Forgot your password?
typodupeerror
User Journal

Journal: Space Patrol: Almost there

Journal by Mr Z

Production continues to move forward. The ROMs are all programmed, and the CPLDs are close behind. PCBs are getting fabbed right now, and when they come it, they'll be stuffed shortly thereafter. If we hurry, we may be able to ship something by year's end.

Manual and overlay are finalized, box layout's finalized, trial box looks good... It's all becoming real!

Classic Games (Games)

Journal: Space Patrol: The Final Stretch!

Journal by Mr Z

Space Patrol, as mentioned here is entering the home stretch. I figure I have a week of solid coding left to do on it, meaning I'll probably be finished with it by the end of January. (It is a hobby.) Currently, I'm adding the final few missing pieces:

  • Scoring, including high-score tracking.
  • Game over and continues.
  • Splash screen.
  • General cleanups, including maybe a few cycle-count reductions.

Once I get the game done, I'm going to begin working on an extra secret special feature that, if I can get it to work, will make this one of the most unique Intellivision games of all time. Whoo hoo!

User Journal

Journal: Space Patrol! 3

Journal by Mr Z

I'm actually working on Space Patrol again! It pretty much sat on the shelf for 2-3 years. I've dusted it off and really put some work into it.

I made a post about it on my LiveJournal. David Harley, who is helping me with level design, has his own page up on it on his website.

So far we have 2 complete worlds finished. We're adding 6 more. The engine itself is 90-95% complete. Now it's for all the work around the edges--sound effects, level data, transitions, scoring and dying.

User Journal

Journal: Good, working code heading for the wastebin *sniff* 2

Journal by Mr Z

Shameless rambling about a personal project...

I'm working on a video game that is similar to Moon Patrol. It's a game for the venerable, underpowered Intellivision video game system. You can find more information and a developer's kit on the Intellivision on my SDK-1600 page.

Well, I spent Thanksgiving up through now working on various aspects of the game. One large piece was working out a method for controlling the 'bad guys.' I figured I'd write a small interpreter that would read a "Bad Guy Motion Program", and move the guys around. I even wrote a very simplistic assembler to assemble the command words for the interpreter. The whole thing actually works pretty reliably.

Turns out, though, that I need something else. My current approach is 'timer based.' That is, the motion programs are essentially of the form "Set velocity to dx,dy", "wait XX tics", "set velocity to dx,dy", "wait XX tics", "fire", "wait XX tics", "leave the screen." The problems are twofold:

  • In Moon Patrol, at least, the bad guys are cued into the level and cued out of the level by location. That is, when I cross a particular marker, the bad guys show up. When I cross a different particular marker, they leave. I'd like similar behavior for my game, but the timer method causes the bad guys to leave too soon if the tank is driving slowly, and too late if the tank is driving quickly.
  • The bad guys always move in EXACTLY the same pattern. In Moon Patrol, at least, the bad guys move consistently only if you move consistently. That is -- their movements seem rule based, not hardcoded.

So, I'm going to have to scrap my current interpreter. I'm torn between writing a new interpreter (or extending this one) to allow me to write rule-based motion programs for the bad guys, or just writing the rule-based programs directly in CP-1600 assembly. The latter is less convenient, I think, but will likely be smaller and is more likely to fit in my cycle budget.

So it appears I will probably scrap the interpreter.

The interpreter wasn't much, but it was something. And it worked. And I was proud of it. It provided some neat concepts, such as a looping construct that combined initialization, decrement, and branching into a single instruction. FWIW, here's how the looping instruction worked:

I offer two loop-counters per program. The LOOPA instruction uses the 'A' loop counter, and LOOPB uses the 'B' loop counter. The counters are initialized to 0 at the start of the program. Upon reaching an instruction of the form "LOOP[AB] label, count", the loop instruction checks the corresponding loop counter. If it's 0, it's initialized to the specified count and a branch is made to the label. If it's non-zero, it's decremented, and if the result is non-zero the branch is taken. That's it.

Here's an example motion program that causes a bad-guy to move in a box-shaped spiral:

####
## sample motion programs
####
MP_BOX PROC

@@box
. . . . GO (3) rg1
. . . . GO (3) dn1
@@boxb GO (3) lf1
. . . . GO (2) up1
. . . . LOOPA @@box, 5 # do the box 5 times

. . . . GO (2) rg2
. . . . GO (2) dn2
. . . . LOOPB @@boxb, 4 # slide over and do this 4 times total.
. . . . GO (9) lf3
. . . . DIE
. . . . ENDP

The names "lf1", "dn2" represent directions and velocities. The number in parentheses is number of ticks to wait after the instruction executes (at a 30Hz tick-rate). Nothing particularly mind-bending.

