Music Player Network
Previous Thread
Next Thread
Print Thread
Page 5 of 7 1 2 3 4 5 6 7
Joined: Jun 2001
Posts: 175
G
GD Offline
Senior Member
Offline
Senior Member
G
Joined: Jun 2001
Posts: 175

Joined: May 2000
Posts: 477
S
Senior Member
Offline
Senior Member
S
Joined: May 2000
Posts: 477
All I know is this. When i record at higher sample rates, to me I lose less sonically as compared to what i am hearing from the preamp. I suspect that it is all about converter desin as opposed to sample rates, and that is exactly the reason we are all headed for higher rates. It IS cheaper to build a beter sounding converter wen using highr sample rates, and that is what 90% of manufacturers will do. What sucks is that we will be having this argument for a good amount more time I should think, and all we have to go on is personal preference of sonics ( not that I am against that as a decision maker) when we went to 24 bit, it was obiously beter, and it was obivious why.. this is all so subjective.

Joined: Mar 2000
Posts: 7,950
MP Hall of Fame Member
Offline
MP Hall of Fame Member
Joined: Mar 2000
Posts: 7,950
Quote:
Originally posted by MrDGreen:
Sorry, show me once in any of my posts where I said it was "perfect".
I'm stating that the current A/D and D/A technology is linear and well spec'ed to very precise degrees, to values that are essentially non-perceptable (inperceptable?).

To exemplify this, as I've mentioned before current CODECs usually spec in the range of 10Hz to 20k/40kHz FR, Bandpass ripple of +/-0.005dB, 90-100+dB S/N, 90-100+dB DR, etc.

If we feel that the A/D/A portion is the weakest link, and that this is the worst specs one has to worry about in the signal chain, then they better check the specs on the mics, monitors, effects, etc.
you just said it again, i made it bold. play symantics if you want but what you are essentially saying is that the ADAC conversion is imperceptible/perfect/flawless/et al.

otoh, i think there is still error, either by lack of resolution or some other factor [of which there are many im sure]

can we take an impulse and it perfectly be represented digitally? then send that impulse back out to analog and back in again and it still be perfect? and again? if you cant, there is still error in the system.


alphajerk
FATcompilation
"if god is truly just, i tremble for the fate of my country" -thomas jefferson
Joined: Mar 2002
Posts: 20,318
G
TPS cook & bottle washer
20k Club
Offline
TPS cook & bottle washer
20k Club
G
Joined: Mar 2002
Posts: 20,318
Quote:
Originally posted by alphajerk:
you just said it again, i made it bold. play symantics if you want but what you are essentially saying is that the ADAC conversion is imperceptible/perfect/flawless/et al.

otoh, i think there is still error, either by lack of resolution or some other factor [of which there are many im sure]

can we take an impulse and it perfectly be represented digitally? then send that impulse back out to analog and back in again and it still be perfect? and again? if you cant, there is still error in the system.
Well, there is a difference between the ADA conversion specifications and the ADA implementations

I feel you are correct as far as the idea that the implementation is still very much a flawed ideal.

No company has managed to step forward with a perfect A/D or D/A converter, regardless of the science behind the codecs.

Until this perfection is acheived, we still have error.

Superior to analog? Statistically, yes. No analog format can acheive the SNR that digital is capable of.

Problem is, we humans still prefer analog noise to digital noise.

Joined: Aug 2000
Posts: 1,242
Platinum Member
Offline
Platinum Member
Joined: Aug 2000
Posts: 1,242
Quote:
Originally posted by Griffinator:
This is why I'm trying desperately to scratch up the funds necessary to outfit my studio with DSD AD and DA converters and the DAW to utilize them.

Unfortunately, it's about a $10,000 upgrade.
Didn't you just bashed SACD on the other thread?


"There's no right, there's no wrong. There's only popular opinion" Jeffrey Goines
Joined: Feb 2000
Posts: 2,184
gm Offline
MP Hall of Fame Member
Offline
MP Hall of Fame Member
Joined: Feb 2000
Posts: 2,184
Quote:
Originally posted by Loco:
Quote:
Originally posted by Griffinator:
This is why I'm trying desperately to scratch up the funds necessary to outfit my studio with DSD AD and DA converters and the DAW to utilize them.

