Which Intellivision games are "winnable"?

[+thegoldenband]

Inspired by this topic for the 2600 (and a more recent version), I thought it might be interesting to make a list of which Intellivision games can actually be completed or beaten. I suppose there are a few different cases, ranging from very clear-cut to highly debatable:

 

- games that have an absolute ending and clear winning conditions (example: AD&D Cloudy Mountain)

- games that have an ending, but no clear winning condition per se (example: Skiing)

- games that have an ending, but which keep going afterwards or loop (example: Ice Trek)

- games that have no ending, but which might have a point maximum, rollover, or difficulty maximum (example: Venture)

 

Well, let's begin by listing the 2-player only games that can be 100% ruled out, since they only offer competitive head-to-head scoring:

 

Armor Battle

Boxing

Major League Baseball

NASL Soccer

NBA Basketball

NFL Football

NHL Hockey

Sea Battle

Tennis

Triple Action

 

I think that's all of them, though I haven't touched the ECS games or homebrews. In my next post I'll start listing some of the clearcut cases.

[+thegoldenband]

Games with a clear ending and winning conditions:

 

AD&D: Cloudy Mountain - complete the Crown of Kings on the highest difficulty level, and you've beaten the game

AD&D: Treasure of Tarmin - beat the Minotaur on the highest difficulty level (though you can optionally bypass him and keep going)

Atlantis - does this end after a certain number of days? If so, it seems to me that if you survive that period on the highest difficulty, you've beaten the game

Bomb Squad - disarm the bomb on the highest level, and beat the game

Fathom - complete all eight "seas" (levels), and I believe the game ends

Space Battle - fend off the computer on the highest difficulty level and you win; be sure to play the harder ROM variant

Stonix - beat all 100+ levels; are there any twists or hidden endings?

Super Pro Decathlon - successfully pass all events on the highest difficulty level

Vectron - play through all 99 levels and get a hidden message

 

(to be continued and edited)

[+thegoldenband]

Games with a clear ending, but with debatable winning conditions:

 

ABPA Backgammon - I assume that if you beat the computer on the highest difficulty level, you've essentially beaten the game. Confirm?

 

Checkers - see Backgammon. Does it matter which side you take?

 

Championship Tennis - beating the computer on the highest difficulty should equate to beating the game; I don't think the court surface changes the gameplay, so there's no reason to worry about all three surfaces -- is there?

 

Microsurgeon - no matter what, you get paid and get an ending, but to claim victory you should probably have to restore the patient to "good" status in every body part and exit safely, on the highest level (good luck with that!)

 

Pitfall - get all the treasures before time runs out; alternative winning condition: get them all before time runs out and get a perfect score

 

USCF Chess - see Checkers. If you want to be super-picky, you could say you have to beat the computer on the highest difficulty with the Black pieces.

 

(to be continued and edited)

[cparsley]

I question #3 and #4... if the game doesn't end, it's not "beatable" per say...

[+thegoldenband]

Understood. I'm trying to sketch out a continuum, starting with games that are unambiguously beatable like the AD&D games, to games where at a certain point you've seen everything there is to see, like Venture.

 

Or then you have a game like Blockade Runner, which has an ending but keeps going (and I don't think it gets any harder?) vs. River Raid, which ends when you reach a certain score (at least on the VCS version), but doesn't have an ending. Are there any other games that deliberately or inadvertently end when you reach a particular score? Do any Intellivision games have kill screens?

[+thegoldenband]

Games with a clear ending, but that are complicated by multiple winning conditions:

 

Reversi - like other board games, beating the computer on the highest difficulty is the first step, but there are three different boards and two different orders of play (you go first vs. computer goes first). If I understand correctly, beating the 10x10 board and going first (not second) is the hardest task, so that's a good minimum.

 

Royal Dealer - there are multiple card games, and you can play against up to three computer opponents. Presumably the game is harder with more adversaries, so winning a full session of every card game against three opponents probably equals beating the game.

 

Star Strike - there are five difficulty levels with an ending, and then a sixth that never ends. I'd say difficulty #5 counts as beating the game, and #6 is just a bonus.

 

Tower of Doom - there are a whole pile of scenarios and characters. The hardest is the Grail Quest, so that's one baseline, but does it matter which character you use? A super-hardcore player might say you have to beat all the scenarios with the weakest character (Waif) to win.

