is correct from what I can gather from the manuals and
having used mine like this for a few years.
--
Ken O'Meara
http://www.btinternet.com/~unsteadyken/
Jim Lesurf
<ja...@evij.com.invalid> wrote:
[snip]
Afraid I can't make any comment wrt MiniDisc as I've never used one.
One general comment, though: Might be a good idea for you to copy your
posting to uk.rec.audio as there will be various people there who can
probably provide info and ideas. I'l make a few points below, though...
> Since Christmas, I've digitised H2G2 and all the MDs, and have nearly
> finished the first pass through the vinyls. However, and hence this
> post, I've had some totally unexpected problems ...
> H2G2 (ACs) ...
> I first recorded this via the Terratec, using the optical arrangement
> described above. However, on playback, the recordings sounded slightly
> but definitely, noticeably speeded up, vaguely suggestive of Mini Mouse
> on 'speed' and helium. When I simply moved the optical input to the SB
> Live, everything was fine.
> Why?!
It does seem like the recording was made at one sample rate, but the files
are being played back at a higher rate. What sample rate do the headers of
the LPCM Wave files specify for their content? What rate are they being
played back at?
IIRC you are using Linux. If so you can experiment with the ALSA commands
and the AlsaPlayer player. These let you overrule the sample rate
specified. Or you can bodge as I tend to by editing the header of the file.
:-)
> Vinyls ...
> Here the big problems are
> Finding a stable surface for the deck(s) Dust and gunge on the vinyls
> Mains Hum
[snip]
> My conclusion from all of this is that a vinyl record-deck should be
> wired as follows, but I'd be interested to see if others agree ...
> If the deck is driven from mains voltages, then its metalwork should be
> earthed via the earth in a three-core mains lead connected to the earth
> in a 3-pin plug. In this case, the metalwork of the arm should be
> insulated from the deck and a seperate earth point provided to connect
> the arm to the amp earth.
> Whether or not the deck is driven by mains voltages, the cartridge
> should be connectioned to the outputs via screened cable over the entire
> distance. The problem is that screened cables are stiffer than the tiny
> wiring commonly used in an arm, and their stiffness might affect the
> tracking if the job's not carefully done.
> I did once wire up an old Garrard deck like this, as an experiment. My
> recollection is that it was hum-free. I got the shielded cabling from
> RS or Maplin's - can't remember which, but I do remember that I used
> the same cabling in my TR when I swapped out some worn heads. It was two
> cores, about the same thickness as you'd find in a pick-up arm, with a
> common braiding outside, then a thin outer insulation. It was quite
> flexible, I do not recall ever noticing tracking problems with it.
> Thoughts on this reasoning?
The difficulty here is that in practice it comes down to an "it all
depends..." situation. In theory you ground the system in the kind of way
you describe. But in practice you may have to experiment. And some amps,
cartridges, arms, etc, work better if you alter details which would make
other combinations hum louder.
Personally, I to ground the TT only via a flying lead to the preamp, and
use the coax outers for the signal grounding to the preamp. No earth wire
to mains directly from the deck. That works fine with the deck I use. But
could easily be the wrong advice for something else!
You probably also know that if you change the coax leads and are using a MM
cart you need to ensure the leads have the optimum capacitance as well as
being well shielded.
> Next came recording via the laptop.
> Having had more success with the SB Live than the Terratec, I thought
> I'd better stick with that. Although the docking station was really
> bought so that I could connect the laptop instantly to my KVM and office
> network upstairs, it has a single PCI slot, so I bad farewell to
> convenience upstairs, I put one of the SBs in it, connected up to the
> HiFi, and started recording. However, when I checked the playback,
> there was a problem with distortion - some albums in particular were
> ruined by it.
> WTF? Why? And why some albums consistently so much more than others?
Hard to say without some measurements or a short file to examine. Could you
put up something like a very brief flac file of an example of the
distortion and one that isn't?
But my guess is a combination of the ADC clipping and some of the LPs
having more pressed in rumble or LF which you may not hear but cause large
signal excursions and so shove the signal into clipping.
FWIW I've found cases where the sound card clipped one sort of path whilst
set fine for another. Due to idiotic internal designs by the kinds of
clueless dimwits who design some computer hardware. They simply assume no
one will notice or care.
> The card has been working perfectly well for years in a desktop PC
> running Windows v5.0, why not in the docking station of a laptop running
> v5.1? Thinking, however that it might be something to do with driver
> versions under XP rather than 2000, I checked for the latest driver
> downloads, but those I've been using all along for 2000 are also the
> correct and latest for XP.
> So I tried the Terratec. Success at last!
