Play Balance regarding Russia

[Shannon V. OKeets]

ORIGINAL: amwild

ORIGINAL: lavisj
In the case of the Mulligan and the auto save redoing the rolls for the combat, I was under the impressions that random numbers generated by computers were not random but the product of a complex mathematical function whose result is unpredictable but certain.
Which would mean that if no other roll is being made, you could add the move, and the rolls for the combat should be the same. Am I wrong on that?

Given the same seed value, the sequence of results from a psudeorandom number generator is always the same. However, most applications that use psudeorandom numbers use some sort of variable seed - often based on the system clock - so that the sequence of psudeorandom numbers cannot easily be predicted. Also, since psudeorandom number generators use their previous result as the seed for the next result, sequences of numbers do not readily repeat themselves.

So, under normal circumstances, going back and repeating a move will not result in the same die roll and the same result. In some games, you can save the game, make a move, and if you don't like the result, you can load the game and make the move again (and again...) until you achieve a result you like.

However, I seem to recall that Steve is adding code so that if you try this, you will get the same result every time - the die roll is tied to the event it is rolled for - so if you went back, you would have to try something else to get a different result.

No. At least I hope I didn't say something like that.

As you noted, it is merely a question of whether the current 'seed' number is saved or not. However, MWIF will use a system I have used for over 30 years, where 2 random number generators are used, the first fills a table with random numbers and the second selects values from that table. So saving the "current seed" would involve saving a bunch of numbers.

I am only really concerned about games between human players and preventing cheating. And even then it is only relevant for Internet and PBEM games. For those, the system I intend to use will inform the opposing player(s) automatically when a player causes a random number to be used. In fact, I will strive to have the opposing players informed before the player who causes the event is informed (only milliseconds before, but still ...).

[Shannon V. OKeets]

ORIGINAL: Incy
Point is that once I roll that first naval combat in pacific it's to late to send out that forgotten escort in the atlantic, resupply malta with that CP, or send the dutch TRS to pick up a south african FTR (Because there's no realistic way for me to recreate the outcome of a naval battle). So I can't really have a fair mulligan, because the dice have allready been thrown and it's not fair to go back and throw them again (you took Gibraltar? Wait a minute, I think I forgot to rebase this russian LND4 to Vlad..).
So I end up spending an extra minute or 15 CONSTANTLY, so that I don't miss to much stuff.
But my opponent would rather have me play faster, and then grant me a mulligan whenever I miss something obvious, such as the dutch TRS picking up a FTR in some outlaying theathre. If I could set the dice how I want *in cheat mode*, I could go to an autosave, reimplement stuff that allready happened (providing the same dicerolls), and then start from where the mistake was dicovered *with cheat mode off*.

I'll try not to lose sight of your concerns on these points as I continue programming.

My first defense would be a good player interface which minimizes mistakes and makes reviewing decisions before committing to them not extremely painful.

The more I program all the details of WIF FE RAW into MWIF the more I am astonished at the complexity of the rules. For example, one of the things that the program does is to check whenever a land unit moves for whether the hex entered changes control, and if it does, whether it is a city, and if so, whether there are any militia units associated with that city, and if so, whether those militia units should to be moved from the FutureForce pool to the Force pool (or vice-a-versa). That is for every hex a land unit enters. I doubt that anyone playing over the board has ever bothered to make those adjustments.

My point is that a lot of the mistakes made during over the board play are due to the high rule complexity of WIF FE. As part of that, I include the 76 phases of the game. Automation should simplify play enormously, and enable players to devote their attention to accomplishing tasks during a game, rather than double checking whether they are playing the game according to the rules.

[Neilster]

ORIGINAL: Shannon V. OKeets

ORIGINAL: amwild

ORIGINAL: lavisj
In the case of the Mulligan and the auto save redoing the rolls for the combat, I was under the impressions that random numbers generated by computers were not random but the product of a complex mathematical function whose result is unpredictable but certain.
Which would mean that if no other roll is being made, you could add the move, and the rolls for the combat should be the same. Am I wrong on that?

