Scott Crabtree
United States
Davis
California
flag msg tools
Avatar
mbmbmbmbmb
Poll: How Many Components to Print For Final Product?
I am told producers evaluate games in part based on the number of components (i.e. cost to produce). When it comes to components that are not necessarily played every game or can be added and removed during game play, how do you decide how many components to include?
The maximum number of components that could theoretically be played.
The highest number of components actually used in game testing.
The average number of components used in game testing plus 1 or 2.
Some other process (presumably... magic)?
6. Always 6.
      8 answers
Poll created by Jaundice101
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jeremy Lennert
United States
California
flag msg tools
designer
Avatar
mbmbmbmbmb
This is basically a question of how much it costs versus how much benefit it provides.

For instance, in For the Crown you could theoretically have something like 36 queens in play (more with expansions), but almost all games are going to have 4 or less, so that's what we included. Including 32 extras would have been a huge cost (especially when you multiply that by ~20 types of pieces) with only a trivial benefit (arguably a negative benefit when you consider weight and storage).

On the other hand, Gem Rush actually has slightly more point tokens than are strictly necessary, so that you won't have to change denominations back and forth as much. It was a small additional cost (only a few tokens; didn't require an additional counter sheet) and it makes the game noticeably more convenient in high-player-count games.

If you just want a number, I'd say typically aim to provide enough pieces for about 99.9% of games (that is, one game in a thousand will need more pieces than you included). But I think that's an overly simplistic view.
6 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Adrian Schmidt
Sweden
Malmö
flag msg tools
Avatar
mbmbmbmbmb
Antistone wrote:
This is basically a question of how much it costs versus how much benefit it provides.

For instance, in For the Crown you could theoretically have something like 36 queens in play (more with expansions), but almost all games are going to have 4 or less, so that's what we included. Including 32 extras would have been a huge cost (especially when you multiply that by ~20 types of pieces) with only a trivial benefit (arguably a negative benefit when you consider weight and storage).

On the other hand, Gem Rush actually has slightly more point tokens than are strictly necessary, so that you won't have to change denominations back and forth as much. It was a small additional cost (only a few tokens; didn't require an additional counter sheet) and it makes the game noticeably more convenient in high-player-count games.

If you just want a number, I'd say typically aim to provide enough pieces for about 99.9% of games (that is, one game in a thousand will need more pieces than you included). But I think that's an overly simplistic view.


+1

"The average number of components used in game testing plus 1 or 2" is clearly not a good measure to go by. If you ignore the "plus 1 or 2"-part for now, "the average number of components used in game testing" means that 50 % of games required more components (unless you used the exact same number every time, but then you probably wouldn't have asked in the first place).

Adding a static number like 1 or 2 makes it no better. If between 1 and 3 pieces were used in every game, the average is 2. Adding 2 more would mean that you ship a piece that is never used. But if between 50 and 100 pieces was used every game, the average is 75. Adding 2 extra would make no significant dent in that 50 % of games that will be lacking pieces.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Scott Crabtree
United States
Davis
California
flag msg tools
Avatar
mbmbmbmbmb
Ha ha. The answers were a bit tongue in cheek. I thought that would be more obvious -- I wasn't exactly going for subtlety -- but regardless the whole point of the poll was to stimulate the discussion of how many components is enough. A number that would be sufficient for 99% of game play, as suggested by Antistone is, I believe, essentially equivalent to "the highest number of components actually used in game testing" as unless you've game tested hundreds of games it's unlikely there will be a difference in those two numbers.

Getting away from statistics and pedantry, how do you decide how many components you are going to include in a final mock up SpecularRain?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jeremy Lennert
United States
California
flag msg tools
designer
Avatar
mbmbmbmbmb
Jaundice101 wrote:
A number that would be sufficient for 99% of game play, as suggested by Antistone is, I believe, essentially equivalent to "the highest number of components actually used in game testing" as unless you've game tested hundreds of games it's unlikely there will be a difference in those two numbers.

I said 99.9%, not 99%. One in a hundred, one in a thousand, and one in a million might all seem like minuscule numbers, but there's a noticeable difference between them when you're selling thousands of copies of your game.

And it obviously depends on the game and your playtesting strategy, but it's entirely plausible that number would be higher than the maximum you actually see in playtesting.

I am assuming that you understand your game well enough to make reasonable extrapolations. If you literally have NO insight into your game and can't go by anything other than playtesting data, then I suppose you'd probably have to go with the max amount you see (but then, "the theoretical maximum" wouldn't be an option, either).


