Sign in with
Sign up | Sign in
Your question

Recommendation for sound card for recording quality.

Last response: in Home Audio
Share
Anonymous
April 28, 2005 8:46:56 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Hi,

I have a collection of analogue audio cassettes. I have a Hifi
"Nakamichi cassette deck 2". I want to go through the process of
turning all the cassettes into mp3 files. I have creative I-teague
speakers. I currently have a creative sound blaster gold sound card.
When I plug the I-teague speakers straight into the Nakamichi cassette
deck it gives better sound quality than exsiting mp3s I have playing
from the Sound blaster gold sound card (in comparison it sounds dull
and you're less able to pick out individual instruments).

Please could you recommend me a sound card that will give
significantly better quality than a Soundblaster gold card that might
be able to match the sound quality of the Nakamichi cassette deck? I
obviously don't need any fancy bells and whisles just good sound
quality. Would I be better buying an older card from ebay that is
reknowned for good audio quality that may be a bit cheaper than a
modern card with lots of facilities?
Anonymous
April 28, 2005 3:10:11 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

William,

> Please could you recommend me a sound card that will give significantly
better quality than a Soundblaster gold card that might be able to match the
sound quality of the Nakamichi cassette deck? <

I agree with Bob that a $25 SB Live card is head and shoulders above any
cassette deck. Even an MP3 file will beat the pants off any cassette if you
use a reasonably high bit rate.

--Ethan
Anonymous
April 28, 2005 4:46:51 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

On 28 Apr 2005 04:46:56 -0700, wrreisen@yahoo.com (William.R.Reisen)
wrote:

>Hi,
>
>I have a collection of analogue audio cassettes. I have a Hifi
>"Nakamichi cassette deck 2". I want to go through the process of
>turning all the cassettes into mp3 files. I have creative I-teague
>speakers. I currently have a creative sound blaster gold sound card.
>When I plug the I-teague speakers straight into the Nakamichi cassette
>deck it gives better sound quality than exsiting mp3s I have playing
>from the Sound blaster gold sound card (in comparison it sounds dull
>and you're less able to pick out individual instruments).
>
>Please could you recommend me a sound card that will give
>significantly better quality than a Soundblaster gold card that might
>be able to match the sound quality of the Nakamichi cassette deck? I
>obviously don't need any fancy bells and whisles just good sound
>quality. Would I be better buying an older card from ebay that is
>reknowned for good audio quality that may be a bit cheaper than a
>modern card with lots of facilities?

Just about any 16-bit sound card should be much better than
analog tape: Lower noise, better frequency response, no
print-through, etc. The tape will definitely be the limiting
factor here, even with the best of tape decks. (And the
Nakamichi was one of the better ones, as I recall.)

Do a test with your present sound card, and record as
a normal .WAV instead of MP3. Sample at 44100 Hz.
Make sure the recording level is set high enough; you
want it as high as possible, but you must make certain
that it doesn't clip. (Digital clipping is far, FAR worse than
tape saturation!)

I think you will find that this produces excellent results.
Then you can experiment with MP3 bandwidth to see
what level of compression you can tolerate.

Best regards,






Bob Masta
dqatechATdaqartaDOTcom

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Home of DaqGen, the FREEWARE signal generator
Related resources
Anonymous
April 28, 2005 4:46:52 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"Bob Masta" <NoSpam@daqarta.com> wrote:

> Just about any 16-bit sound card should be much better than
> analog tape: Lower noise, better frequency response, no
> print-through, etc. The tape will definitely be the limiting
> factor here, even with the best of tape decks. (And the
> Nakamichi was one of the better ones, as I recall.)
>
> Do a test with your present sound card, and record as
> a normal .WAV instead of MP3. Sample at 44100 Hz.
> Make sure the recording level is set high enough; you
> want it as high as possible, but you must make certain
> that it doesn't clip. (Digital clipping is far, FAR worse than
> tape saturation!)
>
> I think you will find that this produces excellent results.
> Then you can experiment with MP3 bandwidth to see
> what level of compression you can tolerate.

Bob is right, don't jump to conclusions, but personally I hate sound blaster
cards. I think you'd you yourself a big favor by getting a cheap pro card
like the M Audio, or Echo MIA MIDI or one of the Edirol products. Then do
what Bob says and you should be able to make some very nice MP3's from your
Nakamichi.

Julian
Anonymous
April 28, 2005 4:50:20 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

William.R.Reisen wrote:
> When I plug the I-teague speakers straight into the Nakamichi cassette
> deck it gives better sound quality than exsiting mp3s I have playing
> from the Sound blaster gold sound card

Are you sure that isn't just becuase they're MP3's?

> Please could you recommend me a sound card that will give
> significantly better quality than a Soundblaster

M-Audio Audiophile 2496
Anonymous
April 28, 2005 4:56:37 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

On Thu, 28 Apr 2005 08:12:08 -0700, "Julian Adamaitis"
<nospamJulianPA@Access4Less.net> wrote:

I think you'd you yourself a big favor by getting a cheap pro card
>like the M Audio, or Echo MIA MIDI or one of the Edirol products. Then do
>what Bob says and you should be able to make some very nice MP3's from your
>Nakamichi.

Best advice so far...

Al
Anonymous
April 28, 2005 11:21:14 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

In article <c254b7f4.0504280346.50afae08@posting.google.com> wrreisen@yahoo.com writes:

> I have a collection of analogue audio cassettes. I have a Hifi
> "Nakamichi cassette deck 2". I want to go through the process of
> turning all the cassettes into mp3 files. I have creative I-teague
> speakers. I currently have a creative sound blaster gold sound card.
> When I plug the I-teague speakers straight into the Nakamichi cassette
> deck it gives better sound quality than exsiting mp3s I have playing
> from the Sound blaster gold sound card (in comparison it sounds dull
> and you're less able to pick out individual instruments).

I think that a modern Soundblaster card won't cause you to lose
anything coming off the cassette. However, MP3s can easily sound not
as good as a good cassette. If you want to maintain the cassette
quality, record PCM format, uncompressed. Then you can experiment with
different encoder programs and different amounts of compression until
you find something that doesn't sound worse than the cassette.

The most important thing you can do when transferring your cassettes
is to adjust the azimuth alignment for each tape, at least check it in
a few places and tweak it for the best compromise if you don't want to
transfer the tapes in sections. The Nakamichi is pretty good, but you
don't know what the tapes were recorded on, and the azimuth (the
perpendicularity of the head relative to the tape) tends to wander
quite a bit with cassettes.


