Tron HD

[lamprey]

Hope this wasn't posted before, but I can across this today..

http://spaceparanoids.net/tron/

Tron HD was created in Microsfot .NET with the XNA framework. The full game is about 4000 lines of C#. I wrote it from scratch, utilizing no plug-ins or existing game engines.

The program was designed by capturing output from MAME running the original arcade software, then analyzing it frame-by-frame to reverse engineer the game logic. Adjustments were made to double the framerate of the gameplay, and the graphics were upressed from 480x512 to 1200x1600--the native resolution of the UXGA LCD display panel in the cabinet.

[wp34]

Nice. Too bad he won't release the code.  I would run that on my cab.

[pbj]

Looks pretty neat.  I can't imagine that spinner hack works all that well.  Wonder why he didn't tap into the optics on the original PCB....

[Vigo]

Looks awesome! Is it just me, though, or does it look like it was made easier than the original?

Too bad he won't release the code.

"Hey Sjaak...wanna take on another major project?"

[wp34]

Looks awesome! Is it just me, though, or does it look like it was made easier than the original?

Yeah I thought that as well.  That might not be a bad thing though.   

[nitrogen_widget]

Looks awesome! Is it just me, though, or does it look like it was made easier than the original?

Too bad he won't release the code.

"Hey Sjaak...wanna take on another major project?"

Probably because he just wrote a Tron game and didn't actually reverse engineer game logic from video captures?

I mean how many hrs of recorded game play would you need in order to figure out the AI?
How many hrs of analyze time would it take a computer to do?

Can someone even do this?

Not trying to be a jerk or anything but to me it seems a little far fetched.

[lamprey]

Nice. Too bad he won't release the code.  I would run that on my cab.

No prob.. I'll grab some of my guns and cash in some air miles and see what happens.. j/K

yeah, it would be cool to be able to run it on a cab.

[IAmDotorg]

Looks awesome! Is it just me, though, or does it look like it was made easier than the original?

Probably because he just wrote a Tron game and didn't actually reverse engineer game logic from video captures?

I mean how many hrs of recorded game play would you need in order to figure out the AI?
How many hrs of analyze time would it take a computer to do?

Can someone even do this?

Not trying to be a jerk or anything but to me it seems a little far fetched.

I could be done... keep in mind its a simple little game written for a Z80 released in 1982. Its AI isn't going to be any more sophisticated than, say, Pac Man. As someone who was writing games on equally limited hardware at the time, I can tell you there was a lot of tricks we used to make "AI" look smarter than it was. (Like how Pac Man basically runs the identical logic for all the ghosts, but uses four different target locations relative to the player to make them have "personalities" -- but its a handful of lines of code to do it all!)

I haven't looked that closely at the Tron logic, even though I've spent countless hours playing it, as its one of my favorites from that era, but the logic does seem fairly simplistic. Going frame-by-frame would help you figure out things like the timing and grid granularity it uses for tracking the player. It might be hard to figure out details like when the tanks fire and stuff, but you could get it pretty close.

IMO, its not the way I'd go about doing it -- the ROMs are in MAME, and they use the mcr driver. I'd just grab one of the many Z80 disassemblers and look at what the code is doing. It'd take less time. The only downside, because its not a "clean" reverse engineering, I couldn't legally distribute it.

But doing it the way he claims to have done it seems perfectly possible, and perfectly valid, if you weren't comfortable or familiar with working with old 8-bit assembly.

[ark_ader]

It is not that hard, but time consuming.  I would take that project on if it wasn't for my current workload.  Just get a book on C# and XNA and you will be half way there.  This is BYOAC after all....

[Vigo]

I have to disagree that AI would be fully accurate using video tape recordings.

Yes, you could probably come up with something close, but it wasn't until a few years ago that the actual Pac-Man ghost AI was dug out of the code and revealed and still competitive players debated about it being correct. Being that Pac-man is probably the #1 game of all time that was studied for winning strategy, you would think that someone would have it all down before 2008 if all it took was video study. I am sure you can make assumptions that might be close or you might get lucky and guess it right, but if there is "random" commands in the mix, I doubt there would be a way to know for sure if the code is completely accurate.

http://www.twingalaxies.com/showthread.php/119946-Pac-Man-ghost-behavior-revealed

[eds1275]

I didn't bother to watch the video, but what purpose did it serve to create this if he isn't releasing it?

