Category Archives: CoCo

Tandy/Radio Shack TRS-80 Color Computer (CoCo)

What you PSET is not always what you PPOINT

3 Replies

Let’s consider this a part 2 for this 2024 post:

The 9 colors of the CoCo’s high resolution screens…

Read that first, then come back. I’ll wait.

Ready… PSET… GO!

In Microsoft Extended Color BASIC on the Radio Shack Color Computer, high resolution graphics command were added. There were fancy commands like DRAW for drawing complex designs, then simpler command like LINE and CIRCLE.

Even simpler was the PSET command which would set an individual pixel on the screen. This was the high resolution equivalent of SET in Color BASIC for the text screen (64×32 blocks).

I had a Commodore VIC-20 at the time, but remember getting a call from my Radio Shack sales guy, Don, to tell me they had just gotten in a new Extended BASIC that I should come out. I did, and fell instantly in love with being able to do things with simple commands versus confusing POKEs on my VIC.

I read through the manual in the store, and created programs on their CoCo. One of them used PSET to randomly place dots on the screen:

0 'RNDPSET.BAS
10 PMODE 3,1:PCLS:SCREEN 1,1
20 PSET(RND(256)-1,RND(192)-1,RND(4)-1)
30 GOTO 20

Oh the fun we had back then.

Use the SCREEN 1,0 color set and you get different colors:

The PMODE 4 screen (256×192) was only two colors, white and black, but TV set (composite video, NTSC) created artifact colors that made it look weird. The emulators try to simulate this effect:

Depending on the mode the CoCo power up, even column pixels would be red and odd would be blue, or the reverse. While the CoCo 3’s video hardware could control this, on the CoCo 1 and 2 it was “random.” You could keep hitting the rest button until the colors flipped. Thus, many CoCo 1/2 games started up to a solid red (or blue) screen, expecting you to know what to do to get it the color the game wanted, or, if the programmer was fancy, it might even tell you what to do:

Paper Route by Diecom

And if the programmer was really fancy they could just handle it in software based on what you told it you saw (“Press 1 if the screen is RED, 2 if it is blue” kind of thing.)

NOTE: I wanted to include a screenshot of this but I cannot remember which games worked like that. If you know, leave a comment and I’ll update this post.

Get to the PPOINT!

PSET would put a pixel on the screen using a specific color:

The syntax is:

PSET (X, Y, C)

The C was the color, and it accepted a value of 0-8. But, you did not get nine colors on any of the high resolution screens. You got either 2 (black/green or black/buff or using the alternate color set), or 4 (green/yellow/blue/red or buff/cyan/magenta/orange using the alternate color set). You had to know the range of the 4 color values to use for the mode you were using.

Or did you?

No. The C value could be 0-8 on any graphics screen, so most programs I saw used 0-1 for two color, or 0-3 for four color. And that’s not at all how Radio Shack described it in the manual… but it worked and kept us from having to memory color ranges based on modes. (See my 2024 article, link at the top of this one, for a table showing how this worked.)

And my point is … PPOINT returns the value that “should” be used — not the value you PSET there! If you were using a screen that wanted colors 5-6-7-8 and you used colors 1-2-3-4, you could PSET color 1, but when you would PPOINT that pixel, you got back a 5 — the same color, but the value you were supposed to be using.

Thus, anyone who made use of PPOINT learned this. I did not use it, and I never learned it until 2024 in the comments to that earlier post.

So for fun, I wrote this program that cycles through each PMODE (0 to 4) and then PSETs each color value (0-8) to a pixel and then reads the pixel color back using PPOINT. It prints it out so we can see this:

The first value of the column is the C value that was used in PSET. The number after it is the return value from PPOINT. So “1) 5” means PSET(0,0,5) and P=PPOINT(0,0).

PMODE 0, PMODE 2 and PMODE 4 are a two color modes, so the range of colors you can use (0-8) are just the same two colors over and over – 0 and 5. PMODE 1 and PMODE 3 are four color modes, so you see the range repeating the same four color values over and over, different depending on the mode.

