Author Topic: [DON'T] Binary win conditions  (Read 3983 times)

0 Members and 1 Guest are viewing this topic.

Offline Strato Incendus

  • The King of Shimmiers (crowned by Flopsy ;D )
  • Posts: 1746
  • #RIP Spearer/Grenader (2020 - 2021)
    • View Profile
[DON'T] Binary win conditions
« on: September 24, 2020, 11:11:41 AM »
I've been noticing a pattern recently across various games that seems to spark frustration in players, both in me and others. Once I had become aware of this, I started thinking back whether I had encountered it in Lemmings level as well - and yes, I did. It just didn't occur to me back then that this design choice might be a general flaw, if not to say a cardinal sin of game design, instead of something specific to (custom) Lemmings packs.

Thus, most of my "ranting" in this thread will refer to other games than Lemmings ;) . And I'm not planning to hit on any specific custom level pack here. I'm just sure that we all can think of levels where we encountered this type of obstacle.

I'm talking about what I've heard being described as "binary win conditions".
A case in which the difference between success and failure comes down to a single aspect that also happens to be hard to control.
Consequently, the chance of hit or miss basically amounts to the toss of a coin. In other words: It's random. ;) As a result, frustration arises with both outcomes.
- In case of failure, the player is frustrated because they feel there is little they can control to improve the outcome. All they can do is try again and hope for the best.
- In case of success, the player can't really attribute it to their own skill, because (unless they're extremely lucky and succeed on the first attempt) through the previous failures they know the outcome is basically random. Meaning, they externalise their success as "I've just been lucky this time". They will be glad it's over, yes, but it's not something they will enjoy looking back on as a proud achievement.



Here are some examples of games that made me aware of this general pattern:

Magic: The Gathering (click to show/hide)


Star Wars Episode I Racer (click to show/hide)

With examples 2 and 3, because both are video games, knowing that these binary win conditions appear at certain fix points of the game is also hugely detrimental to their replay value: Because you always know you have this huge roadblock ahead of you that basically comes down to chance. This divides every repeated playing experience into a "before" and "after". Kind of like what traumatised people report about their lives... :lem-mindblown:
During the "before" stage, you can't really enjoy the game, because you know the binary win condition is still coming, the dreaded thing is still ahead of you.
During the roadblock, you don't enjoy it, for the reasons stated above - you're just all tense the entire time and simply want to get it over with (again).
And afterwards, while you may be relieved about it being over, a good chunk of the game is now already over, as well. Usually the best part (because most developers tend to put more effort into the early parts of a game, while at the end, they're under time pressure to finish up the product and start losing their creativity). But the best part of the game you couldn't enjoy as much as you would have liked to, because at that time you still had the known roadblock ahead of you.



In original Lemmings, the classical example of a binary win condition is the aptly named Mayhem level "All or nothing":

Spoiler (click to show/hide)

In NeoLemmix
, we rarely have such binary win conditions in terms of execution difficulty. We do however have them in terms of puzzle difficulty.

In particular, this comes up when a certain obscure skill  trick is required, and, similarly to "All or nothing", this trick is the main or even the only thing the level requires:

- If the player happens to know the trick is possible, the level is an auto solve.
- If the player doesn't know the trick is possible, and the level isn't set up in such a way that the player is bound to find out eventually, the level becomes a complete roadblock. Because the knowledge required for the intended solution isn't even part of the player's considerations.

Examples from original Lemmings would famously "I have a cunning plan" and "No added colours or Lemmings".
Spoiler (click to show/hide)
"No added colours or Lemmings" usually gets a pass, because it has a clever solution overall that even makes it the best level in original Lemmings in the eyes of many.
"I have a cunning plan" is the worse offender here, because it's really just about knowing or not knowing the trick. There is an alternative solution, the one that I've been using at a kid. But it basically just requires you to have knowledge of the "No added colours or Lemmings" trick at this earlier level, plus it's much harder to execute:
Spoiler (click to show/hide)

In custom packs, I've encountered binary win conditions e.g. on two levels that both used the Blocker-Basher turnaround / cancel trick. Even though I was later told by the level designer that neither of the two levels where I saw somebody else use this trick actually required it, to this day I still can't think of any other solution for it than using this trick.

