193 replies to this topic

[krewat]

Hello all...

I am Arthur Krewat - the guy who did the final conversion of the C64 Impossible Mission to the Atari 7800.

It was originally started by another contractor in the company (Brian Richter, Computer Magic, LTD), but I had to finish it up after he quit. I had to re-work major sections of the game because of performance - mostly graphics.

For the record, I did supply Atari with a game that was able to be completed. I played it myself all the way to the end a few times before giving it over to the Atari testing department and they signed off on it so Computer Magic would get paid for the game.

However, they either used an earlier version from during development, or did something else to reduce the amount of RAM it required and introduced that "unfinishable" bug.

Does anyone have any info about the fix?

Does anyone have a copy of the IM ROM? I have one I found on the net, but not sure it's not corrupt.

I checked it against some HEX files I have of the version I supplied to Atari, and it appears to be intact, except there are areas that do not match WHATSOEVER. Yes, I have the source code, or at least 99% of it. I have to reconstruct one file from an earlier version I had, which is also why I am hunting for a good ROM image. I want to examine that ROM image and reconstruct one ASM file from it.

Thanks all.

DO NOT ask for the source code. Yet. I signed a non-disclosure agreement, granted that was 20+ years ago, but I am still bound by it. I want to figure out what the problem with Atari's version was, and see if it's my fault or not

PS: Anyone know anything about 2500 AD x6502 cross compiler for MSDOS (command was x6502) plus it's linker (link2) ?

Edited by krewat, Fri Dec 5, 2008 4:36 PM.

[krewat]

By the way, I updated the Wikipedia section for Impossible Mission:

http://en.wikipedia....ossible_Mission

[krewat]

Duh, found the ROM image here on this site - same as the one I found on the web elsewhere (really eMule).

Will set out soon to figure out if it's any different than the HEX file I have.

[supercat]

krewat, on Fri Dec 5, 2008 4:33 PM, said:

I am Arthur Krewat - the guy who did the final conversion of the C64 Impossible Mission to the Atari 7800.

Greetings. There are quite a few things about the 7800 Impossible Mission that impress me. I was very happy to see that you used 320 mode where possible on the control panel; one of my pet peeves with the 7800 is that a lot of games use 160 mode even when it's totally silly (e.g. the opening text screen on Klax). Your use of 320 mode is much appreciated.

I found it curious that the music room screen was smaller than on the 64 version. Was that because the 7800's TIA can't manage the same range of notes that the C64 can?

Were you given a RAM and ROM target to start with, and did you meet or beat the original goal that was set? Would there have been any possibility of getting enough ROM space for a little crude speech or something resembling speech? Probably not enough for Melvin's opening monologue, but the monologue was a great part of what set the atmosphere for the C64 game.

[the.golden.ax]

Welcome to AA! I hope you will be inspired to take up where you left off with 7800 and perhaps work with it again. Lot's of us fans still play it.

Thanks for everything!
AX

[DracIsBack]

Welcome! Always nice to see someone who developed for the 7800 back in the day! I always thought IMPOSSIBLE MISSION was a blast to play ... just wish it was able to be finished. I saved long and hard for that game only to find I could never beat it.

The experience must have been interesting working with the Tramiels! They always seemed so cheap with everything. I always wondered why you didn't attempt the digitized sound effects (there are 2600 and 7800 games that pull this off with TIA) but when you gave the space example, it becomes very clear.

Thanks for sharing your story.

Did you ever work on any other 7800 titles besides the Epyx titles? Always wondered by the 7800 SUMMER GAMES had fewer events than the XEGS version. :-(Seemed like Atari being cheap again.

Edited by DracIsBack, Fri Dec 5, 2008 5:53 PM.

[krewat]

Thanks for the welcome guys!

I only worked on the Epyx titles IM, Summer and Winter games. I have 7800 code for California Games, and One on One (no idea what that is) but I don't know WHY. I don't remember working on them, but then, I was a supervisor for a few other people at the time, I might have just been collecting backups of what they did. Haven't had the chance to get to those backups yet. Too many bad sectors to do a simple DOS restore. Yet.

I'm not too sure of the whole business arrangement with Atari and Epyx. I worked for a contactor called Computer Magic, LTD (they did Robot*Link which I also created the PC version of). Atari themselves did not really give any real targets for RAM or ROM space, but when they found I wanted to use 256Kbit Static RAM chips they puked. Static RAM at the time was very expensive. So much so that I think the 256Kbit one was more than the cartridge would sell for (or close enough they wouldn't make any money).