> "Great!", I thought, "I don't need the docking station downstairs
> anymore!". So I got the normal PSU for the laptop, and removed the
> docking station. Big, big hum! Replaced the docking station. Back to
> normal. Removed it again. Big, big hum! Replaced it again. Back to
> normal.
> I conclude that the DS has a decent PSU, while that supplied for the
> laptop is crap. Or is it more subtle than that? Is there a mains loop
> being formed that I can do something about? I really would like to have
> the docking station back up here. Any thoughts?
If the PSU is lousy it may well inject hum in ways that can't then be fixed
by looking for a 'loop'. May simply be shedloads of ripple on the power
rails. Or producing large local fields. When I tried using the analogue on
my old laptop I could clearly hear the disc activity producing electrical
noises as it fizzed the rails. In general computer PSUs are rubbish so far
as audio is concerned.
For me the solution is always to use *external* DAC/ADC arrangements
specifically designed for audio. Not trust anything sold by 'computer'
makers. Note BTW that some more modern USB sound items by audio firms use
well loop-isolated USB connections to break any loops or hf injections.
Ideally something with a ground/lift switch gives you a choice.
You may be able to sort the problems by experimenting with wiring and
levels. But my reaction is to recommend getting some explicitly audio items
to do the ADC part.
FWIW I used Pioneer home CDRW audio recorders for years to digitise audio.
They work very nicely indeed. You can still find them secondhand. And other
CDRW audio recorders are made by other people. The snag is they are limited
to the CDDA format. But they work nicely if you are willing to then take
the CDRW to the computer to rip the results.
More recently I've used the Tascam HDP2 which I regard as excellent. But it
is costly. There are cheaper solid-state recorders. Some may be excellent.
But I haven't tried any. If you ask on uk.rec.audio someone there may be
able to diagnose you specific problem above or suggest an alternative.
Jim Lesurf
> In article <pn89v65ta3r6e8gto...@4ax.com>, Java Jive
> <ja...@evij.com.invalid> wrote: [snip]
> > "Great!", I thought, "I don't need the docking station downstairs
> > anymore!". So I got the normal PSU for the laptop, and removed the
> > docking station. Big, big hum! Replaced the docking station. Back
> > to normal. Removed it again. Big, big hum! Replaced it again. Back
> > to normal.
One question occurs to me, to clarify:
Which of the items - preamp / docking station / laptop psu - have their own
ground to the earth pin of the mains plug?
John Legon
> Java Jive wrote...
>
>> I
>> put a blank MD in the player, pressing MD Record but also Pause, which
>> allows the analogue source to be routed out via the MD's optical to be
>> recorded by the PC;
>
> On Sony MD decks you can skip the insert disc business, pressing Record
> with no disc inserted will work as above and if a digital input is
> selected instead of analogue then the deck will perform as an external
> DAC.
I've used this feature to listen to the digital output from my satellite
box. With no disk inserted, my Sony MD deck displays "AD - DA" when an
analogue source is selected, and just "DA" for a digital source.
>> I assume, perhaps incorrectly, that this enables
>> me to convert the analogue to digital without incurring the
>> compression lossiness of recording it onto MD
>
> I think this is correct from what I can gather from the manuals and
> having used mine like this for a few years.
Same here, though I don't use the digital output.
Java Jive
uk.tech.digital-tv for the duplication of previous posts that this
necessarily entails ...
On Mon, 13 Jun 2011 09:58:58 +0100, Jim Lesurf <no...@audiomisc.co.uk>
wrote:
>
> One general comment, though: Might be a good idea for you to copy your
> posting to uk.rec.audio as there will be various people there who can
> probably provide info and ideas. I'll make a few points below, though ...
...
> IIRC you are using Linux.
Not for this Jim. I tend to use Windows generally, Linux for
particular purposes.
In article <pn89v65ta3r6e8gto...@4ax.com>, Java Jive
<ja...@evij.com.invalid> wrote:
:Long ago I digitised my Audio-Cassette and my VHS collections, but I
:still have some outdated audio recordings in the form of:
:
: ACs: 'The Hitchhikers Guide To The Galaxy' first 2 radio series
: About 60 MiniDisks.
: About 125 LPs
: About 5 x 45 rpms
:
:At one time there were over 400 LPs and over 100 45s. Most of these
:have been either replaced by CDs, already digitised, a few wanted
:tracks found elsewhere, or just become superfluous to my artistic well
:being!
:
:Then of course, there is the attendant hardware:
:
: Denon HiFi (seperates) around DRA-275RD amp downstairs in lounge
: Sanyo DC-007C Midi Tower (seperates) upstairs in bedroom/office.
