The Trouble With Balance

I’ve been having balance issues lately. It’s not due to the giant bottle of gin I have just bought, either. Rather, it’s all about the problem of balance with the cost of stuff, crystals, and the value of the gidrahs when you destroy them.

First of all, I have a feeling that there are too many crystals on most worlds, which let you harvest an awful lot of cash that way. Secondly, I have another feeling that silos may be just too damned useful. The difference between having silos around all your refineries and not having silos is pretty drastic. So I think they’ll be first up against the wall when the nerfing revolution comes. Which will be around the time of Saturn. Silos will become somewhat less effective – I think I’ll reduce their efficiency increase to just 10% (down from 25%) per silo. And there will be a smallish reduction in crystals. In fact there will also be a reduction in crystals if it appears you’ve got a whole lot of cash in the bank, too, as you’re clearly doing rather well.

Also, the amount of money that can be harvested by shooting gidrahs is pretty massive. This sort of lends itself to a gameplay balance issue whereby the player doesn’t have a whole lot of incentive to bother collecting crystals, especially if you get the aliens cornered in a tight spot and can keep them at bay using minimal firepower and expense. Or alternatively the collection of crystals can be trivial if they’re behind the choke point on the map, meaning you’ll come away from a level with $30k in the bank.

The difficulty range was supposed to work best in the range of $0-$15k or so, but clearly it seems I’ve got to extend it to cater for up to $30k. And so I shall πŸ™‚

Work on Saturn continues, with suicide exploding gidrahs working nicely, and the code for flying aliens in, but they don’t yet drop bombs. And we can’t seem to get a good graphic for it yet, so to distract himself, Chaz is doing the scenery instead, which revolves around toxic waste and pumping machinery. I tried a nice simplex noise fog shader for some cool weather effects but discovered it was too slow, so that’s on the back burner again and unlikely to see the light of day.

And finally, I got around to fixing the three older games, Droid Assault, Titan Attacks, and Ultratron, after leaving them broken for over a year (the wouldn’t work on ATI cards, and also wouldn’t install on Windows Vista or Windows 7 properly). I’ll upload nice fixed versions this weekend with any luck.

67 thoughts on 'The Trouble With Balance'

  1. I think the problem is that the cash you gain from doing missions is variable. If cash earned was fixed then you can balance better, and the gameplay would still be the same.

    1. What I mean is… starting cash should always be the same (depending on the level). Cash earned by completing a level should always be the same (depending on the level). Cash to spent on upgrades and cash to spend in game should be separate if you want better balance.

      1. no, no
        and no thats not how it should be, you should get money on how well you play the level nuff said

        1. That’s the immediately intuitive solution except that it creates a positive feedback loop, that is, the better you play a level, the more money you’ll have in the next level, therefore the better you’ll do, and so on, until you’re racking up ridiculous money and it’s not getting any harder and it’s not fun any more.

          1. unless you make the dudes get harder, or make some disaster happen that saps your money say, our base died, we need to steal 5000 to survive the next wave

          2. The simplest solution would be to reduce the amount of crystals on a level depending on how well you’re doing, if a person starts a level with more than 10k in the bank and they’ve researched something every level they should find very few crystals scattered around. Conversely, people doing poorly will find that a level will have more crystals on it and they can catch up in terms of turrets and research.

    2. How about having for every three level, one level without any resource on the map ?

      Of course the player would be warned 3 turns in advance so he may stock some cash.

      This may make the player focus even more on optimally managing his funds.

      And kudos for the great game, played with the demo, then bought it the next hour

        1. I’ve made so many balance changes for Saturn this won’t really be necessary I think. There’s a greatly increased emphasis on mining now, and plenty to mine. I have also somehow got to get across the notion that you do not have to research something every level if you can’t easily afford it.

  2. If you released enemy and building stats we can work out balance issues for you with good old spreadsheet. Currently I think not enough emphasis is put on harvesting as you can pretty much get buy with the money you get from killing enemies. On one level I harvested all the minerals and spent all my cash, by the end of the level I had 10k in the bank. It’s right that tougher enemies should offset the cost of the bigger guns but the overall income from enemies should take a back seat to crystals. Maybe have a second set of more expensive crystals on Mars that are red to keep up with the curve.

