Current NEON 32k binary with full EXTEND plus EASYFILE. Load this up with Spin tool or BST etc and it will automatically handle 5 or 10MHz crystals to operate at 80MHz. Connect a serial terminal (TeraTerm or minicom etc) at 115200-8N1.
If you want to enable the SD card on your board just specify the pin numbers prefixed by & and each pin separated by a decimal point in IP format as &ce.miso.mosi.clk so that if CE = P25 and card data out = P24 and card data in = P27 and clock = P26 then enter:
&25.24.27.26 SDPINS
and this will be locked into EEPROM so it will automatically mount on reboot or manually type "MOUNT"
The card detect is built into every SD card as a pull-up on CE so there is no need for a separate card detect input.
Because this is a standard 32k image it is missing the ROMS but most casual users won't need these although it is easy enough to start with a fresh compile etc.
I thought I'd play a little bit and convert my VGA Breakout game across to V5. Turns out that the code fits into 1kB and the source is less than 250 lines long.
TACHYON V5
{ BREAKOUT for 32x15 VGA text
Updated for V5 NEON
Code bytes used = 1020
}
TIMER duration
--- square wave output audio pin
pri audio 10 ;
--- hitting a wall sound
pub BONK audio APIN 300 HZ 50 duration TIMEOUT ;
--- palette to suit game
TABLE gamepal
$0800.0800 , $0808.0000 , $F000.F000 , $F0F0.0000 ,
$8000.8000 , $8080.0000 , $2000.2000 , $2020.0000 ,
$8800.8800 , $8888.0000 , $2800.2800 , $2828.0000 ,
$5400.5400 , $5454.0000 , $A800.A800 , $A8A8.0000 ,
--- palette colors
0 := red
1 := grn
2 := blu
3 := yel
4 := mag
5 := cyan
6 := gry
7 := wht
--- back of the brick wall - leave room for the ball to bounce behind
3 := backwall
$7F00 := bram
$03FC := bval
byte score
byte balls
pri BRICKS? ( -- cnt )
score C~
backwall cols * 2* screen + FROM 2 BY
cols 4* FOR I C@ 14 <> IF score C++ THEN NEXT
;
--- calculate how many bricks have been removed and display
pri .SCORE
BRICKS?
grn HUE cols 7 - 0 VXY
VGA score C@ .AS" ###/128"
<CR> balls C@ .AS" # balls"
;
--- ball position variables
word bx
word by
word abx
word aby
pri bdir 0 >N ;
--- initialize the brick wall with 4 rows of random colored bricks/tiles from backwall (gap)
pub WALL 0 backwall VXY cols 4* FOR RND 8>> 7 AND HUE 14 VCHAR NEXT ;
{ The ball is drawn with a single character but by using codes which force the driver to access RAM
for the font we can have finer movement by drawing a ball character dynamically in x 16x32 matrix
Use code $3FA to access table at $7E80
This routine is essentially the same as VCHAR but optimized for directly writing without scroll
Execution time: 48.8us
--- 10-bit character
}
pub VCHAR! ( ch -- )
--- (color << 1 + c & 1 )
DUP 1 AND color C@ 2* +
--- form color field
2* 1+ 9 <<
--- merge 7 msb of data
SWAP 1 ANDN +
--- current screen position
row C@ 5 << col C@ +
DUP 960 <
--- write to the screen word & inc column
IF 2* screen + W! col C++ ELSE 2DROP THEN
;
TIMER balldly --- timeout leads to next ball movement
word speed
--- rebound translation table looks up the corresponding rebound direction ( 7 into left returns with 9 )
TABLE rebound
--- 0 1 2 3 4 5 6 7 8 9
0 | 7 | 8 | 9 | 6 | 5 | 4 | 1 | 2 | 3 | --- bottom/top rebound
0 | 3 | 8 | 1 | 6 | 5 | 4 | 9 | 2 | 7 | --- left/right rebound
--- lookup the table for the rebound action & sound
pub BOUNCE ( table-offset -- ) rebound + bdir + C@ ' bdir C! BONK ;
pri BALL@ ( -- word ) bx W@ 2/ 2/ by W@ 3 >> cols * + 2* screen + W@ ;
--- read the contents of the screen cell/tile where the ball will be next
pub NOBALL
bx W@ 2/ 2/ by W@ 3 >> VXY
$20 VCHAR!
;
pri !BALL
NOBALL
pri NEWBALL
64 bx W! 60 by W!
50 speed W! 1 ' bdir C!
--- delay serving 1st ball for 1.5secs
1500 balldly TIMEOUT
pub BALL
--- limit ball x
bx W@ cols 2* 2* MIN bx W!
--- limit ball y
by W@ rows 2* 2* 2* MIN by W!
--- bounce off special tiles (not blank or alphanumeric)
BALL@ >B 20 'z' WITHIN NOT IF 0 BOUNCE EXIT THEN
--- bounce off bottom
by W@ rows 3 << 1- => IF 0 BOUNCE EXIT THEN
--- bounce off top
by W@ 1 < IF 0 BOUNCE EXIT THEN
--- bounce off left wall
bx W@ 0= IF 10 BOUNCE EXIT THEN
--- bounce off right wall
bx W@ 2/ 2/ cols => IF 10 BOUNCE EXIT THEN
\ pub DRAWBALL ( -- )
--- plot ball as 4x4 block - 185us
wht HUE bx W@ by W@
--- address the screen position for the tile - 17.2us
OVER 2/ 2/ OVER 2/ 2/ 2/ VXY
--- maps to programmable char at $7E80 - 63.6us
bval VCHAR!
--- generate a dot in the correct x position - 7.8us
SWAP 3 AND 2* 2* 2* $AA SWAP <<
--- wipe the character clean - 64.6us
bram 128 ERASE
--- calculate y position in character - 8.4us
SWAP 7 AND 4 << bram + ( mask addr )
--- create the pattern
SWAP OVER ! DUP 4 + 12 CMOVE
;
--- setup constants for ball x and y movement increments which contain 2 fractional bits
1 := abxcon
1 := abycon
pri BALL+! ( x y -- )
NOBALL
by W+! bx W+!
