How Games Use Feedback Loops
ฝัง
- เผยแพร่เมื่อ 22 พ.ค. 2024
- 🔴 Get bonus content by supporting Game Maker’s Toolkit - gamemakerstoolkit.com/support/ 🔴
Playing Pyre over Christmas got me thinking about feedback loops: the reward structures in games that can reinforce or balance out winners and losers. In this episode I’ll explain what this all means, and talk about the design of Pyre’s positive and negative loops.
=== Sources and Resources ===
- Sources
Solving XCOM’s Snowball Problem | Pentadact
www.pentadact.com/2016-02-25-s...
Git along there, little doggies | Team Fortress 2
www.teamfortress.com/post.php?...
- Additional resources
Level 4.4: Feedback Loops | Canvas
learn.canvas.net/courses/3/pa...
Designer's Notebook: Positive Feedback | Gamasutra
www.gamasutra.com/view/featur...
[PPT] Feedback Systems and the Dramatic Structure of Competition | Mark LeBlanc
algorithmancy.8kindsoffun.com/...
=== Chapters ===
00:00 - Intro
00:35 - What is a feedback loop?
00:58 - Positive Feedback Loops
01:55 - Negative Feedback Loops
02:43 - Benefits of feedback loops
03:47 - Disadvantages of feedback loops
06:13 - How Pyre fixes feedback loops
08:16 - How loops balance themselves out
10:55 - Dampening feedback loops
12:50 - Patreon Credits
=== Games Shown ===
Bastion (Supergiant Games, 2011)
Transistor (Supergiant Games, 2014)
Pyre (Supergiant Games, 2017)
Call of Duty: Infinite Warfare (Infinity Ward, 2016)
Call of Duty 4: Modern Warfare (Infinity Ward, 2007)
Chess Ultra (Ripstone Ltd., 2017)
Mario Kart 8 Deluxe (Nintendo, 2017)
Tekken 7 (Bandai Namco, 2017)
Splatoon (Nintendo, 2015)
Forza Horizon 3 (Playground Games, 2016)
Titanfall 2 (Respawn Entertainment, 2016)
Team Fortress 2 (Valve Corporation, 2007)
Arms (Nintendo, 2017)
XCOM 2 (Firaxis, 2016)
Battlefield 4 (EA DICE, 2013)
SteamWorld Dig 2 (Image and Form, 2017)
Final Fantasy XV (Square Enix, 2016)
The Witcher 3: Wild Hunt (CD Projekt, 2015)
Resident Evil 4 (Capcom Production Studio 4, 2005)
NBA 2K18 (Visual Concepts, 2017)
Civilization V (Firaxis Games, 2010)
Dark Souls (From Software, 2011)
Call of Duty: Modern Warfare 2 (Infinity Ward, 2009)
Call of Duty: Black Ops (Treyarch, 2010)
Devil May Cry (Capcom, 2001)
Donkey Kong Country: Tropical Freeze (Retro Studios, 2014)
=== Credits ===
Music used in this episode
Pyre soundtrack
k. Part 2 - 01 untitled 1, animeistrash
Other credits
MW2 - 6 KILLS WITH ONE PREDATOR MISSILE | Volound
• MW2 - 6 KILLS WITH ONE...
TACTICAL NUKE with AKIMBO Model 1887 - Modern Warfare 2 | TheKoreanSavage
• Video
=== Subtitles ===
Contribute translated subtitles - amara.org/v/C3BEl/ - เกม
Personally I'm not a fan of the GMTK feedback loop where the more videos you upload the better you get.
I was about to leave a comment of the same sort. But I am stuck in this loop of appreciation.
BRUH TURBO all you gotta do is upload some videos for a change. Train up 😤
If you're upset at this guy's improvements, you should avoid this youtube channel Turbo Button. He's on the same improving feedback loop as well
The good thing about GMTK feedback loop is that they're all beneficial to the channel. He gets a positive feedback loop for the better videos (more viewers, subs, money, leading to more determination towards future videos), and negative feedback loops for his worse videos (constructive criticism from friends, leads to fewer poor quality videos).
Literally a win-win in this scenario. Not that I've seen the latter possibility yet, but it'd still be beneficial for him to fail.
ill check it
When we were working on Elite: Dangerous, during the alpha we found that players could die and end up in situations where restarting the game was a better option than continuing, due to losing their ship and cash. In the end we made a 'debt' player tag, where players can always buy back their ship in the early game, but end up in debt and will be offered missions to clear the debt. This might be humiliating missions, like transporting dung, or questionable/dangerous missions like high profile assassinations. It was a nice way to let a player turn around a death spiral situation.
Transporting dung! Brilliant
But is restarting quicker and less humiliating?
Nazareadain so the idea is that completing the debt mission always leaves you substantially better off than just restarting.
A game providing an option for a way out of a bad situation, a way that exists *within* the fourth wall (meaning instead of reloading a save, or restarting a game, or creating a new character) is awesome.
> Transporting dung! Brilliant
It's possible that this phrase has never been uttered before in human history :D
I remember as kids my sister was absolutely unbeatable at Mario Kart.
She had a hustle where she would bet candy or pokemon cards or whatever and she would act like she sucks. She would hang back, and take advantage of the reward system, get a good item and save it while she utilized the rubber band system to launch herself toward the front and take out the winner right at the last moment.
She didn't understand the mechanics fully as we do now, but she had an inherent understanding of the rubber band and item system. She wasn't sure how it worked, but knew how to take advantage of it anyway!
Those were good days.
She won me an awesome World Industries skateboard that had flame boy and wet willy on it in like a cool yin-yang type design.. it was a really cool board in the late 90s-early 00s! Haha
My sissy is the best.
I had one of those skateboards, really liked it :D
That is actually big brain
Amazing childhood memories. Hope yall are doing well today.
@@Tr0lliPop HUGE BRAIN
Rubber banding is only for AI that was something else she was doing
9:28 “This is similar to real world basketball” *shows basketball video game*
Ironically, because he said that, I thought for a moment it was real world basketball but with everyone moving as if they had a plague. Then I realised what was happening and was reminded of the depressing inverse relationship between progress in graphics and average quality of available games.