With a bunch of skills - and a bunch of tricks regarding their combinations - being available by now, binary win conditions also explain the split that we've been observing in the player base for a while now: Meaning between very skilled and very casual players. And this has been reflected by a split in pack difficulty, between a lot of very hard packs (designed by players who know all the tricks) and very easy packs (designed by newcomers), with few packs in the middle or intermediate difficulty category:

Once you know about the tricks, they become self-evident. You start looking for all kinds of affordances in the level landscape where you can employ them.
However, as long as you don't know them, levels requiring such tricks from you as part of their intended solution (and doing a good job at enforcing them, i.e. being backroute-proof) will be complete roadblocks until you read up on the fact that trick X is even possibly in the first place.

As someone who likes inventing solutions revolving around obscure skill tricks and combinations, this is something I should probably take to heart more frequently
- considering how much such binary win conditions frustrate me whenever I encounter them in other games. ;)

I think the main reason I resort to levels based on skill tricks so much is because I suck at hiding things in plain sight. I can of course create "resource-conservation" levels, but such levels are generally open-ended and will usually leave the player with a lot of remaining skills, instead of all skills being accounted for. This is basically not much different from the issues I've been criticising about X-of-everything levels, yet this is what even my own resource-conservation levels often end up looking like.

Meanwhile, some of the most amazing puzzles I've seen are technically just "additive" challenges, meaning all the skills are used in fairly standard ways, not in obscure skill combinations. And yet, when you approach these levels in a standard manner, you always find yourself one skill short, and you can't figure out why, until you do. This is explicitly not a binary win condition ;) - the level just somehow makes you overlook the obvious, but there are many ways you can ultimately arrive at the correct conclusion.

I've tried to bridge the gap between the "trick-aware" and "non-trick-aware" players via the Noisemaker rank in my pack Lemmings World Tour. However, some of the hints instructing the trick where clearly too vague in the beginning. Thus, I probably created a roadblock of a binary win condition as early as level 08 in the pack, "Glide and joy", which I could only remove by spelling out the trick in the pre-level text explicitly.


So it seems like this is something we all occasionally do. ;) Maybe we should try doing it less in our future levels?
« Last Edit: September 24, 2020, 11:28:04 AM by Strato Incendus »
My packs so far:
Lemmings World Tour (New & Old Formats), my music-themed flagship pack, 320 levels - Let's Played by Colorful Arty
Lemmings Open Air, my newest release and follow-up to World Tour, 120 levels
Paralems (Old Formats), a more flavour-driven one, 150 levels
Pit Lems (Old Formats), a more puzzly one, 100 levels - Let's Played by nin10doadict
Lemmicks, a pack for (very old) NeoLemmix 1.43 full of gimmicks, 170 levels

Offline kaywhyn

  • Global Moderator
  • Posts: 1846
    • View Profile
Re: [DON'T] Binary win conditions
« Reply #1 on: September 24, 2020, 07:06:35 PM »
I've been noticing a pattern recently across various games that seems to spark frustration in players, both in me and others. Once I had become aware of this, I started thinking back whether I had encountered it in Lemmings level as well - and yes, I did. It just didn't occur to me back then that this design choice might be a general flaw, if not to say a cardinal sin of game design, instead of something specific to (custom) Lemmings packs.

Thus, most of my "ranting" in this thread will refer to other games than Lemmings ;) . And I'm not planning to hit on any specific custom level pack here. I'm just sure that we all can think of levels where we encountered this type of obstacle.

