Topography, the silent killer
By LucusLoC
I recently played a game against the AI that highlighted what I see as a problem. I had selected a rush AI because I wanted to test out some new build orders to see how they would work when under pressure. That is when I found a deficiency with how the game displays some very important, if often overlooked, information. You see, I had built my fist defensive line with a shallow hill to the northern flank. Now I don’t know about you, but I very rarely zoom in close enough to see the topography. Usually I zoom in just enough to see the green highlight for a building, then zoom back out again to find out where my attention is needed next. At that zoom level only the large and steep hills are apparent. This hill was small enough that the ground still appeared flat. Needless to say that the first enemy wave was only half obliterated on my defensive line. The other half waltzed into my base unmolested thanks to the cover of a hill I did not notice.
Fortunately for me I was gearing up for my first assault and had a decent army built up that was able to deal with the threat before it did any real damage. But the point is that I should not have had to resort to this. If I had noticed the hill you can bet that I would have had a point defense or two on top of it, and that would have made all the difference. What I propose is a means to be able to tell the elevation while in strategic view, such as a toggleable colored topography map. This would allow an aspiring commander to get a very good and quick look at the surrounding land and allow him to plan his defenses at a glance, without having to muck about with zooming in to each location and tilt the camera to get a look at any low hills that may be in the way. [Ed. note by Cyde Weys: The topography map should auto-scale the range it uses for coloration based on the highest and lowest points currently visible on the screen. So when completely zoomed out, the full color range represents the entire topographical range of the map, but at closer zoom levels, smaller terrain features, like slight hills, become much more visible.]
Of course this is not the only problem that topography caused me this game. That blasted little hill allowed a second wave to get through, though not in the way you would suspect. You see, one of the enemy units from the first wave stopped on the way in to shoot at one of my turrets, one that he could not hit. So the little T1 tank just sat there, shooting at the ground. The annoying part was that he was in range of four of my turrets, all of which decided to shoot at him since he was the only threat left in range. They couldn’t kill the unit though because the ground blocked their shots. Then the second enemy wave rolled in from a little further south. They immediately obliterated the one unoccupied turret and proceeded to decimate the other four while the turrets stubbornly continued to shoot at the one T1 tank they could not hit.
It was at this point that I came back to check in and see what was going on on this front. (I had to catch all the details of what went down prior to this on the replay.) I hastily gathered together my third wave forces (I was sending smaller surgical strikes at the enemy, plus I was outproducing him, but it was only an AI, so this is really not surprising) and beat off the attack, managing to save one of the distracted turrets in the process. So I got curious and when I rebuilt my defensive line I did so in a way that gave the two battling units a nice little free place to just sit and shoot at the ground. They did so for the rest of the game, and that little T1 tank was the last unit I killed, still shooting the ground between the turret and itself.
This brings to light a second, more difficult problem. Turrets (and units in general) need to be able to tell when they cannot hit a particular target so that they do not waste time shooting at it. [Ed. note by Cyde Weys: This is very computationally expensive.] It would also be nice to have the range view automatically adjust so that you could see where the turret can actually hit, so that if a hill blocked part of the field of fire the doted line would bend to follow the contour of the hill or obstruction. Hell, it would be nice if that also worked for movable units as well, though I suppose that for them it’s less of an issue. I realize that this game actually has hit boxes for units, and that while a T1 tank may be hidden by a hill, a monkey lord will be in full view and be able to be hit. I think if we made it the convention that we would show only where the turret was actually capable of hitting the *ground* then that would give a pretty decent indicator of where it would be most likely to hit most units. But that’s just me.