:
: DMD-1300 MD deck in the downstairs HiFi (may be faulty)
: MDG-007 MD deck in the upstairs HiFi
:
: Dual 601 turntable, aged, some rumble, no preamp.
: Phono inputs on Denon DRA-275RD downstairs.
: NAD Phono Preamp.
: Project TK38 turntable with inbuilt pre-amp.
: Vinyl washing machine
:
: 2 x Desktop P4s W2k (still), each with ...
: SB Live with Digital IO dongle (SPDIF Coax & Optical In & Out)
:
: Dell Latitude 610 laptop + docking station
:
: USB Terratec Aureon 5.1 MKII (SPDIF Optical In & Out)
:
:In order to reduce the chances of picking up hum and stray e-m, I
:always prefer to use optical digital connections whereever possible. I
:can do this upstairs by connecting optically between the MD output and
:an optical input on the PC - for AC or any other analog source, I
:put a blank MD in the player, pressing MD Record but also Pause, which
:allows the analogue source to be routed out via the MD's optical to be
:recorded by the PC;
On Mon, 13 Jun 2011 10:31:49 +0100, John Legon
<jo...@nospam.demon.co.uk> wrote:
>
> UnsteadyKen wrote:
> >
> > On Sony MD decks you can skip the insert disc business, pressing Record
> > with no disc inserted will work as above and if a digital input is
> > selected instead of analogue then the deck will perform as an external
> > DAC.
>
> I've used this feature to listen to the digital output from my satellite
> box. With no disk inserted, my Sony MD deck displays "AD - DA" when an
> analogue source is selected, and just "DA" for a digital source.
:I assume, perhaps incorrectly, that this enables
:me to convert the analogue to digital without incurring the
:compression lossiness of recording it onto MD and then playing it
:back, but in truth this is an informed guess, I don't actually know
:this for certain, and don't recall actually testing this belief, which
:I suppose I ought to have. Bit late now though.
> > I think this is correct from what I can gather from the manuals and
> > having used mine like this for a few years.
>
> Same here, though I don't use the digital output.
:Unfortunately, there is no such solution for the vinyls.
:
:I record everything to LPCM wave files.
:
:For some time, I've been thinking that it would be good to:
:
: :-) Forestall irreplacable, obsolescent equipment going down
: :-) Digitise these remaining vinyls before further deterioration
: :-) Be able to play the vinyls without the associated hassle
: :-) Lose this significant pile of junk, but keep the recordings.
:
:Since Christmas, I've digitised H2G2 and all the MDs, and have finished a first pass through the vinyls. However, and hence this post, I've had some totally unexpected problems ...
:
:
:H2G2 (ACs) ...
:
:I first recorded this via the Terratec, using the optical arrangement
:described above. However, on playback, the recordings sounded
:slightly but definitely, noticeably speeded up, vaguely suggestive of
:Mini Mouse on 'speed' and helium. When I simply moved the optical
:input to the SB Live, everything was fine.
:
:Why?!
On Mon, 13 Jun 2011 09:58:58 +0100, Jim Lesurf <no...@audiomisc.co.uk>
wrote:
>
> It does seem like the recording was made at one sample rate, but the files
> are being played back at a higher rate. What sample rate do the headers of
> the LPCM Wave files specify for their content? What rate are they being
> played back at?
According to the headers, there's no difference ...
Terratec Aureon Wave File: PCM 44,100Hz, 2 channels, 16bit
SB Live Wave File: PCM 44,100Hz, 2 channels, 16bit
... BUT on average each Terratec file is 23.67Mb smaller than the
equivalent SB Live version! There must be a reason for that, so I
still think that somehow you are right. The precise mechanism is
beyond me, however.
:MDs ...
: Earthing via the Earth terminal of the amp, and not
: Using the seperate preamp, and the phono inputs on the amp
:
:Finally, in desperation, I took the turntable out of its plinth, and discovered that the grounds of the cartridge connections were connected to the metal of the deck. I cut these connections and rewired them so that the metal of the deck is earthed via the mains lead, and seperately the grounds of the cartridge, like the signals, come straight out of the back to the outputs. This is a HUGE improvement: there is no build-up of static, and it's effectively hum-free until I start the motor. With the motor running, there is some residual hum, which I'd like to get rid of, but at least the result is listenable.
:
:When I tried the new arrangement with the NAD pre-amp, there was just
:a little more hum, so I've been using the HiFi phono inputs.
:When I tried it additionally connecting the metalwork to the Earth
:connection on the amp either there was either no difference or it was
:worse, I can't now remember which, only that no benefit was obtained.
:
:My conclusion from all of this is that a vinyl record-deck should be
:wired as follows, but I'd be interested to see if others agree ...
