As far as a multiplayer preview goes, it will work over the 3G network so you won't be limited to wifi. The AI will take over in cases where the player drops out to keep the game flowing.
*****Official Texas 42 Discussion Thread******
171,492 Views |
1444 Replies |
Last: 22 days ago by Philip J Fry
Outstanding!
quote:
I'm sure the bidding algorithm could use some tweaking. It is in no way an exact science
Neither are human beings bidding.
My partner is kdrunk....
Same as earlier with "punting"
Partner gets bid (30)
Follow me
First hand
Plays 5/2.....
Opponent has 5/5
I'm forced to play 5/0
[This message has been edited by Bonfirewillburn (edited 8/1/2012 10:56a).]
Same as earlier with "punting"
Partner gets bid (30)
Follow me
First hand
Plays 5/2.....
Opponent has 5/5
I'm forced to play 5/0
[This message has been edited by Bonfirewillburn (edited 8/1/2012 10:56a).]
Gets bid, or forced to bid?
PJF, what kind of coding experience did you have before this? If you had an estimate, how many hours did you put in to get to the point of the first iPhone submission?
I've got two ideas for apps, but have no coding experience and no Mac computer. One would basically be a data input app with options to see the information based on filtered scenarios. The other would be a multiplayer game, but a fairly simple one as far as those go I would think. I wouldn't mind wasting my time trying to code them, but I don't necessarily want to spend even $600 on a Mac Mini if the learning curve is too steep and the time to learn in addition to the time to code is going to be a big mountain to overcome.
I've got two ideas for apps, but have no coding experience and no Mac computer. One would basically be a data input app with options to see the information based on filtered scenarios. The other would be a multiplayer game, but a fairly simple one as far as those go I would think. I wouldn't mind wasting my time trying to code them, but I don't necessarily want to spend even $600 on a Mac Mini if the learning curve is too steep and the time to learn in addition to the time to code is going to be a big mountain to overcome.
I had zero experience other than a little VBA and fortran homework I copied in college.
To be honest, when I first started the project I was easily putting in 40 hours or more a week trying to get up to speed and I am still far from an expert. It is a VERY steep learning curve, but you can do it if you put in the effort.
It took me about 10 months from start to finish and I took a long break/gave up on it around month 4. The only thing pushing me at that point was the voice in my head saying "I need to at least payoff this MacBook Pro."
[This message has been edited by Philip J Fry (edited 8/1/2012 2:58p).]
To be honest, when I first started the project I was easily putting in 40 hours or more a week trying to get up to speed and I am still far from an expert. It is a VERY steep learning curve, but you can do it if you put in the effort.
It took me about 10 months from start to finish and I took a long break/gave up on it around month 4. The only thing pushing me at that point was the voice in my head saying "I need to at least payoff this MacBook Pro."
[This message has been edited by Philip J Fry (edited 8/1/2012 2:58p).]
Damn, I don't remember if it was forced.....
Either way leading with a 5 off is not a good idea.
Either way leading with a 5 off is not a good idea.
I don't disagree. It just helps me identify if it was a bad bid to begin with or if it was just a flat out bad move.
I've found that the AI almost spins out of control when it has a really bad hand and without actually knowing what the hand is, I suspect it's holding on to 5 counts and playing 10-count offs instead. Like I said, that's the sort of thing that needs a bit of work.
I can say without a doubt in 1.0.2 if he bids follow me, he will lead with the double if he has it. If he doesnt have it and gas nothing but offs and count that are losers, he'll play the off. That bit of code could use some work.
[This message has been edited by Philip J Fry (edited 8/1/2012 5:11p).]
I've found that the AI almost spins out of control when it has a really bad hand and without actually knowing what the hand is, I suspect it's holding on to 5 counts and playing 10-count offs instead. Like I said, that's the sort of thing that needs a bit of work.
I can say without a doubt in 1.0.2 if he bids follow me, he will lead with the double if he has it. If he doesnt have it and gas nothing but offs and count that are losers, he'll play the off. That bit of code could use some work.
[This message has been edited by Philip J Fry (edited 8/1/2012 5:11p).]
Sometimes a 5 off lead will allow your partner to trump the double.
Why would the AI (opponent) hold 11 and play the 66 if the 01 was previously played and the 64 was still not played
Just happened :
I led last trump
Left plays 65
Partner playes 32
Right playes 66
I then walk the 64
Left plays....?