    Glad to hear you’re still supporting your other games.

    1. I am +1 for this, I want to help out and chip off the stress of the balance from cas’s shoulders
      and as for the red crystals another +1

  3. cas Just wondering why wernt you in notch’s private smp test? I noticed I didnt see you :l
    you did know about private smp right?

    1. too busy, even if I did know πŸ™‚ I’ve got a full-time job which I’m desperate to escape and a 19 month old daughter to look after!

  4. You could slow down refinery extraction too. I finished the Mars level quite easily without ever making research investments in refinery efficiency or silo development or anything on that technology branch. Never built a silo so I don’t know what they do.

    maybe instead of instant mining -> resource credits account, you could implement little trucks to drive the ore from the refinery to the HQ before it goes into the player’s account

    1. That’s a great idea, it would make more sense to have to ‘transport’ the harvested crystals back to the base to be used (i.e. turned into $) by way of some small vehicle* than having the crystals suddenly converted into $ from the refineries… I suggested something like this in an earlier post… plus, the benefit of this is that it would make another target for the titans to attack, thereby making the game more strategic I think…..
      I think the problem of the game being somewhat too easy with regard to accumulating large amounts of $ (and being able to purchase lasers etc early on in the game) is a pretty simple one to fix: namely, a ratio whereby (as others have pointed out)
      1)more $ in the bank = less crystals placed on a level (or less big patches of crystals)
      2)more tech researched (i.e. bigger and more efficient guns etc) = more unfriendly placement of crystals on a level
      3) combining points 1 +2 would result in patches of smaller crystals spread further out from each other making them harder to defend…

      *this would also serve as a positive reason why you have to click on a refinery to ‘collect’ the harvested crystals!

      1. Ah it already does this with the crystals (although the formula’s a bit more complex). Though it doesn’t shrink the individual crystals. Good idea πŸ™‚

        Adding a transporter vehicle is a lot of work and probably makes the game just a bit too complex. I dunno. I like it the way it is – simple – and the next tweak I shall apply is attempting to balance the amount of money available on a level and the ratio of crystal money available to gidrah money.

        1. yeah, I can understand what you mean about making it too complex, I just thought that it would be a good visual mechanic. I still think its an excellent idea! πŸ˜‰
          Maybe one way to balance the amount of $ obtained from shooting the titans should be related to how far away from your base they are* …for example, if you’ve got a weapon supported with scanners (increasing the weapons range) that can take out titans from a distance then decrease the $ amount you get from slaying them; whereas if your suffering from up-close and personal onslaught of the little devils then increase the amount of $; or perhaps that would change the game dynamic too much??
          * or even make it dependant on what type of weapon kills them???

  5. I have just found a HUGE bug on Mars! One of the big titans that came out of a crater was standing still. I thought he was just planning his assault on my base until he became the last guy left. My robot army didn’t move into attack him (the titan is still standing still), so I decided to put a $250 blaster in front of him. It didn’t fire, I put one behind him it didn’t fire, I drop 3 Smart bombs, AND he still lived. So I had to restart the level. It was level 27, but that probably doesn’t matter because it’s randomly generated (which I love).

    1. Damn, that’s that “gidrahs not dying when they’re supposed to” bug rearing its ugly head again. I suppose the crater spawners do spawn in a slightly different way… I’ll investigate.

  6. Regarding the money problem:

    One thing I really liked about this game is that the better you play it, the easier it does get. To see an example of a game that would play opposite, take a look at a game like Dune : Emperor. In that game you would attack or be attacked in different maps but it when back and forth based on rounds. In one level you could mine a TON of spice and build an indestructible base, completely crushing the enemy. However, if you were attacked THE NEXT ROUND on THE SAME MAP, your base, all of your resources and anything you had built up disappeared. I considered that to be anti-rewarding, in fact, it’s almost sadistic. What it says to the player is that you should play a level as best as you can, but in reality you have no control over the outcome of the game.