[yotsuya]

I didn't bother to watch the video, but what purpose did it serve to create this if he isn't releasing it?

"Lookie here what I got! Jealous?"

[bulbousbeard]

I really don't understand the point of this. Why wouldn't you just play it in MAME? It's not like he actually REDREW the art, right? It's just upscaled. You could run MAME through a filter, too, so who cares?

Seems like a waste of time, but whatever.

[Xiaou2]

I didn't bother to watch the video, but what purpose did it serve to create this if he isn't releasing it?

"Lookie here what I got! Jealous?"

 Hes a guy who works in a commercial industry, and has a family.   He is probably scared of getting into legal troubles.

 

I really don't understand the point of this. Why wouldn't you just play it in MAME? It's not like he actually REDREW the art, right? It's just upscaled. You could run MAME through a filter, too, so who cares?

Seems like a waste of time, but whatever.

 You didnt watch the video, obviously.. which makes you look like a fool for your comments.

 He made a custom version of the game, due to his PCB being broken.

 The graphics are not merely scaled.   Hes added and changed things.   The closeups show some really cool added details. Tasteful changes, that fit the rest of the graphics and style.


 I really dig the remake, in that it stays true to the real gameplay and look.   Its not a FPS, or a completely different game.
There are some quips I have about it...  as I think he removed the limitation of bullets on screen at once.   I see some other differences, but Ill refrain from pointing them out until I verify.  I never got too far into the real game.. mostly because of that darn LightCycle stage.  The Cycle stage was far more random than skill based.  I know there are patterns you can memorize.. but I have a pretty bad memory.

 

[Xiaou2]

[SavannahLion]

Ah jeez.... I think I've run into him in RL. 

[Le Chuck]

I totally agree with X2 for once. Wowzers. Well not totally. Yot and X2 are both right. Yes he's scared of getting sued but yes he wanted the attention. Nothing wrong with that. If we didn't want some attention we'd never post our projects.

I think a black bag operation might be in order to go and um rescue that software. Yeah. Rescue.

If it gets out due to theft he can't be sued. We need a mole.

[bulbousbeard]

Bleh, the original art looks better. You really can't go in and update classic game art. Radically different resolutions require radically different designs.

[Vigo]

  I have to agree with X2. The art changes were incredibly tasteful.

[Leapinleny]

He should call his game Tron Legacy after the new movie 

[IAmDotorg]

It is not that hard, but time consuming.  I would take that project on if it wasn't for my current workload.  Just get a book on C# and XNA and you will be half way there.  This is BYOAC after all....

XNA is basically a dead platform. If someone's going to try to take on this kind of project, C# and Unity is a better bet.

A game like this would be a weekend project with some Unity experience, 2/3 of it swearing at your computer about how terrible the UI components are and debating if you want to just leave the score off.

Literally a couple dozen textures and a few hundred lines of code to wire it all up.

I threw together a 4 player modern take on Asteroids for my cabinet in a couple evenings, but that was more complicated because it doesn't have a fixed play space. Now if only I had the cabinet together to pay it on!

A Tron remake using a top-down orthographic view and full 3D models would be *awesome* and equally simple from a coding standpoint. But beyond my modeling skills.

[Le Chuck]

It is not that hard, but time consuming.  I would take that project on if it wasn't for my current workload.  Just get a book on C# and XNA and you will be half way there.  This is BYOAC after all....

XNA is basically a dead platform. If someone's going to try to take on this kind of project, C# and Unity is a better bet.

A game like this would be a weekend project with some Unity experience, 2/3 of it swearing at your computer about how terrible the UI components are and debating if you want to just leave the score off.

Literally a couple dozen textures and a few hundred lines of code to wire it all up.

I threw together a 4 player modern take on Asteroids for my cabinet in a couple evenings, but that was more complicated because it doesn't have a fixed play space. Now if only I had the cabinet together to pay it on!

A Tron remake using a top-down orthographic view and full 3D models would be *awesome* and equally simple from a coding standpoint. But beyond my modeling skills.

