"And the winner is . . .

[Tris]

. . . Beetle Bomb!"

"You heard that right folks. Just moments ago, in the annual CMBO Sprint Sweepstakes, run over 1000 meters on grass somewhere in the heart of Europe, the most unlikely of nags. . . . "

So here's the poop.

I selected 16 M4 Sherman tanks with Elite crews to run on parallel tracks over dry grass. In lanes 1-4 we had four Shermans ordered to complete the course at Fast, in lanes 5-8 the next four Shermans were under orders to Move over this course, lanes 9-12 we given the Hunt command, and the final four entrants were rotated around 180^ before the start and ordered to go it in Reverse.

The fastest time recorded over the 1K meters was 3 minutes 31 seconds by the Sherman in lane #2. At this juncture the fastest Sherman ordered to Move sat at the 330m marker. The fastest Sherman in Reverse came in under the wire at 4:51. The fastest Hunter completed the course in 6:20. The fastest Move Sherman had by this time reached the 600m marker, and this tank finally crept in under the wire at 10:43.

Perhaps this should come as no surprise to us. Appendix B in the manual, p. 157 clearly states that the Move order will direct a vehicle to move at walking speed, whereas a Hunt order will direct the same vehicle to proceed at a medium speed. Now while it isn't clearly stated which speeds will be the highest--except that Fast will move a vehicle at maximum speed--it seems now that our clue to have picked up on was "walking speed." I must say all along I'd supposed that Reverse would be the slowest of the speeds, with Hunt slower than Move. Now I know better.

Test Parameters:

16 M4 Sherman tanks w/Elite crews divided into four groups of four Shermans each with orders to move over a dry grass course of 1000 meters at group speeds of Fast, Move, Hunt and Reverse.

Results:

Speed Time

FAST 3:31

FAST 3:31

FAST 3:32

FAST 3:32.5

MOVE 10:43

MOVE 10:44

MOVE 10:45

MOVE 10:46

HUNT 6:20

HUNT 6:20.5

HUNT 6:21

HUNT 6:21

REVERSE 4:51

REVERSE 4:51

REVERSE 4:53

REVERSE 4:53

[This message has been edited by Tris (edited 11-29-2000).]

[Joeri]

That's funny, I did the exact same (almost) test last night as part of my bogging chances analysis. I came to the same conclusion as you (also expecting the hunt command to be slower as the move). I didn't do it as accurate as you by the way, I thought hunt and reverse were the same speed.

Try it in mud (you might be surprised).

I tested it to get the chances for different movement orders in bad conditions and I was quite surprised about the results. You can read about it in my previous post 'statistics on bogging' if you're interested.

Joeri

[Tris]

I thought I would next examine the speed of the M4 Sherman tank ordered to Move in relation to foot travelers.

Test Parameters:

Eight unique classes of Elite foot travelers (all in command) and one M4 Sherman tank over a dry course of 1000 meters ordered to Move.

Results:

Time

Platoon HQ 10:40

M1919 MMG 10:41

M4 Sherman 10:43

Sharpshooter 10:43

60mm Mortar 10:44

.50 cal HMG 17:46

Flamethrower 17:46

81mm Mortar 17:48

M1917 HMG 17:50

Note: As you can readily see this test delineated two distinct groupings for speed, with all classes of travelers closely matched with one another within each of the two speed groupings. Platoon HQ led all the way and reached the 500m mark in five minutes and 34 seconds. The M1919 MMG was close behind (5:34.5). Of the slower grouping, the .50 cal team led by a nose at the 500m mark, recording a time for halfway over the course of 9:16.

Of interest is the fact that all of the travelers completed the second half of the course in a considerably faster time, so much faster that it is hard to account for it by any possible delay caused at the startup of the race due to a CC issue. (My observation was that all travelers seemed to get out of the blocks in good order.)

[This message has been edited by Tris (edited 11-23-2000).]

[Tris]

Say, Joeri, don't either one of us sleep?

(andimustvisitmybrotherlaterandpretendilikevickisturkey)

[Joeri]

Do I understand correctly that you are saying that they move faster the second part of the course?

If I make a quick calculation from your post, the HQ does the first part in 5.34 and the second part in 5.06. Those 28 sec difference are partly due to command delay. Say 8 sec for elite HQ's. Leaving 20 sec difference??

That's strange. How did you measure the distance? Did you use the terrain tiles with gridlines?

Maybe it has something to do with the unit fatigue level. Of course then you would expect it the other way around. Maybe if the troops are rested they can't walk that fast because they have stiff legs. When they get warmed up (ready) they walk a little faster

By the way, I do get sleep. I just don't have a social live (well, at least on tuesday and wednesday evening )

Joeri

[Tris]

Do I understand correctly that you are saying that they move faster the second part of the course?

That was the case in this test.

If I make a quick calculation from your post, the HQ does the first part in 5.34 and the second part in 5.06. Those 28 sec difference are partly due to command delay. Say 8 sec for elite HQ's. Leaving 20 sec difference??