BALL
;
--- Check if ball is ready for next movement and proceed
pub ?BALL
--- ready yet?
balldly TIMEOUT? 0EXIT
--- yes, set next timeout period
speed W@ balldly TIMEOUT
--- proceed using the ball direction
bdir SWITCH
7 CASE abxcon NEGATE abycon NEGATE BALL+! BREAK
9 CASE abxcon abycon NEGATE BALL+! BREAK
1 CASE abxcon NEGATE abycon BALL+! BREAK
3 CASE abxcon abycon BALL+! BREAK
;
--- paddle xy
byte px
byte py
--- two characters that make up the paddle
$8E88 := mypad
$2020 := nopad
--- draw the paddle
pub PADDLE ( shape -- ) yel HUE px C@ py C@ VXY W>B VCHAR VCHAR ;
--- PADDLE MOVEMENTS
pub L px C@ IF nopad PADDLE px C-- mypad PADDLE THEN ;
pub R px C@ cols 2 - < IF nopad PADDLE px C++ mypad PADDLE THEN ;
--- check to see if the ball has hit the paddle and which change it's direction based on the edge hit
pub ?PADDLE mypad PADDLE ;
pri NEWPADDLE rows 1- py C! 15 px C! mypad PADDLE ;
pub GAMEOVER red " GAME OVER" ( 11 8 TEXTBOX ) CON CONSOLE ;
pub NEWGAME
VGA CLS NEWPADDLE NEWBALL WALL
7 0 VXY 2 HUE PRINT" TACHYON BREAKOUT "
3 balls C! score C~
;
pri FASTER speed W-- ;
pri SLOWER speed W++ ;
--- game control keys ---
pub GAMEKEY
KEY ?DUP 0EXIT
SWITCHES
'Z' L 'X' R
'z' L 'x' R
'+' FASTER '-' SLOWER
'B' !BALL $0D NEWGAME $1B GAMEOVER
;
pub ACTION
--- make it speed up proportionaly to the score
130 score C@ - 3 / speed W!
GAMEKEY
?BALL
?PADDLE
duration TIMEOUT? IF MUTE THEN
.SCORE
--- testing: autostart a new game
score C@ 128 = IF NEWGAME THEN
;
pub BREAKOUT
!VGA
audio APIN MUTE 0 duration TIMEOUT
gamepal colors 64 CMOVE
NEWGAME
pub RESUME
30 FLOAT
VGA BEGIN ACTION AGAIN
;
I've loaded up the v5.2 binary, and run the refactored stepper motor code, works fine.
Thanks for your advice about refactoring, by the way structuring the code in a semi-naturalistic way is one of the intriguing things about Forth. I'm very much a Forth noob of course, so it will take some time for me to acquire any degree of skill. I have an electronic copy of Brodie's Starting Forth, so that should help
One thing I am having trouble with is applying the v5.2 EXTEND code (I want to remove the LCD, VGA and KEY modules as I don't need them). I commented out the first line with \ <space> as directed, Ctrl-A to select all the code, and pasted into the minicom windows as I did with v4.7. I noticed the system throwing up errors as the file was processed:
... \
...
... TACHYON V5 Propeller .:.:--TACHYON--:.:. Forth V5r2 NEON 520180323.2200
0100 ok
0200 ok ??? ??? ok
0300
0400
0500 ok ??? ok
0600
0616 ??? in s at CLKFREQ
0632 ??? in M at 1M �
0700 ok �
0726 ??? in $= at FALSE
etc.
My line delay is set to 30ms (I upped it from 15ms to see if it had any effect).
A> Now that looks like you might actually have to load in the kernel with the Prop tool or similar, and then paste EXTEND. Perhaps I've made some change but at least try that. I even tried COLD (or ^Z^Z) and loaded EXTEND back in without incident.
Of course you could simply FORGET VGATEXT and that will remove all the other modules including the extra tools, which you could reload again and BACKUP.
Which terminal emulator are you using btw? I run minicom in Linux with 8ms line delay.
I have to mention though that V5 doesn't try to convert case, at least I haven't decided yet if it is as good as an idea it once seemed anyway. But for larger blocks of source it is more efficient to have them beginning with TACHYON and terminated with END.
Don't miss clicking on my "Use the Forth" logo to take you to the links and intros etc.
Cheers!
I have a fresh build, I had to compile as I am using a 12MHz xtal.
I loaded Extend without a problem but when I try to load EASYFILE it get errors and it then a hang.
I have tried different serial delays, different serial programs, different editors to clean up files (I was concerned that there may have been hidden chars) but not joy.
I have even tried archive versions of EASYFILE etc.
I am using Terraterm from windows atm.
The only thing from standard is that I did not load VGA from EXTEND as I am using only a serial console.
Any suggestions?
...
Propeller .:.:--TACHYON--:.:. Forth V5r2 NEON 520180323.2200
*** MODULES *** Propeller .:.:--TACHYON--:.:. Forth V5r2 NEON 520180323.2200
3386: TOOLS DEV TOOLS
30DE: CHARLCD BUFFERED CHARACTER LCD
1982: EXTEND Primary extensions to TACHYON V5.2 kernel - 180322-1300
AUTORUN BOOT 3076
FREQ = 96.00MHZ
*** INITS ***
*** ROMS ***
E01C: VGA32x15 848
E37C: UART 464
E55C: HSUART 560
E79C: F32 1900
*** I2C ***
$A0 EEPROM
I/O = 31 :UHUU 27 :~~~~ 23 :~~~~ 19 :~~~~ 15 :U~~~ 11 :~~~~ 7 :~~~~ 3 :~~~~
INTERCOM:
CODE:$3A2A = 14378 bytes
NAME:$5A2E = 6610 bytes
DATA:$75FF = 239 bytes
FREE: = 8196 bytes
Data Stack (0)
Mon, 01 Jan 2001 00:00:00 UTC
--------------------------------------------------------------------------------
...