    On the other hand, you are right, too many resources can make the game way too easy to play and if there is no chance of losing then there is nothing to be gained.

    What I would recommend in this situation is to have the game setup so that the player can unlock most (save maybe the biggest 3 guns) by the end of Mars. Have resources be plentiful in the beginning of the game so the player feels they are getting some reward if they play levels correctly and get extra cash. Then, starting with either Mars or Saturn, reduce the point value of enemies from earlier worlds especially smaller enemies. With less cash coming in, players will still be challenged to run a tight ship. In fact, would want to consider increasing the crystal resources and removing points for killing enemies all together. Especially repeat bosses. This will increase the importance of mining in the game while keeping the overall resources down…

    I mean c’mon, whens the last time you got a cookie for actually blowing up an enemy…from the enemy, anyway?

    As for cash problems. I don’t know if you have played Eternal Red which is another tower defense game, but one weakness that game has is that once you figure out the secret of how to beat it, the game becomes incredibly dull except for the levels which that secret doesn’t work due to limitations…

    (The secret, by the way, is to build one machine gun as close to the door as possible then max it out, maybe two guns tops, after that, all you have to do is max them out and then add and max out additional guns one at a time. Making multiple guns is completely counter-productive as is using acid guns or grenades.)

    Anyway, to finish, I think having a lack of resources in later levels is a much better solution than trying to balance the build trees and resources to be linear throughout the game. It’s more practical, should be easy to implement by reducing points for enemies, and won’t reduce the importance of mining.

    I’m on a horse.

    1. I’m actually in favour of completely removing the points values earned from killing gidrahs anyway. Except for one small problem, which is that if you haven’t got any refineries or crystals left (or indeed on level 1 where you don’t have refineries anyway!) there’s no way to earn any more money until the level ends! So perhaps I should just make all gidrahs worth a basic $10 and $50 for the big ones and $1000 for bosses with occasional bonuses arriving in the form of pickups. This will more or less mean I don’t have to worry about the balance issues from the fact that the amount of cash from gidrahs on later levels far outstrips that available from crystals. I should get it so that it’s about 50/50. Some twiddling with the difficulty curve to make it accommodate difficulties when you’ve got up to $30k in the bank should help a bit too.

      Where I want it to get to is that beyond the tutorial levels (1-10 or so), you should pretty much always be on the edge of your seat in a state hovering between smug zenlike satisfaction and unexpected panic. The tide of the battle should be able to turn in the blink of an eye.

      Experiments tonight! But first I’ll be attempting to upload the fixed versions of Droid Assault, Ultratron and Titan Attacks. I was thwarted last night by an Eclipse hiccup that got me stumped for 2 hours, grr.

      Btw, the only tower defence game I’ve ever played is Defense Grid, which is top fun, if a little difficult and sometimes monotonous compared to the flowing, shifty nature of RoTT πŸ˜‰ I heartily recommend it.

      1. Oh, Defense Grid, I’ve played that yeah. It was a pretty good game. Probably the best tower defense game ever made considering the almost ridiculous level of 3D it had for a simple tower defense game. I still prefer the graphics/replay value of ROTT though. I played through the first 10 levels trying to get as much unlocked as possible. I still enjoy the replay value and that the levels are randomized even if they are sometimes easier or harder.