--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
Anonymous
April 29, 2005 3:50:47 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

> Please could you recommend me a sound card that will give significantly
>better quality than a Soundblaster gold card that might be able to match the
>sound quality of the Nakamichi cassette deck?

Ethan Winer <ethanw at ethanwiner dot com> wrote:
>I agree with Bob that a $25 SB Live card is head and shoulders above any
>cassette deck. Even an MP3 file will beat the pants off any cassette if you
>use a reasonably high bit rate.

I don't think so. His Nakamichi cassette deck isn't just any cassette
deck, and the original SoundBlaster Live! cards have terrible resampling.
He's going to want at least a Audigy to do the recording.

Ross Ridge

--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rridge@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca/u/rridge/
db //
Anonymous
April 29, 2005 6:17:57 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"William.R.Reisen" wrote:

> Hi,
>
> I have a collection of analogue audio cassettes. I have a Hifi
> "Nakamichi cassette deck 2". I want to go through the process of
> turning all the cassettes into mp3 files. I have creative I-teague
> speakers. I currently have a creative sound blaster gold sound card.
> When I plug the I-teague speakers straight into the Nakamichi cassette
> deck it gives better sound quality than exsiting mp3s I have playing
> from the Sound blaster gold sound card (in comparison it sounds dull
> and you're less able to pick out individual instruments).
>
> Please could you recommend me a sound card that will give
> significantly better quality than a Soundblaster gold card that might
> be able to match the sound quality of the Nakamichi cassette deck? I
> obviously don't need any fancy bells and whisles just good sound
> quality. Would I be better buying an older card from ebay that is
> reknowned for good audio quality that may be a bit cheaper than a
> modern card with lots of facilities?

The better Soundblaster cards are actually not that bad.

Nakamichi cassette decks were also rather good. Why would you want to
trash the potential quality of your cassette recordings by saving them as
mp3s?

I'd recommend ( subject to disk space ) that you try saving them as wav
files. Use uncompressed linear PCM format if that means anything to you.

The coments you make about mp3 quality ring *so* true to me.

/ quote

in comparison it sounds dull and you're less able to pick out individual
instruments

/ endquote.

This is what mp3 does and how it makes the files smaller. It's one of the
'lossy compression' style formats that degrade detail. I've experienced
exactly the same - even when making the comparison on the *car stereo* !

Try again and save as wav files. I bet that'll fix it.

Btw - you won't find any good old sound cards on ebay unless you also tend
to win the lottery regularly.

Graham
Anonymous
April 29, 2005 6:33:16 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Bob Masta wrote:

> On 28 Apr 2005 04:46:56 -0700, wrreisen@yahoo.com (William.R.Reisen)
> wrote:
>
> >Hi,
> >
> >I have a collection of analogue audio cassettes. I have a Hifi
> >"Nakamichi cassette deck 2". I want to go through the process of
> >turning all the cassettes into mp3 files. I have creative I-teague
> >speakers. I currently have a creative sound blaster gold sound card.
> >When I plug the I-teague speakers straight into the Nakamichi cassette
> >deck it gives better sound quality than exsiting mp3s I have playing
> >from the Sound blaster gold sound card (in comparison it sounds dull
> >and you're less able to pick out individual instruments).
> >
> >Please could you recommend me a sound card that will give
> >significantly better quality than a Soundblaster gold card that might
> >be able to match the sound quality of the Nakamichi cassette deck? I
> >obviously don't need any fancy bells and whisles just good sound
> >quality. Would I be better buying an older card from ebay that is
> >reknowned for good audio quality that may be a bit cheaper than a
> >modern card with lots of facilities?
>
> Just about any 16-bit sound card should be much better than
> analog tape: Lower noise, better frequency response, no
> print-through, etc. The tape will definitely be the limiting
> factor here, even with the best of tape decks. (And the
> Nakamichi was one of the better ones, as I recall.)

I can assure you that's not true for motherboards with 'integrated sound'
that are often sold advertised as if they had the actual original sound
card installed.


> Do a test with your present sound card, and record as
> a normal .WAV instead of MP3. Sample at 44100 Hz.
> Make sure the recording level is set high enough; you
> want it as high as possible, but you must make certain
> that it doesn't clip. (Digital clipping is far, FAR worse than
> tape saturation!)
>
> I think you will find that this produces excellent results.
> Then you can experiment with MP3 bandwidth to see
> what level of compression you can tolerate.

I've listened to mp3 @ 320 kbits/sec and it's still obviously inferior to
the raw uncompressed file. The effect varies with the complxity of the
music btw. A simple guitar / soloist song may sound quite good while
complex passages can be trashed.

I was quite startled after purchasing a certain CD after 'previewing' it
courtesy of an mp3 file to discover that I could hear detail on the CD that
had simply 'vanished' on the mp3 - to the extent of masking low level
backing vocals even !


Graham
Anonymous
April 29, 2005 6:39:20 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Ethan Winer wrote:

> William,
>
> > Please could you recommend me a sound card that will give significantly
> better quality than a Soundblaster gold card that might be able to match the
> sound quality of the Nakamichi cassette deck? <
>
> I agree with Bob that a $25 SB Live card is head and shoulders above any
> cassette deck. Even an MP3 file will beat the pants off any cassette if you
> use a reasonably high bit rate.

You obviously never owned a Nakamichi ! They are / were the real 'Rolls Royce'
of the breed. Including such features as auto line-up ! You know what a line-up
is ?

About the only cassette deck ever to approach the quality of 1/4" tape. Damn
fine machines. I used a Nakamichi portable as my reference source for setting up
sound rigs many many yrs ago.


Graham
Anonymous
April 29, 2005 6:44:05 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"William.R.Reisen" wrote:

> Hi,
>
> I have a collection of analogue audio cassettes. I have a Hifi
> "Nakamichi cassette deck 2". I want to go through the process of
> turning all the cassettes into mp3 files. I have creative I-teague
> speakers. I currently have a creative sound blaster gold sound card.
> When I plug the I-teague speakers straight into the Nakamichi cassette
> deck it gives better sound quality than exsiting mp3s I have playing
> from the Sound blaster gold sound card (in comparison it sounds dull
> and you're less able to pick out individual instruments).
>
> Please could you recommend me a sound card that will give
> significantly better quality than a Soundblaster gold card that might
> be able to match the sound quality of the Nakamichi cassette deck? I
> obviously don't need any fancy bells and whisles just good sound
> quality. Would I be better buying an older card from ebay that is
> reknowned for good audio quality that may be a bit cheaper than a
> modern card with lots of facilities?

