decle

Hey all, 2022 brings a new version of the Intellivision development overview, you can download it here: intellivisionDevelopmentBackInTheDay-20220101.pdf This update focuses on two topics: The Magus, or should that be MEGAS. The tools put together by Joe Jacobs and Dennis Clark. The details of the changes are... Magus / MEGAS information (p23-p27) First images of a MEGAS board Lots of technical details Full circuit diagram of a 16K MEGAS More information on Joe Jacobs and Dennis Clarke's development tools (p42-p49) PlayCable development kit and Vector board Details of their software tools such as Foreground, Pick-A-Doodle and DISLX Information regarding a cartridge reader created by Joe Jacobs If you're interested in the work done by Joe and Dennis at Technology Associates, you might also want to have a look at both the write up of their story and the PlayCable Technical Summary. Cheers decle

Happy 2022 everyone, In the ongoing tradition of publishing something to welcome in the New Year, I have an update to the PlayCable technical summary, which you can download here: playCableTechnicalSummary-20220101.pdf This brings you significant new material in the following sections: 6 PlayCable Hardware - All circuit diagrams have been checked and corrected against Jerrold schematics (p13-p19) 7.3 PlayCable Game ROMs - more information about which games were broadcast and when (p28-p31) And did you know there was a quarterly PlayCable newsletter, that got to at least issue 3? 8.1 PlayCable Adapter Emulator - full circuit diagram of the Jerrold version of the PlayCable (p37-p38) 8.2 PlayTape Adapter (p39-p43) Much more information on PlayTape and the changes made to support it A description of DIRTAP a PlayCable demonstration that may have been similar to the menu program used in the broadcast service 8.3 CYBER Development Kit (p47-p48) Information on the changes made and Joe Jacob's Vector board modification to the Master Component Details on PlayComp, Joe and Dennis' CYBER based personal PlayCable service, quite possibly the best PlayCable variant of all. I've also posted a brief description of DIRTAP and PlayComp, which includes a video and some interesting details of the unreleased games. As always any additions or corrections please let me know. And if this lot interests you, then you might want to take a peek at an update to the Intellivision development document here. Cheers decle

Hi all, In amongst the programs in Joe and Dennis' archive is this beauty. It's a jukebox that plays five tunes. Three of these arrangements were broadcast to subscribers homes as part of the PlayCable game menu. We know this because @dave1dmarx posted some contemporary recordings on YouTube that he made as a child. Well, now we can hear these tunes in glorious JzIntv hi-fi. The Entertainer @dave1dmarx 1980s recording: https://www.youtube.com/watch?v=WvZp_k6s33o 12th Street Rag @dave1dmarx 1980s recording: https://www.youtube.com/watch?v=lDfvzo4vAsA Music Box Dancer @dave1dmarx 1980s recording: https://www.youtube.com/watch?v=_0CX0vEg-WE We also get to see what the original PlayCable logo looked like on the Intellivision, and perhaps a glimpse of the colouring of the menu splash screen shown in this clipping: You will notice that The Entertainer has some "bubbles" at the end of the tune. These artefacts are also heard at the end of @dave1dmarx's recording. We originally thought that this might be caused by the omission of an "end of tune" marker, and the music engine subsequently ploughing through memory, blinding trying to play program code and data. However, with access to the jukebox source code, we can now see that for some reason someone has tacked a specific sound effect, called BEBOP, onto the end of the tune. Presumably, this was just some fiddling that accidentally made it to production. @dave1dmarx says that the "bubbles" were removed somewhere between late 1982 and early 1983. The source code for all three of these songs dates from March 1982 and uses a set of macros to define the tune. These macros are played by a common three part harmony music player called TRIHRM that was written by @David Rolfe at APh. David says that this was originally written to be part of the EXEC, but was relegated to an optional library to save on space during development. The other two tunes in the jukebox are also quite interesting. The first is Ride of Valkyries... This is the same tune used to signal victories in both Checkers and Sub Hunt. It also appears in the Easter Egg found in Conversational French. @David Rolfe explains that "it's a Caltech tradition to blast 'Ride of the Valkyries' through the dorms as a wake-up call when final exams come. This is no small thing; no one who attended Caltech can hear 'Ride' without feeling a rush of serious PTSD. So it wouldn't be surprising if a 'Ride' rendition were to come out of APh". My belief is that this tune was written by APh and found its way to Jerrold in a Checkers ROM image or cartridge. The source code for Ride of the Valkyries dates back to September 1979 is encoded using regular CP-1610 instructions rather than macros. This might suggest it was constructed by ripping it from a ROM disassembly, rather than it being from built from legit source code. Further evidence for this hypothesis is that the version of TRIHRM found in the jukebox differs from David's original EXEC version. Instead, it seems to have been taken from Checkers and then tweaked slightly to integrate it into the music player framework. The final tune is an arrangement of Take Me Out To The Ball Game... This is the oldest of the tunes, dating back to June 1979. Like Ride of Valkyries it is encoded using regular CP-1610 instructions. Perhaps some of the instructions were not decoded correctly and this is why there are some odd notes? I was wondering whether this song was intended to be used as the title screen or victory tune in MLB Baseball. Talking to David, he remembers working on the tune with Hal Finney, but he doesn't recall what it was to be used for. On a related note, there are also a number of sound board programs that seem to have been used to test out sound effects that were taken from games. Most of these effects are the bleeps and bloops familiar to us all, such as YER OUT! However, a there are a couple of interesting ones. The first is GALSONG, the theme tune to Battlestar Galactica: We know from this interview with Hal Finney that Space Battle originally had a version of the BSG theme in it, which was removed when Mattel didn't get the licence for the name. Perhaps this is that version of the tune? We also know that Jerrold had access to a pre-release version of Las Vegas Roulette and Slots, maybe they had an early copy of Space Battle too? Before anyone gets excited, although there are PlayTape versions of Baseball and Space Battle in the archive, they're just the regular production versions. Finally, there is SUPSONG, a version of the Superman theme... The tunes for both GALSONG and SUPSONG date from mid 1981, and are encoded as data statements, rather than being macros or regular CPU instructions. Although the sound board version of SUPSONG dates from 1981, references to it can be found in other code going as far back as August 1979. David thinks this tune may also have been written by Hal, but he doesn't know what it was intended to be used for. So there we have the Jerrold Music Jukebox and sound board. As someone who likes a banging 8-bit chip tune, I think they're pretty cool. For those that got this far, your reward is a ROM image for a new rendition of 12th Street Rag, complete with 6-voice ECS support... 12thStreetRag.rom Cheers decle

Happy Intellivision Day 2021! Following our work on Conversational French, we've captured and digitized Spelling Challenge. Like Conversational French and Jack Lalanne, Spelling Challenge was written by APh for Mattel. We know from the 2004 CGE panel that Peter Kaminiski was one of the programmers (24:40). Now don't get your hopes up! It's unlikely that Spelling Challenge is going to blow you away. The premise is that by correctly spelling words you help a monkey to feed a crocodile with coconuts. What is it with Mattel's educational primate fixation? Having shaken the coconuts from the tree, they have to be floated into the air using balloons, before they're picked off by a passing bird which ferries them to the crocodile!? Why this convoluted process? Don't know! Sounds like an idea generated at the end of a particularly long and liquid lunch to me. The result is that you have to spell each word correctly three times to complete a level. The game comes with five predefined lists, each containing 20 words. The words to be spelled in the first two lessons are represented by sounds, in the remaining lists they're just spoken. The player can also record their own lesson of up to 20 words with associated noises or spoken cues, providing unlimited spelling entertainment! Here's a quick "highlights" video: OK, so calling these highlights is probably stretching things. Even for an edutainment title Spelling Challenge is pretty lack lustre. Having to spell each word three times turns each list into a bit of a slog, it's missing the competitive, two player aspect of Math Fun or Word Fun, and it doesn't provide a timer, score or track your progress. For the completionists, here are the full set of lessons (apologies for the changes in exposure on scene changes, this is caused by my cheaping out on an AV capture device). For Animal Sounds we've provided a full play-through of all three game phases so you can see all the animations. However, for the other lessons we only show shaking the coconuts from the tree. The balloon and bird sections have the same animations as Animal Sounds, and once the coconuts are on the ground, you've heard all the pre-recorded audio content: List 2 - Animal Sounds - https://www.youtube.com/watch?v=g4wyZ95WZ7o List 3 - Sounds Like... - https://www.youtube.com/watch?v=FOwUUB3CiV0 List 4 - Spell It Right! - https://www.youtube.com/watch?v=RzalCbEH4ZE List 5 - Toughies - https://www.youtube.com/watch?v=a-vetPWAVpE List 6 - Words In Space - https://www.youtube.com/watch?v=X1owRxV5yN0 And finally, you can record your own lists like this: Edit Home Spelling List - https://www.youtube.com/watch?v=9z6fP67kIlk As was mentioned in the Conversational French "making of" video, the Kitty Faker cannot record audio, so we can't demonstrate using a home word list. So far, we've not found Easter Eggs in Spelling Challenge. The nearest to one is the rather cryptic text "SIGNAL for condition not ENABLEd" found in one of the program records (the capitalisation is as it's written in the program). We've not worked out how this message is triggered or what it means yet. As noted by @Lathe26, the K/C software was pretty expensive, with recommended prices between $50 and $70. As a comparison, a regular Inty cart cost $25 at the time, and a tape from the album chart was between $7 and $10. Spelling Challenge's program and audio amounts to about 8 minutes of content on the tape (1:15 of program and 6:30 of audio samples). Although, as @intvnut has pointed out, at 12K decles of program, the game is quite a bit bigger than something like Math Fun. Was it value for money? We leave it to you to decide. Again, credit to @Knarfian, @Ron The Cat, @Lathe26 and @intvnut for their help in bringing this to you. Right, on to Jack Lalanne! Cheers decle

