Jaguar game developement and pre-existing graphic engines

[Thunderbird]

quote:

---

Originally posted by Glenn Saunders:

The development time of Battlesphere is a perfect case study on how long it takes to get a sufficiently state of the art game done part-time on the Jag.

---

 

Please remember that we had a working version of "Free-For-All" with sound, music, explosions, enemy AI, limited networking, and even some titles / menus in 1995, which was 1.5 years after we started. We had to build all of our own "engines" for 3D, sound, music, network, etc. which took about a year of that time...

 

So, actual game-specific coding and artwork for the early version of the game took 2 people only 6 months to do.

 

Each additional play-variation took 4-6 months to complete and test.

 

What I'm saying is that any single play-mode of BS would qualify as a nice game by itself and would have made a nice "homebrew" type of title.

 

I personally would love to see some cool arcade ports of easy games like Gyrus, or Sinistar, or Dig-Dug, etc. Very easy for a single hobby person to make for the Jag.

 

But nobody has accepted the challenge yet.

[Starcat]

quote:

---

Originally posted by Tyrant:

Which just proves my other point, there isnt much info easily availible to wannabie jag developers. I did look into the possibility about a year ago, I tried to find as much documentation and info as I could on the net, and came up practically empty, I found a few of atari's docs,

---

At that time the Jaguar Development Manual was already available on my site.

 

quote:

---

which I seem to remember were insanely complex and very low level,

---

ISANELY COMPLEX ??? What are you talking about ? If I understood the parts I needed to without ANY previous programming knowledge it can't be so complicated.

 

You can also learn Jag coding without even reading the manual. (a lot of the underground coders didn't have it)

Just by reading examples and trying to understand them.

quote:

---

>but on the whole, I couldnt find much >>>practical newbie advice at all.

---

It is a manual for coders. It's not "Jag coding for dummys" book.

They expected that the coder knows the basics of coding.

 

quote:

---

>Not to mention that quite a lot of the reports are conflicting, you say the jaguar is a nice system to code on, and well documented, yet I've read in several places that its one of the hardest systems there is.

---

A lot of people (especially the ones who failed) claim that the Jaguar is incredible hard to code. You can be sure that it is easy to code as long as you don't try to push the system to its limits.

quote:

---

>What we really need is an easy cheap way to get development kits, and distribution setups,

---

There are enough ways: BJL, JUGS, Jag Server, JagOs, JagFree CD, JUGS DD etc.

 

quote:

---

We also then need a dedicated site that is well known just for jaguar programers,

---

There is one. www.atari-jaguar64.de

I think my site has the biggest collection of Jag Coding related infos and downloads on the net.

 

quote:

---

>with totorials on how to write simple games, moving on to the more complex articles on 3d engines and such,

---

 

Just like Thunderbird said, there ARE tutorials and examples by Atari. Also for a 3D Engine.

I could do tutorials for simple games anytime. But there were too many people, who took my program code and claimed that they did it. All of these wannabe "developers" failed and left the scene so far, but there is still a chance that it happens again and again.

 

And why should somebody do an tutorial how to do a whole game ????

After all that is what coding is all about. If you can't think how to code something yourself you can't code.

Tutorials can help you to understand how a system works, but it doesn't have to be a whole game. A little display example and things like that are enough.

quote:

---

>Im sure if all the active jag developers put in a little time to help out the newbies, and if the efforts were all coordinated and availible through a single easy to use website, we could have a whole flock of new jag games.

---

 

No. This doesn't work because:

 

1) I tried that since I entered the Jag scene. With little to no success at all.

2) There are only very few coders

3) Those coders have hardly time for their projects, how could they work on examples and tutorials ?

[Starcat]

Hi !

 

Sorry about the post format. Accidently pressed "send" instead of "preview"

 

Regards, Lars.

