The best game designers are non-casual gamers. Programmer, writer, artist - it all doesn't matter. People who play games every day since childhood will always be the best game designers, since they will learn on others devs' mistakes. Munchkins/powergamers will be the best in balancing and combat mechanics design, people who played every RPG in the world will be the best quest-designers, etc.

Also game design is very broad definition.

For RPG it's quests, choices and consequences and non-boring and non-repetative combat - I'm yet to see an RPG with this, 100% of current RPGs in our universe has idiotic combat made by some game-designer-wannabes with one or two "the best" character build and the same universal tactic which will work in every situation, even the most modern RPGs, who claim to be balanced and diverse, have this problem. EVERY party-based RPG in the world suffer from the same game design mistake made over and over and over by amateurs from AAA studios or indie game devs alike. Just make a party of ranged damage dealers and focus fire / AoE fire everyone to death. Works every time. From DnD-based games on infinity engine to Might and Magic series or even NWN2 or Fallout Tactics or Divinity Original Sin or Wizardry or Pillars of Eternity, Wasteland, whatever else I don't remember, every party-based RPG has this problem, one universal tactic which will work on highest difficulty with the very first playthrough without game knowledge, I guarantee.

For RTS it's unit composition balance, timing, etc.

For 4X strategy it also includes AI. 4X strategies are usually played single player due to the amount of time you need to dedicate to the game. If an AI is an idiot, then the game will be piece of trash. And one of the ways to have good AI is to design the game itself for AI, rather than player. Because player can adapt. AI in 4X strategy games is usually cheaply made cheater, because this game genre is too complicated for good algorithm, because devs make it for humans. It can be considered as a game design mistake, unless they know how to make really good AI who can compete with human.

You need experience with those games as a gamer yourself. This is what makes you a good game designer.

Programmers or writers - it doesn't matter at all, it's completely different skills which are unrelated to game design. A person, who plays one particular game genre whole his life will be the best game designer for this particular game genre. Be he a writer or a programmer or none of it. Because he'll learn from others' mistake and see what actually works and what doesn't. And what makes this particular genre fun. He doesn't need to analyze anything or listen to feedback, because he's the gamer - he knows it.

I don't think programmers are inherently better at game design than others.

Game design is definitely it's own skill set independent from programming.

There's some overlap, especially with games that lean heavy on the numbers and systems side of gameplay, but really there's not any magic connection between being a good programmer and being a good game designer.

Of course, historically many game designers (and lots of good ones at that) have been programmers but I think that's just because up until very recently programmers have been the ones in the best position to realize their designs.

And to  邪悪's point about players making good designers, some of the best designers have also come out of QA where their experience playing games all day looking for faults has honed their design sense.  Again though, QA is a different skill set from game design, there's overlap, but being good at one does not necessarily make one good at the other.

> sometimes I'm just like "what is wrong with this game is design, but its like I can't put my finger on exactly why it sucks.

Definitely know that feeling, had it with my own work more than once! ;)
Actually, in my experience identifying why a game doesn't work is one of the toughest challenges in game development.   Of course, when a game's not working, anyone can rattle off the 50 things that are all wrong and they hate about it, but identifying the one or two things that actually need to be changed to make the game work?  That's a much harder skill to come by.

i agree, playing games would give an advantage towards game design, but at the same time I think player expectations could tend to learn towards what is bad game design. If your playing a platformer for example and played lots of them you would generally have an expectation on what they feel like, or what you think they should play like because of previous experiances, however should a programmer try something different in that genre and make a few changes or just 'break the Mold' in some way, this could be seen as bad game design, because it's not what you would expect, but actually it's not that bad at all. What's confusing is when trying different things with games, The ' ingredients' which we have come to learn what makes up a genre get 'muddled' or actually missing/replaced by something else, and the identity can get a bit lost, so can give that "can't put my finger on why it sucks" feeling.:) this could be from any direction of the programmer,game designer or indeed a experianced gamer. 

I recently played a game that was stunning looking, sounded great, good story line which claims it's a platformer with generated levels. It was designed well in my opinion and for all the lore and ingenuous mechanics it had it lacked any feeling of 'achievement' or that sense that I had beaten it's challenging nature, which for me platformers must have. This was lost because of its randomly generated platform Levels. I could say this was a design flaw but I think it's more my expectation on what a platformer is. Bad design I think will only depend on the players opinion of the playing experiance or expectations. What works for some may not work for others. :) but you got to try right.:)

thats my OP, God I waffle on a bit.

