#321209 - 07/24/03 11:04 PM
RDAC quality analysis results.
|
bbadger
Planeteer
Registered: 07/24/03
Posts: 117
Loc: Colorado Springs, CO
|
All,
It has been many years (ca 1997.. Doug remembers me...) since I last commented on the RDAC algorithm. I had an opportunity to sit down with the patent text, a cup of coffee and my Magic Japlish Decoder Ring, and ran up some theoretical characterizations of the RDAC coding scheme.
As a waveform coding algorithm, it is similar in behavior to ADPCM, but is significantly more complicated. It exhibits quantization errors which increase in proportion to the frequency of the signal being encoded. The result of RDAC encoding should be a noise spectrum that is heavily weighted to the >10kHz band.
The basic premises of the RDAC coding appears to be that 1) low-level quantization errors (read that "noise") are psychoacousticaly masked by loud signals, and 2) most program content (read that "music") has all or most of it's pure tone signals below 10kHz.
Having worked up the theory, I built two test cases to analyze the actual performance of MTP using my newly acquired, still smells like the factory, VS2400CD.
First testcase:
1) Setup project, M24/44.1k.
2) Generate pink noise, record to track 1
3) Export track 1 to .wav
4) Save project
5) Setup project, MTP/44.1k
6) Import wav to track 1.
7) Export track 1 to wav.
8) Open original project
9) Import wav to track 2.
10) Set track 2 to INVERTED playback.
11) Set track 1 and 2 to +6db
12) Turn down the monitors, press play
13) Slowly, carefully turn up the monitors.
Result:
The signals of both tracks cancelled, leaving the MTP quantization errors behind in the mix as noise. The noise is high pitched -- sounds like tape hiss. The level is very low, and didn't show up on the meters. Used the analyzer function, which showed perfect cancellation up to about 10kHz, and then white noise at about -48dB from 10kHz to 20kHz.
Second Testcase:
Repeat first testcase with new step 2:
2) Generate 12kHz pure tone, record to track 1
Result:
Again, the signals of both tracks cancel, leaving behind low level broadband noise from 10kHz to 20kHz. Measured S/N at 12kHz = ~45dB.
What does it all mean:
1) RDAC is definitely lossy. (Same conclusion as last time, different millenium)
2) The psychoacoustic assumptions in the RDAC design appear to be correct. Pure tones above 10kHz are useful for dog whistles and pest repellers. Speech and musical instruments have content above 10kHz, but this content is almost never pure tone in nature. Since the source content abve 10kHz can be expected to be noise sources, quantization errors in this band are largely inaudible. The "hiss" is only present when full magnitude signals are present, and is largely masked by the signal.
3) I continue to use RDAC. I now know what it sounds like when it's working... it's not offensive to me, and the "tape hiss" like character may be partly responsible for the Roland VS line's "musical sound".
_________________________
Brian Badger
|
|
Top
|
|
|
|
|
#321210 - 07/24/03 11:26 PM
Re: RDAC quality analysis results.
|
LakeStone Karl
Planeteer
Registered: 06/26/02
Posts: 15821
Loc: Eastern PA, USA
|
Great post! I do not know you, but your logic and test methods seem sound.
So, it's almost a digital tape machine?! (  )
Thanks for the technical write up.
[ 07-24-2003: Message edited by: LakeStone Karl ]
|
|
Top
|
|
|
|
|
#321211 - 07/24/03 11:41 PM
Re: RDAC quality analysis results.
|
White Rabbit
Planeteer/Artist # 100
Planeteer
Registered: 08/04/00
Posts: 3828
Loc: Born in Scotland, but live in ...
|
so what your saying is that the RDAC really only effects frequencies above 10Khz, the majority of the signal goes (almost) unaffected by the encoding process.
If thats the case then all is well because I'll bet half of us are deaf by now and cant hear above 10k anymore
|
|
Top
|
|
|
|
#321212 - 07/24/03 11:54 PM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Originally posted by bbadger:
First testcase:
1) Setup project, M24/44.1k.
2) Generate pink noise, record to track 1
3) Export track 1 to .wav
4) Save project
5) Setup project, MTP/44.1k
6) Import wav to track 1.
7) Export track 1 to wav.
8) Open original project
9) Import wav to track 2.