:
:If the deck is driven from mains voltages, then its metalwork should be earthed via the Earth in a three-core mains lead connected to the Earth in a 3-pin plug. In this case, the metalwork of the arm should be insulated from that of the deck and a seperate Earth point provided to connect the arm to the amp Earth.
:
:Whether or not the deck is driven by mains voltages, the cartridgeshould be connectioned to the outputs via screened cable over theentire distance. The problem is that screened cables are stiffer thanthe tiny wiring commonly used in a pick-up arm, and their stiffness mightaffect the tracking if the job's not carefully done.
:
:I did once wire up an old Garrard deck like this, as an experiment. My
:recollection is that it was hum-free. I got the shielded cabling from
:RS or Maplin's - can't remember which, but I do remember that I used
:the same cabling in my TR when I swapped out some worn heads. It was
:two cores, about the same thickness as you'd find in a pick-up arm,
:with a common braiding outside, then a thin outer insulation. It was
:quite flexible, I do not recall ever noticing tracking problems with
:it.
:
:Thoughts on this reasoning?
On Mon, 13 Jun 2011 09:58:58 +0100, Jim Lesurf <no...@audiomisc.co.uk>
wrote:
>
> The difficulty here is that in practice it comes down to an "it all
> depends..." situation. In theory you ground the system in the kind of way
> you describe. But in practice you may have to experiment. And some amps,
> cartridges, arms, etc, work better if you alter details which would make
> other combinations hum louder.
>
> Personally, I to ground the TT only via a flying lead to the preamp, and
> use the coax outers for the signal grounding to the preamp. No earth wire
> to mains directly from the deck. That works fine with the deck I use. But
> could easily be the wrong advice for something else!
Well at least I'm along the right lines, and not saying something that
is actually wrong. It would be nice to be able to give more specific
advice on my webpage linked above though.
> You probably also know that if you change the coax leads and are using a MM
> cart you need to ensure the leads have the optimum capacitance as well as
> being well shielded.
Can you expound further on this?
The Dual 601 has a Shure VN35MR, which by recollection I'm fairly
certain is a magnetic cartridge. I'm not aware of problems with the
signal except those that are most easily explained by the earthing
problems I've outlined, and the use of unshielded cable in the pick-up
arm. Connected as it currently is, normal listening level is between
a quarter and a third round the amp's volume dial. With the arm at
rest and the motor not running, I can hear white noise and hum if I
turn the volume all the way up, but if I were to play any signal at
that level, it would probably demolish the house. The only
appreciable hum at listening levels is when the motor is running.
:Next came recording via the laptop.
:
:Having had more success with the SB Live than the Terratec, I thought
:I'd better stick with that. Although the docking station was really
:bought so that I could connect the laptop instantly to my KVM and
:office network upstairs, it has a single PCI slot, so I bad farewell
:to convenience upstairs, I put one of the SBs in it, connected up to
:the HiFi, and started recording. However, when I checked the
:playback, there was a problem with distortion - some albums in
:particular were ruined by it.
:
:WTF? Why? And why some albums consistently so much more than others?
:The card has been working perfectly well for years in a desktop PC
:running Windows v5.0, why not in the docking station of a laptop
:running v5.1? Thinking, however that it might be something to do with
:driver versions under XP rather than 2000, I checked for the latest
:driver downloads, but those I've been using all along for 2000 are
:also the correct and latest for XP.
On Mon, 13 Jun 2011 09:58:58 +0100, Jim Lesurf <no...@audiomisc.co.uk>
wrote:
>
> Hard to say without some measurements or a short file to examine. Could you
> put up something like a very brief flac file of an example of the
> distortion and one that isn't?
To my ears, it sounds similar to a worn CD deck, what perhaps you
might term digital jitter.
Terratec:
http://www.macfh.co.uk/PrivTest/HouseBand-HeresTheTenderClean.wav
SBLive:
http://www.macfh.co.uk/PrivTest/HouseBand-HeresTheTenderDistort.wav
> But my guess is a combination of the ADC clipping and some of the LPs
> having more pressed in rumble or LF which you may not hear but cause large
> signal excursions and so shove the signal into clipping.
Looking at the actual waveform, I can't see any signs of clipping.
> FWIW I've found cases where the sound card clipped one sort of path whilst
> set fine for another. Due to idiotic internal designs by the kinds of
> clueless dimwits who design some computer hardware. They simply assume no
> one will notice or care.
Does seem odd that the same card with the same drivers works perfectly
well in a desktop.
:So I tried the Terratec. Success at last!
:
:"Great!", I thought, "I don't need the docking station downstairs
:anymore!". So I got the normal PSU for the laptop, and removed the
:docking station. Big, big hum! Replaced the docking station. Back
:to normal. Removed it again. Big, big hum! Replaced it again. Back
:to normal.
