Comprehensive Wishlist

[Panama]

ORIGINAL: Curtis Lemay

ORIGINAL: Panama

ORIGINAL: Curtis Lemay


Depends upon how much the river meanders. They don't really move in straight lines right on the hex grid. But, this has been discussed to death. I'm not going to rehash it.

Please refer to bottom line of sig.

I'm just saying that I've said my piece about river hexsides and haven't anything else to add. You guys can knock yourselves out, though.

Understood. [;)]

[Panama]

ORIGINAL: Curtis Lemay

ORIGINAL: Panama

So then you'll also fix the problem where a unit has to pay river crossing costs even if it doesn't cross the river, yes? Simply moving parllel to a river costs the same as if you crossed it. If you're going to clean it up might as well clean it all up or not at all. [;)]

Also, you don't move "parallel" to the river. You move upon the river. So the game can't know if you cross it or not.

Thank you. Finally, some agreement about the problems. I missed that. I probably blacked out when I read it. [X(][X(][X(]

And that is the problem. The game doesn't know if you cross or not. Exactly the same as the game not knowing if a road/rail crosses or not. It just knows it's in the same hex and treats it all as if it's a bridge. So actually, there are no bridges in TOAW. There are just roads/rails and rivers that just happen to be in the same hex making it all arbitrary and pointless to have a bridge at all.

So, let's make it simple. Since bridges don't exist in the present form of TOAW we'll just do away with them all together. After all, they probably made no difference in a battle or campaign since the game treats them as non existant.

I don't know why Norm designed it this way instead of hex side rivers. He probably doesn't care since he now has bigger fish to fry. It really does need to be changed.

[Curtis Lemay]

ORIGINAL: ColinWright

It seems to me that the program merely needs to identify when a road or rail line connects to a non-river hex.

This would create the possibility that both the road and the rail bridge could be blown when only one had left the river -- but that would still be an improvement over the current situation. So it can still be blown in that case. Oh well. Still an improvement over the current situation -- where both can be blown even when neither has exited.

I just want to point out (using this quote from near the start) that there was a time in this discussion, that even Colin was at least partially sane. Then things went south for him and, as usual, he went stark raving mad. I'm used to it.

He found his one artificially contrived counter example and now all roads are Interstate highways weaving through canyons. So any road running parallel to a river must be assumed to be switching back and forth from one bank to the other. Absolute rubbish in 99.99% of cases, but that doesn't matter to Colin. It's his straw and he's clinging to it.

Note that this was the original issue:

ORIGINAL: Oberst_Klink

I might be off subject, but - the roads/RR parallel to rivers. Are there any plans for a routine that simply prevents that they're treated as bridges? When blown up (because they're treated as bridges) you need RR engineers to fix'em. This would be a good thing to change.

Note that the Matrix fixes this. Cheaply. Nothing else does.

[Curtis Lemay]

ORIGINAL: Panama

The game doesn't know if you cross or not.

For the case where the unit is not moving by road. When it does move by road, we can know whether it crossed or not. Just take the map literally. Assume the map maker meant what he made.

[ColinWright]

ORIGINAL: Curtis Lemay

ORIGINAL: ColinWright

It seems to me that the program merely needs to identify when a road or rail line connects to a non-river hex.

This would create the possibility that both the road and the rail bridge could be blown when only one had left the river -- but that would still be an improvement over the current situation. So it can still be blown in that case. Oh well. Still an improvement over the current situation -- where both can be blown even when neither has exited.

I just want to point out (using this quote from near the start) that there was a time in this discussion, that even Colin was at least partially sane. Then things went south for him and, as usual, he went stark raving mad. I'm used to it.

He found his one artificially contrived counter example and now all roads are Interstate highways weaving through canyons. So any road running parallel to a river must be assumed to be switching back and forth from one bank to the other. Absolute rubbish in 99.99% of cases, but that doesn't matter to Colin. It's his straw and he's clinging to it.

Note that this was the original issue:

ORIGINAL: Oberst_Klink

I might be off subject, but - the roads/RR parallel to rivers. Are there any plans for a routine that simply prevents that they're treated as bridges? When blown up (because they're treated as bridges) you need RR engineers to fix'em. This would be a good thing to change.