Unfortunately, it's about a $10,000 upgrade.
Didn't you just bashed SACD on the other thread?
Guys, there's a WORLD of difference between "SACD" and "DSD". SACD is a Sony/Philips release medium (with all that entails, including copy protection). "DSD" is the recording technology, and it's alot of different things in and of itself. For instance, DSD format on SACD is a single-bit, 2.8-ish megaHertz stream (or 64X, as it's commonly termed). most pro production is on workstations utilizing DSD-Wide, or 64X, 8-bit wide streams. There have been numerous suggestions as to how to move it up to a truly invariant professional format, and this has been stalled on occasion by the idiots at Sony in Japan for solely political reasons. Any of these (128X, 192X, 256X, 8 X 16, 16 X 16 & etc) would eventually be released in 64FS single-bit, but represent various degrees of precision, and (much more importantly) noise floor.

You don't have to hate DSD to loathe Sony and their ham-handed, greedy marketing methods.

George


George Massenburg

http://www.massenburg.com
Joined: Mar 2002
Posts: 20,318
G
TPS cook & bottle washer
20k Club
Offline
TPS cook & bottle washer
20k Club
G
Joined: Mar 2002
Posts: 20,318
Loco, I did nothing of the sort. I am quite impressed with the sonic fidelity of SACD vs. PCM of any sort (DVD-A, CD, HDCD, DTS)

George - the frustration for me is that I despise Sony, but I love DSD and the SACD format.

Joined: May 2003
Posts: 4,836
S
MP Hall of Fame Member
Offline
MP Hall of Fame Member
S
Joined: May 2003
Posts: 4,836
Yes, I know Mr. Burr's nifty little company was purchased by TI for around 4 billion dollars. Whether they are still the undisputed world's champs in AD/DA is another question, but I believe they still use the name "Burr-Brown."

Nevertheless, if everthing were right, there would be no debate.

When things are not right, you begin to question the "underlying assumptions" and go back and do good, objective science.

The "time smear" problem is quite real according to Hutch at Manley Labs and Julian Dunn was apparently working on trying to explain the "smear problem" before his untimely death.

Clearly, the goal of great A-to-D and D-to-A is to represent the real world. Equally clear, people are not satisfied with many of the current implementations. That means going back to the drawing board.

To remove all the black magic and all the gobble-dee-gook, the order-of-the-day is for more "applied research" and more "objective science."

Joined: Feb 2002
Posts: 479
Senior Member
Offline
Senior Member
Joined: Feb 2002
Posts: 479
There seems to be a lot of talk about why is it necessary to have bandwidth beyond 22Khz with 88.2 and above?
I think two far more important factors with using high sample rates are higher resolution at audible frequencies and lower DSP latency as well.

Joined: Mar 2002
Posts: 20,318
G
TPS cook & bottle washer
20k Club
Offline
TPS cook & bottle washer
20k Club
G
Joined: Mar 2002
Posts: 20,318
Quote:
higher resolution
RED HERRING ALERT! \:D

Joined: Nov 2002
Posts: 108
Senior Member
Offline
Senior Member
Joined: Nov 2002
Posts: 108
George,

Like Mark W, I am also utilizing the LynxTWO; & have been getting great results recording (in SONAR 2.2) @ 24/96; bouncing to stereo .wav (@24/96); & then using WaveLab 4.0 to convert to CD Quality (by 1st converting the sample frequency to 44.1KHz; & then converting the bit depth to 16, rendering the file with a high-quality dither.) In most cases (& I've performed a few tests on this), the files recorded at 96; & SCR'd to 44.1, sound much better than originally recorded at 44.1

So I guess my question is: what about the basic known advantages of starting high (e.g., spreading the noise to the higher freqs > increasing S to N ratio, post-processing benefits, etc.)? Are you saying that these advantages are lost in downsampling...or that there not all they're cracked up to be in the first place?

I'd like to know (if only for the benefit of saving disc space.) But my ears are telling me the 96 downsamples are cleaner (&...they're 50 year old ears!)

Thanks,

mark4man

Joined: Nov 2002
Posts: 108
Senior Member
Offline
Senior Member
Joined: Nov 2002
Posts: 108
--------------------------------------------------------------------------------------------------
Originally posted by MrDGreen:

has anyone performed similar tests to what I have on my SRC page using D/A/D instead?
--------------------------------------------------------------------------------------------------
David,

Where exactly is this SRC page (couldn't find it at lilchips)...I'd be interested in the tests.

Thanks,

mark4man

Never mind...found them (sorry for spot reading.)

Joined: Jun 2001
Posts: 175
G
GD Offline
Senior Member
Offline
Senior Member
G
Joined: Jun 2001
Posts: 175

Joined: Nov 2002
Posts: 108
Senior Member
Offline
Senior Member
Joined: Nov 2002
Posts: 108
Quote:
Originally posted by MrDGreen:
If you are going 96->44.1k, I'll guarantee that 88.2->44.1 will give better results unless the SRC method you are using is not optimized
For real?