Very cool Super Pro Intellivision Kartridge Reader?

Hi all, This is a big one, in some ways it's the culmination of several years of work by @Knarfian, @Lathe26, @intvnut, @Ron The Cat and myself. We have finally completed the restoration of Conversational French, probably the most technically advanced of all the Intellivision Keyboard Component software. To share it with you we have put together 21!? videos showing the full hour and three quarters of Mimi Schroeder's language course. Let's start with a highlights reel for those who just want to see the best bits and don't want to get bogged down in the details... Why is Conversational French the most advanced Keyboard Component tape? Whilst many of the PowerPoint style slides it contains, such as the Mona Lisa, are quite cool for the time, for me the most impressive content is the lip sync'd animation of Mimi. I would have been blown away if I'd seen it running on a home machine in 1980. I know this is the case because I did see Dancing Demon at my local Radio Shack at around that time (it was written in 1979), and I thought that was really cool. It should be noted that our videos probably don't do this animation justice. There is quite a bit of wow and flutter on the recordings we're working from, which means the lip-sync wanders a bit. Then there's the use of audio, both pre-recorded and home recorded. Only Conversational French and Spelling Challenge make use of the K/C microphone and clearly it's totally integral to Mimi's course. Finally, there's something that's not obvious. French Tape A manages to squeeze an hour of content onto a 30 minute cassette. It does this by having lessons recorded on both the Read Only and Read/Write tracks. The program and audio content are carefully aligned so that when you record your responses to the lessons on the Read Only track, it doesn't obliterate Mimi's audio for the lesson recorded on the Read/Write track. Very cunning, although it complicates our digital K/C tape replacement. Whilst investigating bugs in our initial digitization of Tape A, we came across some hidden credits. At 1.5K decles in size, it's probably the largest Easter Egg in the original Intellivision software library... ...Many thanks to Mark Stroberg for putting these credits in, and to everyone he listed, for their hard work on this piece of software. To trigger the credits, you hold down 1 & 9 on the left controller and 0 & 8 on the right controller as Mimi starts her introduction. This is acknowledged by a bing before she talks. The credits are then triggered by choosing 6 to save any progress and quit. So, how did we archive Conversational French and get it running again? Well, here's an over long and yet strangely unsatisfactory explanation... Finally we have the full contents of the course. In all, there are five lessons split across two tapes. Each lesson is further broken into between two and five parts. We've put together an individual video for each part lesson, linked below... Tape A Mimi's Introduction - https://www.youtube.com/watch?v=JjWO35vE8rs Lesson 1 - Customs Customs I - https://www.youtube.com/watch?v=aEzKvNdcrO8 Customs II - https://www.youtube.com/watch?v=FeqKqz3MbSU Customs III - https://www.youtube.com/watch?v=I7EYZzvzPWk Review - https://www.youtube.com/watch?v=j471b-b2CHQ Taxi - https://www.youtube.com/watch?v=jSMx8SOtHV8 Lesson 2 - Hotel Bedroom I - https://www.youtube.com/watch?v=cLiD2eQlUgI Bedroom II - https://www.youtube.com/watch?v=GIjsoatp7Xc Bathroom - https://www.youtube.com/watch?v=0dapQKVakx0 Money - https://www.youtube.com/watch?v=1s253xgIFN8 Tape B Lesson 3 - Restaurant Restaurant I - https://www.youtube.com/watch?v=EPaC7PKG9yQ Restaurant II - https://www.youtube.com/watch?v=yCVkLyWvih4 Review - https://www.youtube.com/watch?v=mgHwGU04g0Y Lesson 4 - Tour Tour I - https://www.youtube.com/watch?v=wq-R5_gDjPM Tour II - https://www.youtube.com/watch?v=0F-WqLU1hfc Tour III - https://www.youtube.com/watch?v=StChw0lUWEs Lesson 5 - Train Train 1 - https://www.youtube.com/watch?v=EbUceCDCrx8 Train 2 - https://www.youtube.com/watch?v=aejOgOuLPns So there you have it, Conversational French in all its glory. Unfortunately, for copyright reasons we can't share software with you, and even if we were to there is no way to use it at the moment without a Keyboard Component and Kitty Faker. However, we hope you enjoyed seeing what you missed out on at your local Frederick & Nelson all those years ago... We are well on the way to preserving the other Keyboard Component tapes and hope to have more videos showcasing this rare software soon. Special thanks to @Ron The Cat, @Knarfian, @Lathe26, @intvnut, @David Rolfe, Bill & Mimi Schroeder, Mark Stroberg, Tom Boellstorff and Braxton Soderman for their help in preserving Conversational French and their memories of putting it together in the first place. Cheers decle