And, if you used the alternate color set (SCREEN 1,1), you got a different set of color values:

Here is the code:

10 'PPOINT0.BAS
20 'SET PIXEL USING COLOR
30 'THEN PPOINT THE COLOR
40 CLS
50 PRINT@6,"PMODES - SCREEN 1,0:"
60 PRINT STRING$(32,"-");
70 FOR M=0 TO 4
80 PMODE M,1:PCLS:SCREEN 1,0
90 PRINT:PRINT@64+M*7+1,M;
100 FOR C=0 TO 8
110 GOSUB 160
120 NEXT:NEXT

130 GOTO 130

140 'SET PIXEL WITH PSET
150 'GET PIXEL COLOR
160 PSET(0,0,C):P=PPOINT(0,0)
170 PRINT@96+M*7+32*C,USING "#) #";C;P;
180 RETURN

And slight changes for the other color set:

10 'PPOINT1.BAS
20 'SET PIXEL USING COLOR
30 'THEN PPOINT THE COLOR
40 CLS
50 PRINT@6,"PMODES - SCREEN 1,1:"
60 PRINT STRING$(32,"-");
70 FOR M=0 TO 4
80 PMODE M,1:PCLS:SCREEN 1,1
90 PRINT:PRINT@64+M*7+1,M;
100 FOR C=0 TO 8
110 GOSUB 160
120 NEXT:NEXT

130 GOTO 130

140 'SET PIXEL WITH PSET
150 'GET PIXEL COLOR
160 PSET(0,0,C):P=PPOINT(0,0)
170 PRINT@96+M*7+32*C,USING "#) #";C;P;
180 RETURN

It dawns on me now that a BASIC program could detect what screen was being used by PSETting something and then reading it back with PPOINT. Based on the number that returned, you could tell which PMODE and SCREEN was being used.

But that’s a program for a different time.

Until then…

A slightly faster MID$ using VARPTR

2 Replies

TL:DNR – This is stupid. Don’t bother trying it :)

Here is another stupid thing I had to try in Color BASIC. See also my Stupid VARPTR tricks post.

Suppose you want to horizontally scroll a string across the CoCo’s 32-column text screen. You can build a long string (up to 255 characters) and then use a FOR/NEXT loop and MID$ to print 32 characters from within that longer string.

20 CLS:CLEAR 510:DIM A$
30 A$="":FORA=32TO255:A$=A$+CHR$(A):NEXT:A$=A$+STRING$(31,128)
40 TIMER=0
50 FORA=1TO224:PRINT@256,MID$(A$,A,32);:NEXT
60 PRINTTIMER

On a CoCo 3, you get a timer result of around 330-335. On a CoCo 1/2, it is slightly faster showing around 295. (There are less BASIC keyword tokens to look for so Color BASIC is faster than Exended BASIC which is faster than Super BASIC.)

When I first started benchmarking BASIC here years ago, I was using an emulator that only ran CoCo 1/2, and it never occurred to me it would be different on the CoCo 3. #TheMoreYouKnow

As I have discussed in earlier string theory articles (quite a few relating to this topic), MID$, LEFT$ and RIGHT$ all need to create a new string containing data from the original. Suppose you have this:

A$="THIS IS A TEST"

If you trim the string to the first four characters, like this:

A$=LEFT$(A$,4)

…BASIC is basically doing this:

  1. Allocate new string space for a 4 character string.
  2. Copy the first four characters from A$ into the new string.
  3. Make A$ point to the new string, which leaves the original A$ string memory available for cleanup.

This is why you need extra string memory to use these functions. If you clear JUST enough memory to hold a ten character string, you can’t do any MID$, LEFT$ or RIGHT$ with that string.

In order to make that work, you need at least 4 extra bytes reserved for string memory so LEFT$ can make the new string before deallocating the original. And if you add 4 bytes but try to use 5, you get that ?OS ERROR (out of string space) still.

So if you have a 255 character string and you use MID$(A$,0,32) to get 32 characters from that string into a new string (or use it directly with PRINT, which will directly access that string memory without a variable needed), it has to allocate 32 bytes from the reserved string memory and copy those 32 bytes over, then you can use it (PRINT or assign to a variable or whatever).

As my sample program runs, it is allocating 32 byte strings and copying 32 bytes over each time through the loop.

VARPTR, can you help?

After recently learning a weird trick involving changing where a string variable points to for its string data, that made me wonder if you could just manipulate the string address using VARPTR and then PRINT it, and be faster than the overhead of MID$.

I just had to try this…

Here is my test program. It first builds A$ to contain character 32 (space) to character 255 (end of semigraphics characters) and then pads that with 31 black blocks to be the full 255 character string size.

Starting at line 40 is a loop that will PRINT a substring of 32 characters in a loop, starting at the first character and going to the end, making the line scroll to the left.

When done, it prints how many TIMER ticks it took to do this.

Next, starting in line 70, it makes a new B$ that is a clone of A$. (No, it doesn’t need to clone it. I could have just manipulated A$ and saved a few lines.)

It then gets the address of that string descriptor using VARPTR, and saves the size (at the first byte) and the address of the string data (third and fourth byte). This allows restoring the string after the test, since manipulating strings this way could cause problems with strings and garbage collection.

For this specific example I know that the MSB and LSB of the new string are 126 and 1 on my emulated CoCo. That means I could make the string’s starting address be one higher in that memory by incrementing the second value. Had that second value been 250, once I incremented it 5 times to 255, I’d have to increment the first one and reset the second one to 0. (Clear as mud?)