Just so.

That's strange. How did you measure the distance? Did you use the terrain tiles with gridlines?

Yes, I marked off the distance carefully, placing a flags at the 500m mark and finish line. All travelers were routed straight along the grid lines my grass terrain is modified with.

Maybe it has something to do with the unit fatigue level. Of course then you would

expect it the other way around.

Quite.

Maybe if the troops are rested they can't walk that fast because they have stiff legs. When they get warmed up (ready) they walk a little faster

Interesting rationale, Joeri. Assuming Steve's off today you might think about standing in for him. Charles would be pleased.

[coralsaw]

Tris,

Would it be possible to run a test to see how far infantry units can move in one turn.

[Preempt ON]

Yes I know math, I can't isolate C&C delays.

[Preempt OFF]

Thanks Tris.

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

My squads are regular, must be the fibre in the musli...

[Tris]

Yes, Coralsaw, it would be possible to run such a test.

[Tris]

And I think I can now accurately predict that the longer they move (at least over 1000 meters) the faster they will likely progressively travel.

I ran another quick test, again over a course of dry grass for 1000 meters with four M4 Sherman tanks to see if anything different might show with them. The Sherman ordered to move Fast finished the course in 3:30, with a clocking of 1:49 at the 500m mark, which means it completed the second half of the course in 1:41, a savings of eight seconds. The Sherman ordered to Hunt finished in 6:19 with a halfway timing of 3:13 (savings of seven seconds), the Sherman ordered to Move came in with a total time of 10:40 and was 5:21 for the halfway mark, a second-half savings of two seconds.

My fourth test vehicle was ordered to Reverse and finished the first half of the course with a time of 2:58 (this amounted to a considerable lead over the Hunt tank). It then bogged down at the 778m mark, became immobilized and did not finish the test.

[This message has been edited by Tris (edited 11-23-2000).]

[Juardis]

Very interesting. Do way points slow things down? For example, put waypoints every 250m and see if they stop or slow down at each waypoint before continuing on.

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

Jeff Abbott

[Tris]

I have not included waypoints to my test parameters but my observation ingame is that the only delay entailed would amount to the time required for the ordered unit to affect change of direction--call it less than a second for infantry, next to nothing for jeeps on roads, more and dependent on the number of degrees turned for tanks, etc.

[Wilhammer]

This is all very fascinating.

Are you guys compiling this stuff on a spreadsheet?

Some quick reference charts based on your data would be very handy to play with.

BTW,

I always had thought that speed rank was Fast, Hunt, Move.

It was also expected that vehicles would move faster than during the "acceleration phase".

The questions;

What is the distance of the acceleration phase of movement?

What effects do waypints have on this? Specifically, if a vehicle pivots, what effect does this have on momentum?

[Guest Germanboy]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Tris:

And I think I can now accurately predict that the longer they move (at least over 1000 meters) the faster they will likely progressively travel.

<HR></BLOCKQUOTE>

You must have a lot of time on your hands

Now my question would be whether this matters in the game? I can not recall ever to have given a straight 1,000m moving order to anything, be it tank or infantry. I would imagine few other people have, seeing that the max size map is 2,000x3,200m.

Are the differences pronounced at the level of 100-300m movement orders?

If they are, do they favour one side over the other?

And as Juardis pointed out, what about waypoints? The AI sets them automatically for vehicles when encountering obstacles, so on a real map this phenomenon would presumably not exist even for long-range movement orders.

As you can see, I have too much time on my hands too, wondering about this.

But the fact that people are running tests like this just shows what a great game this is, where nothing more important needs to be tested.

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

Andreas

<a href="http://www.geocities.com/greg_mudry/sturm.html">Der Kessel</a >

Home of „Die Sturmgruppe“; Scenario Design Group for Combat Mission.

[This message has been edited by Germanboy (edited 11-23-2000).]

[Tris]

I have not considered the practical ingame implications for the results of this data yet. I merely felt moved to pick up what Joeri started last night.

There are a few surprises.

Wilhammer, re acceleration: by definition this is a period when speed increases, but only to the desired attained cruising speed. I must have missed something with that question. Care to rephrase it?

[Tris]

I agree that this is one great game, Andreas, and a superior tactical war simulator on top of that. I think it's for the latter reason that many people take the time to examine it so closely for flaws. Now there's irony at work there which BTS does not always seem willing to respond to in a completely gracious manner, but then one can't have everything.

[Guest Germanboy]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Tris:

Wilhammer, re acceleration: by definition this is a period when speed increases, but only to the desired attained cruising speed. I must have missed something with that question. Care to rephrase it?<HR></BLOCKQUOTE>

I think what he means is that it is logical that it takes longer for a vehicle to do the first half of a given distance if starting from a stopped position, because it has to accelerate during the first half. It can deccelerate more quickly than accelerate, and therefore it is logical to expect the first half to take longer than the second half.

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

Andreas

<a href="http://www.geocities.com/greg_mudry/sturm.html">Der Kessel</a >

Home of „Die Sturmgruppe“; Scenario Design Group for Combat Mission.