 

(to be continued and edited)

[+thegoldenband]

Games where the winning condition is arguably a certain score:

 

PBA Bowling - bowling a perfect game should equal beating the main game, no? There's also the Pick-Up Spares game, about which the manual more or less says that a score of 75 is equivalent to victory.

 

PGA Golf - as with any golf game, theoretical perfection would be a hole-in-one on every hole, but I doubt that's possible even with computer assistance; is there an absolute lower limit to the score?

 

River Raid - if the Intellivision game is like the Atari 2600 original, then the game freezes when you reach 999,999 pts.

 

Sharp Shot - there's clearly a maximum possible score in all four games, though we don't yet know what it is.

[roberto]

What about Swords & Serpents? It has a clear ending, though you can't actually kill the dragon... still, we can say the game ends when facing the dragon?

[roberto]

USCF Chess - see Checkers. If you want to be super-picky, you could say you have to beat the computer on the highest difficulty with the Black pieces.

)

 

:lol: Loved this!

[intvnut]

I'd put Utopia into an ambiguous category: It's a two player game, with somewhat complex scoring. But you can play it one player, and establish a "highest score" for a given set of parameters (number of turns, length of turn). So, since it's of finite but configurable length, I'd say it's "beatable" in some sense.

 

Interestingly, the scoring function has some strange internal overflows. As your "approval rating" nears 100%, if you have too many of certain items (never figured out the combination), you'll get a random rebel and your approval rating will drop into the 60s instantly. Utopia is a toughy to categorize fairly.

 

AD&D Treasure of Tarmin also doesn't seem to have a proper "ending". You can end the game the first time you catch up to the Minotaur, or you can choose not to engage him and keep exploring (all the way up to dungeon level 255). I have no idea if level 255 actually removes the exits that take you deeper in the dungeon... honestly, I get to bored by then. Also, by then all of the enemies start showing up as the minotaur, so it's next to unplayable.

 

Vectron is another game that technically has an ending, but is likely an unreachable ending. If you somehow manage to clear level 99, the game will tell you "Congratulations, you are very good" or something to that effect. But Mark Urbaneic himself has admitted that it's unlikely that that level is actually reachable, because the game gets too fast to be playable.

[cparsley]

As for some posts..

AD&D Tarmin does not end at 255, there are "Use ladder" exits on that level (and on 256 as well), exiting that level puts you back on level 1.

 

Vectron actually has 100 levels, at the end it just says "congratulations" instead of the promised light show, due to Mark not having enough memory left on the cartridge.

 

I'll have to go and break down every single cart then, it seems

[+thegoldenband]

What about Swords & Serpents? It has a clear ending, though you can't actually kill the dragon... still, we can say the game ends when facing the dragon?

 

I'd been thinking about that one. From memory, the docs and box seem to point to the dragon being essentially the goal of the game, right? Or is something else going on? I haven't played S&S enough to know firsthand.

 

I'd put Utopia into an ambiguous category: It's a two player game, with somewhat complex scoring. But you can play it one player, and establish a "highest score" for a given set of parameters (number of turns, length of turn). So, since it's of finite but configurable length, I'd say it's "beatable" in some sense.

 

Yeah, Utopia is a perfect example of a game that has a clear ending, but no universally agreed-upon winning condition, partly because of the variable starting conditions. The manual makes mention on page 8 of a "Governor's Award" -- specifically, at the top of a list of gameplay hints labeled "How to Win the Governor's Award". Is there any sign of unused text or graphics in the ROM that might point to this being more than just a figure of speech?

 

Interestingly, the scoring function has some strange internal overflows. As your "approval rating" nears 100%, if you have too many of certain items (never figured out the combination), you'll get a random rebel and your approval rating will drop into the 60s instantly. Utopia is a toughy to categorize fairly.

 

I know we talked at one point about a disassembly -- it's still far beyond my skills, but if you ever feel like looking into this and other quirky behaviors, I'd love to know what you find. The mechanics of the scoring algorithm are mysterious: for example, does proximity of certain kinds of buildings (besides the Tower) affect anything, or do they have a uniform effect on the Island?

 

AD&D Treasure of Tarmin also doesn't seem to have a proper "ending". You can end the game the first time you catch up to the Minotaur, or you can choose not to engage him and keep exploring (all the way up to dungeon level 255).

