Movement

[Exel]

Seeing the new screenshots in the SC2 FAQ portraying the highlighted movement zones reminded me of the old discussion we had on the "Tiles / Hexes" thread about the subject.

Judging from the screenies unit movement costs the same (1 AP) to all directions. The situation and the different alternatives were described and illustrated in this post. Based on that I would highly recommend using a system where diagonal movement costs by default 1,5 AP while moving "normally" to adjacent tiles would cost 1 AP. That way the movement pattern would more closely resemble a circle, ie. de facto movement to all directions would be roughly equal.

Hubert, please read the post in question and consider the choices.

[ February 20, 2005, 03:23 AM: Message edited by: Exel ]

[Hubert Cater]

I have and here is an excerpt from my original reasoning in the beta forums:

I've gone with simply having each tile cost 1 action point regardless of direction. I did experiment with using pythagorean distances for diagonal movement but I felt that the complication far out weighed any of the accuracy benefits. For example it is much more difficult to determine distances when just eyeballing the map.

Here is a sample of an HQ with 3 action points using pythagorean distances for diagonal movement:

Edit: I originally cut and paste the wrong post and picture

[ February 20, 2005, 03:01 PM: Message edited by: Hubert Cater ]

[Exel]

I see and understand your point (after all I did use similar reasoning while trying to prove why hexes were superior to tiles ), but the cruder 1,5x cost for diagonal movement system might actually work better than the pythagoran system you had tested. In the example of the lone tank it would cut off the farthest corners from the movement zone (as shown in the picture below with the thin red lines). It is far from perfect, but imho still better than equal movement cost to all directions. The biggest problem with the latter is that diagonal movement is so highly advantageous - you can cover a lot more area with the same movement points by moving diagonally than you would by going straight, as the thick red arrows in the picture illustrate.

In any case I trust your good judgement. My only concern is that whatever system ends up in the final product is well thought out to be the most suitable choice.

[Hubert Cater]

Oops... looks like I made a small mistake and I've edited my original response, thanks.

Hubert

[pzgndr]

It is far from perfect, but imho still better than equal movement cost to all directions.

I agree to a point, but I'll argue that application in this particular game is not a big deal. I've had an opportunity to play around with the SC2 system and I'll offer a couple of insights.

1. Like SC, SC2 has no unit stacking. So jockeying for position and sequencing your battles becomes an art form almost, akin to chess. Terrain and ZOC between enemy units impedes movement. Enemy units block movement. Friendly units limit tiles available to move to. These penalties are generally more significant than any diagonal bonuses, which both sides can use to equal advantage of course.

2. Doing the math is easy enough, but applying it to cases like motorization research and extrapolating out to custom games with different scales presents problems. Believe me, it started to get complicated. Hubert chose to keep it simple, which is probably best in the long run.

And hey, take another look at those screenshots. Note the highlighted movement paths, the ability to set waypoints, and the ability to move/attack in 8 directions instead of 6. You tell me, when you're in the heat of battle on the east front in 1942 trying to move and fight your army, are you really going to fret about Pythagoras?

[Desert Dave]

As originally posted by Exel:

The biggest problem with the latter is that diagonal movement is so highly advantageous - you can cover a lot more area with the same movement points by moving diagonally than you would by going straight

From posted screenshots - including your own example, it is apparent that there is NOT going to be ability to cover

"A LOT more area."

You will, VERY occasionally, have an extra acre of ground that you can (theoretically) advance.

For, as Bill has eloquently reminded:

So jockeying for position and sequencing your battles becomes an art form almost, akin to chess. Terrain and ZOC between enemy units impedes movement. Enemy units block movement. Friendly units limit tiles available to move to. These penalties are generally more significant than any diagonal bonuses, which both sides can use to equal advantage of course.

Not only the above, but - IF you as defender KNOW that the "motorised" or tank regiments CAN possibly break through on blitz and achieve one more tile in extended advance,

Wouldn't you?

As seasoned and adept Strategic Commander,

THEN

Position some forces to act as stop-gaps, or rear guard "security" or better... as MOBILE reserve.

Isn't that how it was done in WW2?

We are not re-playing WW1. The days of a long thin or thick red line are over.

THIS!

Is SC2-BLITZKRIEG! :cool:

And, you can COUNTER ploy every single perceived advantage.

Plotting & planning AHEAD, you can cut to the quick the slanting "bishop's" ability to gain, MAYBE, one extra tile, yes?

Not to mention, it is the tank formations mostly that will be blitzing through, and so, YOU, as wily defensive genius, MAY have a couple Corps just BRISTLING with anti-tank guns.

Which your opponent was not AWARE that you had, due to FOW.

What was once imagined as a deep thrust into enemy territory, NOW becomes a "tank trap" and those Panzers or T-34's end up as smoking hulks.

Hubert chose to keep it simple, which is probably best in the long run.