:
:I conclude that the DS has a decent PSU, while that supplied for the
:laptop is crap. Or is it more subtle than that? Is there a mains
:loop being formed that I can do something about? I really would like
:to have the docking station back up here. Any thoughts?
On Mon, 13 Jun 2011 09:58:58 +0100, Jim Lesurf <no...@audiomisc.co.uk>
wrote:
>
> If the PSU is lousy it may well inject hum in ways that can't then be fixed
> by looking for a 'loop'. May simply be shedloads of ripple on the power
> rails. Or producing large local fields. When I tried using the analogue on
> my old laptop I could clearly hear the disc activity producing electrical
> noises as it fizzed the rails. In general computer PSUs are rubbish so far
> as audio is concerned.
>
> One question occurs to me, to clarify:
>
> Which of the items - preamp / docking station / laptop psu - have their own
> ground to the Earth pin of the mains plug?
Docking station has three-core, clover leaf connector mains lead.
Laptop has inline power-block connected to the mains via a twin-core
lead and to the laptop via a standard coaxial connector
Despite having an Earth connector for the phono inputs, the HiFi amp
actually only has a twin-core lead, so no actual Earth! Thank you,
Jim! I surely should've checked this basic point before, but it
simply didn't occur to me that an amp with an external Earth point
would not itself be earthed! This goes some way to explaining why I
used the Dual turntable for years with an earthed amp without any
problems.
> For me the solution is always to use *external* DAC/ADC arrangements
> specifically designed for audio. Not trust anything sold by 'computer'
> makers. Note BTW that some more modern USB sound items by audio firms use
> well loop-isolated USB connections to break any loops or hf injections.
> Ideally something with a ground/lift switch gives you a choice.
>
> You may be able to sort the problems by experimenting with wiring and
> levels. But my reaction is to recommend getting some explicitly audio items
> to do the ADC part.
>
> ...
Sorry, should've mentioned that I don't want to spend any more money
on this than the minimum necessary. I didn't mind the laptop, because
that is very useful for other things, and will retain its second-hand
value for a while.
Jim Lesurf
<ja...@evij.com.invalid> wrote:
.
> > You probably also know that if you change the coax leads and are using
> > a MM cart you need to ensure the leads have the optimum capacitance as
> > well as being well shielded.
> Can you expound further on this?
> The Dual 601 has a Shure VN35MR, which by recollection I'm fairly
> certain is a magnetic cartridge.
VN35MR is the part number for the MR (Micro Ridge) stylus assembly for the
V15/III. So I assume the cartridge is the V15/III. (FWIW I also use the
same as yourself. But in my case with the Technics SL1200 deck.)
The V15 is called a moving *magnet* cartridge. (More precisely, a moving
iron variable magnetic shunt.)
These require to see a load capacitance of a few hundred pF as well as 47k.
If you vary either the capacitance or resistance of their loading then the
frequency response will change. Since general co-ax of the types people use
for home audio have capacitances in the range from about 50 to a few
hundred pF *per meter* you need to cut the length of the coax chosen so
that its capacitance (plus that of the preamp input) make up the required
capacitance. I've forgotten the optimum value. I can look it up if you
wish, but IIRC it is around 300 to 400 pF. Take off about 100pF for the amp
if you have no value given for it.
This *won't* affect things like hum pickup or have much effect on audible
distortion, though.
The good news is that the V15 is well shielded against magnetic field
pickup and the coils should be grounded back to the co-ax correctly. That
was the era when Shure really did make good cartridges!
> I'm not aware of problems with the signal except those that are most
> easily explained by the earthing problems I've outlined, and the use of
> unshielded cable in the pick-up arm. Connected as it currently is,
> normal listening level is between a quarter and a third round the amp's
> volume dial. With the arm at rest and the motor not running, I can hear
> white noise and hum if I turn the volume all the way up, but if I were
> to play any signal at that level, it would probably demolish the house.
> The only appreciable hum at listening levels is when the motor is
> running.
Yes, that sounds fine. About what I get and would expect.
[snip]
> >
> > Hard to say without some measurements or a short file to examine.
> > Could you put up something like a very brief flac file of an example
> > of the distortion and one that isn't?
> To my ears, it sounds similar to a worn CD deck, what perhaps you might
> term digital jitter.
> Terratec:
> http://www.macfh.co.uk/PrivTest/HouseBand-HeresTheTenderClean.wav
> SBLive:
> http://www.macfh.co.uk/PrivTest/HouseBand-HeresTheTenderDistort.wav
OK. I'll download those and have a listen, etc, when I get a chance.