Note that the Matrix fixes this. Cheaply. Nothing else does.

Mein Gott. First off, I've provided three cases: one being the Thames, which certainly doesn't flow through a mountain canyon. You're provided none -- except your TOAW screenshot, which as pointed out, doesn't demonstrate what you appear to think it does. Secondly, the 'artificially contrived counterexamples' were simply the actual cases of roads running along rivers that I'm intimately familiar with. The Sacramento, for example, is the closest river of any size to my house. Roads routinely cross back and forth over rivers, or don't, but in any case, what's on a TOAW map won't tell you if they do or where they do.

It's amazing that you're revealing so much of yourself. We have what is clearly an untenable proposition -- this 'matrix' idea of yours, and the notion that it can divine where the bridge would be on a TOAW map.

I'll credit you with enough intelligence to see that the idea is obviously faulty. The map simply doesn't provide the information needed to make such a determination. And yet you feel compelled to continue to defend it. Why?

[Curtis Lemay]

ORIGINAL: ColinWright

Mein Gott. First off, I've provided three cases: one being the Thames, which certainly doesn't flow through a mountain canyon. You're provided none -- except your TOAW screenshot, which as pointed out, doesn't demonstrate what you appear to think it does. Secondly, the 'artificially contrived counterexamples' were simply the actual cases of roads running along rivers that I'm intimately familiar with. The Sacramento, for example, is the closest river of any size to my house. Roads routinely cross back and forth over rivers, or don't, but in any case, what's on a TOAW map won't tell you if they do or where they do.

The Thames isn't an example of this at all. Yes, it's got lots of bridges - but for the normal reasons - its at a major road network with roads crossing it all over the place. It isn't a single road running along a river snakeing back and forth due to canyon walls. I expect that's much the same for the Sacramento river near your house. In that case, the TOAW map will have those crossroads, just like the real world - and there will be no problem. (See my France 1944 map).

I-40 travels adjacent to the Arkansas River for about 200 miles (I was on it just last week). It crosses it once - when the river finally turns north. There are lots of bridges over the Arkansas, but due to crossroads, not I-40 snakeing around. You'll see the same thing around the Mississippi River as well. Bridges over militarily significant rivers are expensive. They're only built when necessary.

And, as I've repeatedly pointed out, even for that one artificially contrived example, the map designer would have been compelled to make the roads cross the river at the appropriate places for appearance purposes! (That's assuming the river was even militarily significant enough to include - most canyon rivers can be waded over most of the year).

There is only a problem if the map designer was oblivious to what his map looked like. That's not going to be the case. But, even if it is in some rare case, the matrix can be made optional.

Now for the good news: Yesterday, I took a stab at creating the map for the matrix. Editing it in XML allowed me to complete it in about an hour. (Doing it in the map editor would have taken forever). It's attached if anyone wants to see it. The river permutations run vertically, while the road permutations run horizontally. A bonus is that I'm now familiar with the XML map format (although I don't think I would ever edit a real map this way) [:)]

Next, I have to pick the permutations that have bridges in them. That may take a few hours more. Then, I'll translate the picks to text and the Matrix will be ready.

[ColinWright]

ORIGINAL: Curtis Lemay

ORIGINAL: ColinWright

Mein Gott. First off, I've provided three cases: one being the Thames, which certainly doesn't flow through a mountain canyon. You're provided none -- except your TOAW screenshot, which as pointed out, doesn't demonstrate what you appear to think it does. Secondly, the 'artificially contrived counterexamples' were simply the actual cases of roads running along rivers that I'm intimately familiar with. The Sacramento, for example, is the closest river of any size to my house. Roads routinely cross back and forth over rivers, or don't, but in any case, what's on a TOAW map won't tell you if they do or where they do.

The Thames isn't an example of this at all. Yes, it's got lots of bridges - but for the normal reasons - its at a major road network with roads crossing it all over the place...

...and you lead off with an assertion that simply and obviously ignores the point I made when I discussed the lower Thames. There isn't any point in trying to discuss this issue with you. You don't read the responses.

If you want anyone to take you seriously, go back, find my post on the Thames, read it, and respond to what was said.