Yep, and Bill is right on - again. This is a chess game and it is also a game where you mean to allow - BOTH the "historical imperative" and that which is... "fun to play."

You shall have both.

________________________________

Besides, why would you only wish to have the SAME OLD paradigm - once again?

Isn't that how most game designers do it? Trot out the same maps and resource topography, and perhaps, change the color of the epaulets on the Colonel's shoulders or the shape of the insignia on his collar? Or, some little fiddling around that gives you essentially, the SAME game?

Why wouldn't you wish to have something new! and different and of great challenge, albeit, in a SLIGHTLY altered Gestalt?

I have been playing war-games since the late 1950's, and am VERY accustomed to "hex-games."

Fine, I can always go and set up my cherished A3R, or SC1, and enjoy hexes all the night long.

NOW you have a NEW way to look at your schematic of WW2.

You are... the hunter Hawk flying high and surveying all that array of potential prey.

You are hungry and you are swift and talon-down sharp.

You have the WHOLE of the Western World, from the Potomac River to the distant Ural mountain range to soar around in.

And so, you are going to WORRY UNDULY over - MAYBE, a couple of times a game, you MIGHT gain an extra tile?

You may; I'm not thinking of it that way.

[ February 23, 2005, 05:56 AM: Message edited by: Desert Dave ]

[acrashb]

Simple would be 2 movement points to move laterally, and 3 to move diagonally, with units having double the current ability to move. That gives you close-enough-to-Pythagorean without a lot of math.

Making it 1-per-tile regardless of diagonals will mess things up, particularly at sea and in the air. Long detours will have the same movement impact as a straight line - leading to some weird tactics and the total inefficacy of a point blockade - you'll need a line of ships to do anything useful.

[Edwin P.]

Naval blockades are never perfect. Weather is always a factor. Consider the scale of each Atlantic hex in S2. Consider the ability of Japanese and US Naval Forces to evade each other in the Pacific, look at the ability of German subs to reach Long Island in New York during WWII.

[ February 22, 2005, 01:49 PM: Message edited by: Edwin P. ]

[acrashb]

It's not about a perfect blockade, it's about forcing the enemy to use up time, supply and readiness to go around one - that makes the blockage effective. With 1-per-tile, that doesn't happen.

To show this graphically, a unit that has 10 'movement points' running on the 2-3 system can do a rough circle. This is almost the same 'circle' one would get using (x^2+y^2)^(.5), telling us that accuracy does not materially suffer, at this scale, from approximating Pythagoras:

The empty 'tiles' show the movement with 1-per-tile (at 5 MP's of course). Interdiction (which includes spotting range) of naval vessels would be significantly - and unrealistically - more difficult.

Air movement/attack/spotting would also smell unrealistic once the ranges crept up from the small base at game start.

[ February 23, 2005, 07:15 AM: Message edited by: acrashb ]

[Hubert Cater]

Agreed a circle would be nice but after some considerable gameplay and testing and trying to balance out the benefits of a 3/2 (or pythagorean) system versus the added complications wrt ZOC rules, movement rules, quick surveys of the map I'll just reiterate what Bill and Dave have mentioned so far that the simplier 1 per tile really was the better choice.

Believe me we did fiddle around with it and while there is some give and take with each system, once you are in game and playing and moving units around the chances are you'll forget that your units are not moving in a perfect circle and/or really care for that matter... and of course once the demo is available you will be able to judge for yourself.

[ February 23, 2005, 08:39 AM: Message edited by: Hubert Cater ]

[Desert Dave]

For the sake of argument, acrashb,

Let's re-consider Gestalt, and thus, as this:

You are presuming that every unit, ground, sea or air, is BEGINNING it's move

From - the center point.

And, that NOTHING ever impedes your progress.

Whereas,

I am presuming that each unit begins from THE EDGE.

IE, if you have a HEX, and each unit commences from the edge of the hex, and NOT the exact center, then, surely, you will end your moves in a very "ragged arc" and not a perfect circle, true?

IF you have a square, with each unit commencing as above, then, certainly, you will end your moves as a "ragged squared pattern."

And so, begin from the angled point of the square, and you surely will get further on out there, UNLESS! :eek:

As above mentioned,

(... and leaving out such considerations as "Zeno's fallacy," IE, if you go half way from point A to point B, and then half again, and etc, you'll NEVER get to point

Those fey and ipso-facto-dramatic intruders... somehow impede your progress!

Such as, if you are imitating a ground unit, there are... hills & dales, and swamp-flush swales and also, maybe - pause-causing Tales told you by an Idiot, or a Chaucer type traveler on a mule.

Or, should you be an air unit, what of that head-wind or tail-wind?

Is the ocean or atmosphere EVER free of Gordian knotted wind or wave, or other stray clouds or enemy intruders... that might cause you, at least, caution?

Or, should you be another different ground unit, beginning it's move from another different edge of the hex or square - what of that... sudden stomach-ache suffered by your dauntless Commander, who then orders a brief halt to the long march?