I don't remember anything specific about the number of notes on that "organ" ... I think that was written up by Brian Richter before I ever got hold of the project. One of the many things that I didn't touch. It might have had more to do with the limited graphics ability of the 7800 compared to the Commodore 64. The Atari 7800's graphics chip took memory cycles from the processor to accomplish anything. The more you did with graphics, the less processor time you had. Or at least, that's the way I remember it.

I'll remember more and more stuff about the 7800 as time goes on. As it is, I only found the backups of IM and SG/WG a few weeks ago. After cobbling together a PC with a 5.25" drive and DOS, which took a while...

Tramiel was Commodore, wasn't he? For Atari, I remember working with a few guys, Tom someone, VP of video games (or some such title), and then Mike Katz actually visited my home (when I was 20 living with my mother) and saw a demo of Winter Games I think it was. I remember burning a copy of WG (or was it SG?) onto EEPROMs and hand delivering it to the Hilton in Manhattan for a show

[DracIsBack]

Thanks for sharing! Hope you find that California Games code. Lost 7800 games were the stuff of legend for a long time. Thanks to this site and its visitors, a number of them have surfaced (Plutos, Sirius, Chuck Norris: Missing In Action, Klax, Pit Fighter, Gato, Rescue on Fractalus) and it would be awesome to have more treasures unearthed.

Quote

I'm not too sure of the whole business arrangement with Atari and Epyx. I worked for a contactor called Computer Magic, LTD (they did Robot*Link which I also created the PC version of). Atari themselves did not really give any real targets for RAM or ROM space, but when they found I wanted to use 256Kbit Static RAM chips they puked. Static RAM at the time was very expensive. So much so that I think the 256Kbit one was more than the cartridge would sell for (or close enough they wouldn't make any money).

In the 7800 docs, they expressely forbade the use of extra RAM without written permission from the Tramiels. There are a few stories of games getting killed (an RPG called Time Lords) or developers getting into arguments with the Tramiels over the cost of games in more expensive carts. Did you do any work on some of the other levels in SUMMER GAMES and WINTER GAMES which didn't make the cart?

I always liked how you made the 7800 version of IMPOSSIBLE MISSION more colorful than what I remember on the C64.

Appreciate all the great stories! Hope you'll keep joining us!Check out what some of the homebrewers have been doing on the 7800!

Edited by DracIsBack, Fri Dec 5, 2008 7:04 PM.

[Mayhem]

Welcome Arthur

Impossible Mission is one of my favourite C64 games of all time and I was quite impressed with the 7800 conversion of it. It IS curious that while the NTSC released version of the game is bugged, the PAL version is perfectly fine. I wonder if it's true that it was fixed in transition, or maybe it was converted from a different NTSC release. Your revelation that you delivered a working game to Atari certainly puts the cat amongst the pigeons.

One on One was a basketball game, originally released in 1983 but updated and improved for the 7800 five years later. But you have California Games code for 7800?! Oh... wow... I'm sure other people are gonna chime in that we'd love to see either shots of it in action, or even perhaps the ROM file itself, if possible. The game was never released you see. We've had some unreleased games through the doors via various sources in the past (Rescue on Fractalus, GATO etc) but I'm not sure we've seen anything of this (needless to say someone will correct me if I'm wrong).

Given Summer and Winter Games made it, then it would be natural to assume California Games would follow... but it never did. Another one to chalk up for not being released. How complete do you think it is? I guess you probably won't remember from 20 years ago. Hope you find the code somewhere...!

Edited by Mayhem, Fri Dec 5, 2008 7:07 PM.

[Mitch]

Hey cool, glad to see an original 7800 programmer here.
I'm not sure if you found it but if you search the forums here you will find a copy of the fixed version of Impossible Mission that I posted a few months ago. I managed to pick up copies of the master EPROMs for Impossible Mission along with the paperwork (Checksums, manager signoffs, etc.). The only problem with the fixed version is that the puzzle rooms are buggy in that they don't work ~75% of the time.

If you are able to get the clearance I'm sure we'd all love to see the various source code you have, especially California Games! Some 7800 source code was already released for games like Crossbow and Commando and also the 7800 BIOS.

Mitch

Edit: Found the thread: http://www.atariage....howtopic=131223

[Mitch]

krewat, on Fri Dec 5, 2008 2:47 PM, said:

Duh, found the ROM image here on this site - same as the one I found on the web elsewhere (really eMule).

Will set out soon to figure out if it's any different than the HEX file I have.

Do you need a converter for your hex file? I have a freeware one I've used in the past for the same thing. I can send it to you if you want.

Mitch

[krewat]

Funny, I don't remember Tramiel at all... Just Katz. Who seemed like a cool dude who would go around the world to get the developers what they wanted just so that they could sell a GOOD game. At least, that was my impression of him.

