I think we need to say there are two different types of playtest and I will use Spodley to illustrate my point, and my own games actually.

Stitch was very much a game in early development and it was the basic ideas behind it that I wanted to test. As such i didn't mind discussion or changing things radically that obviously were not working. I will classify these as 'early playtests'.

Reel adventures on the other hand had quite a lot of playtesting prior to spoley and I am approaching the later stages of development with it and beginnging to think about publication. Although the playtest threw up some accesibility problems, and a need for simplification, it ran fairly smoothly with little interruption. This is how I wanted it and it allowed me to get the feedback I needed. I will classify these as 'late playtests'.

So basically what I am saying is that there are 'early playtests' that are to try out ideas and see if the initial concept works. 'Late playtests' are to see if that concept, after many 'interim playtests' to polish a concept into a full mechanic.

Like I said in the mechanic design document I put up a while back, the length of time involved in taking an idea, turning it into a mechanic and testing it can be a few seconds to a few months. It is a matter of gut instinct to know when it is ready.

These different types of playtest could refer to individual mechanics, parts of the system or the system as a whole depending on how you go about your game design.

Maybe in future, we can specify the kind of playtest we are running to avoid any possible confusions. I know I make very clear before any playtest I run the state of the game I am about to inflict on my players!

Cheers

Iain

Lead Developer Mob Justice RPG

Check out my home brew games like 'Reel Adventures'

Except my experiences are the opposite of yours!

My Memories & Madness game was the first game I'd run of it, so definitely an early playtest. But it ran smoothly and any problems we had we brought up at the end.

My Six Bullets playtest was a much later playtest, and when we hit problems we tore it apart mid game (not entirely, just slightly) and patched it then and there. Then we playtested the rules patch for the duration of the game.

I guess that just underlines how different things work for different people, but how making it explicit what you want from a playtest upfront is essential.

Yeah what we should do is say 'This playtest is an early one so feel free to bring up stuff as we go along' or the opposite if it is a late playtest.

I think the designer always has caveat to turn a late playtest into an early if he thinks things are going badly.

Cheers

Iain

Lead Developer Mob Justice RPG

Check out my home brew games like 'Reel Adventures'

I guess it's the question of "what is a playtest?", really.

I figure that with procedures and mechanics I like to figure that out piece-by-piece carefully out of game. I run through mechanical steps in my head, or on paper, but I don't like finding that stuff during a game.

Sometimes my mechanics drift as I forget a rule, sure, but I don't like to debate rules in game, really. Afterwards we can see how it went (even if it went terribly).

Making mechanics on the hoof is not something that I like to put a lot of faith in. Maybe they can get you over a sticking point, but for me they then need to be evaluated further.

I guess I've seen people put rules in for a reason and then junk them at the first point of contact. That doesn't really work for me.

I would say that chucking rules 'at first point of contact' is an entirely valid way to test a mechanic. All the design and guess work in your hand means didly squat when a mechanic actually comes into contact with the players. It could be interpreted differently, cause them to do something you didn't expect or just plain not work.

In my experience, and this may just be the way I design or an indication I need to think things through more, it is very rare to design a game and have it not be incredibly clunky the first time you take it out for a test drive.

Laying out before hand that the game you are about to test is in early stages would avoid anyone being annoyed by the stop start nature those playtests tend to have.

Cheers

Iain

Lead Developer Mob Justice RPG

Check out my home brew games like 'Reel Adventures'

I think you need to know what you are playtesting before you start. Is it the whole system? Is it one specific mechanic?

A while back I ran a one-off of Atlantis for some friends. The point of the game was to have fun, but I'd also been tinkering with the rules to change them from what they had been when I'd last run it (around 10 years ago) so it was a semi-playtest as well. In that game when things went wrong we just changed or ignored them and got on with the fun, because the game was there to be played not examined.

On the other hand there was a particular mechanic that I knew in advance I wasn't certain about, and for that mechanic alone I prepared an alternative, one which I hoped might work if the first one didn't. When that mechanic did indeed fail mid-game we simply switched it for the other one and went on. But we could only do that because I knew in advance what aspects of the system particularly needed testing and what didn't.

I'm guessing I'm saying a Playtest, certainly a later one, needs just as much preparation as a normal game session. In most sessions the GM doesn't go in with no notion of the plot, or the NPCs, or what to do if things don't go as intended, and a playtest needs just as much preparation. You need to know what you are going to test, what you'll do if the rules don't work and so on.

Going back to when I playtested for L5R we played two sorts of games with the cards under test. Sometimes we'd just stick them into an existing deck and see what they did in play, that was the undirected sort of playtest, just getting a feel for things. Much more often though we would look at the cards first, identify what we thought was a potential problem or abuse, and then build decks specifically designed to abuse that and see how they did. They were not just random games, they were games with a specific purpose where we knew in advance what we were testing, and would have alternatives prepared to switch to mid-game if the cards really were broken.

I think an RPG playtest should be just as focussed.

Solipsist RPG, on its way ... eventually

I think the key thing coming through these replies is communicating what you want to achieve with your playtest and how you want to go about it.

I also think once you have your mechanics working you need to just play the game as, you know, play. Because, hopefully, that's what people playing the game out there will be doing.

As Nickelback said: are we having fun yet?