Interesting Journal Article on filtering/differences betwe..

Archived from groups: rec.audio.high-end (More info?)

The current issue of Hi-Fi News (July 2004, pp 80-82) contains a summary
article covering work done by Peter Craven and published in the Journal of
the AES. Citation follows:

P. Craven, "Anti Alias Filters and System Transient Response at High Sample
Rates", Journal of the Audio Engineering Society, March 2004.

Mr. Craven was hired by Meriden apparently to work on improving filtering
for DVD-A at 96khz and 192khz. In his investigations, he found SACD's
asymmetrical analog filter to have a -100db to -100db rise and fall time of
35 micro-seconds (almost all post-transient with no pre-smear) to a
transient pulse maxing at 0 db reference. The convention linear phase
digital filter used in most DVD-A players yielded a symmetrical -100db
to -100db rise and fall time of over 4 milliseconds, even when filtering
192khz. He designed digital, asymmetric filters that managed to bring the
rise-and-fall time down to 300 microseconds for 96khz and 100 microseconds
for 192khz. The author of the summary article concluded that these latter
at least had the potential of putting DVD-A "in the range of SACD as regards
transient capability" (paraphrased quote, since I don't have the article in
front of me).

Interesting stuff that I haven't seen discussed before, at least in the
popular/audiophile press or here on RAHE or rec.audio.tech. There may be
something to SACD aficionados perception of greater macro- and micro-
dynamics after all, if there has been this much difference between SACD (no
pre-smear, short post-smear) and traditional digital linear filtering
(substantial pre- and post- smear).

Craven makes the point that the filter needs to be inserted just once in the
chain (ad,da) to gain the benefit. Sounds like may be in part a key as to
Meriden's DVD-A success.

Harry Lavo
"It don't mean a thing if it aint got that swing" - Duke Ellington

Archived from groups: rec.audio.high-end (More info?)

TChelvam wrote:

> "Harry Lavo" <harry.lavo@rcn.com> wrote in message news:<cceiit019mg@news1.newsguy.com>...
>> The current issue of Hi-Fi News (July 2004, pp 80-82) contains a summary
>> article covering work done by Peter Craven and published in the Journal of
>> the AES. Citation follows:
>>
>> P. Craven, "Anti Alias Filters and System Transient Response at High Sample
>> Rates", Journal of the Audio Engineering Society, March 2004.
>>
>>snip..snip..
>
> Mr. Craven was hired by Meriden apparently to work on improving
> filtering
>>>
>> Interesting stuff that I haven't seen discussed before, at least in the
>> popular/audiophile press or here on RAHE or rec.audio.tech. There may be
>> something to SACD aficionados perception of greater macro- and micro-
>> dynamics after all, if there has been this much difference between SACD (no
>> pre-smear, short post-smear) and traditional digital linear filtering
>> (substantial pre- and post- smear).
>>
>
> Hi Harry, a timely post to help gullible souls like myself from being
> manipulated by the advertisements and write-ups in high End magazine
> and scientific opinion on RAHE.

Can you give some examples of how you have been mainpulated by
scientific opinion on RAHE?

> For a moment, I thought I was
> convinced that SACD and High format resolution was all hype despite I
> was hearing what appeared to my ears to be a more realistic musical
> reproduction. At least that was my first impression of SACD.
>
> However, as time passes by and under my own DBT which will not meet
> the scientific standard set by physicists and engineers, I was unable
> to tell if I am hearing a particular number in SACD format or CD
> unless a immediate comparison done (without level matching, of
> course).
>
> I was beginning to convince myself that High-resolution format is more
> a business strategy and it got nothing to do "perfect sound".
> Especially after reading somewhere here that CD done correctly is more
> than adequate to reproduce all frequencies within human audible range.

Can you hear above 20KHz?

> Since my dog (whose audible frequencies extending beyond 20KHZ) who
> had accompanied me the last 13 years listening music died 5 months
> ago, I hardly bought any SACD software. Why should I since under DBT I
> can't tell the diff.

