Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[vorpal+5]

@Xenoargh

I find you are hitting a bit hard on the knife from Teleglitch! By moving backward and stabbing repeatedly you can manage to hit with a decent probability, and it will save you some bullets.

[xenoargh]

Oh, it "works", sure.  It's just not great; it's a classic example of a mechanic that works, functionally, but doesn't meet the real need.

[Morrokain]

I see RNG as a game design "spice". Sprinkle some here and there in your recipe, and it brings out the flavor of your dish and actually does quite a bit to extend a person's threshold of repeatedly eating at your table.

Overdo it (very very easy to do), and the entire dished is ruined and probably inedible. It should never be the core ingredient, because spices only work to emphasize the core flavor and texture of the main ingredients.

A perfect example is The Binding of Isaac. Everything is so RNG heavy, you can actually spawn in dungeons that are unbeatable. There just isn't a way to get to the exit door. It doesn't matter how good you are at the game.

Your starting upgrades are also the difference between an easy run, and hardmode more often than not- so running into the above scenario is all the more frustrating if you were lucky in the beginning.

The main problem I see with too much RNG is that its already hard to predict and limit the standard starting experience pitfalls of a new player who isn't as well-versed with your system, so adding in the random element just makes it impossible. With that in mind, how likely do you think the player is going to hit the "first time fun" sweetspot under those conditions? lol

[David]

*sees talk of randomness, swoops in*

I simply must share this blog post which breaks down types of randomness and categorizes how randomness can be used in games.

In particular, it discusses input randomness ("procedural generation") vs. output randomness ("noise injected between the player's decision and the outcome")

What it actually does is obscure the outcome. You may have played perfectly, and still lost. The game has now sent you off on a wild goose chase, thinking about where you must have messed up, when in fact your play wasn't the problem; dice rolls were.

Because of that wild goose chase, the game seems more complex than it is. The game provides unreliable feedback, and only after playing many, many games will it become clear which feedback you should ignore. Essentially, random games delay learning - the essential fun part of games - by injecting false signals into the engine.

Now even with that said, I don't think the takeaway should be that output randomness = bad randomness. But it depends entirely upon how the game is handling and contextualizing the result. For example, rolling the dice badly in D&D can lead to Vardox the Barbarbian failing to slay the Necromancer, leading to him getting brainwashed and turned into a minion serving the dark empire of 1000 years, but that can be a fascinating story. If the game simply says "you failed! try again ... and watch this five minute cutscene again before you try", that's rather frustrating. Similarly,

Everything is so RNG heavy, you can actually spawn in dungeons that are unbeatable. There just isn't a way to get to the exit door. It doesn't matter how good you are at the game.

... Doesn't sound great. And that's even the first type of randomness, procedural generation. It just has some problems with output that would be better served by having the generator run better evaluation of the dungeon generation results. ('Course this can be quite difficult/expensive to do - I sure know, what with working on Dredmor. Our solution was to make dying funny so that players didn't mind quite so much if they died. And the way loot/xp worked, you usually flattened out the variable difficulty of procgen within the first two levels of the game, where 90% of playtime and deaths occurred anyway.)

Ahem. Back to work!

Oh, maybe a parting shot for "worst mechanic": Yeah, I think gambling-style gacha mechanics in games that are tied to players spending real money AKA loot boxes is an ethically compromised mechanic that should be regulated by law because it is literally gambling -- but with even worse payout to players than eg. playing slots). Also it should never be in games anyway because it's relying on psychological exploitation to be appealing/addicting (I won't use the word "fun"!)

[Morrokain]

^ This is a fantastic article thank you! I greatly enjoyed reading it and have given it a lot of thought before I wanted to respond because I absolutely love these kind of discussions.

I certainly do agree that the strategy element needs deterministic design to give the player a reason to continue to play. We all want to learn and improve at a game over time. It is our very nature to do so. This was what I was referring to as the core ingredients in a game. It is hard to do, but giving design complexity such depth that it takes more than a single skim over the surface to learn its intricacies is the inherent goal of good design. We want the players to continue to have fun each time they play because they learn to get better and have a unique and enriching experience each time.