Let’s just say I got lucky enough for this test, but even if this approach DID work, it would not work on all strings based on where they are in memory. It would need extra code for that, which would slow it down even further.

I POKE to change the size of B$ to 32. Then I can go in a loop POKEing the second byte of the start address up by one each time, and simply PRINTing the string. It will print 32 bytes from wherever it thinks the string data begins. It has no idea those values are being manipulated.

When done, I print the time it took and restore the original B$ values back.

10 'VARMID.BAS
20 CLS:CLEAR 510:DIM A$
30 A$="":FORA=32TO255:A$=A$+CHR$(A):NEXT:A$=A$+STRING$(31,128)
40 TIMER=0
50 FORA=1TO224:PRINT@256,MID$(A$,A,32);:NEXT
60 PRINTTIMER
70 DIM B$,V,O2,O3,S1,S2,SS
80 ' CLONE A$. GET VARPTR OF B$.
90 ' V POINTS TO SIZE
100 ' O2-O3 POINTS TO START
110 B$=A$:V=VARPTR(B$):O2=V+2:O3=V+3
120 ' SS - SAVE SIZE
130 ' S1-S2 - SAVE START
140 SS=PEEK(V):S1=PEEK(O2):S2=PEEK(O3)
150 ' CHANGE SIZE TO 32
160 POKE V,32
170 ' CHANGE START ADDR THEN
180 ' PRINT.
190 TIMER=0
200 FORA=0TO223:POKEO3,S2+A:PRINT@256,B$;:NEXT
210 PRINTTIMER
220 ' RESTORE ORIGINAL STRING
230 POKE V,SS:POKE O2,S1:POKE O3,S2

Sadly, it is only a tiny bit faster, and if I had to take care of both address bytes in the VARPTR memory, it would be slower and, without testing this, I assume slower than just using MID$.

It was a fun experiment, but “sometimes the juice is not worth the squeeze.”

Until next time…

The CHANGE command on PDP8 BASIC is weird. And neat.

1 Reply

Recently, Rick “Shanghai” Adams pointed me to a BASIC program he was working on for a PDP-8. Much like I now spend my time playing with BASIC from 1980 on a CoCo 1 emulator, Rick is going even further back and playing with a version of BASIC from a machine created in the 1960s.

And its a bit different.

I have previously posted about this machine (well, PDP systems in general) when I was trying to find the origins of the INSTR command in BASIC. Then I was able to dig up a 1971 PDP-11 manual that referenced this function for that BASIC.

Rick sent me a link to a two-player game called CHOMP. I was unfamiliar with it, but see it has its own wiki page.

https://en.wikipedia.org/wiki/Chomp

On that entry, it is presented as a chocolate bar made up of squares you can break off. The top left square is poison, so whoever takes that final piece loses.

By Lord Belbury – Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=86379139

This helps me understand why the came is called CHOMP.

But I digress… This really has nothing to do with that game, though CHOMP will likely return in a future blog post here.

BASIC may have been cooler in the 1960s…

Rick has a BASIC preprocessor that allows him to write easy to read source code, then convert it into messy BASIC code with line numbers later. Here is a Guest The Number game from Rick’s GitHub:

https://github.com/yggdrasilradio/b8pp/blob/master/guess.txt

	' Generate random number
	randomize
	x = int(100 * rnd(0)) + 1

	' Intro
	print "I'VE THOUGHT OF A NUMBER BETWEEN 1 AND 100."
	print "CAN YOU GUESS WHAT IT IS?"

	n = 0
60	n = n + 1
	print
	print "YOUR GUESS";
	input g
	if g <> x then 130

	' Player wins!
	print
	print "YOU GUESSED IT IN"; n; "GUESSES!"

	if n <= 7 then 120
	print "BUT IT SHOULD HAVE ONLY TAKEN YOU 7 GUESSES."

120	stop

130	if g > x then 160
	print "TOO LOW."
	goto 60

160	print "TOO HIGH."
	goto 60

	end

That text is processed by his script and becomes this:

https://github.com/yggdrasilradio/b8pp/blob/master/guess.bas

1RANDOM\X=INT(100*RND(0))+1
2PRINT"I'VE THOUGHT OF A NUMBER BETWEEN 1 AND 100."
3PRINT"CAN YOU GUESS WHAT IT IS?"\N=0
60N=N+1\PRINT\PRINT"YOUR GUESS";\INPUTG\IFG<>XTHEN130\PRINT
61PRINT"YOU GUESSED IT IN";N;"GUESSES!"\IFN<=7THEN120
62PRINT"BUT IT SHOULD HAVE ONLY TAKEN YOU 7 GUESSES."
120STOP
130IFG>XTHEN160\PRINT"TOO LOW."\GOTO60
160PRINT"TOO HIGH."\GOTO60\END

…and that is the code that he loads and runs on the PDP emulator.