10) Set track 2 to INVERTED playback.
11) Set track 1 and 2 to +6db
12) Turn down the monitors, press play
13) Slowly, carefully turn up the monitors.
Second Testcase:
Repeat first testcase with new step 2:
2) Generate 12kHz pure tone, record to track 1
I think a third testcase would be even more interesting:
Repeat first testcase with new step 2:
2) Generate 100 Hz square wave, record to track 1
I don't know about R-DAC but I know that ATRAC (for Mini Disc) has great troubles with square wave. Would be interesting to verify that R-DAC doesn't have the same problem.
It would also be interesting to compare the old MT1, MT2, LIV1 and LIV2, but maybe not that necessary...
Also, when having that wav file available, why not let your PC make a mp3 copy of it at highest bit rate. Just for fun... And then there are wma, mp4, ogg and vqf... Well... this could be a very large project indeed... so I'd better stop here...
/Johnny Andersson
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321213 - 07/25/03 01:07 AM
Re: RDAC quality analysis results.
|
Rick O-Shay
Planeteer
Registered: 03/07/02
Posts: 160
Loc: pacifica, ca
|
bbadger,
Last year I did some tests similar to yours and reached many of the same conclusions. I put together a few documents:
The first is an overview of data compression (written for the non-technical person) and the other sums up my RDAC test results. There are also some audio examples included.
Anyone who is interested can download the info at http://webs.lanset.com/psm/rdac.zip
|
|
Top
|
|
|
|
|
#321214 - 07/25/03 02:00 AM
Re: RDAC quality analysis results.
|
FalconEddy
Retired Mastering Marvel
Planeteer
Registered: 10/16/01
Posts: 10645
Loc: Litchfield, NH, USA
|
I also performed basically the same tests across most the recording modes available, and noted some interesting results.
Unfortunately, that hard drive had died a horrible death, and most of the data from those tests are unrecoverable.
However, I believe I still have some Spectrum Analysis graphs available on one of my other NT Servers (or, backup tape) that show RDAC dB levels for the original stereo samples, MTP, MT1, and LIV2 modes.
I recall the difference (for my particular musical passage tested) averaged out at -40.9dB below the levels of the original sample, in both MTP and MT1, was fairly flat, and the SA was nearly identical for both these modes.
But LIV2 mode was, as expected, poop! (that's a technical term)  With huge losses across the entire spectrum.
. . Falcon
_________________________
. . . but, what do I know?
|
|
Top
|
|
|
|
#321215 - 07/25/03 03:54 PM
Re: RDAC quality analysis results.
|
transfiguration
Chris
Planeteer
Registered: 10/28/02
Posts: 2232
Loc: Lockhart, TX
|
Excellent thread gentlemen! Thank you for doing the homework.
|
|
Top
|
|
|
|
|
#321216 - 07/25/03 09:13 PM
Re: RDAC quality analysis results.
|
$W?FT
Planeteer
Registered: 07/25/03
Posts: 40
Loc: Detroit
|
very nice dialogue gentlemen,
exactly the reasons that i joined this group. i must say that the outcome of your testing is very encouraging. technically, we realize that any type of compression will have some lost, but artistically i love the vs2480's sound. i just finished mixing a project that i feel is industry standard, using only the vs8f-2 effects cards. i personally couldn't ask for better quality. maybe someday soon i'll post a mp3. i'm very happy i selected this daw despite the naysayers.
|
|
Top
|
|
|
|
|
#321217 - 07/26/03 09:16 AM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Originally posted by $W?FT:
i'm very happy i selected this daw despite the naysayers.
"Naysayers"? Is that English? I just wonder, because "nej" in Swedish (pronounced "nay") is "no" in English, so I guess "naysayers" would be "nejsägers" in "Swinglish" (some of the swedish slang words are kind of hybrids between Swedish and English, therefor "Swinglish"...) and "nejsägare" in real Swedish.
Is "naysayers" a real word or is it slang? By the way, "slang" is the same in Swedish and also means tube, tubing or hose.
There are a lot of English words in Swedish, but some of the words were actually Swedish before they were English. Like the Swedish word "klubb" which means "club". It was brought to English by the vikings, and it originate from the word "klubba" which means club, bat or gavel. It also means lollipop, but I guess that is a different story...