This is an interesting topic, i think that in FOSS games what you call good game design crashes agaisnt the will of the creator to make something they will enjoy themselves, but maybe others find boring; while with commercial games its a product intended to be sold as much as possible so you get bucketloads of the same everywhere, they dont risk at all, they stick with things which sale good.

Look at all the recent AAA, they are mostly just another entry in a franquise (gears of war 5, call of Duty XX, Fifa something) all of them made by pros, but that doesnt make them good or fun. So i think we are into a good era since indie teams can innovate and create real jewels, they risk to break the mold, many crash and burn, but the lucky ones get great success (stardew valley, undertale, Isaac).

Getting back on topic, a programmer will have a much easier time at delivering a more refined product, in a faster period but that wont make it better or fun. While an artist can take years and deliver something good because of a unique feel(Ex. Undertale, creator is a musician); that feel mainstream has lost. But it all boils down to the core ideas, some are bound to fail from the start, others are hated by a good portion of players (gachas and energy that plague cell phones), rogue-likes (which i truly depise) or are overused. Or that ideas are good but need more refining, QA is great for that, sadly FOSS developers cant afford/dont get testers/dont care about imput for their games.

To resume, i dont think a programmer has more chances at making a fun game (better coding), nor an artist (unique art or music), being a game addict can help make it fell better to other gamers (but more of the same). Its the refined core idea, aimed at one player group and the combination of all the previous.

I like this discussion. :)

Jastiv "what is wrong with this game is design, but its like I can't put my finger on exactly why it sucks."

Indeed. Same thought all the time. It bugs me so much I usually end up doing in-depth analysis of a game I don't even really enjoy that much. Ugh! Fortunately, I often come to a satisfying conclusion about what parts of the design are bad and how I might go about fixing them in my own games. No idea if my conclusions would actually be effective though. At least not until I publish my games and see how the players feel about them. :/

邪悪 "Programmer, writer, artist - it all doesn't matter. People who play games every day since childhood will always be the best game designers"

I agree that being a gamer is more relevant to becoming a good game designer than being a programmer is. Though I'd say there is still a big gap between "gamer who wants to be a game designer" and "actually decent game designer".

"Just make a party of ranged damage dealers and focus fire / AoE fire everyone to death. Works every time."

I have seen this a lot, but I have played a lot of games where this is definitely a losing strategy. Most of the single-digit Final Fantasies allow you to spec out your entire party as ranged/AoE, but it tends to be a drawback more than an advantage in my opinion. Although DCSS is not a party-based RPG, I challenge you to beat the game as a ranger or AoE mage. Not because the ranged classes are under-powered, but because the game is IMHO properly balanced across multiple play styles. Perhaps that is the key here; If every player can blow through even the toughest enemies with ease using the ranged/AoE party, sure the game is poorly designed. But if some players can, yet others find that strategy undesireable, then I would say the game is well designed; it caters to different play styles in different ways.

You mentioned Might and Magic. Reading advice from other pro gamers on M&M6/7, the conventional wisdom is to avoid playing as druids at all costs; "they're worthless". However, when I tried a party of all druids, they quickly became unbelievably overpowered. Far more so than any other party combination I've heard of. With 4x more mana than health, I was able to do some rediculous things like casting Inferno more than an archer would shoot arrows. (Inferno being rather costly top tier spell) I could clear out entire end-game dungeons without running out of mana. Evil didn't stand a chance. Yes, that does fall into the "ranged damage dealers and AoE fire everyone to death" tactic, but my point is -in this case- it is a path everyone else considered the most difficult. Maybe some of the games are broken, but others are designed to have "different strokes for different folks". :)

withthelove "... identifying why a game doesn't work is one of the toughest challenges in game development.   Of course, when a game's not working, anyone can rattle off the 50 things that are all wrong and they hate about it, but identifying the one or two things that actually need to be changed to make the game work?  That's a much harder skill to come by."

Yes. This. It is easy to say "this is a poor design. I would have made it better." Harder to determine how it could be better without breaking some other game mechanic. Harder still to actually implement such an 'improved mechanic' and have your own players like it more than the other game's version of it.

chasersgaming "...This was lost because of its randomly generated platform Levels."

I see this a lot, actually. Sometimes the unlimited chaotic potential of RNG is more fun for the game dev than it is for the player. As Jastiv hints at, no randomly generated content is ever going be better crafted than what a manually designed level can be. On the other hand, even the most amazingly hand-crafted levels are not going to still be exciting and new after the 4th playthrough. Random levels have the potential to be novel and exciting on repeat playthroughs, but they can still get old pretty fast if the RNG is bland.