Looking through the source code of Math Fun we've discovered some unknown, or at least informally documented bits and pieces. Problem types: As is described in the manual, there are 18 difficulty levels, broken into major colours and minor numbers (Black-1 through Red-2). Internally these levels are numbered 0-17 and each relates to a particular problem type. The types are defined in a table found in PROBLM.ASM (comments are mine): PRBTBL: PTBL SADD ; Black-1 Single digit addition PTBL SSUB ; Black-2 Single digit subtraction PTBL DADDNC ; Black-3 Double digit addition - no carry PTBL DADDC ; Black-4 Double digit addition with carry PTBL DSUBNB ; Blue-1 Double digit subtraction no borrow PTBL DSUBB ; Blue-2 Double digit subtraction with borrow PTBL TADDNC ; Blue-3 Triple digit addition no carry PTBL TADD1C ; Blue-4 Triple digit addition one carry PTBL TADD2C ; Yellow-1 Triple digit addition two carries PTBL TSUBNB ; Yellow-2 Triple digit subtraction no borrow PTBL TSUB1B ; Yellow-3 Triple digit subtraction one borrow PTBL TSUB2B ; Yellow-4 Triple digit subtraction two borrows PTBL SMULT ; Purple-1 Single digit multiplication PTBL SDIV ; Purple-2 Single digit division PTBL M1MULT ; Purple-3 Single x multi-digit multiplication PTBL DMULT ; Purple-4 Double x double-digit multiplication PTBL D1DIV ; Red-1 Multi by single-digit division PTBL DDIV ; Red-2 Multi by double-digit division When you choose a level, you're setting the highest problem difficulty that can be used in a game. This maximum value is stored in addresses $16f and $170 for player 1 and 2 respectively. Problem difficulty adjustment: I was careful to state that the maximum difficulty the player chooses was for the whole game. There is a separate, current difficulty level which is used to generate the next problem. This is always less than or equal to the overall game difficulty. Its value is stored in addresses $171 and $172 for player 1 and 2. The game can choose any problem type up to and including the current problem difficulty when setting a question. By changing the current problem difficulty, the game can adjust the challenge for each player independently. This is mentioned multiple times in the manual. At the start of the game the next problem difficulty is set to be three less than the overall game level. This means that the game starts slightly easier before ramping up to its ultimate challenge. A quirk of this is that you are guaranteed to get a single-digit addition problem as the first question on all the black levels. From here the current difficulty is adjusted based on the player's performance. Get a question right and the difficulty goes up by one, limited by the overall game difficulty. Answer incorrectly and the difficulty is dropped by two. It is possible to drop the current difficulty down from level red-2 to level black-1 by answering nine consecutive questions incorrectly. One page 1.5 (the page numbering is a bit weird in my copy), the manual suggests that the rate of progress answering puzzles may also be a factor in question difficulty... ...however, I've not found any evidence for this. Reverse-entry is not reverse-entry, at least not all the time When I made my original video about the Math Fun prototype, I included this comment from MASTER.ASM... ;*** NOTE *** MODIFICATIONS TO INPUT ROUTINES ALLOWING RIGHT TO LEFT ENTRY ; FOR +,-,* MADE 5/5/82 BY KJM, AFTER RESEARCH DONE BY CJH ; A FEW WEEKS EARLIER What I failed to notice at the time was that this suggests that reverse entry does not apply to division problems. Sure enough, the comment is correct. Forward entry is used for division problems, even in the reverse entry version of the game... ;GIVEN ADDRESS OF FLAGS IN R2, NUMBER PRESSED IN R0, PLAYER IN R1: ;SET THE FLAG TO SAY THAT A NUMBER HAS BEEN PRESSED. FIGURE THE NEW ANSWER ;BASED ON THIS NUMBER BEING USED, AND UPDATE THE ANSWER DISPLAY AND THE ;ANSWER ITSELF IF THE NEW ANSWER IS LEGAL. DIVISION IS ANSWERED LEFT TO ;RIGHT; ALL ELSE ANSWERED RIGHT TO LEFT. IF THIS IS NOT DIVISION, ;INCREMENT THE EXPONENT FOR THE CURRENT ANSWER IF THIS NUMBER IS USED. This is probably evidence that the real intent of reverse-entry was to support solving problems using the step-by-step methods taught in schools. These typically involve writing the problem down and working through it on paper. For addition, subtraction and multiplication these techniques work from units, through tens to hundreds, but for division they work from hundreds, through tens to units. These approaches allow students to break the question down, solving partial problems and reducing the amount of working that has to be maintained in the player's head. This makes solving larger problems using mental arithmetic significantly easier, but makes entering answers to some of the easier problems, notably single digit multiplications like 9x7 rather unintuitive. Personally, I wish we had the reverse-entry version, rather than the forward-entry variant, when we were kids as it would have made it possible to play harder levels without resorting to pen and paper. Perhaps a better name for the reverse-entry variant is the "work-it-out" or "schools" version. Which would make the original forward-entry version the "know-the-answer" or "mental-math-eureka" variant. Blue really is the the colour I was a little unconvinced by the violet ($d) colour used in the background of the prototype. It made me suspect that this was a ROM hack put together by the Jerrold engineers. However, there is some evidence in the production code that it was intentional. In INIT.ASM from the production code we see the following comment... MOV #UANSW1,R5 ; ** NOW IN 8 BIT RAM -LMZ ** MOV R0,@R5 ; " MOV R0,@R5 ; " MOV R0,@R5 ; " MOV R0,@R5 ; " DOBLNK: CALL BLANK ;BLANK SCREEN (LIGHT BLUE) This suggests that the intention was that the colour stack was initially set to light blue ($d), also known as violet. Then later in the code we find a commented out switch to green when setting up the game screen... ;ROUTINE TO SET UP SCREEN SCRNSET: ;** MOV #CH.GREEN,R1 ;** MOV R1,.BCOLTAB ;CHANGE BACKGROUND STACK BACK CALL BLANK ;FILL THE SCREEN WITH BLANKS So, it looks as though the original intention was to have a light blue background for the selection screen, but then green for the game. However, our prototype originates from a period where the switch to green was commented out, but before the colour stack entries in the cartridge headers were changed. Leaving it with a light blue background throughout. Cut the music already We can see that someone really didn't like the Electric Company theme tune. In addition to the second half being cut, Larry Zwick added the following code during game setup: MOV R1,.KEYDSP CLR R5 ; ** AND LAST, TURN OFF THEME MUSIC MOV R5,.SPRIO ; ** -LMZ ** This isn't present in the prototype, and as a consequence the tune plays to completion, even if you get through the selection quickly and onto the game screen. In the production release it is cut short once the Gorillas come into view. Animal switcheroo There seems to have been a change in the animals setting the questions at some point during development. In the code, the antelope is labelled a horse, and the kangaroo is named cat2. What I have always thought is a sheep is identified as a zebra, and seems to have been added as a bit of an afterthought, as it is not defined with the rest of the zoo. On page 2 of the manual there is something that looks suspiciously like a camel, which does not appear in the game at all. This creature also appears in the mock-up (notice the probable violation of the 2 colour rule) shown in all the game catalogues... Although the graphics are the same, the colours of a number of the animals were also changed between the prototype and the final release. I believe the easiest way to get hold of a reverse-entry cartridge is to target the white label INTV version of the game. I don't know that they are all reverse-entry, but certainly my copy is. It may also be a good idea to go after games in the later flip-top boxes that don't feature "The Electric Company" branding on the front...

Sorry, we don't own the rights to the game, so we can't distribute it. I think this is an excellent observation. Vegas II is 3900 decles long, this compares with the released version of Roulette which is 3265 decles. So, as things stand, Craps would have to be 196 decles or less for all three games to fit into 4K (4096 decles). However, we don't know how much optimisation has been done to Vegas II, and therefore, how much more could be squeezed. No, I didn't, but it's not hard to put one together, or at least a first stab at one... Game Programmer(s) Reference(s) ABPA Backgammon Kevin Miller https://www.mobygames.com/game/abpa-backgammon Advanced Dungeons and Dragons Chris Hawley, Tom Soulanille, Tom Loughry http://intellivisionaries.com/episode-6-the-dreadnaught-factor/ Armor Battle Chris Kingsley http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Auto Racing Larry Zwick https://en.wikipedia.org/wiki/Auto_Racing_(video_game) Boxing Tom Loughry https://en.wikipedia.org/wiki/Boxing_(1981_video_game) Checkers David Rolfe http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Frog Bog Peter Kaminski, Tom Soulanille http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Horse Racing Chris Hawley http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Las Vegas Poker and Blackjack David Rolfe http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Las Vegas Roulette Walter Bright, John Brooks https://en.wikipedia.org/wiki/List_of_Intellivision_games Major League Baseball David Rolfe https://en.wikipedia.org/wiki/Baseball_(Intellivision_video_game) Math Fun Kimo Yap http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting NASL Soccer Kevin Miller http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting NBA Basketball Ken Smith http://intellivisionaries.com/episode-18-basketballs/ NFL Football Ken Smith, Kevin Miller, Glyn Anderson https://www.mobygames.com/game/nfl-football, http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting NHL Hockey Ken Smith https://www.mobygames.com/game/nhl-hockey_ PGA Golf Scott Bishop http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Reversi Greg Favour http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Royal Dealer Rich O’Keefe http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Sea Battle Ken Smith http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Sharp Shot Frank Evans http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Space Armada John Brooks, Chris Hawley http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Space Battle Hal Finney http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Star Strike Hal Finney, Brett Stutz http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Sub Hunt John Hershberger, Tom Loughry https://en.wikipedia.org/wiki/List_of_Intellivision_games Tennis Gavin Claypool http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Treasure of Tarmin Tom Loughry http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Triple Action Rich O’Keefe, Shal Farley https://en.wikipedia.org/wiki/List_of_Intellivision_games US Ski Team Skiing Scott Reynolds http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting Word Fun Kevin Miller https://www.mobygames.com/game/abpa-backgammon World Championship Baseball Dan Dickerson, Mike Minkoff http://gdri.smspower.org/wiki/index.php/APh_Technological_Consulting As you can see most of the references come from GRDI, Moby Games and Wikipedia. The GRDI page is especially good, it lists a couple of titles I had not heard of before, for example "More of Theirs" by John Brooks, and Explorer. My understanding is that Walter wrote the core Roulette game prior to leaving APh in June 1979. However, he did not write Slots and doesn't know who did. Perhaps John Brooks did the control and animation overhaul? Its interesting the GRDI doesn't list Slots, but does list Craps. Unfortunately, it doesn't attribute an author, so we can't go a chasing.

Following hard on the heels of our rundown of Vegas-II, the prototype of Las Vegas Roulette & Slots... ...we've put together an overview of what appears to be an early version of Math Fun... I don't think I mention this in the video, but the bulk of Math Fun was written by Kimo Yap while he worked at APh, as you can find out in Intellivisionaries episode 17... http://intellivisionaries.com/episode-17-math-word-and-learning-fun/ It seems that people like Larry Zwick, who were just making small changes, were much more likely to autograph their code. So it's easier to attribute their tweaks.

