Jump to content

Reconsidering email turn sequencing


ianc

Recommended Posts

Greetings all and BTS,

I've just had my first opportunity to engage in an email game with a friend. In exchanging the first few turns, it seemed to me that the sequence in which turns are exchanged is a bit unwieldy. With emails indicated by the dashed lines, the current sequence is:

A - Setup

---------

B - Setup

B - Orders

---------

A - Orders

---------

B - See Playback

---------

A - See Playback

A - Orders

---------

B - Orders

And the sequence begins again. To decrease the number of emails required, I'd like to propose the following scheme:

A - Setup

---------

B - Setup

B - Orders

---------

A - Orders

A - See Playback

---------

B - See Playback

B - Orders

Note that the game is in the same state after both of the above sequences. Letting whomever is last to complete orders for a specific turn view the playback immediately cuts out two emails that delay the prosecution of the game. The proposed sequence has B always giving first orders for a turn (rather than the existing alternating scheme), but this shouldn't cause a problem, should it? Sorry if this has been dealt with before, but I'm just wondering if this had been considered already, and if there's some logic I'm just not seeing in it. Thanks for any info,

Ian

Link to comment
Share on other sites

Sorry... Err, what I did was post before searching. I think that would actually be the opposite of the way it should be done. Yes.

I found Fionn's reason for this posted:

"Why is this done? To stop CHEATING.. Some players would simply replay the resolution phase until the MOST favourable results possible came up and then send it off to you. By not allowing them to see the results of their orders until you see them first no cheating can occur."

I think this could be solved by making the first play-through of the turn unstoppable until it finished. After the first run-through you can rewind as often as you like, you'll just redisplay what was initially generated. Hopefully BTS is still considering some type of streamlining feature for these exchanges. Best,

Ian

Link to comment
Share on other sites

Guest Big Time Software

Hi Ian, you can cheat around this very easily. All you need to do is make a copy of the file to work from first. Play through, see what you need to see, then exit and start up from another copy.

We have given this a whole lot of thought, and we can find no way to do it other than the way it is now. Cheating in PBEM is a big problem that we are avoiding. Don't forget that you will be able to play via TCP/IP, and therefore can avoid PBEM altogether wink.gif

Steve

Link to comment
Share on other sites

BTS...I hope I'm not beating a dead horse but here is a better way to do PBEM and still avoid the cheating. I only post this because there are some turns all I get to do is orders or watch playback..and on some turns I have never been able to give orders. So here goes:

Player A: Setup

---------------

Player B:Setup and place orders for turn 1

---------------

Player A: place orders for turn 1

---------------

Player B: watch playback turn 1 and orders turn 2

---------------

Player A: watch playback turn 1 and place orders for turn 2

---------------

Player B: watch playback turn 2 and place orders for turn 3

and so on until I win the game. This would eliminate cheating and take less time to play. I play a great deal of SP1 by e-mail and that is why this is of such concern for me. I'm sorry if I have posted on this topic before but it is the ONLY thing that I have been disappointed with so far. Other than that the game is perfect and without flaw.

Link to comment
Share on other sites

Guest L Tankersley

<BLOCKQUOTE>quote:</font><HR>Player B: watch playback turn 1 and orders turn 2<HR></BLOCKQUOTE>

This is the problem, the same one Steve pointed out in response to ianc's suggestion. The FIRST player to watch a playback CANNOT be allowed to issue orders before the opponent has also seen the playback. If you allow this, that player can recalculate the turn multiple times to achieve favorable results. [in fact, you need to ensure that the player that calculates the turn results does not VIEW those results before the opponent gets the results file. What people have been leaving out of their proposed email exchange schedules (no doubt because it happens "automatically" with no operator intervention) is the detail of who calculates the turn results and when.]

At least, this is true if the first player to view the playback is also the player that calculates turn results. Hmmm...if you don't have that condition, you can get something like the following:

A setup

-------

B setup

B orders turn 1

-------

A orders turn 1

A calculates turn 1, does not view

------

B views turn 1

B orders turn 2

------

A views turn 1

A orders turn 2

A calculates turn 2, does not view

------

B views turn 2

B orders turn 3

------

etc.