[Tyrant]

Ok, I must admit that it was a long time ago that I looked for the info, and Im not entirely sure what I found even, but that info which I did find was confusing in the extreme. I am a fairly good coder, although I've never done... well essentially nothing in assembily. When I get time I will go through all the resources I can get my hands on and seriously give it a go. Oh and my point on the distribution / development hacks was that there are too many of them, a single, agreed upon system for distributing freeware / shareware games via the net would be useful. As far as I know (and I admit I have been out of the loop on jag development), currently bjl and jugs are the main systems, right? requiring extra hardware in the form of an expensive cartidge game or a new chip. Didnt Carl release a new system again to support cd games? Does that load them into the working ram like the other mods or does it actually run them from cd? It isnt a huge problem, but is there a single agreed upon standard for how to release games and get the most people possible able to play them?

 

Anyway, I cant really add much more to this thread without more research, which I will do when I have time for it. I have presented my thoughts and opinions about how I percieved jag development, if I was wrong, thats great, I'll try to do some developing once I get time (I just started a huge vb project tho so I doubt I'll have much time soon).

[Thunderbird]

quote:

---

Originally posted by Tyrant:

Didnt Carl release a new system again to support cd games? Does that load them into the working ram like the other mods or does it actually run them from cd?

---

 

You must be thinking of the Ingenious J.U.G.S. D.D. system from ScatoLOGIC. It allows anyone with a copy of BattleSphere Gold to load and run any CD game which follows the Atari Standard CD Layout.

 

This means that any prototype CD game as well as any new homebrew CD game can load and run with ease.

 

And just to make it clearm ALL CD games on the Jaguar load into RAM to run. Usually the games will load in a main code loop and this would load level data as needed. Loading into RAM is not a disadvantage, as all CD based consoles must do this as well.

[Tyrant]

quote:

---

Originally posted by Thunderbird:

Actually, there's a lot of things wrong with your statement.

---

 

Which just proves my other point, there isnt much info easily availible to wannabie jag developers. I did look into the possibility about a year ago, I tried to find as much documentation and info as I could on the net, and came up practically empty, I found a few of atari's docs, which I seem to remember were insanely complex and very low level, but on the whole, I couldnt find much practical newbie advice at all. Not to mention that quite a lot of the reports are conflicting, you say the jaguar is a nice system to code on, and well documented, yet I've read in several places that its one of the hardest systems there is.

 

What we really need is an easy cheap way to get development kits, and distribution setups, in a form all the big programers can agree on so you wont need jugs to play some games, bjl to play others, flash carts for yet more, etc, better yet would be if someone could find / crack the cd encription key, so no extra hardware is needed. How many different ways exist to get live debuging working? obviously there's alpine boards, which are too expensive for the amature coder I believe, how many different kinds of hacked jags allow debuging of code? and is anyone selling them completed so as not to have to rewire the boards, most programers dont like arnt really hardware guys hehe.

 

We also then need a dedicated site that is well known just for jaguar programers, with totorials on how to write simple games, moving on to the more complex articles on 3d engines and such, Im sure if all the active jag developers put in a little time to help out the newbies, and if the efforts were all coordinated and availible through a single easy to use website, we could have a whole flock of new jag games.

[Gunstar]