Amongst the files we've recovered from a PDP-11 owned by Technology Associates was the intriguingly titled "Vegas II". This name first appeared on a list of games that was found with a PlayCable adapter bought from a store in NJ... The file we've found is in the PlayTape format used by Jerrold engineers to make game backups that can be played at home on hacked PlayCable adapters. This, and the other oddball names on the list, suggests the piece of paper might detail the contents of one or more PlayTape audio cassettes (unfortunately no cassettes were sold with the PlayCable). Following a bit of tinkering to convert the PlayTape format into something compatible with JzIntv, we have it up and running. The result appears to be a previously unknown prototype of Las Vegas Roulette & Slots. So, for the first time in 40 years, Vegas II... By the way, apologies if you get ads on my videos now. This is not my doing, and I don't get any revenue from it (my channel is waaaayyyy too small to be monetized). This seems to be a policy change by Google. Unfortunately, I didn't notice the switch until recently (I use an ad-blocker).

That does seem to be a fair description. Indeed, which should have meant that Mattel was less concerned about the risk of Joe and Dennis going elsewhere than Keith portrayed. They were only two amongst many. <SpeculativeCity>I wonder whether Jerrold's initial expectation that they would write exclusive games for PlayCable, the fact that this never came to pass, their laissez-fair attitude toward Joe and Dennis' extra-curricular activities, and Mattel's reaction when the pair rocked up with Clone-Man are somehow linked</SpeculativeCity>. We'll probably never know. That is correct. Unfortunately that is a bug that was missed during QA, not the Easter egg. Keep looking... Well, would you be interested in seeing the level / track editor that Dennis wrote for Joe to use? If so, it was on one of the floppy disks and we've managed to get it working... ...Dennis also wrote a tool to help develop the animated cars which was on another disk... ...and here are some of the early player car graphics that Joe and Dennis put together for Bump N Jump (the gif shows the MOBs both linked / overlaid and split)... ...I know it's authentic to the arcade game, but I never understood the "wobbly wheels" aesthetic, I much prefer the rotating effect Joe and Dennis created here. Anyway, next we have what I suspect are some enemy cars...

First up, thanks for the kind feedback, it's much appreciated. Now on to some specific points... Yup, Joe and Dennis were, and are, both professional engineers, not spotty teenage hackers who wrote "Hello World" for the Intellivision. This is perhaps part of the reason Mattel took them seriously. Personally, I wouldn't describe what they did as reverse engineering the PlayCable, because they were part of the team developing it. However, they did use the skills and tools at their disposal to repurpose the PlayCable adapter in an innovative way, and put together a pretty sophisticated development kit using it, kudos to them. Sadly, Keith does seem to have been aware of who Joe and Dennis are, as can be seen here I suspect that any concerns Mattel had over others repeating what Joe and Dennis did were massively overblown. To use the PlayCable to construct a development kit required access to a pre-production Jerrold model. These were used for internal development work, and perhaps for the early field trials. As such, I believe they were probably limited in number and I expect Jerrold restricted access to them. It would not be possible to use a regular PlayCable branded adapter to create a devkit in this way because all its digital logic was consolidated onto a single chip. This list came with a PlayCable adapter that a lucky collector bought from a retro shop in NJ. We are pretty confident that it was one of the modified PlayTape adapters previously owned by a Jerrold engineer. Unfortunately no audio cassettes were with the PlayCable, and therefore, they are presumed lost. The numbers on the list are believed to be the tape counter value on the cassette player at the start of each game. So the list shows the contents of either four single sided PlayTape audio cassettes, or possibly two tapes with recordings on both sides. Notice how most of the 4K games on the list are separated by a count of 6 or 7 , but the larger 6K game Sub Hunt has a length of 9. Pick-a-Doodle is a little animation program originally written by Dennis in 1981. It allows non-programmers to put together up to 20 frames of background cards and play them like a flick-book accompanied by background music. Here's a short video of it in action... ...I forgot to mention in the video that Pick-A-Doodle was meant to be run on the PlayCable devkit. So it stores the animation frames in PlayCable RAM, and this data can be downloaded to the developer's computer once it is complete, so it can be played back later or potentially be incorporated into another program. I am working with Joe and Dennis to recover data from one of their old PDP-11 systems, and as you can see we have had some success. I hope that we might be able to share videos of more of this stuff in due course. We can already see that it contains some cool bits and pieces, especially for dweebs like me.