Jastiv: "...how long that will take depends mostly on how much variety there is to begin with."

Yes, though I would add that it's not just about component versatility. The rules of the RNG have to be carefully crafted and heavily restricted so that any randomly crafted level is still interesting and exciting. Just throwing chaos at a game without restrition results in levels that aren't just uninteresting, they're unplayable. Platforms that appear too far apart to be reached, Exits embedded behind impassable walls, etc. Obviously rules have to be applied that prevent this, but rules also need to be applied that prevent bland level generation.

I think a lot of game devs use RNG to save time and effort, but a properly designed random level generator takes more effort and more time than 10 carefully hand-crafted levels do. Here are 6 different randomly generated mazes:

They are all very different both in their style and the solutions to navigate through them. Yet the fact remains they all feel same-y to the player. There is nothing in them that inspires wonder and exploration. Sure, the solution is different for all 6, but the player experience is the same boring trudge they did the first time through. This is not an argument against RNG. I love games with RNG levels. But they'd better have really well-designed generation rules.

You mentioned Might and Magic. Reading advice from other pro gamers on M&M6/7, the conventional wisdom is to avoid playing as druids at all costs; "they're worthless". However, when I tried a party of all druids, they quickly became unbelievably overpowered. Far more so than any other party combination I've heard of. With 4x more mana than health, I was able to do some rediculous things like casting Inferno more than an archer would shoot arrows. (Inferno being rather costly top tier spell) I could clear out entire end-game dungeons without running out of mana. Evil didn't stand a chance. Yes, that does fall into the "ranged damage dealers and AoE fire everyone to death" tactic, but my point is -in this case- it is a path everyone else considered the most difficult. Maybe some of the games are broken, but others are designed to have "different strokes for different folks".

Either you're kidding / using sarcasm or they were kidding/trolling/using sarcasm, because x4 druids is THE STRONGEST party in MM6-7. Second place is x4 sorcerers. Some will say that sorcerers are better than druids, because they have access to dark magic, which is OP, but dark is overkill and endgame, which means 95% of the game you won't use it and at end game you no longer need it lol, because you'll instakill everyone with insanely high fire/water skill using fireballs and acid something, thus x4 druids is the best combination as they also have access to body magic for healing, what makes game a little easier (enemies will usually die instantly, but you may sill be damaged by traps). With x4 sorcerers you have to sleep to regenerate health after every trapped chest, until you can open them with Telekenesis spell (which completely destroys Thief class, seriously... the best trap disarmers in MM6-7 are sorcerers and druids... who are also the best damage dealers... and they can also cast fly spell and teleport... "balance"), with x4 druids you also have a party of healers, so it's just makes game a little more comfortable.

Also "pro" gamers from forums aren't always pros :D

I always laugh at NMA forums when people talk about Fallout1/2 builds there. There're a lot of fallout veterans, who completed fallout like 10000 times and they still don't know anything about game mechanics. They say how they like to play snipers and then they say... "i don't rely on luck, so I see it as dump stat". *facepalm*. Luck in Fallout1/2/Tactics is the key stat. It affects aimed shot critical chance - the whole reason to play sniper! Character sheet doesn't show the formula, it just shows that 1 point of luck gives +1% crit chance, but it doesn't show the formula for aimed attacks. I don't remember the formula for each body part (for torso it's just your crit chance, for everything else the formula involves luck stat AND then crit chance is added on top), but I remember that with 10 LK you have 70% (10% base crit chance for each point of luck +60% from formula) crit chance in the eyes (85% with x3 More criticals perk, 95% if Finesse trait is also taken), with 2 LK (1 + Gifted trait) something like 30-40%. Plus the perk "Sniper" makes [LK*10]% roll for crit, with 10 LK each your ranged attack becomes critical, what makes it the best perk in the game and makes More Criticals perk and Finesse trait obsolete, as it's a separate roll. Ironically Sniper perk is almost useless for snipers (as they already have at least 70% crit in the eyes), it's OP for bursters. x2 Bonus Ranged Damage perks, Better Criticals and Sniper perk will oneshot everything in the game in point blank burst, since every burst will be critical and deal insane damage. Luck is the most important stat in Fallout1/2/Tactics, even more important than Agility, as it also a stat requirement for the best perks in the game and also affects rare random encounters (alien blaster, the best non-bursting weapon in Fallout1 is attainable only with high LK) and most "pro" gamers see it as dump stat, because "they don't like to rely on luck" lol.

