Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.csd.uwm.edu!cs.utexas.edu!sun-barr!sun!pepper!cmcmanis
From: cmcmanis%pepper@Sun.COM (Chuck McManis)
Newsgroups: comp.sys.amiga
Subject: Re: Interlace and Multiscan Moniters
Keywords: Interlace Multiscan Multisync
Message-ID: <122510@sun.Eng.Sun.COM>
Date: 18 Aug 89 19:29:37 GMT
References: <4321@tekig4.LEN.TEK.COM>
Sender: news@sun.Eng.Sun.COM
Reply-To: cmcmanis@sun.UUCP (Chuck McManis)
Organization: Sun Microsystems, Mountain View
Lines: 64

In article <4321@tekig4.LEN.TEK.COM> (Mike Larson) writes:
> This multiscan performed flawlessly when the Amiga was displaying a
> non-interlaced screen. An interlaced screen, however, showed an 
> incredible amount of flicker, much worse than my 1080.  Upon closer 
> inspection of the CRT I realized that the odd and even frames were 
> being SUPERIMPOSED on each other rather than being displaced by half 
> of a scan line.

>Question: do any multiscan moniters correctly handle interlace, or is it
>just this particular brand that fails?  Or am I doing something else wrong?

And the answer is "not officially." To understand why this is you have to 
understand a bit about how interlacing is accomplished and how the multiscan
feature is accomplished. First interlace :

	The video beam on a CRT travels to the right horizontally, and when
	a "line" time has passed or HSYNC occurs, it moves to the left and
	down by one line thickness. The "line" time is the monitors "scan
	rate." And can be calculated by multiplying the number of lines on
	the screen by the screen refresh rate. For NTSC that is 262.5 *
	60 = 15750 Hz. If one of the first HSYNCs  near the top of frame
	is half as long as normal. Then the beam will only be halfway 
	across the screen and when it returns to the left, because it only 
	goes half as far, it only goes "down" a half a line thickness. 
	And since subsequent HSYNCs occur at the right time (a full line
	width) all subsequent lines are display "offset" by half a line
	thickness, and that makes them appear between the lines that are
	currently on the screen.

And now MultiSyncs :

	Before the NEC multisync, monitors were "hard coded" with the 
	scan rate, and computer hardware matched themselves to the 
	appropriate monitor. However, NEC noticed that many programmable
	CRT controllers were appearing on the market, and these controllers
	could change their video parameters under program control. The
	main reason for doing this was to change the resolution of the
	image they were displaying. And people didn't want to buy a monitor
	for hi-res mono text, and one for lo-res multicolor text. So they
	invented the original MultiSync. The magic was that this monitor
	could change it's internal scan rate *on the fly* and adapt without
	buttons or switches to the kind of video. But how could it know 
	what frequency to set itself to? Well, since the time between 2
	HSYNCS is defined by the scan rate frequency, the tube simply
	monitored that time and calculated what frequency it should be at.
	(This sounds tough, but in fact a simple Phase Locked Loop circuit
	allowed it to identify the proper frequency fairly quickly.) So
	what happens when you send it a short HSYNC so that the beam will
	move halfway down ? The monitor detects this as a change in the 
	scan rate, but notices that it only happened for one HSYNC and
	so writes it off as a glitch. In the mean time *it* keeps that
	video beam right in sync. Poof, overlayed video not interlaced
	video. 

So the moral of the story is, either buy a multiscan monitor that says
"Supports Interlaced video" or try it out on a machine at the store before
you buy it to make sure it doesn't "help you out" by keeping all of those
video lines, lined right up for you.


--Chuck McManis
uucp: {anywhere}!sun!cmcmanis   BIX: cmcmanis  ARPAnet: cmcmanis@sun.com
These opinions are my own and no one elses, but you knew that didn't you.
"A most excellent barbarian ... Genghis Kahn!"