You may be aware from updates to the Intellivision development history that I’ve been in contact with Joe Jacobs and Dennis Clark for a while. They have provided some background to the creation of their PlayCable development system and the subsequent work on Bump ‘N’ Jump. This posting details the story from Joe and Dennis’ perspective and I’d like to thank them for helping me to put it together, and allowing it to be shared. Joe Jacobs & Dennis Clark Circa 1980 Joe Jacobs and Dennis Clark are engineers who worked for Jerrold, the cable television division of General Instrument and manufacturers of PlayCable. Dennis joined Jerrold in the summer of 1978 working in Hatboro PA. Back then, in the early days of PlayCable development, Jerrold anticipated that, like Mattel, it would write games for the Intellivision to be distributed over PlayCable. In part, Dennis was recruited to go into arcades and scout for titles suitable for conversion. As it turned out Jerrold never wrote a game for the PlayCable, and Dennis did not make it to an arcade on company time. Instead, he worked in the Software Department writing firmware for cable boxes and PDP-11 software for Jerrold’s cable head-end infrastructure. Development of the PlayCable hardware was well advanced by the first half of 1979, and over the summer Dennis worked to develop the firmware for the PlayCable adapter. He was also responsible for the music tracker used by the PlayCable menu program, and he arranged the version of The Entertainer that can be heard playing on the splash screen of the menu. In early 1981 Joe Jacobs left Siemens, where he worked on automated test equipment. He was hired by Jerrold to work in their Head-End Division as a Project Engineer to develop hardware and software associated with the distribution of cable services. This equipment was used by cable companies to distribute and control their subscribers' access to channels. The systems that Joe worked on communicated with the converter boxes installed in customers’ homes that Dennis helped to develop. Whereas Dennis is primarily a software specialist, Joe is more of a hybrid engineer, his focus is on hardware development, but he also writes software. Although Joe and Dennis looked after different aspects of Jerrold’s products, they worked in close proximity to each other, and became good friends. Dennis recalls how Joe nicknamed him “Grumpy” because he always had a determined look on his face. Joe explained that “In the early 1980's, Jerrold was still a small to mid-sized company and most of Jerrold's engineering was in one building”. Dennis says that, under the management of Charles Dages, Jerrold’s engineering department was very supportive of engineers’ creativity and fostered collaboration. It should be noted that when Mattel partnered with Jerrold to develop the PlayCable the two companies had a symbiotic relationship. Jerrold brought hardware knowledge specific to the cable industry and Mattel supplied access to the secret sauce for the Intellivision. This included the APh assembler and linker, and details of the EXEC and how to use it. Joe describes a “HUGE listing called the Mattel 'EXEC'. This listing was an assembler list file generated when Mattel compiled the library routines that went into each and every Mattel Intellivision main unit. It was dot-matrix printed, on that wide paper with the holes at each end and was about two-inches thick. It described each and every routine available to the game developer, calling conventions, parameter passing, object creation and interaction, etc”. Dennis noted that the interrupt driven model of the Mattel EXEC was unusual for the time and something he thinks was very innovative. Although General Instrument could provide Jerrold with information about the Gimini chipset on which the Intellivision is built, it needed these Master Component specific resources to write software for PlayCable. Remember that Jerrold had to write the firmware ROM in the PlayCable adapter, the menu program used by customers to select games, and potentially original Intellivision titles. Therefore, Jerrold, like APh, was one of a small number of trusted partners, and Jerrold engineers like Joe and Dennis had an inside track on writing software for the Master Component. Interestingly, Dennis recalls that during the development of the PlayCable he visited APh in Pasadena to learn more about the Intellivision, a trip that led to him meeting Glen Hightower and the Intellivision developers. It seems that at some point in late 1979, one of Dennis’ colleagues, possibly Joe Rocci, realised that the head-end infrastructure could be used to create backups of Intellivision games that could be played at home. PlayCable games were transmitted from dedicated microprocessor controlled cards, housed in a PDP-11 minicomputer. These same cards were also used by cable company head-end systems to communicate with consumers' cable boxes. A side effect of the encoding scheme used to transmit PlayCable titles was that the game data could be recorded directly off the transmission cards onto a regular audio cassette. The image below shows one such a DCX11A (Dual-Channel Xmitter) card connected to an audio adapter that was used to record Intellivision games. DCX-11A DataChannel Transmission Card with Audio Adapter Jerrold engineers could load games into the transmission card, connect the digital output to a tape machine using the adapter box and record the resulting stream. At home, they could then connect a regular audio cassette tape machine to a hacked PlayCable adapter and play the recorded game directly into the PlayCable’s memory. To make this work required some changes to the PlayCable adapter firmware, and for the digital board within the adapter to be connected to an audio input, rather than the normal cable receiver. These hacked PlayCable adapters were based on the earlier, limited-production Jerrold model which, unlike the later PlayCable branded units, had their digital sub-system implemented using standard off-the-shelf components. This made them much more hackable by exposing their inner secrets to those in the know, or with access to oscilloscopes and datasheets (see Sections 8.1 and 8.2 of the PlayCable Technical Summary for more information). Jerrold’s engineers christened these audio backups “PlayTape”. This innovation gave unrestricted access to the entire Intellivision PlayCable games library and was shared amongst some of the members of the engineering department. As Joe says, “all of us engineers had a modified PlayCable setup so we could play Intellivision at home. Remember, at the time, Intellivision was the ‘cat's meow’ of video games, handily beating the Atari 2600; Colecovision had not yet come on the scene”. Dennis believes that the management of Jerrold’s engineering department were probably aware of what their engineers were up to, but turned a blind eye, not seeing any harm in it. Title Screens for the Standard PlayCable (left) and Joe's PlayTape (right) On joining Jerrold in 1981, Joe quickly discovered what was going on and got involved, contributing to the modified firmware that ran on the PlayTape adapters. Before joining Jerrold, Joe had put together a small PDP-11/03 “Frankenstein” system of his own at home. This was compatible with the computers that were used to develop Jerrold’s cable head-end software and write Intellivision games. Through the summer of 1981 Dennis continued to tinker with Intellivision development, stripping sounds from Mattel games and building a sound board application to play them back. Joe’s interest in video games led him to start reverse-engineering the Arcadia Supercharger following its release for the Atari 2600. He figured out a way to read some of his Atari game cartridges and transfer them to the Supercharger replicating the “game-backups-on-tape” principle behind PlayTape. Catalogue of PlayTape Titles Through the fall of 1981 the library of PlayTape games was extended as new titles were released for the Intellivision, the pair also wrote diagnostic programs, and started to investigate the inner workings of the Intellivision’s EXEC. Joe realised that it would be possible to use a specially-modified PlayCable adapter, along with his Frankenstein PDP-11, and the tools he had access to at Jerrold, to develop rudimentary Intellivision games. Inspired, Joe suggested to Dennis that they "try and write a game for the Intellivision". Dennis was up for the challenge and explained the methods Jerrold used for Intellivision development. Joe recalls that the process was pretty simplistic. “It wasn't a whole lot, in my mind, it was basically EPROM burn and crash and burn and crash and... development". By this point Dennis also had a PDP-11 at home, put together from spare Jerrold equipment. Building such home systems was supported by Jerrold, as it allowed engineers to continue to work on company projects in their own time. In the meantime, Joe had started to think about how to improve the development tools, “I was, and still am, an in-circuit emulator kind of guy and prefer to do my software debugging in that environment if possible”. According to Dennis, testing was done using “something like ROM simulators to load the code from the LSI-11 to a modified Playcable type adapter”. This allowed test code to be uploaded from their development machines directly to the PlayCable, bypassing the need to use a broadcast card and audio cassettes. Joe says that “the whole concept was loosely modelled on the then-popular Motorola ExORciser development environment”. In the spring of 1982 Dennis and Joe concluded that they needed a demonstration to showcase their maturing Intellivision development capabilities and grab the attention of Mattel. They tossed some ideas back and forth and settled on writing Clone-Man, a homage to PAC-MAN. At the time PAC-MAN had just been released on the Atari 2600 and was at the forefront of public consciousness. Unfortunately, this next step in the journey coincided with Dennis suffering a back injury. Despite this, Joe and Dennis pressed ahead with Clone-man over the next two or three months whilst Dennis was off work recovering from his back injury. This led to Clone-Man initially being credited to “Bedside Productions”. Within the team, Dennis’ focus was on core software, with Joe sorting out the hardware necessary for their development systems and providing some additional utilities. Dennis says that he saw porting PAC-MAN as “just a challenge to see how to copy an arcade video game onto Intellivision”. Clone-Man - a Glimpse of Dennis and Joe’s homage to PAC-MAN The resulting “Demonstration Program” was a pretty comprehensive recreation of the game, with a landscape version of the original maze, power pellets, bonus fruit, and sound effects. However, the algorithms that drive the movement of Blinky, Pinky, Inky and Clyde were not replicated and there are no intermissions. Overall, the game is clearly superior to the Atari 2600 version, but is not as polished as the Atarisoft version for the Intellivision, for example the sound effects are not replicated as accurately. As Joe says, Clone-Man “came out pretty good. Not good enough for commercial appeal, but good enough”. Dennis’ opinion is that “it would have been hard to tell it from Pac Man”, which is probably stretching things. However, with its more accurate maze, it clearly attempts to be more faithful to the arcade original than either the Atari 2600 or official Intellivision ports. Throughout this period, Joe and Dennis continued to enhance their PlayCable test systems. The modified adapters were linked to their PDP-11 computers using an RS-232 serial connection, and ran enhanced firmware containing a debugger called CYBER. The pictures below show the results of Joe and Dennis’ alterations (see Section 8.3 of the PlayCable Technical Summary for more details). Joe and Dennis’ Development Kit PlayCable Receiver Board Joe and Dennis’ Development Kit PlayCable Digital Board In addition to modifying the PlayCable adapters to support RS-232 communication, Joe added what he calls a “vector” board to their development Intellivision Master Components. These enabled breakpoint and single stepping features to be added to the CYBER debugger being developed by Dennis. A video showing CYBER being used to debug an Intellivision program can be seen here: The modifications made to the PlayCables were pretty extensive, and together with Dennis’ CYBER debugger, they led to the early MAGUS-like ROM emulator turning into a system that had similar features to Mattel’s Blue Whale test harness. This can be seen in the following list of CYBER commands: CYBER Debugger Command Crib Sheet Once Clone-Man was complete, Joe says he “did some checking with Jerrold management about our intentions of writing something for Mattel; they didn't have a problem so I went for it”. He used a Betamax video camcorder to record Clone-Man running on the Intellivision and sent the tape to Don Daglow at Mattel. At this point Joe says that “evidently, the crap hit the fan at Mattel”. Joe doesn’t really remember any fallout at Jerrold over Clone-Man, but the Mattel people were clearly “spinning in their seats”. Given Mattel’s paranoia over industrial secrecy, this was perhaps inevitable. Many phone conferences ensued over the next couple of months as Joe negotiated a deal with Mattel to write a game. This led to an agreement in December of 1982 that Technology Associates, the fledgling computer consulting company founded by Joe in 1981, would write a port of Bump N Jump for the Intellivision under contract to Mattel. Effectively, Technology Associates became a second-party developer for Intellivision, like APh. As might be expected, Mattel seems to have been concerned that Joe and Dennis could take their skills and knowledge to a competitor. However, Joe and Dennis are clear that this was never an option for them and, despite what is reported elsewhere, they did not threaten to do so. In fact, Jerrold was aware that Joe and Dennis had approached Mattel, and seems to have been supportive of their entrepreneurial streak, as they both continued in their day jobs. The reasons for Jerrold’s lack of concern over their game-writing endeavours are unclear, although Joe explains it like this, “We did not work on BNJ during our work hours at Jerrold for obvious reasons. Jerrold was aware of the situation and left us to it. At the time, we were pretty valuable employees... Besides, there was absolutely no negative karma, letting us do our own thing at the time. A benefit of working for a smaller company”. Regardless, like Clone-Man before, the Bump N Jump project was to be an extra-curricular activity for Joe and Dennis that occupied their evenings and weekends. What would have happened if a deal had not been struck? According to Joe and Dennis, they would have continued working for Jerrold at their regular day-jobs, and would have explored the Intellivision on their own time just for fun. Having landed the contract to write Bump N Jump, and with the dust settling, Technology Associates purchased two new PDP-11 systems from Sigma Information Systems, complete with 8” floppy disks and enormous 20MB hard drives. These machines would be used to do the bulk of the subsequent Bump N Jump development. Up to this point, Joe and Dennis only had a single PlayCable development system to test Clone-Man. Joe took the opportunity to rectify this by building a second test harness to use while creating Bump N Jump, and the pair set to it. In all, development of Bump N Jump took around six months of intensive work in the evenings and weekends. Joe suggests that “Dennis was, no question, the brains behind the code. While he worked on game play such as object generation, object interaction, scoring, etc. I was responsible for the entire background”. Dennis agrees, explaining that “Joe did the background and track work”, effectively being responsible for the accurate reproduction of the levels. To help with development, Mattel shipped an arcade version of Burnin’ Rubber (the international variant of Bump ‘N’ Jump) to Dennis' house. Once installed in the basement, Dennis' girlfriend's son played the game for hours and became an expert at it. Joe used his camcorder to record the teenager’s games for use in development. By watching the recordings back, over and over, ad nauseum, Joe was able to transcribe the levels of the arcade game using a level designer written by Dennis. Joe says, “The background of Bump ‘N’ Jump is basically a gigantic table of ‘cards’, with the presentation of those cards handled by Dennis’ level designer code”. As a consequence, the Intellivision port has a faithful reproduction of the playfield of the arcade version, including the track layout, bridges and other obstacles. Meanwhile, in addition to the core game mechanic, Dennis wrote more tools, including a music generator and an animation designer to support development. As Bump N Jump took shape it became clear that the 8K of RAM within their PlayCables was not going to be enough to hold the full game. Sadly, the limits of their homebrew development kit had been exceeded. So, Joe “contacted Mattel to ask what was available to get past the 8K limit, and their answer was a board called the 16K Megas board". Mattel sent a couple of Megas (aka MAGUS) test harnesses for end-to-end play testing and Joe sorted out the hardware necessary to interface them to their PDP-11s. This he did by customising a Heathkit parallel interface board. Joe explains that during use “you had to tell the Megas board to 'freeze' the CPU from accessing the Megas ram, load the RAM, un-freeze the CPU and then tell the CPU where to start executing. Basically, it was a RAM-based burn and crash idea, but instead of burning an eprom or rom, you 'burned' the Megas RAM and it was pretty quick. A lot quicker than burning chips. The Megas wasn't really for troubleshooting/debugging but more an end-to-end play/test of the game you were working on”. As was mentioned by Keith Robinson at Classic Game Fest in 2016, David Warhol acted as the liaison between Mattel and Technology Associates. Unfortunately, the relationship between the two organisations was not easy, as Joe observed, “I think the Mattel developers were definitely leery of us and certainly didn't voluntarily share anything on their own. If we had a particular question [that] needed answering they did answer but only the exact answer, nothing more, nothing less. We were still 'outsiders'”. Mattel’s attempts to limit the flow of information to Technology Associates can be seen as part of their ongoing attempts to hold their cards close and prevent third parties developing games for the Intellivision. Joe and Dennis finished the core game of Bump N Jump at the end of May 1983 and shipped the source code containing two levels to Mattel HQ in Hawthorne. Once there, it entered the Intellivision QA process. A BSR review meeting in the first week of June highlighted that game play tuning was required. The most significant points raised were that the game required a greater sense of speed, with the enemy cars needing to be easier to bump and kill, but also requiring more personality and aggression to increase the intensity of the game. A number of developers requested the inclusion of an engine sound, to provide auditory feedback of the player’s speed. It was at this point that Mattel decided a change to the title screen was also required. The original received mixed reviews, with some confusion about whether it depicted a road or a mountain. Regardless, it was felt to be too similar to the introduction of Buzz Bombers and needed an update. The final animated titles were developed by Daisy Nguyen and seem to have been added sometime in early July. As always, there were also some bugs found that were subsequently fixed. Although Joe and Dennis don’t recall Mattel requesting much work after the code was shipped, a message from David Warhol suggests that the updates were split between Mattel and Technology Associates, with Mattel looking after graphical tweaks and Daisy’s title screen, while Joe and Dennis focused on game play tuning. It’s clear that not all Mattel’s suggestions were included, for example, music wasn’t added to Daisy’s title screen, and the requested engine sound isn’t present in the released version. The final game with its full set of levels was accepted for production by Dale Lynn and Traci Glauser on August 1st 1983 as can be seen in the QA report below. Mattel Bump N Jump QA Record At around this time it normally took Mattel about three months to get from acceptance of the final code to a game hitting the stores. Roughly two months of this time was ROM production, with the last month typically being consumed with finalising printed materials, packaging the game and distribution. The advert below for Bump N Jump was run in the October and November issues of games magazines across the US, and according to The Video Game Update, the title was one of the last games Mattel released when it hit store shelves in November 1983. Bump N Jump Print Advertisement Joe and Dennis are rightly proud of Bump N Jump and they feel that the title really pushed the capabilities of the hardware. The game play is very similar to the arcade, with the original levels and background music both faithfully reproduced. Unfortunately, interest in the Intellivision dwindled rapidly with the closure of Mattel Electronics at the start of 1984, and there seems to be very little about Bump N Jump in the press after its release. The Video Game Update did review Bump N Jump in their January 1984 issue, giving the title two and a half out of four stars for both graphics and gameplay, rating it as fair to good, but questioning the game’s depth, and therefore not recommending it. Video Game Update Bump N Jump Review However, history has been rather kinder to Bump N Jump, the title is now consistently rated amongst the Intellivision’s best games. This includes the current generation of Intellivision gamers placing it in the top 10 Intellivision titles in 2014, and the top 15 games in 2019. Reviewers such as The Intellivision Library, Intv Funhouse and Video Game Critic all rate the game highly, noting the quality of both graphics and sound, and the accuracy of the conversion. Overwhelmingly, the prevailing wisdom is that Bump N Jump deserves a place in your Intellivision collection. In late June 1983 Mattel Electronics announced the first round of redundancies that would mark the start of a death spiral for the division. Unsurprisingly given the timing of the completion of Bump N Jump development, Joe and Dennis didn’t receive offers of additional Intellivision work. With hindsight, the decision to continue to work for Jerrold whilst developing Bump N Jump on their own-time can be seen as an excellent one! Later, at the end of September David Warhol wrote to Joe and Dennis explaining the situation, and expressing the hope that more projects might be on the horizon with Mattel’s new focus on software; unfortunately, this future never materialised. Although they were initially unaware of the turmoil at Mattel, it was clear to both Joe and Dennis that they would always be considered outsiders at Hawthorne. In addition, Dennis explained that he enjoyed his work at Jerrold, and whilst writing Bump N Jump was profitable as a side-line, the money they made writing it wasn’t good enough to tempt the pair into giving up their day jobs. They also decided against pursuing opportunities with other games companies. Instead, they continued working for Jerrold and went back to just hacking for fun. Having grown tired of his lengthy commute to Hatboro, Joe left Jerrold in 1984 for a new role working for Omnidata (later Singer-Link Simulation) on power plant simulators, used to train control room engineers. However, Dennis continued with Jerrold, rising through the ranks to become Director of Project Management before retiring in the mid 2000s. So there we go, the story of the development of Bump N Jump and the mythical PlayCable development system from the perspective of Joe and Dennis. Incredibly, their whole Intellivision adventure lasted less than 30 months. It would be great to get the recollections of Mattel people like Don Daglow and David Warhol, and the management at Jerrold to complete the picture. Hopefully one day. One last thing before I go… A little birdy tells me that there is an Easter egg buried in Bump ‘N’ Jump that has gone undiscovered since the game’s release. Can the players and developers of the Intellivision Brotherhood find it? The challenge has been issued, just for kicks. Once again, thanks to Joe and Dennis for giving their permission to share their story and for their help in putting it together.