...
...
...
...
...
...
...
.. IFDEF EASYFILE
...
...
...
... TACHYON V5 Propeller .:.:--TACHYON--:.:. Forth V5r2 NEON 520180323.2200
0085 ??? in *SDCS at cspin
0100 þ
0198 ??? in SDDAT! at =dtk
0200 þ
0248 ??? in SDWR at =dtk
0250 ??? in SDWR at BLKSIZ
0284 ??? in RDSECT at BLKSIZ
0300 þ
0312 ??? in SECTOR at =dtk
0352 ??? in FILE at #files þ
0361 ??? in FILE# at #files
0394 ??? in FSIZE at @FSIZE þ
0400 þ
0500 ??? ??? ok
0550 ??? in @BOOT at fatptr
0636 ??? in MOUNT at #files þ
0639 ??? in MOUNT at BLKSIZ þ
0700 ??? þ
0834 ??? in FSTAMP at @FTIME
0835 ??? in FSTAMP at @FDATE þ
0910 ??? in FreeClusters? at BLKSIZ
0925 ??? in ClaimClusters at endcl þ
0931 ??? in FirstCluster at @FCLST þ
ÞÞÞäin FirstCluster at @FCLSTH
0980 ??? in FPUTB at BLKSIZ þ
0986 ??? in FPUTB at BLKSIZ þ
1000 þ
1043 ??? in FCREATE$ at @ATR
1045 ??? in FCREATE$ at @CDATE
ÞÞÞäin FCREATE$ at @CTIME
1049 ??? in FCREATE$ at @FCLSTH þ
ÞÞÞäin FCREATE$ at @FCLST
1100 þ
1153 ??? in -FERASE at BLKSIZ
1154 ??? in -FERASE at BLKSIZ þ
1189 ??? in APPEND at BLKSIZ þ
1200 þ
1234 ??? in _FCOPY at BLKSIZ
1300 þ
1306 ??? in cd$ at @ATR
1447 ??? in .DIR$ at @ATR
1464 ??? in .FNAME at @ATR
1468 ??? in .FNAME at @ATR þ
1483 ??? in .DIR at @ATR þ
1487 ??? in .DIR at @FSIZE
1489 ??? in .DIR at @FDATE þ
ÞÞÞäin .DIR at @FTIME
1491 ??? in .DIR at @CDATE þ
ÞÞÞäin .DIR at @CTIME
1500 þ
1512 ??? in lsdirs at BLKSIZ
1516 ??? in lsdirs at @ATR þ
1558 ??? in .UTIME at @FTIME
1574 ??? in (.LIST) at @FSIZE
1575 ??? in (.LIST) at @FDATE þ
1600 þ
1607 ??? in .FILES at #files
1760 ??? þ
Ahhh, that's what happens when I get distracted in the middle of changes. I decided to rename some common Tachyon words such as == which I use for constants and rename it to := instead but I hadn't updated EASYFILE. So try it now as it compiled fine after that. If you want to free up some memory then do a RECLAIM and then a BACKUP.
Yes, you're right - I compiled it with BST, no problems. Loaded it into my DAQiri system and pasted in EXTEND (minus VGA/LCD/KEY), again no problems. 9k free!
I may be misunderstanding how to read this but I think it is telling me it comes from EXTEND?
I appreciate the assistance.
Hi Lee,
Some versions may have certain words, others don't, either because there are many ways of doing this, or they were't used often enough. However there are definitely core words that are essential too that are always there. ESC? is not of the latter but because it is so simple I can add it back into EXTEND
It is not a problem for me to re add now as you have shown me the code, I was just trying to work out if there is a master list with the code so if any are missing I can add as needed.
I have gone back to the V4 and had a look around but could not find but again I am a babe in the woods so may be looking in the wrong places.
It is not a problem for me to re add now as you have shown me the code, I was just trying to work out if there is a master list with the code so if any are missing I can add as needed.
I have gone back to the V4 and had a look around but could not find but again I am a babe in the woods so may be looking in the wrong places.
No problems, I added it back in and it is useful for terminating listings and test loops though. Each new version of Tachyon is done for a reason, and V5 builds on the V4 which changed over from bytecode to wordcode, mainly for handling the address space more efficiently for larger programs. But V5 compacts the dictionary a fraction more, simplifies some other things while making more hub space available for code. This way V5 can squeeze in EASYFILE and EASYNET while avoiding having to compress the dictionary into upper EEPROM. Besides, it has a simple decompiler built-in! Type "HELP ESC?" and it will decompile the word and pause on any EXIT it comes across where you can decide to let it continue or hit a key to stop it.
I wrote this as a test, I was expecting the first 13 but I tried other keys after the program was running and it does not update lastkey, from what I read in the docs, lastkey is filled from the RX buffer, have I misunderstood?
@lmclaren - my bad, I see that previous versions directly wrote to the lastkey variable from the serial cog but I cut this bit of overhead out for later versions which then incorporated the ping-pong Propeller networking software referred to as the "intercom".
The advantage of reading lastkey was that it was non-invasive by not using up a character from the serial receive buffer. However if you don't mind that then define ESC? as:
Do you have a 'roadmap' for where you want to take v5? The reason I ask is that I think Tachyon deserves a much higher profile (along with the Propeller itself!), and I wonder if the key to that might be documentation. I note that you have a couple of documents, "Introduction to Tachyon" and the Tachyon glossary, which would be an excellent place to start for a full manual. I'd be happy to assist. What are your thoughts?
I'm currently in the middle of designing a new hardware revision of my data acquisition system, but I expect to be making much fuller use of Tachyon soon!