This feedback loop system in Pyre gives me an interesting idea for solving the Xcom problem you mentioned. You could have it so troops that become promoted to a certain level in combat become non-combat officers that provide bonuses for lower ranked troops, which would easily have an in-fiction explanation. You could make it so it's a constant forward progression of making your entire army more powerful, rather than the current state in Xcom where it's a rather static affair. I've had playthroughs of Xcom 2 where I had a group of extremely powerful troops fall prey to some dome tactical mistake I made, only for that to turn into the entire game being a miserable failure because I simply wasn't prepared to continue on with rookies or low ranked squaddies.
Haha, I really like that idea of promoting a soldier into a non-combat role if you use them too much. It would need to be carefully balanced (and it would probably be a controversial mechanic that many would mod out of the game!) but could be very interesting...
I think ideally the bonuses would be something the player would actively seek out, otherwise you'd run into issues of players benching their best troops for important missions and avoiding ranking them up. You'd probably have to balance the game to be slightly harder as a result but I think the outcome would be at least interesting.
The Non-Combat troops could also end up having powerful one-use abilities in a battle or something like that, such as an airstrike. Something that players genuinely wouldn't mind giving up a soldier for.
You have to make sure not to make the officers too powerful either, otherwise failing players, who don't manage to promote a unit that far, miss out on the bonus buffs/abilities and fall behind the desired strength curve once again.
Maybe the officers get their own command and you can call in a very powerful one-time-only favour which can help you crack a level you are struggling with. This way the day to day gameplay isn't impacted too much but it would still be worthwhile to send off powerful units.
It would also be fitting for a possible XCOM 3 scenario. Imagine this: After defeating the aliens on our hometurf (once again) it is now time to take war back on these suckers! This is why we need more troops now, an army even. And as you are invading the aliens' home planet these officers become increasingly important because they can lead strike teams that go on missions on their own which in turn rewards you with important resources.
So when can we expect to see a Chess 2 with better gameplay?
You can play Checkers, which is like Chess, but streamlined for a wider audience.
There actually is a game called "Chess 2". Whether it has better gameplay is up for debate.
Chess: Origins. Pre-order now to get the free "Bishop's Remorse" DLC for free
I would hate to be winning at Chess 2 and getting my King Blue Shelled :D
I'm playing Chess Evolved Online, which is free to play. It's basically chess crossed with a collectible card game, with many new pieces and some new game mechanics to go with them.
One thing that is a very good point is that there is *no* energy mechanics that prevents you to play to push you to spend money.
Furthermore, you can get both currencies by simply playing, so using money will just help you build up your army faster (and thank the dev for his work). The way the game limits what pieces you can put in your army also makes it so that you're not going to end up constantly chasing to catch up to paying players. Well, there are new pieces added and a rebalance done with each patch (there is one every few months), so you can end up constantly reworking your army and being in need of money, but it's not inherent with the game, just with the fact that it's still being worked on.
I recommend it.
Wait, Tekken 7 HAS a negative feedback loop of of sorts, though. When you’re low on life, not only do you get a damage boost, but you also gain access to the game’s version of Supers and EX moves (called Rage Art and Rage Drive respectively), all designed to allow the losing player to amount a comeback.
Granted, Tekken 7 isn’t the only fighting game to have this sort of mechanic; most fighting games since Street Fighter 4 has one (for better or worse, depending on who you ask). But it still has one all the same.
Ack, you're right! I didn't know that. Luckily the major point still stands that nothing carries over from round to round. But thanks for the correction.
Mark Brown Oh absolutely, your point still made perfect sense. I just wanted to clarify that there WAS in fact a negative feedback loop there, just not in the way you where using the game to make your point :).
And as a fun side note, there ARE fighting games where the damage incurred on the winner of a round is carried over to the next round (Darkstalkers and Killer Instinct comes to mind). But that’s almost more like having one big permanent health bar that stretches between rounds, rather than a negative feedback loop per say. Still a design decision I always found interesting.
basically like the rage effect in smash 4
Almost ALL fighting game designs have a feedback loop inherent in juggles/combo systems that revolves around the stun locks for chaining consecutive hits. Of course all good fighting games will cap combo/juggles at some point (usually by the insertion of uninterruptible cooldown frames in the animation, or offer an invulnerability frame for the damage recipient at the end of a long combo chain to give a chance to escape.).
I do believe there have existed some perpetual slippery slope exploits in older, archaic fighting games that make escaping some juggles basically impossible. Artifacts of when management of feedback loop mechanics in games were less well understood.
Elia, forse Yes. It’s even called the same thing: Rage. Possibly because Namco was a co-developer on the game, but I don’t know that for a fact.
I just tried Shogi for the first time in Yakuza 0 and man... Shogi sure is a lot more brutal than chess in this regard. Losing a piece not only means you have less tactical options, it also gives the opponent and incredible number of options, as every "captured" piece can be put back on any free space on your side.
That's true, but it also counteracts the standard Chess follow-up, where the player with more material trades pieces aggressively, because whatever advantage they have is amplified when there's less total stuff on the board.
In Shogi, you have to be careful, because trading pieces gives your opponent more ammo to exploit any weaknesses in your castle, and could be the thing that creates those weaknesses in the first place. And in the endgame, almost any army can kill you eventually, so an extra piece has to make your attack faster or slow down your opponent to be of any use.
To that end, if you're behind, you're incentivized to attack, both to create opportunities to turn the tables and win back material, and to push towards the endgame where material doesn't matter as much. The player with the advantage wants to keep the situation stable, so they can find ways to build their lead safely.
If you want some in-depth discussion on the topic, there's a translation of Tanigawa's 'Lightning Speed Endgame Technique' available online.
It really is
Ahhh, Pyre is one of my fave games of 2017 for so many reasons, and this is one of them for sure! One little thing that wasn't mentioned is that the very first time you lose, the characters that lost that Rite get a 'Valuable Lesson', in the form of a huge exp bonus - bigger than you would normally get from a standard victory I believe. This both gives players a bit of a boost now that it has confirmed that they are struggling, but it also helps convince the player to *not* restart the rite again like so many would do in this genre of game, and instead take the loss as canon - and so give the game the chance to demonstrate how losing isn't the end of all things in the process.