          1. When the endless survival feature the tech tree, and will we be able to play it in 1.5?

            1. I’m not sure that sentence made grammatical sense πŸ™‚ Endless and survival are two different game modes. Endless is like campaign mode, where you play level after level. The levels scenery and colours are picked at random from all 5 worlds; the size will gradually increase from the minimum size to the maximum size, and the level duration gradually increases from minimum to maximum duration. The aliens are introduced one at a time and follow a simple pattern of appearance. And it’s called endless because once you’ve reached maximum duration/level size, the game simply carries on generating levels at this size and duration, gradually getting harder. Your eventual “score” will be the final level number you reach. Endless games are likely to feel … endlessly long πŸ™‚

              Survival mode generates a single random level. The aliens spawn from any available edge location in a constant stream; the list of aliens, time, and difficulty determine which alien is spawned next. The crystals spawn randomly on the map over time such that there is always a constant amount of crystal to harvest. You begin with only the blaster; to research an item you simply click on its icon (we’ll probably colour it differently to show it needs researching) and that unlocks it for the required amount of hoarded cash. Research cost is likely to be about 5x the basic value of the building. There won’t be any of the “intermediate” research options. Your score will effectively be the length of time you have survived for before being overrun. I have a feeling this mode will be a lot of fun and really replayable.

              Both modes will be available only in the full version of the game (not the demo).

      2. How about instead of money from enemies, the base produces a stead income every few seconds so that the player is never unable to build a blaster and you wont have to try to balance enemy income. Then that way the amount of money you can earn is fixed based on the amount of crystals on a stage.

  7. For the record, I have Windows 7 Ultimate 32bit and Ultratron + Droid Assault installed & are running fine on my system :F

      1. I just downloaded Droid Assault again, but it still wouldn’t start. So… how’s fixing those games, then?

  8. Defense Grid was actually the first tower defence game I played; admittedly the graphcs got my interest initially but I really enjoyed it, and got sucked into it;excellent game! Subsequent games in that genre that I’ve played haven’t really gained my attention, until RotT that is….

  9. Have played through the game a few times and am loving it. One comment about the auto-adjustment of the difficulty upon restarting a level — the level becomes too easy too quickly! Sometimes after losing, I would like to try a different strategy after losing, but the game seems to immediately adjust the layout/enemies to be easier. How about an option to redo a level precisely the way it is?

    1. I’m fine tuning that so that it’s a little more subtle. Right now it lops a whole 25% off the difficulty factor every time you restart a level. I will adjust it a little more slowly – barely noticeable – by 10%. It’s a bit tricky to replay the level exactly as it is in any case because the randomness can account for half of the difficulty anyway.

      1. Don’t do that! One of the things I love about this game is that if I loose horribly the first round I can simply hit restart, and get a level that is more in my favor.

        1. Or, just restart a couple more times until it becomes easy enough to pass.
          How’s this, restarting lets you keep the crystals you’ve mined before you lost?… maybe not

          1. Its always a delicate balancing act (regarding difficulty in games) but I gotta agree with Chi on this one; I think it creates an unfair advantage if a player can restart a level and the difficult is reduced… that panders too much to casual gamers, and at the same time creates a dis-incentive by rewarding players without enough skill (which has the knock on effect of reducing a players motivation to think of different strategies to the one that they’ve been using unsuccessfully)
            Btw Cas, I’ve noticed that gidlets seem to target my toughest weapons once they’ve burst from the queens… is this their a.i. (they seem pretty clever buggers!)
            Oh, and I still HATE the mid-level speedup… but will admit that it does add to the challenge! πŸ˜‰

            1. Certainly people have different styles of play. Some people mostly want to be entertained — think theme park rides, guided tours, etc. — and other people like to be challenged — think mountain climbing, brain teasers, etc.

              The auto-difficulty suits the first group of people better. For them, getting stuck on a level ruins the flow of the game and interrupts their enjoyment. On the other hand, for those who want puzzle it out and try, try, try, again, the auto-difficulty deprives them of that challenge.

              Given that both styles of play are equally valid, though perhaps with more people preferring auto-difficulty, how about making that the default, but give folks an option to turn off the auto-difficulty? Or maybe just a slider that controls the amount of tuning — from a lot to none at all?

              1. How’s this: When you lose a battle you get 2 options, Retry and Abandon, if you pick Retry you start the level again exactly as it was, if you pick Abandon you get taken to the level select screen and a new, easier level is generated. That way everybody wins.

                  1. I do like the solution; however, “Abandon” is not obvious. I would assume that it means “stop playing”.
                    Also, having to go to a menu/level select is not good either.