Now it looks as though I might bin all this code and write up motion programs in straight assembly. *sigh*

Ah well, here's the interpreter (or at least the heart of it) for posterity.

;; ======================================================================= ;;
;;. UPBGM -- Update Bad Guy Motion.. Called on a 30Hz tick.. . . . . . . . ;;
;; ======================================================================= ;;
UPBGM. PROC
. . . . PSHR. . R5

. . . . MVII. . #BGMPTBL, R4
. . . . CLRR. . R1

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. Iterate through the motion programs . . . . . . . . . . . . . . ;;
. . . . ;;. R0 -- scratch. . . . . . . . . . . . . . . . . . . . . . . . . ;;
. . . . ;;. R1 -- Slot # for MOB (within group 1). . . . . . . . . . . . . ;;
. . . . ;;. R2 -- Program counter for current record . . . . . . . . . . . ;;
. . . . ;;. R3 -- scratch. . . . . . . . . . . . . . . . . . . . . . . . . ;;
. . . . ;;. R4 -- Pointer to BGMP state tables . . . . . . . . . . . . . . ;;
. . . . ;;. R5 -- scratch. . . . . . . . . . . . . . . . . . . . . . . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
. . . . B . . . @@first_bgmp. . ; skip first INCR R1

@@next_bgmp:
. . . . INCR. . R1. . . . . . . ; next slot #

. . . . CMPI. . #5,. . R1. . . ; Is this the end?
. . . . BGE . . @@done_bgmp . . ; Yes, done.

@@first_bgmp:
. . . . SDBD
. . . . MVI@. . R4,. . R2. . . ; Get program counter
. . . . TSTR. . R2. . . . . . . ; If it's zero, motion program is inactive
. . . . BNEQ. . @@active_bgmp. ; Non-zero, it's active.

. . . . ADDI. . #3,. . R4. . . ; Skip rest of record.
. . . . CMPI. . #BGMPTBL+25, R4 ; Is this the end?
. . . . BLT . . @@next_bgmp . . ; No, do the next one.
. . . . B . . . @@done_bgmp . . ; Yes, done.

@@active_bgmp:
. . . . MOVR. . R4,. . R3
. . . . MVI@. . R4,. . R0. . . ; Get delay counter for this motion program
. . . . ADDI. . #2,. . R4. . . ; point to next record
. . . . DECR. . R0. . . . . . . ; Count down the delay counter.
. . . . MVO@. . R0,. . R3. . . ; Store the updated delay counter

. . . . BPL . . @@next_bgmp. . ; If not expired, go to the next program.

@@expired:
. . . . MVI@. . R2, . . R0. . . ; Get next program word.
. . . . INCR. . R2
. . . . MVO@. . R0, . . R3. . . ; Store out lower byte as delay count.
. . . . XOR@. . R3, . . R0. . . ; Clear delay byte
. . . . SWAP. . R0. . . . . . . ; Extract command byte.

. . . . SUBI. . #64,. . R0. . . ; Less than 64?. It's a motion command.
. . . . BMI . . @@cmd_move. . . ;
. . . . SLL . . R0,. . 1
. . . . ADDR. . R0,. . PC. . . ; Otherwise, vector to the other commands.
. . . . B . . . @@cmd_loopa . . ;. 0 : LOOPA
. . . . B . . . @@cmd_loopb . . ;. 1 : LOOPB
. . . . B . . . @@cmd_jump. . . ;. 2 : JUMP
. . . . B . . . @@cmd_fire. . . ;. 3 : FIREA
. . . . B . . . @@cmd_fire. . . ;. 4 : FIREB
. . . . B . . . @@cmd_fire. . . ;. 5 : FIREC
. . . . B . . . @@cmd_die . . . ;. 6 : DIE
. . . . B . . . @@cmd_attr. . . ;. 7 : ATTR
. . . . B . . . @@cmd_seths . . ;. 8 : SETHS
. . . . B . . . @@cmd_clrhs . . ;. 9 : CLRHS
; . . . B . . . @@cmd_wait. . . ; 10 : WAIT aka. NOP.