The above seems like it would defeat the recalculate cheating problem, but it puts all results calculation onto the same computer for the entire game. There might also be other issues (for one thing, once you get past the first couple emails EVERY exchange contains film data along with orders for the next turn instead of this data being separated - that might not be attractive).

Leland J. Tankersley

[This message has been edited by L Tankersley (edited 11-15-99).]

[This message has been edited by L Tankersley (edited 11-15-99).]

Link to comment
Share on other sites

It's as if a light bulb went off in my tiny head. I now have the information I need to stop my whinning. I never accounted for how and where the calulations for each turn were done. In theory an unscrupulous swine could receive the movie....watch it and without plotting his moves and sending them simply leave the game and go back and watch the movie again and again until he gets the results he likes most and THEN plot he moves and send them to the unsuspecting sap on the the other end. I will no longer bitch about the way the PBEM works. I must become sneakier in my thinking because that never occured to me....this is something I will work on (becoming more devious)

Link to comment
Share on other sites

Guest Scott Clinton

A setup

B setup

A generate/view setup (possible LOS)

A plot turn one

B view setup (possible LOS)

B plott turn one

A generate/watch turn one movie

A plot turn two

B view turn one movie

B plot turn two

A generate/watch turn two movie

A plot turn three

B view turn two movie

B plot turn three

etc...etc...

------------------

The Grumbling Grognard

Link to comment
Share on other sites

Guest L Tankersley

A setup

B setup

A generate/view setup (possible LOS)

A plot turn one

B view setup (possible LOS)

B plot turn one

A generate/watch turn one movie

A doesn't like results he got

A generate and watch new turn one movie

A likes this result better

A plot turn two

etc.

----

The problem is that the orders files themselves don't contain all the information needed to generate the results due to randomness in the execution/calculation - issuing the same orders does not necessarily produce the same results. For this reason, the player that generates the turn results MUST send the results to the opponent before viewing.

Leland J. Tankersley

Link to comment
Share on other sites

Guest Scott Clinton

Leland:

> The problem is that the orders files

> themselves don't contain all the

> information needed to generate the

> results due to randomness in the

> execution/calculation - issuing

> the same orders does not necessarily

> produce the same results. For this

> reason, the player that generates

> the turn results MUST send the

> results to the opponent before viewing.

ahhhh, yes exactly. That is why I labelled the sequence as "VIEW" and "GENERATE". I can clarify as:

A setup

A send to B his set up file

B setup

B send both setup files back to A

A generate/view combined setup (possible LOS only no movie)

A plot turn one

A send to B compiled setup and his orders for turn one

B view combined setup (possible LOS only no movie)

B plott turn one

B send to A his orders for turn one

A generate/watch turn one movie

A plot turn two

A send to B compiled movie for turn one and his orders for turn two

B view turn one movie

B plot turn two

B send to A his orders for turn two

A generate/watch turn two movie

A plot turn three

A send to B compiled movie for turn two and his orders for turn three

etc...etc...

------------------

The Grumbling Grognard

Link to comment
Share on other sites

Guest L Tankersley

<BLOCKQUOTE>quote:</font><HR>A generate/watch turn one movie<HR></BLOCKQUOTE>

If A generates the results for a turn and watches those results before sending them to B, then A has the opportunity to cheat by reloading the prior email file and regenerating the turn until a favorable result is achieved. Plus with your sequence, ONLY A has the chance to cheat because only A generates turn results.

Leland J. Tankersley

Link to comment
Share on other sites

Guest Scott Clinton

ahhhh I see now. I stand corrected, I was wrong (see its not hard to say!)

I do think this would be a nice 'low security' option though.

I was more worried about folks changing orders after they viewed the files but I see your point.

Thanks

------------------

The Grumbling Grognard

Link to comment
Share on other sites

A setup

B setup

A orders turn1

B orders turn1

B generate movie1

A view movie1

A orders turn2

B view movie1

B orders turn2

B generate movie2

A view movie2

A orders turn3

B view movie2

B orders turn3

B generate movie3

and so on...

Sure, the same player has to generate all the movies, but so what?

Link to comment
Share on other sites

Just to add my 2 cents worth:

1) Can't you save the random-number generator's seed value in the PBEM file so that even "random" sequences can be regenerated, and hence there's no problem to solve?