Well, I am sorry for not mentioning the VS-2480 this time, I just wondered about that word. I guess I just have to learn English some day.
Maybe I should just get out of bed and have some breakfast right now and then start up my 2480 and try to get the bass guitar of my version of "Diamond Head" to sound properly.
See you later, guys...
/Johnny Andersson
(Yes, Andersson is originally a Swedish surname and is originally spelled like I just did. It means Anders son which in English is Anders' son or maybe son of Anders. Anders is a common first name in Sweden. Some decades or so ago, children had their surnames after their parents' first name, but nowadays the kids just get the same surname as their parents, just like in most other countries, so my father's first name was not Anders...)
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321218 - 07/26/03 09:23 AM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Originally posted by guraknugen:
Some decades or so ago, children had their surnames after their parents' first name
Hehehe... what was I thinking of...? Of course the kids' surnames came from their FATHER'S first name. It doesn't happen very often that both parents have tha same first name, does it...? So it was always the father. There are no surnames like "Mariasson" or "Ingridsson", I least not as far as I know...
/Johnny Andersson
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321219 - 07/26/03 10:26 AM
Re: RDAC quality analysis results.
|
Boray
Planeteer
Registered: 10/15/01
Posts: 4019
Loc: Gothenburg, Sweden
|
Thanks for your test!!!
What I would like to know is if MT1 has less hiss than MTP?
Or maybe they have the same amount of hiss, just that MT1 SHOULD have more quantisation noise being a 16 bit mode...?
/Anders Persson
(Anders, son of Per )
|
|
Top
|
|
|
|
|
#321220 - 07/26/03 12:07 PM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Originally posted by Boray:
Or maybe they have the same amount of hiss, just that MT1 SHOULD have more quantisation noise being a 16 bit mode...?
On the other hand, according to the Owner's manual page 98 (of 452), MTP seems to compress 3:1 and MT1 2:1.
The recording time is the same, but since MTP is 24 bits and MT1 is 16 bits and since MT1 is 2:1, MTP have to be 3:1... 2×24/16=3...
Or (page 98 again): Recording time at 44.1 kHz at M24=1442 minutes. At MTP: 4329 minutes. 4329/1442=3.00.
M16=2165 minutes, MT1=4329 minutes. 4329/2165=2.00.
Well, just a thought...
Johnny Andersson
Son of a bitch..?
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321221 - 07/26/03 12:17 PM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Originally posted by Boray: /Anders Persson
(Anders, son of Per )
Hi Anders! You aren't my father, are you...? Hehehe
That Swedish flag of yours, shouldn't it be rotated 90°?
/Johnny Andersson
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321224 - 07/27/03 05:58 PM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Originally posted by Boray:
About MTP vs MT1... With RDAC, what's mainly recorded is the difference between every sample. And this is stored as a kind of floating point number. The bigger difference, the less accurate will it be. And when will there be big differences between every sample? - The higher (and louder) frequency recorded, the less accurate will RDAC be. And as RDAC originally was designed for 16 bit, it could be that way that for the highest and loudest frequencies, the accuracy of MTP will be more that of 16 bit (or less) than that of 24 bit. I think that's a reasonable assumption.
That would also mean that a square wave at 0 dB would be very different from the original in both cases, wouldn't it?
/The Guraknug Viking from Sweden...
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321225 - 07/28/03 09:46 AM
Re: RDAC quality analysis results.
|
Boray
Planeteer
Registered: 10/15/01
Posts: 4019
Loc: Gothenburg, Sweden
|
Originally posted by guraknugen:
That would also mean that a square wave at 0 dB would be very different from the original in both cases, wouldn't it?
I don't think it has any trouble generating a more or less perfect square wave, but maybe it won't get exactly the same amplitude. RDAC adjusts to the content all the time, trying to find a way to store the differences between samples the best way it can but using the same bitlength storing it (for a small block of data). I think it also stores a real value for every new block. So I guess it can generate a square wave, just that it won't get as exact as 24 bit, but more like 16 or less. Noone will hear the difference. Not on a loud square wave. Where I think you in theory maybe should be able to hear a difference, would be on things like cymbals.
