AA Homebrews for 2600 Nearing Activision Titles

[Jacob Rose]

After seeing Al's cool presentation for the Midwest Gaming Classic, I did a search for Activision titles in the AA database. I count 47 original titles (is that right?). AA currently has 41 unique homebrews in the store (again, if I count correctly). In very short order, AA will have produced more original titles for the 2600 than Activision!

[Happy_Dude]

Thats cool.

 

Now I'm really going to finally finish Master Mind so I can get into the first 47

 

 

That or Squares (hush hush tb game)

[bjk7382]

AA currently has 41 unique homebrews in the store (again, if I count correctly).

Did you count the ones that aren't in the store? Such as Edtris, Qb, Cubis, Tetris 26, and others.

[Happy_Dude]

I don't think AA ever sold Edtris or Tetris 26.

 

 

edit: nope didn't think so http://www.atariage.com/cart_page.html?SoftwareLabelID=866

Edited June 8, 2006 by Happy_Dude

[Albert]

I don't think AA ever sold Edtris or Tetris 26.

That is correct.

 

..Al

[Cybergoth]

I think there's just two titles that were once sold in the 2600 homebrew section that are missing now: Qb and Mondo Pong.

[Jacob Rose]

I think there's just two titles that were once sold in the 2600 homebrew section that are missing now: Qb and Mondo Pong.

 

Oh, yeah - why wasn't Mondo Pong renamed (a la Joust Pong)? That looked like a neat game.

 

So...is 43 the right count?

Edited June 8, 2006 by Jacob Rose

[Cybergoth]

I think 43 is one too many:

 

- There's 42 entries in the store.

- Remove both Ebivision packs

- Add Qb and Mondo Pong.

[vdub_bobby]

Depends on if you count Holiday carts also.

[Jacob Rose]

I think 43 is one too many:

 

- There's 42 entries in the store.

- Remove both Ebivision packs

- Add Qb and Mondo Pong.

 

Ah, I missed the second Ebivision pack. Also, I guess the new games released this weekend weren't on the list, were they? They're all original except Wolfenstein VCS II (it's an extensive hack, not original, like its predecessor, right?), so that's three new original titles (Superbug, Rainbow Invaders, and Conquest of Mars), isn't it? Bringing the total to 45?

 

How about the other number? 47 original titles for Activision? I didn't include Kabobber and Thwocker (and Venetian Blinds), since they weren't released (kinda like not counting AA's "In Development" page...).

[Jacob Rose]

Depends on if you count Holiday carts also.

 

Isn't yours is the first original game to be released that way? Holiday Qb is just Qb with graphic changes. There was one previous Holiday Cart that I know of (I'm a relative newcomer), but that was just a graphic demo. Then again, I guess the Boing! cart is just a graphic demo, too (would it be possible to hack it to play Pong with the ball? ). So, 45 - Boing! + Reindeer Rescue = 45 again?

[Cybergoth]

Depends on where you want to draw the line. I personally keep a list where I don't count non-games, compilations, demos, and hacks and I also make a difference between basic and assembly titles.

[Jacob Rose]

Depends on where you want to draw the line. I personally keep a list where I don't count non-games, compilations, demos, and hacks and I also make a difference between basic and assembly titles.

 

Ah, compilations, that's an interesting exception. So, no mini-game 2005, but what about Swoops!?

 

So how many do you count as "original titles," Manuel?

[Albert]

I think 43 is one too many:

 

- There's 42 entries in the store.

- Remove both Ebivision packs

- Add Qb and Mondo Pong.

 

Ah, I missed the second Ebivision pack. Also, I guess the new games released this weekend weren't on the list, were they? They're all original except Wolfenstein VCS II (it's an extensive hack, not original, like its predecessor, right?), so that's three new original titles (Superbug, Rainbow Invaders, and Conquest of Mars), isn't it? Bringing the total to 45?

Correct, Wolfenstein VCS: The Next Mission is an extensive hack. The other games have not yet been added to the store (I'll be working on that tonight), and they'll be added to the main database sometime over the weekend.

 

..Al

[vdub_bobby]

Also, I was under the impression that Fred wasn't going to sell Superbug on carts...? So I don't know that you should count that particular egg before it hatches, so to speak.

 

EDIT: And how many homebrew compilations have been sold? SWOOPS! and the 2005 Minigame Compo and...? Any others?