It seems there is no job that I can't do in a half-arsed manner, and converting Killer Bees to as1600 is an excellent example of this. 15 years was just not long enough for me to generate 375 lines of bug free Fatal Anthophila code. How do I know this? Well, a copy of the APh CP-1610 assembler and linker has come into my possession, which means we can use a PDP-11 emulator to rebuild a definitive version of @David Rolfe's original code and look at the results... ...and the binary clearly does not match the one I created in 2016. Admittedly the differences were not that significant, but if something's worth doing... As you can see in the video, the same 8" floppy has also provided a previously unseen version of Killer Bees. I believe this is contemporaneous to David's original, as the source file date is April 4th 1979. Having shown the results to David, he provided the following explanation regarding the history of Killer Bees and the possible origins of this new, Mattel variant: So, perhaps this new variant was written by an APh summer hire and somehow made its way to Mattel, where David's Talkin Horse branding was removed. I got to wondering if Mattel's Killer Tomatoes and Crazy Clones tutorial programs might also be derivatives of David's original Killer Bees? Whilst I don't think we have access to Killer Tomatoes, Crazy Clones was released by Intellivision Productions on the Intellivision Lives CD. So, using the binary comparison tool I wrote to look for similarities in Coleco and Parker Bros games we can compare it with Killer Bees. We find that there are a total of 35 instructions shared by David Rolfe's version of Killer Bees and Crazy Clones, split into two sections. This includes the following 25 instruction run taken from David's source code... Now this might not seem like much, but Killer Bees is tiny, less than 400 decles in length, and it only contains about 150 instructions (over 130 decles are graphics data). Therefore, nearly 25% of David's Killer Bee code is found in Crazy Clones, a pretty high hit rate. As a consequence, I think there is a reasonable possibility that the ancestry of Intellivision tutorial games might look like this... As always, things are rarely straight forward, and bringing the new Mattel branded version of Killer Bees into the mix complicates things a bit. Although it is very similar to David's version (over 100 decles of shared graphics data in addition to 37 shared instructions), it shares less in common with Crazy Clones (two runs, totalling 28 shared instructions). So the relationship between the programs might not be as linear as is depicted here. Anyway, for those that are interested, the following zip archive contains the source code and binaries for the known versions of Killer Bees, including David's original, the new Mattel variant, and both my 2016 borked and a corrected 2021 version that can be built using as1600. killerBees.zip Cheers decle