You may find this useful when comparing sound cards.

http://www.pcavtech.com/soundcards/compare/index.htm

It rates a SB 64 Gold as 'acceptable' for many. I'll bet the mp3
compression is at least if not more culpable of trashing the detail.


Graham
Anonymous
April 29, 2005 2:45:04 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Ross and Graham,

> His Nakamichi cassette deck isn't just any cassette deck, and the original
SoundBlaster Live! cards have terrible resampling. <

> They are / were the real 'Rolls Royce' of the breed. Including such
features as auto line-up ! You know what a line-up is ? <

I've been around a while, and I know all about Nakamichi cassette decks. And
Yes, I know what "line-up" means. :->)

I agree those are/were great cassette recorders, but that's still not saying
much. Even the best cassette deck is inferior to an SB Live card in any way
you care to measure:

Dynamic range
THD and IM distortion
Frequency response
Wow and flutter

If you still think this is wrong, please list the *specific audio
attributes* where even the best cassette deck could possibly be superior to
an SB Live card. I don't mean "resampling" because that doesn't say anything
specific. And it's expressed better as distortion or noise, or whatever
other standard spec you believe is harmed by "terrible resampling."

BTW, I'm not trying to be argumentative here. I just want to keep things
focused on fact instead of subjective opinion.

--Ethan
Anonymous
April 29, 2005 5:53:43 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

> Please could you recommend me a sound card that will give
>significantly better quality than a Soundblaster gold card that might
>be able to match the sound quality of the Nakamichi cassette deck?

What you are experiencing is the slurring of a cheap analogue audio
curcuit on the sound card.
it is designed for fitting into the computer
if you want a better sound you need to
remove all audio curcuits from the computer
exspecially the analogue sections.

the nak has much better audio then most any sound card
it's designed for its musicality.
most sound cards are designed for test specs that confuse
the issue of aesthetics (musicality)
with digital facts.

check out m audio ,
firewire protocol is desirable

dale
Anonymous
April 29, 2005 7:20:50 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

In comp.sys.ibm.pc.soundcard.tech Anahata <anahata@treewind.co.uk> wrote:
> William.R.Reisen wrote:
>> When I plug the I-teague speakers straight into the Nakamichi cassette
>> deck it gives better sound quality than exsiting mp3s I have playing
>> from the Sound blaster gold sound card
>
> Are you sure that isn't just becuase they're MP3's?
>
>> Please could you recommend me a sound card that will give
>> significantly better quality than a Soundblaster
>
> M-Audio Audiophile 2496

So my question is: how much better is the 2-channel recording ability
of this card vs. the Revolution 7.1, or the Audigy 2 ZS? I'm about to get
a new card, and I want surround capabilities, but I'd like to make sure that
the recording quality doesn't suffer (much).

For that matter, how does this compare to my venerable but unsupported
Diamond Monster MX-300?

Thanks,
Colin
Anonymous
April 29, 2005 7:34:09 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

On Fri, 29 Apr 2005 02:17:57 +0100, Pooh Bear
<rabbitsfriendsandrelations@hotmail.com> wrote:

>
> Btw - you won't find any good old sound cards on ebay unless you also
> tend
> to win the lottery regularly.
>
> Graham
>
>
>

I've noticed that there are a few M-Audio Audiophile cards on Ebay at the
moment - around 60 quid new or 30 quid used. It looks like they're about
to replace them with a new model so now is a good time to buy the older
version.

Cheers.

James.
Anonymous
April 29, 2005 10:14:01 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

In article <427250aa@news.nucleus.com> cbigam@somewhereelse.nucleus.com writes:

> So my question is: how much better is the 2-channel recording ability
> of this card vs. the Revolution 7.1, or the Audigy 2 ZS? I'm about to get
> a new card, and I want surround capabilities, but I'd like to make sure that
> the recording quality doesn't suffer (much).

You really can't make "how much better" comparisons without defining
what "better" actually means. Any of those cards will work for your
purposes. Make a decision and get to work.


--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
Anonymous
April 30, 2005 12:19:28 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Ethan Winer <ethanw at ethanwiner dot com> wrote:
>I agree those are/were great cassette recorders, but that's still not saying
>much. Even the best cassette deck is inferior to an SB Live card in any way
>you care to measure:
>
> Dynamic range
> THD and IM distortion
> Frequency response
> Wow and flutter

You haven't measured any of these values for either piece of equipment.

>If you still think this is wrong, please list the *specific audio
>attributes* where even the best cassette deck could possibly be superior to
>an SB Live card. I don't mean "resampling" because that doesn't say anything
>specific.

Resampling error is quite specific. The original SoundBlaster Live! cards
did a terrible job of this, and it's quite audible. It's the reason
why the original poster think his MP3's sound so bad, worse than his
tape deck.

>BTW, I'm not trying to be argumentative here. I just want to keep things
>focused on fact instead of subjective opinion.

You haven't posted any facts. You don't even have any actual, real,
obseverations to form a rational "subjective opinion". You're just
speculating based on assumptions. The fact that I posted however is
both based on reality and it's one that's easily verifiable.

Ross Ridge

--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rridge@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca/u/rridge/
db //
Anonymous
April 30, 2005 12:48:20 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

In article <d4u4sg$bgn$1@rumours.uwaterloo.ca> rridge@csclub.uwaterloo.ca writes:

> You haven't posted any facts. You don't even have any actual, real,
> obseverations to form a rational "subjective opinion". You're just
> speculating based on assumptions. The fact that I posted however is
> both based on reality and it's one that's easily verifiable.

So what? Why can't the guy just use what he has if it works for his
project? I could buy a better car than the one I have, but my present
car can get me to the grocery store just fine. I don't race every
week. In fact, I never race.

--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
Anonymous
April 30, 2005 7:26:44 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Mike Rivers <mrivers@d-and-d.com> wrote:
>So what? Why can't the guy just use what he has if it works for his
>project?

I never said that he couldn't. It's the original poster who doesn't
think what he already has will work for his project.

Ross Ridge

--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rridge@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca/u/rridge/
db //
Anonymous
April 30, 2005 10:17:50 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

> Why can't the guy just use what he has if it works for his project?

he does not like what he hears
he wants to improve his setup

dale
Anonymous
April 30, 2005 11:19:01 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"William.R.Reisen" wrote ...
> I have a collection of analogue audio cassettes. I have a Hifi
> "Nakamichi cassette deck 2". I want to go through the process of
> turning all the cassettes into mp3 files. I have creative I-teague
> speakers. I currently have a creative sound blaster gold sound card.
> When I plug the I-teague speakers straight into the Nakamichi cassette
> deck it gives better sound quality than exsiting mp3s I have playing
> from the Sound blaster gold sound card (in comparison it sounds dull
> and you're less able to pick out individual instruments).
>
> Please could you recommend me a sound card that will give
> significantly better quality than a Soundblaster gold card that might
> be able to match the sound quality of the Nakamichi cassette deck? I
> obviously don't need any fancy bells and whisles just good sound
> quality. Would I be better buying an older card from ebay that is
> reknowned for good audio quality that may be a bit cheaper than a
> modern card with lots of facilities?