I also appreciate the many genres of games used as examples too, it really is helpful to see how the concept of feedback loops translates across games like that. :D Looking forward to the next Boss Keys!
An additional aspect about the Valuable Lesson mechanic that I just thought of: if the reason you lost is beacuse you decided to experiment with new, out-of-comfort-zone characters and don't do well as a result, those characters that you found it hard to play with now get a significant boost that could make them easier to play in the future as compensation.
Where can i find/play Pyre? Maybe Mark said it in the video but i missed it.
Pyre's on PS4 and PC. So check the PlayStation Store or Steam. Enjoy!
Steam works for sure, probably the consoles too.
@@daud3171 Bae is bae
Positive Feedback Loops are the bread and butter of most roguelike games, many of them do however fix this issue by a simple trick: By being short. If you get powerful early on and steamroll through the game, then there is no issue if the game is over after one or two hours and you have to start fresh again. These loops can be really exciting too, especially if you are struggling with the game and suddenly have a shot at winning.
That's also why they are an issue for Xcom, Xcom is essentially a Roguelike (without fail states), but a playthrough will take the average player several days.
I found the negative feedback loop In darkest dungeon very interesting that when a champion became high enough level they would refuse to play the easier fights because it was "below them" meaning that you couldn't power level one squad but to continue to get rewards from the lower tires you would have to use newbies also a smart choice because it forced the player to level backup champions for when the others got permanently killed this loop also didn't seem to have the effect of makeing me feel punished for playing good or rewarded for being bad
That and also you still had access to lower level missions to train up new adventurers, and the Heirloom currency helped develop the Hamlet to make it easier to train equip and recruit new(and even better) people, while the game is painfully punishing at higher tiers of dungeons and quests its never "Game Over".
Yes, exactly. And also, as far as I am not mistaken, the amount of missions you have available of each difficulty level depends on your current roster. So if you have a lot of high level heroes sitting there, you'll be getting more champion difficulty missions popping up and less apprentice ones. But if you have more rookie heroes in the roster and hardly any high lvl ones, then you'll be seeing more apprentice missions available (and hence more choice of what to do with them).
I quickly modded that and their refusal to go with an abom. I don't care what they think, I am the boss in that town lol. I also did not abuse one team, just changed something that annoyed me
I was hoping someone would bring up Darkest Dungeon. Personally I didn't like having to go back and beat the things I already did so much, because yeah, losing your units is so punishing. I didn't hate it and it could certainly be interesting to use new classes and squads, but well I certainly never finished the game.
Just a note with TF2, and that it has both! When the attacking team captures points, their spawn moves closer to the objective and pushes the team defending team back, but soon the points that need to be captured end up getting closer and closer to the defending teams base where they have less distance to travel and more map control! Great video as always, you're always expanding my shopping list :)
Hm, I've noticed systems like this in other games too. Like League of Legends, where doing well and pushing towers means the enemy now has to travel less far to engage with your team, meaning it'll take a lot longer for your team to get back to pushing than your enemy.
I wonder what these types of systems are called? I would call them Distance Feedback systems, since most of the it's simply a matter of you and your enemy getting closer.
Robin Rai Ah, finally a man of culture.
Edit: spelling mistake.
another TF2 positive feedback loop is that tf2 has random crits (yes, in a shooter game) and the amount of random crits you get increases when you do well
This actually reminds me of a really clever Negative Feedback Loop in XCOM: Long War, a community-made mood for XCOM: Enemy Unknown which makes the game far more complex and - you guessed it - long. Anyways, the mod has a Fatigue System where soldiers that go on too many missions in a row become Fatigued. Whilst it is still possible to bring them along for missions, they always come back wounded no matter how well you perform, encouraging players to let even their most elite troops take a break every now and then. This results in them trying out B Teams, C Teams, D Teams or - dare I say - E Teams, allowing members of these squads to level up, making them more capable and thus convincing you to bring them on more missions to level them up even further, advancing them from Rookie to Expendable to Least Concern to Veteran to Protect Immediately to Elite to God or somewhere in between. As a result, you start bringing them on much more missions, making them fatigued and ending the cycle. In fact, the game’s entire roster system is based around this - you have a few squads of elites, some get tired so you bring along some fresh meat, eventually turning them into elites making them fatigued making you try out more and more rookies until nearly your entire roster is made up of tired elites and you have to hire a crapton of rookies. It’s also why Long War is considered to be a vastly different game to XCOM and its subgenre. The official games are based around avoiding losses as much as you can whereas Long War is about examining the losses you will be presented with and choosing which ones to cut, making Long War stories darker and about expendability whereas offical XCOM stories more generic and about personal strife.
EDIT: speling erur
If you watch Resident Evil 4 speedruns, players deliberately exploit the dynamic difficulty by taking hits so the following areas will have fewer enemies.
Dark Souls 2 deaths actually decrease your overall health capacity. That was a tough feedback loop.
And let's not forget Demons' Souls where enemies got more numerous the more you died.
I never understood that mechanic (DeS). It would make much more sense to add stronger enemies after you win a boss fight to increase replayability of an already finished level while not punishing the players who are struggling.
Well, some positive loops allow for people that overcome larger challenges to get more power, which means that losing more means harder challenges that will make you stronger if you surpass than.
That only happens if you die as a Human tho. As a spirit, you dont get any bonuses/penalties. World tendecy was a weird system admittedly.
(my previous comment was meant as a response for Payadopa)
Anyway, Dark Souls 2 also had some negative loops in regard that enemies would only respawn a limited number of times. As a mid to bad player, you would enventually empty out places with enough patience.
Piece of advice, always keep close to the person in second place. When the blue shell comes along, pull back to second. If the shell isn't circling you yet, it'll actually target the person you just put in first place instead of you
I don't know why, but at the end I was expecting:
"...using feedback loops effectively to keep multiplayer games fair, and single player games challenging" is another invaluable asset in the game maker's toolkit.
Dunno why I was expecting it, since Mark doesn't do that, but man, the way he was saying it and the ending music just made me feel that.
what do you mean with mark does do that
@@dontspikemydrink9382 Uhhhhh.... This was 4 years ago apparently, but from what I can tell, Mark hasn't ended essays by saying his channel's name at the end, which is a quick but clever way to bring the video back home.