Although since we are now also posting to uk.rec.audio I suspect someone
else there may well beat me to it! :-)
> Does seem odd that the same card with the same drivers works perfectly
> well in a desktop.
Can't really comment on that beyond making the general point that the rest
of the hardware, rail condition, etc, may well matter. No guarantee that
the same card and driver will work exactly the same in every 'host' system.
[snip]
> >
> > Which of the items - preamp / docking station / laptop psu - have
> > their own ground to the Earth pin of the mains plug?
> Docking station has three-core, clover leaf connector mains lead.
> Laptop has inline power-block connected to the mains via a twin-core
> lead and to the laptop via a standard coaxial connector
FWIW You can sometimes alter hum behaviour by unplugging a 'figure of
eight' plug, rotating it 180 deg, and plugging it back in again. So
changing over which part of the socket gets neutral and which live. In
theory this never makes any difference. But in practice it can. :-)
> Despite having an Earth connector for the phono inputs, the HiFi amp
> actually only has a twin-core lead, so no actual Earth! Thank you, Jim!
> I surely should've checked this basic point before, but it simply
> didn't occur to me that an amp with an external Earth point would not
> itself be earthed! This goes some way to explaining why I used the Dual
> turntable for years with an earthed amp without any problems.
The point to bear in mind is that 'ground' and 'chassis' and 'earth' are
all often used as if they meant the same things. They don't. :-)
The purpose of the 'earthing' from the POV of the signal is to remove loops
and connect the system into a 'box' which surrounds the signal with an
equipotential shield of metal. So the coax, box of preamp, etc, all make a
'screen' for E-field and there is a controlled set of current paths to
avoid loops.
So in theory you could float the entire system with no 'earth' at all. This
is fine... except for the snag that often then turn up in practice. One of
these is the risk of killing yourself if there is a fault. Hence, for
example, the UK enthusiasm for having an 'earth' connection. And in
practice unwanted field couplings and paths via the PSU may mean you need
an 'Earth'.
FWIW I once worked on a telescope which we had to 'float' to avoid loops
and other signal problems. The snag was that the potential of the entire
telescope and all its kit wandered up to a few kV if left to itself. This
was via a big isolation resistance from its surroundings and psus.
So we used to take turns to go out and adjust anything. The first move of a
hand towards the telescope produced quite a noticable spark from fingers to
telescope in the dark. Accompanied by and "Ouch!" from the owner of the
fingers. :-) If you then kept holding on to the telescope whilst working
you were fine. The current was tiny. But if no-one had touched the
telescope for 20mins the next user would pay the entrance fee. ;->
Bit like a big supermarket trolley. :-)
> > For me the solution is always to use *external* DAC/ADC arrangements
> > specifically designed for audio. Not trust anything sold by 'computer'
> > makers. Note BTW that some more modern USB sound items by audio firms
> > use well loop-isolated USB connections to break any loops or hf
> > injections. Ideally something with a ground/lift switch gives you a
> > choice.
> >
> > You may be able to sort the problems by experimenting with wiring and
> > levels. But my reaction is to recommend getting some explicitly audio
> > items to do the ADC part.
> >
> > ...
> Sorry, should've mentioned that I don't want to spend any more money on
> this than the minimum necessary. I didn't mind the laptop, because that
> is very useful for other things, and will retain its second-hand value
> for a while. --
That's fair enough. But it does have the 'price' you are encountering. That
you then have instead to spend some time trying to sort out problems. Good
way to 'learn' if you have the time and patience, though. :-)
Terry Casey
Jim Lesurf <no...@audiomisc.co.uk> wrote:
>
> FWIW I once worked on a telescope which we had to 'float' to avoid loops
> and other signal problems. The snag was that the potential of the entire
> telescope and all its kit wandered up to a few kV if left to itself. This
> was via a big isolation resistance from its surroundings and psus.
>
> So we used to take turns to go out and adjust anything. The first move of a
> hand towards the telescope produced quite a noticable spark from fingers to
> telescope in the dark. Accompanied by and "Ouch!" from the owner of the
> fingers. :-) If you then kept holding on to the telescope whilst working
> you were fine. The current was tiny. But if no-one had touched the
> telescope for 20mins the next user would pay the entrance fee. ;->
>
Not on the same scale but this reminds me of moving into a new building in 1970
fitted throughout - including the corridors - with synthetic carpets. You
couldn't move far without getting a belt - the size of which, of course, was
determined by the size of the metal object you touched.
Metal door handles were a particular pain (pardon the pun) - probably because
you'd usually walked some way before you got to them and built up quite a hefty
charge!
The worst, in general, were metal filing cabinets, coming only second to the
rare occasions when you came into contact with metal that was directly earthed
- surprisingly uncommon in the 1970 office.