This would also apply to the range of artillery pieces. All too often I have noticed that artillery perched on a hill or cliff is capable of shooting far past its indicated range marker. But you can’t target anything outside the circle, so often times this added range is lost, unless you manually target the ground at the extreme range limit. This is not always desirable, especially if the artillery is used defensively. And it is very micro-intensive. It would be nice if the computer could adjust the maximum range to also take into account any added benefit altitude gives you. Maybe just calculate how far the furthest shell could go, and move the dotted line there. This would allow the artillery to automatically target anything that wanders into its true range. It would have an effect on point defenses as well. If these changes were implemented it could bring a new dimension to the game, making it important to secure the high ground outside your base so that your enemy could not set up an arty station there that is outside the range of yours.
August 17th, 2007 at 3:22 pm
I think that you’re right about the line around the pd updating because of the terrain. If the line wasn’t a circle when you go to build the pd then you’d know that something was up and not end up building a pointless pd.
August 17th, 2007 at 4:06 pm
Great article. The fact that we can even have discussions about terrain, range, hitboxes, etc. is a testament to the greatness of this game. I love how physics truly affects gameplay, as opposed to “this defensive turret can hit anything 8 squares away from it.” I definitely agree that there should be a shaded relief overlay option like you proposed. I disagree with the notion of having range lines for pd’s curve around the terrain. Not only would it indeed by cpu-intensive, it also doesn’t make sense that the pd would somehow “know” what it could hit before even being built. Wasn’t there an update in one of the patches a while ago that said it added a check to pd’s to see if their hits would intersect with terrain before shooting? I could’ve sworn I read that in some patch notes.
August 17th, 2007 at 4:29 pm
If you think about it, a PD [i]would[/i] know just about where it would be able to shoot, since all it would have to do is know its range and then it would be able to see any obstacles in its way. you as the builder cannot though, because you are looking at it from god mode, not from the perspective of the pd.
i do see the problem with being computationally expensive, but it would only be that way when you were actually placing the pd and the calculations were taking place. after that all the comp has to do is remember the answers (for units it would be more of an issue, because the calculations have to be redone every time it moves, which is pretty much all the time). of course if all your doing is checking to see if the shot hits the ground its not that big of a deal, since your doing that anyway. just tell the unit to stop shooting/move if all of its shots impact the ground.
of course with a topography map this would be less of a problem, because you would be able to make the judgment call yourself if a hill was going to be in the way.
and yes, just the fact that we can gripe about stupid pd shooting the ground repeatedly because there shots are simulated is a great testament to how good sup com is, and that is not meant in a sarcastic way, i really mean it. all we need to do is refine how it works to make it better.
August 17th, 2007 at 4:31 pm
Psssst Lucus, WordPress uses HTML, not bbCode.
August 17th, 2007 at 4:41 pm
yeah, i saw, too bad i cant edit. . . lol
August 17th, 2007 at 6:44 pm
I think cartography mode was supposed to help you see terrain elevation, but no-one that I know uses that anyway.
August 17th, 2007 at 8:01 pm
This is making me think–in the supcom universe, how are units, pd’s, etc. made and how do they work? Does each unit and building have built-in optics linked to AI, enabling it to determine where to shoot and if it hit or missed its target? Or does the human in the ACU remotely control all of that? It would definitely make sense that if the pd’s were real, they would be able to predict terrain in the path of its shot (or at least be able to tell that shot after shot is hitting the ground). But since it’s a game, I think cpu usage is the reason this isn’t implemented. The whole workflow of a single shot fired would probably double in complexity, since the results of a shot would have to be reported back to the unit/pd, then be analyzed. I have a feeling that if the devs did implement this, people would be complaining that their pd didn’t shoot at an enemy they could swear was in range.
August 17th, 2007 at 8:11 pm
well, it seems to me that you have to calculate the result of every shot anyway, to see if it hit the hitbox of the unit it was aiming at. and then if it did hit you have to subtract the damage done from the units hp, if you did not hit then you don’t subtract the damage. so your are already calculating if you hit or not. all i’m saying is put in a check, and if you are not hitting, then find another target.
but then again, i’m not a programmer, so i don’t really know. maybe if a dev reads this he could give a good reason why not.
August 17th, 2007 at 9:54 pm
computationally intensive my arse. i reckon i could mod this. might give it a shot if i get a break from relaxing.