You are assuming your sound card is the cause of the "dull" sound.

You have violated a basic principle of scientific experimentation.
You have changed more than one variable in a single step, so you
can't differentiate between the effect of each of the two (or more)
variables.

Do the experiment of recording to un-compressed WAV files and
playing back. Then at least you can do the "differential diagnosis"
to determine whether it is really the sound card (doubtful IMHO)
or the MP3 compression (much more likely).

SoundBlaster doesn't have a great reputation, but when you casually
mentioned "MP3", you drew suspicion away from the sound card.
You will need to reveal what software you are using to do the MP3
conversion if you want complete responses.
Anonymous
April 30, 2005 1:05:51 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

> I agree those are/were great cassette recorders, but that's still not
saying
> much. Even the best cassette deck is inferior to an SB Live card in any
way
> you care to measure:

and the SBLIVE is SHITE anyway
Anonymous
April 30, 2005 3:12:39 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

In article <d4uttk$v3j$1@rumours.uwaterloo.ca> rridge@csclub.uwaterloo.ca writes:

> >So what? Why can't the guy just use what he has if it works for his
> >project?
>
> I never said that he couldn't. It's the original poster who doesn't
> think what he already has will work for his project.

And I say that it probably will. He's just too lazy or lacking the
confidence in his own judgement to give it a try. If he had, he'd tell
us what he thought was wrong so we could better suggest a path for
improvement. Most people who get into this for the first time don't
even realize that they can adjust the record level (and need to do so).



--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
Anonymous
April 30, 2005 6:23:41 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

On 30 Apr 2005 06:17:50 -0700, "dale" <dallen@frognet.net> wrote:

>> Why can't the guy just use what he has if it works for his project?
>
>he does not like what he hears
>he wants to improve his setup
>
>dale
>
The OP stated that he did not like what he hears, but
he jumped to the conclusion that it was due to his
sound card. He first needs to rule out the strong
possibility that what he dislikes is the MP3 encoding.
Buying a different sound card won't help that!

Best regards,


Bob Masta
dqatechATdaqartaDOTcom

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Home of DaqGen, the FREEWARE signal generator
Anonymous
April 30, 2005 10:26:44 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

>So what? Why can't the guy just use what he has if it works for his
>project?

rridge@csclub.uwaterloo.ca writes:
> I never said that he couldn't. It's the original poster who doesn't
> think what he already has will work for his project.

Mike Rivers <mrivers@d-and-d.com> wrote:
>And I say that it probably will.

So why did you start arguing with me?

>He's just too lazy or lacking the confidence in his own judgement to
>give it a try.

There's no reason to believe that's true.

>If he had, he'd tell us what he thought was wrong so we could better
>suggest a path for improvement.

He's already told us what he thinks is wrong.

Please take the time to read posts in their entirety before responding.

Ross Ridge

--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rridge@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca/u/rridge/
db //
Anonymous
April 30, 2005 11:35:32 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

In article <d50il4$jpk$1@rumours.uwaterloo.ca> rridge@csclub.uwaterloo.ca writes:

> He's already told us what he thinks is wrong.
> Please take the time to read posts in their entirety before responding.

Would you care to regurgitate it to me? I'm too lazy and too
uninterested to go back to the server and try to look it up? As far as
I recall, he said he didn't care for how it sounds. That doesn't tell
me anything useful.

It's been suggested that the MP3 process may be causing whatever
problem he perceives. We haven't heard back from the original poster
yet that he tried linear recording with his present sound card (after
carefully setting the record level) to see if he still dislikes what
he hears.

No point in arguing among ourselves about what we think he thinks is
wrong when we know he has one big strike (MP3) against him and he's
looking to change his sound card. He may not need to do so.

Or he could just buy a LynxTWO, get the SoundBlaster out of the way,
and then start aking which MP3 encoder will give him results that he
thinks sounds good. But if he takes that approach, he'll never know
whether or not he actually had to spend that $1,000.



--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
Anonymous
April 30, 2005 11:35:33 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"Mike Rivers" <mrivers@d-and-d.com> wrote in message
news:znr1114887334k@trad...
>
> In article <d50il4$jpk$1@rumours.uwaterloo.ca>
> rridge@csclub.uwaterloo.ca writes:
>
>> He's already told us what he thinks is wrong.
>> Please take the time to read posts in their entirety before
>> responding.
>
> Would you care to regurgitate it to me? I'm too lazy and too
> uninterested to go back to the server and try to look it up? As far as
> I recall, he said he didn't care for how it sounds. That doesn't tell
> me anything useful.

The OP's "presenting symptoms" (as they say in the medical biz)...

"I have a collection of analogue audio cassettes. I have a Hifi
"Nakamichi cassette deck 2". I want to go through the process of
turning all the cassettes into mp3 files. I have creative I-teague
speakers. I currently have a creative sound blaster gold sound card.
When I plug the I-teague speakers straight into the Nakamichi cassette
deck it gives better sound quality than exsiting mp3s I have playing
from the Sound blaster gold sound card (in comparison it sounds dull
and you're less able to pick out individual instruments)."

> It's been suggested that the MP3 process may be causing whatever
> problem he perceives. We haven't heard back from the original poster
> yet that he tried linear recording with his present sound card (after
> carefully setting the record level) to see if he still dislikes what
> he hears.
>
> No point in arguing among ourselves about what we think he thinks is
> wrong when we know he has one big strike (MP3) against him and he's
> looking to change his sound card. He may not need to do so.
>
> Or he could just buy a LynxTWO, get the SoundBlaster out of the way,
> and then start aking which MP3 encoder will give him results that he
> thinks sounds good. But if he takes that approach, he'll never know
> whether or not he actually had to spend that $1,000.

Precisely.
Anonymous
May 1, 2005 1:03:43 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Bob Masta wrote:

> Just about any 16-bit sound card should be much better than
> analog tape: Lower noise, better frequency response, no
> print-through, etc. The tape will definitely be the limiting
> factor here, even with the best of tape decks. (And the
> Nakamichi was one of the better ones, as I recall.)