Partner plays ...?
Right plays 11
The honest answer is that I don't know. The 66 is coded to be a more valuable tile than the 11 and in it SHOULD have chosen the 11. This would be Fry or Philip?
[This message has been edited by Philip J Fry (edited 8/1/2012 6:46p).]
[This message has been edited by Philip J Fry (edited 8/1/2012 6:46p).]
Bonfire, can you send me an email? I want to get your help with something when you have a chance.
maroonentertainment@gmail.com
[This message has been edited by Philip J Fry (edited 8/1/2012 6:56p).]
maroonentertainment@gmail.com
[This message has been edited by Philip J Fry (edited 8/1/2012 6:56p).]
DDP
[This message has been edited by Bonfirewillburn (edited 8/1/2012 10:45p).]
[This message has been edited by Bonfirewillburn (edited 8/1/2012 10:45p).]
....changed all the names so I am not sure. was the player to my right. Player 4
Honestly BOTH players misplayed, looks like they both glitched and played their high 6's.
Email sent
[This message has been edited by Bonfirewillburn (edited 8/1/2012 10:49p).]
Honestly BOTH players misplayed, looks like they both glitched and played their high 6's.
Email sent
[This message has been edited by Bonfirewillburn (edited 8/1/2012 10:49p).]
One thing that will be necessary soon in the AI is to mark dominoes as worthless or trash so they get sloughed over good tiles.
Also the reverse. Dominoes must be marked as must win (or the opponent bid makes). Similarly, a step through each legal tile to see if it will push the partner over their bid.
Also the reverse. Dominoes must be marked as must win (or the opponent bid makes). Similarly, a step through each legal tile to see if it will push the partner over their bid.
Oh, some of that is already in there, but admittedly not all. A better question is why what is in there isn't it working. The code was written with the intent of holding on to the 66 and 65 for the 64. Once the 64 is played, they are free to be sloughed...at least as the first choice of sloughes go. That is of course, if the code is operating the way I intended.
Usually it goes something like this.
Usually it goes something like this.
did you proceed to play the 65?
From a helpful beta tester before the release. I'm sure the email was something along the lines of "Not sure WTF just happened here"
Old complaint on the AI:
Jay still refuses to take the lead on a low trump.
Hands:
Me: 11 15 14 10 55 43 22
Philip: 13 x x x x x x
Jay: 16 12 44 x x x x
Fry: doesn't matter.
Lead 15: Jay plays 12
Lead 11 to pull 16 from Philip. Nope, Jay played wrong. I'm now forced to lead a 4 when Jay could have taken the lead and led once as I needed.
Now have 4 winners and a 4 off. No way to pull 64. 43 is a sinker.
I got lucky that Jay could catch the only off I had.
Better is having the bidders partner win the first trick if possible. No exceptions.
Jay still refuses to take the lead on a low trump.
Hands:
Me: 11 15 14 10 55 43 22
Philip: 13 x x x x x x
Jay: 16 12 44 x x x x
Fry: doesn't matter.
Lead 15: Jay plays 12
Lead 11 to pull 16 from Philip. Nope, Jay played wrong. I'm now forced to lead a 4 when Jay could have taken the lead and led once as I needed.
Now have 4 winners and a 4 off. No way to pull 64. 43 is a sinker.
I got lucky that Jay could catch the only off I had.
Better is having the bidders partner win the first trick if possible. No exceptions.
I bid 84 in 2s
22 26 24 21 33 55 53
Fry holds 65 50 to the end, then plays backwards.
22 26 24 21 33 55 53
Fry holds 65 50 to the end, then plays backwards.
Do I understand right that you lead with the 15, Philip played the 12, and Jay played the 13 and didn't pull the lead from you?
One of the issues I'm struggling with, is making Jay determine that you really are trying to punt to him in situations like this. Jay thinks that Fry has the 11 at this point so he doesn't want to waste throwing out the 16 to a losing hand.
And I agree on the 84 bids. I still have some "playing for count" artifacts in the code that need to be bypassed for marked bids.
[This message has been edited by Philip J Fry (edited 8/2/2012 7:13p).]
One of the issues I'm struggling with, is making Jay determine that you really are trying to punt to him in situations like this. Jay thinks that Fry has the 11 at this point so he doesn't want to waste throwing out the 16 to a losing hand.