I have gathered this from the RDAC patent, but I don't think I have the link any more, and VSPlanet has lost the thread it was in....
/Anders
|
|
Top
|
|
|
|
|
#321226 - 11/14/03 06:13 PM
Re: RDAC quality analysis results.
|
$W?FT
Planeteer
Registered: 07/25/03
Posts: 40
Loc: Detroit
|
I just had a hearing test during a physical. They don't even test you above 8 kHz. So if the only content lost is above 10 kHz then I don't think most people miss it. For those who speak negatively about RDAC, I wonder if they can even hear frequencies above 10 k. Most musicians can't, cuz we like our music too loud.
|
|
Top
|
|
|
|
|
#321227 - 11/14/03 06:40 PM
Re: RDAC quality analysis results.
|
Brianonymous
Planeteer/Artist # 99
Planeteer
Registered: 07/18/02
Posts: 5635
Loc: Laptop
|
I can hear 10k as a pure tone, but more often I can hear it when its missing from a mix. Especially in the crispy stuff, like cymbals.
If you ever play with winamp sometime on a good set of speakers, and a wav file, give the 16khz EQ bar a crank and check out the overall difference.. You mostly likely wouldnt hear the 16khz tone if it was by itself, but added in context can make a very interesting overall difference. I do this a lot to figure out what frequencies im missing or have too much of in a mix.
This RDAC stuff sounds very simular to layer 3 audio codecs, in the fact that it masks the inaudible. Very interesting stuff. When I get back home, I think Im going to experiment and see how much the upper end gets scrambled up.
|
|
Top
|
|
|
|
|
#321228 - 11/14/03 07:02 PM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Brian,
A few years ago I could hear 15 kHz (I haven't tried it since then, maybe I still can) but I can't really hear anything defective which I can relate to R-DAC. I can hear a lot of defects on mp3 at 160 kbps and wma at 96 kbps, and it doesn't seem to be comparable to R-DAC at all. On the other hand there's quite a difference between 2:1 and about 10:1...
About Winamp EQ, you didn't mention the value of Q. The lower Q the more frequencies are affected, so even if it says 16 kHz it will still affect 10 kHz a lot at low values of Q. The more frequency bands on an EQ, the higher Q is needed to obtain linerar frequency response when all frequency bands are set to 0 dB.
So affecting the 16 kHz doesn't mean that only 16 kHz is affected, that's why you can hear such a difference.
[ 11-14-2003: Message edited by: guraknugen ]
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321229 - 11/14/03 07:19 PM
Re: RDAC quality analysis results.
|
Brianonymous
Planeteer/Artist # 99
Planeteer
Registered: 07/18/02
Posts: 5635
Loc: Laptop
|
gura: oh yeah! Dont get me wrong, im not complaining about the RDAC, I think it sounds great. Im just curious about the implications it could have: like how much of a difference would a high end pre really make (striving for the ultimate pure duplication) if the spectrum is ultimately different in the end. Im sure they help, no question.. But what is the upper limit of quality before it begins to not matter.
Winamp is a broad example of what I was trying to imply too.. I have also used 32 band EQs and what not and can hear a presence difference in the higher higher band frequencies. Yeah, the lower band resolutions of winamp would definitely effect the overall in the standard curve configs, im just really saying that those frequencies are indeed helpful and noticable.. Im sure RDAC does very very minimal mangling.
I love my 2480 :-D
|
|
Top
|
|
|
|
|
#321230 - 11/14/03 09:35 PM
Re: RDAC quality analysis results.
|
mike buzz
Planeteer
Registered: 04/28/99
Posts: 1958
Loc: paso robles ca usa
|
Acually this point is exactly the complaint ( outside of this board ) about the RDAC compression routine , any compression has loss it just happens that RDAC has it in the hi end .
LAter
Buzz
|
|
Top
|
|
|
|
|
#321231 - 11/15/03 03:10 AM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Originally posted by mike buzz:
any compression has loss it just happens that RDAC has it in the hi end
Well... ZIP, CAB etc are also kinds of compression and they are lossless. However they don't compress audio very much... but it IS compression and it IS indeed lossless...
Yes, I know, a bit out of subject, but I couldn't resist writing that...
By the way, ZIP, CAB, ARJ etc are data compression. R-DAC, ATRAC, MP3, WMA, VQF etc are data reduction.
