@MartinEnder got somebody to play it on stream. Stuff I noticed:

@NathanMerrill Are you sure about the last one? I don't see how wearing one at the beginning would affect it. I think walking through a buzz saw loses you the one on your head and the one in your hands

And even if you got through there's still the bricks

yes, I saw it happen

you can break the bricks, and they stay broken on enter/exit of pipe

and if you wear the shell, you damage boost through the first one while holding the red helmet, the red helmet let you damage boost through the second as well as the plant

But how do you take the red one through with you? Just carrying it normally?

you wear the blue one

and bounce the red one up and grab it and carry it through

Oh I think I get it. I assumed that walking through a buzz saw would make you drop what's in your hands

If that's the case, you could cheese it without even taking the shelmet through. Just qb gur fnzr guvat va gur bgure beqre (jrne erq, obhapr oyhr hc, pneel vg guebhtu sbe n qnzntr obbfg)

yeah, I don't think it's a big cheese.

but there are lots of shenanigans you can pull in SMM that you have to watch out for, especially around pipes and doors

@NathanMerrill great feedback, thank you! Do you have a link to a VoD?

@AdmBorkBork Oh yeah, I'll probably wait until it leaves EA though.

@DJMcMayhem got it, nice! :)

Have you also tried that guy's 1 Screen Puzzle Uno? I think that one's even better. S4W-LL5-SGG

Although apparently that one can't be cheesed and I'm not sure how

@MartinEnder I picked it up and played it some last night. It's pretty faithful to the board game in that it is distressingly difficult. For the little bit I played, I was not able to complete a scenario (though, also, I was playing two completely unfamiliar characters -- they weren't ones that we had picked for our gaming group).

I still need to find enough people to actually start Pandemic Legacy :\

Oh that reminds me. Co-op boardgame fans (I guess that also includes @DJMcMayhem and @NathanMerrill): this is probably your last chance to get the KS version of 7th Continent, which I've been playing for the last couple of weeks and it's amazing.

There will be a stripped down retail version of the base game at the end of the year though.

I've played two sessions 2p and most of it solo and either way is great.

I'm hearing 3p is also really good, but 4p gets a bit too much.

@MartinEnder twitch.tv/videos/454289074?t=02h23m21s

@MartinEnder how heavy are the rules?

Haven't sat down to play Gloomhaven but physical game I heard is pretty fiddly. A rule engine might make a world of difference however

@NathanMerrill not very. definitely not casual (especially since some of the scenario take many sessions to complete, and survival is not actually easy), but the focus is definitely more narrative and exploration.

Can you compare it to an existing game?

dice tower has a great spoiler free review

TIME stories without dice?

I was able to convince my family to play pandemic legacy season 1, but season 2 was too heavy on the rules

I haven't played any pandemic legacy unfortunately

@MartinEnder is it Legacy?

I'm watching the video right now

and I can't tell

no

@MartinEnder Oh yeah, if you to try my friend's Die Hard level, here's another course of his (because he only texted me two of them labeled): XLK-05H-8VG

I don't think I did, did you actually share the code?

oh wait "to try" -> "want to try" and not "tried". nvm

Ha ha, missing key words

Whoops

@MartinEnder is this like Hollow Knight the board game? Storytelling and exploration, and when you die, you tell yourself how you'll do better next time?

I guess the item mechanics are kinda metroidvania... ish.

But I haven't died yet, so I'm hoping I can finish the first scenario without doing so :P

I've only played 2 story telling games: Mansions of Madness and Betrayal at House on the Hill.

That said, I'm not a huge fan of the horror theme, so this is much better on that front. However, with those games, I quickly become bored of the mechanics

Betrayal would be miles better with a rules editing pass. Some of the sloppiest shit I ever did read

Yeah. I think they intentionally favored storytelling over balanced rules

@Veskah Beat that one (with the key). That was good!