I'm talking about what I've heard being described as "binary win conditions".
A case in which the difference between success and failure comes down to a single aspect that also happens to be hard to control.
Consequently, the chance of hit or miss basically amounts to the toss of a coin. In other words: It's random. ;) As a result, frustration arises with both outcomes.
- In case of failure, the player is frustrated because they feel there is little they can control to improve the outcome. All they can do is try again and hope for the best.
- In case of success, the player can't really attribute it to their own skill, because (unless they're extremely lucky and succeed on the first attempt) through the previous failures they know the outcome is basically random. Meaning, they externalise their success as "I've just been lucky this time". They will be glad it's over, yes, but it's not something they will enjoy looking back on as a proud achievement.

This is interesting. Certainly I myself can think of several examples from video games other than Lemmings. The biggest offender to me that I'm also currently playing myself is Candy Crush Saga. Yes, I'm still playing it and I'm all caught up with the most recent episode releases. In the very recent episode releases, the levels themselves have been beyond frustrating due to absurdly low number of moves AND the random candy generator, where you have absolutely no control over what color candy will spawn to take the place of the candies that got removed through matching and setting off of special candies and hence the levels have felt extremely luck based rather than the player needs to rely on skill to pass them. Consequently, I think what King is really trying to do is make people buy boosters to pass the levels, but the thing about them is that even when you do have them, they're still no guarantee that you'll win the level. As a result, I have plenty of boosters all stocked up, mainly because I refuse to use any to pass them and will still sit through each level and fail them repeatedly until the game decides to be nice enough and allow me to pass.

Sounds just like how I am with Lemmings, where I'm willing to sit through long and difficult levels and solve them. Now you know why I don't like viewing solutions before I have solved the level myself first and how I just have a high level of patience when it comes to solving levels in Lemmings. ;)

Quote

In original Lemmings, the classical example of a binary win condition is the aptly named Mayhem level "All or nothing":

Spoiler (click to show/hide)

Partly correct. In Dos, it's not possible to select skills with the mouse while the game is paused, but you can still select skills via the F3 - F9 keys. Although, as you correctly stated, in this level you're only given bashers AND the RR is already 99, and so it's not possible to select anything else in the skill panel even with the Fn keys. And yea, at least much of the rage in the level is eliminated by how it's a very short level and is over quickly despite having to bash in the correct direction 3 times in a row.

Quote
"I have a cunning plan" is the worse offender here, because it's really just about knowing or not knowing the trick. There is an alternative solution, the one that I've been using at a kid. But it basically just requires you to have knowledge of the "No added colours or Lemmings" trick at this earlier level, plus it's much harder to execute:
Spoiler (click to show/hide)

I too also used this solution as a kid. I don't think I used a pass-through blocker, though. Instead, I blocked the first lemming that came back to the left and then did the exact same thing as you did in releasing it, except this time I now have two lemmings to contend with to build and make them turn back towards to the exit. Also, from what I understand, the last lemming out is the one that gets prioritized when it comes to assigning a skill in the crowd, and so I think assigning a miner to release the blocker is all down to luck because there's still a lot of lemmings that need to come out of the entrance by the time it comes to releasing the blocker. Of course, nowadays you can make it easier by simply waiting for all lemmings to come out and then build with one of the last few lemmings to turn around. Or, just delay a few lemmings and simply let the first two bash. As others have mentioned, this level does a very poor job of requiring the builder wall, as well as no level in the game requires this trick. Now that I mentioned it, I'm not sure how I found out about this builder wall trick. Might had been a video I saw on Youtube, or, more likely, I probably just stumbled upon it by accident in a custom level by experimenting.
https://www.youtube.com/channel/UCPMqwuqZ206rBWJrUC6wkrA - My YouTube channel and you can also find my playlists of Lemmings level packs that I have LPed
kaywhyn's blog: https://www.lemmingsforums.net/index.php?topic=5363.0

Offline Strato Incendus

  • The King of Shimmiers (crowned by Flopsy ;D )
  • Posts: 1746
  • #RIP Spearer/Grenader (2020 - 2021)
    • View Profile
Re: [DON'T] Binary win conditions
« Reply #2 on: September 24, 2020, 09:42:04 PM »
Quote
Now you know why I don't like viewing solutions before I have solved the level myself first and how I just have a high level of patience when it comes to solving levels in Lemmings.