[This message has been edited by Germanboy (edited 11-24-2000).]

[Tris]

If that's what he meant, Andreas, his logic is flawed within the context of this test as it pertains to vehicles; it has no bearing whatsoever when it comes to foot travelers.

[Bastables]

I'm sure Steve said a long time ago taht although CM models acceleration it does not model deccelerate, I mean watch how fast a Hellcat comes to a stop.

[Guest Germanboy]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Tris:

If that's what he meant, Andreas, his logic is flawed within the context of this test as it pertains to vehicles; it has no bearing whatsoever when it comes to foot travelers.<HR></BLOCKQUOTE>

This is fun - a totally pointless discussion

Sorry, but I don't understand why it is flawed. I agree that for foot travelers it has no relevance, unless you want to argue about them getting comfy after the first 50m, when they have balanced themselves out. Which would be a bit far-fetched, to put it mildly.

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

Andreas

<a href="http://www.geocities.com/greg_mudry/sturm.html">Der Kessel</a >

Home of „Die Sturmgruppe“; Scenario Design Group for Combat Mission.

[This message has been edited by Germanboy (edited 11-24-2000).]

[Tris]

Pointless discussions are the most fun. And how long do you suppose it takes a Sherman to "accelerate" all the way up to "walking" speed? I'd estimate we're talking less than a second.

[Maj. Bosco]

Look at the horse power on an M4 Sherman, it takes it a bit to get up to walking speed. I don't understand why this would be for infantry though. Maybe you could have a start marker and then put the test group behind it. Then have the walk to the end of the course, but start timing at the start marker. That should take care of any acceleration considerations.

[Guest Big Time Software]

Hehe... totally missed this discussion until it got bumped.

The reason why REVERSE is faster than Hunt or Move is that we assumed the driver to be trying for max manuever speed in that direction since it is most often used (at least for any significant distance) for emergency moves to better cover.

The CM top speed of Reverse is based on the vehicle's top Forward speed. This is an abstraction because different vehicles had different gear ratios and number of gears for reverse, which might mean that some of CM's vehicles move a little faster or slower than they would in real life. But the difference really shouldn't make much difference in game terms.

And obviously a tank shouldn't be able to drive backwards for 1000m at top speed. I am certain that something would break/overheat long before then. But CM doesn't put a cap on it because, practically speaking, one isn't needed. I mean, I personally would LOVE for someone to drive their Panther's at me in Reverse

As for the observed foot movement being faster for 2nd half than first half, I don't know what the deal is what that. I'll ask Charles.

Steve

[Tris]

The apparent increase in the speed of foot travellers over the second half of a 1000m course is the only thing I found truly amiss from the few tests I ran the other night, Steve; at that, my sample is small enough to where the results might not be (though they probably are) representative.

I haven't tested this yet but I'm (mildly) curious to see whether if the course were increased to 2000m the speed of travellers would increase even more still--or perhaps the speed would begin to decrease at some mark along the course. For that matter I'd like to see if fatigue might eventually become a dynamic, if uphill/woods/marsh travelling affects the issue, etc.

What effect any such data derived would have in practical game terms is something else. This is the primary reason the speed decision for reverse doesn't bother me much, the other being that I'm just not sure what the reverse speeds for tanks were. And yes, it would be a happy Allied player, indeed, who met an opponent who chooses to drive panzers in reverse solely to prevent these from becoming bogged.

[This message has been edited by Tris (edited 11-29-2000).]

[Tris]

Well, you sparked enough curiosity within me to seek an answer to the question you posed, Maj. Bosco.

So, it's basically the same test as before, but this time we will only run four M4 Sherman tanks over the 1000m dry grass course. Also, for this test each tank was afforded an acceleration of phase of approximately 30m.

The M4 Sherman tank ordered to Fast required 12 seconds to reach this new nominal start mark, Reverse required 16 seconds (Hunt :19 and Move :24).

Results

Speed Time Td

FAST 3:23 :08

REVERSE 4:45 :06

HUNT 6:12 :08

MOVE 10:40 :03

KEY: Td Time differential from previous test

The result for the Fast and Move ordered tanks looks about right, as we would expect a longer acceleration period for a tank trying to gear all the way up to top end than for one merely inched along. I do wonder, however, if :03 is an accurate gauge of how much time it requires a tank standing still to lurch into "walking" gear. Of more surprise is the fact it required the faster Reverse-ordered Sherman less time to reach its ordered speed than the slower Hunt-ordered Sherman. The discrepancy isn't great, and keep in mind that this test was performed by only one ordered tank at each speed, but we would have expected to see a :02 difference between the two times rendered in reverse order.

What we do not know is whether a 30m acceleration zone is sufficient for tanks to reach their maximum speeds for the respective orders. I think it is safe to say that this distance is sufficient for Move and probably sufficient for Hunt; it might not be for Reverse and Fast.

[This message has been edited by Tris (edited 11-29-2000).]