So like, feedback loops, difficulty scaling, color coding, they're all tools. And the channel's called Game Maker's Toolkit. Like, it just seemed a natural way to close out the video.
I could imagine a really interesting positive-negative system for XCOM where your top soldiers, upon reaching max rank, can be upgraded into NPC commanders/generals that can carry out missions on their own, and you can send rookies with them.. but they can't be used by the player anymore. One could pair this by necessarily increasing the number of missions beyond the scope of what the player is able to carry out on their own.
I do the same thing as pyre for pokemon. Retiring an MVP after important fights (gyms, rival fights and Team X's final scenario fights) was an awesome way of diversifying my team throughout the game
that's an interesting idea! Pokemon does kinda suffer from having a million critters to collect, and yet most just stick with the same six-or-so for the whole game. Maybe I should try a Nuzlocke challenge...
Mark Brown yeah a nuzlocke is one of the best ways to use pokemon you would have never used otherwise, you could try with a randomizer as well if you really want to change things up.
So basically Resident Evil is every bitter dice-and-paper GM ever
"Awesome, we slayed the dragon! Let's start splitting up the loot!"
"Well.. damnit.. Wait! Now there's zombies!"
"How many?"
"ALL OF THEM"
You mentioned there being a positive feedback loop in Splatoon in how a team with more ink coverage has more opportunities available to them, but something else in Splatoon (1 & 2) that I think is very important is how negative feedback loops are built into the rules for many of the game's ranked modes. In the King-of-the-hill-like Splat Zones mode, when a team takes control of the point away from the other team, the other team is given an amount of penalty points proportional to how long they held on to the point before losing it. These penalty points must be burned through before the team can start ticking down their timer. This is great because it means that, if a team captures the point right away, almost gets the timer to zero and then loses the point, the opposing team still has a fighting chance because the first team won't be able to just recapture the point for another 5 seconds to win. The new ranked mode in Splatoon 2, Clam Blitz (which is much harder to explain all the rules of) also has a negative feedback loop wherein a team is awarded a free "super clam" (which is carried to the other team's base to score) immediately after the opposing team scores. Kinda like hitting people back right away in Bloodborne to regain health, Clam Blitz always gives a team the chance to strike back after losing some points. Not to mention how each ranked mode includes a generous overtime that allows the losing team to pull through in one last-ditch effort. Those are just some examples, but they contribute to an overall design philosophy that makes Splatoon the single most fun and exciting multiplayer game I've played. The rules are designed so that quick comebacks are always a possibility which means that all players need to give it their all for the entire duration of the match. The losing team can never give up because they always have a decent chance of turning things around, and for the same reason, the winning team can never get complacent.
Cameron Tauxe I think one interesting part of splatoons negative feedback loop is how supers are earnt: its easier for a losing team to get supers quickly, as they have more turf to ink and gain super meter. Balanced out by the fact that they have a harder time movinf around the map. Splatoon is great imo
The stages, too, are (mostly) supermassive negative feedback loops. Most stages have the middle ground be the lowest area in the stage, and the closer you get to one teams spawn the higher up you are. So if your team falls behind you then get the higher ground to help give you an edge in turning the tide.
I've been watching you for about 2 years and watching that TOP PATRONS list grow is actually really cool. Good job Mark, you've truly deserved it!
Thank you! But it's actually becoming a problem - the sheer number of names has crashed Photoshop at times!
I think Tekken deserves a touch more discussion here. Sure, winning or losing a round doesn't directly impact how likely you are to win or lose the next round, but within the individual rounds, landing a hit will (almost always) get you frame advantage and often times wall carry, making it more likely for whoever landed the hit to land the next hit-a textbook positive feedback loop. Also of note is that many fighting games today (tekken included) have introduced negative feedback loops in "rage" systems that grant players at low health extra damage, defense, meter, or additional options in general.
Regardless, this was still a great vid! I just wanted to comment on the one thing that I noticed.
I was looking for this. Fighting, platforming action, and hack/slash have the positive feedback loop of stagger and air combos, and some have the negative feedback loop of reduced damage for longer combos or attacking downed units.
Exactly. The use of frame data in combos create some sort of short-lived, self-contained feedback loops.
Holy cats what an entertaining video! I've been rewatching your older stuff basically daily for the last week or so, and when i saw this one pop up, i stopped everything i was doing just to watch it.
Your videos are incredibly insightful and open my eyes to even more design tricks and ideas than i thought was possible.
It also warms my heart to have watched a video from 3 years ago and have seen a handful of patrons, and then watching your video today, seeing a gigantic list of patrons scrolling almost faster than I could follow. You deserve it, man. Congratulations!
Thanks for watching! Lots more to come :)
Lol same here they are so god damn good
I thought the same thing when I saw the list of patrons after older videos. And that's only "Top Patrons" which means there were more?
Armageddon is here, cats know how to read the Bible.
After watching this video a question popped up:
In CS:GO, there is an economy system for every team, for buying guns, grenades etc, necessary equipment to win the match. Since the game is round based, every round win nets you $3000 and every loss $1400 (per player).
A mechanic of the economy is that if a team is on a "losing" streak, they get more money each round, from $1400 all the way to $3400 (a _Negative_ feedback loop, since you get more for losing). But if you chain wins, the reward doesnt get bigger, its the same.
On the one hand, chaining wins helps you win the game faster, but on the other hand, the losing team can get _more_ money than the winning, and if economic damage has been done in previous rounds (players getting killed ,losing their gear), even if a team wins it can be weaker in the following round. So is this a positive or negative loop. You get the same (big) amount of money every round (~positive) but the opponent can get more eventually and be more prepared in a round (~negative).
Maybe Im just confused, but opinions are welcome.
NextDoorGuy so winning one round gives you the short term advantage next round, but going on a winning streak will eventually give the other team a bigger advantage? I like that idea of a slow curve negative feedback loop
charlie rose Yes thats the tl;dr of the system.
Essentially it evovles into a negative loop is what you're saying. Interesting. It makes sense to be a negative loop ,but I was confused with the winning rounds part since its, you know, the goal of the game lol.