2) Doubling the time between moves is really really annoying. If the 2 players trust each other not to cheat, then perhaps you could offer a simple option to allow us to turn on "fast email turnaround" which doesn't involve all this faffing around.

3) If you generate the move at the end of the placing-orders phase, rather than at the start of the viewing-movies phase, then what's the problem? So long as you have to give the file back to the other player *before* you can see the result of your orders, there is no chance of cheating.

i.e.

A - Setup & 1st orders

B - Setup & 1st orders (turn 1 resolved now)

A - Playback turn 1. Orders 2

B - Playback turn 1. Orders 2 (turn 2 resolved now)

A - Playback turn 2. Orders 3

B - Playback turn 2. Orders 3 (turn 3 resolved now)

...

Aside from the technical issues of what you put in the PBEM file when (which might make any of the above difficult/time-consuming to code), this would be a vast improvement, as I'm currently only able to make 2 moves a week rather than 5.

Zot!

Link to comment
Share on other sites

This has got to be the 38th thread about this PBEM stuff! I still say we can do a turn in two emails and keep the current cheating precautions in, but I've already posted that about twelve times and won't rehash it here.

Coffee--The reason PBEM is so important to me is because this is not a RTS game like "Age of Empires" or "Sid Meier's Gettysburg" where both players constantly have something to do and which lend themselves well to TCPIP play. With CM, one player is always going to wind up waiting for the other player to finish issuing orders after watching the film X times before they can do anything--especially if you're the defender.

Another reason for PBEM and, especially, streamlining the PBEM down to two emails/turn is because you have Yanks, Kiwis, Aussies, Brits, Scots, Swedes, Krauts, Frogs, etc. playing each other all over the world and it may be 3:00 AM for your opponent when you're ready to send them your turn.

So I don't expect this topic to get dropped for some time until the PBEM scheme is improved.

Just my two cents back atcha! smile.gif

Dar

Link to comment
Share on other sites

Amen to the time zones bit Dar.

I too would like to see the email procedure streamlined, but I feel it should be an all or nothing thing. ie I would not like to see an "I trust this player" option.

IMO its kind of a delicate thing when establishing a game: "Hi, lets play a game, but I think you're a cheating bag of pus so I'm going to insist on the 'I think my opponent is a cheat' option". Granted you can phrase it better, but really I feel that's what it boils down to.

If there is only one option - painless, secure PBEM - then there is no problem. smile.gif

Jon

------------------

Quo Fas et Vino du Femme

Link to comment
Share on other sites

Now I'm back to wondering why the PBEM was set up the way it has been. Couldn't the turn resolution be generated BEFORE the e-mail is sent then the first person to view the movie couldn't watch it again for better results because it would have been generated before he received it. This would eliminate cheating and cut down the number of e-mails to play a game. While TCP/IP will be the preferred method of playing CM, I do not have the time to play for 3 hours everyday. I do have the time to do the 5-10 PBEM a couple of times a day. This is why it has been of some importance to me.

Also, has the game locking up after you have done your end of the e-mail bug been worked out yet for the final release?

Sorry if I'm beating a deadhorse and if Steve,Charles,Fionn or Moon want to tell me to stop my whinning "it is what it is" I will.

Link to comment
Share on other sites

  • 6 months later...

Resurecting a dead post:

Zot and Jarmo bring up a good point that was not (at least as I could find) refuted. PBEM currently works like this (ignoring setup and the like):

A--view action9, turn10

B--turn10, process

A--view action10

B--view action10, turn11

A--turn11, process

B--view action11

A--view action11, turn12

B--turn12, process

A--view action12

B--view action12, turn13

A--turn13, process

B--view action13

A--view action13, turn14

B--turn14, process

etc

The suggested new method

A--view action9, turn10

B--view action9, turn10, process

A--view action10, turn11

B--view action10, turn11, process

A--view action11, turn12

B--view action11, turn12, process

A--view action12, turn13

B--view action12, turn13, process

A--view action13, turn14

B--view action13, turn14, process

etc

The newly proposed method is about 1/3 faster (10 emails vs 14 for the given example). Because B always processes and A always sees the movie first there is no way B can proccess again and again from a saved file. Is there a problem with having the same person process repeatedly? If the processing process can be "cracked" then PBEM is insecure anyway, with BTS's method it is about half as insecure, but still. Could someone explain why my logic is faulty. Thanks!