Given the same seed value, the sequence of results from a psudeorandom number generator is always the same. However, most applications that use psudeorandom numbers use some sort of variable seed - often based on the system clock - so that the sequence of psudeorandom numbers cannot easily be predicted. Also, since psudeorandom number generators use their previous result as the seed for the next result, sequences of numbers do not readily repeat themselves.

So, under normal circumstances, going back and repeating a move will not result in the same die roll and the same result. In some games, you can save the game, make a move, and if you don't like the result, you can load the game and make the move again (and again...) until you achieve a result you like.

However, I seem to recall that Steve is adding code so that if you try this, you will get the same result every time - the die roll is tied to the event it is rolled for - so if you went back, you would have to try something else to get a different result.

No. At least I hope I didn't say something like that.

As you noted, it is merely a question of whether the current 'seed' number is saved or not. However, MWIF will use a system I have used for over 30 years, where 2 random number generators are used, the first fills a table with random numbers and the second selects values from that table. So saving the "current seed" would involve saving a bunch of numbers.

I am only really concerned about games between human players and preventing cheating. And even then it is only relevant for Internet and PBEM games. For those, the system I intend to use will inform the opposing player(s) automatically when a player causes a random number to be used. In fact, I will strive to have the opposing players informed before the player who causes the event is informed (only milliseconds before, but still ...).

Being able to return to a saved game to try again for a better roll is a good way of achieving "what if" scenarios. I'm sure this is quite common but in CWIF I've done this until Japan had a stunning victory at Midway to see how this affects the Pacific theatre thereafter.

Cheers, Neilster

[amwild]

ORIGINAL: Shannon V. OKeets

ORIGINAL: amwild

ORIGINAL: lavisj
In the case of the Mulligan and the auto save redoing the rolls for the combat, I was under the impressions that random numbers generated by computers were not random but the product of a complex mathematical function whose result is unpredictable but certain.
Which would mean that if no other roll is being made, you could add the move, and the rolls for the combat should be the same. Am I wrong on that?

Given the same seed value, the sequence of results from a psudeorandom number generator is always the same. However, most applications that use psudeorandom numbers use some sort of variable seed - often based on the system clock - so that the sequence of psudeorandom numbers cannot easily be predicted. Also, since psudeorandom number generators use their previous result as the seed for the next result, sequences of numbers do not readily repeat themselves.

So, under normal circumstances, going back and repeating a move will not result in the same die roll and the same result. In some games, you can save the game, make a move, and if you don't like the result, you can load the game and make the move again (and again...) until you achieve a result you like.

However, I seem to recall that Steve is adding code so that if you try this, you will get the same result every time - the die roll is tied to the event it is rolled for - so if you went back, you would have to try something else to get a different result.

No. At least I hope I didn't say something like that.

As you noted, it is merely a question of whether the current 'seed' number is saved or not. However, MWIF will use a system I have used for over 30 years, where 2 random number generators are used, the first fills a table with random numbers and the second selects values from that table. So saving the "current seed" would involve saving a bunch of numbers.

I am only really concerned about games between human players and preventing cheating. And even then it is only relevant for Internet and PBEM games. For those, the system I intend to use will inform the opposing player(s) automatically when a player causes a random number to be used. In fact, I will strive to have the opposing players informed before the player who causes the event is informed (only milliseconds before, but still ...).

When I mention the saved seed, I was talking about the basic random number generators ( i.e. the rnd() function or similar) built into most application development tools, not MWiF's particular implementation of a random number generator.

As to the same result comment, my memory has been known to be fuzzy, but I do seem to recall Steve saying something about the dice rolling program located on a third-party server that will be used to prevent players in PBEM games fishing for better dice rolls by remembering what roll was made for each event, and in the event that a player discards his turn's moves and and starts his move again, the dice rolling app would feed the player the same rolls for his move. I suppose I thought that it would also have the effect of preventing dice roll fishing in all aspects of the game. Since I am a WiF novice, I may be making some incorrect assumptions about the sequence of play here too...