Do you have a 'roadmap' for where you want to take v5? The reason I ask is that I think Tachyon deserves a much higher profile (along with the Propeller itself!), and I wonder if the key to that might be documentation. I note that you have a couple of documents, "Introduction to Tachyon" and the Tachyon glossary, which would be an excellent place to start for a full manual. I'd be happy to assist. What are your thoughts?
I'm currently in the middle of designing a new hardware revision of my data acquisition system, but I expect to be making much fuller use of Tachyon soon!
If you want you can edit those documents and others, I'd be happy to have some help. I can pm you an edit link if you like.
But now that P2 is so close I wonder whether I would use P1 in many new designs. This may be the end of the Tachyon road for the P1 but a whole new super expressway for the P2. Nonetheless, Neon is a powerful and refined implementation of Tachyon for the P1 and even if it doesn't have the profile it should have along with the Propeller, we are not the ones missing out
A couple of questions regarding your VGA Breakout code:
1. TABLE - this keyword doesn't appear in the glossary. Looking the code, I'm assuming it copies the a hub memory address to a variable, which can be used as a reference to the series of compiled literals that follow? And that the address corresponds to the address of the first compiled literal in hub memory?
So in this case, the variable 'gamepal' holds the address of the compiled literal $0800.0800, and you access elements in the table by adding an offset to gamepal?
2. colors - appears to be the destination for a copy of the gamepal table, but I don't see a definition for colors anywhere?
TABLE creates a variable in code memory without any extra memory allocated, a bit like the traditional 0 VARIABLE -4 ALLOT would. So all that gamepal comes back with is the address in code memory of that array and it's up to code to index into it but in this case it overwrites the color table that is defined in the VGATEXT module in EXTEND.
Thanks for your help thus far. I took some time out from PCB routing to sit down with Tachyon again. I've modified the stepper code to be able to step in either direction, e.g.
100 CCW MTR.STEPS
or
25 CW MTR.STEPS
I'm also driving the stepper in half-step mode as the manufacturer recommends, which means there are 8 phase patterns per step. I created a table of the phase patterns, and index from the start to the end of vice versa, depending on direction:
0 := CCW
1 := CW
--- utility function to write a byte to an I2C chip register
pub IO! ( data reg -- ) $40
pub TCA! ( data reg dev -- ) <I2C I2C! I2C! I2C! I2C> ;
word ratems
--- set the step but also make sure the I/O is initialized
pub MTR.RATE ( ms -- ) ratems W!
--- Set I2C bus pins for stepper card, initialise TCA9535 ports 0 & 1 to output, set all outputs low
pub MTR.INIT ( -- )
7 6 I2CPINS
0 2 IO!
0 3 IO!
0 6 IO!
0 7 IO! ;
--- Bit patterns for driving 28BYJ-48 in half-step mode
TABLE phases
$02 | $06 | $04 | $0C | $08 | $18 | $10 | $12 |
--- set motor phase outputs on port 0 of TCA9535 and delay for programmed speed
pub MTR.PHASE ( phase -- ) 2 IO! ratems W@ ms ;
--- Do a single complete step
pub MTR.STEP ( -- ) 8 SWAP 0= IF phases FROM 1 ELSE phases 7 + FROM -1 THEN BY FOR I C@ MTR.PHASE NEXT ;
--- Turn on activity LED for motor, run for <steps> steps in the specified direction (CW or CCW), at set rate between phase outputs
pub MTR.STEPS ( steps dir -- )
3 HIGH
SWAP
FOR
DUP
MTR.STEP
NEXT
DROP
0 MTR.PHASE
3 LOW ;
I would be interested in your comments, as I guessing this can be done more efficiently than the approach that I have taken.
Cheers
Andrew
P.S. I'm compiling a list of glossary edits, once it is sufficiently long we might go over it...
Well done so far, please bear with me as I try to share some of the things I have found that helps with writing Forth software.
Compiled languages encourage you to use all_kinds_of_long_names but one of the things I like about Forth is that the name becomes part of the language. More correctly though, the name becomes the language in that you end up programming with these new names. If well chosen and thought out then it will feel and read natural but all too often the structures of conventional languages that were forced upon us are in turn forced upon Forth so that the Forth vocabulary has not been "extended" or enhanced at all. What we end up with is "code" in need of decoding.
So you find that to move the motor slowly in a clockwise direction you end up typing/coding:
20 MTR.RATE
200 CW MTR.STEPS
Now that's not so bad, It works, but could you express what you want to do more naturally like:
SLOW 200 RIGHT
then perhaps
FASTER 400 LEFT
This is also a lot easier to type in interactively and now the language has been extended and enhanced. The code can also be its own comments.
--- utility function to write a byte to an I2C chip register
pub IO! ( data reg -- ) $40
pub TCA! ( data reg dev -- ) <I2C I2C! I2C! I2C! I2C> ;
long mdir word ratems byte phase
--- Bit patterns for driving 28BYJ-48 in half-step mode
TABLE phases $02 | $06 | $04 | $0C | $08 | $18 | $10 | $12 |
--- set the step but also make sure the I/O is initialized
pri THROTTLE ( ms -- )
1 MAX ratems W!
--- Set I2C bus pins for stepper card, initialise TCA9535 ports 0 & 1 to output, set all outputs low
7 6 I2CPINS
0 2 IO!
0 3 IO!
0 6 IO!