If you followed some of the recent threads, you would realize that there
could be major differences in the resulting sound due to differences in
mastering.

>
> Now, you have thrown me into another dilemma. Who is this Peter
> Craven? Is he being impartial? Or was he doing marketing for Meridian?
> And why it took the Engineer to find out that filters and 30 micro
> seconds actually make a difference? Don't they know that already?

Did anyone conclude that those differences are audible?

>
> If only some one who has the expertise in this field could provide an
> answer. But as in the past, tough questions usually do not catch the
> eyes of experts in RAHE.
>

There have been many posts on differences between SACD and CD. Do a
google search on this topic in rahe.

Archived from groups: rec.audio.high-end (More info?)

On 10 Jul 2004 14:40:18 GMT, Codifus <codifus@optonline.net> wrote:

>Harry Lavo wrote:
>> The current issue of Hi-Fi News (July 2004, pp 80-82) contains a summary
>> article covering work done by Peter Craven and published in the Journal of
>> the AES. Citation follows:
>>
>> P. Craven, "Anti Alias Filters and System Transient Response at High Sample
>> Rates", Journal of the Audio Engineering Society, March 2004.
>>
>> Mr. Craven was hired by Meriden apparently to work on improving filtering
>> for DVD-A at 96khz and 192khz. In his investigations, he found SACD's
>> asymmetrical analog filter to have a -100db to -100db rise and fall time of
>> 35 micro-seconds (almost all post-transient with no pre-smear) to a
>> transient pulse maxing at 0 db reference. The convention linear phase
>> digital filter used in most DVD-A players yielded a symmetrical -100db
>> to -100db rise and fall time of over 4 milliseconds, even when filtering
>> 192khz. He designed digital, asymmetric filters that managed to bring the
>> rise-and-fall time down to 300 microseconds for 96khz and 100 microseconds
>> for 192khz. The author of the summary article concluded that these latter
>> at least had the potential of putting DVD-A "in the range of SACD as regards
>> transient capability" (paraphrased quote, since I don't have the article in
>> front of me).
>>
>> Interesting stuff that I haven't seen discussed before, at least in the
>> popular/audiophile press or here on RAHE or rec.audio.tech. There may be
>> something to SACD aficionados perception of greater macro- and micro-
>> dynamics after all, if there has been this much difference between SACD (no
>> pre-smear, short post-smear) and traditional digital linear filtering
>> (substantial pre- and post- smear).
>>
>> Craven makes the point that the filter needs to be inserted just once in the
>> chain (ad,da) to gain the benefit. Sounds like may be in part a key as to
>> Meriden's DVD-A success.
>>
>> Harry Lavo
>> "It don't mean a thing if it aint got that swing" - Duke Ellington
>So SACD does have a technical advantage over DVD-A.

No, DVD-A at 24/192 will always be *vastly* superior to DSD. Filtering
is an entirely separate issue.

>I've always been a
>fan of DVD-A . . . .and still am. Even if it isn't perfect at the
>beginning, it definitely will mature as a medium, like CD did. SACD, on
>the other hand, despite being able to handle transients better, was
>technically inferior to even CD in at least 1 aspect. Seeing that it
>failed to oust the inferior Redbook CD on all technical merits, and even
>though I most probably would not hear that difference in its failing, I
>dismissed SACD. It's just the principle of the thing. If you're going
>to replace CD, then you'd better be better than it in everything. DVD-A
>does that from the beginning, and it will just get better as it matures.
>
>I was about to jump into buying a high-end audio unit, but after seeing
>this, I'll keep enjoying my Redbooks and wait to see players come out
>with Craven's filter enhancement for DVD-A.

Actually, Craven's stuff isn't at all new. Wadia and Sony have been
using similar algorithms for more than a decade, so perhaps you should
simply buy a European market Sony players (Wadia headed them off at
the pass in the US with patent suits), and enjoy the best of both
worlds.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

Archived from groups: rec.audio.high-end (More info?)