The first thing you will notice is the use of a backslash. Initially, I thought this was just in place of a colon. For example, I see things like:

PRINT "TOO HIGH"\GOTO 60

…and that looks like it is just…

PRINT "TOO HIGH":GOTO 60

…but once I started converting PDP CHOMP to work on the CoCo, I realized that was not the intent of the backslash there. It appears to be a way to have multiple lines without line numbers. The logic does not continue to flow through those slashes:

60N=N+1\PRINT\PRINT"YOUR GUESS";\INPUTG\IFG<>XTHEN130\PRINT

Above, if G is not equal to X, it goes to line 130. There is a PRINT after that which would NEVER be reached if these were just colons:

60N=N+1:PRINT:PRINT"YOUR GUESS";:INPUTG:IFG<>XTHEN130:PRINT

To port this to Microsoft Color BASIC, I would have to do something like this:

60 N=N+1:PRINT:PRINT"YOUR GUESS";:INPUTG:IFG<>XTHEN130
61 PRINT

And guess what? That is also not important to this blog post. But might be later to a follow up blog post about porting PDP BASIC CHOMP to CoCo BASIC CHOMP.

CHANGE is … good?

Rick has entered the Logiker Christmas Challenge in the past with entries done in this much earlier BASIC. I have seen him comment on the far more limited STRING handling in that version of BASIC which required some very different approaches to the challenges.

Now I understand what he means.

While looking at his CHOMP source code, there is a keyword I had never seen before – CHANGE. He explained it to me:

“The CHANGE statement converts strings to and from an array, and the zeroth element of the array is the length of the string”

– Rick Adams

I saw it used like this:

40	print
	print "YOUR MOVE";
	input m$
	change m$ to m

Or, in the converted BASIC:

40PRINT\PRINT"YOUR MOVE";\INPUTM$\CHANGEM$TOM

So if you did something like this:

M$="ABC"
CHANGE M$ TO M

…then you would get this:

FOR I=0 TO M(0):PRINT M(I):NEXT
3
65
66
67
  • M(0) is the length of the converted string.
  • M(1) is the first character of the string (“A”)
  • M(2) is the second character of the string (“B)”
  • M(3) is the third character of the string (“C”)

That’s kinda neat. To do the same thing on Color BASIC, we could make subroutines like this:

2000 'M TO M$
2010 M$="":FORZ=1TOM(0):M$=M$+CHR$(M(Z)):NEXT:RETURN

3000 'M$ TO M
3010 M(0)=LEN(M$):FORZ=1TOM(0):M(Z)=ASC(MID$(M$,Z)):NEXT:RETURN

And then we could change “CHANGE M$ TO M” to “GOSUB 3000” and “CHANGE M TO M$” to GOSUB 2000.

10 INPUT M$:GOSUB 3000
20 FOR I=0 TO M(0):PRINT M(I):NEXT:END

Or reverse it, if we created one manually:

10 M(0)=3:M(1)=65:M(2)=66:M(3)=67:GOSUB 2000
20 PRINT M$:END

Though Rick pointed out with MID$() and such available, there would be far easier ways to do this. But those subroutines are what I will be using so I can more directly run his PDP BASIC code with minimal changes.

Have you ever worked on a BASIC that used the CHANGE command? If so, what was it? Leave a comment, please.

Until next time…

Revisiting PUT defined by ASCII strings

Leave a reply

This is a followup (and correction) to my PUT defined by ASCII strings article. That article, itself, was a followup to my Extended Color BASIC PUT from DATA article. They both discuss a method to use the GET/PUT graphics commands with pre-loaded data to PUT on the screen, rather than having to render/draw something first and then GET it so you could PUT it later.

I post such articles for two main reasons:

  1. So I can learn from all the much-smarter programmers who leave comments with advice/suggestions/corrections.
  2. So I can find it again when I’ve forgotten how it worked.

#2 is what led me to my article as I wanted to use the code for another experiment. And that’s when I realized my original PUTDATA.BAS program had a bug: it PUTs the tanks flipped vertically!

The program defined 8×8 characters as ASCII strings. I used a tank with four versions – one for each direction. They looked like this:

580 'TANK UP
590 DATA "   XX   "
600 DATA "   XX   "
610 DATA "XX XX XX"
620 DATA "XXXXXXXX"
630 DATA "XXXXXXXX"
640 DATA "XXXXXXXX"
650 DATA "XXXXXXXX"
660 DATA "XX    XX"

A routine READs those strings from the DATA statements, then walks through the characters looking for an “X”. If it finds one, it sets a bit in a number variable. At the end of the scan, it now has a byte value which can be loaded into the PUT data.

