Crazy idea!

[Koro]

So what if, when all of the battles were completed, you glued them all together in to one grand map and then let the AI and player loose on it with say 150.000 troops? I realize the maps might not combine perfectly but this then truly give a sense of moving an army in to good positions, trying to scout for the enemy forces and allow flanking opportunities etc.

Have the winter maps in the middle, perhaps turning some of them to make the rivers connect.

Yeah, that would be incredibly fun I think :).

[Wandering1]

If we had a higher camera viewpoint to manage those armies, it would certainly be interesting.

At the moment, with wide fronts like Chancellorsville, already shows you the command limitations of only being able to reasonably control 1 Corps at a time, due to how much area you're actually looking at.

This, being before you run into technical issues like toasters crying at the giant map size with all of the units it's trying to process.

[Don't Escrow Taxes]

I think a random skirmish generator that used portion of existing maps would be a lot of fun.  Love Koro's idea.

[Nox165]

2 hours ago, Koro said:

So what if, when all of the battles were completed, you glued them all together in to one grand map and then let the AI and player lose on it with say 150.000 troops? I realize the maps might not combine perfectly but this then truly give a sense of moving an army in to good positions, trying to scout for the enemy forces and allow flanking opportunities etc.

Have the winter maps in the middle, perhaps turning some of them to make the rivers connect.

Yeah, that would be incredibly fun I think :). I like the idea, I would love to be able to put to massive armies on a map and let them fight it out to the death.

 

[Andre Bolkonsky]

You will fry out half the computers that play this game. 

 

But i like it otherwise. 

[Wandering1]

Just now, Andre Bolkonsky said:

You will fry out half the computers that play this game. 

 

But i like it otherwise. 

See, my reference to toasters. It's a euphemism for old computers. 

[Andre Bolkonsky]

8 minutes ago, Wandering1 said:

See, my reference to toasters. It's a euphemism for old computers. 

I did, I totally stole your comment, but I rephrased it to seem 'cooo' and edgy. 

Primarily because I totally agreed with what you said. 

[Koro]

4 hours ago, Andre Bolkonsky said:

You will fry out half the computers that play this game. 

 

But i like it otherwise. 

So 150000 might be too much but the size of the map shouldn't really matter, should it? 

[Wandering1]

4 minutes ago, Koro said:

So 150000 might be too much but the size of the map shouldn't really matter, should it? 

I can be argued for either way.

The map consumes memory at the end of the day, so as long as you have a relatively recent computer, the size of the map by itself shouldn't be an issue. So long as you aren't hitting the 4 GB limit, because the client is a 32-bit client.

The implication of having a gigantic map though, is the space for units to fill it. And the more units there are, the more computations it has to chug, whether the units are in combat or not. And, I could be wrong, but my inclination is that the game is not multi-threaded, meaning only one CPU is working on the game.

[Andre Bolkonsky]

6 minutes ago, Koro said:

So 150000 might be too much but the size of the map shouldn't really matter, should it? 

I would say it would incrementally escalate the problem of an older computer trying to keep detail on a map, all of its components, and massive army traversing a map that is ten times larger than the largest map it currently handles. But I'm not a coder, so what do I know?

[Jamesk2]

If the lagging comes from number of units, then make it Company of Heroes style: You have a small force at the beginning, and have to go out and capture strategic points to buy more units. Will also offer a nice dynamic to the map.

[James Cornelius]

The way to make that work on a larger scale is that it is not subdivided into brigades, but divisions (or potentially even corps if that proves too daunting). Then, when the battle is actually joined, it rezooms to enable the brigade level command. Yes, it will lose some precision, but from a practical standpoint it is both 1) close enough and 2) accurate in that on a campaign scale, from 1862 onwards the division was the primary sub-unit for unit movement.

 

While still operating in realtime, it would be more of a chess match, as your army is strung out. In addition to the default historical maps, a few others could be included of various areas near, and units would arrive (or  not arrive haha) from the direction they were in relation to where the two opposing armies met.

[Andre Bolkonsky]

6 hours ago, Jamesk2 said:

If the lagging comes from number of units, then make it Company of Heroes style: You have a small force at the beginning, and have to go out and capture strategic points to buy more units. Will also offer a nice dynamic to the map.

No. Company of Heroes is a fine game, i enjoy playing it and always have. But it is not the model  you want to emulate with this game. 

[Wandering1]

Sounding more like a Hearts of Iron mod for the American Civil War, where the difference is basically resolving combat through an RTS match instead of automated AIs fighting each other. Can one imagine how much time one would require to finish a campaign? 

[Keepbro]

You propose creating a map of the eastern portion of the North American Continent then simulate the entirety of the American Civil War in real time? I'm sold.

 

You might go actually mad with power or the micro might well cause you to slit your wrists. It would necessitate accurately mapping some (and Im just using rough figures here) 8 million square kilometers of terrain, oh and accurately modeling it as it was back in 1860s (so as to get the right feel what with all the wood, mud and shit that was everywhere). Then you'd have to add in the some 3 million odd poor fuckers who actually fought the war and accurately portray all the economics of the time (fuck tons of farms) so as to allow a realistic portrayal of supply and technology that made such an impact during the war. Oh and other small systems like accurate weather patterns and other little things like health since most of the Civil War's dead died due to disease rather than actual combat.

