How long would it take you to make a level of your favourite retro game in Unity?

That would have been real handy for getting the level right, although it is a remake so just as any bad movie remake director/producer I can throw away the original 'classic' stuff that makes it great and mess it up!

Also the idea is to use a classic as a self learning tool. It's amazing how much you can get done when you are not having to worry about what you should do and can just work from an original classic in a genre.

 

Are you actually learning anything though? It seems like you're just doing stuff you already know how to do cause it's fun. Nothing wrong with that but I wouldn't call it an excercise in anything really.

 

Basic Lo Res Texturing and Gun Firing, Dust Strike from miss in Distance.

Some level fixing up and rework needed, the level has two elevators! DOH!

Learning a lot more about the doom game and the detail that went into making it, amazing work for the time.

PS Selfie cam live feed, so shows enemies behind you!

 

It's looking good and coming together quickly. I'm interested in how it plays. Hopefully you'll throw up a web version or at least desktop download.

I know I asked about your decision to use models instead of quads / sprites but I get it. You're doing your version of Doom. A heavily Doom-inspired game completely in 3D. Nothing wrong with that.

I think your thread title was a bit misleading so people had different expectations. As in literally recreating the exact game in Unity. Which there is little point in doing. What you are making can stand on its own and even be expanded as a game for sale. Probably want to change level layout a bit.

 

I actually see more point in that, for learning purposes at least. It'd be an exercise in attention to detail and sticking with that frustrating polishing phase (and not just the presentation but also the exact same controls, enemy behaviour etc). I don't see the point in throwing something together sloppily vaguely resembling a different game and calling it a recreation of it.

 

Freedom planet started off as a sonic mod and became a great sonic game (apparently). But that would actually require arrowx to finish a game.

 

I see a few great benefits to it...

Experience. The more a person develops... and focuses on maximizing speed identifying common patterns, solve bottlenecks in workflow and find other shortcuts... the more efficient of a developer they will be. Also just DOING like he is is beneficial because a person can learn how NOT to do things. Or how best to do things (within the context of each way a person tests). All of this is beneficial to future game projects to one degree or another.

Creating a solid base for a new game inspired by an older game. It seems close enough I think most people can easily see the influence of the earlier game. Doing it this way he can make a new game that can be released as its own for sale even. Obviously he should change some things a bit more to go that route.

I do agree it would be even better to take it to a point of completion. Finishing the gameplay of the level in full with all of the little details, polish, etc. But he may do that. It isn't time in the project to be focusing on things like that yet. Right now he is still getting everything in there period.

That said I do think there is a good chance he'll stop at the point where everything is in there and not polish it up at all. I think he mainly enjoys prototyping and not spending all of the time required on polish. But there is always a tiny possibility this will be the project that sees that change.

I think there is also a good chance he doesn't typically do the polish stage because by the time he reaches that point he is burnt out and needs a break. Because it is obvious the speed he develops these prototypes is above normal. Pushing that hard can only be sustained for just so long.

 

.... I'm thinking about giving it a whirl as well, but if I keep getting distracted at this rate I'll never finish the other project.

Spoiler

 

Just realised that an Elevator is just a vertical door!

Do any of the monsters in doom use elevators, as Unity's navmesh does not support none static navmeshes (hopefully new one in beta will make elevators easy?).

 

Yes, all of them. They don't call them, but all monsters walk over elevators all the time.

 

Emphasizing this a bit further I dug up the postmortem for Pitfall. David Crane mentions that GDDs back then were basically just rough sketches on paper of the gameplay elements that the hardware (mainly the Atari 2600) was capable of supporting with every other idea it wasn't thrown out.

The original Pitfall game only took him 10 to 15 minutes to design but it took about 1,000 hours of programming.

 

That's goddamn terrifying.

Working within your abilities just to make something is always useful, it's a skill in its own right. But if we're talking about prototyping in the interest of meaningful growth, you're shooting yourself in the foot here.

The benefit of recreating something that already exists is that you don't have to make up or design anything, you just buckle yourself in and do the work. It's purposeful neglect of creative skill in favor of learning to make something exactly right, whether that's exactly as has been requested by a client or exactly as you imagined.

If we're interested in the best way to practice, my suggestion would be to challenge yourself either by recreating the most difficult parts of the game verbatim or just gettin' funky wit it in the imagination department.

I'm obviously just thinking aloud here. "Do it how I want it because I said so!"