--Chris

Link to comment
Share on other sites

Guest tom w

Hi

I think it is not so bad the way it is.

Could it be better?

Well some people here seem to think so.

I'm very happy that BTS has identified the the possibility of cheating as the most important thing to guard against. I would like to know the game has been designed to prevent cheating. I like the attention to detail that also avoids non realistic "gamey" play.

I'm Sure that the TCI/IP patch/update will be the answer to all our dreams. In the meantime I'm happy to know that the PBEM protocol has proven (so far) to be cheat proof and tamper proof, I think this should be the most important consideration because we all know at least one wargamer, that loves to cheat anyway they can. (NOW before anyone on this board gets paranoid, I was refering to guys we all knew or used to know that fiddled and diddled with 2D board game pieces until there original start position was uncertain and them they claimed they had enough movement factors left to "make it" to their new "new" objuective which would have been imposible to reach from their original position before it was "accidentally" moved or forgotten or diddled the unit to its now "new" location)

Some 2D board wargamers are KNOWN to cheat anyway they can.

I'm glad BTS has made every effort to eliminate such clearly reprehensible behaviour!

(Case I point I introduced the "friend" I was just refering to, to CM and said that this was a NEW way of fighting WW II tank battles because you could not be sure were the enemy units are as both players see only the perspective of their own units and what is the VERY first thing he tries to do? Get this....

HONEST! I caught him being a Damn weasel and walking behind me "pretending" not to look but I caught staring right at my screen, so I caught him red handed, weasle that he is he calmly says "well the game doesn't really account for this level of intelligence and I had spies in the local area that were feeding me recon info" !!! Serioulsy smile.gif

Cheaters who are masters will try ANYTHING to cheat and gain an advantage. I'm so glad that BTS has made every possible effort to prevent ANY form of cheating.

Thanks Steve and Charles for KNOWING some people would really want to actually try to seriously cheat to win.

Again I am not refering to anyone on this board, as know of no confirmed cheaters here,

(no suspected cheaters for that matter either) but I'm sure you all know someone like my "friend"

-tom w

Link to comment
Share on other sites

> The suggested new method

> A--view action9, turn10

> B--view action9, turn10, process

> A--view action10, turn11

> B--view action10, turn11, process

Hi Chris,

The problem with the new way explained in really basic terms is:

When A sends his turn after viewing 9 and doing 10, that saved file would have to contain both the movie for 9 _and_ the moves for 10. The way things are done now is that a saved game file contains either moves or a movie but never both. It's the simplest way of doing it. The old way forces the game to hold on to too much old info that you would, as a programmer, rather get rid of as soon as possible.

It sounds deceptively easy to reprogram things so that they work the way outlined in your method, but if you've ever programmed a decent sized app you'd probably realize that it would be a major pain in the butt to do it. Comes down to whether the programmers decide that it's worth the trouble to save us a few e-mail exchanges - so far the answer has been no and it's not too difficult a decision to live with considering how cool the game as a whole is.

p.

Link to comment
Share on other sites

Guest Scott Clinton

I too would dearly love to see the email process sped up. I can see no reason (other than time/more work for Charles) that the process can not be streamlined as outlined above.

As for people cheating...get over it guys!

tongue.gif

I can't throw a rock in the 'Cubical Hell' where I work and NOT hit a programmer that could crack ANY pc game on the market...if they really wanted to.

Streamlining the email process will not make the game any easier to 'crack' at all.

------------------

Please note: The above is solely the opinion of 'The Grumbling Grognard' and reflects no one else's views but his own.

Link to comment
Share on other sites

Guest entec

<BLOCKQUOTE>quote:</font><HR>Originally posted by Scott Clinton:

I can't throw a rock in the 'Cubical Hell' where I work ...

<HR></BLOCKQUOTE>

Ha Ha I've got a corner office, nice view of the river and the ladies walking by. Now if I could only get a better computer so I could play CM at work.

[This message has been edited by entec (edited 05-26-2000).]

Link to comment
Share on other sites

×
×
  • Create New...