Coo

Buckets and You're a Tank are both quite silly

Finished Die Hard as well. :)

You can soft lock there though. The first time you need to use the big shell, if you're too greedy you can send the shell back out of the one-way wall.

Ah, I'll make sure to pass on any feedback

Tank is hard

I agree, it's a bit of a bastard

is it even possible to reach the last two bros without the dry shell?

Pretty sure you have to keep it the whole way

got it

it's a lot easier if you just stay in the lava

the claws in Bucket D:

Heh

okay yeah, apart from that I liked the level

Cool, he'll appreciate it. Homie's addicted to SMM

I like the concept of the multiplayer race, but the first pipe is so incredibly superior to the others that it slightly defeats the point

Sp3000 just sent me this amazing puzzle level: 1NK-VVL-F1H (not his)

Friend has in fact responded that he's glad someone likes his stuff. And apparently Cape fall is 3x faster than the rest apparently

hmmm okay, will try that one again

I guess you can damage boost the last two thirds or something

I haven't played it and am merely the messanger

hm, I did beat my record with that path, but it doesn't seem that much faster

maybe a couple of seconds

@MartinEnder are there any expansions you'd recommend?

all of them

Might be a lot faster with YOLO strats or buddy is full of hot air, either are true

I only got the big expansion and I'm regretting it

I'm hoping to snatch the others once the KS-leftover shop opens for my region

That makes the game 2.5x more expensive

-.-

@Veskah yeah, I managed to save a couple more seconds with YOLO strats. can't really do that in a blind race though, and unless you play it locally, you're going to play it blind.

@NathanMerrill true. which is why I only got the big expansion back then.

it's also going to increase the resale value if you have the complete package though.

What is this "re-sale" of which you speak?

I think it's like buying the 2nd edition of it, Adm

Oh gotcha

@MartinEnder Apparently he's going to remake the Mad dash scramble and take your feedback in mind. He's not that proud of it. He also says thanks

cool, looking forward to it! :)

@NathanMerrill oh there's another cheese in puzzle 3. I didn't realise you could spin jump on the piranhas. the idea was that you kill them first with the fire flower

oh, didn't realize that. That's a mechanic I've seen so many times, I didn't even think about it

@martin Got somebody to play your next level. Not a huge fan of it. It's far too easy for players to spam jump without learning the patterns. I'd do something where jumping creates some enemy and the only way you can survive is using the mushroom given to you

No vod this time, sorry

oh! better idea that requires less tech. There's a hidden block on every square, all but one of them will kill you

you don't even have to kill them. If you did the lava jumpers (that fly up and down), you make the level harder to traverse, so it's a punishment, but not an immediately-deadly one

A lava jumper?

Ooh, yes

Any enemy, really, will work. Ideally enemies you can't kill though

@NathanMerrill yeah, that might work. I specifically wanted to avoid being able to die though. I guess I just assumed that jumping around randomly would be less fun :P

If you really wanted to avoid death, make bigger levels. The 3rd level was the best, but the guy I watched still beat just by jumping everywhere

Literally the only level he followed the pattern for was the coin level

Oh, speaking of the third level, I accidentally went through the door before checking out the pattern, so I had to reset the level because the door is above the ground.

Not sure if there's a reason the door is above the ground or not

@MartinEnder I don't know how much of the story mode you've tried out, but have you done levels 12 and 15? Those are some really really well done puzzle levels

I've completed story mode.

Whoops, meant 9 not 12

@DJMcMayhem Well yeah, the idea of the puzzle is that you need to memorise the pattern. I tried to make it as clear as possible that that's going to be the case (the previous rainbow door is supposed to act as a warning that you can't re-enter those), but it was hard to make it fool-proof without using words.

Oh, I see

That's fair. Restarting wasn't too obnoxious

@MartinEnder Oh cool. 9 is "The Keymaster", where there are a bunch of levels and you have to choose if you need a pow, spring, or p switch. 15 is "No jumping allowed". I'm not very far into story mode yet

Ceave's No Jumping Allowed mechanism in SMM1 is neat

I don't think I've seen that one yet. Man, I really need to go back and binge all of Ceave's videos.

