The Perfect Blend...

[Dullstar]

Eh, it's a bit of an involved field. A proficient programmer not only knows how to instruct a computer to do certain things, but also understands how those things happen on the internal circuitry and knows how to make the necessary considerations accordingly. For the time being, I recommend you find some programming tutorials via your favorite search engine. Once you're comfortable making things that are functional, I can help you dig into the deeper stuff and get really good at it.

BASIC and its variants are by far the easiest to learn if you've never programmed before. C and C++ are nearly universally supported, so if you learn one of them, you'll be able to write programs for virtually any computer. What I suggest is finding an old copy of Microsoft QBASIC and learning some of the fundamentals to programming through that, then move to C to see how most of the world does things. You can dip into C++ if you want, but be advised that there are many features in that language that are geared towards productivity at the loss of performance--which results in programmers not really knowing what the computer does when they tell it to do things--so I tend to avoid it like the plague.

Well, I'll try and learn either C, or, if I can figure out what it is, the one used in Cave Story, because if I can find a way to look at the Cave Story script, I could change things a bit and see what happens to learn what effects what.

[Dullstar]

What would I love to see in a new Lemmings game...

The original 8 skills, and no more. Oversaturation killed Lemmings 2. Having ONLY 8 skills meant you had defined constraints, the same for each level. You could go down, across, diagonally, up, or dead in your tracks. What more do you need? Anything else just makes hero Lemmings easier to come by, and the need to think becomes absent.

Quirky themes. Crystal, Hell, Earth, Pink (aka Roman), Yellow (aka Egypt)... They're all rather surreal, except for maybe Yellow. I mean, little men running through the pits of hell and jumping into a giant pig... that's just brilliant. Waltzing around an ice-palace outlined with giant crystals the size of blue whales? Great! ... It fits Lemmings. Changing that prospect causes a failure to maintain an aspect of the spirit of Lemmings.

Lemmings need to die. If there's a need to save them all, then... no. There needs to be bombers, they need to die when they fall from a height, they need to drown, they need to burn. Otherwise, what are we saving them from? ... Just the clock, it seems.

That's all.

Maybe that's holding too close to the original? Well... it is. The original is the best, and say what you like, the sales said it all. The multitude of ports say it all. The fangames which only recreate the original say it all. The forums that are dedicated to all things Lemmings but 99% of the time only touches on the originals or Xmas or Oh No say it all. The fansites on the original say it all.

Don't mess with success. (Or else you get a PS3...)

Hey, I didn't say ALL lemmings would be able to do that, I just said there would be certain special (but rare) types that COULD!

[Clam]

Maybe it's a little early to be thinking about fun little extras, but this thought just crossed my mind. What if you could customize your lemmings? LittleBigPlanet has huge scope for customization (I don't have the game, but I've heard a lot about it), and other games do as well to a lesser extent, and I think this character design concept could be transferred very well to Lemmings.

If the game has a strong multiplayer aspect - which I doubt this new project could really do without - then this becomes an even better idea. You'd have a good incentive to play the singleplayer levels, and maybe even make some levels of your own, rather than jumping straight into multiplayer. And of course there'd be rewards for doing well in multiplayer too .

What do you think?

[Dullstar]

I don't know about what the developer thinks, but it's a good idea to me!

[Clam]

I should add that this idea could be extended to trapdoors and exits too. I think it would add a lot of flavor to multiplayer when you (and your opponent) control your own little tribe that you design yourself, rather than the generic green and blue guys.

[GuyPerfect]

From a technical standpoint, the method I've devised is to draw the Lemming animation cells by hand (whoever the super-ultra-talented artist happens to be), scan them into the computer, then trace over them digitally to produce vector graphics. On game load-up, the vector graphics are rendered into texture data, which allows them to be scaled indefinitely in either direction.

In addition to this fun hand-drawn stuff, there will be multiple layers of texturing (probably two or three) for each Lemming, allowing the shirt/hair to change color depending on what the Lemming is doing. The reason for the multiple layers is to enable color changes without preparing an entire set of animation cells (which will be quite plentiful) for every possible combination of colors.

Under these conditions, the potential for customizability is somewhat limited. But since the Lemmings will be 2D models of sorts, I suppose it will be possible to introduce optional "costume pieces" to overlay on top of them when the texture data is rendered.

[Clam]

How big do you plan to make the lemming sprites? I guess the visual effect of custom items is going to depend quite heavily on this.