OK, I was going to quote someone here, but I couldn't decide the post, so...you guys are talking about all these difficulties in programming, and I'm not a coder, so I don't know anything about this stuff in the way you guys do, but this was my whole point of starting this thread...since it does take so long to start a REAL project for the Jag (not space invaders), not being a coder, I assumed it would be much more helpful for you coders to start with pre-existing engine than from scratch. I'm still hereing two different view points here though; 1) the Jag's a buggy beast and hard to program-wouldn't a pre-existing engine help this? 2) the Jag's actually easy and the tools are great-I would still think something is better than starting with nothing. But you guys are telling me that it's really not any easier to start with a pre-made "engine" than starting from scratch. I naturally assumed the opposite because you do see so many developers using pre-existing engines and companies like iD who license them out. An example is earlier in this thread when it was discussed that the 32X DOOM is the Jaguar DOOM at it's core...why did they do this if it was no easier than starting from scratch? Why such a market for licensing out 3D engines if it's no easier than making your own? I mean, Most fps games these days are running on a Quake or Half-life engine, I'm sure Halo's engine will probably be licensed eventually too, if it has not already been...so why would people license these "engines" if there is no advantage to writing their own? There must be SOME advantage, no? I mean, if I wanted to do a Jag game, not knowing coding, I think it would be great if I had this completely developed 3d engine and all I had to do was design levels and design my own inhabitants of the world and throw in my textures and such, this would be perfect for me as an artist, this is done all the time with the shareware DOOM stuff out there...I just assumed that there would be a similiar circumstance for developers with pre-existing Jag engines, obviosly not as easy as designing new levels and monsters for DOOM, but easier than starting from scratch...

[Gunstar]

quote:

---

Originally posted by Tyrant:

Which just proves my other point, there isnt much info easily availible to wannabie jag developers. I did look into the possibility about a year ago, I tried to find as much documentation and info as I could on the net, and came up practically empty, I found a few of atari's docs, which I seem to remember were insanely complex and very low level, but on the whole, I couldnt find much practical newbie advice at all. Not to mention that quite a lot of the reports are conflicting, you say the jaguar is a nice system to code on, and well documented, yet I've read in several places that its one of the hardest systems there is.

 

What we really need is an easy cheap way to get development kits, and distribution setups, in a form all the big programers can agree on so you wont need jugs to play some games, bjl to play others, flash carts for yet more, etc, better yet would be if someone could find / crack the cd encription key, so no extra hardware is needed. How many different ways exist to get live debuging working? obviously there's alpine boards, which are too expensive for the amature coder I believe, how many different kinds of hacked jags allow debuging of code? and is anyone selling them completed so as not to have to rewire the boards, most programers dont like arnt really hardware guys hehe.

 

We also then need a dedicated site that is well known just for jaguar programers, with totorials on how to write simple games, moving on to the more complex articles on 3d engines and such, Im sure if all the active jag developers put in a little time to help out the newbies, and if the efforts were all coordinated and availible through a single easy to use website, we could have a whole flock of new jag games.

---

 

There are several sites with TONS of stuff, that I've come across just surfing...not even looking for developer's stuff...One is the "BADCODER" site with a guy that has been running a tutorial and shows with great demos just how it's done and he has a ton of info...unfortunately I forget the site here so someone else will have to help...and what about the German site here:

Jag developer's haven

 

[ 05-12-2002: Message edited by: Gunstar ]

[mos6507]

quote:

---

Originally posted by Thunderbird:

1) Writing in assembly KICKS @$$! It's more fun than any other way to code! The VCS and 7800 and 5200 seem to ONLY be programmable in Assembly, yet they have droves of developers.

---

 

I wouldn't say droves. The 5200 has only recently got a couple people homebrewing on it. There was nothing before.

 

Sure, there are a lot of people on Stellalist but the output is few and far between due to the technical challenges in doing even the simplest games on the VCS. The average coder, let alone an assembly novice like me has the odds stacked against them. (We do, however, have a tightly knit community which may be lacking for the Jag. To a great extent all VCS homebrews are collaborations.)

 

I think with the Jag the expectations are definitely higher. Sure, some people are writing simple applets and shareware games, but these people aren't getting the kind of attention someone who writes a Vectrex or 2600 game gets with a similarly minimalistic game.

 

Even given the Jag's hardware compared to today, I think people (including the developers themselves) expect more to be done than is generally possible as a part time hobby affair.

 

People are not going to get that excited about 2D Jag homebrews.

 

The development time of Battlesphere is a perfect case study on how long it takes to get a sufficiently state of the art game done part-time on the Jag.