Wait, if your team looses for multiple rounds in a row you get more and more money to compensate? I never new it, and always wondered how the hell CS can be balanced when there is this obvious positive feedback loop of winners having more money for arms and equipment.
7:54 These little guys to the right and left side look a lot like a badass Hollow Knight
haha yes I thought the same thing
The imps are pretty badass. Ti'zo is bae
Ti'Zo the wise, ever faithful to the cause of the Nightwings - he stayed true to the end!
How all you're videos always circle back on themselves is truly amazing, your writing skills are unparalleled.
Yeah I really like how in both defeats and victories, Pyre has a way of encouraging you. Getting XP when losing prompts you to keep going. Letting your best teammates go means moving on to strengthen overs. I like how it basically uses both loops in different situations, more the negative loop though.
I like how, similar to your dark souls examples, there's an increased incline when it comes to levelling up in shooters and how in a way it helps newcomers. Each time you level up, the next level will take that extra bit of time/XP to earn, so high level players have a steeper climb while a newcomer is jumping through the levels for a while, generally working as a confidence builder. It's not quite a balance, but it encourages play.
i think the game Massive Chalice solves the Xcom pretty well by having move incredibly quickly and have it basically become the income for the game. Over time your best troops retire or die just because of old age but instead of just losing them as Pyre does you can match your best players to create new and interesting team members.
I really liked Massive Chalice overall. The only issue with that game is that the difficulty gets a bit flat about the midpoint through the campaign. Since the entire game is about playing defense, and there are only so many enemy types they ever introduce, you're always only ever reacting on the battlefield and you're always only ever facing a limited number of enemies that you eventually learn how to counteract hyper-effectively.
There's just no good endgame to thing, ultimately. They needed a final quest to actually set off the chalice where you had to go into the void and defeat a unique boss or something and return to set off the bomb.
Real life needs more negative feedback loops. u.u
Tax the rich
Supposedly, this is the idea of welfare, the people most in need of help getting a little boost to get them back on their feet is a good thing. People, however, are not as simple in their behaviours as AI and game players: despair and misfortune drive people to drink and drugs, which are expensive and detrimental, causing a positive feedback loop of further poverty.
A compromising, socialist (some might say communist) solution potentially lies with Universal Basic Income which lets there be a baseline for people to live at with job income on top of it, Kurzgesagt did a cool video on it here
th-cam.com/video/kl39KHS07Xc/w-d-xo.html
Ooh interesting point: income tax is literally a negative feedback loop. The trouble is that people dodge it
I wouldn't want to take away the positive feedback loops. I think that people who work for it and use their money well should be rewarded. All I want is for people at the other extreme to still be able to live happy, dignified lives. UBI would be nice for this. I'm gonna watch that video.
this comment thread is surprisingly good! i was worried where it would go :P
Your video reminds me of Faster Than Light, in that it also has a positive feedback loop system with every encounter you take, so if you make too many mistakes early on, you'll never defeat the boss at the end of the game. It also has chance encounters where the right decision will boost you and the wrong one will screw you for the rest of the game. However, it felt like the game was balanced by the fact that each run, no matter how far you get, is pretty short. You're meant to die a bunch and start from the very beginning all over again until you figure out how to make it through without dying.
I really love the built in comeback systems in Splatoon 2. Almost every map (I'm looking at you Port Mackerel) is designed such that players have easy mobility on their side of the map but extremely limited mobility on the enemy's side.
Usually the maps are a bit of a grove where both spawn points are higher up than the centre so and there are a lot of cliffs and drops you you can jump down effortlessly but can't get up, forcing detours to get around and get near the enemy's spawn. Inkblot art academy has an especially ruthless one that almost seems impossible you you don't know one specific jump.
Mark,
I've noticed that in several of your videos you've mentioned speedrunning, and you seem to be familiar with it. Have you considered making a video about games which are designed with speedrunning in mind? For example, you could cover games which incorporate individual level time trials, leaderboards, replays, dedicated speedrunning modes, or even just speedrunning integrated into the level design itself. Many indie games are developed with these in mind, and some bigger games do as well. In the speedrunning event Games Done Quick, some runs even have commentary from the developers that provide insight into the design. I think it would be a very interesting topic to make a video about.
Thanks for reading, and thanks for the wonderful video!
Cheers! I might do something on speedrunning down the line, though SnomanGaming has a great vid on the subject
I haven't seen Snoman's video; I'll have to check that out. Thanks for the reply! Love your content :D
The feedback loop in XCOM is actually a selling point for me and I don't want them to get rid of it. The game has a reputation for being a brutal meatgrinder for new recruits, which is fun in its own way. Then, when your guys survive, it's satisfying to have them level up and become tougher at the same time you grow attached to them. It's the best of both worlds and helps you construct your own narrative.
The issue is the 'when your guys survive' part. The aliens never tone down their assault, meaning if your ace squad gets wiped because of some unlucky crits, or mechanics that aren't explained (muton parries for instance), then your rookies are not only stuck with limited options, but thrown against the brand-new units that have weaknesses your rookies can't exploit, likely resulting in another wipe. Then another wipe. Then another wipe.
For instance, quite early in the game, enemies with armor appear. And quite notable armor sometimes, with 3 on MECs being the most abrupt spike. Sure, grenadiers can shred the armor with a specific skill, or with their long-range grenade launchers, but if you have rookies? You're screwed. Plinking away at heavy armor while mission timers tick down. Sectoids camp far back and are weak to melee, just fine for an established Ranger with boosted movement distance from doing special operations, but for a brand new one? He can't even keep up with the rest of the squad, nevermind flank ahead to exploit the melee weakness.
Add in that soldier aim/evasiveness/armor/health/movement distance/will are all tied to essentially how long they are alive, and you run into some huge problems. While your ace squad can tank more hits, dodge shots that should have hit them, hit targets they had no right to, and reach advantageous positions like flanks or cover in a turn earlier than anyone else, you quickly hit the downward spiral if you lose them. Can't tell you how many times my rookies on ambush missions get killed by exact damage that they could have tanked if they had armor or +1 health from a mission. Or missed a shot that was really important because they don't have good aim yet, or the movement capacity to gain height advantage.