On Sun, 11 Jul 2004 18:19:04 GMT, B&D <bromo@ix.netcom.com> wrote:

>On 7/11/04 1:32 AM, in article Ch4Ic.66683$Oq2.2493@attbi_s52, "Stewart
>Pinkerton" <patent3@dircon.co.uk> wrote:
>
>>>> Harry Lavo
>>>> "It don't mean a thing if it aint got that swing" - Duke Ellington
>>> So SACD does have a technical advantage over DVD-A.
>>
>> No, DVD-A at 24/192 will always be *vastly* superior to DSD. Filtering
>> is an entirely separate issue.
>
>If we as a group seem to be leaning towards CD as a "perfect" sound - or the
>faults are inaudible somehow - why would either format matter?

A perfectly fair point, since not one single person has yet offered
*proof* that, given no other differences, SACD or 24/192 DVD-A
actually *sound* different from 16/44 'Red Book' CD made from
*exactly* the same master.

--

Stewart Pinkerton | Music is Art - Audio is Engineering

Archived from groups: rec.audio.high-end (More info?)

"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
news:Ch4Ic.66683$Oq2.2493@attbi_s52...
> On 10 Jul 2004 14:40:18 GMT, Codifus <codifus@optonline.net> wrote:
>
> >Harry Lavo wrote:
> >> The current issue of Hi-Fi News (July 2004, pp 80-82) contains a
summary
> >> article covering work done by Peter Craven and published in the Journal
of
> >> the AES. Citation follows:
> >>
> >> P. Craven, "Anti Alias Filters and System Transient Response at High
Sample
> >> Rates", Journal of the Audio Engineering Society, March 2004.
> >>
> >> Mr. Craven was hired by Meriden apparently to work on improving
filtering
> >> for DVD-A at 96khz and 192khz. In his investigations, he found SACD's
> >> asymmetrical analog filter to have a -100db to -100db rise and fall
time of
> >> 35 micro-seconds (almost all post-transient with no pre-smear) to a
> >> transient pulse maxing at 0 db reference. The convention linear phase
> >> digital filter used in most DVD-A players yielded a symmetrical -100db
> >> to -100db rise and fall time of over 4 milliseconds, even when
filtering
> >> 192khz. He designed digital, asymmetric filters that managed to bring
the
> >> rise-and-fall time down to 300 microseconds for 96khz and 100
microseconds
> >> for 192khz. The author of the summary article concluded that these
latter
> >> at least had the potential of putting DVD-A "in the range of SACD as
regards
> >> transient capability" (paraphrased quote, since I don't have the
article in
> >> front of me).
> >>
> >> Interesting stuff that I haven't seen discussed before, at least in the
> >> popular/audiophile press or here on RAHE or rec.audio.tech. There may
be
> >> something to SACD aficionados perception of greater macro- and micro-
> >> dynamics after all, if there has been this much difference between SACD
(no
> >> pre-smear, short post-smear) and traditional digital linear filtering
> >> (substantial pre- and post- smear).
> >>
> >> Craven makes the point that the filter needs to be inserted just once
in the
> >> chain (ad,da) to gain the benefit. Sounds like may be in part a key as
to
> >> Meriden's DVD-A success.
> >>
> >> Harry Lavo
> >> "It don't mean a thing if it aint got that swing" - Duke Ellington
> >So SACD does have a technical advantage over DVD-A.
>
> No, DVD-A at 24/192 will always be *vastly* superior to DSD. Filtering
> is an entirely separate issue.
>

Still a matter of opinion. Yours, of course.