August 18th, 2007 at 11:54 am
ALL units already check if they are hitting their target. If not then they proceed on to another target within a certain arc of their current target. So they will never make a full 180° turn to fire on something if a unhittable target is in front of em.
And the topography map is already in Press Y ingame to open the menu.
August 19th, 2007 at 3:28 pm
Yes, the topo map is already in the game, but it does kind of suck a lot. Last I tried to use it I drowned in green. Cyde’s idea where it would reference the min/max height based on what you can currently see would be awesome. It’d also be nice if instead of opening a menu you could have it so that while holding Y you’d get the topo thing, so you could quickly and easily see how things look while playing something.
August 19th, 2007 at 10:23 pm
first the thing everyone has missed…. how the new build order go? and was it any good???
next
computationally expensive…. i doubt it…
all it has to do is say im here @ this hight…. shade everything inside my range thats above my hight… thats all u need… not its not 100% accurate… but its close enough….
ive many a time build PD or arty to find out that only 1 in 5 shots would make it over the hill…. or around the hill…
i will admit it makes a nasty trick when u build a line of arty beside a mountin to shoot @ the mountin most of the time and sometimes shoot over or around the moutian… again a realy nasty trick… (wastes a lot of mass and energy but who cares!)
REQ: NEXT supcom (either expansion (doubt it) or Supcom 2 have destorable treain!!! and relaimable mountians!
August 20th, 2007 at 2:50 am
Sorry about the not commenting, i saw the post about the topography map and decided i had to give it a try (didn’t even know it was in the game, guess thats what the manual is for. is it even in the manual?) anyhoo, the build order was kinda a bust. actually that’s kinda what i find frustrating about the game, its very econ centric. i love going out and managing my troops and building forward fire bases, and i love the scale. but too often i find myself winning on the front lines, and horribly loosing on the econ side, which means that i loose the game cause my opponent managed to get out a t3 arty and kill my base, even though i was constantly pushing him back on the map. the problem is that there is just no real reward for having map control. i guess what i really want is a war of attrition, not one where there is a single decisive blow that is impossible to recover from. such as your opponent getting t3 arty when you don’t quite have enough to over run his base. or a decisive blow that takes place on the battlefield. that would be ok with me. but getting t3 arty for-the-win just means that you beat me in the econ. especially if you got your t3 while i was grinding away at your base defense. i guess thats why i liked Rise of Nations so much. octopus strat was mandatory in that game. there just really isn’t incentive in this game to do it. it is much more efficient in this game to turtle your main base, never expand and put the resources that you would have put into defending satalite bases into your econ. maybe ill write an article about that.
and your right, the current topography map sux. a lot. you still have to zoom in half the time to see if its a hill or a valley.
August 21st, 2007 at 8:34 pm
well LucusLoC id suggest u get into 2v2 maps….. me and a mate play well together… i normaly go econ… and turtle the hell out of both of our bases and he just builds and attacks… al i have to do is keep positive mass…. and lots of it… and he just has to keep spending it… so if im @ +100 mass hes @ -110 mass…. its great when u look @ the scores @ the end…. he spends like 2.5x what mass he produced… and i wasted sooooooo much mass….. (that all went to him anyway)
takes a diffrent type of thinking…. and good team mates… oh and u NEED live chat… typeing just doesnt work…
also what works well… is ill give him SCU’s then upgrade another 2 and give him another 1… etc etc…
also id like to see the option (think its in AOE) where if u both choose the SAME color… u both can control the same units and the same resources…. made some intresting games in the past because u had 1 base and 2 pplz working on it… giveing u twice the micro!