I don't necessarily think the first part is particularly special about you; that should be any Lemmings player's default state. :P At least particularly in NeoLemmix, where figuring out the puzzle usually is the sole challenge, and its practical implementation / execution is entirely non-challenging.

Watching replays, from a player's standpoint (not a level creator's standpoint), is really just for when you're absolutely stuck. Even then, I usually just watch up to a point where I can derive a first hint from the replay, and then stop watching so that I can still figure out the rest of it myself.
This works especially well with IchoTolot's replay videos on YouTube, because you can just halt them in place. ;) With a replay file within the NeoLemmix player, you always have to switch between cancelling to try your on solution, then probably reloading the other user's replay again etc.

The second part, however - your high level of patience - that is probably what sets you apart much more from the average level solver. ;) As they say "when the going gets tough, the tough get going". :thumbsup: My patience as a level solver definitely wears thin quickly, especially when I encounter one of my "immediate turn-offs". :evil:
My packs so far:
Lemmings World Tour (New & Old Formats), my music-themed flagship pack, 320 levels - Let's Played by Colorful Arty
Lemmings Open Air, my newest release and follow-up to World Tour, 120 levels
Paralems (Old Formats), a more flavour-driven one, 150 levels
Pit Lems (Old Formats), a more puzzly one, 100 levels - Let's Played by nin10doadict
Lemmicks, a pack for (very old) NeoLemmix 1.43 full of gimmicks, 170 levels

Offline Proxima

  • Posts: 4562
    • View Profile
Re: [DON'T] Binary win conditions
« Reply #3 on: September 24, 2020, 10:38:31 PM »
As you might guess, I'm going to firmly disagree with you here :P

The first part of the rant, I do agree with -- it's not much fun if you put in hard work to get an achievement and it can be stolen from you by sheer luck at the end. ("All or Nothing", while extremely pointless, isn't really that bad because there's no hard work and it takes a very short time to retry.) I can't comment on the other specific examples as I don't know those games.

Trick-requiring levels are not even remotely in the same category, firstly because there's no luck involved.

More importantly, they are not a complete roadblock because it is always possible to work out the solution, even if you don't know the trick in advance. Lemmings came out in 1991, before the internet was constantly available as it is now, and most of us solved all the levels with little or no help. Sure, some of us found out about releasing blockers before "No added colours" by accident or by experimentation on other levels, but I'm sure all of us have had the experience of being stuck and needing to work out the trick to a level at some point or other, whether it was "No added colours", "Compression Method 1", "Dolly Dimple" or any of several others.

And here's the thing: it's frustrating to be stuck, but that is precisely what makes it really satisfying when the "aha" moment kicks in. Custom levels that provide those moments of insight are often the ones that feel the most satisfying when you solve them -- "Don't catch me if you can", "Bipolar Maniac" and "A Matter of Perspective" come to mind (all from the Lix community set).

(I strongly recommend the game Baba is You. Not only is it an excellent puzzle game by any standards, it might well soften your obstinacy on this topic. Most levels in that game seem completely unsolvable until you figure out their tricks, but it feels great when you get them. Of course, Baba can get away with more in that regard because it's completely grid-based, so the space of possibilities available to try is much smaller. Still the same principle, though.)

Quick replies to a couple of other points you made:

Quote
In custom packs, I've encountered binary win conditions e.g. on two levels that both used the Blocker-Basher turnaround / cancel trick. Even though I was later told by the level designer that neither of the two levels where I saw somebody else use this trick actually required it, to this day I still can't think of any other solution for it than using this trick.

In general, once you've seen one solution to a level, it takes special effort to mentally block that solution out and look for others, so that's really not a good argument. You were told that those levels weren't intended to require that trick, and that should close the matter.

Quote
With a bunch of skills - and a bunch of tricks regarding their combinations - being available by now, binary win conditions also explain the split that we've been observing in the player base for a while now: Meaning between very skilled and very casual players. And this has been reflected by a split in pack difficulty, between a lot of very hard packs (designed by players who know all the tricks) and very easy packs (designed by newcomers), with few packs in the middle or intermediate difficulty category.