The postiest of scripts: I think my favorite retro game is Bomberman, my estimate is 2 days to do a semi-competent clone.

 

That's awesome stuff. Never did any programming for an Atari 2600 but I knew about the hardware and how you had to literally draw every scanline worth of the display. Those were the days. Working with the technical challenges like they had. What an awesome time to be a game creator! One of these days I will retire from my job and when I do I think I will seriously spend my days writing games for the Atari 2600, C64 and Amiga.

It doesn't surprise me that Pitfall would take 1,000 hours to create on the Atari 2600. However, on the C64 games (simpler games like this, Space Invaders, etc) often had dev times of a few weeks to maybe 12 weeks from what I've read. But that is because it was just so much easier to work with than an Atari 2600. You had 8 hardware sprites and the character sets could be redefined as tiles and in those early days people generally didn't iterate endlessly on graphics Also the graphics chips took care of all of the display and you only messed with display interrupts to do things like multiplexing to reuse the sprites or change the colors of objects, etc.

1982 C64 games




1984 C64 game


1990 C64 games



Of course, as the years passed the dev times grew longer and longer as the scope of the games and the time spent on graphics increased.

Often I wish I had made some simple games & released them or sent to a publisher back in those early days. One nice thing is with the Indie thing, popularity & renewed appreciation for simpler graphics & games we kind of have a second chance.

And speaking of that stuff I think I will do a little work on the combat piece of my current game.

 

@Ryiah - great video.

Very similar to C64 - most enjoyable programming. I remember this very well:
- "racing the beam" / stable raster interrupt
- counted every CPU cycle
- counted every byte

 

Weeks?

I threw together my first FPS camera system in about 3 hours and I'm not very skilled.

 

Yes, two weeks. Art + sprites + music + polish. Prototyping is easy and is maybe 5% of final result. "FPS camera" system is "prototyping".

 

Ah, I thought you meant weeks for the camera.

@Arowx you should just bust out some action figures for the sprites to put @neginfinity in his place.

And maybe take some pics of your hands for the gun.

 

I'd still have to texture/paint them (see below), and they are discontinued.

 

It's good fun trying to recreate old games like this Doom level in Unity.
What's much more interesting, to me anyway, is to compare how you go about achieving it with how it needed to be done in the original. What different techniques were used, and why.
Otherwise in this case it's basically copying an FPS tutorial.

 

Running over initial time estimate and as graph shows long way from getting it done!

Going to put out the current WIP build soon, just so you can have a laugh at how bad it is!

 

What do you think of it so far!? -> https://arowx.itch.io/doom-retro-remake-level-1

 

Controls are bad. Doom had much more fluid movement, and you couldnt' get stuck moving horizontally.

Doom shotgun sounds and textures are copyrighted and are not in public domain. Logo is trademarked, you can't use it either.

The only part that resembles original doom is texture near the door, but that's about it. It does not look like original and does not feel like original.

 

Yay it works. Still working on it thanks for the great feedback.

Could copyright be OK it it's for learning under fair useage and it's only a tiny fraction of the total work which is also OK for music and video (I think).

 

On topic of the controls, do something about inertia effect. In retro FPS, characters accelerate and stop movement instantly. The built-in CharacterController component is specially made for old-school, not physically based movement, try to use it instead of Rigidbody (though you'll have to code your own gravity). In case of Rigidbody, there are also workarounds to kill the inertia, like setting a bigger Drag value or bringing velocity to zero each frame before you apply movement forces.

Also I found a bug. Running full speed into a wall (noticed it twice after running into the green armor room) sometimes put the player on the ceiling and reverses controls as well https://gyazo.com/907f07500e56d6d694dfc6cd09832188

 

I'm starting to think you've only seen Doom played in videos in an updated engine like ZDoom or the like.

1. First off, you can't look up and down in Doom. At all. Mouselook simply wasn't a technical possibility given how the engine was made.
2. There's no indication that you're being shot by enemies. No knockback, no screen indication, your health just drops.
3. The controls, my GOD the controls. Even a like 20 line character controller codebase would be better than this. Everything is so floaty and you get caught on every possible surface.
4. The level itself is a pale comparison to E1M1. The ceiling is too low and everything is all squatted down.

Advice: Sit down and play Doom. The original, no mouselook or anything like that. Take in the controls, the level itself, and all the nuances of how the game is played. Right now this is like somebody gave you a vague description of Doom and you tried to recreate it from that.

 

As far as I know, it does not fall under fair use.