In addition, I was thinking about some sort of "achievements" system. These are very common in recent games, and would tie in very well with customization (each achievement earned opens up a new costume piece or whatever), as well as (hopefully) being easy to implement.

I wonder also if we could introduce a scoring system for single player levels, to encourage players to find different solutions. This would be tricky though - it has been tried on these forums before, without much success.
(EDIT) Link to the thread where a scoring system was discussed. Personally, I think they tried to make this way too complicated, right from the start.

[GuyPerfect]

How big do you plan to make the lemming sprites? I guess the visual effect of custom items is going to depend quite heavily on this.

In the event you're asking about relative size compared to the screen, I figure the Lemmings 3 scale is a good reference, though they may end up a bit bigger or smaller depending on what looks and feels best. Otherwise, if you're asking about the pixel dimensions of the Lemming sprites, read my previous post and get back to me.

In addition, I was thinking about some sort of "achievements" system. These are very common in recent games, and would tie in very well with customization (each achievement earned opens up a new costume piece or whatever), as well as (hopefully) being easy to implement.

I'd already planned to unlock certain things like custom level elements and background musics. Achievements would be a good way to go about doing this.

I wonder also if we could introduce a scoring system for single player levels, to encourage players to find different solutions. This would be tricky though - it has been tried on these forums before, without much success.
(EDIT) Link to the thread where a scoring system was discussed. Personally, I think they tried to make this way too complicated, right from the start.

I've considered various methods for scoring level completions in single player. My ideas are kinda complicated as well, but they're very functional. This is what I'm considering:

Every level has a number of given Lemmings, a required number of Lemmings to save, and a maximum possible number of Lemmings that can be saved.  During level playtesting before release, the testers can devise a Max% solution for each level to determine what the maximum possible count happens to be.

For example, imagine a level where you get 35 Lemmings and the intended solution involves sacrificing 3 of them. This makes the intended number of saved Lemmings 32, though to give the player some slack to make mistakes or what-not, let's say that the required number of saved Lemmings is 30. However, the playtesters found a solution that sacrifices only one Lemming, which makes the maximum number of saved Lemmings 34.

Three scores are calculated to determine the value of your solution: Completion, Skills and Time. These need to be kept separate, as there is no way to elegantly combine them. Solutions are sorted by order of Completion > Skills > Time. If you save more Lemmings than someone else, your solution is ranked higher. If you save the same number of Lemmings but you used fewer skills, your solution is ranked higher. Lastly, if you both saved the same number of Lemmings with the same number of skills, but you did it faster, your solution is ranked higher.

The Completion score is calculated linearly between the required number of Lemmings and the maximum number of Lemmings; up to 100 points for max saved. Completion = RoundDown((Saved - Required) / (Max - Required) * 100)

The Skills score is calculated by summing all remaining available skills. If you have 2 Builders, 1 Blocker and 3 Runners left at the end, you get 6 points for Skills.

The Time score is simply the amount of time remaining on the clock.

So saving 33 Lemmings in our example (yielding a Completion of 75) with 7 skills left and 4:38 on the clock, you could notate your solution as 75-7-4:38, which would then be inserted into the leader board.

[Dullstar]

About the achievements:  I'm kinda neutral about that.  Would that be a good idea or not?

[GuyPerfect]

I'm thinkin' the Wario Land: Shake It! style of achievements. Imagine every stage having its own set of achievements that apply only to that stage. "Complete the stage without using any Blockers" or "Save all Lemmings" or what-not. The Kirby Air Ride/Super Smash Bros. Brawl style of checklist would go good as well, as a master set of challenges accessible from the main menu that become unlocked as you accomplish them anywhere in the game. "Complete all standard difficulty stages" and the like.

They certainly don't take away from gameplay or even get in the way, so I see no reason not to put them in. They give incentive to replay completed levels, at the very least.

[Clam]

Yeah, sorry, I meant the size relative to the screen (obviously their absolute dimensions can be anything you want).


Re scoring (wall of text coming up - skip to the summary if you're too lazy to read):
I had an idea that instead of basing your score for a level on a single playthrough (and therefore having to weigh up lemmings vs skills vs time, which proved to be near impossible), you could have separate records for lemmings saved, skills remaining, and time remaining, and combine these to give you a score for the level. This would make things simpler for the player (since they don't have to work out what's more valuable on each level), and encourage them to play the levels again and look for different solutions. The "lemmings > skills > time" system makes speedrun-type solutions practically worthless, and I think such solutions should be encouraged even if they save fewer lemmings or use more skills.