Not true at all. There certainly are very hard packs out there, but there are a lot of intermediate packs, and there are a lot of intermediate players. Namida is an example of someone who does know a lot of tricks, but sometimes deliberately designs a pack below the difficulty he's capable of, such as LPVI. Coming from the other direction, WillLem's Lemminas is a pack by a newcomer that is definitely not "extremely easy".

Offline kaywhyn

  • Global Moderator
  • Posts: 1846
    • View Profile
Re: [DON'T] Binary win conditions
« Reply #4 on: September 24, 2020, 11:07:11 PM »
I don't necessarily think the first part is particularly special about you; that should be any Lemmings player's default state. :P At least particularly in NeoLemmix, where figuring out the puzzle usually is the sole challenge, and its practical implementation / execution is entirely non-challenging.

Watching replays, from a player's standpoint (not a level creator's standpoint), is really just for when you're absolutely stuck. Even then, I usually just watch up to a point where I can derive a first hint from the replay, and then stop watching so that I can still figure out the rest of it myself.

Ok, I agree with you about players using replays when they're stuck. However, what sets us even further apart is this: I don't watch replays, even when I'm stuck :P I only watch after I solve the level myself first, and so really at that point I'm simply comparing my solution to that of another player(s) or, if available, the intended solution from the pack author. Also, I prefer to not ask for hints if I can, but I have done that plenty of times for United. Even then, this is different, as the hints I've been given are vague enough so that they allow me to think and still allow me to figure out the solution for myself. This is exactly what I appreciate, as my system of giving hints to help other players is the same, as my goal is to help players improve their critical thinking and solving ability while preventing spoiling of one's enjoyment of figuring out the solution. With a replay, however, I can simply interrupt it or imitate where to place the skills to solve the level, which in my opinion isn't fun for me at all, as the solution has now been spoiled but in visual form, and I prefer the satisfaction of figuring out the solutions for myself. Thus, I don't ever watch replays to levels, even when I'm stuck. I only watch after I have solved the level to compare solutions.

It's like you said in your Paralems topic: "You can't un-see what you have seen."
https://www.youtube.com/channel/UCPMqwuqZ206rBWJrUC6wkrA - My YouTube channel and you can also find my playlists of Lemmings level packs that I have LPed
kaywhyn's blog: https://www.lemmingsforums.net/index.php?topic=5363.0

Offline Dullstar

  • Posts: 2092
    • View Profile
    • Leafwing Studios Website (EXTREMELY OUTDATED)
Re: [DON'T] Binary win conditions
« Reply #5 on: September 24, 2020, 11:50:58 PM »
On the topic of tricks where you either know them or you don't, I think a pinned topic listing pretty much all the known ones might be a good idea. People who don't want them spoiled could simply not read the topic, while people who are stuck on a level and want to know if there's a trick they don't know about can peruse the topic to see if any of the tricks might work for the level they're stuck on.

I constantly see ones referred to that I didn't know about: e.g. the Blocker-Basher turnaround / cancel trick mentioned earlier.

Offline Simon

  • Administrator
  • Posts: 3860
    • View Profile
    • Lix
Re: [DON'T] Binary win conditions
« Reply #6 on: September 25, 2020, 07:40:25 AM »
Very nice article.

Definitions. Win/loss conditions are always binary, and single-factor by nature. I'll be wary of how you use the term binary win condition; you mean approachability of the solution from many angles. Yes, that makes for nice levels, they induce careful economic decisions about which parts of the attempt to ditch.

I doubt that there is a definable sharp division between single-trick levels and levels with large-scale economic decisions.

Aha effect. Proxima is right about how the aha effect is not dependent on multi-angle approachability. The aha effect will come from a nontrivial economic decisions as well as from insight of skill interaction for tricks, as long as it feels fair.

I believe the strongest aha effect comes from single-trick levels whose tricks feel entirely fair, but look impossible at first. This is extremely hard to design on purpose for stronger players, and risks being fidgety and obscure.