0 7 IO! ;
--- speed control
pub SLOW 10 THROTTLE ;
pub FAST 1 THROTTLE ;
pub FASTER ratems W@ 2/ THROTTLE ;
pub SLOWER ratems W@ 2* THROTTLE ;
--- set motor phase outputs on port 0 of TCA9535 and delay for programmed speed
pri COILS ( phase -- ) 2 IO! ratems W@ ms ;
--- Do just one tiny phase change step
pub STEP ( -- ) phase C@ 7 AND phases + C@ COILS mdir @ + phase C+! ;
--- Turn on activity LED for motor, run for <steps> steps in the specified direction (CW or CCW), at set rate between phase outputs
pub STEPS ( steps -- ) 3 HIGH FOR DUP STEP NEXT 0 PHASE 3 LOW ;
pub LEFT 1 mdir ! STEPS ;
pub RIGHT -1 mdir ! STEPS ;
pub DEMO
SLOW 200 LEFT FAST 200 RIGHT
SLOWER 500 LEFT FASTER 500 RIGHT
;
Add TURNS or DEGREES so you can say LEFT 45 DEGREES or 8 TURNS if positioning is more important than speed and movement.
First we define these I/O lines that we need as constants and pass this to the PING routine to keep it general-purpose so we can use any I/O.
#P16 == TRIG --- high output to trigger ping
#P17 == ECHO --- signal received input, measure high time
Now we can create a general-purpose PING function that accepts the pins we want to use and returns with the time in microseconds.
pub PING ( trig echo -- us )
MASK 3 COGREG! --- setup WAITPxx mask (in Tachyon cog registers )
DUP HIGH >R R> LOW --- 10us trigger (dummy cycles)
(WAITPEQ) (WAITPNE) --- detect high period as these also capture the CNT into cogregs
0 COGREG@ 1 COGREG@ - --- calculate high period from lo to hi and hi to lo captures
CLKFREQ #1,000,000 / / --- convert cycles to us
;
I am getting errors with the COGREG! and the (WAITPEQ) and (WAITPNE).
I can not see where COGREG! is defined, I can find references in the documentation but it does not seem to work.
I am not sure if the WAITPEQ etc have changed names, are they now WAITHI and WAITLO?
Comments
there might still be some alphas left in V5.2 - here just comment
... searched - did not find others
' malphae the exponents equal if_c shr manA, expA if_nc shr manB, expAIf you want to enable the SD card on your board just specify the pin numbers prefixed by & and each pin separated by a decimal point in IP format as &ce.miso.mosi.clk so that if CE = P25 and card data out = P24 and card data in = P27 and clock = P26 then enter: and this will be locked into EEPROM so it will automatically mount on reboot or manually type "MOUNT"
The card detect is built into every SD card as a pull-up on CE so there is no need for a separate card detect input.
Because this is a standard 32k image it is missing the ROMS but most casual users won't need these although it is easy enough to start with a fresh compile etc.
TACHYON V5 { BREAKOUT for 32x15 VGA text Updated for V5 NEON Code bytes used = 1020 } TIMER duration --- square wave output audio pin pri audio 10 ; --- hitting a wall sound pub BONK audio APIN 300 HZ 50 duration TIMEOUT ; --- palette to suit game TABLE gamepal $0800.0800 , $0808.0000 , $F000.F000 , $F0F0.0000 , $8000.8000 , $8080.0000 , $2000.2000 , $2020.0000 , $8800.8800 , $8888.0000 , $2800.2800 , $2828.0000 , $5400.5400 , $5454.0000 , $A800.A800 , $A8A8.0000 , --- palette colors 0 := red 1 := grn 2 := blu 3 := yel 4 := mag 5 := cyan 6 := gry 7 := wht --- back of the brick wall - leave room for the ball to bounce behind 3 := backwall $7F00 := bram $03FC := bval byte score byte balls pri BRICKS? ( -- cnt ) score C~ backwall cols * 2* screen + FROM 2 BY cols 4* FOR I C@ 14 <> IF score C++ THEN NEXT ; --- calculate how many bricks have been removed and display pri .SCORE BRICKS? grn HUE cols 7 - 0 VXY VGA score C@ .AS" ###/128" <CR> balls C@ .AS" # balls" ; --- ball position variables word bx word by word abx word aby pri bdir 0 >N ; --- initialize the brick wall with 4 rows of random colored bricks/tiles from backwall (gap) pub WALL 0 backwall VXY cols 4* FOR RND 8>> 7 AND HUE 14 VCHAR NEXT ; { The ball is drawn with a single character but by using codes which force the driver to access RAM for the font we can have finer movement by drawing a ball character dynamically in x 16x32 matrix Use code $3FA to access table at $7E80 This routine is essentially the same as VCHAR but optimized for directly writing without scroll Execution time: 48.8us --- 10-bit character } pub VCHAR! ( ch -- ) --- (color << 1 + c & 1 ) DUP 1 AND color C@ 2* + --- form color field 2* 1+ 9 << --- merge 7 msb of data SWAP 1 ANDN + --- current screen position row C@ 5 << col C@ + DUP 960 < --- write to the screen word & inc column IF 2* screen + W! col C++ ELSE 2DROP THEN ; TIMER balldly --- timeout leads to next ball movement word speed --- rebound translation table looks up the corresponding rebound direction ( 7 into left returns with 9 ) TABLE rebound --- 0 1 2 3 4 5 6 7 8 9 0 | 7 | 8 | 9 | 6 | 5 | 4 | 1 | 2 | 3 | --- bottom/top rebound 0 | 3 | 8 | 1 | 6 | 5 | 4 | 9 | 2 | 7 | --- left/right rebound --- lookup the table for the rebound action & sound pub BOUNCE ( table-offset -- ) rebound + bdir + C@ ' bdir C! BONK ; pri BALL@ ( -- word ) bx W@ 2/ 2/ by W@ 