Hmm ... about the first I did when testing the Mdiman 2496 Audiophile
was to listen to the change as I changed sampling frequency. I did that
test with a tape that was recorded on the original Philips compact
casette recorder, albeit with a Sennheiser MD211 rather than the ex
works mic.

> Do a test with your present sound card, and record as
> a normal .WAV instead of MP3. Sample at 44100 Hz.

That would be plain foolish, especially btw. with a soundblaster that
resamples to and from 48 kHz and uses that internally no matter what, a
possibly incomplete rendition of various tales of soundblaster woes. 48
kHz is likely to sound notably better on the small number of converters
I am familiar with and I prefer to use native 32-bit recording format,
and then dither down after relevant repair work.

> Make sure the recording level is set high enough; you
> want it as high as possible, but you must make certain
> that it doesn't clip. (Digital clipping is far, FAR worse than
> tape saturation!)

Your statement contradicts my actual recording experience - up to some 4
dB of clean clipping of Chamber music recorded with high quality
equipment is easily fixable in the digital domain and completely
inaudible even if unfixed. Digital clipping is no better nor no worse
than any other clean clip.

> I think you will find that this produces excellent results.

If he wants to use ex works sound cards, then he should perhaps take a
look at the Echo range that Arny Kr├╝ger is so fond of.

> Then you can experiment with MP3 bandwidth to see
> what level of compression you can tolerate.

192 kbit/sec stereo, 96 kbit second mono, fixed bit rate is the format I
selected for my cassette drawer using my preferred encoder.

> Bob Masta


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************
Anonymous
May 1, 2005 1:03:44 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Bob Masta wrote:
> Do a test with your present sound card, and record as
> a normal .WAV instead of MP3. Sample at 44100 Hz.

Peter Larsen <SPAMSHIELD_plarsen@mail.tele.dk> wrote:
>That would be plain foolish, especially btw. with a soundblaster that
>resamples to and from 48 kHz and uses that internally no matter what, a
>possibly incomplete rendition of various tales of soundblaster woes.

The SoundBlaster Live! always resamples to 48 kHz even when the input
is already 48 KHz. It makes no difference if you use 44.1 kHz or 48
kHz with a original SoundBlaster Live!, you get the same bad resampling.

Ross Ridge

--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rridge@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca/u/rridge/
db //
Anonymous
May 1, 2005 1:13:12 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Ethan Winer wrote:

> I agree those are/were great cassette recorders, but that's still
> not saying much. Even the best cassette deck is inferior to an SB
> Live card in any way you care to measure:

> Dynamic range
> THD and IM distortion
> Frequency response
> Wow and flutter

> If you still think this is wrong, please list the *specific audio
> attributes* where even the best cassette deck could possibly be
> superior to an SB Live card.

The sound ... O;-)

> BTW, I'm not trying to be argumentative here. I just want
> to keep things focused on fact instead of subjective opinion.

I am a member of a tape recordists club here in Copenhagen. The only one
to prefer the sound of his extigy over the M-audio Delta 44 or any other
M-audio card, modded or ex works, is the owner of the extigy.

> --Ethan

I would disagree less on the relevance of a usable sound card if it was
not about someone with a Nak and if I had not made the sampling rate
test with that old recording made on an original portable Philips.


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************
Anonymous
May 1, 2005 1:15:52 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Bob Masta wrote:

> The OP stated that he did not like what he hears, but
> he jumped to the conclusion that it was due to his
> sound card. He first needs to rule out the strong
> possibility that what he dislikes is the MP3 encoding.
> Buying a different sound card won't help that!

On this I agree fully.

> Bob Masta


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************
Anonymous
May 1, 2005 2:08:48 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Ross,

> You haven't measured any of these values for either piece of equipment. <

I've measured - or seen specs for - tons of audio gear over the years. I
have a very good handle on what the specs mean, how they translate audibly,
and at what amounts they become objectionable.

> Resampling error is quite specific. <

But "resampling error" is not a spec! It's an opinion. Either you can quote
which *already established audio spec* is affected, and by how much, or you
cannot.

> It's the reason why the original poster think his MP3's sound so bad,
worse than his tape deck. <

How can you claim to know what the OP heard, or his level of understanding
on audio matters? I think Mike nailed it when he said, "He's just ...
lacking the confidence in his own judgement." This is a huge problem in
these groups. Every day I see newbies post that they're not happy with their
home-made recordings, and ask us what tube compressor or outboard word clock
they should buy to make their lame efforts sound like a commercial CD. They
think the problem is missing some key piece of gear, all the while
overlooking that their gear is fine and it's their skill that's lacking.

> You haven't posted any facts. <

I did! I explained that "Even the best cassette deck is inferior to an SB
Live card in any way you care to measure," then listed the four audio specs:
Dynamic Range, Distortion, Frequency Response, and Wow & Flutter. So in
which of these specs is a Nakamichi cassette recorder better than an SB Live
card? Or have you discovered a new audio parameter that nobody else yet
knows about?

I repeat again that my point is not to argue or flame, but rather to keep
things focused on what really matters and what is known by science.

--Ethan
Anonymous
May 1, 2005 2:18:06 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Peter,

> The sound ... O;-) <

Of course, that explains nothing.

I am not saying an SB Live is equal to the best sound card ever made. But
it's not nearly as bad as a lot of people seem to think. In fact, it beats
even the best cassette recorder in any way you care to measure. :->)

Also, the OP is using speakers that cost $99 including two satellites and a
powered sub, no doubt in an untreated room. And you're fretting over the
difference between 0.01% and 0.001% distortion?

--Ethan
Anonymous
May 1, 2005 6:13:52 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Hi William,

> I have a collection of analogue audio cassettes.
> I have a Hifi "Nakamichi cassette deck 2".
> I want to go through the process of turning all the
> cassettes into mp3 files. [...]
> Please could you recommend me a sound card that will give
> significantly better quality than a Soundblaster gold card
> that might be able to match the sound quality of the
> Nakamichi cassette deck?

the secret advice for that task is to pick up
a used Philips DCC tape deck from Ebay or the like.
They are no longer manufactured and go away for small
change since the DCC tapes are no longer available.

The tape decks themselves, however, are of good quality
and for backward compatibilty they can play analog tapes
as well. Plus, they have a digital out!

So with any soundcard that has a digital input you can
easily transfer your old analog tapes to digital audio
of whatever format you prefer. No hassling with hum or
the proper recording level. Very easy. I picked up a
Philips DCC 450 (compact version) for about $25.

All the best,

Dieter Michel
Anonymous
May 1, 2005 6:16:02 PM

Archived from groups: rec.audio.pro (More info?)