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. WAIT/NOP:. Just set the delay counter and go to next command.. ;;
. . . . ;; --------------------------------------------------------------- ;;
@@cmd_wait:

@@finish_cmd:
. . . . SUBI. . #5,. . R4. . . ; Rewind to program pointer in record.
. . . . MVO@. . R2,. . R4. . . ; \.
. . . . SWAP. . R2. . . . . . . ;. |-- Store program counter as double-byte
. . . . MVO@. . R2,. . R4. . . ; /. . data
. . . . ADDI. . #3,. . R4. . . ; point to next record.

. . . . B . . . @@next_bgmp. . ; No, do the next one.

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. LOOPA/LOOPB:. Count down the loop counter.. If it goes -ve, . . ;;
. . . . ;;. we're just starting the loop so initialize it and branch. . . . ;;
. . . . ;;. If it goes 0, the loop is just finishing, so fall through.. . . ;;
. . . . ;;. Otherwise, store updated counter and branch.. . . . . . . . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@cmd_loopb:
. . . . INCR. . R3. . . . . . . ; Point to loop counter B
@@cmd_loopa:
. . . . INCR. . R3. . . . . . . ; Point to loop counter A

. . . . MVI@. . R2,. . R0. . . ; Read argument to loop into R0
. . . . INCR. . R2. . . . . . . ;

. . . . MVI@. . R3,. . R5. . . ; read loop counter
. . . . DECR. . R5. . . . . . . ; count it down
. . . . BMI . . @@init_loop . . ; if it goes negative, we need to init it.
. . . . BEQ . . @@done_loop . . ; if it goes to zero, it's expiring

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. +ve case:. Store updated counter and branch.. . . . . . . . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@exec_loop:. . . . . . . . . . ; otherwise it was an active loop.
. . . . MVO@. . R5,. . R3. . . ; store updated loop counter

@@loop_jump:
. . . . ANDI. . #$FF,. R0. . . ; execute the jump.
. . . . XORI. . #$80,. R0. . . ;
. . . . SUBI. . #$80,. R0. . . ;
. . . . ADDR. . R0,. . R2. . . ;

. . . . B . . . @@finish_cmd. . ; finish this command.

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. -ve case:. Initialize the loop counter and branch.. . . . . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@init_loop:
. . . . SWAP. . R0. . . . . . . ;
. . . . MVO@. . R0,. . R3. . . ; initialize the loop counter.
. . . . SWAP. . R0. . . . . . . ;
. . . . B . . . @@loop_jump

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. Zero case:. Zero out the counter and don't branch.. . . . . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@done_loop:
. . . . MVO@. . R5,. . R3. . . ; store updated loop counter
. . . . B . . . @@finish_cmd. . ; and that's it.

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. JUMP:. Just read the argument into the program counter.. . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@cmd_jump:
. . . . MVI@. . R2,. . R2
. . . . B . . . @@finish_cmd. . . . . . ; Done.

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. MOVE:. Read the X velocity, Y velocity into SPXYV table.. . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@cmd_move:

. . . . MOVR. . R1,. . R5. . . . . . . ;\
. . . . ADDR. . R5,. . R5. . . . . . . ; |-- convert slot number to sprite
. . . . ADDI. . #SPXYV, R5. . . . . . . ;/. . X/Y velocity table pointer

. . . . ADDI. . #@@xvel_tbl + 64, R0. . ; Index into X velocity lookup table
. . . . MOVR. . R0,. . R3

. . . . MVI@. . R3,. . R0. . . . . . . ; Get X velocity from lookup
. . . . MVO@. . R0,. . R5. . . . . . . ; Write it to X vel for this sprite

. . . . ADDI. . #@@yvel_tbl - @@xvel_tbl, R3. . ; Index into Y vel lookup tbl
. . . . MVI@. . R3,. . R0. . . . . . . ; Get Y velocity from lookup
. . . . MVO@. . R0,. . R5. . . . . . . ; Write it to Y vel for this sprite.

. . . . B . . . @@finish_cmd. . . . . . ; Done.