Edited June 8, 2006 by vdub_bobby

[Albert]

Also, I was under the impression that Fred wasn't going to sell Superbug on carts...? So I don't know that you should count that particular egg before it hatches, so to speak.

 

EDIT: And how many homebrew compilations have been sold? SWOOPS! and the 2005 Minigame Compo and...? Any others?

Those are the only two I've had any hand in publishing. I personally think these two should count, since the individual games in each collection have not been published individually, and both compilations have unique labels and manuals.

 

..Al

[+batari]

Depends on where you want to draw the line. I personally keep a list where I don't count non-games, compilations, demos, and hacks and I also make a difference between basic and assembly titles.

 

Ah, compilations, that's an interesting exception. So, no mini-game 2005, but what about Swoops!?

 

So how many do you count as "original titles," Manuel?

Agreed... I can see not counting Swoops! as 3 and the Minigame Multicart as eightseven, but they should at least be counted as one each!

[vdub_bobby]

...and the Minigame Multicart as eightseven...

[+batari]

Also, I was under the impression that Fred wasn't going to sell Superbug on carts...? So I don't know that you should count that particular egg before it hatches, so to speak.

I fully plan for Superbug to be sold on cart.

 

However, Maze Craze 2 will very likely never be sold, since it's a Supercharger game. Selling cassettes/CD's seems pointless, and I expect that the Chimera will have a fairly high price point, making a cart release impractical. Though if anyone wants to make a Chimera-cart out of it anyway, they can sell it royalty-free, or even include it on their own Chimera-cart releases as a bonus if they want.

[supercat]

However, Maze Craze 2 will very likely never be sold, since it's a Supercharger game. Selling cassettes/CD's seems pointless, and I expect that the Chimera will have a fairly high price point, making a cart release impractical. Though if anyone wants to make a Chimera-cart out of it anyway, they can sell it royalty-free, or even include it on their own Chimera-cart releases as a bonus if they want.

 

I wonder if Maze Craze 2 could work with a maze stored in EEPROM? Adding an EEPROM to one of Al's carts would be pretty easy (just an 8-pin DIP, a resistor, and some wires)--certainly a lot simpler than a full-fledged 4A50 or Chimera cart. Playing the maze probably wouldn't be too hard. Generating it would be the fun bit.

[vdub_bobby]

Also, I was under the impression that Fred wasn't going to sell Superbug on carts...? So I don't know that you should count that particular egg before it hatches, so to speak.

I fully plan for Superbug to be sold on cart.

 

However, Maze Craze 2 will very likely never be sold, since it's a Supercharger game. Selling cassettes/CD's seems pointless, and I expect that the Chimera will have a fairly high price point, making a cart release impractical. Though if anyone wants to make a Chimera-cart out of it anyway, they can sell it royalty-free, or even include it on their own Chimera-cart releases as a bonus if they want.

Whoops, my mistake. Guess I got the two confused.

 

I would buy a Maze Craze 2 CD, though.

[+batari]

Whoops, my mistake. Guess I got the two confused.

 

I would buy a Maze Craze 2 CD, though.

I hereby grant you the rights to make a Maze Craze 2 CD, then you can sell it to yourself.

 

I wonder if Maze Craze 2 could work with a maze stored in EEPROM? Adding an EEPROM to one of Al's carts would be pretty easy (just an 8-pin DIP, a resistor, and some wires)--certainly a lot simpler than a full-fledged 4A50 or Chimera cart. Playing the maze probably wouldn't be too hard. Generating it would be the fun bit.

The game could probably be adapted to RAM-less operation without too much trouble. It doesn't lean too heavily on the SC except for the 2k maze itself.

 

Timing will probably not be much of an issue in the game itself, as the game operates in an 8-frame loop so there's lots of time available for bit banging 3 or 4 bytes.

 

I suppose the practicality of maze generation would depend on how time-consuming a random access read or write would be. Right now maze generation will do > 4,000 writes and perhaps > 20,000 reads. How long do you suppose each EEPROM access would take?

[supercat]

How long do you suppose each EEPROM access would take?