In article <aKKdnZWHgc2Te-nfRVn-vw@giganews.com> "Ethan Winer" <ethanw at ethanwiner dot com> writes:

> Also, the OP is using speakers that cost $99 including two satellites and a
> powered sub, no doubt in an untreated room. And you're fretting over the
> difference between 0.01% and 0.001% distortion?

I didn't say anything about that, but I indeed noticed it. It's no
particular credit to his speakers that he can tell the difference
between an MP3 (of unspecified data rate, sample rate, and encoder)
and direct playback from a decent cassette deck. "Dull" was the only
real descriptive word he used, and that's more likely a function of
sample rate and encoding than the A/D converters and internal sample
rate conversion of the SoundBlaster.

So, we're still just pissing in the wind. But if it would make him
feel better to get another sound card, it might as well be an M-Audio
Audiophile or Echo Mia. At least they're generally little or no
trouble to install and don't have the SoundBlaster's inherent
resampling that seems to scare people around here.

I've never been so unlucky as to have a SoundBlaster except for an
early one which I never used for anything but computer audio, so I
really don't know just how bad a modern one can be at its worst, but I
still support your contention that it's as good as the cassette. I'll
bet that if the original poster made a PCM recording at the proper
record level, he'd find that it sounded essentially the same as the
cassette source. If it didn't, it would be time to troubleshoot
settings to be sure he wasn't recording at an 11 kHz 8-bit default or
something like that.

--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
Anonymous
May 1, 2005 10:35:50 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

> You haven't measured any of these values for either piece of equipment.

Ethan Winer <ethanw at ethanwiner dot com> wrote:
>I've measured - or seen specs for - tons of audio gear over the years. I
>have a very good handle on what the specs mean, how they translate audibly,
>and at what amounts they become objectionable.

Irrelevent.

>> Resampling error is quite specific. <
>
>But "resampling error" is not a spec! It's an opinion.

Just because you don't know what resampling error is doesn't make it
an opinion.

>> You haven't posted any facts.
>
>I did!

No, you haven't posted any facts.

> I explained that "Even the best cassette deck is inferior to an SB
>Live card in any way you care to measure,"

That's just your opinion. An opinion which you have no measurements
nor any obsevervations to substatiate.

>... then listed the four audio specs: Dynamic Range, Distortion,
>Frequency Response, and Wow & Flutter. So in which of these specs is
>a Nakamichi cassette recorder better than an SB Live card?

In which of these "specs" can you provide any evidence, any facts
whatsoever, that an SoundBlaster Live! is better than the orignal poster's
Nakamichi cassette deck?

>I repeat again that my point is not to argue or flame, but rather to keep
>things focused on what really matters and what is known by science.

Your speculation is not science.

Ross Ridge

--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rridge@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca/u/rridge/
db //
Anonymous
May 2, 2005 3:25:20 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"Ethan Winer" <ethanw at ethanwiner dot com> wrote

> But "resampling error" is not a spec! It's an opinion. Either you can
> quote
> which *already established audio spec* is affected, and by how much, or
> you
> cannot.

There is indeed an "error", or at least a "change" caused by resampling
between 44.1 and 48. If it is audible and if so to what extent is it
significant is an entirely different question.

I just took a short 44.1 audio file and made an inverted copy. When I sum
the 2 together, there is absolutely no resulting sound because they are bit
accurate, only inverted. If I sum my original to one that has been
converted to 48 and then inverted, there is a significant resulting signal,
which proves something is most definitely changed. A Sound Blaster card
that resamples everything to 48 definitely changes the sound compared to a
bit accurate card that comes in at 44.1 and stays there. You cannot state
this is an opinion. It is a fact. How significant it is, is indeed an
opinion until someone devises a test to evaluate.

Julian
Anonymous
May 2, 2005 3:27:36 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"Dieter Michel" <dmichel@prosound.de> wrote

> the secret advice for that task is to pick up
> a used Philips DCC tape deck from Ebay or the like.
> They are no longer manufactured and go away for small
> change since the DCC tapes are no longer available.
>
> The tape decks themselves, however, are of good quality
> and for backward compatibilty they can play analog tapes
> as well. Plus, they have a digital out!
>
> So with any soundcard that has a digital input you can
> easily transfer your old analog tapes to digital audio
> of whatever format you prefer. No hassling with hum or
> the proper recording level. Very easy. I picked up a
> Philips DCC 450 (compact version) for about $25.

But then you are dependent on the cassette tape quality of the $25 recorder
which probably isn't as good as the nakamichi in question.

Julian
Anonymous
May 2, 2005 5:00:33 AM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

In comp.sys.ibm.pc.soundcard.tech Mike Rivers <mrivers@d-and-d.com> wrote:
>
> In article <427250aa@news.nucleus.com> cbigam@somewhereelse.nucleus.com writes:
>
>> So my question is: how much better is the 2-channel recording ability
>> of this card vs. the Revolution 7.1, or the Audigy 2 ZS? I'm about to get
>> a new card, and I want surround capabilities, but I'd like to make sure that
>> the recording quality doesn't suffer (much).
>
> You really can't make "how much better" comparisons without defining
> what "better" actually means. Any of those cards will work for your
> purposes. Make a decision and get to work.

I would have thought that 'better' was self-evident, given the purpose for
which I was putting the card.

Better meaning lower distortion figures?
Better meaning flatter frequency response?
Better meaning fewer digitisation artifacts?
Better meaning NOT resampling to a different frequency?

Or maybe NOT better, meaning that any of the cards I mentioned will
do all of these things comparably, within human audibility.

Put it this way: For straiht two-channel recording to digital
(16bit/44.1kHz), minimal processing if any, and burning to CD-R, will any
of the cards cause audible artifacts? If so, will any of them NOT produce
audible artifacts? If so, which problems are the lowest level, based on
listening criteria?

Colin
Anonymous
May 2, 2005 1:27:06 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

In article <42757b88@news.nucleus.com> cbigam@somewhereelse.nucleus.com writes:

> I would have thought that 'better' was self-evident, given the purpose for
> which I was putting the card.

As I understand it (I hope I'm taking about the right thread here),
you're making MP3 versions of a cassette tape. In my book, "better" is
irrelevant because you'll never hear better. The cassette and the MP3
conversion are "worse" than all but the worst sound cards.

> Better meaning lower distortion figures?
> Better meaning flatter frequency response?
> Better meaning fewer digitisation artifacts?
> Better meaning NOT resampling to a different frequency?

