|
Richard Andrew x64 wrote: if modern displays had a vector mode
Honestly, from a certain perspective that is true. The data passed between the CPU, the video adapter, and even the monitor is rarely simple raster lines of pixels. Higher-level constructs are used throughout the pipeline to reduce the amount of information required per frame, which helps increase the overall frame rate possible with the device.
The other appeal of the original vector displays was the lack of pixelation in the image. Frankly, modern displays have such incredible pixel density that pixelation is only an issue in certain contrived scenarios.
Software Zen: delete this;
|
|
|
|
|
I agree with that. Vector displays definitely have a certain appeal.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
Vectors are used heavily by illustrators in packages such as Adobe Illustrator. They're used on the web in SVG's.
|
|
|
|
|
I guess what you mean is that they are used in the mathematical sense to generate the shapes, but vector display hardware is no more, according to CodeWraith.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
|
WPF, WInUI, etc... all use vector graphics..
Also, inspired by work, I started using SVG for the icon of my own home app.
modified 29-Jan-22 8:31am.
|
|
|
|
|
I guess I was referring to the kind of graphics that is characterized by shapes being made up of lines only.
CodeWraith said that vector display hardware no longer exists.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
If you are thinking of true, continuous and smooth lines on the screen with no hardware-induced limited resolution, like the old Tektronix storage tube displays: You couldn't make a color display with that technology. In theory, you could make stripes of R, B, G phosphor, with a shadow mask just like on plain color CRTs, but that would introduce a resolution limit; lines would not be continuous and smooth on the screen.
I will not rule out that there existed some sort of continuous line color screens, but I never ever saw any such thing out in the marketplace. If it existed, it seems to never have escaped from the development lab.
In the really old PC screens, the display driver was responsible for generating the signals to the coils do the horizontal and vertical deflection of the electron beam. While a proper driver would scan the screen line by line, in principle it could let the beam trace any path, drawing "true" vector lines, in the same pattern as you would on a Tektronix screen. Note that the Tektronix had a mechanism for keeping the phosphor "lit", once lit, until the (entire) screen was erased; no refresh was required. If you wanted to emulate X/Y plotting on a standard PC screen, you would have to refresh the figure at at frequency of at least 30 times a second. (Even 30 Hz gives significant visible flicker.) If you try this on a color screen, the mask would of course limit the resolution.
Lots of systems/standards specify representation of figures (both wireframe and surface), and processing of them, in a continuous, resolution independent vector format. Figures often remain in that format all the way into the GPU; lines are not digitized (pixelized) until immediately before the signals go out on the cable to the display. As long as the pixelized signal is created according to the true physical display resolution, you get the best possible image from that screen.
Side remark / old memory:
Deep down in my archives is a copy of one of the first issues of "Dr. Dobb's journal". At that time (second half of the 1970s) microcomputer owners didn't expect very much. This DIY article about "How to build a true 3D display" looked quite serious, in the opening paragraphs, describing how you could "draw" a line, like on Tektronix screens, with a LED that could be moved left and right, up and down, by small motors pulling the LED along a horizontal and a vertical rail. If you furthermore made a front-back rail, the LED could trace a wireframe figure in the cubic space delimited by the movements of the the three rails.
Of course the LED would have to retrace the wireframe figure at a high frequency. To simplify the mechanics, rather than a single LED, the horizontal rail could consist of a dense row of LEDs, to replace the horizontal movement. The next step would be to build a pile of such horizontal rails, forming a dense matrix of LEDs, to avoid the vertical movement as well.
Connecting the LED matrix to the computer would require thousands of wires, and a complex control logic. But we've got a matrix of lights that is far easier to handle, if we replace the LED mesh with a CRT. So all it takes to make a true 3D display is to build a mechanism for flipping a CRT back and forth 30 times a second ...
In those days, people built the craziest things for their micros, and the article was well written, so you could read half of it before realizing that it was all a big joke from the very beginning. Maybe I should dig up that old magazine and copy the article to CP. (It would probably be thrown out due to copyright infringement, even if I stated the source in all details.)
|
|
|
|
|
trønderen wrote: but I never ever saw any such thing out in the marketplace.
Are you serious? Look up the old arcade game Tempest. That had red, green and blue objects.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
But were the lines continous, with no breaks caused by as shadow mask?
If they were: Can you provide a reference to a description of the technology, how they did it?
|
|
|
|
|
I'll try looking it up a little later and post back if I find something.
But as I recall, the lines were continuous.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
trønderen wrote: In the really old PC screens, the display driver was responsible for generating the signals to the coils do the horizontal and vertical deflection of the electron beam. While a proper driver would scan the screen line by line, in principle it could let the beam trace any path, drawing "true" vector lines, in the same pattern as you would on a Tektronix screen.
Not entirely accurate. You could do that only on CRTs with electrostatic deflection. The more common ones used as PC monitors had electromagnetic deflection with those big coils installed on the neck of the CRT. Trying to move the beam randomly on those was impossible due to the high induced voltage. For regular scans (CRTs and TVs) the horizontal fly-back was used to produce the high voltage for the secondary anode.
The electrostatic CRTs (like the ones used in scopes) had the drawback of a limited deflection angle, at least for any pretense of linearity and that made those very long necks.
Mircea
|
|
|
|
|
I am old enough to remember Vector Graphics, the PC's and the company. I was a dealer until they went out of business.
|
|
|
|
|
Oh wow. Do you recall a home entertainment device named Vectrix? I think it was a game console with its own CRT display.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
|
Shirley, you must be joking!
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
If you remember how they were drawn then you will likely understand why vector graphics displays no longer exist. Originally, the vectors actually aimed the electron gun to draw the image. Instead of left-to-right, one line at a time drawing as a traditional CRT did, vector graphics displays actually aimed the beam to draw each of its lines. Since we no longer have CRT displays manufactured in any significant number, if at all, that just doesn't happen any more. As mentioned, the mathematical constructs are still with us but image drawing is done in a completely different way.
"They have a consciousness, they have a life, they have a soul! Damn you! Let the rabbits wear glasses! Save our brothers! Can I get an amen?"
|
|
|
|
|
"Real" vector graphics only work on (special) CRT displays (and other hypothetical beam-steered displays), the lines are never converted to pixels, they're drawn with the screen itself. If drawing lines on a grid of pixels is also "vector graphics", then you're looking at it right now.
|
|
|
|
|
after 8 years, I dumped current modem and bought a new one.
it is ARRIS G34 with WIFI router. I need faster Internet service now: download speed up to 1Gbs...
diligent hands rule....
|
|
|
|
|
Congrats! Can your internet service supply such a high download speed?
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
yes, absolutely. My monthly fee keeps going up and I called the internet service company.
if I upgrade my internet service package, I can lower my cost...
diligent hands rule....
|
|
|
|
|
Southmountain wrote: if I upgrade my internet service package, I can lower my cost...
Beware of those kinds of deals. I bought such a deal, and when the promotional period expired, my price went up tremendously.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
it's their marketing trick. I have to call them every year, otherwise my fee keeps going up.
also I need faster service to download financial data for my data mining application.
diligent hands rule....
|
|
|
|
|
With XFinity at least (US) you can negotiate with them when this happens. Make sure you have an alternative service to jump to and then tell them you will switch. They'll extend the "promotion"
You can do this every year or so.
Real programmers use butterflies
|
|
|
|
|
Good to know, thanks!
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
