Tuesday, January 26, 2010

TRC XNA Frustration

Bounced frustration

I have been in this games creation industry for many a long year. I have seen machines come and go. But one thing that i have seen turn up (Sony started it!) is the TRCs and with the coming of the TRCs we have had to do some bending over backwards to get some games out.

Now TRCs are not bad in themsleves. It is nice to have some semblnce of order and design. The TRC that states the screen cannot remain black for more than a few seconds at a time is excellent. There were many games in the past that had crashed, and you couldn't tell. Fice minutes later you would get bored of the loading and just switch to another game. Along with useful TRCs came the nitty gritty ones. The ones that were supposed to make the life of the gamer a little easier, and therefore the game a little nicer to the player.

One such TRC is the Pulled Joystick Pause. When a joystick is pulled while in the middle of a game, the game should pause. Excellent idea, and i agree with the idea. But in practice this adds cost to the manfacture of the game. Suddenly the creation process has to make sure this rule is adhered too. I know most of you are thinking that this cannot be that bad. At times though it has been, with multiplayer games coming along this awkward testing got worse, and then we had inetrnet playing. Remote players could then inadvertantly pause your game when they accidentaly (yeah right!) pulled their joystick out. Just how many times in a persons life are they actually likely to pull out a controller. I know i have done it a few times, and i never cared if the game paused. What am i on, my latest high score. Well if i could do it once, i can do it again etc.So now the testing of a simple rule is rfaught with annoying edge cases. It is edge cases that really cause pain. The main rules themselves are still good to my mind (even if it sounds like i am completly trashing them!).

So onto my final rant..
Pellmell just failed peer review. Why ? Well because of one of these edge cases. In this case it is the fact that you can actually crash my game by pulling the MU just as you select it. You have to be quick and get the timing right, but after a few tries it gets easier to time. WHO THE HELL DOES THIS? I mean, who removes their MU while selcting it? I know peer reviewers do, as they have made these nice guidelines to test peoples games with. I really hoped that XNA would test stuff for sure, but not have all the freakin silly edge case TRC testing that i face on my other jobs.

I have no real reason to complain, everyone gets tarred with the same brush and testing patterns. Well they might do, haha. As i personally do not test edge cases i do not care about. Does that make me a bad tester? I dunno. I do test quite thoroughly the odd combination stuff that can happen with signed in profiles and joystick switching etc. Thos are things that a player would get very confused about. I do nnot believe they would get confused as to why the game crashed when performing MU pulling exercises.

Now i have to catch the error. A stupid error in my opinion that MS should have taken care of by returnigng real test codes and not forcing me to Try..Catch surround my good code. BAH! Will i bother is the only real question now on my mind. Should i bother fixing this and uploading this improved version of the game to go live? After all the version that is already up there to be bought has this bug.

Ok, my rant is over.. for now. I hope you all have better luck with these TRC edge cases.
If you want to comment about some of the ones that frustrate you, please feel free.

No comments:

Post a Comment