 

I remember that, but it was never clear to me whether it was an oversight or an intentional design decision. Either way it still seems to me that killing the Minotaur is the intended goal of the game.

 

Vectron is another game that technically has an ending, but is likely an unreachable ending.

 

I could've sworn that someone claimed to have reached it, but maybe I misunderstood (or they were full of it).

 

BTW, do you know of any other Intellivision "kill screens" or score-based gameplay limits besides (maybe) River Raid? And also -- does Space Patrol have an ending?

 

I'll have to go and break down every single cart then, it seems

 

Well, that's sort of the idea of this thread!

[+thegoldenband]

Time-based games with an end, but no clear winning condition:

 

Auto Racing - no computer opponent; manual lists no target times; one could establish the equivalent of an "Activision patch" time for each of the five courses

 

Motocross - beat the computer on courses 1-3, I suppose; number of laps is probably irrelevant

 

Mountain Madness Super Pro Skiing - basic gameplay is the same as Skiing, below, but manual lists no target times -- and, of course, there are 32 different courses!

 

Skiing - manual says that "Olympic gold medalist" equates to a downhill time under 38, or a slalom time under 105; reaching both those times is probably a viable claim to victory

 

(to be continued and edited)

[+cmart604]

What about the 1 player vs computer options on the various Super Pro games, would beating the computer on the hardest setting not count...I do this all the time with World Championship Baseball.

[+thegoldenband]

What about the 1 player vs computer options on the various Super Pro games, would beating the computer on the hardest setting not count...I do this all the time with World Championship Baseball.

 

It absolutely would count IMHO. I just haven't added those games yet because for most of them, I don't know how many options there are. I felt pretty sure that WCB only had a difficulty setting and no other significant variables, but I know Slam Dunk Super Pro Basketball gets a lot more complicated.

[intvnut]

I'd put Utopia into an ambiguous category: It's a two player game, with somewhat complex scoring. But you can play it one player, and establish a "highest score" for a given set of parameters (number of turns, length of turn). So, since it's of finite but configurable length, I'd say it's "beatable" in some sense.

 

Yeah, Utopia is a perfect example of a game that has a clear ending, but no universally agreed-upon winning condition, partly because of the variable starting conditions. The manual makes mention on page 8 of a "Governor's Award" -- specifically, at the top of a list of gameplay hints labeled "How to Win the Governor's Award". Is there any sign of unused text or graphics in the ROM that might point to this being more than just a figure of speech?

 

I'm pretty sure it's just a figure of speech. The game is only 4K decles. It's tiiiny.

 

 

Interestingly, the scoring function has some strange internal overflows. As your "approval rating" nears 100%, if you have too many of certain items (never figured out the combination), you'll get a random rebel and your approval rating will drop into the 60s instantly. Utopia is a toughy to categorize fairly.

 

I know we talked at one point about a disassembly -- it's still far beyond my skills, but if you ever feel like looking into this and other quirky behaviors, I'd love to know what you find. The mechanics of the scoring algorithm are mysterious: for example, does proximity of certain kinds of buildings (besides the Tower) affect anything, or do they have a uniform effect on the Island?

 

Assembly isn't that hard to read.

 

Vectron is another game that technically has an ending, but is likely an unreachable ending.

 

I could've sworn that someone claimed to have reached it, but maybe I misunderstood (or they were full of it).

 

Perhaps you're thinking of Mike Hayes triggering the Mark Urbaneic easter egg? Now, looking at his page here, http://spatula-city.org/~intvlib/inty/emulator.html , he does have a save state for level 99. What I don't know is whether that was achieved under normal circumstances, or by slowing the emulation down and/or disabling collision detection, etc. I guess we need to ask Mike if that level 99 save state is a "legit" state.

 

I need to dig up my gms2rom converter so y'all can experience the easter egg, too...

 

BTW, do you know of any other Intellivision "kill screens" or score-based gameplay limits besides (maybe) River Raid?

 

I'm not sure I know what you mean. You mean games with obvious, finite endings? I can't think of any off-hand. Most games seemed to aim for endless gameplay.

 

And also -- does Space Patrol have an ending?

 

Each of the 8 courses has a "point Z." If you get there, you've beat that course. Even the "continues" are limited. You only get two continues, and using a continue resets your score to 0. So, no, it doesn't have "an ending." It has 8 endings, one for each course.