 

All in all its the type of game that I think a video game developer would rather chew his/her reproductive organs off than make. But then you do get the occasional crazy type who does games like this (the Dwarf Fortress guy springs to mind).

[James Cornelius]

 

57 minutes ago, Keepbro said:

You might go actually mad with power or the micro might well cause you to slit your wrists. It would necessitate accurately mapping some (and Im just using rough figures here) 8 million square kilometers of terrain, oh and accurately modeling it as it was back in 1860s (so as to get the right feel what with all the wood, mud and shit that was everywhere). Then you'd have to add in the some 3 million odd poor fuckers who actually fought the war and accurately portray all the economics of the time (fuck tons of farms) so as to allow a realistic portrayal of supply and technology that made such an impact during the war. Oh and other small systems like accurate weather patterns and other little things like health since most of the Civil War's dead died due to disease rather than actual combat.

I don't think he's suggesting that level of detail, though it would certainly be welcome from the standpoint of a hard core player, but there is some very solid middle ground.

[James Cornelius]

2 hours ago, Wandering1 said:

Sounding more like a Hearts of Iron mod for the American Civil War, where the difference is basically resolving combat through an RTS match instead of automated AIs fighting each other. Can one imagine how much time one would require to finish a campaign? 

I actually think the better model here would be (ironically enough since that is where Nick got his start) the Total War campaign map, except it would be realtime instead of turn based. The terrain would be generalized as "mountain/hill/open/river" etc with (hopefully) accurate battlefield maps for the area based on, say, Google Earth. Yes, that doesn't necessarily represent how the ground looked in 1861 to 1865, but it is a starting point.

 

As with my previous suggestion. you would be moving divisions or corps on the campaign map, not individual brigades.

 

And, as far as an existing example goes, I actually think AGEOD's "Civil War II" is a better representation of this than "Hearts of Iron", though the various incarnations of HoI are a superior game to Civil War II. If you haven't checked it out, I encourage you to do so. I would love to see UG:CW somehow end up (roughly) as if the best parts of Total War and Civil War Generals 2 had a child that was then adopted by Civil War II and a fine historical RTS game...or something.

[Wandering1]

As a little side note, I played Robert E Lee's Civil War General when I was about 9 or 10 years old. Did a biography report on Stonewall Jackson in 4th grade using the encyclopedia entry that was in it.

Anyways, the bigger reason that I use Hearts of Iron instead of Total War: Total War's campaign map is turn-based, whereas Hearts of Iron is already developed for real-time. Really, it's just how granular you want to go with regards to creating territories which indicate different terrain types to map it out on. And it already has an interface for manpower and factory production (which are significant parts of the American Civil War, as far as the confederates are concerned). Manpower and factory production are glossed over as total income in most Total War frameworks.

 

[MikeK]

Crazy idea? Yes, it is.

And "skirmish" mode using existing maps? Would be a nice add, but it would lead to agitation for a whole dynamic campaign and that is the road to crazytown since the actual war needs to be covered first, and more wrinkles can be added there.

 

[vren55]

I love the idea... except there are problems.  irregardless of the computers needed... there's an interface issue.

Supreme commander was able to allow you to zoom in and manage stuff, so I don't think that'd be a problem.

However, the whole purpose with the Battle deployment/army organization screen was so you can select and equip your troops...But then you'd have to figure out recruitment as well and on a real time map it can get really complicated very quickly. Plus all the historical programming to create an AI smart enough but historical enough to have a Civil war Experience... yeah... lots of challenges.

Frankly a total war style thing or a... Star Wars Empire at War style thing would be much more manageable... you'd actually be able to take advantage of the imaginary lag time between the battle map and the strategy map to sort out your army stuff before needing to zoom out and think about strategy.

 

[James Cornelius]

I think in the suggestions noted, there would still be time to organize at least the forces immediately engaged before entering the battle from the campaign map, or whatever it would be called.

 

Like I said, I think a Total War style campaign overview but in real time rather than turn based would probably work just fine - but the time would be suitably compressed or adjustable just like it is in the current battle maps to allow for more efficiency.

[waldopbarnstormer]

I used to have a game called Napoleon 1813 which featured a real time campaign map and battles which played out on a zoomed in hex battle map, again in real time. The smallest unit you could give orders to was a division. The campaign map consisted of towns and cities connected by roads and armies could only move along the road network. This added some realism because historically armies then and during the civil war could only move along roads because of wagons and supplies.

Using this feature would reduce the complexity for programming the ai as it limits the movement options. I never really liked the  way total war made use of the campaign map. Towns linked by a road could be used as supply dumps and if cut by raiding cavalry reduced the effectiveness of your army, or increased desertion.

[A. P. Hill]

Not really so crazy.

I've said all along that I want an overview map of the entire part of the U.S. where the ACW was fought with all the battlefields as selectable locations that once selected take you to the maps that exist in game to battle it out. 

[Nox165]

Massive battles please

[Fred Sanford]

Even if existing maps are used as the basis, I think the implementation hurdle is scripting the AI unless they can build a general 'destroy the enemy army' AI that can function on any map.