Something else that I disagree with, which you just mentioned and which was discussed at length in the linked thread, is the idea of a "max %" score on each level, on which part of the score is based. We still can't be 100% sure what the max % scores are for the original levels - it's possible that someone could find a new bug in the game that lets you save an extra lemming in one level, or something similar. Now, a new game built from scratch is gonna have its own list of bugs and exploits, some of which won't be picked up before release, and so when someone later discovers a new trick, your so-called "max" won't be a max any more. Instead, I suggest we start with the minimum requirement and give bonus points for lemmings saved beyond that.

But wait, won't this inflate the scores for the Fun and Tricky levels where you could sometimes save 50 or 60 more than needed? Not necessarily - you could scale the bonus points based on the difficulty rating of the level. An extra lemming saved in Mayhem is worth much, much more than an extra saved in Fun, and the score should reflect this.

What about the ridiculous amounts of surplus skills and time that some levels had? *cough* Tricky 4 *cough* Well, since we'll be making the levels, we can avoid this problem altogether - I'd suggest that any such levels that we make should go into the easiest rating. If we were to create a system for the original game where you get bonus points for skills and time, we'd have little choice but to make these bonuses zero for Fun and Tricky - anything else would be ludicrous. What we end up with here will depend on how far overboard you want to go with skill/time availability on the easier levels.

In addition to all this, there could be a bonus "challenge" on each level, in the same vein as the ones in the challenge threads on this forum (as well as the old one - you can find a link to that at the start of my challenge thread (see sig) if you're interested). These would require solutions that wouldn't set a %/skills/time record, but are difficult nonetheless, eg. "builders only" or "no builders".


SUMMARY
So, my proposed system would work something like this. I won't put numerical values on it now because we don't have any levels to base it on yet (though I'll have a go at scoring the original levels if you like :undecided:).

- All components of the score for each level would scale with the difficulty rating, with separate scales for each component (you could easily draw up a table for this).

- The majority of your score would be based on passing levels and progressing through the game.

- For each level you have three records (lemmings, skills, time), and you break these records individually - NOT all in one playthrough with one solution.

- Over the whole game, these three categories have a roughly equal amount of bonus points available (or perhaps more for lemmings saved).

- Each level has a challenge (maybe the challenge goals could be revealed after a difficulty rating is cleared) which gives you extra points.

- Instead of using division and other more complicated mathematical functions to calculate the score (these plagued the other thread, and also crept into your post), you end up with a formula based purely on addition and multiplication, so you can display a nice little animation on the level completion screen that shows the score calculation, and the player can follow this easily.

- You get a large bonus (perhaps half of the total of the "level completion" bonuses) for completing a difficulty rating. Keep in mind that you could use Revolution-style level unlocking, and (for example) unlock some Mayhem levels before you complete (or even fully unlock) the Taxing ones.


With all this considered, you should be able to compare players based on:
- their records within a level
- their overall score for each level
- their total score for each rating, and
- their total score for the whole game.

So, to get the highest score, you have to be a well-rounded player, and an expert at ALL parts of the game.

Please tell me if you disagree with any of this, or if you have any ideas on how to improve it.

[GuyPerfect]

I had an idea that instead of basing your score for a level on a single playthrough (and therefore having to weigh up lemmings vs skills vs time, which proved to be near impossible), you could have separate records for lemmings saved, skills remaining, and time remaining, and combine these to give you a score for the level.

Reconsidering things, I agree that each record should be seperate, though I'm still against trying to combine all three values into a single score.

Something else that I disagree with, which you just mentioned and which was discussed at length in the linked thread, is the idea of a "max %" score on each level, on which part of the score is based. We still can't be 100% sure what the max % scores are for the original levels - it's possible that someone could find a new bug in the game that lets you save an extra lemming in one level, or something similar. Now, a new game built from scratch is gonna have its own list of bugs and exploits, some of which won't be picked up before release, and so when someone later discovers a new trick, your so-called "max" won't be a max any more.

The response here is that if you manage to save more than the determined maximum number of Lemmings, the game will prompt you to submit your solution to the devs to see if the level meta data needs to be adjusted or if the game program needs to have a bug fixed. We can do this in the world of internet distribution.