[intvnut]

Vectron is another game that technically has an ending, but is likely an unreachable ending.

 

I could've sworn that someone claimed to have reached it, but maybe I misunderstood (or they were full of it).

 

Perhaps you're thinking of Mike Hayes triggering the Mark Urbaneic easter egg? Now, looking at his page here, http://spatula-city.org/~intvlib/inty/emulator.html , he does have a save state for level 99. What I don't know is whether that was achieved under normal circumstances, or by slowing the emulation down and/or disabling collision detection, etc. I guess we need to ask Mike if that level 99 save state is a "legit" state.

 

I need to dig up my gms2rom converter so y'all can experience the easter egg, too...

 

Ok, so I took Mike Hayes' INTVPC save states for Vectron and converted them to .ROM format, whereupon I discovered I was misspelling Mark Urbaniec's name. My apologies, Mark. For the Easter Egg ROM, press 7 to see the Easter Egg. For the Level 99 ROM, press any key to start level 99.

 

Enjoy!

 

And for those of you too lazy to download the save-state, see the attached GIF.

 

(Apparently, I cannot attach .ROM format files to this post, so get them here: http://spatula-city.org/~im14u2c/intv/dl/vectron99.rom and http://spatula-city.org/~im14u2c/intv/dl/vectron_egg.rom )

 

Steve Orth's INTV Fun House captured the details of the Vectron egg here: http://www.intvfunhouse.com/games/vect.php#eastereggs I'm pretty sure I posted the exact disassembly details to INTVPROG back in the time-frame mentioned.

[+thegoldenband]

I'm pretty sure it's just a figure of speech. The game is only 4K decles. It's tiiiny.

 

Wow! OK, that doesn't leave much room for anything hidden.

 

Assembly isn't that hard to read.

 

Well, neither is music, and you know how that goes! More seriously: I can (at least potentially) understand individual instructions, and I'm sure I could write a "HELLO WORLD" program, but I don't have the understanding of the whole I would need to make sense of what those instructions are doing (let alone identifying data tables and so forth). I have no basis for pattern recognition, so to speak.

 

Someday I'll sit down, devote a few consecutive days to improving my understanding of assembly, and come out with at least a rudimentary grasp of it. At least Intellivision ROMs don't use the kind of compression and copy-protection-through-obfuscation seen in Genesis ROMs.

 

I guess we need to ask Mike if that level 99 save state is a "legit" state.

 

I need to dig up my gms2rom converter so y'all can experience the easter egg, too...

 

That'd be great, on both counts. (EDIT: And there it is, thanks!)

 

For now I'll keep Vectron in the list unless it's proven that it's literally impossible to reach level 99 without cheats. (If it is, then we'd have a scenario like Impossible Mission II on the 7800: a game with an intended ending that can't actually be reached!)

 

BTW, do you know of any other Intellivision "kill screens" or score-based gameplay limits besides (maybe) River Raid?

 

I'm not sure I know what you mean. You mean games with obvious, finite endings?

 

Not exactly -- by "kill screen", I mean a game that either crashes or comes up with an impossible level once a certain score/level threshold is reached. Examples include Donkey Kong and Pac-Man (in the arcades), and the original release of Demon Attack on the VCS. There are also games which deliberately stop when you reach a certain score, like River Raid on the VCS which changes to "!!!!!!" and stops play when 999999 points is reached. Do any Intellivision games have fatal overflows past a certain score or level?

 

Each of the 8 courses has a "point Z." If you get there, you've beat that course. Even the "continues" are limited. You only get two continues, and using a continue resets your score to 0. So, no, it doesn't have "an ending." It has 8 endings, one for each course.

 

Got it, thanks! Once you finish a course, do you keep the same score and # of lives when you return to the main menu, or do you begin the next course as if you were starting from a cold boot? In other words, is the player expected (or allowed!) to beat all eight courses successively?

[intvnut]

I'm pretty sure it's just a figure of speech. The game is only 4K decles. It's tiiiny.

 

Wow! OK, that doesn't leave much room for anything hidden.

 

Yeah, not really. I did a quick cursory check for add'l text and it turned up nothing. I'm rather impressed they fit what they did in 4K!

 

Assembly isn't that hard to read.