Quite by chance, I discovered one day that if I first touched the office door
handle with the backs of my fingers, so that my finger nails made the initial
physical contact, I would hear the crack of the discharge but didn't feel
anything!
I became quite adept at flicking my nails against metal objects before touching
them - to the extent I sometimes caught myself doing it elsewhere, such as at
home! After that, I never had any shocks.
Also, when I got married a little later, I found that making initial contact
with my wedding ring had the same effect although, by this time wear and the
frequent application of antistatic sprays seemed to have diminished the problem
somewhat.
I've never been able to work out why such a simple change could change the
discharge of the same high static voltages from a painful to a painless
operation.
Sorry if this is a bit late for your telescope problems, Jim ...
--
Terry
Jim Lesurf
> In article <jjbcv6d1aunt2i7uc...@4ax.com>, Java Jive
> <ja...@evij.com.invalid> wrote: .
> > To my ears, it sounds similar to a worn CD deck, what perhaps you
> > might term digital jitter.
> > Terratec:
> > http://www.macfh.co.uk/PrivTest/HouseBand-HeresTheTenderClean.wav
> > SBLive:
> > http://www.macfh.co.uk/PrivTest/HouseBand-HeresTheTenderDistort.wav
> OK. I'll download those and have a listen, etc, when I get a chance.
> Although since we are now also posting to uk.rec.audio I suspect someone
> else there may well beat me to it! :-)
I've now had a chance to listen to the above. It does sound to me like some
kind of buffering or timing problem with the 'Distort' version. I've not
yet examined the audio data. But to me it sounds like the kind of effect
where a buffer keeps under- or over-flowing so there are bursts of
lost/repeated/altered values until the system sorts itself out - only for
the same thing to duly happen again. May also be due to interrupts or
other transfers regularly disrupting the data flow.
If I get a chance I'll see if I can find any diagnostic pattens in the
data. But the 'distortion' does seem to be a sampling timing/transfer
problem.
I would not call this 'jitter' in the normal hifi sense. That is where all
the samples are present, correct, and in the right order. and the timing
variations are generally much smaller than the intersample period. So
rather more subtle than what can be heard in the above!
Bill Wright
> In message <51e31f7...@audiomisc.co.uk> on Tue, 14 Jun 2011 09:20:02 +0100
> Jim Lesurf <no...@audiomisc.co.uk> wrote:
>> FWIW I once worked on a telescope which we had to 'float' to avoid loops
>> and other signal problems. The snag was that the potential of the entire
>> telescope and all its kit wandered up to a few kV if left to itself. This
>> was via a big isolation resistance from its surroundings and psus.
>>
>> So we used to take turns to go out and adjust anything. The first move of a
>> hand towards the telescope produced quite a noticable spark from fingers to
>> telescope in the dark. Accompanied by and "Ouch!" from the owner of the
>> fingers. :-) If you then kept holding on to the telescope whilst working
>> you were fine. The current was tiny. But if no-one had touched the
>> telescope for 20mins the next user would pay the entrance fee. ;->
discharge the static slowly, and then you don't get a shock when you
touch the metal.
Staff in new TV shops often complain that they're getting a shock from
the TV aerial plugs. The TV system is earth bonded and the carpets are new.
Bill
Java Jive
It doesn't seem as though there's much to be done about it, but at
least the Terratec sounds alright!
On Tue, 14 Jun 2011 18:02:05 +0100, Jim Lesurf <no...@audiomisc.co.uk>
wrote:
>
> I've now had a chance to listen to the above. It does sound to me like some
> kind of buffering or timing problem with the 'Distort' version.
Brian Gaff
live actually sample at 48khz, then do a conversion to the required rate on
the fly, so to speak, resulting in awkward ones, like to cd standard, in
some odd effects.
Brian
--
Brian Gaff....Note, this account does not accept Bcc: email.
graphics are great, but the blind can't hear them
Email: bri...@blueyonder.co.uk
______________________________________________________________________________________________________________
"Java Jive" <ja...@evij.com.invalid> wrote in message
news:jjbcv6d1aunt2i7uc...@4ax.com...
Jim Lesurf
<Bri...@blueyonder.co.uk> wrote:
> On the sample rate thing, did I not read that many cards, including the
> sb live actually sample at 48khz, then do a conversion to the required
> rate on the fly, so to speak, resulting in awkward ones, like to cd
> standard, in some odd effects.