I do remember a SMM1 level where every single space had an invisible block, so that was a no jumping allowed mechanism. Just not a particularly inventive one

tl;dr: You're one tile below an off-screen Thwomp who's above a p-switch and the floor is gold blocks

an off-screen thwomp's detection range

> shiny, yet deadly coins

I've grown to love how consistent Ceave is with his phrases.

Hoo ray

Eeevil coins

s/coins/whatever/

@DJMcMayhem oh yeah I really liked the keymaster

and there's a bunch of really neat "no jumping" levels (although some of them are actually just "keep bouncing on the enemies" levels)

Did you figure out the secret ending? I got seven keys, but I think I was still missing one

I think I did but I'm not sure. It's been too long already :D

Have you played the puzzle level I linked above? That's probably the best I've seen so far.

The one-room uno or something?

Not yet

No, this one: 1NK-VVL-F1H

Ah, I'll try 'em both out later

Do you enjoy platformer/difficult levels as well? Or do you mostly stick to the puzzley levels?

I did a hard speedrun the other day (probably took me about 30 tries), and that was a lot of fun (even though it was pretty frustrating)

I do enjoy hard levels, but it's tricky to find some that are both my skill level and well designed. The really well-designed ones tend to be the crazy kaizo levels.

Haha, Epic keep stepping up their free-games... game? After switching from fortnightly to weekly, next week will have two games for free :D (And they're Moonlighter and This War of Mine.)

Ooh!

TWoM is great, and Moonlighter has intrigued me for a while, but not enough to buy it

@MartinEnder how do they have money to do this?

this is baffling to me

Fortnite?

@NathanMerrill Fortnite.

oh, right. A huge amount of initial capital

Also, UE, I guess.

well, thank you Fortnite players for all of these free games, I guess

Also Tencent

@DJMcMayhem Do you have the ID for that? I'd love to try it. I did complete both of GPB's kaizo training levels yesterday.

moonlighter next week looks pretty good

It's a slightly weird game loop, but I've enjoyed the bits I've played of it enough to want to revisit it at some point.

@MartinEnder Hmm, is there a way to see games I've played recently?

Also, the action gameplay is a good way to get your Binding of Isaac fix without all the shit and blood.

@DJMcMayhem yeah, on your Maker page, there's a Played Courses tab

@MartinEnder are you talking about moonlighter?

yes

ok, making sure

Ah, perfect. I'll send you the id in 5 hours :P

yeah sure

@MartinEnder Everything I've seen about that game makes me think that it's the perfect game for my wife

Totally up her alley

Not sure whether I've mentioned this, but I've been working on a zachlike game. The mechanics allow for randomness, but I'm still unsure how to actually do verification of random levels. I'd really like to avoid the issue of 7 Billion Humans where you can build a barely working solution and then just run it often enough until the RNG is on your side.

It might help to just seed the RNG. Then at least you'd have to modify the level to get a different series of random results, which would probably already discourage something like this. Or it would just make this grind even more annoying.

But the other issue is verifying the correctness of a random level with lots of valid outputs. Something like a coin flip or dice roll is easy enough to verify (run it enough times and make sure all results get output... the distribution doesn't even have to be perfect). But if you want a level like "shuffle this list" if the list is more than a few elements long, it's starting to get tricky.

Seeding is definitely a solid option.

Another option I rarely see is something that Dota 2 does.

Specifically, when an action says it has x% chance of happening, it's not actually true

not a game I would have expected to come up in this discussion :D

rather, the likelyhood of it happening is based on the # of attempts since the last success

so, for example, the first time you try to do something, it has, say, a 10% chance of succeeding. the next time, it's something like 30%, then 70%, then 99%

Overall, it may happen over 25% of the time, but what this leads to is an effect where the luckier you are in the first couple of rolls, the less likely you are to be lucky in later rolls

Now, this isn't always feasible, as it means you have to define "lucky" in game code

but it mitigates the effect of 7 billion humans where you roll over and over...because if you're trying to get that 1% chance to happen, it won't happen

this seems useful for action games and stuff where "luck" is more relevant, I think. and people would get frustrated by true randomness.