 

Well, neither is music, and you know how that goes! More seriously: I can (at least potentially) understand individual instructions, and I'm sure I could write a "HELLO WORLD" program, but I don't have the understanding of the whole I would need to make sense of what those instructions are doing (let alone identifying data tables and so forth). I have no basis for pattern recognition, so to speak.

 

True enough. Really, the trick comes down to identifying what the main variables are and how it's actually using them. Once you've identified, say, the "multiply" routine or "square root" routine or "divide" routine, you don't care what the instructions are anymore inside them. But in general, reading a disassembly is the whole mountain-climbing process of abstracting away each piece of code to work your way toward higher and higher levels of abstraction.

 

Someday I'll sit down, devote a few consecutive days to improving my understanding of assembly, and come out with at least a rudimentary grasp of it. At least Intellivision ROMs don't use the kind of compression and copy-protection-through-obfuscation seen in Genesis ROMs.

 

I still insist assembly itself isn't difficult, almost by definition. After all, you can only do very simple things, such as copy values between registers, add them, subtract them, shift them and compare them, among other things. The real difficulty comes in understanding how a high level goal gets broken into tiny tiny pieces.

 

For example, you have no problem walking, I'm certain. Put one foot in front of the other, more or less. Now, describe "going to the store to buy some toothpaste" in terms of all of the footsteps you would make. Both are easy concepts... connecting them is where the difficulty is.

 

I guess we need to ask Mike if that level 99 save state is a "legit" state.

 

I need to dig up my gms2rom converter so y'all can experience the easter egg, too...

 

That'd be great, on both counts. (EDIT: And there it is, thanks!)

I still need to ask Mike to remind me of the details of how he got to level 99. He was rather surprisingly good at Vectron. I don't want to say he got there with a cheat unless he says he got there with a cheat.

 

BTW, do you know of any other Intellivision "kill screens" or score-based gameplay limits besides (maybe) River Raid?

 

I'm not sure I know what you mean. You mean games with obvious, finite endings?

 

Not exactly -- by "kill screen", I mean a game that either crashes or comes up with an impossible level once a certain score/level threshold is reached. Examples include Donkey Kong and Pac-Man (in the arcades), and the original release of Demon Attack on the VCS. There are also games which deliberately stop when you reach a certain score, like River Raid on the VCS which changes to "!!!!!!" and stops play when 999999 points is reached. Do any Intellivision games have fatal overflows past a certain score or level?

 

Ah, ok. Well, on Space Spartans, if you get to level 6, you can reposition the enemy bases as if they were your own. That's documented as a bug. Not really a kill screen, but certainly a bug.

 

Otherwise, so far as I know, all Intellivision games seem to largely muddle on. Some you can get into unsolvable situations. For example, there's a bug in AD&D Cloudy Mountain where you can march your party off the upper left corner of the screen and then warp to the middle of the map. Depending on the details of the map, you may not get the resources you need to complete the journey. Kill screen? Not exactly...

 

I know B-17 Bomber has an outright bug, if you drop a bomb from too far left and above center, you could hang the game for overflowing the input to the sqrt() routine (so that it looks negative). Again, not really a kill screen, but just a bug.

 

Each of the 8 courses has a "point Z." If you get there, you've beat that course. Even the "continues" are limited. You only get two continues, and using a continue resets your score to 0. So, no, it doesn't have "an ending." It has 8 endings, one for each course.

 

Got it, thanks! Once you finish a course, do you keep the same score and # of lives when you return to the main menu, or do you begin the next course as if you were starting from a cold boot? In other words, is the player expected (or allowed!) to beat all eight courses successively?

 

I treated all 8 courses as separate, with no carry-over between them. That's different than, say, Moon Patrol, which connects the Beginner and Champion courses, and loops back to Champion after that forever. I decided that for Space Patrol, that it made more sense to treat all 8 courses as separate, and allow you to tackle them all separately. Your score doesn't carry between them. Each has different challenges, and so I decided to make them all separate and unconnected.

[cparsley]

The only two games I can think of unpassable bugs (kill screens) would be B-17 bomber (divide by 0 error that crashes the game), and Atlantis (the floating & sign at day 12-14 depending on how well your playing). Other than that, rest you can theoretically play forever (if there isn't a exacted defined goal, like Cloudy Mountain)...

