Pre-WWI Possibilities?

[ColinWright]

ORIGINAL: a white rabbit

ORIGINAL: ColinWright

ORIGINAL: Curtis Lemay



Not if I'm modeling it operationally. How is that different from any other operational scenario. "France 1944" doesn't have to get all the tactical minutia of Omaha Beach detailed. It abstracts it.

Yeah -- but Omaha Beach wasn't the nub and the gist of the entire 1944 campaign. That's one of the many areas where OPART falls down when it comes to the pre-modern era: it can't accomodate the difference in scale between the arena of strategic movement and that of actual battlefield decision. Strikingly, this remains true even when you have an arena as sharpy confined as that of the Waterloo Campaign.

..it does if you set the ground scale at some 500m, approx the effective range for artillery, or at 250m to allow for smoke and give a possible cannister effective of 1 hex for a two hex total range..

..300*500=150,000m=150k sides (or 75k at 250m). admitedly this doesn't give you the invasion of Russia but otherwise..

You're kind of missing the point. For twentieth century warfare, a single-screen engine works quite well, as a rule. Do the Western Front 1918 and the first Kaiserschlacht will eat up a quarter of the map. Do the Eastern Front 1943 and Kursk and the associated battles will take up a fifth of the front.

Now: do Napoleon's invasion of Russia. Borodino is a one-hex battle. Do Grant's campaign against Vicksburg: Champion Hill is a one-hex battle. Do whatever campaign of Marlborough's it was: Ramillies will be a one-hex battle. Do the Mongol invasion of Poland: Leignitz will be a one-hex battle. Need I go on?

Among a considerable list of other things, OPART would require a two-screen system to even start to be a satisfactory engine. Look at Forge of Freedom: the American Civil War game. What does it have? Two screens. Look at Medieval Total War. What does it have? Two screens.

Not that these games are perfect, but they do have one essential element to modeling warfare in their era that OPART lacks: two screens. The gap between the scale of the campaigns and the scale of the battles is just too great to dispense with this.

Now, you can use OPART for whatever you want, and the results may even be entertaining -- but really, you're using a table knife to unscrew the door hinges. It's not the right tool, and it's exasperating to listen to people try to insist it is.

[golden delicious]

ORIGINAL: Curtis Lemay

Is that just the French?

No. That's the whole of the battlefield from the French rear to the British rear, and including the extended flank with Dutch troops.

Not if I'm modeling it operationally.

You claimed in a discussion we had elsewhere that a tactical model would be useless for simulation because you have no idea where the battle will be fought out. If we follow your logic, then if the operational map shows the units in four hexes then they must be occupying a front of not less than 8km, because the operational map isn't an abstraction of where the battle actually takes place- it's the one and only definitive representation. If the battle frontage was smaller then they would only cover three hexes.

In any case, if the forces are spread over too many hexes, this artificially reduces the unit density, making the model invalid.

[a white rabbit]

..Funny, the following from an article on the Ionian Revolt made me think of you, Colin..
 
..Sparta, the greatest military power within the Hellenic world, had no great liking for distant expeditions, and having little or no knowledge of events outside her own European Greece, refused to help.. 

[golden delicious]

ORIGINAL: a white rabbit

..it does if you set the ground scale at some 500m, approx the effective range for artillery,

I rather think direct fire artillery can fire more than 500 metres;

http://en.wikipedia.org/wiki/Field_Arti ... _Civil_War

The chart on this page gives ranges varying from 1000 to nearly 3000 yards. Obviously this is a later period- but the technology is not that different. In any case, at Waterloo the guns were set up around a mile from the enemy main line.

[Curtis Lemay]

ORIGINAL: ColinWright
Yeah -- but Omaha Beach wasn't the nub and the gist of the entire 1944 campaign.

I could make a short subset of France 1944 that only dealt with the first few days of Normandy.

That's one of the many areas where OPART falls down when it comes to the pre-modern era: it can't accomodate the difference in scale between the arena of strategic movement and that of actual battlefield decision. Strikingly, this remains true even when you have an arena as sharpy confined as that of the Waterloo Campaign.

Actually, as my AAR shows, there's plenty for players to do in Waterloo 1815. There's at least as much to do as in similar sized WWII scenarios. Not all scenarios have to be huge monsters. There's room for small, short ones too.

[Curtis Lemay]

ORIGINAL: golden delicious
You claimed in a discussion we had elsewhere that a tactical model would be useless for simulation because you have no idea where the battle will be fought out.

Actually, I said it would add nothing except problems.

If we follow your logic, then if the operational map shows the units in four hexes then they must be occupying a front of not less than 8km, because the operational map isn't an abstraction of where the battle actually takes place- it's the one and only definitive representation. If the battle frontage was smaller then they would only cover three hexes.