I believe that all of those will be overshadowed by what you're doing
otherwise. The only one that I can't attest to is the resampling
issues which so many people seem to be bothered by. I really don't
know how bad this is, but I do know that it's very easy to tell the
difference between even a mediocre sound card's direct PCM recording
and a 128 kbps MP3 with most of the music that I choose to devistate
with MP3 conversion.

> Or maybe NOT better, meaning that any of the cards I mentioned will
> do all of these things comparably, within human audibility.

That's pretty much what I've been saying. I'm sure that when comparing
PCM recordings of good sources, many humans will be able to rank some
cards as being "better" than the others, but since you have a source
of questionable quality (you probably don't have anything above 10
kHz, and with less than 1% THD, so why worry about tenths of a dB of
non-flatness or a few hundreths of a percent distortion?) and you're
further buggering your recording with data reduction, the quality of
the sound card is pretty much a wash.

> Put it this way: For straiht two-channel recording to digital
> (16bit/44.1kHz), minimal processing if any, and burning to CD-R, will any
> of the cards cause audible artifacts? If so, will any of them NOT produce
> audible artifacts? If so, which problems are the lowest level, based on
> listening criteria?

Everything produces artifacts. I believe that FOR YOUR PURPOSES the
differences, while perhaps distinguishable, will be insignificant. If
you were talking about forensic or archival (that is, putting it away
for history, not just listening to it next month) work, then I would
recommend nothing but top shelf products.

Take this as you wish. You can pay more money and get better accuracy.
It's your decision, though, as to where to make that trade-off. If
spending $100 on a new card will make you feel better about the whole
thing, then by all means do so. But I wouldn't advice you to spend
$1000 for what probably won't make a tenfold improvement in sound
quality of your final product.



--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
Anonymous
May 2, 2005 4:32:56 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Ross,

> In which of these "specs" can you provide any evidence, any facts
whatsoever, that an SoundBlaster Live! is better than the orignal poster's
Nakamichi cassette deck? <

Which specs? Okay, for starters, the signal to noise ratio is about 20 dB
better on an SB Live card. And THD and IM distortion are much better too,
but several orders of magnitude. Frequency response is also much better on
an SB Live card. And wow and flutter are infinitely better on the SB Live
because a sound card has no wow or flutter. Finally, an SB Live has no
modulation noise either, and modulation noise is particularly bad (-40 dB or
higher) on all cassette recorders.

Now, if you have any hard facts with which to refute this, I'd love to hear
it.

--Ethan
Anonymous
May 2, 2005 4:47:57 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

On Sun, 1 May 2005 23:25:20 -0700, "Julian Adamaitis"
<nospamJulianPA@Access4Less.net> wrote:

>
>"Ethan Winer" <ethanw at ethanwiner dot com> wrote
>
>> But "resampling error" is not a spec! It's an opinion. Either you can
>> quote
>> which *already established audio spec* is affected, and by how much, or
>> you
>> cannot.
>
>There is indeed an "error", or at least a "change" caused by resampling
>between 44.1 and 48. If it is audible and if so to what extent is it
>significant is an entirely different question.
>
>I just took a short 44.1 audio file and made an inverted copy. When I sum
>the 2 together, there is absolutely no resulting sound because they are bit
>accurate, only inverted. If I sum my original to one that has been
>converted to 48 and then inverted, there is a significant resulting signal,
>which proves something is most definitely changed. A Sound Blaster card
>that resamples everything to 48 definitely changes the sound compared to a
>bit accurate card that comes in at 44.1 and stays there. You cannot state
>this is an opinion. It is a fact. How significant it is, is indeed an
>opinion until someone devises a test to evaluate.
>

Excellent test idea! The point Ethan was making is that resampling
error (or anything else) is more meaningful when expressed with
some sort of numerical value. The mere fact of resampling does
not in itself cause distortion, it has to do with the algorithm used.
(Specifically, the filter used.) Typically the resulting distortion
is inharmonic, so it sounds worse than an equivalent percentage
of harmonic distortion. (I don't have a handy way to quantify how
much worse, since it has to do with psychoacoustics.)

But your test should at least give an accurate numerical value,
since the difference signal is a measure of the distortion.
To make readings really comparable, everyone would need to agree
on a standard input signal and a method to average the
resulting difference signal. But even without a standard, we
might come up with some simple description, such as "difference
peaks up to X% of original signal peaks on solo brass instruments"
or whatever. I like it!

Best regards,





Bob Masta
dqatechATdaqartaDOTcom

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Home of DaqGen, the FREEWARE signal generator
Anonymous
May 2, 2005 4:47:58 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"Bob Masta" <NoSpam@daqarta.com> wrote in message
news:42761e84.1436493@news.itd.umich.edu...
> On Sun, 1 May 2005 23:25:20 -0700, "Julian Adamaitis"
> <nospamJulianPA@Access4Less.net> wrote:
>
>>
>>"Ethan Winer" <ethanw at ethanwiner dot com> wrote
>>
>>> But "resampling error" is not a spec! It's an opinion. Either you can
>>> quote
>>> which *already established audio spec* is affected, and by how much, or
>>> you
>>> cannot.
>>
>>There is indeed an "error", or at least a "change" caused by resampling
>>between 44.1 and 48. If it is audible and if so to what extent is it
>>significant is an entirely different question.
>>
>>I just took a short 44.1 audio file and made an inverted copy. When I sum
>>the 2 together, there is absolutely no resulting sound because they are
>>bit
>>accurate, only inverted. If I sum my original to one that has been
>>converted to 48 and then inverted, there is a significant resulting
>>signal,
>>which proves something is most definitely changed. A Sound Blaster card
>>that resamples everything to 48 definitely changes the sound compared to a
>>bit accurate card that comes in at 44.1 and stays there. You cannot state
>>this is an opinion. It is a fact. How significant it is, is indeed an
>>opinion until someone devises a test to evaluate.
>>
>
> Excellent test idea! The point Ethan was making is that resampling
> error (or anything else) is more meaningful when expressed with
> some sort of numerical value. The mere fact of resampling does
> not in itself cause distortion, it has to do with the algorithm used.
> (Specifically, the filter used.) Typically the resulting distortion
> is inharmonic, so it sounds worse than an equivalent percentage
> of harmonic distortion. (I don't have a handy way to quantify how
> much worse, since it has to do with psychoacoustics.)
>
> But your test should at least give an accurate numerical value,
> since the difference signal is a measure of the distortion.
> To make readings really comparable, everyone would need to agree
> on a standard input signal and a method to average the
> resulting difference signal. But even without a standard, we
> might come up with some simple description, such as "difference
> peaks up to X% of original signal peaks on solo brass instruments"
> or whatever. I like it!