However, my routine to scan the strings does it backwards and the image is reversed left-to-right:

380 '
390 'READ DATA AND POKE AT L
400 '
410 PRINT
420 'READ STRINGS AND CONVERT
430 'TO BYTES.
440 FOR Z=0 TO 7:V=0
450 READ Z$:PRINT Z$,;
460 FOR BT=1 TO 8
470 'FASTER WAY
480 IF MID$(Z$,BT,1)="X" THEN V=V+BT(BT-1)
490 'SLOW WAY
500 'IF MID$(Z$,BT,1)="X" THEN V=V+2^(BT-1)
510 NEXT:PRINT INT(V);HEX$(V)
520 POKE L+Z,(NOT V)+256:NEXT:RETURN
530 NEXT:PRINT V;HEX$(V):NEXT
540 RETURN

I did not notice this until I tried to use this code for a new experiment, and it was not PUTting what I expected. But for the tanks, being symmetrical, I did not catch it in my original demo. The UP and DOWN tanks looked correct, and I failed to notice my LEFT and RIGHT tanks were reversed.

The bug is from the loop at line 460 that walks through the bits from 1 to 8, and uses that value as the index into the string using MID$. A bit from 1 to 8 is rightmost to leftmost, but on a string, 1 to 8 is leftmost to rightmost. One of them needs to work backwards.

Since I already do some math to calculate the bit (to base-0 counting, 0-7 instead of 1-8), I just made my change there:

470 'FASTER WAY
480 IF MID$(Z$,BT,1)="X" THEN V=V+BT(8-BT)
490 'SLOW WAY
500 'IF MID$(Z$,BT,1)="X" THEN V=V+2^(8-BT)

Now I get the tanks facing the proper directions my test program PUTs them on the screen: UP, DOWN, LEFT then RIGHT.

I have added this code to my GitHub repository in “CoCo Programs” under “basic”:

https://github.com/allenhuffman/CoCo-Programs/tree/main/basic/tank

And here is the full version with the fix:

0 'PUTDATA2.BAS (FIXED BUG)
1 '2026-05-06
2 'SHOWS HOW TO DO IT USING
3 'STRINGS RATHER THAN #S
4 '

10 'TO SPEED UP STR-TO-BYTE
20 DIM BT(7):FOR BT=0 TO 7:BT(BT)=2^BT:NEXT
30 '
40 'VARIABLES USED IN THE
50 'POKE-TO-ARRAY ROUTINE MUST
60 'BE PRE-ALLOCATED OR THEY
70 'WILL CAUSE ARRAY MEM TO
80 'SHIFT WHEN IT GET USED.
90 '
100 DIM L,V,Z,Z$
110 '
120 'EACH ARRAY ENTRY CAN STORE
130 '5 BYTES. AN 8X8 SINGLE
140 'COLOR CHARACTER IS 8, SO
150 'WE NEED TWO ARRAY ENTRIES
160 '(10 BYTES) TO FIT THE 8
170 'BYTE CHARACTER.
180 '
190 DIM TU(1),TD(1),TL(1),TR(1):GOSUB 320

200 '
210 ' TEST PROGRAM
220 '
230 PMODE 4,1:PCLS 1:SCREEN 1,1
240 PUT (0,0)-(7,7),TU
250 PUT (16,0)-(16+7,7),TD
260 PUT (32,0)-(32+7,7),TL
270 PUT (48,0)-(48+7,7),TR
280 GOTO 280

290 '
300 'LOAD SPRITE CHARACTERS
310 '
320 PRINT "LOADING DATA";
330 L=VARPTR(TU(0)):GOSUB 390
340 L=VARPTR(TD(0)):GOSUB 390
350 L=VARPTR(TL(0)):GOSUB 390
360 L=VARPTR(TR(0)):GOSUB 390
370 RETURN

380 '
390 'READ DATA AND POKE AT L
400 '
410 PRINT
420 'READ STRINGS AND CONVERT
430 'TO BYTES.
440 FOR Z=0 TO 7:V=0
450 READ Z$:PRINT Z$,;
460 FOR BT=1 TO 8
470 'FASTER WAY
480 IF MID$(Z$,BT,1)="X" THEN V=V+BT(8-BT)
490 'SLOW WAY
500 'IF MID$(Z$,BT,1)="X" THEN V=V+2^(8-BT)
510 NEXT:PRINT INT(V);HEX$(V)
520 POKE L+Z,(NOT V)+256:NEXT:RETURN
530 NEXT:PRINT V;HEX$(V):NEXT
540 RETURN