3 >> cols * + 2* screen + W@ ; --- read the contents of the screen cell/tile where the ball will be next pub NOBALL bx W@ 2/ 2/ by W@ 3 >> VXY $20 VCHAR! ; pri !BALL NOBALL pri NEWBALL 64 bx W! 60 by W! 50 speed W! 1 ' bdir C! --- delay serving 1st ball for 1.5secs 1500 balldly TIMEOUT pub BALL --- limit ball x bx W@ cols 2* 2* MIN bx W! --- limit ball y by W@ rows 2* 2* 2* MIN by W! --- bounce off special tiles (not blank or alphanumeric) BALL@ >B 20 'z' WITHIN NOT IF 0 BOUNCE EXIT THEN --- bounce off bottom by W@ rows 3 << 1- => IF 0 BOUNCE EXIT THEN --- bounce off top by W@ 1 < IF 0 BOUNCE EXIT THEN --- bounce off left wall bx W@ 0= IF 10 BOUNCE EXIT THEN --- bounce off right wall bx W@ 2/ 2/ cols => IF 10 BOUNCE EXIT THEN \ pub DRAWBALL ( -- ) --- plot ball as 4x4 block - 185us wht HUE bx W@ by W@ --- address the screen position for the tile - 17.2us OVER 2/ 2/ OVER 2/ 2/ 2/ VXY --- maps to programmable char at $7E80 - 63.6us bval VCHAR! --- generate a dot in the correct x position - 7.8us SWAP 3 AND 2* 2* 2* $AA SWAP << --- wipe the character clean - 64.6us bram 128 ERASE --- calculate y position in character - 8.4us SWAP 7 AND 4 << bram + ( mask addr ) --- create the pattern SWAP OVER ! DUP 4 + 12 CMOVE ; --- setup constants for ball x and y movement increments which contain 2 fractional bits 1 := abxcon 1 := abycon pri BALL+! ( x y -- ) NOBALL by W+! bx W+! BALL ; --- Check if ball is ready for next movement and proceed pub ?BALL --- ready yet? balldly TIMEOUT? 0EXIT --- yes, set next timeout period speed W@ balldly TIMEOUT --- proceed using the ball direction bdir SWITCH 7 CASE abxcon NEGATE abycon NEGATE BALL+! BREAK 9 CASE abxcon abycon NEGATE BALL+! BREAK 1 CASE abxcon NEGATE abycon BALL+! BREAK 3 CASE abxcon abycon BALL+! BREAK ; --- paddle xy byte px byte py --- two characters that make up the paddle $8E88 := mypad $2020 := nopad --- draw the paddle pub PADDLE ( shape -- ) yel HUE px C@ py C@ VXY W>B VCHAR VCHAR ; --- PADDLE MOVEMENTS pub L px C@ IF nopad PADDLE px C-- mypad PADDLE THEN ; pub R px C@ cols 2 - < IF nopad PADDLE px C++ mypad PADDLE THEN ; --- check to see if the ball has hit the paddle and which change it's direction based on the edge hit pub ?PADDLE mypad PADDLE ; pri NEWPADDLE rows 1- py C! 15 px C! mypad PADDLE ; pub GAMEOVER red " GAME OVER" ( 11 8 TEXTBOX ) CON CONSOLE ; pub NEWGAME VGA CLS NEWPADDLE NEWBALL WALL 7 0 VXY 2 HUE PRINT" TACHYON BREAKOUT " 3 balls C! score C~ ; pri FASTER speed W-- ; pri SLOWER speed W++ ; --- game control keys --- pub GAMEKEY KEY ?DUP 0EXIT SWITCHES 'Z' L 'X' R 'z' L 'x' R '+' FASTER '-' SLOWER 'B' !BALL $0D NEWGAME $1B GAMEOVER ; pub ACTION --- make it speed up proportionaly to the score 130 score C@ - 3 / speed W! GAMEKEY ?BALL ?PADDLE duration TIMEOUT? IF MUTE THEN .SCORE --- testing: autostart a new game score C@ 128 = IF NEWGAME THEN ; pub BREAKOUT !VGA audio APIN MUTE 0 duration TIMEOUT gamepal colors 64 CMOVE NEWGAME pub RESUME 30 FLOAT VGA BEGIN ACTION AGAIN ;ENDMike
I've loaded up the v5.2 binary, and run the refactored stepper motor code, works fine.
Thanks for your advice about refactoring, by the way structuring the code in a semi-naturalistic way is one of the intriguing things about Forth. I'm very much a Forth noob of course, so it will take some time for me to acquire any degree of skill. I have an electronic copy of Brodie's Starting Forth, so that should help
One thing I am having trouble with is applying the v5.2 EXTEND code (I want to remove the LCD, VGA and KEY modules as I don't need them). I commented out the first line with \ <space> as directed, Ctrl-A to select all the code, and pasted into the minicom windows as I did with v4.7. I noticed the system throwing up errors as the file was processed:
My line delay is set to 30ms (I upped it from 15ms to see if it had any effect).
Any ideas?
A>
Of course you could simply FORGET VGATEXT and that will remove all the other modules including the extra tools, which you could reload again and BACKUP.
Which terminal emulator are you using btw? I run minicom in Linux with 8ms line delay.
I have to mention though that V5 doesn't try to convert case, at least I haven't decided yet if it is as good as an idea it once seemed anyway. But for larger blocks of source it is more efficient to have them beginning with TACHYON and terminated with END.
Don't miss clicking on my "Use the Forth" logo to take you to the links and intros etc.
Cheers!
I am seeing the same problem as A>
I have a fresh build, I had to compile as I am using a 12MHz xtal.
I loaded Extend without a problem but when I try to load EASYFILE it get errors and it then a hang.
I have tried different serial delays, different serial programs, different editors to clean up files (I was concerned that there may have been hidden chars) but not joy.
I have even tried archive versions of EASYFILE etc.
I am using Terraterm from windows atm.
The only thing from standard is that I did not load VGA from EXTEND as I am using only a serial console.
Any suggestions?