 

I'm not sure exactly how I'd set up the EEPROM access. If I add another resistor and cap beyond what would otherwise be necessary, I could make things pretty fast. If it would fit in a 22V10 (might require board layout change, rather than a hand patch-mod), the optimal code sequence would probably be something like:

; Write byte stored in Y
cmp  $8FF ; Raise data then raise clock
cmp  $700 ; Leave data alone then lower clock
cmp  $800,y ; Set data from Y.7 then raise clock
cmp  $700  ; Leave data alone then lower clock
cmp  $900,y ; Set data from Y.6 then raise clock
cmp  $700   ; Leave data alone and lower clock
cmp  $A00,y ; Set data from Y.6 then raise clock
cmp  $700   ; Leave data alone and lower clock
cmp  $B00,y ; Set data from Y.6 then raise clock
cmp  $700   ; Leave data alone and lower clock
cmp  $C00,y ; Set data from Y.6 then raise clock
cmp  $700   ; Leave data alone and lower clock
cmp  $D00,y ; Set data from Y.6 then raise clock
cmp  $700   ; Leave data alone and lower clock
cmp  $E00,y ; Set data from Y.6 then raise clock
cmp  $700   ; Leave data alone and lower clock
cmp  $F00,y ; Set data from Y.6 then raise clock
cmp  $700   ; Leave data alone and lower clock

That's eighteen CMP instructions at four cycles each; 72 cycles. To read...

cmp $800 ; Assert data then raise clock
cmp $700 ; Leave data alone and lower clock
cmp $8FF ; Release data then raise clock
cmp $600 ; Do magic bankswitch (bit 7) and lower clock
lda $1FE0 ; Raise clock
cmp $600 ; Do magic bankswitch and lower clock
ora $1FE1 ; Raise clock
cmp $600 ; Do magic bankswitch and lower clock
ora $1FE2 ; Raise clock
cmp $600 ; Do magic bankswitch and lower clock
ora $1FE3 ; Raise clock
cmp $600 ; Do magic bankswitch and lower clock
ora $1FE4 ; Raise clock
cmp $600 ; Do magic bankswitch and lower clock
ora $1FE5 ; Raise clock
cmp $600 ; Do magic bankswitch and lower clock
nop
ora  #1 (or #0)
cmp $600 ; Do magic bankswitch and lower clock
ora  #1 (or #0)

A total of 74 cycles. The actual addresses may have to be worked out later, and I don't know if I'll be able to go quite that fast, but it's a nice target.

 

Reading a byte from a specified address requires a start cycle (not shown, but short), two or three byte writes (two for 2Kbyte or smaller EEPROMs; 3 for larger ones); a restart (also short), another write, and then a read. Successive bytes may be accessed with repeated read operations.

 

Writing to a specified address requires a start cycle, two or three byte writes for the address, and then a byte write for each byte of data (up to a max of 8-64, depending upon chip size), followed by a stop cycle.

 

As you can see, sequential access is pretty fast. Random access is very slow. So some creativity would be needed in the maze algorithm.

[+batari]

Reading a byte from a specified address requires a start cycle (not shown, but short), two or three byte writes (two for 2Kbyte or smaller EEPROMs; 3 for larger ones); a restart (also short), another write, and then a read. Successive bytes may be accessed with repeated read operations.

 

Writing to a specified address requires a start cycle, two or three byte writes for the address, and then a byte write for each byte of data (up to a max of 8-64, depending upon chip size), followed by a stop cycle.

 

As you can see, sequential access is pretty fast. Random access is very slow. So some creativity would be needed in the maze algorithm.

Probably the best way would be to use the 2600's RAM as a cache. If I can spare 32 bytes for this, it would give me a 16x16 bit chunk to work with at any time.

 

It appears that the access routine runs at 8 cycles per clock, or 150 kHz. However, I recall from messing with I2C drivers a few years ago that most wouldn't work reliably at > 100 kHz. I take it you can get faster devices now?

 

Anyway, maybe I'll do some experiments with the Memcard first to see how fast I can get maze generation to run.

[Cybergoth]

Depends on where you want to draw the line. I personally keep a list where I don't count non-games, compilations, demos, and hacks and I also make a difference between basic and assembly titles.

Ah, compilations, that's an interesting exception. So, no mini-game 2005, but what about Swoops!?

 

So how many do you count as "original titles," Manuel?

 

I count 37, that is including CoM, Rainbow Invaders and AStar, but not MG2005, Swoops, Boing!, Testcart, Synthcart and Solar Plexus. I also don't count Qb, Mondo Pong or Reindeer Rescue, as they're not unlimited available.

 

Agreed... I can see not counting Swoops! as 3 and the Minigame Multicart as eightseven, but they should at least be counted as one each!

 

Well, it's just me counting like that. Just add two for your own count

The official library never saw something like that (maybe you want to consider Pigs in Space coming close.) , so in a rally against Activision I'm not counting them.