550 '
560 '8X8 SPRITE CHARACTERS
570 '
580 'TANK UP
590 DATA "   XX   "
600 DATA "   XX   "
610 DATA "XX XX XX"
620 DATA "XXXXXXXX"
630 DATA "XXXXXXXX"
640 DATA "XXXXXXXX"
650 DATA "XXXXXXXX"
660 DATA "XX    XX"

670 'TANK DOWN
680 DATA "XX    XX"
690 DATA "XXXXXXXX"
700 DATA "XXXXXXXX"
710 DATA "XXXXXXXX"
720 DATA "XXXXXXXX"
730 DATA "XX XX XX"
740 DATA "   XX   "
750 DATA "   XX   "

760 'TANK LEFT
770 DATA "  XXXXXX"
780 DATA "  XXXXXX"
790 DATA "   XXXX "
800 DATA "XXXXXXX "
810 DATA "XXXXXXX "
820 DATA "   XXXX "
830 DATA "  XXXXXX"
840 DATA "  XXXXXX"

850 'TANK RIGHT
860 DATA "XXXXXX  "
870 DATA "XXXXXX  "
880 DATA " XXXX   "
890 DATA " XXXXXXX"
900 DATA " XXXXXXX"
910 DATA " XXXX   "
920 DATA "XXXXXX  "
930 DATA "XXXXXX  "

Until next bug … I mean, time. Until next time…

CoCo smooth scrolling.

Leave a reply

Can someone please “explain this to me like I’m five”:

https://www.6809.org.uk/xroar/online/?-run=512scroll.cas

Click that link to see it run in the web browser. This screen shot does not really show how it looks running:

It shows a CoCo 32×16 text screen smoothy scrolling semigraphics blocks and text characters up the screen. The note that the SAM is not involved is also something I need a child-like explanation of. What does the SAM actually do that would be part of such a technique?

GFXFLIP2.BAS – scrolling a PMODE 4 screen.

Leave a reply

In a post to the CoCo mailing list, Juan Castro said this:

“… I was trying to comment on this post – https://subethasoftware.com/2026/05/19/where-does-cocos-pmode-graphics-memory-start/#comment-42465 – and I was trying to say this:

In GFXFLIP.BAS, instead of the ‘flip through them’ code, try this:

1000 ‘GO DO THE ROTATION
1010 GOTO 1050
1020 FOR P=2 TO 5:GOSUB 1070:PMODE 4,P:SCREEN1:NEXT
1030 IF INKEY$=”” THEN 1030
1040 FOR P=4 TO 1 STEP -1:GOSUB 1070:PMODE 4,P:SCREEN1:NEXT
1050 IF INKEY$=”” THEN 1050
1060 GOTO 1020
1070 FOR TT=0 TO 40:NEXT:RETURN
“

– Juan Casto via CoCo Mailing list

So let’s take a look and see what it does. Replacing my code from 70-110 with a GOSUB to this routine gives us this:

0 'GFXFLIP2.BAS
1 'Juan Castro
10 PCLEAR 8
15 ' DRAW FIRST SCREEN
20 PMODE 4,1:PCLS:SCREEN 1,1
30 CIRCLE(128,96),50
35 ' DRAW SECOND SCREEN
40 PMODE 4,5:PCLS:SCREEN 1,1
50 LINE (10,10)-(245,171),PSET,B
60 ' FLIP THROUGH THEM
70 GOSUB 1000
110 GOTO 70

1000 'GO DO THE ROTATION
1010 GOTO 1050
1020 FOR P=2 TO 5:GOSUB 1070:PMODE 4,P:SCREEN1:NEXT
1030 IF INKEY$="" THEN 1030
1040 FOR P=4 TO 1 STEP -1:GOSUB 1070:PMODE 4,P:SCREEN1:NEXT
1050 IF INKEY$="" THEN 1050
1060 GOTO 1020
1070 FOR TT=0 TO 40:NEXT:RETURN

…and when you run it, you will see it takes the PMODE 4 screen (which is 4 blocks of graphics memory) then goes in a loop where it starts the displayed screen at block 2, then 3, then 4, then 5. With a small delay in-between each change, it shows (initially) the first PMODE 4 screen (blocks 1-4), then it toggles the memory displayed through each block until the start is block 5. This makes the screen appear to scroll up in chunks (1/4th of the screen each time). The process is then reversed to scroll back down.

Neat trick.

Stupid VARPTR tricks

Leave a reply

A recent e-mail exchange with CoCo user Torsten D. inspired me to try something stupid.

The Radio Shack Color Computer gained the VARPTR command in the Extended BASIC ROM. As has been discussed here many times, VARPTR returns the memory location of a 5-byte variable descriptor. There are actually seven bytes total, since the two bytes before the VARPTR address is the two byte variable name.