Data compression - lossless
Data reduction - lossy
Right?
[ 11-14-2003: Message edited by: guraknugen ]
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321232 - 11/15/03 10:58 AM
Re: RDAC quality analysis results.
|
Brianonymous
Planeteer/Artist # 99
Planeteer
Registered: 07/18/02
Posts: 5635
Loc: Laptop
|
Originally posted by guraknugen:
By the way, ZIP, CAB, ARJ etc are data compression. R-DAC, ATRAC, MP3, WMA, VQF etc are data reduction.
Data compression - lossless
Data reduction - lossy
Right?
[ 11-14-2003: Message edited by: guraknugen ]
100%!
|
|
Top
|
|
|
|
|
#321233 - 11/15/03 05:28 PM
Re: RDAC quality analysis results.
|
bbadger
Planeteer
Registered: 07/24/03
Posts: 117
Loc: Colorado Springs, CO
|
Originally posted by mike buzz:
Acually this point is exactly the complaint ( outside of this board ) about the RDAC compression routine , any compression has loss it just happens that RDAC has it in the hi end .
But let me qualify the analysis that you have summarized as "RDAC has loss in the high end."
RDAC only has loss in the high end when there are loud signals in the high end. That is, the noise level is not 48db below maximum signal level... its 48db below the CURRENT signal level. The dynamic range is actually 144db, because the noise level with no signal is -144db.
The well-established science of psychoacoustics shows us that the loud signal (the REAL signal) will greatly mask our ability to hear quiet signals (the noise) in the same band.
Have a read at:
http://is.rice.edu/~welsh/elec431/psychoAcoustic.html
_________________________
Brian Badger
|
|
Top
|
|
|
|
|
#321234 - 01/12/04 10:17 PM
Re: RDAC quality analysis results.
|
Gurari
Planeteer
Registered: 03/27/03
Posts: 92
Loc: New York City
|
Thanks guys, much appreciated.
|
|
Top
|
|
|
|
|
#321235 - 01/13/04 12:16 AM
Re: RDAC quality analysis results.
|
Monkey
Planeteer
Registered: 11/13/02
Posts: 1873
Loc: Kansas City
|
First of all naysayer is a word meaning one who says nay or no.
Secondly i have a friend from Iceland and he tells me they still name thier children after thier parents. since my name is Mark my sons last name would become Markson. As I understand it a daughter would be Marksdaughter or the Icelandic equivilent of that anyway.
BTW thanks for all the info in testing of RDAC.
|
|
Top
|
|
|
|
|
#321236 - 01/13/04 12:56 AM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Originally posted by Monkey:
First of all naysayer is a word meaning one who says nay or no.
Yes, that's what I thought since "nay" is pronounced exactly (almost...) the same way as "nej" in Swedish... "Nejsägare" = "No-sayer"... The reason I asked was that "nay" sounds soooo Swedish!
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321237 - 01/13/04 04:09 AM
Re: RDAC quality analysis results.
|
Scoobie Doo
Planeteer
Registered: 12/01/00
Posts: 814
Loc: Miramichi, N.B. Canada
|
What about MT-2? Is it good enough to record in this mode? Most use MTP but I wonder if MT-2 adeqate?
_________________________
May Gods Peace Be with you!
|
|
Top
|
|
|
|
|
#321238 - 01/13/04 05:40 AM
Re: RDAC quality analysis results.
|
guraknugen
Planeteer
Registered: 10/04/02
Posts: 800
Loc: Norsborg (too far from Götebor...
|
Originally posted by Scoobie Doo:
What about MT-2?
Well, I used it every time when I only had the VS-880 and it wasn't so bad, but remember that MT2 is only 16 bits, MTP is 24. You might not hear that big difference, but if you are going to do a lot of processing, maybe after exporting tracks to wav for use on a PC, you might need the 24 bits. I guess the only reason for MT2 to be included at all is for compatibility with those older VS machines, like the 880 and 1680.