Basically, what you're doing is ripping game assets and using them into your game. Usually this kind of thing gets cease and desists if it gets IP owner attention.

 

Great feedback, and it's a Remake, you know like Ghostbusters the movie.

 

So much like the new Ghostbusters movie it's a pale comparison to the original that leans too far into nostalgia while having no real identity of its own? Yeah, the comparison seems apt.

 

Dude, when you started the thread I thought that MAYBE you'll try to replicate look and feel of original properly. You know kinda the way Axiom Verge did it in its secret worlds:

^^^ This almost perfectly mimics NES view (aside from insane amount of moving sprites)

If you did THAT, it would be amazing.

However, what you're doing right now - vague approximation of original - is significantly less interesting.

And there's already a remake, released last year.

 

I'm pretty sure in the original you didn't walk around on the ceiling with reversed controls...


It just started like this. I'm not sure what's going on here.

 

LOL what you haven't tried DOOM in inverted mode!


For the fans of retro lo res!

 

There used to be a great app on the app store called 2600 magic, but it doesn't seem to exist anymore. It was sort of an interactive e-book that showed how graphics were programmed on the Atari.

 

I think axiom verge also started off as a metroid mod. So maybe someone could use doom as a good template to make something good

 

Update with shotgun and Imps throwing fireballs very slowly, should they be faster?

Try it here -> https://arowx.itch.io/doom-retro-remake-level-1

Think I've fixed the inversion problem.

And super low resolution to keep the retro doom enthusiasts a bit happier.

Still needs texture work to more closely match the original e.g. Computer room/Final Room

But I think I'm about done with this as it's working OK. Any major missing features or flaws, please let me know.

Thank You and Good Night!

 

1. Yeah make those fireballs faster.
2. Try to find a more suitable music track
3. Add some sound effects for when you get hit, die, when enemies get hit, die. Maybe add a sound effect when the door opens.
4. I dont like the fact I have to stand by the door to keep it opening (and get shot).

​

 

1. Low resolution is not what makes Doom. You have low resolution rendering but you do things like have ultra glossy doors.
2. Enemy AI causes enemies to clump up at the door instead of their initial positions
3. The controller is STILL a nightmare, with the player flying down downward slopes and flying off the edge of upward slopes.
4. Enemies don't leave bodies when they die, nor do they gib. They just vanish and leave their loot on the floor.

 

So long as he does not engage in large scale distribution and is not commercial, it falls under fair use.

You're allowed to draw your favorite super heroes and share your work with friends and potential employers.

Same goes for games.

The only real issue would be if having it posted online to freely download might be considered distribution, if you have really on edge lawyers.

 

Hah!

Nope. You are not allowed. Fanart does not fall under fair use.

http://chrisoatley.com/fan-art/

http://www.blendswap.com/page/fan-art

Enjoy the copyright law.

P.S. Besides, "fanart" and "Ripping game's assets" are two different things, don't you agree?

 

So a summary of @neginfinity's feedback.

No no make it more like DOOM!!
No make it more like DOOM!
It's still not enough like DOOM!
Ha Ha you're in breach of copyright now!

I'm taking the copyright breach issue as positive feedback.

Check out fair use, you can copy other people's work when it's for educational usage. Otherwise art professors would not be able to show their students photos of great art, youtube game commentators would be silenced and we would never find out if Arowx can remake the first level of doom in Unity or how it plays or doesn't.

The point of this thread is How long and can you remake a level of your favourite game in Unity.

Side note if we didn't have fair use would that block adverts?

 

Still got quit a bit to do on this to get closer to the original.

OK This is probably obvious to most of you but when you do an exercise like this one, you start to see game development from a different perspective. Because you're not tirelessly working on your dream game you are working to someone else's game design (or my take on what DOOM is).

Everything interacts, this is the core complexity of game development the exponential growth in complexity from every game element interacting with every other.

Adding one new thing adds n complexity where n is the number of existing things in the game it can interacts with.

Or why just adding one more thing can kill/break a project.

Is that be chaos theory for game developers?

 

Hmm maybe I should have read this DOOM source code review before starting?? -> http://fabiensanglard.net/doomIphone/doomClassicRenderer.php

 

Fair Use does not allow ripping other game's asset and incorporating them into your game.
Youtube commenters fall into "news reporting" category and are allowed. Same applies to education and demonstrating artworks. However, Fair use does not allow fanart, and does not allow making a derivative.