[ColinWright]

ORIGINAL: Curtis Lemay


Now for the good news: Yesterday, I took a stab at creating the map for the matrix. Editing it in XML allowed me to complete it in about an hour. (Doing it in the map editor would have taken forever). It's attached if anyone wants to see it. The river permutations run vertically, while the road permutations run horizontally. A bonus is that I'm now familiar with the XML map format (although I don't think I would ever edit a real map this way) [:)]

Next, I have to pick the permutations that have bridges in them. That may take a few hours more. Then, I'll translate the picks to text and the Matrix will be ready.

This is almost as exciting as hearing that my broker has worked out how to pick my stocks with an ouija board.

Actually, it's worse. I could easily change brokers. How do I move you someplace safe?

...and I see no more interest to look at your attachment than I would to look at the program my broker worked out for picking my stocks. Both are based on a transparently silly premise. No doubt you can write a program that will label some hexes 'bridges' and others not, but the premise -- that the TOAW map offers the information necessary to make this determination -- is clearly invalid.

As explained, there is one and only one case where from the information contained in a TOAW map, you can determine that a particular hex does contain a bridge. That is when the road enters the river from a hex that does not lie on that river and immediately exits to another hex that does not lie on that river. Then you can know that there is in fact a bridge in that hex.

But the program already assumes that there is a bridge there -- so the Matrix could hardly offer anything there. For all other cases, there might or might not be a bridge, and a TOAW map contains no information allowing us to definitively decide one way or the other. Any 'program' that decides otherwise is simply working with a faulty premise.

Take your Arkansas River. I've driven along it, and actually, it follows a progression similar to the case of the Sacramento I discussed. It starts out flowing down a mountain canyon, then enters flatter country. So the highway crosses your Arkansas only once, and at a point your Matrix would pick out.

So what? The highway crosses my Sacramento at five points, none of which your Matrix would divine. And it gets worse. As it happens there isn't a bridge at either of the two points where your Matrix would decide there was a bridge.

It's pure nonsense. Worse, mischievous nonsense. I have no doubt that my broker's Ouija board would make correctly decide to go long or short 50% of the time. It's what it would do the other 50% of the time that disturbs me.

[ColinWright]

ORIGINAL: Curtis Lemay

ORIGINAL: Panama

The game doesn't know if you cross or not.

For the case where the unit is not moving by road. When it does move by road, we can know whether it crossed or not. Just take the map literally. Assume the map maker meant what he made.

Nonsense. The map maker meant that the road lies in the same hex as the river. That's all. Any other meaning is entirely speculative.

And that's all. In some cases, you can lay the road one hex off the river -- but then why have the Matrix at all? The designer must have wanted there to be a bridge if he put the road in the same hex as the river. Ergo, all road/river hexes are bridges -- which is the situation now.

Worse, laying the road one hex off the river has at a minimum the drawback that it distorts the map. Actually, the road is not there. Some of us find that disturbing.

More importantly, laying it off a hex may move it above escarpments and into mountains or whatever rather than flat ground, simply put it on another river, or cause there to be a road junction where in fact there wasn't one.

I resort to the device, but it causes a number of problems, and often, the best choice is to lay the road along the river even when there wasn't a bridge. Certainly in most scenarios the roads clearly run along rivers without regard to whether a bridge existed at that point or not.

Really, and if this discussion was happening with anyone else, the choices long ago would have been narrowed down to three:

1. Do nothing. The current situation can't be improved.

2. Go with hex-side rivers. Personally, I find them ugly, but the arguments are strong...

3. Allow the designer to label which hexes -- river or non-river -- contain roads that can be 'destroyed.' This has the advantage of (a) allowing the designer to partially solve the problem we're supposedly discussing, and (b) allowing designers to extend the 'bridge destroyed' effect to cover any other situation in which they feel a road can be rendered useless.

I like (3). Others like (2). There's always an argument for doing nothing. It's all good, and whatever people say will at least in theory contribute towards making the best decision. But no...

Instead, we've got you clinging like a leech to your nonsensical 'matrix,' and the argument has become utterly sterile. It's a complete waste of time, and I'd ignore it except for the fact that you potentially have some say in how this game is modified, and I'd just as soon not see it made worse.