Baba. Mid-to-late Baba is strongly single-trick based. Potential issue: Late Baba is not honest as a purely logical puzzle. It requires lateral thinking on what the game's rules might be. It doesn't feel unfair, but it tastes different.

Zendo. After the pandemic, everyone is cordially invited to play lots of Zendo with me. I miss visiting people from the community. I want to do that more again, finally having a stable job.

-- Simon

Offline WillLem

  • Posts: 3345
  • Unity isn't sameness, it's togetherness
    • View Profile
Re: [DON'T] Binary win conditions
« Reply #7 on: September 27, 2020, 01:28:30 AM »
Interesting topic!

In original Lemmings, the classical example of a binary win condition is the aptly named Mayhem level "All or nothing

Just as a fun fact, there is actually a way around this, at least in the Amiga version:


I get your point though: not knowing this trick essentially reduces the level to random chance. It may not give the same level of satisfaction as solving it using skill instead of luck, but solving such a level can come with its own sense of satisfaction for having been "lucky enough" to get all 3 bashers in a row correct.

Also, random chance is a big part of many games. Any game involving dice throws or shuffled cards is based largely on chance, with the player only able to influence the game once the dice have been rolled, or the cards dealt. Admittedly, I generally tend to prefer games that involve a greater degree of skill than luck, but I can see why games of chance are popular. And I love a game of Monopoly occasionally ;P

binary win conditions also explain the split that we've been observing in the player base for a while now: Meaning between very skilled and very casual players. And this has been reflected by a split in pack difficulty, between a lot of very hard packs (designed by players who know all the tricks) and very easy packs (designed by newcomers), with few packs in the middle or intermediate difficulty category

I'd agree that sometimes simply knowing a particular trick can mean the difference between a relatively straightforward level solve and a seemingly impenetrable puzzle. However, I'd disagree that this has anything to do with a player's level of skill, and/or whether or not they play casually. Both very skilled and very casual players are capable of knowing or not knowing any given trick or skill application. Furthermore, being casual and being skilled are not necessarily mutually exclusive.

I do see where you're coming from though: it's more likely that highly skilled players who play regularly will know more tricks, and be more generally practiced at puzzle solving, so levels requiring more obscure tricks seem to have a better chance of polarising a given group of players.

I'd probably describe myself as an "intermediate" player. My puzzle solving capability is probably below the community average, but I do know a few of the more advanced tricks. Also, when playing on a platform that requires good execution as well, I can be a bit of a wizard (check out my pause-free 100% solve of Cascade, for example!)

Also, when I'm going through a Lemmings phase I play a lot, so I'm not sure I'd describe myself as a "casual" player, but I definitely do have a more laid-back approach to the game than others, and I'm happy to move on quite quickly if a level's solution is a bit beyond me (which happens a lot!). I think it's a spectrum really, and any player can be anywhere on it from one day to the next.

Offline Dullstar

  • Posts: 2092
    • View Profile
    • Leafwing Studios Website (EXTREMELY OUTDATED)
Re: [DON'T] Binary win conditions
« Reply #8 on: September 27, 2020, 09:58:27 AM »
To be fair, if I recall correctly, All or Nothing is technically not random*; just frame perfect.

It still sucks.

*I'm pretty sure there's a priority order to decide which lemming gets chosen, but the lemmings are effectively shuffled, so selecting one lemming from a crowd might as well be random.

Offline mobius

  • Posts: 2747
  • relax.
    • View Profile
Re: [DON'T] Binary win conditions
« Reply #9 on: September 27, 2020, 02:24:54 PM »
thanks for bringing this to my attention. It makes sense when you think about it and I've thought about this before but I never heard it termed this way.

There are definitely a LOT of games that have this problem. Even many overall good games have them in spots.
everything by me: https://www.lemmingsforums.net/index.php?topic=5982.msg96035#msg96035

"Not knowing how near the truth is, we seek it far away."
-Hakuin Ekaku

"I have seen a heap of trouble in my life, and most of it has never come to pass" - Mark Twain