I can't comment on SB, but yes, I've encountered this problem. Indeed, the
machine I'm typing this on has a 'soundcard'[1] that is locked to 48k and
simply drops or repeats samples to 'get any other rate'. This is ridiculous
in audio terms, but seems to have been par for the course for many years in
home computers. The assumption is that users will always take "I can hear
something" to mean "works fine" and that awful levels of distortion, etc,
won't bother the cloth-eared user. Indeed, I do wonder how many makers of
computer hardware have themselves had any clue about audio quality or have
ever heard a decent audio setup.
In other cases, though, a fixed rate may be being imposed by the
OS/software running. Both the Linux boxes I use on a daily basis started
off forcing 48k onto everything - although fortunately with decent
resampling. But after experimenting with the software and OS I overcame
this and the rate follows the source material as it should. The problem
here seems to be programmers who thoughtlessly assume '48k is standard' or
even '48k is better than 44.1k'.
The good thing about USB audio is that it makes it easier to sidestep some
of the rubbish hardware that gets shoved into domestic general purpose
computers.
Slainte,
Jim
[1] Actually an AC97 mimic chip stuck on the motherboard.
Andy Burns
> Indeed, the
> machine I'm typing this on has a 'soundcard'[1] that is locked to 48k and
> simply drops or repeats samples to 'get any other rate'. This is ridiculous
> in audio terms,
And if the chipset doesn't do that, the drivers (ALSA or pulseaudio,
Windows kernel mixer) are likely to do it instead, for bogstandard PC
usage I suppose this makes sense, allows the user to alter volume of
individual programs rather than all sounds, allows system "beeps" to
play at one sample rate while iPlayer is running in the background at
another rate.
Terry Casey
Bill Wright <bi...@invalid.com> wrote:
>
> Terry Casey wrote:
> > In message <51e31f7...@audiomisc.co.uk> on Tue, 14 Jun 2011 09:20:02 +0100
> > Jim Lesurf <no...@audiomisc.co.uk> wrote:
> >> FWIW I once worked on a telescope which we had to 'float' to avoid loops
> >> and other signal problems. The snag was that the potential of the entire
> >> telescope and all its kit wandered up to a few kV if left to itself. This
> >> was via a big isolation resistance from its surroundings and psus.
> >>
> >> So we used to take turns to go out and adjust anything. The first move of a
> >> hand towards the telescope produced quite a noticable spark from fingers to
> >> telescope in the dark. Accompanied by and "Ouch!" from the owner of the
> >> fingers. :-) If you then kept holding on to the telescope whilst working
> >> you were fine. The current was tiny. But if no-one had touched the
> >> telescope for 20mins the next user would pay the entrance fee. ;->
> The solution is to touch the metal with a short glass rod. That will
> discharge the static slowly, and then you don't get a shock when you
> touch the metal.
>
>
> Staff in new TV shops often complain that they're getting a shock from
> the TV aerial plugs. The TV system is earth bonded and the carpets are new.
>
--
Terry
Jim Lesurf
Burns
Yes, I'd agree that for general use that is likely the sort of presumptions
the providers of the OS/software have tended to make in the past. However I
don't think it really 'makes sense' these days when the same systems can
easily switch rate with source when only one source stream is playing at a
time. And when increasing numbers of people are using a home computer to
play music more seriously. There is also the problem that the assumptions
of the OS/software providers are imposed without giving you any warning, or
flagging up what is happening. Also in some cases the 'resampling' and
'mixing' is done in an astonishingly incompetent way!
Slainte,
Jim
Albert Ross
wrote:
>The maxim of the Royal Society (removing the fancy latin) is "Take no-one's
>word for it". Good advice in my experience.
>
>FWIW A number of times during my professional career I was told words to
>the effect of "Why investigate that, there isn't anything there to be
>found!"... and then discovered something interesting that helped move some
>part of science or engineering along a bit.
Improbably, in 1945 Oxford University disbanded their research into
nutrition on the basis that "everything was already known."
The whole field is now in the middle of a paradigm shift and divided
between appeals to Authority vs. facts.
"Everything You Know Is Wrong" (Firesign Theater c. 1970)
Norman Wells
> On Sun, 12 Jun 2011 16:01:00 +0100, Jim Lesurf <no...@audiomisc.co.uk>
> wrote:
>
>
>> The maxim of the Royal Society (removing the fancy latin) is "Take
>> no-one's word for it". Good advice in my experience.
>>
>> FWIW A number of times during my professional career I was told
>> words to the effect of "Why investigate that, there isn't anything
>> there to be found!"... and then discovered something interesting
>> that helped move some part of science or engineering along a bit.
>
> Improbably, in 1945 Oxford University disbanded their research into
> nutrition on the basis that "everything was already known."
Quite right too!
"Everything that can be invented has been invented"
- Charles H. Duell, Commissioner, U.S. patent office, 1899