[Thunderbird]

quote:

---

Originally posted by Gunstar:

I assumed it would be much more helpful for you coders to start with pre-existing engine than from scratch. I'm still hereing two different view points here though; 1) the Jag's a buggy beast and hard to program-wouldn't a pre-existing engine help this? 2) the Jag's actually easy and the tools are great-I would still think something is better than starting with nothing. But you guys are telling me that it's really not any easier to start with a pre-made "engine" than starting from scratch. I naturally assumed the opposite because you do see so many developers using pre-existing engines and companies like iD who license them out. An example is earlier in this thread when it was discussed that the 32X DOOM is the Jaguar DOOM at it's core...why did they do this if it was no easier than starting from scratch?

---

 

Most of the "pre-existing" engines were specifically coded for specific games. A person making a FPS would find it hard to use the IS2 or BS engine. They're suited for specific types of games, making them less generic and a LOT of work to convert to something else.

 

The reason 32X Doom (and others) was based on JagDoom was NOT because it was an easy to use "engine" that was easier to make then starting from scratch, it's because the code for JagDoom was a complete and functional total source for DOOM. Having a complete source for a game is different than just having an "engine". The complete source has all of the graphics and all of the detail worked out already and just needs to be tweaked to run on different hardware.

 

You keep perperating the myth that the Jaguar is so buggy and so unstable that it's totally impossible to program for. As was said before, doing a straighforward game on the jaguar would not expose the programmer to any of the bugs or timing issues at all. A game could run entirely on the 68000 in the Jaguar and cause no issues with the hardware.

 

Also, as I mentioned before... the Atari Devkit provides a COMPLETE and DOCUMENTED 3D and Sound Engine and nobody has done a thing with it. Native's Engine is freely available too. Nobody has done anything with that. If things were being done with these engines, I'd be a lot more hopeful for future developments. Remember, the BS and IS2 engines are totally unducomented and have no sample code I know of. A person looking to use one of these engines would be taking a bad approach when there already existed the Atari engines which are documented and have tutorials.

[Thunderbird]

quote:

---

Originally posted by Glenn Saunders:

Even given the Jag's hardware compared to today, I think people (including the developers themselves) expect more to be done than is generally possible as a part time hobby affair.

 

People are not going to get that excited about 2D Jag homebrews.

---

 

If that were the case, then how do you explain the popularity of new 2D games like "Hyper Force", "Soccer Kid", and "Protector"?

 

A small group of collaborators could reproduce games of this quality working part time in a matter of months.

 

2D games seem to be pretty popular still, and are not rocket science to create.

 

The good news is that we might see completed versions of Slam Racer and Asteroids coming from Sinister in the future.

[Songbird]

Yes, Protector SE includes JagFree CD, which supports the Atari standard layout for Jaguar CDROMs. It appears B&C will also be publishing a CD Bypass cart, and BattleSphere Gold (as Doug has indicated) includes support for new CD games as well.

 

I hope we see more CD games in the coming months and years on the Jaguar as a result of all these options! I really do think the main problem we're seeing with so few new Jaguar games written from scratch is simply a matter of lack of free time. If you can only afford 4-8 hours a week on a hobby project, it can easily take a year to complete. If you can afford 10-20 hours a week, then 6 months isn't unreasonable but that's a pretty big commitment to undertake.

[Tyrant]

Are all of these systems interoperable tho? Will a game writen for one system manage to load and run on another? Will game developers have to release different copies of their games, thus adding more work for them and extra confusion for the non-techie jag fans?

 

P.S. Remember that Protector SE and BattleSphere are both very expensive, and lots of people cant afford to get both, lots of people wont even be able to get either. It seems to me that a market like this dosnt benifit from competetion about delivery systems.

 

[ 05-13-2002: Message edited by: Tyrant ]

[Songbird]

All the systems will boot CDs that adhere to the Atari standard layout. So there should be no problems there. The only potential problem is saving game data (high scores, etc.) to cartridge, since there is not consistency among the three options available. So to answer your question, I would expect only one version of each CD game to be released, since it will play on any Jaguar unit which has any bypass cartridge installed.