So then the difference would be Synchronous vs. Optimized...Synchronous requiring direct multiples?

Where can I find white papers on the two?

Thanks,

mark4man

Joined: Jun 2001
Posts: 175
G
GD Offline
Senior Member
Offline
Senior Member
G
Joined: Jun 2001
Posts: 175

Joined: May 2003
Posts: 4,836
S
MP Hall of Fame Member
Offline
MP Hall of Fame Member
S
Joined: May 2003
Posts: 4,836
Here, you can get some free white papers and download them. You may have already read them, I dunno. Hope it helps. \:\)

http://www.nanophon.com

Joined: Mar 2000
Posts: 7,950
MP Hall of Fame Member
Offline
MP Hall of Fame Member
Joined: Mar 2000
Posts: 7,950
the whole SRC debate is pointless if convertors were built RIGHT. one should simply be able to output analog and back in with no quality loss at a lower sample rate. otherwise, the traditional method of digital recording is filled with error that must be corrected.

its like running an analog board and the path from the input to the EQ is faulty or to the compressor, or through the faders and pan functions. the insert point generating a shittier signal out and even worse back in nevermind what was put in line. we should be able at any time to send DAC to analog outboard with no error all they way through. the signal chain from the source should be EXACT from the point it hits the ADC back through to the monitors via the DAC. the live signal should be undisguinshible from the fully ADAC'd output.

SRC is meaningless.


alphajerk
FATcompilation
"if god is truly just, i tremble for the fate of my country" -thomas jefferson
Joined: Jan 2000
Posts: 1,032
Platinum Member
Offline
Platinum Member
Joined: Jan 2000
Posts: 1,032
Quote:
Originally posted by alphajerk:
the whole SRC debate is pointless if convertors were built RIGHT. one should simply be able to output analog and back in with no quality loss at a lower sample rate. otherwise, the traditional method of digital recording is filled with error that must be corrected.

its like running an analog board and the path from the input to the EQ is faulty or to the compressor, or through the faders and pan functions. the insert point generating a shittier signal out and even worse back in nevermind what was put in line. we should be able at any time to send DAC to analog outboard with no error all they way through. the signal chain from the source should be EXACT from the point it hits the ADC back through to the monitors via the DAC. the live signal should be undisguinshible from the fully ADAC'd output.
In a perfect world. As it is, even the best, most expensive converters can't deliver the same output as input.


Lynn Fuston
3D Audio Inc
Home of 3dB
Joined: Nov 2002
Posts: 108
Senior Member
Offline
Senior Member
Joined: Nov 2002
Posts: 108
David,

I am assuming your site is the same site as recommended by one; Scott Garrigus, going all the back to the original "Cakewalk Power"???

Unbelievable!...good stuff, very good stuff (actually, great stuff!)

Your SRC link wouldn't open for me yesterday. Today, it opened; & I read. Right out of the gate, you answered a question that's been nagging me for weeks. You write:

"Audio file bit-depth is also an issue, as lower bit-depths will increase the error percentage since they can't properly represent the interpolation values with sufficient accuracy"

This tells me my conversion sequencing is correct [SRC prior to bit depth conversion (w/ dither.)]

Then, I checked out the 16-Bit test tones. In the second tone (reSRDed), the freakin' upper harmonics were gone ...(as in, not present!)

I now know what I have to do. I've already determined (on my system) that 96 SRC'd to 44.1 sounds better than original 44.1 (utilizing identical source material.) Since my LynxTWO provides for recording at an 88.2KHz SR; & since WaveLab can generate clinical test tones, I intend to perform my own tests (with both audio material & test tones.) If it turns out that the 88.2 downsampled to 44.1 sounds better than the 96 > 44.1 (which theoretically it should), I will have become a convert.

Thanks much (& for the stanford.edu links, as well),

mark4man

BTW - Johnny B...
The Julian Dunn links are excellent. Thanks. This has been one hell of a thread. Eyeryone's input has been tremendous (Oh man, the anti-depressants are kickin' in.)

Joined: Jul 2002
Posts: 487
Senior Member
Offline
Senior Member
Joined: Jul 2002
Posts: 487
I agree that mixing/mastering at the highest sample rates available shall compensate for artifacts but record at 192k when your microphones, monitors, mic pres hardly reach 24khz?

Hey folks, check this poll I started thi wek:

http://www.recording.org/cgi-local/ubb/ultimatebb.cgi?ubb=get_topic&f=21&t=001198#000016


Eng. Al
Joined: Jul 2002
Posts: 487
Senior Member
Offline
Senior Member
Joined: Jul 2002
Posts: 487
As you can see in the top of the page, click on "results" or "vote" if ya wish. At this moment, 65% of the users still work at 44.1k/24.