However, there is an assumption being made here. It fails to adequately appreciate experience as a factor in fun alongside the strategy element. The article responds to several counter arguments. Specifically, it mentions the simulation of real life- which indeed contains much randomness to account for, and I feel this should not be discounted as much as it is as simple "noise". It makes the mistake of assuming the only reason of playing a game is to learn to be the best at it, and that truly does makes sense and should form the foundation of the design. It is born from our inherent competitive nature that willed us to survive in the wild to learn to be better or die. It is a powerful attraction, but therefore so is our imagination that allowed innovation.

I believe there are many reasons to play a game beyond this that can strengthen the overall design. It can be a simulation approach (as the article stated had separate parameters) that can bring a lot of depth in its own right because it gives the ability to inject yourself into a scenario and be fulfilled in the idea of the possibilities from the start to the outcome which can change with each attempt. It provides a unique experience each time even within the framework of skilled play leading to better results, kind of like the DnD reference to dice rolls leading to an interesting story.

Roleplaying has been popular for decades because the inherent sense of building a character and acting out a play from your own imagination is fun! If the outcome was always the same with the same input, it would no longer be quite as fun. It was entirely predictable after a certain threshold of learning. Whereas if the goal was the experience rather than a fulfillment through increasing knowledge of the system, more weight would be given to differing outputs because they provide inherent replay-ability even under the same starting context as long as the variance isn't high enough to ruin the fun of the strategy element.

There can be a certain element of fun to the idea of taking a risk in the hopes of gaining an advantage or losing an advantage as long as two things are present in all cases:

Spoiler

1) The player understands the rules of when its random. If it is under the surface it will just confuse the player. The player must be willing to accept the outcome, even if unfavorable, because they were hoping for the alternative and were willing to sacrifice the cost regardless. But, one has to consider the psychology of such an action. Would it be when they are comfortable? Of course not. One would only resort to a gamble when all other alternatives have been attempted. Their back is against the wall and there is nothing left to lose. But the ability to then gamble makes all the difference. Because it adds an extra element when the alternative is simply losing outright. The system can still be deterministic. It gives rise to the "underdog story" that permeates most of our imaginative thinking and resonates within unique experiences where a player otherwise probably doomed anyway received a saving grace through a calculated risk that could have made defeat certain.

2) I feel that giving the player levers to influence the outcome adds complexity in its own right too, because now the player can somewhat determine the odds in some way. Do you have the resources? Then use them when it matters. It provides its own sense of strategy. Admittedly, this can lead to high highs and low lows since giving the highest sacrifice of risk and still getting the worst outcome will make anyone a little salty. I love X-Com, and the times I missed on 95% seem impossible..  But, again, as long as that is communicated upfront players who enjoy that unpredictability will still remember the highs when they achieved something against the odds.
^
Still, even the above stated randomness cannot be effective as a whole due to the emotional response of either extreme- which in too high of doses would be either stressful or even boring without a deterministic shell. I would call this nested randomness, which translates to:

Spectrum:
High End: Deterministic - Or at least very very close

Middle  High - Slightly random output

Middle - Fully random output

Middle Low - Slightly random output

Low End: Deterministic - Or at least very very close

[close]

-----
So the idea here is that the player can still control the outcome, but only on the edges of the spectrum that the player also has control over. The player has to invest other resources in order to obtain- or lose their ability to bargain from their actions. But each experience in between those two extremes remains unique and true to the immersion of the setting. The setting is a very important context for the choices a player makes. It is only the convergence of immersion and solid design that leave imprints deep enough to warrant future investment into the learning attempt in the first place, after all.

Anyway this is already rambling on so thanks for the input! 

[ciago92]

Minigames are bad. But often optional and ignorable.
Minigames that are gates to actual content are awful.

'Warscore' is one of my (incredibly sujective) pet hates. And the reason I bounced off Stellaris (and every other paradox mapgame) hard.
If I'm playing an empire game and end up in a war, the rules are as follows:

• Take what I can
• Give nothing back

Being forced to specify what you want to take beforehand, and then give back the rest is utterly baffling to me in a game.
It's one of those fun > "realism" things.

THIS^ You keep what you kill! This, war weariness and forced peace treaties is why I just can't get into Stellaris. I came to fight space battles, blow s*** up and take over the universe! Not sit on my paws and kiss political a**

It makes sense in settings where there is an overarching legal framework within which all rulers operate; i.e. 'rules of war'.
The 'game' then becomes focused on the politics & intrigue behind acquiring the necessary claims & CBs to enact the wars you desire.