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. FIREA:. Make this bad guy drop a bomb. . . . . . . . . . . . . ;;
. . . . ;;. FIREB:. Make this bad guy fire a crater-maker. . . . . . . . . ;;
. . . . ;;. FIREC:. Make this bad guy shoot horizontally to the left.. . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@cmd_fire:.
. . . . MVII. . #@@firetype - 6, R3
. . . . ADDR. . R0,. . R3
. . . . MVI@. . R3,. . R0. . . ; read X/Y velocity
. . . . INCR. . R3
. . . . MVI@. . R3,. . R3. . . ; read sprite attribute index
. . . . CALL. . BGFIRE
. . . . B . . . @@finish_cmd

@@firetype:
. . . . DECLE. ((VBIAS + $40) SHL 8) + (VBIAS + $00) ; drop the bomb
. . . . DECLE. SPATBL.b2

. . . . DECLE. ((VBIAS + $40) SHL 8) + (VBIAS + $40) ; drop the crater maker
. . . . DECLE. SPATBL.b1

. . . . DECLE. ((VBIAS + $00) SHL 8) + (VBIAS - $80) ; fire left.
. . . . DECLE. SPATBL.b4

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. DIE:. Time for this guy to die.. . . . . . . . . . . . . . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@cmd_die:
. . . . CALL. . BGKILL
. . . . B . . . @@finish_cmd

. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. ATTR:. Set the Sprite Attribute for this sprite.. . . . . . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@cmd_attr:
. . . . MVI@. . R2, . . R0. . . ; Get attribute
. . . . INCR. . R2
. . . . MOVR. . R1, . . R3. . . ; \
. . . . ADDI. . #SPAT,. R3. . . ;. |-- Store attribute number into slot
. . . . MVO@. . R0, . . R3. . . ; /. . in sprite attribute table.

. . . . B. . . @@finish_cmd

. . . .
. . . . ;; --------------------------------------------------------------- ;;
. . . . ;;. CLRHS:. Set the horizontal-scroll bit for this sprite. . . . . ;;
. . . . ;; --------------------------------------------------------------- ;;
@@cmd_clrhs:

. . . . ;; ---------------------------------------------------------------- ;;
. . . . ;;. SETHS:. Set the horizontal-scroll bit for this sprite. . . . . ;;
. . . . ;; ---------------------------------------------------------------- ;;
@@cmd_seths:
@@do_hs:
. . . . MOVR. . R1, . . R3
. . . . ADDI. . #@@shl_tbl, R3
. . . . MVI@. . R3, . . R3
. . . . MOVR. . R3, . . R0
. . . . COMR. . R3
. . . . AND . . SPHSCR, R3
;;. . . ADCR. . PC
. . . . XORR. . R0, . . R3
. . . . MVO . . R3, . . SPHSCR

. . . . B . . . @@finish_cmd

@@shl_tbl:
. . . . DECLE. 1, 2, 4, 8, 16
. . . .

@@done_bgmp:
. . . . PULR. . PC

@@xvel_tbl:
. . . . DECLE. VBIAS - $20, VBIAS - $40, VBIAS - $60, VBIAS - $7F. ; lf0..lf3
. . . . DECLE. VBIAS + $20, VBIAS + $40, VBIAS + $60, VBIAS + $7F. ; rg0..rg3
. . . . DECLE. VBIAS,. . . VBIAS,. . . VBIAS,. . . VBIAS. . . . ; up0..up3
. . . . DECLE. VBIAS,. . . VBIAS,. . . VBIAS,. . . VBIAS. . . . ; dn0..dn3

. . . . DECLE. VBIAS - $20, VBIAS - $40, VBIAS - $60, VBIAS - $7F. ; lu0..lu3
. . . . DECLE. VBIAS + $20, VBIAS + $40, VBIAS + $60, VBIAS + $7F. ; ru0..ru3
. . . . DECLE. VBIAS - $20, VBIAS - $40, VBIAS - $60, VBIAS - $7F. ; ld0..ld3
. . . . DECLE. VBIAS + $20, VBIAS + $40, VBIAS + $60, VBIAS + $7F. ; rd0..rd3

. . . . DECLE. VBIAS,. . . VBIAS - $08
. . . . ;; directions 33 .. 63 not defined yet.. in no hurry....