So... what I'm basically seeing here is you volunteering to make a TRON clone in a weekend using Unity and we'll all test it.  I'll be set to test on monday

[IAmDotorg]

So... what I'm basically seeing here is you volunteering to make a TRON clone in a weekend using Unity and we'll all test it.  I'll be set to test on monday

Unfortunately I'm spending most of the weekend swearing at Unity for how lousy their GUI capabilities are for an entirely different game. That one isn't currently designed to work on an arcade cabinet / controls, but when its done I may do a special version of it for mine. Its intended to be an ID@Xbox game, and maybe a PC game, but if I end up selling it I may look into selling a special cabinet-compatible version.

Seriously, though, its pretty easy to pick up. I did a Pac Man clone as my first experiment of doing a 2D game in it, and for those simple early 80's type games, its really pretty easy to throw them together because the original game logic is so simple.

Why do you think there's so many total junk Android/iPhone games out there? Unity makes it simple to crank them out over and over and over.

[Vigo]

Less time swearing at unity. More time making us Tron clones. 

[wp34]

I watched the video again this time with the sound on.  He has an LCD mounted in the back so that the MCP translite is animated.  That is very cool.  It is so subtle I missed it the first time.   

There are some quips I have about it...  as I think he removed the limitation of bullets on screen at once.   

It looks like you can fire many more bullets at once which makes it quite a bit different game---particularly on the later tank levels.   This also contributes to the game appearing to be much easier than the original.  It reminds me a little of the "remix" titles on the Wii U.  I'd love to get my hands on this game.

[Xiaou2]

So... what I'm basically seeing here is you volunteering to make a TRON clone in a weekend using Unity and we'll all test it.  I'll be set to test on monday

Unfortunately I'm spending most of the weekend swearing at Unity for how lousy their GUI capabilities are for an entirely different game. That one isn't currently designed to work on an arcade cabinet / controls, but when its done I may do a special version of it for mine. Its intended to be an ID@Xbox game, and maybe a PC game, but if I end up selling it I may look into selling a special cabinet-compatible version.

Seriously, though, its pretty easy to pick up. I did a Pac Man clone as my first experiment of doing a 2D game in it, and for those simple early 80's type games, its really pretty easy to throw them together because the original game logic is so simple.

Why do you think there's so many total junk Android/iPhone games out there? Unity makes it simple to crank them out over and over and over.

 
 Im curious to see what you are capable of.

 On the other hand...  I have a feeling you are underestimating things.

 Yes.. there are a LOT of Clone games out there...  but how many people prefer them to the Originals?   How many people, especially those less likely to be age-biased (younger people tend to say new stuff is always better)..  actually Rave that the remake outshines the original by Miles  ?

 To date, Ive played Countless clones that looked interesting visually...  only to be completely disgusted with the gameplay.    Poor difficulty (almost always WAY too easy).  Poor difficulty ramping  (If you dont mind falling asleep through 30 levels before the game has a pulse).   Overall, Poor game balance.   Artificial Intelligence routines are are 'way' off the mark..  too easy too fool...  buggy...   worthless.     Poor use of mathematical formulas..  so things like a Mario Bros jump.. is absolutely nothing like the original.   Additional 'features'  like  Powerups that are too powerful.  Things that make the game look, play, feel, and even sound...worse.    Horrible low-poly models, that are shaded very badly.  Wretched animations that look very artificial and 'wrong'.  Bad use of colors..  and the typical "eye burning flash explosions.. that stay onscreen for about 200x longer than they should... and totally mess up the ability to SEE whats happening.   Reduction of projectile speeds..  making the screen littered with dog slow bullet fire... without any adrenalized reaction needed or felt, to speak of...    And things like 3d FPS versions of 2d games, that do not play anything like the original games... to the point where it cant even really be called a clone.  As well as the typical flippy camera angles / camera in the way.   Artwork thats actually drawn with a worse 'flat' look...  not using proper artistic shading with good Contrast in the shades  (see Original Rayman -vs- new Rayman)...
And lastly.. remakes that Butcher the Controls.   Like Midways remakes that have Spy Hunter in them..   Theres no way you can play that game on a gamepad!  lol    Games like Supersprint..  never play anything like supersprint..  and part of its probably due to the fact that none of them have a spinner... and dont design their games to properly support one.   No trackball?  Use a joystick to play Marble Madness..   Uhh.. No.   Not ever worth wasting your time.