0 'MAKE42-1.BAS
10 DIM A,B,C:A=42:B=VARPTR(A)
20 FOR C=B-2 TO B+4
30 PRINT C,PEEK(C):NEXT

Running this will show the full seven bytes required to represent the A variable:

VARPTR is returning 9792 as the location of the start of the 5-byte floating point value for the numeric A variable. The two bytes before it are the variable name: 65 (uppercase “A”) and 0 (no second letter). Had the variable name been “AA” that would be 65 65.

The five bytes of 134, 40, 0, 0 and 0 are the floating point representation of 42.0.

If you got the VARPTR address of one numeric variable, and of a second numeric variable, you could just copy those five bytes and clone the variable value:

0 'MAKE42-2.BAS
10 DIM A,B,I
20 A=42:B=0
30 PRINT "A =";A,"B =";B
40 FOR I=0 TO 4:POKE VARPTR(B)+I,PEEK(VARPTR(A)+I):NEXT
50 PRINT "A =";A,"B =";B

If you run this, you will see output showing “A = 42” and “B = 0”, followed by output showing “A = 42” and “B = 42”. The code in line 40 copies the five bytes (offset 0 to 4) from the VARPTR address of variable A into the VARPTR address of variable B.

But why would you want to do that?

B=A is would have been much simpler.

Here’s where it gets more stupid…

And if you can clone a variable, you could also change a variable if you knew what five bytes represented the number you wanted. In the first example, we saw that the five bytes that represent 42.0 are 134, 40, 0, 0 and 0. Knowing that, you could do something like this:

0 'MAKE42-3.BAS
10 DIM A,V
20 PRINT "A =";A
30 V=VARPTR(A):POKE V,134:POKE V+1,40: POKE V+2,0:POKE V+3,0:POKE V+4,0
40 PRINT "A =";A

The “DIM” is not really needed in this example, but it is a good habit to get into if you want to control the order your variables exist in the variable table. Variables that need to be the fastest should be at the front of the list, and variables use infrequently or ones that can be slow can be at the end.

But I digress.

The point of this stupid code is that an “A” variable is created with 0 as its default value, then VARPTR is used to see where that variable is in memory. Five POKE commands put that floating point representation of 42.0 into that variable’s storage… So when you print A, you get a value of 42, even though A=42 (or indeed, A= ANYTHING) was nowhere in the program.

I have no idea why you would want to do that.

Until next time…

Clock Signal emulator

Leave a reply

I just learned about the Clock Signal emulator:

https://github.com/TomHarte/CLK

The author recently joined the Cocopedia.com and has updated some technical sections with information he has learned in adding support for the CoCo. Clock Signal is described as:

“A latency-hating emulator of: the Acorn Electron, BBC Micro and Archimedes, Amstrad CPC, Apple II/II+/IIe and early Macintosh, Atari 2600 and ST, ColecoVision, Enterprise 64/128, Commodore Vic-20 and Amiga, MSX 1/2, Oric 1/Atmos, early PC compatibles, Sega Master System, Sinclair ZX80/81 and ZX Spectrum, and Thomson MO5/6.”

– https://github.com/TomHarte/CLK

The author is adding CoCo to this list. As someone who started out with a Commodore VIC-20, then went to the CoCo, I am intrigued to find another emulator that will do both. (MAME/MESS also does both.) There is currently early support for the CoCo in there, but it only boots a CoCo Color BASIC 1.2 ROM:

I was able to download Clock Signal from GitHub, then open the Xcode project on my Mac, change one setting, then build and run. Neat!

When you first run one of the emulators, it will prompt you to drag in the needed ROMs:

Drag them into that window, then it’s ready to go. Pretty easy.

I know little beyond this, but plan to play with this a bit and try to get some of the other systems up and running.

One downside is that it looks like there may not be a Windows version.

To be continued…

Where does CoCo’s PMODE graphics memory start?

8 Replies

When I got my first Radio Shack TRS-80 Color Computer 1 back around 1983, I dove into the Getting Started manuals and learned all the new wonderful commands that EXTENDED COLOR BASIC offered me that my Commodore VIC-20 did not have. Commands like PLAY for music, SOUND for a beep, and graphics commands to draw CIRCLEs, LINEs and even DRAW complex objects were … amazing.

On my original tape based CoCo (EXTENDED COLOR BASIC) the high-resolution graphics memory started just after the memory used for the 32×16 text screen:

DEC     HEX     DESCRIPTION
-----   ----    -----------
0       0000    Color BASIC Use
512     0400    Text Screen
1536    0600    Hi-Rez Graphics

Knowing that, as I learned some 6809 assembly language I wrote routines to scroll a PMODE 4 256×192 graphics screen. I used this to do video titles for my dad. I’d create a screen using a graphics program, and load that into the second half of the graphics memory, then let my routine smooth scroll that screen into view.