[ 01-12-2004: Message edited by: guraknugen ]
_________________________
Johnny Rosenberg (Andersson until 2007-07-31)
VS-2480HD (2001-09-25 – ∞)
VS-880VX (1996-03-27 – 2001-09-25)
|
|
Top
|
|
|
|
|
#321239 - 01/13/04 03:24 PM
Re: RDAC quality analysis results.
|
rwl6935
Planeteer
Registered: 01/01/01
Posts: 731
Loc: New Jersey
|
I gave myself a hear test one time using a sweep generator. I could definately hear up to 16.5 khz. I thought I heard 17khz, but it's probably more that I felt it (the pressure) rather than heard it. The right ear is alittle weaker than the left due to standing in front of a full stack all those years-
Bob
_________________________
rwl6935
|
|
Top
|
|
|
|
|
#321240 - 06/03/04 02:12 PM
Re: RDAC quality analysis results.
|
hannes
Space Cadet
Registered: 10/08/03
Posts: 10
Loc: Austria, Vienna
|
I came to the result: There is a herable difference between MTP and M24! For about one month I only record in M24. I first always recorded MTP and always recognised that the sound isn´t full enough - it´s hard to explain - I tried to make it "full" with turning the compressor on (each channel) but it wasn´t the solution.
Then I recorded for a project the accustic guitars in M24 and - wow. There was a room behind and a warmth and it was easy to mix and it was something like a strong hot sound.
Now I think I can describe: It is the dynamic.
In MPT the sound jumps from one level to the next, when it´s getting louder! There is always a kind of hard attack in the dynamic. You don´t hear the resonance as beautiful like in M24. If you record Overheads for drums you hear a big difference between MTP and M24. I also recorded live a suffi-band (Ahura project - many percussive instruments) twice. One time MTP and one time M24. It was much easier to mix in M24.
But there are solutions to increase the quality in MTP:
1) Send the master into valves (I´ve got dbx 386) they make it much warmer and bring a nicer dynamic
2) Give every channel a compression AND put the compressor-level as high as you can!
3) Track export: Make Wave-files of the tracks and import them into an M24-Song! Yes it sounds better!
4) Record with 48 kHz. instead of 44,1! Then again export tracks to Wave files and import in an 44,1 Song! (I never tried but somebody told me it works)
This is for MT1: I think it sounded better! I once recorded in MT1 and didn´t use the analog inputs. I used my dbx 386 via digital input. That sounded great! I think the MT1 has its compression in the Inputs!
_________________________
I english not good speak
|
|
Top
|
|
|
|
|
#321241 - 06/03/04 03:05 PM
Re: RDAC quality analysis results.
|
White Rabbit
Planeteer/Artist # 100
Planeteer
Registered: 08/04/00
Posts: 3828
Loc: Born in Scotland, but live in ...
|
2) Give every channel a compression AND put the compressor-level as high as you can!
|
|
Top
|
|
|
|
|
#321242 - 06/03/04 03:54 PM
Re: RDAC quality analysis results.
|
motown59
Planeteer/Artist # 238
Planeteer
Registered: 03/01/04
Posts: 17885
Loc: California
|
Cool info! thanks so much for the research.
as far as hearing above 10k, I think most of us can hear above that and there is such a thing as pshycoacoustics as stated above. we can feel things up to 16-20K, especially as overtones/harmonics in the sounds of instruments. Doug Sax and Robert Hadley have mastered a few things for me and they frequently add a few db @22k bell curve EQ on their desk. Doug says it's there as overtones and felt things, even if going to 16 bit CD. "The dog knob" as Robert said.
_________________________
kel
"I love what you guys are trying to do up there" ...from an audience member at one of my gigs.
Gear: Fender Medium pick
|
|
Top
|
|
|
|
|
#321243 - 06/03/04 04:27 PM
Re: RDAC quality analysis results.
|
kidfear
Planeteer
Registered: 12/14/01
Posts: 340
Loc: Bowling Green, KY
|
on more than one occasion, I've been tested as being able to hear 18k tones. the last time, the woman thought it was weird because she could tell i was unsure about some tones lower than 18k, but that i would dart my hand up everytime she played the 18k tone.
_________________________
***4 8 15 16 23 42***
The man in black fled across the desert, and the gunslinger followed.
|
|
Top
|
|
|
|
|
21370 Members
26 Forums
160092 Topics
1854143 Posts
Max Online: 386 @ 01/18/23 04:57 AM
|
|
|
