Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site dsd.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!umcp-cs!gymble!lll-crg!dual!amdcad!fortune!dsd!riner
From: riner@dsd.UUCP (john riner)
Newsgroups: net.audio
Subject: Re: Wanted:  cassette deck advice!
Message-ID: <417@dsd.UUCP>
Date: Tue, 5-Mar-85 14:17:25 EST
Article-I.D.: dsd.417
Posted: Tue Mar  5 14:17:25 1985
Date-Received: Sat, 9-Mar-85 19:24:40 EST
References: <4846@cbscc.UUCP> <> <316@cubsvax.UUCP> <8731@brl-tgr.ARPA>
Organization: AMPEX DSD, Redwood City, CA.
Lines: 63

> > 
> > Several years ago I read an article, though I forget where, in which a Nak
> > engineer claimed that Nak used to do something different from most other
> > cassette deck manufacturers -- I don't remember whether it was head
> > alignment,
> > bias, or what -- but that there is now a standard that all mfgrs. adhere to.
> > 
> I would expect that it is EQ.  Bias affects what is recorded on the tape
> and shouldn't affect the playback at all (i.e. if the bias is wrong it
> should sound equally bad on all decks).  I have a hard time believing
> that head alignment would do it either.  When you adjust head alignment
> you go for nice right angles and centered on the track.  When you adjust
> EQ, you have to pick some standard.  In many radio stations, they are free
> to pick whatever they want since they matching record and playback EQ's and
> never deal with any serious outside tapes.

	Years (many) ago recorders were set up by measuring the record head 
current and setting the bias and record currents to a set level which (due
to the limited performance achievable with the heads and tape then available)
was supposed to set the flux level on the tape. The reproduce equalization
was then set for the response/noise level desired.
	When incompatability became a problem (as mfrs and users began to
understand more and the heads and tape became better) standards were 
developed which specified a reproduce response curve which was fixed and
the record current, bias and equalization (yes there is record equalization)
were adjusted to make the frequency response and distortion meet the
system specifications. Several different schemes were developed to meet
various system performance requirements. Eventually the NAB developed a
set of standards for equalization and these are fairly well adhered to
today.
	The present method used is to reproduce a "standard" tape (a tape
made with a known flux level recording) and adjust the azimuth for peak output
at the highest frequency on the test tape then adjust the reproduce response
for flat output versus frequency. Then the bias is set for peak output
of a signal at a mid frequency (rule of thumb is to use the same frequency
in KHz as the tape speed in Inches per second up to 7.5 Khz at 7-1/2, 15,
and 30 ips) and then set the record level for the distortion level at
1KHz or so. The bias and record level are then "trimmed" to compromise
frequency response and distortion. Basically this applies to cassettes
as well as reel to reel recorders. 
	I have heard many comments on and off the net about Nak incompatability
and I have found no definative answer but here is what I believe is the
case. In order to optimize their product they used some record equalization
which was meant to operate with their heads (which from what I have been
able to deduce are very good) better than "standard" equalization. Since
their products are generally used with high performance tape (less distortion
at higher record levels), they could reduce the record level slightly at
higher frequencies and get wider response at higher record levels which
enhanced the signal to noise level and the spec sheet. This would produce a 
rolloff in the high end when played back on another mfrs tape deck. To
my knowledge this is no longer the case and any incompatability is probably
due to misalignment in on or the other of the machines.
	I don't know if this clears up any of the confusion (it probably
only creates more) but I remain convinced that the notion of standards
is only an illusion anyway (have you ever found a "standard" RS-232 
connection to anything)
	John Riner	AMPEX Corp.
-- 
	John Riner		UUCP: !fortune!dsd!riner
	AMPEX Corp
	Redwood City, CA.
		Nobody knows what I am talking about,
		so these must be my opinions and not theirs.