 

Vectron IS beatable, but just tougher than HELL (and I mean HELL!!!) I was there in the 1980s, think the best I can do today is level 30 or so.

[+DZ-Jay]

 

Assembly isn't that hard to read.

 

True, Assembly is actually rather easy to read. Comprehension is the difficult part.

 

-dZ.

[intvnut]

 

Assembly isn't that hard to read.

 

True, Assembly is actually rather easy to read. Comprehension is the difficult part.

 

It's ironic, but the most complicated thing about assembly code is how little each instruction does. We humans are accustomed to thinking in much, much larger steps than a computer is natively capable of. For example, if you see:

 

   ADDR    R0,  R1

 

That's really easy to understand on one level: It adds R0 to R1. It's just that simple! What does it mean in the context of a larger program, though? Ok, yeah, that's where all the fun is. :-)

[+DZ-Jay]

Assembly isn't that hard to read.

 

True, Assembly is actually rather easy to read. Comprehension is the difficult part.

 

It's ironic, but the most complicated thing about assembly code is how little each instruction does. We humans are accustomed to thinking in much, much larger steps than a computer is natively capable of. For example, if you see:

 

ADDR	R0,  R1

 

That's really easy to understand on one level: It adds R0 to R1. It's just that simple! What does it mean in the context of a larger program, though? Ok, yeah, that's where all the fun is. :-)

 

Yes, that why I said "comprehension" rather than "understanding," implying subtly the grasp of the larger concept (I was also trying to be funny .

 

That's the trouble I personally have with studying a disassembled binary (or any source code in Assembly, even my own): I know what the instructions do, but can't easily grasp their intended purpose within the bigger system. For me to do so, I have to discern--like you said--higher abstractions, and build higher ones above those.

 

That is where all the fun is, indeed.

 

It never ceases to fascinate me that games with such rich behaviour can be created merely by pushing numbers from one memory location to another, and applying simple arithmetic on them. It seems almost like magic.

 

-dZ.

Edited March 5, 2012 by DZ-Jay

[intvnut]

Interestingly, the scoring function has some strange internal overflows. As your "approval rating" nears 100%, if you have too many of certain items (never figured out the combination), you'll get a random rebel and your approval rating will drop into the 60s instantly. Utopia is a toughy to categorize fairly.

 

I know we talked at one point about a disassembly -- it's still far beyond my skills, but if you ever feel like looking into this and other quirky behaviors, I'd love to know what you find. The mechanics of the scoring algorithm are mysterious: for example, does proximity of certain kinds of buildings (besides the Tower) affect anything, or do they have a uniform effect on the Island?

 

Out of curiosity, I did take a peek at Utopia's disassembly. I haven't a clue what it's doing yet, and I'm not really going to spend any time on it. But, one thing that did strike me is the sheer number of times it calls the random number generator. I see no fewer than 15 different places it calls the random number generator, sometimes from within a loop.

 

It reminds me of a joke I heard at an ACM programming contest, referring to challenges writing a game AI (Mechmania at UIUC): "When in doubt, 'rand' it out."

[+DZ-Jay]

Interestingly, the scoring function has some strange internal overflows. As your "approval rating" nears 100%, if you have too many of certain items (never figured out the combination), you'll get a random rebel and your approval rating will drop into the 60s instantly. Utopia is a toughy to categorize fairly.

 

I know we talked at one point about a disassembly -- it's still far beyond my skills, but if you ever feel like looking into this and other quirky behaviors, I'd love to know what you find. The mechanics of the scoring algorithm are mysterious: for example, does proximity of certain kinds of buildings (besides the Tower) affect anything, or do they have a uniform effect on the Island?

 

Out of curiosity, I did take a peek at Utopia's disassembly. I haven't a clue what it's doing yet, and I'm not really going to spend any time on it. But, one thing that did strike me is the sheer number of times it calls the random number generator. I see no fewer than 15 different places it calls the random number generator, sometimes from within a loop.

 

It reminds me of a joke I heard at an ACM programming contest, referring to challenges writing a game AI (Mechmania at UIUC): "When in doubt, 'rand' it out."

 

Funny!

 

By the way, Utopia uses the EXEC model, right? Do you happen to have the EXEC API mapped out, or are you just discerning the PRNG function from the disassembly?

 

-dZ.