I’d love to find that old source code and see how awfully inefficient it was. I bet one of you could really show me a faster way to do it.

But I digress…

DISK EXTENDED BASIC changed everything!

The next major leap in home computing for me was getting a disk drive for my CoCo. Imagine being able to store up to 156K of data on a floppy disk, and load things at such blazing speed (compared to the tape player and it’s 1500 baud rate).

It was nice… but it broke my assembly code! It turned out, when DISK BASIC was added, it used memory after the text screen for its own purposes, and shifted the high resolution graphics memory 2K further down in the memory map:

DEC     HEX     DESCRIPTION
-----   ----    -----------
0       0000    Color BASIC Use
512     0400    Text Screen
1536    0600    Disk BASIC Use    
3584    0E00    Hi-Rez Graphics

Learning this, I adjusted my assembly routines to work on graphics screens starting at 3584 (disk systems) rather than 1536 (tape systems).

How do we know?

I wondered if there was some programmatic way to tell where the screen started. I don’t think this even dawned on me back in the 1980s, but I asked this question to the new Color Computer mailing list and quickly got an answer:

Word at $BC (GRPRAM) is start of graphics RAM. Word at $BA (BEGGRP) gets you the start of the current view window.

Juan Castro
Enviado do meu Olivetti Programma 101
http://retropolis.com.br

Bonus. Not only can you tell where graphics memory starts, but you can tell which page is displayed. With EXTENDED BASIC, you have 8 1.5K pages of graphics memory you can use. You reserve them with the PCLEAR command (it defaults to 4 pages). You can learn more about PCLEAR in this article.

PMODE 3 (128×192 4-color) and PMODE 4 (256×192 2-color) both need 4 pages, so you can have two screens in those modes. PMODE 0 uses 1 page, so you can have 8 pages in that ode.

  • PMODE 0 – 128×96 2-color (1536 bytes)
  • PMODE 1 – 128×96 4-color (3072 bytes)
  • PMODE 2 – 128×192 2-color (3072 bytes)
  • PMODE 3 – 128×192 4-color (6144 bytes)
  • PMODE 4 – 256×192 2-color (6144 bytes)

When you use PMODE, the first parameter is the graphics mode, and the second is which page for the screen to start on. For PMODE 4 you can do “PMODE 4,1” to get one screen of 4 pages starting at page 1, and “PMODE 4,5” to get a second screen of 4 pages starting at page 5. You can set PMODE to the first screen and draw something, then set it to the second screen and draw something different, then flip back and forth between them using PMODE. Here is a silly example:

0 'GFXFLIP.BAS
10 PCLEAR 8
15 ' DRAW FIRST SCREEN
20 PMODE 4,1:PCLS:SCREEN 1,1
30 CIRCLE(128,96),50
35 ' DRAW SECOND SCREEN
40 PMODE 4,5:PCLS:SCREEN 1,1
50 LINE (10,10)-(245,171),PSET,B
60 ' FLIP THROUGH THEM
70 PMODE 4,1:SCREEN 1,1
80 FOR A=1 TO 100:NEXT
90 PMODE 4,5:SCREEN 1,1
100 FOR A=1 TO 100:NEXT
110 GOTO 70

That program will reserve all 8 pages, then set a PMODE 4 screen starting at page 1. It draws a circle on that page. Then it sets a PMODE 4 screen starting at page 5. It draws a box on that screen. After that it just toggles between showing PMODE 4 starting at page 1, then at page 5, and the image will flicker back and forth between the circle screen and the square screen.

With low-resolution PMODE 0, you can do 8 screens of animation this way.

I wrote a second program that will print out where screen memory starts, and where the current viewed page starts.

0 ' SCRSTART.BAS
1 ' THANK YOU, JUAN CARLOS!
10 PCLEAR 8
20 FOR P=1 TO 8:PMODE 0,P
30 PRINT PEEK(&HBC)*256+PEEK(&HBD),PEEK(&HBA)*256+PEEK(&HBB)
40 NEXT P

When I run this on an tape-based EXTENDED BASIC CoCo (emulator) with no Disk Controller, it shows graphics memory starts at 1536 and then it toggles through each PMODE 0 page to show where each one would start:

And when DISK EXTENDED BASIC is used, the same program shows the memory locations starting 2K higher in memory:

Thank you, Juan Carlos, for telling me this. I wish I had known about this back then. I could have made my assembly program automatically find the start of the graphics screen rather than having to assembly separate versions for tape or disk systems.

Until next time…