Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!utgpu!water!watmath!clyde!rutgers!husc6!cmcl2!nrl-cmf!ames!amdahl!oliveb!sun!pepper!cmcmanis From: cmcmanis@pepper.UUCP Newsgroups: comp.sys.amiga Subject: Re: That MAC II Monitor Message-ID: <33382@sun.uucp> Date: Mon, 9-Nov-87 16:07:26 EST Article-I.D.: sun.33382 Posted: Mon Nov 9 16:07:26 1987 Date-Received: Thu, 12-Nov-87 02:31:59 EST References: <253@PT.CS.CMU.EDU> <642@louie.udel.EDU> <267@gethen.UUCP> <33188@sun.uucp> <749@neoucom.UUCP> Sender: news@sun.uucp Reply-To: cmcmanis@sun.UUCP (Chuck McManis) Organization: Sun Microsystems, Mountain View Lines: 110 Keywords: scan rate, resolution, video once again [Another in the Chuck's Video tutorial series ... ] In article <749@neoucom.UUCP> wtm@neoucom.UUCP (Bill Mayhew) writes: Thanks Bill for posting the specs, here are some formulae one can use to determine the maximum resolution this monitor is capable of and the required parameters of the appropriate display interface. Note that it is the *computer* that determines the display resolution not the monitor. >Picture tube: >13 inch viewable diagonal >.25 mm aperature grille pitch >Trinitron (tm) CRT >Active display area: >235 millimeters horizontal by >176 millimeters vertical >(remainder of display >is used for border) These numbers give the physical limits to displaying pixels on the tube. Because the tube is not *really* a continuous surface, rather it is a bunch of phosphor dots that are excited by the electron gun(s). So dividing out the numbers horizontally, you can put on this tube with a .25 mm dot pitch (.25 mm between dot 'centers') 235 mm of active disply/.25 mm between pixels = 940 pixels per line. Vertically, you can divide 176 mm vertical active area by .25 mm dot pitch and come up is 704 lines. Unfortunately, it is not a perfect world so you will not see this sort of actual resolution out of this tube. The tube is driven by a bunch of electronics that control the electron guns inside, and the fastest they can respond is given by this next specification : >Video bandwidth: >23 MHz <---- please note If you looked at a picture that consisted of white and black dots you might not normally associate this with a frequency, but if you looked at the input to the electron gun that caused this pattern to be displayed on an oscilloscope you would see a 'square' wave. This is due to the relationship between voltage on the electron gun and the brightness of the spot it just hit on the screen. So changes in intensity on the screen are really just a time varying waveform to the electron gun. And the above specification says that this wave form cannot exceed 23 MHz. If it does, then the electron gun cannot respond fast enough and rather than black and white dots you only get grey dots. Now we can combine this value with the next one to find out what the maximum number of pixels the *electronics* can display. >Scan rates: >35.0 kilohertz horizontal This means the beam sweeps across the screen 35,000 times a second. Since we know from the above that the fastest the pixels can change is 23 million times a second the most you could put on a line before the beam moves on to the next line is 23,000,000 / 35,000 = 657.14 pixels. However, since the beam doesn't jump immediately to the next line, you have to allow some time for it to mave back to the beginning of the next scan line. This time varies on different monitors and is usually supplied in the extended specifications as Horizontal Retrace Time. Most monitors I've worked with have a retrace time in the range of 4 to 6 uS (micro seconds). The time the beam spends one line is 1 / 35,000 or about 286 uS. Using the worst case of 6 uS retrace time means 280 uS would be available for displaying pixels, and the ratio (280/286) = (657.14/x) yields 643.35 pixels that can be displayed. >66.7 hertz vertical This is also sometimes called the 'frame' rate since it is also the number of times per second that the entire 'frame' is displayed. Using this information we can calculate how many times the beam will trace out a line before the beam gets dragged back to the top of the display. Simply, (35,000 lines/second) / (66.7 frames/second) = 524.73 lines/frame. Again time has to be allowed for the beam to go from the bottom of the screen to the top which is usually specified as the Vertical Retrace Time. Since we don't have it here we have to pick 'reasonable' values pretty much out of the air. Due to the way monitors are built this number is often related to the horizontal retrace time. However because of the increased distance the beam has to travel it is a couple of orders of magnitude larger. So a monitor with a horizontal retrace time of 6 uS might have a vertical retrace time of 600 uS (.6 mille second) For now we will assume this number is a good estimate and calculate from there. Again, since the frame rate is 66.7 Hz, the time spent in one frame is 1/66.7 = .0149 second and subracting off the vertical retrace time gives us (.0149-0.0006) = .0143. Using the ratio technique again (.0143/.0149) = (524/x), x = 503. Or a maximum of 503 lines. >Resolution: >640 dots horizontally by >480 dots vertically So given that the monitor *can* display 643 X 503 pixels the designers of the Mac Display hardware *chose* to display 640 X 480 pixels. When deciding how many pixels to display other factors besides the monitor's capability come into play. Factors such as memory use, and the monitor's aspect ratio. 640 X 480 is a 4:3 aspect ratio which is the same aspect ratio of the tube, thus the pixels displayed are 'square' when the monitor controls are adjusted properly. So now when you look at the spec sheet for a monitor you can calculate for yourself what the maximum resolution that the monitor is likely to display. Since you can adjust the number of rows and columns displayed on your Amiga, if you know the specs for your monitor you can adjust the Amiga to display the largest possible display on it. Note however that the horizontal scan rate and frame rate are fixed at 15,750 Hz and 60 Hz respectively. --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.