 Asterioids remakes?   They almost always look like garbage, with cut and pasted graphics that dont match well..  dont animate well.. dont break into pieces well.  IMO, This also includes Atari's own remake, blasteroids...  which is clunky at best..  and not really anywhere near as fun as Asteroids Deluxe in intense gameplay.    Often the thrust of this game is reproduced very poorly.. and so its a lot less about to be controller as smooth and well as the original.    And out of all of the remakes, I dont think Ive found any to be even close in approximate value.  None of the added gimmicks have ever worked... and have only made the games worse.


 Now, lets be honest though...
 As much as I loved Tron,  the original movie,  as well as have love for the Cabinet and Game...   Tron isnt really a shining example of a "great" game...

 
 - Lightcycles:   As cool as they look in the movie...  the game version is infuriating.  Its almost completely random that you will survive it... unless you memorize some patterns... and to me, thats not good at all.
 - Grid Bugs:  The concept is cool.. but the bugs are too slow, and your fire rate is too slow.   The difficulty is pretty stable, because either you have time to shoot them... or your outnumbered and its impossible.
 - Tanks: Love the Tanks. My 2nd Favorite.  But... it gets way too unfair with the shear number of tanks, and your limited ability to hide from them all.
 - MCP: Love this one.  It can get really tough in the higher levels... but its possible to accomplish... and this challenge actually adapts in difficulty far better than the others.

 But... in the sum total,  even with its money eating ploys and its quirks..  its still quite good in its hardcore-challenging fun.   

 The sounds are epic. 
  Its got really cool interpretive art / design... and cool use of color cycling.

 The cabinets design, is really great... with its use of Blacklight glowing Trigger Stick, and ultraviolet glowing lines works on the Artwork.  The Sound reverberates so well
in that cabinet.. and off the sidewalls.  Really draws you in, and keeps you focused on the tasks at hand.

  It captures the spirit and soul of the Tron experience.   I dont think any other Tron game has come close to doing that so completely.


 Discs of Tron was cool too...  but it suffers from a lack of intensity and variety.   
 Admittedly, the Env. Cabinet is the most insanely awesome  cabinet, ever produced  (non-moving).
The discs fly a bit too slow...  and theres times when you can volley discs for far too long.
The added Sark powers help liven things up a bit..  but it kinda cheapens the overall experience you want to have.   Sark is pretty good with his AI in the later
levels..  but its almost like most of the time... either you are getting him on accident, or because the game is 'giving' it to you.  Its ok for its day.. but still a bit too clunky.

 Now, the game is pretty hard.  But it never feels like much fun.   The up and down section didnt help either.  Seemed to make an ok game even worse.
I do love dual control games..  and the spinner is unique and cool... but sadly, the games doesnt really make it fun to use.  Unlike Mad Planets.. which is a
completely blast, start to finish, without the additional axis.  heh

 I liked the wave with the scrolling blocks.  But, it was too difficult to get through at times, and made it less fun than if it were eased up and
allowing more stuff to pass, only to get a random block to deflect that perfectly timed and aligned shot.

 I do often wonder if a Two Player hack would make the game much more fun.  One player would hack into Sark..  but
all of Sarks extra powers stripped, in that mode.   Add faster discs, maybe more of them too... and faster character movements.

 Maybe hack together a way to link two games together as well.  4 player would be interesting    (two dual hack cabs linked)
(probably have to reduce the projectile count to 1, on 4p.. but just move it much faster   heh)

 A StereoScopic 3d hack/version  would also really add to that experience vastly.  Able to increase difficulty due to better perception
of spatial depth.

 
 Actually, Ive got an even better design in the works..  for the original Tron..  (Ive almost finished the the Disc battle section,
and its leaps and bounds in gameplay, over the DOT version.  Top-view / and or in the typical arcade Tron Style Perspective)   
but Id really need a capable team behind it.
Could be quite awesome...

 Maybe he would be interested in an expansion co-op project  ...   

[Le Chuck]

You know what I'd like to see?  DOT put into TRON as the fifth level it was originally designed to be.  Play the four others and the fifth sequence before difficulty increase is a DOT level.  We've got all kinds of people here saying that how easy it would be in Unity to do one or the other, somebody do an update of both. 