You CAN view the World, or a game , as a linear logically positive exercise, you bet!

[... anybody else out there? Ever play a computer game and have the abrupt strange notion, all at once, that ALL you are doing is shoving one stack of statistics at another stack of statistics?]

Or, you can view the world, or the game, as a work of... "imperfect as it MUST be"... ART. :cool:

For instance, you might consider the ROSE that you see from your near-garden seat.

From where you sit, perhaps sipping your ale or ade, it is nearly PERFECT (... RE: as Plato's "perfect forms").

Yet, unlimber yourself from that chair, and meander over and - what do you notice, if only eventually?

That bit of blight (... RE: William Blake's poet-formed "worm") on the furthest most petal, eh?

Look.

You are playing a GAME.

You are not interested in duplicating some PERFECT expression of linear logic, on some precisely concised GRAPH, are you?

You would rather have FUN!! (... all the while remaining fair faithful to "historical realities") wouldn't you?

Well, I have played many IMPERFECT games (... how could they not be? Cardboard or pixels, there is NO WAY to duplicate ANYTHING exactly) and had the time of my life!

Yipee! Yay! Hooray! :cool:

I have NEVER, however, EVER played any game on a PERFECT medium, or as PERFECT expression of...

What is.

What was,

Or, what will be.

Who would want to?

[ February 23, 2005, 08:51 AM: Message edited by: Desert Dave ]

[acrashb]

Originally posted by Desert Dave:

You are presuming that every unit, ground, sea or air, is BEGINNING it's move from - the center point.

It's a reasonable presumption. As for the rest of your creative post, perhaps you have a career in street-corner beat poetry

Thanks for the response, Hubert. I'll just have to wait for the demo.

[Exel]

I'm not going to argue about the reasons you gave that work against a system where diagonal movement has a different cost than normal movement in the game engine itself, since I obviously don't know how it is built or how it works. But I have to contradict one argument: it is not harder but actually easier to quickly estimate movement/action distances on the map if the movement patterns were more circle-like than squares, at least with my perception. That is in fact the main reason I am for the 3/2 diagonal/straight movement cost system. It just is more natural.

Still, the more technical reasons are also important. If the engine works better with an equal-in-all-directions movement system, then so be it, I wont argue about it any further. But I'm going to give one more example to highlight the possibly biggest flaw with the aforementioned system:

If movement to all directions has the same cost, diagonal movement becomes very advantageous. In fact so advantageous that moving along the squares makes - in theory - no sense. The longer the distance a unit can move, the more this behaviour is emphasized. With ground units it is not so much of a problem since the movement distances are limited and the fronts are usually cluttered with units. But it is ever more important for air and naval units that often have wide, unblocked operation range. It is perhaps most exploitable in scouting an area, as demonstrated here:

If diagonal movement is 1:1 to normal movement, the red and orange paths cost as many APs. But with the straigth red path the ship can only detect the area surrounded with the dashed red line (with 1 tile detection range, 24 new tiles uncovered), whereas with the orange diagonal zig-zag path it could see 12 more tiles. Yes, even though it may not look like it in the picture, it's 50% more area!

[Hubert Cater]

Originally posted by Exel:

But I have to contradict one argument: it is not harder but actually easier to quickly estimate movement/action distances on the map if the movement patterns were more circle-like than squares, at least with my perception.

Hmmm... to each his own I guess, it was just felt that with a 1-per tile system there would be no need to actually estimate as you can count out the distances accurately (and in mind a little more easily) including ZOC penalties. For example, how far does my Air Fleet intercept range cover, can that tank reach Warsaw in the next turn? etc.

Sure you can do it with a 3-2 per tile system but it initially seemed more involved. Keep in mind this was not a philosophical design decision or a choice that I believe to be superior in all cases, rather just the one I went with that I felt was best suited for this type of game

Yes, even though it may not look like it in the picture, it's 50% more area!

Wasn't the navy always instructed in the advantages of moving in a zig zag versus a straight line?

[Iriemon]

Why are there concerns about calculating movement costs? That might have been a concern in the old cardboard days. Since SC had the excellent feature of highlighting where a unit could move, there is no human calculation involved.

The only question about whether diagonal movement is 1, 1.5 or its actual legnth, 1.414 (sq rt of two) should be how it practically affects the movement of the pieces. If the cost is "1" for a diagonal move, you can effectively move due N S E W 41% faster than you can move NW-NE-SW-SE. As a concept that doesn't make a lot of sense, but as a matter of gameplay maybe it doesn't make any difference

[Hubert Cater]

This is true and my hat is also off to Excel for his excellent analysis, yet there are still the cases where, despite the highlighting features of SC and newly incorporated ones of SC2, you might be interested in what you and your enemy will be capable of in the next turns etc., like can his or your units reach a certain tile and so on.

So, like I've mentioned, besides game engine mechanics this was of some concern as well.

Hubert