That's why ck2 is such a great game, and one of the reasons why Stellaris is such a dismal failure.
It's also why everyone will eventually outgrow Civ & Total War games.

If only Stellaris had been set in the Dune universe.....

Two points I'd like to address here. One, when did Stellaris become a dismal failure? It's still growing and thriving as far as I"m aware.
Two, you can make empires that ignore standard diplomacy and instead just go "yes I'd like to take everything you've ever owned, kthxbye". I get that it's not perfect to your vision, but I feel like it's a reasonable substitute at least.

[Midnight Kitsune]

Two points I'd like to address here. One, when did Stellaris become a dismal failure? It's still growing and thriving as far as I"m aware.
Two, you can make empires that ignore standard diplomacy and instead just go "yes I'd like to take everything you've ever owned, kthxbye". I get that it's not perfect to your vision, but I feel like it's a reasonable substitute at least.

Paradox has actually lost quite a few character through out the years due to updates that completely changed the way the game was played and or locked away a much needed QoL feature in one of their MANY DLCs. (Or so I've heard) The worst one I can think of is the 2.0 update that removed the unique ways of traveling space and just made everyone use hyperlanes instead.

[David]

The worst one I can think of is the 2.0 update that removed the unique ways of traveling space and just made everyone use hyperlanes instead.

Oh man, I've got the total opposite take here. (No idea about the rest of Stellaris mechanics lately, haven't played it in quite some time, though I think overall it's got issues with scale and feeling generic. I do recall having the thought that the tile-building part of the game didn't matter because placement had essentially no meaning, so removing those was probably a good move) -- right, but going hyperlanes fixed the problem of the map's geography having little meaning!

If anything is accessible at any time (normal warp drive), then there are no map chokepoints. Nothing matters except timing and raw distance. If it's the same deal but you need to build warp stations, then you get the same problem plus more micromanagement. And if you're the poor fool who selected hyperlanes, you have to follow rules that everyone else breaks and therefore get to play whackamole with enemy fleets who go anywhere - or get shut out by chokepoints. Making everyone follow the same rules means there's interesting geometry to the map, and systems then have meaning based on where they are in addition to what they are.

Now that said, I don't think the combat mechanics quite hold up, but it was a good move! (Perhaps alternately they could have put a ton of design work into making the travel methods interesting w/re to some notion of map geometry and terrain, and interact together interestingly, but that'd be exponentially more work. So: still a good move to make the cut, I feel.)

[Midnight Kitsune]

The worst one I can think of is the 2.0 update that removed the unique ways of traveling space and just made everyone use hyperlanes instead.

Oh man, I've got the total opposite take here. (No idea about the rest of Stellaris mechanics lately, haven't played it in quite some time, though I think overall it's got issues with scale and feeling generic. I do recall having the thought that the tile-building part of the game didn't matter because placement had essentially no meaning, so removing those was probably a good move) -- right, but going hyperlanes fixed the problem of the map's geography having little meaning!

If anything is accessible at any time (normal warp drive), then there are no map chokepoints. Nothing matters except timing and raw distance. If it's the same deal but you need to build warp stations, then you get the same problem plus more micromanagement. And if you're the poor fool who selected hyperlanes, you have to follow rules that everyone else breaks and therefore get to play whackamole with enemy fleets who go anywhere - or get shut out by chokepoints. Making everyone follow the same rules means there's interesting geometry to the map, and systems then have meaning based on where they are in addition to what they are.

Now that said, I don't think the combat mechanics quite hold up, but it was a good move! (Perhaps alternately they could have put a ton of design work into making the travel methods interesting w/re to some notion of map geometry and terrain, and interact together interestingly, but that'd be exponentially more work. So: still a good move to make the cut, I feel.)

Oddly enough, another game did the different drive systems and worked out quite well: Sword of the Stars. Humans had the hyperlanes but they were also the fastest mode of transport and they were still able to "slow boat" if they needed to

[Semondice]

Weapon Degradation and Gun Jamming. Impossible to do well, never should be put into your game. Right? Wrong.
It turns out it's actually just been implemented badly this whole time.