August 22nd, 2007 at 1:15 am
The idea from AoE where two people control the same exact units would, I think, work very well in SupCom, to the point that it would surely be considered cheating if it weren’t specifically sanctioned. Just think, one guy controls the early build order until 2 land factories are built, then each person takes a land factory and starts building their own engies. One guy focuses on the rest of the build order (economy mostly), while the other guy micros the early T1 push. Two players acting as one would surely be able to dominate a single player, since the two-player side could micro those early T1 combat units so well while still keeping the economy going back home. All that said, it may not work out, since in AoE2 it was an absolute nightmare unless you had VERY good communication with the other guy. I tried it once at an AoE2 LAN party, with a guy I didn’t know who was sitting on the other side of the room, it was a nightmare having to type stuff to him–we wasted so much time and resources on duplicate buildings (martkets, blacksmiths, etc.)–but we did manage to hold off the Goth Rush for quite a while, until our teammates got wiped out and we ran out of gold. End rambling.
August 22nd, 2007 at 8:54 am
I think the ability to have multiple people controlling each side would add an incredible aspect to the game, especially if you have a large group of friends who enjoy playing, but only 8 max are allowed in a game, fewer on most maps. Sure, there may only be 8 spots, but with two people per spot, now you’ve got a 16 player game. The required computational resources are increased due to the additional networking, but not as much as having 16 seperate players, since you technically have the same number of units for the engine to keep track of as a standard game.
I know I’m probably in the minority here as a casual player who doesn’t have time to devote to playing online; my friends and I enjoy teaming up for a comp stomp about once a week. When you have four friends playing, you can play a 4 on 4. With five friends, a 5 on 3 is much to easy. But with this method, as the number of allies grows, we can just double up and keep playing.
In a comp stomp, the best part is communicating with your buddies, coordinating attacks, that sort of thing. This would be just what the doctor ordered. (I think TA Spring had this.)
August 23rd, 2007 at 10:40 am
to the article writer :
maybe you should check out a 2nd monitor and enable topographics on it ?!?
i use that since some months and now i know exactly where defensive points will shoot into hills…
August 23rd, 2007 at 5:11 pm
i do use a second mon, and i set it to topography. it helps a little, as at least now i can see where i need to zoom in to have a look-around. but as i said earlier, you cant really tell if your looking at a hill or a valley, unless its really pronounced. in that case you would be able to tell anyway from the regular map. its nice to see that GPG actually realized that you needed a topography map, but its current implementation leaves much to be desired. . .
@meddish: your idea sounds good. . . if one guy wanted to play “sim city” in the sup com world. maybe i can get a friend who doesn’t like rts to give it a try against the comp. maybe my girlfriend. she has not tactical or strategic sense, but she is better at managing stuff. as it stands now, all my friends that play would want to be the ones in the commanders chair.
August 24th, 2007 at 9:42 am
You’r right !
I’have a suggestion for another article or just a reaction.
I’m french so be indulgent about my english :)
There’s a thing that bug me sometimes. It’s about mixed engies like T1 + T2.
If you select a group of engies mixed like that, you can’t build anything why the “structure bar” do not show a mix of the structures that can be built by both of these engies ?
If there’s a T2 engi mixed with 5 T1 engies you would have been able to build a T2 pgen then the T1 engies would start assisting the structure that is being built.
When your base is full of engies mixed together it’s alway a unnerv to select one type of engi at a time, then assist manualy an other group, this is unnecessary micro i think.
What’s your opinion ?
August 24th, 2007 at 12:30 pm
Life: You are right! A mix of T1 and T2 engineers should are able to build any T1 or T2 structure and should be able to be ordered to do so. That has annoyed me many times and would seemingly be a simple fix that would reduce some of the micro. And we all want to reduce micro, right?
August 24th, 2007 at 1:09 pm
Since we got on the topic of engineers, I have trouble with the selection of engineers. I know you’re supposed to be able to hit ctrl + period or some similar key combination (I forget the keys exactly as I’m not home to look in the manual files) but I’ve never gotten them to work right.
Can someone explain the engineer selection controls to me?
August 25th, 2007 at 5:20 pm
To the OP- use Cartographic view? This is a complaint from many patches ago, use Cartographic.