I hope I haven't just been embarrasing myself by dredging up memories of other people's wish lists. I also hope that I haven't just discovered a whole lot of work for Steve that he has forgotten about...

[lavisj]

Ok, what about this. For solving the Mulligan problem.
How difficult would it be to allow the oposing player to move your units after you unlock one. For exemple, let's say you forgot a move and ask the other player if that move could be granted. Then you could unlock your unit, allowing your oponent to move it in his movement phase, and then you could agree to the move and relock it?
Would that solve the problem?

[Shannon V. OKeets]

As to the same result comment, my memory has been known to be fuzzy, but I do seem to recall Steve saying something about the dice rolling program located on a third-party server that will be used to prevent players in PBEM games fishing for better dice rolls by remembering what roll was made for each event, and in the event that a player discards his turn's moves and and starts his move again, the dice rolling app would feed the player the same rolls for his move. I suppose I thought that it would also have the effect of preventing dice roll fishing in all aspects of the game. Since I am a WiF novice, I may be making some incorrect assumptions about the sequence of play here too...

What I plan here is for eMWIF (hosted by a neutral 3rd party system) to send emails to all the PBEM players at the same time (i.e., Copy All). Everyone gets the same die roll at the same time. Alternatively, players could replay a game, causing MWIF to read the game record log, and use whatever die rolls had been stored therein.

[Shannon V. OKeets]

ORIGINAL: lavisj
Ok, what about this. For solving the Mulligan problem.
How difficult would it be to allow the oposing player to move your units after you unlock one. For exemple, let's say you forgot a move and ask the other player if that move could be granted. Then you could unlock your unit, allowing your oponent to move it in his movement phase, and then you could agree to the move and relock it?
Would that solve the problem?

This would be comparably difficult to implement, because of its violation of the rules. To some degree, any solution to "correcting mistakes and omissions" would be comparable to implementing a "house rule".

[Incy]

ORIGINAL: Shannon V. OKeets

What I plan here is for eMWIF (hosted by a neutral 3rd party system) to send emails to all the PBEM players at the same time (i.e., Copy All). Everyone gets the same die roll at the same time. Alternatively, players could replay a game, causing MWIF to read the game record log, and use whatever die rolls had been stored therein.

You might consider checking out tripleA.

It's an opensource wargame engine mostly used for Axis and Allies. They use a very similar approach to what you're proposing.
They started out using IRONY as a dice-server, but ended up providing another one themselves, after there were some trouble with IRONY.

If you post at the developer forum I'm sure you'll get good advice from the developers.

[Shannon V. OKeets]

ORIGINAL: Incy

ORIGINAL: Shannon V. OKeets
What I plan here is for eMWIF (hosted by a neutral 3rd party system) to send emails to all the PBEM players at the same time (i.e., Copy All). Everyone gets the same die roll at the same time. Alternatively, players could replay a game, causing MWIF to read the game record log, and use whatever die rolls had been stored therein.

You might consider checking out tripleA.

It's an opensource wargame engine mostly used for Axis and Allies. They use a very similar approach to what you're proposing.
They started out using IRONY as a dice-server, but ended up providing another one themselves, after there were some trouble with IRONY.

If you post at the developer forum I'm sure you'll get good advice from the developers.

Thanks. When I get to writing eMWIF, I'll be sure to do that. It should be a simple little program, but there may be some 'gotchas' involved I don't know about.

[38special]

Back to topic.

I have played wif as the russians many times and I believe that good CW play with multiple lend lease routs into russia keeps the Germans from beating russia alone. With Japan's aid it is a little harder.

IMO, russia should invade persia turn 1 and possibly iraq as well. Oil resources are always good.

The all armor russian builds looks like a strategy I will have to playtest. I would not think it would work, but I have never tried it either.[:)]