Home » R/E/P » Dave Hecht - Master Tech » Analog and Digital Specs.
| Analog and Digital Specs. [message #341178] |
Tue, 13 May 2008 21:53  |
brendan james
Messages: 3
Registered: May 2008 |
Welcome |
|
|
A simple question that i cannot find the collective answer to. If i am in the wrong forum feel free to send me to the right one.
when recording onto analog tape, what is the frequency cut off (high and low) when recording at the following speeds:
-7 1/2".
-15".
I ask because- is recording at even a low speed like 7 1/2" ips going to give me a better quality frequency and harmonic response than recording onto digital.
|
|
|
| Re: Analog and Digital Specs. [message #341280 is a reply to message #341178 ] |
Wed, 14 May 2008 08:32  |
 ssltech
Messages: 1877
Registered: May 2004
Location: Florida (previously UK) |
Platinum Member |
|
|
Define 'better'. Everyone's version may be significantly different.
Define 'cutoff'. Unlike digital, there IS no sudden cutoff.
If you're asking "Is it going to sound better" then listen and decide for yourself... -There really is no other way.
Keith
| MDM (maxdimario) wrote on Fri, 16 November 2007 21:36 |
I have the feeling that I have more experience in my little finger than you do in your whole body about audio electronics..
|
|
|
|
| Re: Analog and Digital Specs. [message #341572 is a reply to message #341280 ] |
Wed, 14 May 2008 23:28 |
brendan james
Messages: 3
Registered: May 2008 |
Welcome |
|
|
thanks for the reply keith.
right you are. my preference upon listening to the two is for the analog recording. using the word 'better' in a question that was asking for a non subjective response wasn't right. my bad.
cd quality digital recordings are capped between 20 Hz to 20 kHz. which is "sufficient" for our hearing range.
my question is: does changing the ips to a lower speed restrict the "in-audible" frequencies and harmonics recorded onto the tape? does it have a rolling off effect? this is a concept that has been mentioned in forums. i wanted to know if anyone can qoute some figures and sources on this idea.
thanks
|
|
|
| Re: Analog and Digital Specs. [message #341632 is a reply to message #341178 ] |
Thu, 15 May 2008 08:55 |
drknob
Messages: 41
Registered: February 2005
Location: Montreal, QC |
Member |
|
|
In 'general', 7.5ips = good LF, lousy HF, tape noise, high distortion. 30ips = lousy LF, good HF, less tape noise, lower distortion. Many engineers preferred 15ips because it offered a compromise between the high noise of 7.5ips and the poor low end of 30ips. All other things being equal, bandwidth is a function of tape speed and head gap. A tweaked out ATR-100 running at 30ips can give you 20kHz at the top, but the low end is a bit of a bumpy mess. That's not to say it can't sound quite good, but it ain't linear amplitude or phase. Oh yeah, let's not forget wow & flutter.
It takes some knowledge and attention to squeeze performance out of the tape medium - critical alignment is essential at any speed. Tape machine maintenance used to be a full time job for some of us. Unless you're willing to take that on, I would stick to digital. A good A/D converter will always give you better linearity, less noise and lower distortion. Of course, if you like the 'sound' of tape....
Harold Kilianski
CIRMMT, McGill University
|
|
|
| Re: Analog and Digital Specs. [message #341644 is a reply to message #341178 ] |
Thu, 15 May 2008 10:04 |
ssltech
Messages: 1877
Registered: May 2004
Location: Florida (previously UK) |
Platinum Member |
|
|
| brendan james wrote |
cd quality digital recordings are capped between 20 Hz to 20 kHz.
|
Actually there is NO lower cutoff on CDs: -they can and will record AND reproduce down to DC. I think that you've possibly made an assumption there. There IS indeed an upper cutoff, since -due to the Nyquist principle- the highest reproducible frequency has to be just under HALF of the 44.1kHz sampling rate in order to prevent aliasing. So in order to keep the passband as flat as possible, the rolloff curve is sudden and precipitous.
Following are some 'crib notes' on analog stuff: It's a little simplified, but I'm not aiming for super-detail so much as a basic description of the general principles.
The issue with tape bandwidth is physical-wavelength related. The head rolls off HF slowly at first and then progressively more rapidly as the wavelengths get so short that a half-wave occupies the same length as the playback gap. (hence dedicated repro heads have as tiny a gap as possible). The lowest reproducible frequency is basically largely determined by the size of the pole-shoe on the playback head. (Note that -once again- you CAN record DC on an analog machine... it's just that a "coil-and gap" playback head can't REPRODUCE it.)
Now, on the subject of changing analog tape speeds:
Let's first consider low-frequency rolloff.
Slower speed makes all of the wavelengths shorter. at 30ips, a 30Hz cycle is one inch in length... (30 inches per second, 30 cycles per second, one cycle per inch... -easy math!) At 15ips, that cycle is only a half-inch long, at 7½ips, that same cycle is now only a quarter inch long, at three and three-quarter inches per second, it's down to an eighth-inch in length... EASILY short enough to reproduce by even a comparatively small pole-shoe repro head.