Share the sentiment, but for to very good and very notable exeptions, namely the S.T.A.L.K.E.R. saga and Farcry 2.
In S.T.A.L.K.E.R. is not particularly fleshed out in animations or sprites in vanilla; probably it's a combo with the atmosphere, the gameplay, the difficulty and the rest, but it works pretty good.
Farcry 2, in that sense, is extremelly good. Could teach a lesson or two to many other present works; maybe a little slower on the degradation speed, though.

[Kat]

Escort missions, in almost any kind of game that features them, always seem to be very hit or miss in their implementation.

The ones I detest are the heavily scripted ones where enemies spawn out of thin air, but only when the escorted entity reaches a particular point, on their unalterable path. You can't sweep the area ahead, you have to do close escort.

Example: there was an escort quest in one of the Elder Scrolls games that I played, Oblivion it was. The npc I had to escort moved along a non-direct route, and every so often enemy archers or spellcasters would spring into existence, and of course they targeted the npc first.
So it was a case of having to reload several times, to work out the pattern of spawn points.

Contrast example: bomber escort mission in European Air War, you could use the autopilot function which would skip you ahead to the next enemy encounter, where your fighter wing, the bomber wing, the enemy fighters would all have scripted positions. But you could if you so chose, fly manually in which case you could encounter the enemy fighter wing at a different point relative to the bombers, and you could choose to gain more altitude prior to enemy contact, than the scripted autopilot positions & altitudes.

Weird example: Goat Simulator's MMO mode. It has an escort mission, where the npc moves extremely slowly. But you can subvert it, by licking the npc and stuffing them into your inventory, and then using your goat's own abilities to quickly move to the destination and then throw the npc out of your inventory.

[Midnight Kitsune]

Escort missions, in almost any kind of game that features them, always seem to be very hit or miss in their implementation.

The ones I detest are the heavily scripted ones where enemies spawn out of thin air, but only when the escorted entity reaches a particular point, on their unalterable path. You can't sweep the area ahead, you have to do close escort.

Bonus points: No mid mission checkpoints, the NPC moves slower than your run but faster than your walk speeds and has an aggressive/ reckless personality

[Grievous69]

Escort missions, in almost any kind of game that features them, always seem to be very hit or miss in their implementation.

The ones I detest are the heavily scripted ones where enemies spawn out of thin air, but only when the escorted entity reaches a particular point, on their unalterable path. You can't sweep the area ahead, you have to do close escort.

Bonus points: No mid mission checkpoints, the NPC moves slower than your run but faster than your walk speeds and has an aggressive/ reckless personality

I started shaking just by reading this, curse you 'Nam flashbacks 

[Sinosauropteryx]

this blog post

Thanks for sharing, this was a good read.

I agree with the author's breakdown of output randomness and what it does to a game, but I think he's being needlessly specific in his language use, particularly the word "game" itself, as he defines here. In common parlance, what he describes as Toys, Puzzles, and Contests are all "games," especially when played on a screen and referred to as a video game. But the article takes a specific definition of Game in which understanding the system is the core value of the experience, and anything that detracts from that (like RNG) necessarily makes it less Game-like, and thus a worse Game.

This overly specific definition does a disservice to the whole article, as @Morrokain's post demonstrates. The author might argue that, by his definition of game, immersion and roleplaying do not enhance a game, even if they do enhance the interactive experience of which that game is a part. (Or to paraphrase, there is a set of values for strategy games which we can separate from the set of values for a roleplaying experience.) The problem is that no one uses that narrow definition of game; the interactive experience IS the game. The author could have defined terms at the beginning of the article, but instead I had to click on a small link in the rebuttal section to see his working definition.

Ultimately, an interesting look at strategy games from a purist perspective, but less applicable for designing games people actually want to play.

[Noviastar]

I got to throw jumping and climbing in most games.  *cough Elder scrolls* 

Seriously.  a rock is a barrier to advancement.   I can jump in some and climb some places in others.  A simple pile of wood or rocks can be a wall of total impassibility. 

This then gets to world/level design.  When you make things you can't get over.  that IRL you totally could.  Kinda breaks things... 

But i have acrobatics at 100.  so no big deal.