As for RAM in the cartridge, I don't remember any restrictions at all, but then, no one told us that it was possible either. I just remember someone hit the roof when IM needed a 256Kbit RAM - or maybe it was SG or WG - I don't know. I haven't delved back into the IM code enough yet to remember all that yet. I was using 32Kbit or even 64Kbit static RAMs for at least one of the games and didn't get any guff for it. But the 256Kbit one, I remember. Which might be why IM got a bug.

I lost my blueprints for the 7800 long ago. As well as any programming docs. Funny, reading some stuff from this site, I see references to the JAN chip. Which I did see stuff for in what, 1986? 1987 maybe? But I know Maria quite well - or did I just found the development docs on this site. Great stuff!

Anyway, forgive my bad memory, I might have some of it (or all of it) backwards, upside down, or reversed in some way, but ... it's coming back slowly but surely.

Ahh... just looked at the "development card" schematic for the 7800. I remember it being dynamic RAM that required refresh cycles?

I made a static RAM card instead - a LOT less wiring. I remember they were quite interested in my static RAM card at Atari

And they didn't like the fact that I kept supplying game images on PC floppies instead of Atari ST media (ROTFL)... And, using the IBM's parallel port to talk through the joystick ports (in 8 bits) to my own EEPROM.

Needless to say, I didn't use much of their "development kit".

Too bad I seem to have lost my Atari transfer program for the IBM-PC - Really, I was using my own XT at the time, and at some time later, an IBM-AT. A much nicer platform than the Atari ST (no offense).

As for One-On-One, I found source code that included some stuff that has my fingerprint on it, not game-wise, though. I think I might have been getting ready to work on it, but didn't? Have to keep looking...

[krewat]

For reference, the latest file I have in the backup I have is July 5, 1988

Mitch, that one set of EEPROMS you found are dated July 8th 1988. Coincidence? I don't think so.

Still checking on whether or not that code is the same as the HEX files I have. Will report back tomorrow.

[PacManPlus]

Very nice

...and Welcome! I really is great to see a 7800 programmer here.

Bob

[krewat]

Well, it's official.

The transfer files I have from my backup that are dated July 5, 1988, MATCH the ROM image available here byte for byte.

The only difference is in the encryption area from 1FF80 through 1FFF7 in the ROM (after removing the header from the ROM file).

Oh well, guess it WAS my fault.

[HammR25]

oops, you made the game live up to it's name.

[DracIsBack]

HammR25, on Sat Dec 6, 2008 12:18 AM, said:

oops, you made the game live up to it's name.

There's also the issue of Atari doing testing and verification before production.

[Atarifever]

And having seen you try to track the mistake down 20 years later, I imagine some reviews will have to be re-written on fan pages, as it's obvious it wasn't carelessness. I mean, I make mistakes at work and forget them five minutes later.

[Mayhem]

So in easy layman's terms here... the bug was caused by an incorrect encryption procedure for the NTSC 7800? And because the PAL version didn't need encrypting, the bug was not introduced? Is that a good summation?

[krewat]

No, the encryption is there so that the cartridge will play in an unmodified 7800. A copy-protection of sorts. Doesn't really stop anything, but it's there nonetheless.

What I'm saying is that the code I supplied to Atari is right there in the ROM, byte-for-byte, bit-for-bit. The ONLY difference is where Atari put the encryption data.

Diffs between Mitch's 8-11-89 copy, and the released ROM:

At location 14000 in the ROM file (offsets below are relative to location 14000 in the ROM file):

< denotes 8-11-89 copy, > denotes original ROM

< 1f80: d0 f8 9d 40 67 ca 68 9d 40 67 ad 84 2 ea ea ea ...@g.h.@g......
< 1f90: ea 29 3 c9 3 f0 f3 ea ea 9d 70 67 ee 12 6e a9 .)........pg..n.
---
> 1f80: d0 f8 9d 40 67 ca 68 9d 40 67 ad 0 20 4a 4a 4a ...@g.h.@g.. JJJ
> 1f90: 4a 29 3 c9 3 90 2 a9 0 9d 70 67 ee 12 6e a9 J)........pg..n.

< 3ff0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 0 ................
---
> 3ff0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................


At location 10000 in the ROM file:

< ec0: c d8 a9 7 20 ca 8e 60 a9 7 48 ad 7 21 ea ea .... ..`..H..!..
---
> ec0: c d8 a9 7 20 ca 8e 60 a9 7 48 ad 7 21 d0 fb .... ..`..H..!..