                    I realize “Restart this map+this difficulty” and “Restart with a lower difficulty on a random map” are a little wordy.. πŸ˜›

                    1. I agree, it would get really annoying to have to go through three different menus instead of pressing a button that said restart.

    1. All you got to do is set up some rockets and spreader cannons with rangers, and your good to go! Also just put little blasters in front of the ghosts if you’re like me and don’t know what to research to get the thing that shoots them. Unless you are trying to complete it without rockets cause I haven’t done that yet either :).

      1. you get capacitors with reactor and battery, I use rockets and multiblasters, but I never have enogh baricades to go around :c

    1. Chaz has done all the scenery and gidrah graphics for Saturn now. We’ve got the suicide gidrahs in, and last night I got the flying bomb-dropping gidrahs working, though they both need minor tweaks. Chaz is currently working on the boss for Saturn.

      I’ve also made tons and tons of subtle balance changes and fixed a load of bugs. Wraiths and armour are handled very slightly differently now so you’re never down a completely dead end research wise – wraiths can be targeted by blasters but only when they’re really close to their target, and all bullets have a small chance of penetrating even the toughest armour to cause minimum damage. Obviously it’s very suboptimal to use the wrong weaponry so a cunningly thought out research path is still important for success.

      When Chaz is done with the boss, we’ll work on the story screens and I’ll probably end up doing loads more playtesting to make sure it’s not too hard, and then we’ll have Saturn done, I think. It’s looking like 2 weeks or so from now ish.

      In the meantime, sales == 0 πŸ™ Which will leave us a bit buggered. I wonder why no-one’s buying it any more?

      1. Thanks for the quick update, sounds like it’s coming along great! You might want to look into selling your games on Steam though. They always have a lot of advertisement for indie games.

        1. I ought to do a tangential blog post about Steam. Lots of people aren’t really aware of the deal with Steam. Steam is more or less by invite only. They don’t just take on any old game! And of course being on Steam will totally stifle our growth, although it might keep the wolf from the door a few more months.

          1. You really do need more advertising to get your game seen by the public. It was by sheer luck that I noticed your game when someone mentioned RoTT awhile back in one of the Steam’s subforums. The only real advertisement I have seen to date was a mini article on Rock Paper Shotgun. Is there any chance you can get some coverage on sites such as Bluesnews, Neogaf, Gamespot, et cetera?

            It is a shame to hear about invite only for Steam, but if games such as Harvest: Massive Encounter are accepted, there is no way Puppygames would be turned down. As for sales, how would that stifle growth for your company? Sure they would take a percentage of each purchase, but the amount of sales through Steam would more than make up for the loss of a few bucks on each copy sold. Add in a little Steamworks support such as Steam Cloud for savegames (these tools I believe are readily available to developers) and the promise of future updates such as Endless / Survivor mode, and watch your fanbase grow exponentially (which in itself seems like it would be worth the percentage cut on each sale). How many millions of people around the world log into Steam each day? Their community chart tends to average around 2 million concurrent users during normal hours. Toss in ease of access to your demo, the amazing 50% off pre-order sale which would be front page (could even run it with a catchy 1337 title), and I would imagine your worries about sales for this game would be over.

            I hate to sound like a broken record bringing up Steam over and over, but it really is that massive in digital distribution, and it really does help out indie developers such as Puppygames. If you don’t believe it, take a look at companies such as Tripwire Interactive, DiezelPower, Cadenza Interactive, Frozenbyte, and Hidden Path Entertainment to name a few. It appears that one just needs a solid product and some playerbase interaction (both of which you have in abundance) and Steam will pretty much takes care of the rest.

            RoTT is truly a hidden gem of a game with its unique art style, animations, sound, ideas, and charm, there is absolutely no way it would not appeal to the base defense crowd.

            1. We seriously know all about Steam, don’t worry πŸ™‚ There is an ominous background issue to Steam though, which is that of customer leakage. Our mailing list won’t grow through Steam, though the bank balance might.