I'm not sure you would notice this in a zachlike

I'm having trouble combining "zachlike" and "randomness" in my head

at least 7BH and Exapunks have randomness in them

and I think at least 7BH expects you to use it

I don't know 7BH. How does Exapunks have randomness?

I think there's a command that gives you a random value, IIRC

yep, there's RAND

(good thing I have the zine right next to my desk :D)

Oh, so you're trying to allow players to have a "random" in their simulation?

and I think I may have actually used it in a very late game level to break deadlocks

yes, pretty much. in fact, avoiding randomness would make the system a lot more awkward.

And I'm assuming you can't objectively define "lucky"?

aka, which RNG output is better than the other

not really, that depends on how the player uses it. the game's own distributions are all uniform, but the assignment of the results depends on the program.

Something you could consider is doing a normal curve random distribution.

actually, are these random elements binary?

Aka, they either spit out 0 or 1? (Or left/right, etc)?

one is binary, others are a random ordering of 2-4 events.

hmm, I'd have to know more about the game to give a more feedback. I'm having trouble imagining "random ordering of 2-4 events"

how familiar are you with my esolang Labyrinth? :P

with the concept, never used it :)

the game is based on the language. so the binary random event is the same as in Labyrinth (if you have corridors only to the left and right, and a zero on top of the stack, there's no natural choice where to turn, so the choice is random).

as for the others: the main difference between the game and the language is the game lets you use up to 4 robots. but some commands have race conditions if they're performed multiple times in the same cycle (in particular, reading input and writing output, and another one that doesn't exist in the language). if that happens, the order inputs are read/outputs are written is random.

Why wouldn't that be deterministic?

Like, in spacechem, there's an ordering to all of the pieces that makes those kinds of things deterministic

Why would it be? :D

Any order you assign is arbitrary

If one robot is red, and the other blue, that's also arbitrary

sure, but that doesn't have any impact on the mechanics

giving the red 1 and blue 2 gives meaning to the different robots, because they are all identical anyways

I'm 90% a fan of how spacechem does that. It means that you can reorder the different pieces to make everything run smoothly.

the bad 10% is because the ordering is a hidden mechanic, hard to identify which pieces have a higher ordering

I don't know, I'm not a big fan. particularly because of those 10%. it seems inelegant, but also I do like the fact that randomness is possible, it opens an additional interesting puzzle space.

I guess it really depends on the tasks you are giving them

For a game like spacechem that requires exact outputs, it obviously would never work

it sounds like yours allows for variable outputs, but the question is how problematic "getting the wrong ordering" is

I don't think it's going to be a problem for the deterministic levels. The binary direction choice thing doesn't normally come up unless you explicitly set it up. And simultaneous inputs and outputs can easily be avoided.

Basically, the question is less "how do I avoid randomness" (because I really want to keep it), but rather "how do I make random levels work well". I'll probably just seed the RNG for now and stick to levels where the number of possible outputs for a given input is within reason and then just run the program X times and require the solution to produce every output.

But it feels like there should be a cleaner solution that doesn't result in RNG scumming.

@DJMcMayhem Figured I'd play a little bit of NSMBUD and found 3 hidden exits xD (although two of them were actually the levels' proper exits)

You'll have an issue with seeding the randomness: You may have players depend on the seed that is set

Like, if I know that the bots will get it in the order 2,1,3, I may design it to work with that exact input

Hmmm, fair point.

Actually, that's probably not an issue. I wouldn't reset the seed between test cases, so it would only work for the first one.

@MartinEnder 92B-SLP-5BG

Thanks, I'll give it a go tomorrow.