[ColinWright]

ORIGINAL: Curtis Lemay

I-40 travels adjacent to the Arkansas River for about 200 miles (I was on it just last week). It crosses it once - when the river finally turns north. There are lots of bridges over the Arkansas, but due to crossroads, not I-40 snakeing around. You'll see the same thing around the Mississippi River as well. Bridges over militarily significant rivers are expensive. They're only built when necessary.

Amusing. As it happens, I-40 joins the Arkansas at Little Rock. No bridge there: the road came in from the north and stays on the north bank. It stays on the north bank until south of Tulsa, where it crosses above a major impoundment and continues west while the Arkansas heads north.

From what's on a TOAW map, how can you know the road crosses there and there only?

Maybe it crossed at Little Rock and has been on the south bank all this while -- so no bridge when the river turns north. Maybe it ran along the north bank until there was a town to serve on the south bank and crossed at that point. Maybe it stayed on the north bank until a good site for a bridge presented itself.

How can you know? And by what alchemy will your Matrix decide there is a bridge at the western end but not at the Little Rock end? Will it put bridges at both points? In point of fact, I-40 does not cross the Arkansas at Little Rock. Other highways certainly do -- but you already ruled those out.

Bridges may well be built only when necessary -- but you have no means of divining where they will be built.

Colin is wandering along a trout stream back towards camp. Which side of the stream is he on? The same side as camp? The opposite side? If the opposite side, when will he cross? Immediately, and then work his way back towards camp? Only when he is opposite camp? At some point in the middle, when an attractive fishing point or some safe-looking rocks, or a stretch of wadeable water presents itself?

Unless it's a very small stream, or I see something big boiling the water, I'll probably cross only once. Colins don't like to cross rivers any more than highways do. But where? You don't know. You can't.

And something big may indeed boil the water -- or a stretch of boggy ground will present itself along the bank that I'm on, or one bank will obviously offer multiple fishing points while the other does not. I could indeed cross several times. You have no way of deciding that either.

Let's take that southernmost crossing on the Sacramento I discussed earlier. Why there? From a TOAW map, no particular reason. The road's been running happily along the west bank for some time. Why cross now and run along the east bank? It would certainly be possible to keep running along the west bank -- the country's getting to be low hills, but nothing to speak of.

Well, as it happens, the river does veer slightly to the west -- nothing to take it out of the TOAW hex, but a couple of kilometers looks more impressive on the ground than it does in TOAW.

Probably more to the point, the communities to be served -- Corning, Red Bluff -- had been on the west bank. Now they're going to be on the east bank. Anderson, Redding.

So the highway crosses the Sacramento. It's not a decision to be taken lightly -- the Sacramento is a deep, wide, fast-flowing river at that point -- but it does have to be done somewhere. And it is time to cross over. And there is solid rock for bridge piers. And the jog in the river means the highway is approaching the river at a right angle to begin with.

And so -- for a number of reasons that wouldn't show up on a TOAW map -- the highway crosses the Sacramento there. Not down at Red Bluff. Not up at Redding.

There. And there's no way that can be picked out from the TOAW map.

[Panama]

This is not a strictly road/rail situation. So long as rivers, cities, roads and rail can occupy the same hex there will always be a problem making an accurate map. The only way to truely resolve this is to have rivers occupy hex sides instead of hexes. There are many cities in the world that occupy one side of a river and not the other yet you can't portray that either.

It's obvious that as long as where the game places rivers is not resolved this will not go away. Changing where rivers/wadis/canals are placed and using a legacy button so as not to render older scenarios useless would be the proper solution. Anything else is duct taping things. I think this will never come to pass. No sense making a matrix, just more duct tape.

[ColinWright]

ORIGINAL: Panama

This is not a strictly road/rail situation. So long as rivers, cities, roads and rail can occupy the same hex there will always be a problem making an accurate map. The only way to truely resolve this is to have rivers occupy hex sides instead of hexes. There are many cities in the world that occupy one side of a river and not the other yet you can't portray that either.

Yeah. See Stalingrad. It's wandering into my current project -- fortunately only as a bit of eye candy off on the edge of things.

At 10 km/hex, it's indubitably in the same hexes as the Volga. And yet -- if it had to be fought over -- having it there would put the defenders at a considerable disadvantage precisely where they acquired a considerable advantage.