And I agree on the 84 bids. I still have some "playing for count" artifacts in the code that need to be bypassed for marked bids.
[This message has been edited by Philip J Fry (edited 8/2/2012 7:13p).]
No, Philip was forced to play 13. Fry just played under the 15 with the 12 instead of winning with the 16
Even if the 11 comes out, chances are excellent that I have the 14 (next high trump) so holding the 16 in this situation is a huge error. (but not against the rules)
Even if the 11 comes out, chances are excellent that I have the 14 (next high trump) so holding the 16 in this situation is a huge error. (but not against the rules)
Ah. Got my numbers mixed up, but I understood the situation.
Is it a true statement to say that if you lead off with anything other than the double that you want your partner to play his highest tile that beats yours? That would certainly simplify the code a little.
[This message has been edited by Philip J Fry (edited 8/2/2012 7:34p).]
Is it a true statement to say that if you lead off with anything other than the double that you want your partner to play his highest tile that beats yours? That would certainly simplify the code a little.
[This message has been edited by Philip J Fry (edited 8/2/2012 7:34p).]
Pretty much. Unless he knows he can't win it. Then it probably doesn't matter much.
nm. Going to mull it over a bit.
[This message has been edited by Philip J Fry (edited 8/2/2012 11:12p).]
[This message has been edited by Philip J Fry (edited 8/2/2012 11:12p).]
Just had an interesting hand that caught my attention. I'm sure this type of situation is hard to code for, but I figured I would let you know about it.
Partner wins bid for 36, trumps 4's.
He had 6 trumps minus the 4:0 with the 3:2 for his off. He played all 6 trumps in order, then threw the 3:2, opponent threw 3:3, I threw 5:3, other opponent threw 5:5. Mostly bad luck I guess, but had he thrown the 3:2 on the second trick it would've forced the last opponent to play one of his 3's rather than the 5:5.
Partner wins bid for 36, trumps 4's.
He had 6 trumps minus the 4:0 with the 3:2 for his off. He played all 6 trumps in order, then threw the 3:2, opponent threw 3:3, I threw 5:3, other opponent threw 5:5. Mostly bad luck I guess, but had he thrown the 3:2 on the second trick it would've forced the last opponent to play one of his 3's rather than the 5:5.
6th play
Partner has lead, 1 trump left in his hand (5s are trump)
Partner plays 6/0
Opponent right plays 0/0
I play 1/6
Opponent left plays 6/6
Then opponent left plays 6/4
Partner trumps
We win
We would have been set(6/4 would have won previous hand) then play big 6
Partner has lead, 1 trump left in his hand (5s are trump)
Partner plays 6/0
Opponent right plays 0/0
I play 1/6
Opponent left plays 6/6
Then opponent left plays 6/4
Partner trumps
We win
We would have been set(6/4 would have won previous hand) then play big 6
I'm going to have 1.1 ready to submit to the store tomorrow. I'd like to have some more beta testers check it out before it goes live. That will give you about 10 days to find any show stoppers. If any of you want to check out the Nello options (and other bug fixes I've been working on) please send me an e-mail to
MaroonEntertainment@gmail.com
I will need your UDID. You can download a free app from the app store that looks this up for you. E-mail this to me and I'll give you instructions on how to install the beta version.
[This message has been edited by Philip J Fry (edited 8/4/2012 1:43p).]
MaroonEntertainment@gmail.com
I will need your UDID. You can download a free app from the app store that looks this up for you. E-mail this to me and I'll give you instructions on how to install the beta version.
[This message has been edited by Philip J Fry (edited 8/4/2012 1:43p).]
I'm about to have a 10 hr flight to South America. Plenty of time. Am I still on the list?
You are. What's the latest I can get it to you?
Monday around 4
Had some pretty heavy personal issues come up, but I may still be able to get something to you late tonight.
Anyone out there with an iPad 1? Customer claims they can't use the bid/trump wheel and I'd like to see if I can reproduce whatever is happening with her.
I've got one and haven't had any issues with it in that department.
Featured Stories
See All
10:53
3h ago
1.2k
5 Thoughts: No. 23 Texas A&M 97, Lamar 71
by Luke Evangelist
Brauny's Best: Highlighting top high school performances from Week 11
by Ryan Brauninger
riverrataggie
Quinn Ewers begs girl for nude photos in wild leaked messages
in Texas A&M Football
115