> >I've always been a
> >fan of DVD-A . . . .and still am. Even if it isn't perfect at the
> >beginning, it definitely will mature as a medium, like CD did. SACD, on
> >the other hand, despite being able to handle transients better, was
> >technically inferior to even CD in at least 1 aspect. Seeing that it
> >failed to oust the inferior Redbook CD on all technical merits, and even
> >though I most probably would not hear that difference in its failing, I
> >dismissed SACD. It's just the principle of the thing. If you're going
> >to replace CD, then you'd better be better than it in everything. DVD-A
> >does that from the beginning, and it will just get better as it matures.
> >
> >I was about to jump into buying a high-end audio unit, but after seeing
> >this, I'll keep enjoying my Redbooks and wait to see players come out
> >with Craven's filter enhancement for DVD-A.
>

Why don't you go audition a Meriden...their the one's who hired the
consultant? Perhaps its already been implemented in their new G series.

> Actually, Craven's stuff isn't at all new. Wadia and Sony have been
> using similar algorithms for more than a decade, so perhaps you should
> simply buy a European market Sony players (Wadia headed them off at
> the pass in the US with patent suits), and enjoy the best of both
> worlds.

I actually won a bet. Thanks Stewart. You are getting very predictable in
your pronouncements.

Archived from groups: rec.audio.high-end (More info?)

"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
news:TwzIc.61985$MB3.51394@attbi_s04...
> On Mon, 12 Jul 2004 02:22:46 GMT, "Harry Lavo" <harry.lavo@rcn.com>
> wrote:
>
> >"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
> >news:Ch4Ic.66683$Oq2.2493@attbi_s52...
> >> On 10 Jul 2004 14:40:18 GMT, Codifus <codifus@optonline.net> wrote:
>
> >> >So SACD does have a technical advantage over DVD-A.
> >>
> >> No, DVD-A at 24/192 will always be *vastly* superior to DSD. Filtering
> >> is an entirely separate issue.
> >>
> >Still a matter of opinion. Yours, of course.
>
> No, the *fact* that 24/192 DVD-A is technically superior to SACD, is
> certainly not a matter of opinion.

Yes it is...there is no engineering consensus on this. Their are other
parameters than "noise" to be considered...including slew rate and transient
response.

> >
> >> >I've always been a
> >> >fan of DVD-A . . . .and still am. Even if it isn't perfect at the
> >> >beginning, it definitely will mature as a medium, like CD did. SACD,
on
> >> >the other hand, despite being able to handle transients better, was
> >> >technically inferior to even CD in at least 1 aspect. Seeing that it
> >> >failed to oust the inferior Redbook CD on all technical merits, and
even
> >> >though I most probably would not hear that difference in its failing,
I
> >> >dismissed SACD. It's just the principle of the thing. If you're going
> >> >to replace CD, then you'd better be better than it in everything.
DVD-A
> >> >does that from the beginning, and it will just get better as it
matures.
> >> >
> >> >I was about to jump into buying a high-end audio unit, but after
seeing
> >> >this, I'll keep enjoying my Redbooks and wait to see players come out
> >> >with Craven's filter enhancement for DVD-A.
> >>
> >Why don't you go audition a Meriden...their the one's who hired the
> >consultant? Perhaps its already been implemented in their new G series.
>
> That's Meridian - but good advice.

Thanks for the correction. Too tired or lazy to look it up.

> >
> >> Actually, Craven's stuff isn't at all new. Wadia and Sony have been
> >> using similar algorithms for more than a decade, so perhaps you should
> >> simply buy a European market Sony players (Wadia headed them off at
> >> the pass in the US with patent suits), and enjoy the best of both
> >> worlds.
> >
> >I actually won a bet. Thanks Stewart. You are getting very predictable
in
> >your pronouncements.
>
> That's because the facts don't change, Harry.

First, similar is not identical. Second, you've spent a goodly amount of
time pooh-poohing the Wadia approach...but now you tell me Sony (which you
seem to admire) has essentially copied them in the rest of the world, not
doing so only in the U.S. because of patent issues. Seems to me you've just
disproved your claim a) that all CD players competently designed sound the
same, and b) that the CD standard as typically executed over the last decade
cannot/is not bettered by SACD or DVD-A. I believe it is called being
hoisted on your own petard?