Now let's take a look at high-frequency rolloff.
At 30ips, a 30kHz wavelength is a thousandth of a inch. (thirty thousand cycles per second, thirty inches per second...that's one thousand cycles per inch, so each cycle is therefore a thousandth of an inch long). -That's getting to the limits of what a repro head can reasonably be expected to playback. At 15ips, 15kHz is a thousandth of an inch long. at 7½ips it's down to 7.5kHz, and so on.
mercifully, the rolloff isn't 'brickwall' like digital, and so curves can be played with and biasing amounts can be 'furkled and fettled' a little bit to play with the HF self-erasure characteristics as well as moderate the noise floor character... anyway the upshot is that the rolloff is gentler and forgiving, but at the root of it all is that there are physical factors which tend to give low speeds big, strong, warm bottom end and a soft HF rolloff, and give high speeds an extended, 'effortless' top end before starting to gracefully diminish well beyond the spectrum (given a nice, high-quality repro head) but make them struggle to keep the bass end from falling off.
Summary: High speeds clear and pristine at the expense of some sonic 'strength', low speeds strong and mighty at the expense of rather more noise and a loss of some 'clarity'.
If you're recording an orchestral tune which drops suddenly from a tutti fortissimo with all the strings digging in with the bow and the brass players belting it out, to a subito pianissimo solo flute, then pretty much ANY halfway decent digital system is going to be judged to be 'better' (meaning more acceptable on the whole) than a quarter-track machine running at 7½ips because the tape hiss is going to be louder than the music at times, but for heavy, multiband-limited powerpop, the judgment could just as easily go the other way, if people have a 'warm and fuzzy' attachment for "warm and fuzzy" sound.
Keith
| MDM (maxdimario) wrote on Fri, 16 November 2007 21:36 |
I have the feeling that I have more experience in my little finger than you do in your whole body about audio electronics..
|
|
|
|
| Re: Analog and Digital Specs. [message #341664 is a reply to message #341178 ] |
Thu, 15 May 2008 11:01 |
amorris
Messages: 495
Registered: June 2004
Location: florida |
Active Member |
|
|
as far as any "cutoff" freq of analog tape, I just reworked a dozen or so analog 1/4 2 tracks, Otari MTR12's 30 ips and was showing the instructor typical roll off and every machine had a -3db roll off at about 29khz. now that is where the roll of starts. 33khz was down more but was still present. and so on up the scale. at low freqs the roll off at 30 ips was in the 40 hz area, now, 10 hz was still there, just down. and doing lf adjustments, there is a natural (or un-natural) lf bump that is very narrow but very prominent, so usually the bump is found and lf adjusted for a + .5 db bump. ~50hz for 15ips and 90 hz for 30. so linearity is a compromise with analog. digital is inherently more linear, but bandwidth compromised. So, as an engineer, many compromises are made and the best all-around recording is sought after. never will a perfect recording happen, so perfect is not an option.
|
|
| |
| Re: Analog and Digital Specs. [message #341732 is a reply to message #341673 ] |
Thu, 15 May 2008 14:53 |
ssltech
Messages: 1877
Registered: May 2004
Location: Florida (previously UK) |
Platinum Member |
|
|
You and me both Jay.
however about 4 years ago I had to work for an engineer who got sounds that -to my ears- were just AWFUL...
..until about 2 days in, when he finally rolled tape. -All the way down with the red lights on, it sounded nasty. -Then he wound back and it sounded...
...well... -it sounded like a record!
Now if I had been on the session it wouldn't have been HALF as nice, because I don't KNOW how to do it. -I don't KNOW how much to 'pre-bend' sound so that the tape "bends it back straight again"... -It definitely takes experience. -And this guy bent the needles round so hard that they "read the maker's name" most of the time! (all three red peak lights lit up on ALL the active meters on ALL THREE studer A827's!!!)
But it takes all sorts I suppose. -For what it's worth, the record took a whole year to record, with some of the greatest session cats on it, and it's staying in the can. -I bet it never comes out... -I think the artist has 'moved on' and isn't interested any more, but it was VERY interesting seeing it all get done like it was back in the 1970's again... but this time with a lot more tracks! (with 56 channels of Dolby 'A' no less!)
Keith
| MDM (maxdimario) wrote on Fri, 16 November 2007 21:36 |
I have the feeling that I have more experience in my little finger than you do in your whole body about audio electronics..
|
|
|
|
| Re: Analog and Digital Specs. [message #341925 is a reply to message #341178 ] |
Fri, 16 May 2008 06:16 |
brendan james
Messages: 3
Registered: May 2008 |
Welcome |
|
|
thats great info. exactly what i was after.
thanks. you each put a piece of the puzzle in for me.
brendan.
|
|
|
Goto Forum:
Current Time: Sun Jul 6 00:40:23 CDT 2008
Total time taken to generate the page: 0.05442 seconds |
Recording Engineering & Production: The nuts and bolts of making records better, not just louder and cheaper. Some of the brightest minds from the creative and scientific world of audio discuss the elements of audio production that interest you most.
Members
Search
FAQ
Register
Login
Home