... Propeller .:.:--TACHYON--:.:. Forth V5r2 NEON 520180323.2200 *** MODULES *** Propeller .:.:--TACHYON--:.:. Forth V5r2 NEON 520180323.2200 3386: TOOLS DEV TOOLS 30DE: CHARLCD BUFFERED CHARACTER LCD 1982: EXTEND Primary extensions to TACHYON V5.2 kernel - 180322-1300 AUTORUN BOOT 3076 FREQ = 96.00MHZ *** INITS *** *** ROMS *** E01C: VGA32x15 848 E37C: UART 464 E55C: HSUART 560 E79C: F32 1900 *** I2C *** $A0 EEPROM I/O = 31 :UHUU 27 :~~~~ 23 :~~~~ 19 :~~~~ 15 :U~~~ 11 :~~~~ 7 :~~~~ 3 :~~~~ INTERCOM: CODE:$3A2A = 14378 bytes NAME:$5A2E = 6610 bytes DATA:$75FF = 239 bytes FREE: = 8196 bytes Data Stack (0) Mon, 01 Jan 2001 00:00:00 UTC -------------------------------------------------------------------------------- ... ... ... ... ... ... ... ... .. IFDEF EASYFILE ... ... ... ... TACHYON V5 Propeller .:.:--TACHYON--:.:. Forth V5r2 NEON 520180323.2200 0085 ??? in *SDCS at cspin 0100 þ 0198 ??? in SDDAT! at =dtk 0200 þ 0248 ??? in SDWR at =dtk 0250 ??? in SDWR at BLKSIZ 0284 ??? in RDSECT at BLKSIZ 0300 þ 0312 ??? in SECTOR at =dtk 0352 ??? in FILE at #files þ 0361 ??? in FILE# at #files 0394 ??? in FSIZE at @FSIZE þ 0400 þ 0500 ??? ??? ok 0550 ??? in @BOOT at fatptr 0636 ??? in MOUNT at #files þ 0639 ??? in MOUNT at BLKSIZ þ 0700 ??? þ 0834 ??? in FSTAMP at @FTIME 0835 ??? in FSTAMP at @FDATE þ 0910 ??? in FreeClusters? at BLKSIZ 0925 ??? in ClaimClusters at endcl þ 0931 ??? in FirstCluster at @FCLST þ ÞÞÞäin FirstCluster at @FCLSTH 0980 ??? in FPUTB at BLKSIZ þ 0986 ??? in FPUTB at BLKSIZ þ 1000 þ 1043 ??? in FCREATE$ at @ATR 1045 ??? in FCREATE$ at @CDATE ÞÞÞäin FCREATE$ at @CTIME 1049 ??? in FCREATE$ at @FCLSTH þ ÞÞÞäin FCREATE$ at @FCLST 1100 þ 1153 ??? in -FERASE at BLKSIZ 1154 ??? in -FERASE at BLKSIZ þ 1189 ??? in APPEND at BLKSIZ þ 1200 þ 1234 ??? in _FCOPY at BLKSIZ 1300 þ 1306 ??? in cd$ at @ATR 1447 ??? in .DIR$ at @ATR 1464 ??? in .FNAME at @ATR 1468 ??? in .FNAME at @ATR þ 1483 ??? in .DIR at @ATR þ 1487 ??? in .DIR at @FSIZE 1489 ??? in .DIR at @FDATE þ ÞÞÞäin .DIR at @FTIME 1491 ??? in .DIR at @CDATE þ ÞÞÞäin .DIR at @CTIME 1500 þ 1512 ??? in lsdirs at BLKSIZ 1516 ??? in lsdirs at @ATR þ 1558 ??? in .UTIME at @FTIME 1574 ??? in (.LIST) at @FSIZE 1575 ??? in (.LIST) at @FDATE þ 1600 þ 1607 ??? in .FILES at #files 1760 ??? þI'm having trouble building v5.2 from the dropbox .SPIN file (TACHYON5r2.SPIN) in PropellerIDE. Error message is:
v4.7 builds fine, v5.1 reports the same error. I'm using PropellerIDE version 0.33.3 on linux.
Cheers
A>
Regarding: "pub ESC? ( -- flg \ true if an escape has been pressed ) "
From a few examples and the reference page ESC? sets a flag when ESC is pressed, but if I try to use it it get:
ESC? ??? in PROX at ESC?
I have looked through the EXTEND code but can not see it, has it changed? Is there a current reference?
My current config:
*** MODULES *** Propeller .:.:--TACHYON--:.:. Forth V5r2 NEON 520180323.2200
3A2A: EASYFILE SDHC card + FAT32 Virtual Memory File System V1.2 171024-0000
3386: TOOLS DEV TOOLS
30DE: CHARLCD BUFFERED CHARACTER LCD
1982: EXTEND Primary extensions to TACHYON V5.2 kernel - 180322-1300
thanks
I'm not sure what code you may be compiling but you can easily define an ESC?
I am still getting my head around Forth, from what i can tell ESC? is in quite a few examples and is mentioned in the "Glossary of Tachyon words"
https://docs.google.com/document/d/1gkSgKPYidRnhLaqgT7gfNyuhiTuWt-9tNRQC4epr7k4/edit#heading=h.i7dx3hi7lcst
I may be misunderstanding how to read this but I think it is telling me it comes from EXTEND?
I appreciate the assistance.
best regards
Lee
Hi Lee,
Some versions may have certain words, others don't, either because there are many ways of doing this, or they were't used often enough. However there are definitely core words that are essential too that are always there. ESC? is not of the latter but because it is so simple I can add it back into EXTEND
I have gone back to the V4 and had a look around but could not find but again I am a babe in the woods so may be looking in the wrong places.
No problems, I added it back in and it is useful for terminating listings and test loops though. Each new version of Tachyon is done for a reason, and V5 builds on the V4 which changed over from bytecode to wordcode, mainly for handling the address space more efficiently for larger programs. But V5 compacts the dictionary a fraction more, simplifies some other things while making more hub space available for code. This way V5 can squeeze in EASYFILE and EASYNET while avoiding having to compress the dictionary into upper EEPROM. Besides, it has a simple decompiler built-in! Type "HELP ESC?" and it will decompile the word and pause on any EXIT it comes across where you can decide to let it continue or hit a key to stop it.