Also depending on your game, testing it hundreds of times might be entirely reasonable. You should likely be testing it for hundreds of hours (if not more), although a lot of those tests won't be on the final version.
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Adrian Schmidt
Sweden
Malmö
flag msg tools
Avatar
mbmbmbmbmb
Antistone wrote:
Jaundice101 wrote:
A number that would be sufficient for 99% of game play, as suggested by Antistone is, I believe, essentially equivalent to "the highest number of components actually used in game testing" as unless you've game tested hundreds of games it's unlikely there will be a difference in those two numbers.

I said 99.9%, not 99%. One in a hundred, one in a thousand, and one in a million might all seem like minuscule numbers, but there's a noticeable difference between them when you're selling thousands of copies of your game.

[…]

Also depending on your game, testing it hundreds of times might be entirely reasonable. You should likely be testing it for hundreds of hours (if not more), although a lot of those tests won't be on the final version.


This.

I'm no game designer, at least not yet, but my gut-feeling is that if you haven't tested your game hundreds of times, you have no business printing it in the first place. But maybe that gut-feeling is unrealistic?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jeremy Lennert
United States
California
flag msg tools
designer
Avatar
mbmbmbmbmb
SpecularRain wrote:
I'm no game designer, at least not yet, but my gut-feeling is that if you haven't tested your game hundreds of times, you have no business printing it in the first place. But maybe that gut-feeling is unrealistic?

Well, some games are just incredibly long. For instance, I doubt that Descent: The Road to Legend had 100 playtests of the full campaign. Pandemic Legacy: Season 1 might have--Pandemic is famous enough that they could probably have recruited 100 independent playtest groups if they really wanted--but I doubt it.

But for a shorter game like 7 Wonders or Codenames, I'd expect hundreds of playtests. And Donald X has told stories about how he and his friends playtested Dominion for years.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Carel Teijgeler
Netherlands
Vlaardingen
flag msg tools
Avatar
mbmbmbmbmb
Antistone wrote:
But for a shorter game like 7 Wonders or Codenames, I'd expect hundreds of playtests. And Donald X has told stories about how he and his friends playtested Dominion for years.

WHAT! Has 7 Wonders been play tested? Who were those play testers? At least they lack critical mass.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jeremy Lennert
United States
California
flag msg tools
designer
Avatar
mbmbmbmbmb
anijunk wrote:
Antistone wrote:
But for a shorter game like 7 Wonders or Codenames, I'd expect hundreds of playtests. And Donald X has told stories about how he and his friends playtested Dominion for years.

WHAT! Has 7 Wonders been play tested? Who were those play testers? At least they lack critical mass.

I'm not quite sure what you're trying to say, but to answer your direct question, here's the playtesters block from the credits of a 7 Wonders rules PDF linked on the BGG wiki:
Quote:
PLAYTESTERS :
Mikaël Bach, Françoise Sengissen, Matthieu Houssais, Michaël Ber-
trand, Mathias Guillaud, Dominique Figuet, Jenny Godard, Maïa Houssais, Florian
Grenier, Bruno Goube, Julie Politano, Bruno Cathala, Ludovic Maublanc, Milou,
Fred, Cyberfab, Mimi, Thomas Cauet, Yves Phaneuf, the members of the « Jeux
en Société » club of Grenoble, the members of the Dragons Nocturnes playnight
meetings, the Belgo-Ludiques 2010 players, the participants of the « Offs » of the
Cannes FIJ, the players from the Gathering of Friends 2010, the players from the
ludopathique meetups, the play-sitors of the Toulouse festival.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Martin Windischer
Austria
Innsbruck
flag msg tools
designer
Avatar
mbmbmbmbmb
The highest number actually used in playtesting plus 10-20%. If the deviation in the playtests is high I use bigger percentages.
And normally I round this number up to the next one easy to produce for me (or easy to count for wooden components). If I have to cut 23 square counters I will arrange them in a 4x6 rectangle anyway and the 24th counter is produced without any additional work.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Scott Crabtree
United States
Davis
California
flag msg tools
Avatar
mbmbmbmbmb
Thanks for the answers.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Johnathan Ness
United States
Texas
flag msg tools
I would say it depends on your type of game. For example, nobody (in their right mind, anyway) has ever played Apples to Apples long enough to go through all the cards in one sitting. I've created a game where a couple of the decks have more possibilities than have been used so far in any game to keep the randomness of the cards higher. Worker placement games, on the other hand, have little reason for more pieces than could theoretically be played, except, as mentioned above, to prevent changing them in often.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.