 

Note that there is at least one published CD game, World Tour Racing, that does not use any saved game data. I don't think VidGrid does, either. So it's actually quite likely that other CD games could be published that do not require saved game data.

[Paolo]

Hi, Carl!

As far as I know, according to the CDs I have, I can tell you this:

VidGrid does save data (hiscores, and player profile)

Braindead 13 SHOULD save, but doesn't work

Blue lighting YES

Highlander YES

Space Ace NO

Dragon's Lair NO

Battlemorph YES

IS2 YES

Hover Strike 2 YES

WTR NO (sad!)

Primal Rage NO (very sad!)

 

Bye.

[Paolo]

quote:

---

Originally posted by Thunderbird:

Native's Engine is freely available too. Nobody has done anything with that.

---

 

T'Bird,

did you had a look at Native engine?

How complete the game is?

Is anybody working on it now?

While playing the demo found on JUGSDD DEMOCD, I die after a few wawes of 'spheres' in the 'tech background'; since I can't take powerups, it is difficult to survive! So I can't even check how many graphics levels are there

 

Paolo

[Thunderbird]

quote:

---

Originally posted by Paolo:

T'Bird,

did you had a look at Native engine?

How complete the game is?

Is anybody working on it now?

While playing the demo found on JUGSDD DEMOCD, I die after a few wawes of 'spheres' in the 'tech background'; since I can't take powerups, it is difficult to survive! So I can't even check how many graphics levels are there

 

Paolo

---

 

 

What you see is what you get.

 

The first level is all that I know of that was ever released.

 

The code is complete enough to present that one level. The engine required to do those graphics is the hard part. Finishing the game would be mostly adding a lot of artwork and much less programming.

 

I am surprised nobody has really made any effort to do this.

[Starcat]

Hi Guys !

 

Yes, there has never been more than that one demo level. There were some more graphics for other levels, but they were never used in the game.

 

I would say that finishing Native is impossible. Because everybody has his own graphic style and level deisgn style, so even if there would be somebody else to work on more levels and graphics it would never be the real Native, but just a game based on the native engine. The real Native could only be done by Duranik.

 

But that is only the graphics/design part.

Another big problem is the lack of memory.

There also can't be sound in the game because there is too little memory left, and also because the DSP is busy all the time decompressing graphics.

Another huge problem would be, how to create levels ??? When you load the level files all you get is a TON of declares, but nobody knows what those declares mean. I guess that even Duranik themselves forgot how they did it because it's already six years ago.

 

I'm not surprised that nobody ever really tried to finish Native.

I mean WHO should do it ? There is nobody.

All the Jag developers I know of have their own projects, are too busy with life and work or lost their interest in the Jag.

 

After looking through the source code and trying to understand how the game works I think that it would still need a lot of changes until it would be a complete game.

It's possible that it can't be done at all.

 

But the biggest problem is that the design files are only little documented and so nobody can create any new levels.

Reaching that high graphic quality would also be a problem I think.

 

To sum it all up, I think a complete Native game on the Jaguar will stay a dream.

 

Regards, Lars.

[Starcat]

Hi Thunderbird !

 

I have no idea who he was.

 

Regards, Lars.

[Thunderbird]

quote:

---

Originally posted by Starcat:

Hi Guys !

I would say that finishing Native is impossible. Because everybody has his own graphic style and level deisgn style, so even if there would be somebody else to work on more levels and graphics it would never be the real Native, but just a game based on the native engine. The real Native could only be done by Duranik.

---

 

I've seen a lot more complex code figured out by a lot fewer people. I think it could be done, just that nobody has the time to try.

 

Someday, perhaps someone will take up the challenge.

[Thunderbird]

quote:

---

Originally posted by Starcat:

Hi Thunderbird !