I believe that @Lathe26 drew up a partial circuit of one of the boards, based on the work done by Tony and his friends at Hard4Games and some images. However, I think this is incomplete, but I'll leave it to @Lathe26 to say more on this. The boards were originally owned by the late Herbert Schmitz. I did some digging into their possible origins at the start of May. We're waiting on the thoughts of Herbert's family as to the veracity of the findings, so although I believe the facts below are accurate, the conclusions should be viewed as speculation. We know that Herbert worked for Sanders Associates and followed Royden Sanders when he left to form Sanders Technology. Herbert's family indicates that in the summer of 1976 Herbert was introduced to the toy manufacturer Milton Bradley (MB) through some work at Sanders. Sometime between then and 1979 Herbert moved to work at MB. In late 1979 the TI-99/4 was released. Its 4 release titles were written by Milton Bradley. This might seem strange, but MB were working on a console version of the Ti-99 called the GameVision. In fact, in late 1978 Mattel had discussions with TI and MB about the Intellivision also being based on the same design, this is described on page 5 of the Intellivision History and Philosophy document at papaintellivision.com. However, Mattel did their own thing with GI, and the GameVision console was dropped by MB. Anyway, one of the GameVision branded TI-99/4 launch titles was ZeroZap, which is credited to Herb. It looks like it was the standout launch title, although the bar was not the highest. The other three were Yahtzee, Hang Man and Connect-4. Interestingly Herbert's son, Karl is also credited as writing a couple of TI-99 games, Blackjack & Poker (no shifty eyed dealer) and Hustle. The trail then seems to go quiet for a bit. However, by 1982 MB fancy a slice of the console market again. They are reported to be working internally on a game console called Gemini this seems to have been intended to be games / educational console that could be expanded into a computer. The Gemini was unrelated to the GameVision, being based on the Motorola 6809 CPU rather than the TI TMS9900. Its unique selling point was that it had voice recognition, and not in a minimalist, TV POWWW! volume switch way. The Gemini had to be trained by the user and recognised different words. Anyway, it seems that MB's management got cold feet when they saw the Colecovision some time in the first half 1982 and pulled the plug. The Gemini got quite close to being released before management cancelled the project, here are some images of the prototype... So in mid/late 1982 MB's engineers were casting around looking to see how they could repurpose the Gemini technology. They hit on turning it into a peripheral for the TI-99/4a of all things, presumably because of their existing links with TI. The result looks suspiciously over complex, like the Intellivision Keyboard Component. It retains the 6809 processor, along with 4K RAM, voice recognition, a speech synthesizer and two joystick ports. It's nearly a console in its own right, but without the video chip, instead it communicates with the host 99/4a using the TI's bi-directional joystick port!? MB demonstrated their TI-99 wonder peripheral at the winter CES in January 1983. Among a load of other companies, Atari see what will become the Milton Bradley MBX and are impressed, so much so that they do a deal with MB to produce a version for both the Atari 2600 and 5200 to be called the Voice Commander. However, like so many things at the great Fuji, the project is cancelled and doesn't make it to market. MB were not happy with this turn of events and ended up suing Atari. MB finally scratched the game console itch in early 1983 when they bought GCE, manufacturers of the Vectrex, after it achieved strong initial sales following its launch in November 1982. It seems likely that MB's relationship with Jay Smith had something to do with this (his company had developed the earlier Microvision which was also sold by MB). Interestingly, although probably coincidentally, the Vectrex is also based around the 6809 CPU used by the MBX. Milton Bradley released the MBX for the TI-99/4a in late 1983, along with 10 games (the Gemini's edutainment roots are apparent in the four "Bright Beginnings" titles released alongside the six "Arcade Plus" games). You can see it working, complete with voice recognition, in this video: At this point the facts end and we head off into speculation... My hypothesis is that whilst developing the MBX from the ashes of the Gemini, and possibly whilst working on the 2600 Voice Commander, MB looked to see what it would take to do a similar peripheral for the Intellivision. I suspect these prototype boards represent their early work on that project. The software for the TI-99 MBX was distributed on cartridges that plugged into the host computer and ran on its CPU, so to do the same would require MB to write Intellivision games. You may ask what the point was of the 6809 in the MBX if the games ran on the TI/99? Well, the 6809 ran baked-in firmware to manage the various peripherals within the MBX. The TI-99 host sent the MBX high level commands to train for a voice command, listen for a voice command, say a phoneme, read a joystick, etc., and then the 6809 program would scurry around doing all the low level bit banging to actually action the request. Undoubtedly creating an MBX for the Intellivision would have been harder than for either the TI-99/4A or the Atari consoles. In addition to the crazy GI ROMs and the proprietary EXEC, the Intellivision's controllers were hardwired and are not bi-directional which would have made interfacing the MBX a bit more challenging. And then there is the fact that the Mattel already had the Intellivoice in the marketplace, its more characterful voices would undoubtedly have eroded some of the advantage of the MBX. Anyway, parking these concerns, MB would need to have written some Intellivision software to get an Inty MBX to market and these prototypes might have been part of that work, with the Gemini name on the title screen being a reference to the shelved console that was the origin of the technology. Now, as Roger Murtaugh might say "that's pretty f'in thin", and I wouldn't disagree with him. It's held together by baling wire of Herbert Schmitz -> Milton Bradley -> Gemini Console -> Gemini Title Copyright and let's be honest, Gemini isn't exactly an unusual name for a failed project. I was hoping that MB might have been known as MB Games & Toys in some context or other, but I can't find any evidence of that (they seem to have universally been known as The Milton Bradley Company). So there we are, it is what it is, a plausible, if "thin" hypothesis. Hopefully we will get some more input from someone who knew Herbert or worked at MB and we can confirm or refute it. If you want to know more about the Milton Bradley MBX, I recommend this AA thread...

After a long and painstaking investigation, I've finally uncovered the truth in Reading, England... ...@cmart604's Intellivision dungeon is funded by a clandestine noodle empire!

Apologies @CRV, although I found the GDRI page on Roklan, I didn't notice your discussion with Dimitri on the discussion page, otherwise I would have referenced it and credited you. Nice work For those that were curious about the possible link between Auto Racing, Astrosmash!, Space Hawk, AD&D, Motocross, Space Spartans and Tron Solar Sailor. It took a little thought, but once you see it, it's kind of obvious... This is exactly the kind of finger-printing of the ROMs I was hoping to find.