I'd recommend rotating TRON to a horizontal orientation as most the the levels wouldn't have a gameplay impact in doing so (MPC Cone withstanding). 

[RandyT]

I had a look at the Unity stuff.  Anyone who thinks they could crank out TRON in a couple of days is delusional.  Sure, it's a rapid game development platform, but just creating the visual assets would take longer than that if quality was of any interest.  Asteroids isn't TRON.  I can see how simple a clone of this game would be for someone who knew what they were doing in Unity.

I think the TRON remake is pretty amazing, and the author did a great job.  I'd love to see it released.

[IAmDotorg]

Actually the assets are pretty simple -- there's no that many in the original game, and most are pretty simplistic. It'd take a couple hours in Illustrator to create high-resolution versions of all the original tiles, and even then only if you're working pretty slowly.   

Asteroids is pretty simple if you keep it 2D. A couple dozen lines of code total for the rock controls, a gameobject and pool of them for the player shots. They'd only need a few lines of code total. One for the player. One for each of the "bad guy" ships -- although I think the original AI for them was the same, only the speed changed, so it'd really be two meshes and one set of code.

Four player Asteroids rendered fully in 3D with dynamic lighting, particle system explosions, a larger-than-screen playfield with dynamic camera is a good bit more work.

As someone who has 35 years of software development experience, written quite a few games back in the 80's when you had to work in 8k ROM and code in 6502 assembly, and written quite a bit in Unity, I'd say anyone who thought they *couldn't* is delusional, but that's not really true. They're just inexperienced. Nothing wrong with that, though.

If you want to get a sense of how little work it'd be, just grab screen shots of the four game states from Google Images. Load them into a layer, and trace over them in Illustrator or using paths in Photoshop. You can "high-rez" them *very* quickly. There's only a few different tiles for the grid bugs, get a shot with a couple dozen on screen and you'll definitely have all of them. The game logo itself people have redone high-rez for cabinet art, grab that. The background of the light cycles is just a fade with a grid over it, it'd take you maybe a half hour to replicate it looking very nice. The tank maze is just built up of a few tiles, they'll be quick, too.

Create a Unity 2D project. You don't need any physics, so its just a matter of creating a few game objects, tossing the textures onto them and starting to code it up.

In fact the more I think through it, I think you could get two people in a game jam and replicate it in a couple hours. One doing art, one doing the coding. The start screens would probably be the most work, using NGui or something, but if you were willing to use a different typeface and could find a pre-baked asset (one of the SciFi GUI ones in the asset store) with something close enough, they wouldn't be too bad.

Its four game play types, but much like Journey and other games with multiple game types, they're each very simplistic.

In fact, if you look at the video, the player and enemy sprites actually don't look updated anyway. I think he just used the original art over new backgrounds, and added some effects on top like the explosions, which would take even more work off it.

[pbj]

Less forum navel gazing, more game coding kthx.

 

[RandyT]

As someone who has 35 years of software development experience, written quite a few games back in the 80's when you had to work in 8k ROM and code in 6502 assembly, and written quite a bit in Unity, I'd say anyone who thought they *couldn't* is delusional, but that's not really true. They're just inexperienced. Nothing wrong with that, though.

Been there, done that.  The tools are infinitely better nowadays, and after seeing Unity, I really feel tempted to play with it.  But again, if you weren't just interested in slapping stuff together by using screen captures and scaling/filtering in PS, it would take a fair bit of time.  Getting all the logic down for each level so that the game plays like the original would also be quite a challenge. 

But as PBJ stated, let's see what a couple of days results in.  I'll be a believer then.

[IAmDotorg]

Right now I'm actually focused mostly on a second version of the Asteroids game (mostly because I already had the assets and physics done for the rocks). This one is for the Oculus Rift, though... still trying to get the framerate down enough to not feel like I'm going to vomit while I play it. Twin-stick, fully immersive VR Asteroids! Its awesome... for a few seconds!

Unity is a nice enough framework, but the APIs are *terrible* in it. It makes it very easy to write very bad code. But for banging out quick-and-dirty mobile games or things like that, its fine. But use it enough and you start to understand why there aren't many AAA games with it, unlike Unreal.

Edit: framerate up enough, not down enough...