Eng. Al
Joined: Feb 2000
Posts: 2,184
gm Offline
MP Hall of Fame Member
Offline
MP Hall of Fame Member
Joined: Feb 2000
Posts: 2,184
Well, I simply can't vote. Work is not so straight-forward for me as I work in several different standards, each for their own reasons.

George


George Massenburg

http://www.massenburg.com
Joined: Jan 2000
Posts: 1,032
Platinum Member
Offline
Platinum Member
Joined: Jan 2000
Posts: 1,032
Quote:
Originally posted by gm:
Well, I simply can't vote. Work is not so straight-forward for me as I work in several different standards, each for their own reasons.

George
As about half of my work is remixing, I have to agree with George. I mix whatever comes in the door. So far, I haven't even had a single mention of anything above 48K. Except for 96K remixes on Masterlink. But even those were sourced from 48K master PT sessions.


Lynn Fuston
3D Audio Inc
Home of 3dB
Joined: Jul 2002
Posts: 487
Senior Member
Offline
Senior Member
Joined: Jul 2002
Posts: 487
Thanks Guys!
Would you plese give examples where you are forcedgoing in different sample rates/bit depths?
video?
2" to digital transfers?

Nice weekend


Eng. Al
Joined: Sep 2000
Posts: 1,277
A
Platinum Member
Offline
Platinum Member
A
Joined: Sep 2000
Posts: 1,277
david Green wrote,

"If you are going 96->44.1k, I'll guarantee that 88.2->44.1 will give better results unless the SRC method you are using is not optimized."

I'm glad to see another vote for this (besides Dan Lavry.). However, as per my earlier post, I have yet to find an src that IS optimized for the "50%" conversion. can you tell me which software src's ARE, that you know of?

Also, David, what tests do you base that stement on? (I also could not access your src page.)

-thanks.

Joined: Apr 2002
Posts: 242
D
Senior Member
Offline
Senior Member
D
Joined: Apr 2002
Posts: 242
Quote:
the whole SRC debate is pointless if convertors were built RIGHT. one should simply be able to output analog and back in with no quality loss at a lower sample rate.
If that's the case, then why even record at a higher sampling rate in the first place?

-Duardo

Joined: Jul 2002
Posts: 487
Senior Member
Offline
Senior Member
Joined: Jul 2002
Posts: 487
Hey, it would be cool to have some nice talent of bud Nika Aldrich on here.
Calling Nika....

\:\)


Eng. Al
Joined: Jan 2002
Posts: 4,649
MP Hall of Fame Member
Offline
MP Hall of Fame Member
Joined: Jan 2002
Posts: 4,649
So we're doing all this work with recording music, usually with the assumption that it is for humans to buy, and then maybe listen to for indefinable reasons... fun? pleasure?

Fact is, we don't know where this is all leading, what the actual point of all this may be, only the assumptions we're currently toiling under.

I won't be surprised when music takes the place of many forms of medicine. I've heard the example of if your nose is congested, listen to some bassoon and see what happens. Or whatever, the point is we don't know and the wisest of us will admit that.

I'm still looking for a physician who knows those words- "I don't know!"


A WOP BOP A LU BOP, A LOP BAM BOOM!

"There is nothing I regret so much as my good behavior. What demon possessed me that I behaved so well?" -Henry David Thoreau
Joined: Sep 2000
Posts: 1,277
A
Platinum Member
Offline
Platinum Member
A
Joined: Sep 2000
Posts: 1,277
Alécio ,

You seem to be assuming that the only difference when recording at higher s-r is an extended HF response. There are obviously audible advantages that exist below 20Khz, although some may be very subtle. This whole thread is about what those other differences might be, how imprtant they really are, and whether or not they translate to the final (44.1K cd or worse) released media..

Joined: Aug 2000
Posts: 1,242
Platinum Member
Offline
Platinum Member
Joined: Aug 2000
Posts: 1,242
If it's an R&B session and Alsihad is the machine on the room, then 44.1/24 is the way I go. Track count is crucial and Alsihad is handicapped on that matter.

For the rest, I go 88.2/24 if available. DP as my first choice.


"There's no right, there's no wrong. There's only popular opinion" Jeffrey Goines
Page 5 of 7 1 2 3 4 5 6 7

Moderated by  gm 

Link Copied to Clipboard
Powered by UBB.threads™ PHP Forum Software 7.7.5