@@Winasaurus while I might be late to this, even difficult missions are possible with low-level units (I’m doing a four-man campaign at the moment and an avenger defense left all my best units tired or wounded by the assassin)
You have to shift your power from abilities to consumables and use them carefully
No shred or run and gun for flanking? Equip a war suit and a shredder gun. Use grenades to break cover and abuse the crits
Bring flashbangs and frost bombs to stall enemies and a refraction field allows concealment to make sure you’re not activating more enemies than you can handle
Always train two psi-ops and keep one in reserve
Etc.
Robot enemies just don’t matter when bluescreen and AP rounds exist
The game has plenty of recovery options: and your rookies and sergeants will level up FAST if they’re hitting above their weight class
@Duke Wang The issue is there's no point in a negative feedback loop that is intentionally crippling. Because it just means players will either restart the campaign when it happens (basically meaning it was a game over), or savescum to avoid it happening in the first place.
There is precisely 0 incentive to keep playing if your ace squad gets wiped. The missions become longer and more difficult, sometimes outright impossible. The only method to avoid game over is to abuse the resistance ring missions to stall the avatar project, and even then you're not always going to be able to take those.
I could understand seeing it as a good thing if there was some sort of unique comeback mechanic, like peril and danger in paper mario, where you ride the razors edge with a glass cannon build as a comeback mechanic, some sort of "crisis mode" on the avenger where recruits lose armor and health to gain movement distance, aim and damage (argue physical training substituted for combat drugs, armor synthesis swapped for high power ammo), then you could have a sort of "last stand", you put everything into a last punch and hope they go down before you do, creating super tense missions where you're still on the back foot but not so much you can dump a whole squads fire into a guy and deal 6 damage as it is now.
As someone who has played xcom a disgusting amount I can say it's literally not fun and a waste of time to play on after the tipping point of losing people, and it may as well just say game over try again.
Here you are providing actually insightful and new stuff for me to learn, and all I can think of during the Pyre footage is "Damn, that art is pretty."
"In chess, losing pieces can lead to fewer tactical options down the line".
Anderssen: *I'm gonna stop you right there*
"Opportunities multiply as they are seized." ---Sun Tzu
sun tzu explained positive feedback loops before they were even intentional game design choices.
This was really cool to watch. I learned a lot, thanks for all the effort it took to make this video!
Great video Mark!!
Thanks for drawing the graphs at the beginning. I think the red sine wave as a visual example is exactly what I needed to understand the negative loop.
A new GMTK video to enjoy! Yeah! Now I can watch it and then inevitably come back to see it over and over again since these videos are so well produced and thought provoking.
I remember encountering a feedback loop similar to Pyre in Mega Man Zero. The game gives you the option to skip almost every mission. On paper this sounds good. However, the game eventually reaches a point when you can no longer skip. Since i had very low skill with this game, I had skipped every mission that was skip-able. This left me in a state of screwadge. If you don't beat any missions or bosses, you don't get any power ups. Though i had skipped almost every mission, I still had played for more than 4 in-game hours. I am convinced that at that point in the game with no power ups, it is impossible to complete it, thus I needed to reset my progress.
Fantastic episode.
Fantastic analysis! This is one of my fave channels. I wish I had more friends that talked about game, art and music design. So interesting! Thanks for the upload!
I came from your “Telling stories with systems” video wanting so hard to mention Pyre’s solution to storytelling, because I found it genius! The compartmentalized format of the story allows for branches and results in a story that holds up quite well without having to merge to a central story!
oh hi mark
Ivan Nasonov that is bullshit
Ivan Nasonov I did not hit her. I did NOT!
Anyways, how's your sex life?
That should be a part of his opening
You're my favorite TH-camr!
At 11:46 I just realized that I needed a Matthewmatosis / Mark Brown collab. That's gotta happen!
Jump Smash Huh?
Matthematosis did a review of Devil May Cry in its entirety. Once I saw the Devil May Cry part of this video, I realized that the two of them should do a collaboration video together.
Thanks for the info, I gotta watch that.
Pyre is my absolute favorite game I played this year! And now thanks to this video I understand that part of why it was so great, was that the feedback loops were so masterfully intervoven with the game's narrative and setting that it didn't stand out like a gimmick; in fact those elements worked together to deliver a great experience.
Man I love how you use so many different games and different types of games to explain your thought process. It really helps sell your ideas makes it really easy to understand in context. Cheers Mark!
Ooooh dang another gmtk is out
How didn’t you mention bonus (green) exp for characters who don’t play in a match? It encourages you to use every character yet not forcing you to do so, which is neat!
Thank you for the suggested video on psychosis. That video taught a lot to me and I appreciate your awareness of many other youtubers that may never get popular. Thanks for your great videos that are continually amazing!
I'm not sure why but I've often struggled to wrap my head around the positive / negative loop terms... your explanation with the graphs just cleared it up in a flash. Thanks!
If you liked Pyre, I would very strongly recommend The Banner Saga. It replaced the ritual minigame with more traditional TRPG fare, and adds back in permadeath, but also places a much higher emphasis on the visual novel decision making portion of the game, adds a resource management element, and has a more dynamic storyline.
Sounds like TH-cam is a positive feedback loop if your video gets a lot of views it'll be recommended to more people and then it gets watched a bunch which will cause your other videos to gain views, meanwhile poor saps sitting at the bottom of the heap aren't getting views meaning YouTuube won't promote them and thus they stay at the bottom of the pile.
It's worth noting that youtube isn't a game, it's a business... It sucks for aspiring youtubers, and I can relate to wanting it to be different, but from Google's perspective, having it this way makes sense.
I remember when youtube have "video response"
It is business choice, not a rule of nature. An algorithm chooses what videos appear when you open this website, not word of mouth. You can program an algorithm to behave however you want: promoting the more popular channels even more or not, giving preference to newer creators or not, showing the option to easily switchig between trending videos or just the most recent uploads per category, or not.
Fantastic analysis and explanation. That Liberation mechanic from Pyre sounds really interesting. Thank you for the consistently great content, Mark!