: D2 BEGIN 16 HIGH 17 LOW 200 ms 16 LOW 17 HIGH 200 ms ESC? UNTIL ;
I am expecting the pins to toggle until I hit ESC but it never stops.
I thought my terminal (TeraTerm) may not be sending escape, i tried CTL[ as well so I added this:
pub XXX? ( -- flg) lastkey C@ 88 = ;
and changed the test to:
: D2 BEGIN 16 HIGH 17 LOW 200 ms 16 LOW 17 HIGH 200 ms XXX? UNTIL ;
I am expecting the loop to break when I hit X but again no joy.
What am I doing wrong?
thanks!
thanks
... : D4 BEGIN lastkey C@ DECIMAL . 200 ms ESC? UNTIL ;
...
... D4 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 1
The advantage of reading lastkey was that it was non-invasive by not using up a character from the serial receive buffer. However if you don't mind that then define ESC? as:
Test it out...
I have it running now, thank you.
I have noticed that the delays are out, eg a 20 mS is measuring as 16mS, a 100 mS is measuring as 80 mS.
I am running a 96Mhz clock.
best regards
Lee
Do you have a 'roadmap' for where you want to take v5? The reason I ask is that I think Tachyon deserves a much higher profile (along with the Propeller itself!), and I wonder if the key to that might be documentation. I note that you have a couple of documents, "Introduction to Tachyon" and the Tachyon glossary, which would be an excellent place to start for a full manual. I'd be happy to assist. What are your thoughts?
I'm currently in the middle of designing a new hardware revision of my data acquisition system, but I expect to be making much fuller use of Tachyon soon!
If you want you can edit those documents and others, I'd be happy to have some help. I can pm you an edit link if you like.
But now that P2 is so close I wonder whether I would use P1 in many new designs. This may be the end of the Tachyon road for the P1 but a whole new super expressway for the P2. Nonetheless, Neon is a powerful and refined implementation of Tachyon for the P1 and even if it doesn't have the profile it should have along with the Propeller, we are not the ones missing out
A couple of questions regarding your VGA Breakout code:
1. TABLE - this keyword doesn't appear in the glossary. Looking the code, I'm assuming it copies the a hub memory address to a variable, which can be used as a reference to the series of compiled literals that follow? And that the address corresponds to the address of the first compiled literal in hub memory?
So in this case, the variable 'gamepal' holds the address of the compiled literal $0800.0800, and you access elements in the table by adding an offset to gamepal?
2. colors - appears to be the destination for a copy of the gamepal table, but I don't see a definition for colors anywhere? Cheers
Thanks for your help thus far. I took some time out from PCB routing to sit down with Tachyon again. I've modified the stepper code to be able to step in either direction, e.g. or
I'm also driving the stepper in half-step mode as the manufacturer recommends, which means there are 8 phase patterns per step. I created a table of the phase patterns, and index from the start to the end of vice versa, depending on direction:
0 := CCW 1 := CW --- utility function to write a byte to an I2C chip register pub IO! ( data reg -- ) $40 pub TCA! ( data reg dev -- ) <I2C I2C! I2C! I2C! I2C> ; word ratems --- set the step but also make sure the I/O is initialized pub MTR.RATE ( ms -- ) ratems W! --- Set I2C bus pins for stepper card, initialise TCA9535 ports 0 & 1 to output, set all outputs low pub MTR.INIT ( -- ) 7 6 I2CPINS 0 2 IO! 0 3 IO! 0 6 IO! 0 7 IO! ; --- Bit patterns for driving 28BYJ-48 in half-step mode TABLE phases $02 | $06 | $04 | $0C | $08 | $18 | $10 | $12 | --- set motor phase outputs on port 0 of TCA9535 and delay for programmed speed pub MTR.PHASE ( phase -- ) 2 IO! ratems W@ ms ; --- Do a single complete step pub MTR.STEP ( -- ) 8 SWAP 0= IF phases FROM 1 ELSE phases 7 + FROM -1 THEN BY FOR I C@ MTR.PHASE NEXT ; --- Turn on activity LED for motor, run for <steps> steps in the specified direction (CW or CCW), at set rate between phase outputs pub MTR.STEPS ( steps dir -- ) 3 HIGH SWAP FOR DUP MTR.STEP NEXT DROP 0 MTR.PHASE 3 LOW ;I would be interested in your comments, as I guessing this can be done more efficiently than the approach that I have taken.
Cheers
Andrew
P.S. I'm compiling a list of glossary edits, once it is sufficiently long we might go over it...
Well done so far, please bear with me as I try to share some of the things I have found that helps with writing Forth software.
Compiled languages encourage you to use all_kinds_of_long_names but one of the things I like about Forth is that the name becomes part of the language. More correctly though, the name becomes the language in that you end up programming with these new names. If well chosen and thought out then it will feel and read natural but all too often the structures of conventional languages that were forced upon us are in turn forced upon Forth so that the Forth vocabulary has not been "extended" or enhanced at all. What we end up with is "code" in need of decoding.
So you find that to move the motor slowly in a clockwise direction you end up typing/coding:
20 MTR.RATE
200 CW MTR.STEPS
Now that's not so bad, It works, but could you express what you want to do more naturally like:
SLOW 200 RIGHT
then perhaps
FASTER 400 LEFT
This is also a lot easier to type in interactively and now the language has been extended and enhanced. The code can also be its own comments.
Add TURNS or DEGREES so you can say LEFT 45 DEGREES or 8 TURNS if positioning is more important than speed and movement.
I am running:
I am trying to run the example:
I am getting errors with the COGREG! and the (WAITPEQ) and (WAITPNE).
I can not see where COGREG! is defined, I can find references in the documentation but it does not seem to work.
I am not sure if the WAITPEQ etc have changed names, are they now WAITHI and WAITLO?
thank you.