I'm not sure if that's logic. No matter how you try to twist it, units in four hexes could be actually spread out over a max of 10km or a little as just over 5km. A 7.5km front would actually have to be perfectly fit into a three hex length - very unlikely.

In any case, if the forces are spread over too many hexes, this artificially reduces the unit density, making the model invalid.

Well, they're not spread over too many hexes. Note the historical initial deployments - max of six divisions in one hex. And one can stack up to nine divisions in a single 2.5km hex. That's pretty dense.

The system fits Napoleonic norms as well as it fits WWII norms. I'm sure there were instances where more than nine regiments were in a 10km hex during WWII.

[a white rabbit]

ORIGINAL: golden delicious

ORIGINAL: a white rabbit

..it does if you set the ground scale at some 500m, approx the effective range for artillery,

I rather think direct fire artillery can fire more than 500 metres;

http://en.wikipedia.org/wiki/Field_Arti ... _Civil_War

The chart on this page gives ranges varying from 1000 to nearly 3000 yards. Obviously this is a later period- but the technology is not that different. In any case, at Waterloo the guns were set up around a mile from the enemy main line.

..guess i didn't know those figures [8|]...

..EFFECTIVE, Ben, effective range, not test case conditions, cf the RHA tests on cannister, 50 plus balls per sq yd at 200yds, 6 slow moving per sq yd at 500yds, the max range. I'll assume you're familar with the tests run on ancient bows, by comparing experts and amateurs on different kg-pull bows...

[golden delicious]

ORIGINAL: a white rabbit

..EFFECTIVE, Ben, effective range, not test case conditions,

So see my second point: the distance from the guns to the enemy line at Waterloo. About a mile- more in some cases.

So now we're weighing your opinion against those of Wellington and Napoleon. I would hazard a guess that the latter two had a little bit more expertise on the subject than you do.

Anyway, cannister is a last-ditch weapon. Artillery of this period would fire shot most of the time. Naturally you can't hit a target at a mile- but you can hit a division.

[ColinWright]

ORIGINAL: Curtis Lemay

ORIGINAL: golden delicious
You claimed in a discussion we had elsewhere that a tactical model would be useless for simulation because you have no idea where the battle will be fought out.

Actually, I said it would add nothing except problems.

If we follow your logic, then if the operational map shows the units in four hexes then they must be occupying a front of not less than 8km, because the operational map isn't an abstraction of where the battle actually takes place- it's the one and only definitive representation. If the battle frontage was smaller then they would only cover three hexes.

I'm not sure if that's logic. No matter how you try to twist it, units in four hexes could be actually spread out over a max of 10km or a little as just over 5km. A 7.5km front would actually have to be perfectly fit into a three hex length - very unlikely.

In any case, if the forces are spread over too many hexes, this artificially reduces the unit density, making the model invalid.

Well, they're not spread over too many hexes. Note the historical initial deployments - max of six divisions in one hex. And one can stack up to nine divisions in a single 2.5km hex. That's pretty dense.

The system fits Napoleonic norms as well as it fits WWII norms. I'm sure there were instances where more than nine regiments were in a 10km hex during WWII.

However, I distinctly recall you noting that it was adviseable to reduce the density before making attacks -- and the successes you managed to achieve largely stemmed from taking advantage of the 'overcrowding' among the defenders. So whilst you can achieve Napoleonic densities in OPART, the results of attempting to fight a battle with such densities would seem to be disastrous.

It's really pretty simple. You can attempt to simulate almost any conflict you want with OPART. However, it's designed for the World War Two era -- and the further away you get from that, the more limitations and problems your simulation will have. Stridently insisting otherwise won't change the truth of this.

[ColinWright]

ORIGINAL: golden delicious

ORIGINAL: a white rabbit

..EFFECTIVE, Ben, effective range, not test case conditions,

So see my second point: the distance from the guns to the enemy line at Waterloo. About a mile- more in some cases.

So now we're weighing your opinion against those of Wellington and Napoleon. I would hazard a guess that the latter two had a little bit more expertise on the subject than you do.

Anyway, cannister is a last-ditch weapon. Artillery of this period would fire shot most of the time. Naturally you can't hit a target at a mile- but you can hit a division.

Really, it's just another fine example of how OPART isn't suited to pre-modern warfare. Napoleonic cannon did become dramatically more effective as the range closed.

That's something that OPART has no way of simulating -- quite reasonably, since modern cannon have more or less the same effectiveness over most of their theoretical range.

I suppose Richard could tun hoops with 'cannon' of different range in each unit -- but really, all he's demonstrating is that he's got the wrong tool.

[Curtis Lemay]

ORIGINAL: ColinWright
However, I distinctly recall you noting that it was adviseable to reduce the density before making attacks -- and the successes you managed to achieve largely stemmed from taking advantage of the 'overcrowding' among the defenders. So whilst you can achieve Napoleonic densities in OPART, the results of attempting to fight a battle with such densities would seem to be disastrous.