This is a long one, so apologies in advance. Whilst floating around Wikipedia I noticed that the List of Intellivision Games identified Roklan as the authors of the Coleco ports. And it seems that they also did the Parker Bros games as well. No doubt this is well known to most, but it was news to me, I hadn't even heard of Roklan Software. It seems they were pretty big in the arcade porting business. Ron Borta, the head of their games business, gave an interview to ANTIC back in 2016. As you might expect it's quite Atari centric, but it's worth a listen, if only for the host having a gentle dig about the quality of Inty Donkey Kong. Anyway, it got me thinking, could we tell that Coleco and Parker Bros games were written by the same company, just by inspecting the ROMs? So I wrote a little script that disassembles two games and then looks for common runs of 10 or more instructions found in both games*. The idea being that if the games were written using the same tools, they might show a greater number of larger matching chunks than titles that weren't. Then I compared every combination of Coleco, Parker Bros and Activision game to see how similar they were. The results can be seen in this diagram which shows the total number of shared instructions for each combination: The number of matching chunks is depicted by the colour of the cell. White means no significant shared code, then as the total size of matching chunks increases the colour changes from red to green. I only report games with two or more common chunks, or where the average length of the chunks is more than 20 instructions. The games are arranged by publisher and then by release date (as reported by Wikipedia). What we can see from this is that there are indications that the Coleco and Parker Bros games were written by the same company. Early Coleco games (Donkey Kong, Carnival, Mousetrap and Donkey Kong Jnr) mutually share code, and they also have some similarities with most Parker Bros games. We can also see that, as expected, these games don't share much in common with Activision games. It seems the later 1983 Coleco games don't have much in common with either their earlier syblings, with each other, or with the Parker Bros games which were also released in 1983. Perhaps Zaxxon represents a transition between the two sets. Could this represent a change in the Roklan tools or the way they developed games? So the short answer to my question is yes, there are some indications baked into the ROMs that the Coleco and Parker Bros games were written using the same tools. Now, you might be wondering what happens if we apply this process to the full 125 game catalogue? Well, this is what what we see (you might want to download and pan around the image)... It looks as though "Kings of Consistency" were Imagic (blue ellipse) and Atarisoft (black ellipse). Within the Imagic titles we can see what might be an interesting change, with the earlier games up to Microsurgeon having less in common than those that follow, and clearly Nova Blast and Truckin' seem to be oddballs. Again perhaps this represents a switch in development tools? It seems that overall there is much less consistency in INTV games... ...and even less for Mattel titles... ...as exemplified by the lower amount of colour. This could be down to the fact that in many cases much of their common library code is found in the EXEC, rather than the cartridges themselves. That said, going back to the full diagram, links between games that are known to be related can be seen, for example Math Fun and Learning Fun 1, Backgammon and Triple Challenge or Hockey and Slap Shot. We can also see other, less obvious connections; like Tron Deadly Discs and Dig Dug, which are related through the hidden game Deadly Dogs, or Checkers and Sub Hunt which are linked through their common use of the tune Ride of the Valkyries. Some common ancestry of games is also visible, for example Dig Dug's origins as an Atarisoft title, and Space Hawk being a derivative of the hidden game Meteor! within Astrosmash! I don't think there are any unknown gems hidden amongst the shared code, but clearly there are some less obvious links, for example Auto Racing shares quite a bit in common with a number of titles including Astrosmash, Space Hawk, AD&D, Motocross, Space Spartans and Tron Solar Sailor. If you fancy a dive through the data for yourself, here's the spreadsheet... intyGameCommonality.ods Cheers decle * The actual process is rather more involved. For the curious...

For those, like @DZ-Jay, @Lathe26, @Steve Jones and @Morpheus, who expressed an interest in how Intelli-talk works, here's a (much shorter) follow up video showing Intelli-talk connecting to a UK BBS using dial-up. It also explains how the posting to AtariAge was done, what was real, what wasn't etc. Naturally, this will contain some spoilers... Cheers decle

Hi everyone, Ever wondered how @nurmix posts AtariAge messages from his K/C? Wonder no more! I have a short view review of @Ron The Cat's K/C running Jack's Conversational Intelli-talk, the software he uses... Let me know what you think. Cheers decle --------------------------------------------------------------------------------- Sent from my Keyboard Component using Jack's Conversational Intelli-talk Cassette

Unfortunately, a little bit of research suggests that properly straightening a warped PCB probably needs both heat and force... https://www.circuitrework.com/guides/3-2.html and there is some scepticism that it can be done without potentially compromising other components or solder joints... https://www.circuitnet.com/experts/88133.html I guess the good news is that given where you're at, you don't have much to lose trying something. Interestingly, despite what the 8-bit Guy says, it looks as though the ECS Music Keyboard was built by a reputable manufacturer, Pratt-Read. It seems they manufactured the keybeds for the Polymoog, Sequential Prophet 5, Oberheim OB-Xa and other famous synths. Their supplying Mattel is mentioned by Gary Moskovitz at 46:30 in his Intellivisionaries interview with @nurmix and is corroborated by this history of Pratt-Read: "A major blow came in 1984 with the loss of a large contract to manufacture keyboards for Mattel Electronics's Intellivision home video system. This was exacerbated by a slump in the piano industry in mid year." I guess the Mattel keybed came from the budget range

I play Utopia loads with my son too... ...so I hope our thoughts aren't too late. We think the biggest thing is a bit of rebalancing of the rules / scoring. As was noted by @mr_me back in 2016 (and at the risk of spoiling the original game)... https://atariage.com/forums/topic/247426-super-pro-utopia/?do=findComment&comment=3408616 ...population size does not influence either the score, or the utility of units, and let's just say that this has a strong influence on strategy. We also find that we tend not to use aggressive units (PT boats and rebels), so something to increase their value might be good. With regard to suggestions made by others, we can see that both a single player mode and random islands would be great. However, as has been noted, the fixed weather pattern and the important place it plays might make random but fair terrain tricky. Perhaps one way round this would be not to tie each player to one island, but instead to allow them to build anywhere (perhaps with some incentive to cluster building to prevent scattering units anywhere). This might make for another ending to the game through territorial domination, and that perhaps might make the rebels more useful as a means of clearing land for you to occupy. Perhaps this would also allow the weather patterns to be changed from one game to another, and there might then a requirement for players to learn the weather pattern to find the optimal placing of crops and units. Anyway, I think that's enough for now. Needless to say we are very interested in progress. Cheers decle & hoalst

Once again @ricardowilmers has come up with some really intriguing pictures! Let's start with this one, which confirms that Dracula is a ROM variant on Shock Vision: However, Mission X does not seem to be on Intelligame: Which is interesting because it is in its Shock Vision version: Then there is his switched AD&D / Dracula Intelligame cart. This is not the only example of a switched Intelligame title, we have one with Swords and Serpents and White Water found by Intellivision Brasil: However, I think this is the first time we've seen one with these titles. Initially it looks as though AD&D might be a ROM variant: However, I think this is just a glitch. Firstly because the Dracula title screen is unchanged, and as we have seen with Shock Vision, tweaking this is easy. Then there is an 1882 copyright date, and Copr becomes Bkpr and Mattel becomes Lapp`h. This data is all located together in the ROM and with the exception of ' becoming e, each switched character is only one bit away from its correct value. There is also another image with what appears to be the correct title screen. Nonetheless, it is a cool cartridge and this is the first time we've gotten to look inside one, so thanks for showing us it's guts @ricardowilmers! Internally it looks to be in great shape and is certainly a much more refined and sophisticated product than my copy of Intelligame Utopia, which seems to have been hacked together very cheaply (unbelievably it does still work!) @ricardowilmers, thanks so much for sharing these images. Before you ship this lot to @cmart604's dungeon of Intellivision , could you plug each Shock Vision and Intelligame cartridge in and take a picture of its title screen with the console and cartridge in view, just like this one you did for MissionX please? An image like this is enough for us to confirm which game is which, and also whether it is likely to be a ROM variant or not. Thanks once again.

This is my current list of the 50 known Shock Vision and Intelligame pirate carts, gathered from images on the internet. Those with entries in the variant columns have been dumped and their status as ROM variations is confirmed (or refuted). Known ROM variations only differ by a hacked copyright message on the title screen. As a consequence, it is unlikely that any Mattel titles with stock title screens will be ROM variants. Ricardo's images added quite a few new titles to the list, for example Astrosmash, Frog Bog, Golf and Sharp Shot. However, whilst his collection is certainly the most comprehensive seen by some margin, it is still some way from being complete . I still find it interesting that for a football mad country no version of Soccer has turned up yet on either Shock Vision or Intelligame. @ricardowilmers, it is not clear from cartridge label in the image what game this is... Just for completeness, could you let us know what it is please? Thanks.

Looking forward to seeing the end result! I think the toughest thing to source is a good connector / cable. I've discovered the connector I used does not fit the ECS (and possibly the Intellivision II) because the controller ports are heavily recessed. Like you I've found the narrow moulded plugs on the Genesis extension cables fit fine. However, because the cables have 9 cores the wires are very fine (and fragile) which is less than ideal. It looks as though original 2600 joysticks may have beefier cabling, I may try to pick one up and see: Edit: It looks as though a 2600 joystick cable is no good because pins 5 and 7 are not used and therefore won't be wired to the connector. Back to the drawing board. If you find a better solution, please let me know. Personally I'd go with 3 AAA's for the power source, the slightly small form factor opens up one or two cool options for packaging