Don't read forums for character build tips, study game mechanics yourself :D Most people who post on forums don't even know what they're posting. If we talk about pure efficiency, then MM6-7 has only one best party composition - x4 druids. Everything else is for roleplay/fun/challenge/etc. So in truth you were playing the game the easiest possible way with x4 druids contrary to forum pros.

Good designers are people who can collaborate in a team of people and accept constructive criticism.  In my experience, that's less common among programmers than other disciplines.

Its true that programmers tend to be more close minded in regards of what they do. Sometimes all you get when pointing out something is "im doing it this way because i want to"

Level designers - a discipline that requires both artistic and programming knowledge - seem to be the ones most likely to transition into the production/direction/leadership gigs.

In my experience, programmers seem to silo a bit much to be effective game designers.

I think that in FOSS games what you call good game design crashes agaisnt the will of the creator to make something they will enjoy themselves

I am actually a professional game programmer. And a *good* designer is definitely better at game design that a coder!

In the same way I read a fair bit, so I am pretty sure I could write a book. I know most of the "rules" of English, I know what type of story I like.

However I couldn't write a *good* book. Just having consumed lots of games/books/films doesn't mean you could design a better one.

Coding & Art & Design are different diciplines, and while most people can have a go at the others most will be only good at one. Some people are good at a couple, and some very lucky people can do them all.

^Agreed. I think at first glance many assume games are created primarily by Coders, Artists, and Audio engineers. (In reality there are obviously many subsets to these, but I am intentionally oversimplifying) but Designer is ommited.

I feel this is due to Game Design being more susceptible to the Dunning-Kruger effect. If you're a bad coder, the game just won't run well. If you're a bad Artist, it's pretty obvious the graphics clash. But if you're a bad game designer, it always seems like a great idea to the person who had it. It isn't until other objective players test out the game that you'll hear (constructive?) criticism on the design... which necessarily only happens later in the project.

Designer is the job everyone wants. Or it is the job many feel anyone can do? Not sure. Even I'm guilty of this; My team doesn't actually have an official designer position. All members are encouraged to suggest designs. Each department has a Lead (lead programmer, lead graphics, lead music, etc.) that acts as a design filter and weeds out the bad ideas... plus a Producer that runs the whole project from a big-picture perspective and ensures it's all cohesive.

We did have a Designer position, but so many people applied, and they were all pretty terrible IMHO. Again, maybe "My Honest Opinion" is skewed by the same factor; I think they're terrible designers, but it's acutally me that sucks at it. :P

Design is definitely one of the most important aspects of a game, right up there with coding, art, audio, and writing (this last point depends on the genre, for example it's important for RPGs but not for Tetris).

The problem with design is that you can't really study it as a career in the same way as you can study the other disciplines mentioned. As far as I know, you can't get an M.A. in game design, let alone a PhD. But even if we think of education outside of universities, like for example ateliers in the case of the arts, there is nothing comparable for the case of game design.

It's difficult to find professionals when there is no formal education to be had, be it a universities or otherwise.

I see 2 options for the aspiring "idea-guys" (exclusively designers):-
Either also be the cash-guy and fund the project - or have enough coding skills to do playable mockups (even just something like Clickteam Fusion) AND also be a decent concept artist and writer.

Ideally, it takes multiple people with relatively dedicated skills to make a game and each one of those team members are designers in their own right.  In the AAA industry, someone with the title of designer has worked their way up over years and has dedicated skills with a list of published titles under their belt.  For indies, that's impossible, everyone must pull their weight.

I am of the opinion that gamers are tend to design games more better rather than someone who knows to code.

As a programmer, I'd say they are not! Design is a separate skill from implementation, and regardless of the skill the programmer has at coding, their skill at game design can be quite variable. This is not even getting into the fact that not all programmers even know how to code games since some of them don't really focus on that area of coding.

However, when it comes to indie games, programmers (and game programmers especially) may be more willing to design games that require more complicated or customized game mechanics because they might tend to have more confidence and skill to implement and debug those mechanics. I'd imagine that likewise a skilled artist would be more willing to design a game that leans heavily on its art assets to create its appeal. And a music composer might be more willing to design a game that uses its sound assets to best effect or in some innovative way.

In the end, I think anyone can be a good game designer if they have some experience playing games and can deconstruct for themselves the mechanics that make those games enjoyable for themselves and others. But in order to implement the game, they're going to need to have other skills or enlist the help of others with those skills.