True Distances - Preface to 'True Ground'

Edit: These determinations proved incorrect because initially the map distances weren't synchronized with the LOS distances by editing the map.ini line 'UnitPerYard=30', see Post #7

Determining True Distances

True distances were established by scaling down soldier sprites until their height equaled two yards (6 feet - The average man during the American Civil War was around 5'6" or 5'7", according to Wiki). This criteria was established on the flat Kansas map by the following procedures:
1) the game was started and two infantry opposing units were set so that their engagement range was 100 yards (103 yards - Engagement Distance, attachment below)
2) screenshot - centered camera between both units wide enough to include the flags of both engaged units
3) opened screenshot in Paint.net (or any graphics program that displays pixel measurements)
4) measured the height of an infantry sprite in one of the engaged units (sprite height of 26 pixels, then divided in half, attachment below)
5) measured the distance between engaged units' flags (1294 pixels, attachment below)
6) divided height of infantry sprite in half (this establishes how many pixels per yard, figuring sprite height at 6 feet, or two yards), after dividing that number, now representing 1 yard, divided it into the distance between engaged units (1294 pixels divided by 13 pixels = 99.5, attachment below)
7) the goal then (by scaling down the troop sprite size) is to get that last divided number (now representing yards) as close to 100 as possible.

These seven steps result in a fairly accurate map distance of 100 yards in proportion to the size of troop sprites and map objects; end result setting for troop scaling is 46.5%. Note below the engagement range of 103 yards, the calculator readout of 99.5 yards, resulting in a 3.5 yards difference:




Determining Marching (walk) Speed for 'True Distances', resulting in 'Real Time' play

As noted by a recent post by MarkT, for centuries the accepted normal speed of marching infantry is three miles per hour. The following steps established this ‘walk’ speed:
1) edited the Kansas map to have no movement impediments, then added two objectives from the Devil's Den map, 'The Peach Orchard' and 'Little Round Top', which are one mile apart
2) opened the game and used this edited Kansas test map to march an officer between these two objectives
3) based on the 3 miles per hour speed, it takes 20 minutes to walk 1 mile, the officer’s journey was timed as so, tweaking the walk speed until the 1 mile destination was reached in 20 minutes
4) walk speed was established at 2.6 using this process
5) all other walk and run speeds in the unitglobal file were based on this established walk speed

Presently I'm going through the process of scaling down literally thousands of map objects on the Devil's Den map. If anyone sees a flaw in the above processes to establish 'True Distances', please weigh in to help this mod along, or to just convince me to stop the madness.

Making faster progress now that I found I could drag 'grouped' fence segments via CTR and RightClick hold. I was using LeftClick hold but it kept selecting other objects during the drag process. So, for folks moving around fencing groupings, RightClick hold is the fastest and safest procedure I've found. Also, you can keep the CTR button down the entire route, or release it when necessary to move the camera without consequence.

Question:
Is there any way to select connected fence segments rather than one at a time? Something like group selection between two points holding down the 'shift' button? Since we can't use the 'shift' button here, is there another button that would serve as a 'shift' button by selecting multiple fence segments between two selected points?

Or, can some adept coder out there configure a 'lasso' function in this app to select groups of objects?

Davinci,

Well, that's a good tutorial for other folks, but that wasn't my question as I'm moving multiple selections with no problems. My question is a selection question - How to select multiple fence segments at once, eliminating the "one at a time" selection process.

Still, thanks for the tutorial

True Distances has evolved to the ' True Ground ' mods.

Managed to get all the fences and buildings replaced by last week, followed up with scaling down the multitude of rocks, spending most of that time rebuilding Devil's Den, the suspended 'Table Rock' being the main challenge. Finally figured out how to lower it for the smaller scale, details to follow if anyone asks.

Then, after play testing some Devil's Den map scenarios, saw the necessity of lowering the height and density of the smoke effects, editing the efx.csv file.

And then, another eureka revelation, I messed up!

Went back to double check map distances and assigned a 1760 yards (one mile) movement to an officer's toolbar. When this movement was initiated from LRT to the Peach Orchard (one mile) it was three minutes too long. It was then I realized I messed up by not synchronizing the map distances with the game distances; i.e., Move Forward 100 yards...This is done by tweaking the 'True LOS' key function - 'UnitPerYard=30'. After tweaking it down to 'UnitPerYard=26', the 1760 yards (one mile) movement test synchronized with the LRT to the Peach Orchard one mile distance. Along with this map change came the necessity of the marching speed change, up to 3 from 2.6 to achieve the real time of 3 miles per hour.

Now, we have 'True Ground', with almost true distances, the battle distances being off about 30 yards; i.e., engagement distances displaying 160 yards will actually be 130 yards (hopefully most folks won't notice, or better yet, find this a negative). Can't fix this without scaling down everything again about 25%. And since my little finger is still numb from moving fences (holding down the CTR button), I'll leave this mod as is. Not as accurate as promised, however, this mod still represents The Big Picture of True Ground and Real Time, and, can still be used as a setting for SR1 play.