OK, I can easily tell you that. The original sound source has been
normalized and peaks to 0dB. the mix of the original and the out of phase
44.1 - 48 converted version peaks to -7dB. Not much cancellation! This was
with a mix of acoustic instruments - guitar, piano, violin and bass. The
low frequencies cancel well. It is the treble and upper midrange which is
left over to give the difference signal. I tried doing the same thing with
sine waves and the result is much much less dramatic. Instead of 7 dB of
difference on peaks, there is about 66 dB, so the original and the converted
are very close to identical with sine waves. I tried 20, 400 and 10,000 Hz
sine waves, all about 66 dB. That's a huge difference between the real
music and the test tones. Anyone care to speculate why?

Julian
Anonymous
May 2, 2005 4:47:58 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

In article <42761e84.1436493@news.itd.umich.edu> NoSpam@daqarta.com writes:

> Excellent test idea! The point Ethan was making is that resampling
> error (or anything else) is more meaningful when expressed with
> some sort of numerical value. The mere fact of resampling does
> not in itself cause distortion, it has to do with the algorithm used.

> But your test should at least give an accurate numerical value,
> since the difference signal is a measure of the distortion.

A reasonable numeric expression could be the percentage of listeners
tested who can identify when program material has been "resampled"
with an accuracy better than chance. A number based on measurement of
a single frequency test signal, while real, is going to be pretty
meaningless when it comes to the elusive "audio quality."

You could come up with a number by using pink or white noise as a test
signal, inverting a resampled version, summing with the original, and
expressing the RMS amplitude of the "leftovers" as a percentage of the
RMS amplitude of the original signal. But that wouldn't tell you how
it sounds. Only humans can do that.



--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo
Anonymous
May 2, 2005 4:47:59 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Julian,

> That's a huge difference between the real music and the test tones.
Anyone care to speculate why? <

Did you match the levels to get the deepest null possible? If you just
convert and mix, and a tiny but otherwise unrelated level change is
introduced in the process, that -66 dB can easily rise a lot. It takes *very
little* change in level of one source to reduce the effect of a nearly
complete cancellation. Far more than the 53 dB difference implies.

--Ethan
Anonymous
May 2, 2005 4:48:00 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"Ethan Winer" <ethanw at ethanwiner dot com> wrote

> Did you match the levels to get the deepest null possible? If you just
> convert and mix, and a tiny but otherwise unrelated level change is
> introduced in the process, that -66 dB can easily rise a lot. It takes
> *very
> little* change in level of one source to reduce the effect of a nearly
> complete cancellation. Far more than the 53 dB difference implies.

I converted using "convert sample type" in Audition. Then I inverted. No
volume changes were introduced.

If I take a file and simply convert before mixing, there is zero signal,
nada, zip. Not even a flicker of the meters at - 120dB. I can thereby
show my method accurately predicts if 2 files are identical. If I messed
with volumes after doing my sample rate conversion, I would have no
reference anymore or proof of my technique's accuracy.

I suspect there will be no benefit by nulling levels. I suspect changing of
levels will only increase the difference signal.

I'm just wondering why there is so much more difference signal using music
than test tones. One clue is the volume of difference signal is not
constant using music. it peaks quite heavily on the attack of each
percussive guitar strum with the music I used. the meter readings vary
roughly between - 36 and -24, but on the guitar hits it peaks quite heavily
up to as much as -6dB which is quite significant. the difference signal is
only 6 dB quieter than the original which peaks at 0 at these points. You
got to be able to hear that, don't you think, Mike? It is apparently the
attack part of the envelope that is most effected by resampling. When
looking at the difference signal generated by using sine wave test tones, it
only varies about +/- 1 dB and it doesn't appear the difference is effected
by frequency of the tones. I used 20 Hz, 400 Hz, and 10 kHz and the
difference signal was about the same magnitude in each case.

Julian
Anonymous
May 2, 2005 6:33:55 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

Julian,

> I converted using "convert sample type" in Audition. Then I inverted. No
volume changes were introduced. <

It may be that the differences are very high in frequency. Did you apply the
nulling and listen to the result? What do you hear?

Maybe even more relevant, a small amount of phase shift can preclude being
able to null the files, even if there was no audible or measurable
degradation. So in that case you might hear weird stuff when listening to
the nulled result, but the sample-converted file was not degraded in any
way.

--Ethan
Anonymous
May 2, 2005 6:33:56 PM

Archived from groups: rec.audio.pro,comp.sys.ibm.pc.soundcard.tech (More info?)

"Ethan Winer" <ethanw at ethanwiner dot com> wrote in message
news:wYedndtIDPmd6evfRVn-vQ@giganews.com...
> Julian,
>
>> I converted using "convert sample type" in Audition. Then I inverted.
>> No
> volume changes were introduced. <
>
> It may be that the differences are very high in frequency. Did you apply
> the
> nulling and listen to the result? What do you hear?

As I said in a previous post, it was mostly high end and upper midrange.

> Maybe even more relevant, a small amount of phase shift can preclude being
> able to null the files, even if there was no audible or measurable
> degradation. So in that case you might hear weird stuff when listening to
> the nulled result, but the sample-converted file was not degraded in any
> way.

Any phase shift that was introduced by the resampling. Period.

If I do the exact same thing with no resampling there is no difference
signal. Not even at -120 dB below digital zero. I am well aware that even
the least amount of phase shift can change these results. I taken two files
that exactly cancel themselves out and zooming in to the sample level and
shifting on file by one or two samples. The result is enormous. The
difference signal changes from under -120dB to -27 dB. When you shift a
signal by a sample or two, you hear a lot of phasing. I hear no phasing in
the difference signal generated by resampling. It is just high frequency
and upper midrange. No phasing. It is not out of phase or shifted even one
sample.

My test is bit accurate. I am open to be proven wrong, but if I am right,
you can't possibly claim the sample-converted file wasn't degraded. I can't
claim the degradation is audible but if the "error" signal" is only 6 dB
below the "original signal" on musical peaks, I'd have a hard time arguing
it isn't.

Julian
Anonymous
May 2, 2005 6:39:36 PM

Archived from groups: rec.audio.pro (More info?)

Mike,

> But if it would make him feel better to get another sound card, it might
as well be an M-Audio Audiophile or Echo Mia. <

Agreed. I have a Delta 66, and several of my friends have an Audiophile
2496, and they're both great cards.

> I really don't know just how bad a modern one can be at its worst <

Not bad at all. Certainly nothing like some people seem to think. And
certainly far better than any stupid cassette recorder! :->)

--Ethan
!