              Still… watch this space eh.

              Advertising will resume in a few weeks when Saturn’s done.

              1. I don’t know if any of you have ever been to the site, but it’s a site dedicated to any and all indie games. I’m almost positive that it’s free to post your game on there too. It’s done by the same people who made (an awesome site for game mods). Hope that helps getting the game more well known! πŸ™‚

      2. First a warning and an apology for the rambling post, I tend to do brain spew on my writing and it just forms as things come to me…

        First a blatant plug πŸ˜€

        How about releasing some lite version of your games on iphone? The device is fairly limited but plants vs zombies works great on an iPad (big enough screen). You can link back to your site from those games and you can get a little extra revenue. If you want I can hook you up with a development studio (mine :D) if you’re not familiar with iPhone development.

        Social Networking:

        Social networking and casual gaming is the new black, there’s money to be made here and people are leaving it on the table.
        Try getting your games onto facebook or bebo or other social networks. I’m sure you can whip up a browser based demo version of your game and then you’ll have endless mailing lists instantly and also you can do stuff like wall posts or feeds. Micro transactions, doesn’t need to be anything major, different coloured turrets or something would do, you might even dole out planets in the campaign one at a time for a small fee so people aren’t hit with a single lump sum… worth a shot.


        Have you entered your games into any indie competitions? I haven’t really checked so you might have done so already but those tend to generate loads of exposure. If you get a mention on the indie game festival or maybe even win a prize it’ll do wonders for your income.

        Another obvious one: Send copies to reviewers

        Here’s a blog by Kieron Gillen from 2005 about how to get a game reviewed:

        Some of it is fairly obvious but still worth noting, sometimes you just let things slip or you miss a trick. Getting a review can mean mega bucks, just look at mine craft sales after PCG and TFblog posted about it. 1.5k sales in 24 hours (I was spurred into putting down my 10 euros when Notch state that the price was going up for minecraft XD). It’s also worth sending a copy to web comics like penny arcade or VGcats as they have a huge readership.

        Here’s a summary of Kieron’s talk from world of love:

        1) Send us the game.
        2) Read the press.
        3) Know why your game is interesting.
        4) Seek advocates.
        5) Mobilise your army.
        6) Image counts.
        7) Be Good Copy.

        D) DEMO. LIKE, OBV.

        Encourage your fan base to repost your blog and updates onto various forums and websites. Viral FTW.
        Also, competitions and news letters always gets the juices flowing. Send out free copies or Gidrah plushies (I’d buy a plushy at the drop of a hat XD)

        Do interviews for games websites or podcasts. Let them know why you’d make a good interview and how awesome your games are, maybe give away a copy on the podcast to listeners who can answer a question? (recurring theme)

        Publisher Funding:

        Have you considered funding? There are people who are willing to fund indie projects, they take a cut of the profits but you get to stave off hunger for a while. On the plus side they might help you out with advertisement, even simply getting funded would be news worthy for most game blogs on the internet. Try the indie fund ( ) they might not accept a game that’s half way completed and already on pre-order but maybe for a future release. They are however, run by indies for indies, they wont go running off with your IP or go telling you what you should do at any point unlike big name Publishers.

        It’s been stated already but selling your game on other sites as well as your own will boost sales tremendously. You can then have a blog / website link in the game itself or even a news feed so that people who buy your games will see what new releases or announcements you’ve made here on the site. Maybe do an in-game advert for another game maybe in the corner of the main menu so as not to be too intrusive. Or maybe even pay for advertisement or do an ad swap with other indies.

        Publicity Stunts:

        Cheap and effective way of getting media attention.
        Post a youtube video of mock news report about alien invasions or have a guy dressed as a Gidrah running through the streets of London (or something). If it goes viral you’ve hit advertisement gold.

        Conferences / Conventions:

        I know your time is limited so you might not be able to hit the conventions and things but there are always some game conference going on in London where you can flash your game around to get some industry exposure.

        Hope some of that was interesting. If I think of anything else I’ll post a reply.

Comments are closed.