If you're making games, you should be aware of this stuff.
Also see: https://en.wikipedia.org/wiki/Rogers_v._Koons

 

This is a strange challenge in a way. For me the most amount of time spent would be actually going back and researching the original game to make sure I ´degraded´ the graphics and gameplay enough to represent that era. It is kinda like those post-processing camera effects that recreate artifacts from old movie cameras. I mean if your biggest effort is trying to downgrade the graphics, sounds and physics ( or lack thereof)... then in many cases it is actually harder to make a late 80s game than it is to make a late 90s game.

I think the more interesting concept is how fast we can now make improved versions of those games...where you do have better graphics, lighting, sounds but the same gameplay. And like many have pointed out.. Breakout, Asteroids and Space Invaders can all be done easily in a day.. in fact you can follow tutorials where they actually teach newbies how to do it in like 9 hours.

It is just time consuming to replicate someone else´s work. If a person takes two colors of paint and mixes them together and paints a wall, that can be done very fast. But for a second person to start from scratch another wall using the exact same color would take a ton more time.

 

I don't think there is much value in there for creating the game in Unity. There is a lot of value for people doing it lower level as well as for helping people to appreciate the way the original game was created.

Certainly if you're using a lower level rendering approach... not using Unity... or perhaps using unity and actually using sprites to generate the entire display with each sprite being 1 slice for the raycast then the technical info on the rendering and bsp layout could be quite helpful.

I think non-visual code of the original would be more helpful. The code that defines enemy behavior such as delays, movement and code for the player control in the same way could be used to make things more authentic.

Otherwise working at this level it just doesn't apply. You don't need to cast rays to detect wall slices and build the display a slice at a time. With it being 3D and modern engine you just easily do all of that part as far as building and rendering the view.

Focus on the control, feel and gameplay I think is the best way to approach it.

 

How long to make a platformer?

Will it be as good as Mario?

Why not?

A game development professor got his students as an exercise to mimic mario's platforming jump, it's not simple.

Hint -> http://error454.com/2013/10/23/plat...d-the-3-fundamental-equations-of-platformers/

Painting a wall is simple. Finding what makes a game really fun is not simple.

That's the idea behind this to get into a game or style of game and get a feel for what makes it work and how it works.

And game development for even simple games is more like painting a landscape or for AAA development the Sistine Chapel than a wall.

Spoiler: NSFW Nude Art

 

Professor? He is just some dude with a blog, he isn't a professor or a professional developer. Really, he is way over-thinking the whole thing. That stuff is very simple. And bear in mind that the original mario and games of the era weren't using physics or even complex math, it was assembly.

 

You're right. It's not very easy to use Google. People are proving that all the time on this forum.

 

Yeah that is nuts really. Doing a jump is incredibly straightforward and simple. When I went to that page and saw all of those math formulas I think what is wrong with people? lol This stuff is way simpler than the way many people approach it.

It's like fun. Sure there are morons who try to measure out the components with math but that is just stupid IMO. Don't be a dang robot. Be a human. When the game feels fun it will be very obvious. Test different things for the controls and movements. One will feel far better than the others. Some will be too slippery. Some will be too tight. One will be "just right". Keep that one.

 

Attempting to recreate a classic game is an interesting challenge. But that is how I would approach it, as a challenge. I wouldn't want to spend a lot of time on re-creating an old title as opposed to simply creating a new title that is my very own.

I would direct your attention to the FullSail team, and their game Shovel Knight. This is a prime example of a studio that captured the style and appeal of older, classic titles while still making their own game that wasn't tied down by all the restrictions of what they were attempting to pay homage to. Re-creating classic games is for prototyping, not full production.

As an example, I was able to prototype the lost woods from The Legend of Zelda in Unity. It took me less than an hour and a half. But I did not attempt to continue re-constructing the entire game. I could have, but why bother? Also, it's worth noting that part of the reason I was able to do this so quickly is because of my extensive knowledge and experience with that game, as well as having pre-processed all of the needed graphics. I also tackled this challenge as a means of testing a tool that I had been working on. For those purposes it was great.

 

I think that's what he is doing. Only one leve from the game not the entire game.

And yes there is no sense in completely recreating an old game striving for 100% accuracy if the original game is already readily available for play. Other than as a learning or testing exercise. Even in the latter two cases just one level/section should be enough unless the game varies greatly from one level to another.

Oddly although I enjoy many retro games I didn't get into Shovel Knight as much. Think I may have played it for 1 hour.