I was just thinking about Shadow Of Mordor and how wonderful its feedback loops are. A big part of that is because of how the Nemesis system utilizes randomization to create both positive and negative loops.
Defeating enemy leaders earns you experience and can grant you powerful runes, but that leader may then be replaced by another dangerous enemy that you don't have any information about, or the leader may return with new strengths and immunities that it didn't have before. Frightening an enemy leader can cause it to run away, making your mission easier; that leader may appear later with new fears or vulnerabilities, or he may end up becoming stronger and could seek revenge for his humiliation at an inopportune time. If you die, available missions and events will play out without your involvement, which can cause some changes to the enemy hierarchy that are beneficial or detrimental. The game makes a point of ensuring that all of your successes and failures have consequences, and although it always rewards successes and punishes failures through guaranteed positive feedback loops, the Nemesis system incorporates numerous other positive and negative feedback loops that keep the game challenging for skilled players and fair for struggling players. It also adds a ton of variety, giving each player a unique experience.
Very interesting video. God bless 🙏🙌😇
You know why Donkey King does that? Because Donkey Kong is awesome.
Yep, bonus levels are a great, if not the best reward for completing challenges IMO ^^
And rewarding these bonus stages with the ability to play everything in a more difficult way again really blew my mind (and my sanity at some points too but whatever the game's great :D )
@@NIC040901 As someone who got every shiny gold medal, I can comfirm I am insane.
loved the production values. Your videos keep getting better and better
Happy New Year, Mark. Good to see a superb video of yours again. I didn't realise Pyre was this good in its game mechanics.
Would have been nice to have a video game specific example rather than Chess for the Positive feedback loop causing subsequent losses.
Fire Emblem has the same principle. When you lose a unit you may suffer more losses as your tactical options are reduced.
FEs have had a nice negative feedback to counter this, ever since the first game.
Pre-promotes are basically free units that aren't usually as good as the ones you spent time to train to lv20, but are still mostly viable for the endgame.
So in FE7, even if you lost your best units early, you can still beat the game with late-game pre-promotes like Athos, Pent and Louise.
Similarly, in Shadow Dragon you unlock extra chapters where you recruit new, exclusive characters if your army has gotten too small.
Pretty much every RTS works on the same principle as chess. When a unit dies you have fewer options and are more likely to lose
2 years later, wahetever:
- Tekken has no feedback loops.
- Literally a video of the player with low health having rage, which gives them more damage.
Tbf he was talking aboiut feedback loops _between_ rather than within rounds, but could've picked a better example
This was an incredible video. Thank you for explaining Pyre (a game I've been on the fence about trying) and also for making another damn good video about game design.
This channel is PURE GOLD! all the videos are intresting and full of content!
(sorry for my english if i have made mistakes)
Tekken has a feedback loop on a much smaller level, because of hitstun.
11:45 - 12:27
I can only say... The Legend of Zelda: Breath of the Wild's Trial of the Sword is exactly this. A challenging set of trials that you have to accomplish as a gauntlet, only to... upgrade the effectiveness of what's already the best item in the game.
Players who WOULD benefit from the assistance are those who struggle, and aren't likely to actually complete the Trials.
Players who don't NEED to compete the Trials because they're already so good at the game... are the ones who actually succeed in the Trials.
it's not really the best weapon in the game as you can find weapons with much higher damage in hyrule castle much more easily than completing the trails.
Can the trials be thought of as a tutorial of sort used to help train the players that struggle?
I feel it should remain as a high level challenge which is a great time and if you need to have a tutorial you can go to the easier areas in the game
Zift Ylrhavic Resfear If so, then that's the latest tutorial I've ever seen in any video game to date.
Joe McNamee Higher damage, sure. But it's still the only weapon/shield/bow that doesn't permanently break. Therefore, the most useful and most used.
I found losing against a certain team in Pyre actually kinda helped the story, when you finally battle them again in the Liberation match, you grow a sense of rivalry to them throughout the game and when that match actually comes, it made wanting to beat them even more satisfying even if that meant permanently losing a party member.
I am a new subscriber...I've sat here and binge watched your channel for hour and I just wanted to say your videos are absolutely amazing. Your content is unique, informative and endlessly informative.
I never thought I'd want to watch videos analyzing all of Zeldas dungeons or the controls of star fox but every video engages me.
A + man.
Thanks Chad!
This is why Top Trumps isn't fun. The whole game is just a massive positive feedback loop. As you get more cards, you have better cards which help you get more cards and vice versa.
Getting more cards also means you get a lot more shit cards, more often than not you can get a very good comeback in Top Trumps because you're just cycling through the 2 excellent cards that the opponent hasn't managed to get.
That's never happened to me. You might be right, though.
Yeah, in any Top Trumps game, there's a best card in each category. It's not a perfect measure of advantage, but the player with more of those aces is generally going to have the advantage, regardless of how many filler cards each player has.
For a Chess analogy, if one player has 8 pawns and their king, against an opponent with just king and queen, the player with the queen has a significant advantage (so long as they can prevent their opponent from promoting any of their pawns) - quality of pieces is more important than number.
I've never heard of Top Trumps. The way it's described here, is it the card game "War"?
Soumein Yeah, apparently: en.m.wikipedia.org/wiki/Top_Trumps The difference is that the cards have multiple values you can pick for comparison.
In Darkes Dungeon your high level characters can't do low level missions and they force you to level multiple characters because of the stress system. Why didn't you mentioned it here? :(
He talked about it there : th-cam.com/video/6RHH7M4siPM/w-d-xo.html
Welcome back! Can't wait for more throughout the year!
I think this video may be one of your very best. This one is going into my favorites. Keep up the amazing work, Mark!
I likr your
I like your videos
Thanks!
You need to sleep, buddy, you're trashed
you tried
the simple way to fix xcom exists on the steam workshop... the fatigue system in long war forces you to trade out combatants every single mission unless you wait so long between missions that youre basically losing the game. Now everyone will slowly level up together, and your best soldiers will be replaced by your other best soldiers. But the suggestion that without this the snowball effect is a bad thing is kinda silly. The game is difficult, thats not a bad thing. Real life has consequences for failure, and in life or death situations there are life and death consequences. Why is it so wrong for a game to emulate this? Why is it, that if i want to experience that sort of life or death thrill, that i shouldn't be able to get it from xcom, or darkest dungeon, or xenonauts? The perma death system is not the worst thing in XCOM 2... the quantity of timed missions, specifically ones where you have 5 turns even though you are in concealment and there is no reason the advent should know you are there is ludicrous. The timer is fine, every mission can have it even, that would be something... after i am detected. If i can run through an entire map without ever being seen i shouldnt be punished for taking one turn too long when the enemy has no idea im even there.