Davinci wrote: I was wondering how this was coming along!

The "Rocks \ Boulders " will present another problem that you may or not be aware of!

They have a feature of having the units stand on top of them, but this has a lot to do with the position of the camera, by changing the camera angle, they will or will not be standing on the rocks.

All in All, this is still some good practice for you to indulge in!

davinci

Yeah, the "stand on top" feature is even uglier with the smaller scale. I've tried to turn this off in the rock properties to no avail. Suggestions?

garyknowz wrote: I've had this question for a while. You folks have likely long addressed it, but most of the time, I can't follow what you guys are saying anyway when modding.

What would be the accurate length of a 200 man deployed front, and how to scale is that on the terrain? Did ya'll have to make adjustments for playability or map size constraints, or is it pretty accurate?

Thanks folks! Gary K.

Since I screwed up initially with this mod's development I had to read this thread over again to sort it out, so, I understand your confusion. Only I know what it says.

As to your question, this mod doesn't consider unit widths. This mod establishes true map distances and battle distances. The details of how these distances were established are explained in posts 1 and 7.

Your question of unit width as related to terrain is pertinent when scaling is applied. This was a huge consideration as the SOWGB game was developed, and the maps were adjusted for four to one troop scaling...The width of regiments being the key consideration.

Thanks for your Interest

Hey D, good hearing from you. Yeah, still on this quest, and the new computer has literally made it possible with the performance demands of SR1 development. Have no desire to seek out new games, modding SOW games fulfills my entertainment needs just fine, not to mention distracting my mind away from yet another year of poor showings of the football teams I follow, the Hogs and Dallas.

'True Ground SR1, Scenarios & OOBs' is due to be released before Christmas. I'm presently working on the fourth scenario for this release and plan for at least one more. Thereafter I plan on releasing more scenarios as they are produced. I'm really pleased at the way the SOW engine is handling SR1 play, far far better than that hack-job attempt at SR1 play ten years ago with the TC2M engine.

Still no word from Crikey regarding the 'True Ground Devil's Den Map' upgrade. Hopefully he'll chime in soon regarding any progress with that map.

Davinci wrote:

RebBugler wrote: 'True Ground SR1, Scenarios & OOBs' is due to be released before Christmas. I'm presently working on the fourth scenario for this release and plan for at least one more. Thereafter I plan on releasing more scenarios as they are produced. I'm really pleased at the way the SOW engine is handling SR1 play, far far better than that hack-job attempt at SR1 play ten years ago with the TC2M engine.

So, Can you place approximately 2600 men standing shoulder to shoulder within a mile, or have you had the time to test that ?

davinci

Ha, won't be testing that. Besides, as explained earlier in this thread, 'True Ground' isn't perfect, but close enough for smoothbores.

Palyarmerc wrote: I had no clue as to what Devils Den 'True Ground' amounted to. I knew I had Devils Den, so installed this, compared maps, and could see no difference.
A day later I boot up Fredericksburg and my soldiers had shrunk. However, the map did look more expansive, true to size if you like. Took me 5 minutes to work out the True Ground Mod was the cause.

But now worry set in...what if I'd blundered...
When I started work on Fredsburg, I used the Kansas Fall map as a template. Lstudio set the default for this map at 256,512 but other maps I eventually looked at were 512,512

At one point I was convinced the TC2M Fredericksburg was bigger than the one I'd created. (That is, 512,512 as opposed to my 256,512) which would have made any attempt at Scenario conversion impossible (it mostly is for various reasons though; the 90 degree co-ordinate mismatch; csv's packed in an .mmg file to name but 2)

So I quickly knocked up a '512' variation and added it to the download (except all the ambient sounds and objective locations are now at the wrong co-ordinates, more work, but I'll skip that)

512,512 Lstudio Terrain expands the map scale, which is partly good, since anything beyond Brigade sized battles would otherwise be virtually impossible.
The 512 variation helps to enable Corp vs Corp sized battles, else similarly, I'd need to down-scale all the buidlings and fences to the True Ground idea.

Some more discussion on this is certainly welcomed. In all, a worthwhile project I think.
Map making / modifications need a set rules for correct scale, especially regards enabling realistic battles. Not that I've played much Scourge of War, having too much fun with the SDK

Nice post, bumped it up since it got buried before much discussion. As far as I know, I'm the only modder that's messed with map scaling. Different sized maps are common, and all are scaled for 4:1 soldiers to sprites ratio, so I'm not sure why you're suggesting a set of rules governing map scaling.

Most maps made available in recent years by modders tend to be 5 mile maps or greater. I've revisited the 2.5 mile maps because they handle SR1 (1:1) performance demands best. Obviously, the bigger the map, the greater the demand on performance. Also, the 2.5 mile Kansas map is perfect for testing and establishing true distances and movement speeds with its flat and sparse terrain characteristics.

Welcome to SOW modding, looking forward to your contributions.