Try actually looking at the AAR. Unit densities remain more or less the same as at the start throughout the game. Take the Attack Planner for the attack on Ligny: All the French units that started around Ligny attack it. 10 French divisions supported by three corps artillery units are attacking the three Allied divisions, 1 regiment, and 1 corps artillery in Ligny. It takes multiple rounds, but they are eventually successful. Not because they have a density advantage, but because of the weight of their combat strength, and their flanking advantage.

It's really pretty simple. You can attempt to simulate almost any conflict you want with OPART. However, it's designed for the World War Two era -- and the further away you get from that, the more limitations and problems your simulation will have. Stridently insisting otherwise won't change the truth of this.

Things can seem simple if you don't ever test them. Every theory is always correct by the Colin Wright scientific method:

1. Postulate Theory.
2. It's Miller time!

But for real scientists, it's:

1. Postulate Theory.
2. Test Theory.
3. Adjust Theory based upon results.
4. Repeat until results match Theory.

For that reason, I prefer to test things out before I leap to conclusions.

And the success of Waterloo 1815 as an operational scenario suggests that some pre-20th century subjects can be simulated in TOAW.

[ColinWright]

ORIGINAL: Curtis Lemay

ORIGINAL: ColinWright
However, I distinctly recall you noting that it was adviseable to reduce the density before making attacks -- and the successes you managed to achieve largely stemmed from taking advantage of the 'overcrowding' among the defenders. So whilst you can achieve Napoleonic densities in OPART, the results of attempting to fight a battle with such densities would seem to be disastrous.

Try actually looking at the AAR. Unit densities remain more or less the same as at the start throughout the game. Take the Attack Planner for the attack on Ligny: All the French units that started around Ligny attack it. 10 French divisions supported by three corps artillery units are attacking the three Allied divisions, 1 regiment, and 1 corps artillery in Ligny. It takes multiple rounds, but they are eventually successful. Not because they have a density advantage, but because of the weight of their combat strength, and their flanking advantage.

It's really pretty simple. You can attempt to simulate almost any conflict you want with OPART. However, it's designed for the World War Two era -- and the further away you get from that, the more limitations and problems your simulation will have. Stridently insisting otherwise won't change the truth of this.

Things can seem simple if you don't ever test them. Every theory is always correct by the Colin Wright scientific method:

1. Postulate Theory.
2. It's Miller time!

But for real scientists, it's:

1. Postulate Theory.
2. Test Theory.
3. Adjust Theory based upon results.
4. Repeat until results match Theory.

For that reason, I prefer to test things out before I leap to conclusions.

And the success of Waterloo 1815 as an operational scenario suggests that some pre-20th century subjects can be simulated in TOAW.

What 'real scientists' would do is neither here nor there. The Curtis LeMay method would seem to be more along the lines of

1. Postulate theory.

2. Find a 'test case' that is most likely to yield results validating the theory.

3. Adjust perception of the results thus obtained until they appear to vindicate theory.

4. Leave (1) and (2) untouched. Repeat (3) until perception seems real.

Note what exactly you're fulminating against. My insistence that TOAW is designed to simulate World War Two and later, and that the further you move away from this period, the less satisfactory and the more occasional the successes will be. What could be more obvious?

Your position is largely that your pick-up is the ideal vehicle for all occasions. It's true that you can attempt to use it for all kinds of things besides picking up the gardening supplies from Home Depot -- and even enjoy partial success. You can indeed make your daughter take it to the Prom. You can eventually move your three bedroom house to the next state -- it'll take fifty trips, but you can do it. You can even mount a machine gun on it and take on the US Army -- people have done this.

You'll have some partial success. I don't deny it. I just refuse to agree you've got the ideal vehicle for all these occasions. I think it would be more useful to discuss what a good pre-modern engine should have. I'm open to the likelihood that not all my ideas would necessarily turn out to be the best ones -- but I am certain that your effective insistence that OPART is just the thing for this task is not constructive.

[Curtis Lemay]

ORIGINAL: ColinWright

ORIGINAL: Curtis Lemay
For that reason, I prefer to test things out before I leap to conclusions.

What 'real scientists' would do is neither here nor there. The Curtis LeMay method would seem to be more along the lines of

1. Postulate theory.

2. Find a 'test case' that is most likely to yield results validating the theory.

3. Adjust perception of the results thus obtained until they appear to vindicate theory.

4. Leave (1) and (2) untouched. Repeat (3) until perception seems real.

Even if that were the case, note that it's vastly superior to just postulating a theory then sitting back smuggly assuming it's right without ever even testing to see if it actually is.

I actually tested the concept. I'm quite satisfied with the results. Could it be improved with some code changes? Yes. But it already works as well as most WWII scenarios out there.