@@yvel_tbl:
. . . . DECLE. VBIAS,. . . VBIAS,. . . VBIAS,. . . VBIAS. . . . ; lf0..lf3
. . . . DECLE. VBIAS,. . . VBIAS,. . . VBIAS,. . . VBIAS. . . . ; rg0..rg3
. . . . DECLE. VBIAS - $20, VBIAS - $40, VBIAS - $60, VBIAS - $7F. ; up0..up3
. . . . DECLE. VBIAS + $20, VBIAS + $40, VBIAS + $60, VBIAS + $7F. ; dn0..dn3

. . . . DECLE. VBIAS - $20, VBIAS - $40, VBIAS - $60, VBIAS - $7F. ; lu0..lu3
. . . . DECLE. VBIAS - $20, VBIAS - $40, VBIAS - $60, VBIAS - $7F. ; ru0..ru3
. . . . DECLE. VBIAS + $20, VBIAS + $40, VBIAS + $60, VBIAS + $7F. ; ld0..ld3
. . . . DECLE. VBIAS + $20, VBIAS + $40, VBIAS + $60, VBIAS + $7F. ; rd0..rd3

. . . . DECLE. VBIAS,. . . VBIAS
. . . . ;; directions 33 .. 63 not defined yet.. in no hurry....

. . . . ENDP

And there it is. (The dots are added to preserve the original formatting, although it appears I'm not entirely successful. Ah well... it's pretty close.)

--Joe

User Journal

Journal: This whole Friend/Foe/Fans/Freaks thing 9

Journal by Mr Z

Heh. My journal -- probably a bit like pissing in the ocean. It feels good, but nobody notices. ;-)

Anyway, I was clicking through my 'user' pages today (clicking aimlessly, noticing how much snappier Moz 1.1 feels relative to 1.0 on this Solaris box), and noticed that whoa, I have fans!

Up until now, I've entirely avoided the whole friend/foe mechanism. For awhile now, I've checkmarked "unwilling to moderate", and "hide scores". I also browse at '0'. (I used to browse at -1, but it really was too noisy.) Moderation just doesn't seem all that reliable or meaningful, and so I largely ignore it. Too much $3 crack going around. (I don't mind being modded up occasionally though -- if my post is really worth "promoting" to those who browse at higher thresholds.)

The friend/foe mechanism just seemed like another bad idea. It seemed to have the potential to turn Slashdot into a popularity contest like highschool. Either that, or it'd lead to even greater segmentation of the posting populace, with everyone auto-modding up their friends and ignoring everyone else. A sort of "show me only what I agree with and want to hear" mechanism that does little to encourage intelligent discourse and does everything to silo individual cliques. So I just ignored it.

I've thought it over more, now. I certainly don't mind if others mark me as friend or foe. (Although I'd certainly wonder why someone might choose to mark me 'foe'.) I don't have any qualms about marking someone 'friend' anymore either. I don't plan to use the moderation adjustment facility though. I ignore comment moderations for the most part anyway, and I see no reason to change that.

So, I've marked a couple people 'friends' today. Some are longtime acquaintences (either in meatspace or on /.). Others were in my list of 'fans'. And if someone posts a helpful response to one of my posts, or posts something I think is particularly insightful, I may just mark 'em 'friend.' Or maybe not. Who knows.

Well, enough rambling. This post is so long, I've received about 9 or 10 spam emails in the meantime:

Mutt 1.0i@caffeine:=spool/incoming*(reverse-threads) [3173/3173(1% of 71M)][b=0|
. 1 . NDL Sep 26 Jacelyn0828c77@h ( . .25) [Spam] High investment potential ret
. 2 . NDL Sep 26 Rockne Renate . .( . 117) [Spam] Why Wait... Rates Below 4.75%
. 3 . NDC Sep 26 cedrickmills0123 ( . .11) [Spam] how would you like to increas
. 4 . NDL Sep 27 Zared Lucio . . .( . .71) [Spam] Stop going bald, time is stil
. 5 . NDL Sep 27 Brittany Bond . .( . 130) [Spam] In Debt? Get Help!---Free Quo
. 6 . NDL Sep 26 Frayne Clio . . .( . 167) [Spam] Coral Calcium Discovery Lette
. 7 . NDL Nov 25 refi_loans_2@yah ( . .70) [Spam] dont loose sales - take credi
. 8 . NDL Sep 27 celularshop@hotm ( . .38) [Spam] Fwd: Faith is all that is req
. 9 . NDL Sep 27 brellis10@hotmai ( . .54) [Spam] Attn: Do it yourself.

As Stimpy would say: Joy!

--Joe

Some people claim that the UNIX learning curve is steep, but at least you only have to climb it once.

Working...