Also, just so you are aware, I'm one of those guys who knows so much about what my programs do that if I declare something "bug-free," it will most likely be bug-free. Many people before you have interperated this sentiment as arrogance, but you have my fullest assurance that I won't release any software if I suspect there's any potential for a bug in it. Exploits, on the other hand, need to be addressed on a case-for-case basis.

In addition to all this, there could be a bonus "challenge" on each level, in the same vein as the ones in the challenge threads on this forum (as well as the old one - you can find a link to that at the start of my challenge thread (see sig) if you're interested). These would require solutions that wouldn't set a %/skills/time record, but are difficult nonetheless, eg. "builders only" or "no builders".

Just one bonus challenge per level? )-:

- Instead of using division and other more complicated mathematical functions to calculate the score (these plagued the other thread, and also crept into your post), you end up with a formula based purely on addition and multiplication, so you can display a nice little animation on the level completion screen that shows the score calculation, and the player can follow this easily.

Nothing wrong with division; it was in fact put into my post intentionally. I was transferring 100% to a Max% formula, and "percent" explicitly defines a process that involves division.

Though looking back on it, I figure that if I want to go for Max% based on player solutions, it'd only be fair to get a maximum Skills score and a maximum Time remaining as well, which certainly are pushing it overboard a bit. I'll reconsider scoring to enumerate Lemmings directly instead of transforming the number saved into some other value.

[Clam]

Reconsidering things, I agree that each record should be seperate, though I'm still against trying to combine all three values into a single score.

Why? I think there is a worthy challenge in speedrunning and economy of skills, even if it saves fewer lemmings than can be achieved normally (provided that it meets the minimum requirement of course ). What I am going for here is a single (large) number that reflects the player's overall level of skill and commitment to the game.

The response here is that if you manage to save more than the determined maximum number of Lemmings, the game will prompt you to submit your solution to the devs to see if the level meta data needs to be adjusted or if the game program needs to have a bug fixed. We can do this in the world of internet distribution.

You could do the same if the game finds that your solution ranks outright 1st.

Just one bonus challenge per level? )-:

OK, that was a little short-sighted. There's plenty of room for flexibility here. Some levels might be no good for challenges (eg. All or Nothing), while some might have two, three or even more challenge-worthy goals. Of course we can release more later if we find them.

Nothing wrong with division; it was in fact put into my post intentionally. I was transferring 100% to a Max% formula, and "percent" explicitly defines a process that involves division.

Oops, I managed to miss that somehow. Though I'm in favor of ditching the percentage system altogether since it can get complicated depending on the lemming count.

[Dullstar]

Achievement Ideas:

SINGLE PLAYER
(Name ideas included, not needed to be included).

Backroutes Galore! - Beat at least ten levels by rescuing more lemmings than possible with the intended solution.

"F" Student - Fail thirty times WITHOUT NUKING THE LEVEL.

"A" Student - Get 100% on thirty levels.

"C" Student - Get exactly the required percentage of lemmings thirty times.

Only Sacrifices When Needed - Save as many (or possibly more with a backroute) lemmings as possible with the intended solution thirty times.

Solution Generator - Beat the same level three different ways.

Ouch, That's Gotta Hurt - Lose thirty times purely because of traps.

No Traps Allowed - Get past fifteen different levels with traps without losing any lemmings to them.

[GuyPerfect]

Now comes the part where I have to make a decision.

The short version of the story consists of my life being uprooted and dropped down on the other side of the continent with only a minor semblance of what it used to be like. Following some rather pricy auto repairs, nomadic lifestyle and a day when I had all of 8 dollars to my name, I've got a well-paying job and a place to pay for with it. While I will still persue whatever personal endeavors, the fact now stands that I'm in a position to start work on one of the more major-style projects I've been wanting to work on. Lemmings comes to mind, of course, but it's not quite that simple of a choice.

See, back around the time I started this thread, I started a similar one on a Metroid community site in regards to a fangame based on that property. The gist of it is that the game would be a brainless online shooter with 2D side-scrolling gameplay and Prime-style 3D graphics featuring many weapons and elements from various Metriod games. Perhaps it can best be summed up as a cross between Metroid Prime Hunters and Soldat, provided you've heard of either.

The unfortunate part of this decision is that I'll only pick one project to work on; leaving the community surrounding the other project in disappointment. The reason I bring it up at all is because I'd like some input on which project you guys think I should work on. While you're likely more interested in the Lemmings project, I never know without asking: so here I ask!

I'm still weighing the pros and cons of each project, but do be aware that I've currently got a slight tilt towards the Metroid project at this time.