detecting when the ball leaves the pitch bug!

[Steve Camber]

Ahh - I think I see the difference now. The practice mode bug is when the players go awol whereas the 2-player corner bug is when the corner taker goes awol - is that right ?

Yes!

[gdh82]

Just for the record, another variation of the 'detecting when the ball leaves the pitch' bug occurred at the UKC 2007, in the match between Gianni and Simon B.

I believe the ball went out for a corner but before this was awarded the keeper caught the ball and so a corner wasn't given. Play resumed until the next foul when the program fell over! This lead them having to restart half of the match again!

Have I got that right Simon or Gianni ?

[Bounty Bob]

I think I've seen this bug before. If not this one, then something very similar. Was the nature of the mess up that all pitch boundaries were basically ignored? When I've seen it, goals don't count throw ins don't get awarded etc.

[alkis21]

You should have seen the mayhem it caused when it happened during the Nikos-Gianluca semi final in the Groningen world cup!

[Ely]

Mayhem associated with Nikos, surely not

[alkis21]

Nah, not that kind of mayhem.. they both took it very calmly, but it was a confusing situation until a fair solution was decided.

[gdh82]

Just for the record, another variation of the 'detecting when the ball leaves the pitch' bug occurred at the UKC 2007, in the match between Gianni and Simon B.

I believe the ball went out for a corner but before this was awarded the keeper caught the ball and so a corner wasn't given.

Is the nature of the mess up that all pitch boundaries were basically ignored? When I've seen it, goals don't count throw ins don't get awarded etc.

This is fourth and final instalment of my recent bug reports - which, believe it or not, all occurred in one evenings KO2 - but here are just a few of the effects of what we could call the No-Boundaries Bug.

Although I haven't any footage of what triggered this, I can remember what happened as its virtually the same as the UKC07 incident! I was kicking downwards, trying a narrow-angle shot to the left of the goal near the bye-line. The shot glances off the keeper, hits the post and then fractionally and momentarily goes out of play before finding its way back into the keepers hands, all in one movement!!! At the point it goes out of play a corner is displayed on the screen, but it seems the keeper "re-claims" the ball before the corner code is executed ?

I'm wondering if this is similar to the corner-bug ? i.e the code belatedly takes action on the ball's position after it has rebounded back onto the pitch ? Sound possible, Steve ? Anyone else care to comment ?

[gdh82]

8.) There are different kinds of throw in "bugs". The most common type is the one where you slide and hit the ball, the ball is close to the opponent who takes control of the ball again (as noted on the bottom of the screen) but WITHOUT touching it. The slider is awarded the throw in. This is not changed now, and I don't think it should be changed either (since the ability to look at the bottom of the screen and see who last "controlled" (as opposed to touched) the ball has been an integral part of the game since forever). On the other hand, the bug where the CPU makes a diagonal dribble and changes direction 90 degrees on the sideline, the human gets a throw in, and immediately afterwards the CPU gets the throw in, seems to be fixed. Is this a correct observation?

Thought I'd answer your valid query here.
These two clips posted to this thread a while back should help you understand the throw-in fix, Dagh.




In both cases and in the cpu 90' turn you describe, there is the same underlying problem. The ball returns pitch side of the sideline prior to the throw-in being taken and its the ball's final position that caused the throw-in to be awarded to the wrong team. But not anymore. In other words, your observation is correct!

[Steve Camber]

Well done Garry! This shows the problem perfectly.

[dnielsen]

Thanks Garry. I don't understand what is happening in the second video, but I understand your explanation just fine. And good work, Steve!

[gdh82]

I'm sure this isn't news to anyone here but this matter was raised by others at my last two tournies so thought I'd make a post. The ref calls that the ball is out of play when often the ball hasn't fully crossed the line - its half on and half off it. Given the great game is all about precision and the width of a pixel can sometimes decide a game, this seems an obvious if very minor flaw.

If throw-ins etc were only called when the ball fully crossed the line, it would not only be more accurate, it would allow gameplay to continue that little bit more.

This isn't a big deal and the lull in KO2 discussion recently prompted me to post as much as anything.

[Bounty Bob]

I Think Steve said that the ball is referenced in code as a single point and not as a sphere with proper width. This makes sense with the 'bug' as described above. If the ball is a single point, then it would be logical to draw it with that point at the centre. This then means that when the centre of the ball crosses the line, it will be detected as out of play.

If this is the case, then it probably won't be a simple fix. I guess the easiest way would be to increase the pitch boundaries by the size of half the ball as we see it but leave the graphics drawn where they are.

I'm sure Steve will confirm or deny this theory.

[gdh82]

Cheers for the thoughtful reply, Bbob.

Thinking about it, be interesting to know if the pitch-size could altered not just because of this issue but for the potential JFF 5-a-side option now that KO2CV 1.0 allows us to easily remove players! I have actually played a KO2 5aside already - good fun but makes the pitch seem absolutely GIGANTIC!