I can't see what a truly satisfactory solution would be. Stick Stalingrad out in the steppe where it wasn't? Have the Russians make their stand west of the city when they didn't?

It's obvious that as long as where the game places rivers is not resolved this will not go away. Changing where rivers/wadis/canals are placed and using a legacy button so as not to render older scenarios useless would be the proper solution. Anything else is duct taping things. I think this will never come to pass. No sense making a matrix, just more duct tape.

My point is that Curtis's 'solution' is worse than duct tape; it's going to the herbalist when when you've got appendicitis.

Curtis' solution won't do anything -- except make it obscure what is or isn't a bridge. It certainly won't reliably put bridges in the right places. At least now, one knows when the program is going to call a road a 'bridge' -- when it's in the same hex as a river. We'd lose even that. Unsatisfactory as it is, at least the current situation is visually clear and easy to manipulate. Curtis would manage to add night and fog -- and that's it.

We'd have hexes being labelled bridges according to a paradigm which can only coincide with reality by the laws of chance.

I fully appreciate your arguments for hexside rivers. If they came to pass, I wouldn't mind much. I'll just note that (a) I don't like how they look, and (b) that there are a few related questions that would have to be addressed. Like, how would you 'blow up' the bridge and how would you fix it? That'd take some programming. Then the program would have to be told what was a double-sided escarpment that was a river and what was a double-sided escarpment that was a double-sided escarpment.

And riverine movement? These aren't insurmountable obstacles, but they are obstacles.

I prefer keeping things as they are visually but letting the designer designate what is a road that can be destroyed and what isn't. That addresses some problems your solution doesn't and doesn't address some that yours does. For one, as noted, it allows the 'destroyed bridge' to be used to simulate all kinds of things besides river crossings. It would also seem to offer fewer programming challenges.

[ColinWright]

I'll list the choices and their advantages and disadvantages, as I see them.


Curtis's Matrix.

Disadvantages. Makes it less clear than it is presently what is or isn't a bridge. Can't actually predict which hexes contain bridges with any accuracy at all.

Advantages. None.


Hex-Side rivers.

Disadvantages. Considerable work to program. Alters the aesthetics of TOAW. Perhaps some problems with correctly simulating river valleys lined with escarpment.

Advantages. Allows bridges to be placed with perfect accuracy. Distinguishes between movement along rivers versus across rivers. Permits defenders to always gain an advantage from being behind a river -- even when there's another river in the next hex.


Designer determined 'destroyable roads/rails.'

Disadvantages. Leaves several shortcomings of the current system unaddressed.

Advantages. Should be relatively easy to program. Expands the potential uses of 'destroyed bridge.' Allows bridges to be placed with perfect accuracy. Doesn't alter the appearance of TOAW.

[Curtis Lemay]

Ok, let me see if I have this straight:

Because there is a chance that a road running along (but not crossing) a river in an existing TOAW scenario may (in the real world) have been switching back and forth across it repeatedly AND a chance that the map maker may have not represented this, THEREFORE, players should be FORCED to consider all road-river combinations as blowable bridges. They cannot be given an OPTION to consider such bridges as non-blowable.

In other words, we've all been delusional. There was no problem at all. Colin has solved the problem by declaring it void. Thanks a lot.

Don't worry folks, I'm not going to let this endless deluge of stupidity deter me. The Matrix's irresitable truth marches on.

I'm happy to anounce that I've made the picks on the map (scenario attached). And have converted the picks to a spreadsheet that contains the Matrix (included in the attachment).

I still need to confirm that other river/canal types or the other road type or rail has the same matrix. Stay tuned.

[Curtis Lemay]

And here's a screenshot of the contents of the Matrix. This is ready to become a constant array in the TOAW code. Then, to use it, the test for whether a hex contains a bridge will simply be modified from:

River AND Road

to:

River AND Road AND Matrix(i,j)

Where i = the Road permutation and j = the River permutation.

Trivial.

[Curtis Lemay]

ORIGINAL: ColinWright

b]Curtis's Matrix.[/b]

Disadvantages. Makes it less clear than it is presently what is or isn't a bridge. Can't actually predict which hexes contain bridges with any accuracy at all.