I have no idea who he was.

---

 

Anyone else out there care to take a guess?

[Atari-Jess]

quote:

---

Originally posted by Tyrant:

P.S. Remember that Protector SE and BattleSphere are both very expensive, and lots of people cant afford to get both, lots of people wont even be able to get either. It seems to me that a market like this dosnt benifit from competetion about delivery systems.[ 05-13-2002: Message edited by: Tyrant ]

---

 

one of my problems right there.

 

anyways,

 

Of course, I am interested in programming

for the jaguar, Its a nice looking system

that I am know should be programmed for by

more than who already does.

 

But, I cant afford much and my programming

skills consist of this:

 

BASIC programming about 3 years of serious

work on it

 

Python programming, i am not all too great

at this one yet.

 

Visual Basic: I am alright with this language

 

C: I lately have been teaching myself this

 

so far, C seems to be my best option for making

a jag game. Therefore, I think i will work

on my C skills and look out for how to use

C for the Jag, If someone could just lead

me to the right direction I would be greatful

as well as comments and thoughts about this.

 

I cant say you will see anything coming out

from me anytime soon or possibly ever, But

I do want to help the Jag.

 

:S

[mos6507]

>>

If that were the case, then how do you explain the popularity of new 2D games like "Hyper Force", "Soccer Kid", and "Protector"?

<<

 

I'm not aware of Songbird's sale figures. Are you aware enough to classify these as "popular"? And aren't these mostly unreleased games that were just polished off and published? That's a lot less time and money required to deliver than a game from scratch.

 

>>

A small group of collaborators could reproduce games of this quality working part time in a matter of months.

<<

 

But would they want to? Why write for the Jag if your games are basically equivalent to what you can do on an Amiga 1200?

 

>>

Asteroids coming from Sinister in the

<<

 

The world does not need yet another asteroids clone. It didn't need it in 1984 with the 7800 and it doesn't need one now. I guess it's an obligatory project, but nothing to get that excited about.

 

I can play Asteroids or Asteroids Deluxe in all their glory as emulations on my PC.

 

If 2D games are going to be written, they really should try to be original.

[Thunderbird]

quote:

---

Originally posted by Glenn Saunders:

I'm not aware of Songbird's sale figures. Are you aware enough to classify these as "popular"?

---

 

People paid high ags for these games, and he hasn't gone out of business, so there must be a market for them. Sure, they're not as "popular" as Halo or anything, but this is not the XBox.

 

quote:

---

Originally posted by Glenn Saunders:

And aren't these mostly unreleased games that were just polished off and published? That's a lot less time and money required to deliver than a game from scratch.

---

 

I think only Protector still exists as source code. The others couldn't even be "polished off" because they were just binaries. So, no work was involved in the majority of them.

 

quote:

---

Originally posted by Glenn Saunders:

But would they want to? Why write for the Jag if your games are basically equivalent to what you can do on an Amiga 1200?

---

 

Why are there people out there making and selling 2600 games? Why even bother to own or talk about Atari? Atari's dead, right? Why bother?

 

quote:

---

Originally posted by Glenn Saunders:

If 2D games are going to be written, they really should try to be original.

---

 

Let's try to get some games written at all, then when there's some momentum, we can start talking about original ones...

[Gunstar]

I'll tell you what I would LOVE to see finished on the Jag...that MarbleMadness demo that's been floating around and into J.U.G.S enabled Jags! I was very disapointed the first time I uploaded this "game" and found out it was just a couple of screens with no game play...I wrote the author at one time and he said that he was going to finish it, but it's been a while...I don't even remember off hand who it was...I'll have to hunt down the site again. Although I'd like to see the graphics improved over what I saw, the levels pretty much looked identical to the original, but I wouldn't think it would be too much trouble to actually enhance them over the original with maybe 256 or 65,000 or so color depth...I think the original arcade game and it's home ports are 16 color...