< 2a30: e8 c8 c0 9 d0 f4 60 a2 23 20 1e b9 8a 99 0 60 ......`.# .....`
< 2a40: ca 10 f6 60 a2 0 8a 9d 12 78 e8 d0 fa a2 f 8a ...`.....x......
---
> 2a30: e8 c8 c0 9 d0 f4 60 a2 23 20 6a ae 8a 99 0 60 ......`.# j....`
> 2a40: ca d0 f6 60 a2 0 8a 9d 12 78 e8 d0 fa a2 f 8a ...`.....x......

< 3810: ff ff ff ff ff ff ff ff ff ff ff ff ff ff 0 0 ................
< 3820: 0 0 0 0 0 1 1 1 0 0 0 0 0 0 1 1 ................
< 3830: 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 ................
< 3840: 0 1 1 0 0 0 0 0 1 0 0 0 0 0 1 0 ................
< 3850: 0 0 1 1 0 0 0 0 1 1 0 0 1 0 0 0 ................
< 3860: 0 1 0 0 0 0 0 1 1 0 0 1 0 0 0 0 ................
< 3870: 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 ................
< 3880: 0 1 1 0 0 0 0 0 1 1 0 0 0 1 0 0 ................
< 3890: 0 0 0 0 1 0 0 0 1 1 0 0 0 0 0 1 ................
< 38a0: 1 1 0 0 0 1 1 0 0 0 0 0 1 1 0 0 ................
< 38b0: 1 1 0 0 0 0 0 0 1 1 1 0 0 0 0 1 ................
< 38c0: 0 0 0 1 0 0 0 0 1 1 1 1 1 1 1 1 ................
< 38d0: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ................
< 38e0: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ................
< 38f0: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ................
< 3900: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ................
< 3910: 1 1 1 1 1 1 1 1 1 1 1 1 1 1 20 6a .............. j
< 3920: ae b9 1e b8 d0 f8 a9 0 8d aa 73 60 ff ff ff ff ..........s`....
---
> 3810: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3820: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3830: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3840: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3850: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3860: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3870: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3880: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3890: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 38a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 38b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 38c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 38d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 38e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 38f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3900: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3910: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 3920: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................

< 3ff0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 0 ................
---
> 3ff0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................

[krewat]

And not to backpedal at all, but...

There were instances where because I was running the code in a static RAM development cartridge, after burning real EPROMs a bug would come up that wasn't there initially. If I remember correctly, Atari tested the game with development cartridges. The only time I had a bug like that come up was in Summer Games or Winter Games though, it never affected IM. If I remember correctly.

Also, I remember Atari coming back after releasing IM and saying that there was a bug that made it impossible to finish. I had heard they decided to sell the rest of the cartridges that they had produced in Turkey or Hungary or some other foreign country

ALSO, just because the code is exactly what I gave them doesn't mean that the RAM they put in the cartridge was correct either.

--

Does anyone have information on the RAM in the IM cartridge, and whether it was static RAM, dynamic ram, how much of it there was, what RAM chips they used, and what support chips were included?

[DracIsBack]

krewat, on Sat Dec 6, 2008 12:36 PM, said:

Also, I remember Atari coming back after releasing IM and saying that there was a bug that made it impossible to finish. I had heard they decided to sell the rest of the cartridges that they had produced in Turkey or Hungary or some other foreign country

I remember Don Thomas saying on www.ICWhen.com (great site that seems to be missing) that they released a "few" to collectors. I actually bought my copy @ suncoast games and it looked as widely distributed as the other titles.Always wondered how legit the claim of "only a handful were released in the US" actually was. I didn't have to look hard to get it.

Edited by DracIsBack, Sat Dec 6, 2008 11:46 AM.

[Shawn Sr.]

krewat, on Sat Dec 6, 2008 12:36 PM, said:

Does anyone have information on the RAM in the IM cartridge, and whether it was static RAM, dynamic ram, how much of it there was, what RAM chips they used, and what support chips were included?

It's SRAM inside IM.

[tcv]

krewat, on Sat Dec 6, 2008 11:36 AM, said:

And not to backpedal at all, but...

Look, dude. You can always claim this was by design. The name of the game is IMPOSSIBLE Mission. Were it to be named POSSIBLE Mission, then you could call it a bug. But, seriously, you're not supposed to be able to solve the mission. I was never able to solve the C64 version. Now, this kinda was because I was an impatient jerk, but that's irrelevant. In a sense, I suppose, I did Stay Forever.

[krewat]

Shawn Sr., on Sat Dec 6, 2008 12:50 PM, said:

It's SRAM inside IM.

Any idea what the chip numbers are and what the support chips are?