ID:2091628
 
i use a very specific list of steps when developing and testing games. this is a simple list that should help you go from idea to finished game.

1. get game idea. write a very brief design doc, only for yourself, that details the things about the game you'll find convenient to remember(such as a control scheme, and a step by step way for you to get a playable demo)

DON'T TELL ANYONE ABOUT YOUR GAME AT ALL. KEEP IT A SECRET.

2. tell one trusted homie about your game. for me it's someone like ter13 who isn't afraid at all to tell me if the game sucks, but is also able to see the bigger picture of a demo that might have no art assets. changes you make during this phase should be like wide, sweeping brush strokes. Your game idea sucks. the game itself is what is going to be good, and the game will be the result of tons of feedback. this is the first step to getting feedback.

3. after your trusted homie is cool with the game, release it to like 3 or 4 other trusted peeps. they don't have to be as critical- in fact they can just be those d. rider fans who love everything you make unconditionally. the purpose of this phase is to just make sure the overall idea of your game works at a slightly larger scale, and you'll also get more subtle hints and cues about feedback.

4. at this point you should have something that you'd be comfortable releasing in like, an obscure forum post and you wouldn't be worried about people completely ripping the game apart. the game is probably functional by now, and if you've taken any feedback, at least SLIGHTLY fun. so go ahead. release your demo to the tiny portion of this community that will test it. i'll probably test it. but i'll test anything made with byond that isn't a log punching simulator

5. now you're ready for actual content development. this is where you toss on the meat of your game now that the skeleton is firm enough to hold it. so do that. everything that you do up to this point is laying down the foundation for different things, and at this point you're now able to raise the amount of content on top of all those foundations. for example, you'll have made sure your inventory system is all good and fun and functional, so now you can add a bunch of different items.

6. now that there's meat on your game keep testing it with an audience of around 10 or so people. they'll probably best be fans of your game by now, so that they stick around and are excited to see new additions. trust me when i say that it's very motivation to have people who are excited to see something you made grow. continue in a cycle of testing and developing.

every single time you add something new to the game, test every feature level or area. don't only test the most recent thing you've made. this is a VERY BAD way to go about it and a lot of newbie developers fall into this trap. also, avoid admin or console commands as much as possible. in order to know for yourself that the player experience is fun, balanced, and plays the way you want it to, experience the full game as a player every time you test. if you can't handle playing through the full game every test, perhaps something is wrong with your game.

7. release the game officially. you've got a meaty game that has seen many revisions and has a small handful of supporters to help you test.

8. repeat step 6 a lot and then greenlight the game :)
but i'll test anything made with byond that isn't a log punching simulator

Lol'd.

Solid post though. I've kept up with your other posts so this is more of a reiteration of those for me. Nonetheless, it has proven effective so far, though I'm doing things slightly different.

Steps 5 plus, here I come.
it's specifically the way i do things, and everyone has their own methods. the reason i keep posting this same method though is because i've found it to be really effective; prior to following this i had a lot of trouble releasing games
who isn't afraid at all to tell me if the game sucks, but is also able to see the bigger picture of a demo that might have no art assets.

Bigger picture is the operative word here. A lot of the feedback you get on this website from people is what I call "fluff". "Make it jiggle a bit". "I can escape the bounds in this room". "There's a minor typo in this tooltip.". This is late-stage, worthless feedback that any idiot could come up with.

What you want in early stage testing is feedback on the flow and feel of the mechanics of the game, not little visual/presentation tweaks. Bigger picture is important, because if you are like me, you get bogged down in trivial 1-pixel details all the time. If all of your feedback is on matters that don't affect hard gameplay, you might be sinking hundreds of hours into something that's just not fun.
In response to Ter13
if you are like me, you get bogged down in trivial 1-pixel details all the time.

<----- Has seen ter work on one icon for a good 2 or so hours can attest to this.