War of the chosen had fatigue as well as more variations to the wounded system. You should look into the expansion some time.
Thanks so much for posting this video! It motivated me to finally play Pyre about 7 months ago and now it owns my entire brain. Couldn't have asked for anything better 💜
Really loved this video. Wonderfully clear explanation with plenty of examples to back up your points. Awesome stuff mate!
since when did chees has so great graphics???
That's the recently released "Ultra" chess, it's a game that was released with the exact point of having good graphics. You can find free chess games, or you can pay for a game with great graphics like ultra chess
Cheese always had great graphics, people just make it look like yellow swiss cheese all the time
Ever since it used real-life instead of computers for rendering
Bröther, give me LØØPS
Strong start to hopefully another great year of GMTK
Great material. Thank you Mark for doing what you're doing, and I wish you you'll never have to stop!
Mark really hates xcom lol
Haha I don’t! It just has some weird design at times. But when it works, it works amazingly well.
Pyre actually frustrated the shit out of me. It feels like everything has consequences, but the game gives you no tools to predict them, when combined with the save system means you have to restart the whole game if you get your answer and you're unhappy with it.
Perfect example: you say the game promotes you to get rid if your strongest character. The game never told me that. In fact, the implication were that certain characters would have different impacts on the story and would have unresolved stories if they left purgatory which has nothing to do with the level. And what were the results? Dunno. They even humor the idea that you should be intentionally losing some of the rites. And I did. What happened? Still don't know. You don't seem to be told the consequences of anything until way later.
I had a positive feedback loop of unanswered questions. And no safety in saves to test for answers.
Exactly. You had to take the consequences as they came. Sometimes, there is no safety net. So if it frustrates you, it's only because your hand has been held by other games for years. You gotta take Pyre on its own terms. If not, then you'll have a tougher time with it. Which is fine of course. You don't have to like every game you play. But sometimes our frustration isn't born from a game's flaws. It can be born from our preferences that came from years of the same kind of game design.
It feels like you're actively avoiding the point of frustration in order to voice justifications in your preference toward difficulty. Which is a massive stretch to twist what I said to be both about and against, and is a massive disappointment if you feel the need to do that.
If I flipped 10 coins one after another and asked you to guess the exact sequence of binary outcomes and you had to blow your brains out if you failed, would you find that to be fair? Would you find yourself challenged or would you find the challenge arbitrary in its guesswork? And do you find the punishment to be reasonable, or has life been holding your hand for too long?
Because a series of choices you have no way of predicting the outcome of while at the risk of narratively ruining your save or mechanically breaking your team dynamics is pretty much that.
There is no avoiding the point of frustration when there is no point of frustration to begin with. When you use statements like "narratively ruining your save" or " mechanically breaking your team dynamics", you show a clear lack adaptation skills. You don't like the idea of taking the game on its own terms. So your own frustration comes from a preference built over years of a certain kind of game design. Pyre isn't even all that difficult, so I'm certainly not "voicing justifications in your preference toward difficulty" as you'd put it. But I do adapt to what a game is trying to do. There is a difference between a narratively/mechanically broken game and a game that marches to its own beat. You don't have to like it. But the more you talk about how much your frustrated with it, the more you prove my points.
This is one of the most useful video about game design that I ever saw. Watching it from time to time sometimes.
I am so glad a found this channel! so much insight, and all of it is so eloquently put!
Loved this video Mark! I study sciences of education and game-based learning is one of my favourite parts of this career (and probably the area where I will work once I graduate). I have learnt a lot from your videos and hearing you're going to make more on game desing theory is a joy. Keep it up!
Awesome video man! You did an incredible job explaining feedback loops, much better than the 15 books I read on the subject. Subscribed!
Enjoyed the video and the variety of games mentioned I've never heard of. I quickly thought of the weapon system in Cave Story once I started watching.
This is very important in balancing an experience. Great to hear you talking about game design like this, makes it clearer
Great video as always, all of these mechanics really struck me as feedback loops, never seeing them that way.
Even if I already know the concept it's always a blast watching your videos and seeing great example / new games I wouldn't have heard of otherwise
I learnt about feedback loops in my systems dynamics class and I would have never associated that with videogames hadn't it been for you. This makes the concept so much easier to understand! To think those pesky loops can actually be fun
This has got to be one of your best videos, thanks for everything you do!
Excellent video. Takeaways:
--You can make negative feedback loops rewarding in their own way.
-- The means for performing better should still be available even on failure.
--Multiple layers of different feedback loops can make for nuanced design and engaging player choices.
--A good reward for performing difficult tasks is more difficult "exclusive" challenges, as recognition of achievements and trust their abilities as well as extra content for those most invested.
This is very noticeable in boardgames too. There is one cardgame I love, called "The great Dalmuthi", which has the biggest positive feedback loop I have ever seen. If you lose once, it is very likely, by clever design, that you will lose again, and those who fared better in the first round, will likely continue to do so. The game is to be roleplayed on a meta-layer, the high ups become the great dalmuthis, while the losers become the great servants.
And another card game, Felix, the cat in the bag, called in english i think, has a very nice negative loop, just by player positioning around the table giving players who lost a pick-me-up and so on. clever design.
Thanks as always for the vid, pro quality :)
I'd just like to say I think the aesthetics and audio for this episode. Keep up the good work!
I love your videos and have been re-watching them all lately, thanks heaps your the one who got me into game design and its so fun.
Damn I love the editing in this video, Mark! Nicely done!
You're videos are like a little treats every time they come out.
You had me thinking about the feedback loop problem with XCOM before you even mentioned it in the video, and I didn't even know what a feedback loop was until this video. haha
Hey, this is well-timed; I was just musing on feedback loops yesterday!