False. It will be crystal-clear what is or isn't a bridge: any hex where the road graphically crosses the river. That's good for players, since they will have a visual cue tellling them. In contrast, it's obviously not clear to them which are bridges now - just read this thread. The Matrix works within the current system - there remains one tile type for each river type. Designers won't have to switch back and forth between tile types to model bridges.

And, of course, it does predict which hexes contained bridges in old scenarios with a great deal of accuracy - because the map maker was compelled to graphically show his roads crossing rivers in the correct places for appearance purposes. Your lies don't become truths just because you repeat them.

In fact, since it will be easily made player-optional, it can't have any disadvantages at all.

Advantages. None.

The advantages are huge. First, it is the only option that addresses the hundreds of existing scenarios - the vast majority will never be modified again. Second, it is trivial to code. Third, it gives the designer of future scenarios full control of which river-road hexes will be bridges - without requiring any new tiles.

Hex-Side rivers.

Disadvantages. Considerable work to program. Alters the aesthetics of TOAW. Perhaps some problems with correctly simulating river valleys lined with escarpment.

Also: Doesn't help a single existing scenario. And, at the operational scale, rivers don't just have hexside properties. They also have area properties (due to meandering & tributaries). Those will not be represented by hexside rivers.

Advantages. Allows bridges to be placed with perfect accuracy. Distinguishes between movement along rivers versus across rivers. Permits defenders to always gain an advantage from being behind a river -- even when there's another river in the next hex.


Designer determined 'destroyable roads/rails.'

Disadvantages. Leaves several shortcomings of the current system unaddressed.

Doesn't address a single existing scenario. More difficult to program than the Matrix. Dubious whether all designers would know to use it. Requires switching back and forth between river tile types while mapping. Will not give the players a graphical cue that there is a bridge in the hex.

Advantages. Should be relatively easy to program. Expands the potential uses of 'destroyed bridge.' Allows bridges to be placed with perfect accuracy. Doesn't alter the appearance of TOAW.

Dubious "potential uses". And, it places bridges with no more accuracy than the Matrix.

[ColinWright]

Sure Curtis. You realize your statements have started to become transparently indefensible?

Take 'And, it places bridges with no more accuracy than the Matrix.' Why bother to rebut that?  

[ColinWright]

ORIGINAL: Curtis Lemay

And here's a screenshot of the contents of the Matrix. This is ready to become a constant array in the TOAW code. Then, to use it, the test for whether a hex contains a bridge will simply be modified from:

River AND Road

to:

River AND Road AND Matrix(i,j)

Where i = the Road permutation and j = the River permutation.

Trivial.

Happily, your own lat-long program allows an objective means of mapping. Now, pick a list of rivers via some random technique: say all names with seven letters from a list of 'world's thousand longest rivers' or something.

Then, map at...10 km per hex, shall we say? Use the largest city on the river as the midpoint of a 21X21 array. Put in only whatever roads are shown at a given height on Google Earth.

See where your 'Matrix' puts bridges. Then ascertain where they are along those roads in reality.

Time-consuming, but the results should be comic.

[ColinWright]

ORIGINAL: Curtis Lemay



The advantages are huge. First, it is the only option that addresses the hundreds of existing scenarios - the vast majority will never be modified again. Second, it is trivial to code. Third, it gives the designer of future scenarios full control of which river-road hexes will be bridges - without requiring any new tiles.

It's a pity that the program has little better than a random chance of correctly placing bridges.

Sounds great otherwise.

[Curtis Lemay]

ORIGINAL: ColinWright

It's a pity that the program has little better than a random chance of correctly placing bridges.

I concede that you can repeat your nonsense indefinitely. That doesn't give it any validity.

When a TOAW player sees his opponent blow a long line of road that ran along a river, but didn't cross it anywhere, he doesn't think that the road must have been switching from one bank to the other every hex. He thinks his opponent has just made a gamey abuse of the system. And he's right. Everybody knows this. That's what this entire discussion has been about. If it wasn't true, the issue never would have even been raised.

There are two reasons he's right:

1. Isolated roads running along a river rarely switch back and forth across it, if the river is militarily significant. Your artificially contrived nonsense notwithstanding.

2. The map designers are conscious beings. They are aware of their map's appearance and thereby are compelled to make it look right, for appearance purposes. That means making the TOAW road visibly cross the TOAW river where it physically does so in the real world.