Community Discord discussion
We came up with a couple things that everyone who was in the chat agreed with. The TLDR is first, but I'll elaborate on every point below it
- Aggroed camps give a resistance debuff
- Archive shield based on structures alive (archive towers for example)
- Apocalyte spell becomes a projectile, time until impact depending on range, starting at 1.5 sec up to 3 sec at max range
- Increase the bioshards gained on kill based on level
- Medium and Hard camps earlier(5 / 10 minutes) also reduce respawn time
- Titan scales too much, longer games impossible with good teams
- Tier research longer and more expensive
- T1.5 needs to be stronger, tech rushing too good
- Side farm is not giving enough shards (Hardcamp > 10 minutes of side farm)
Hydros ult -> Plate and small heal for units affected (Quadrapus inspiration)
Aggroed camps give a resistance debuff
------We were trying to think of a way to make camps with more risk and more reward - With the middle medium camps for example being like League's dragon rather than the current form where the safe one is pretty safe to take and the other is risky. This is the consensus we arrived on - It may be easier to stop the opponents from taking their camps and make it more risky to take a camp, but it won't make it harder for small groups of units to take a camp uncontested.Archive shield based on structures alive (archive towers for example)
-----This is sort of like shield batteries used to be. I think it could work better with archive towers better though because it would increase their value and make diving more risky and pushing them first with titans better.- Apocalyte spell becomes a projectile, time until impact depending on range, starting at 1.5 sec up to 3 sec at max range
-----This was an interesting idea. I like this idea and the idea of reducing their range. Make them need to be vulnerable and run in with dervishes to land a destructive prophecy! This would make them riskier without losing their identity. - Increase the bioshards gained on kill based on level
-----This may be a good thing. It feels like higher levels don't give much more than lower levels at all. - Medium and Hard camps earlier(5 / 10 minutes) also reduce respawn time
-----At the current game pace, mediums spawn exactly twice and hards exactly once at most. This doesn't feel right. - Titan scales too much, longer games impossible with good teams
-----This is definitely true, but it may just be a symptom of building skipping being the best strategy. - Tier research longer and more expensive
-----Right now, tier 1 lasts for 2 min, tier 2 lasts for 2 min, and tier 3 lasts for the other 20 minutes. This doesn't feel balanced. My thoughts are: what if instead it was aimed that some games end before tier 3 and you needed to get some low tier units if your opponents felt like being aggressive? Say tier 1 being around the first 10 min,, tier 2 focused play started for the next 10 and for the rest of the game, people start using their late game armies. This would be a rough guideline, so some people would rush for fast high tier, and some people would go for tier 1.5 heavy to try to end. - T1.5 needs to be stronger, tech rushing too good
-----Spending your shards on tech rushing is much better than spending it on early game focused units. Teching should be a risk that you take. - Hydros ult -> Plate and small heal for units affected (Quadrapus inspiration)
-----A lot of games and late game fights are extraordinarily dependent on a good hydros ult. This needs to be changed.
Comments
Thanks for the write up.
I want to reiterate that I believe that the pacing for Tiers pre-patch was fantastic. The problem was mainly that there weren't enough interesting options at Tier 1.5 and Tier 2 for all factions, so some factions felt like Tier1-fests for large portions of the game. This isn't the fault of the pacing! It's a matter of lack of options before Tier 3 that encourage people to deviate from a strict Tier 1 army until lategame.
I think this patch showed that Tier 1.5 and 2 need to have interesting, aggressive, and attractive options to take control of a game and press a lead, and that the game needs to be paced in such a way that this is possible. The game felt a lot less "solved" when this was possible and every game felt different. I want this feeling back!
I'm also not sold on Apocalyte travel time on their nukes, as I still feel the wombo-combo with them and Blue armies is too oppressive or at least uninteresting, and neuters a lot of interesting potential strategies outright.
The existence of Apocs informs just about every decision in army building in the game right now and if your strat isn't at least decent against Apocs you have to abandon it outright because they don't just kinda beat anything they are good against, they wipe it off the map before it can be a factor in a fight at all. Travel time will probably help that and make them less slam-dunk good against a lot of stuff but I don't think it fixes the problem entirely.
I am not in general a fan of their design in a world with such huge AOE stuns/snares and I don't know how I'd fix that without major changes. But I'd rather see SOME change than nothing.
I agree with resistance debuffs while taking camps. Invading your enemies and taking the risk now beyond pushing the early Easy camps on in the Medium bushes happens (I've seen my share of double and triple invades) but pushing Easy, Medium, and Hard camps that are "supposed" to be your enemy's typically happens during a Titan push. This feels more like an intelligent cleanup off of an advantage than a deliberate decision to perform a risky invade. Killing the opponent's Medium camps through snipes does not feel so much as an invade as a steal.
I have not experienced the problems detailed above with Apocalytes. That is not the same as saying that they don't exist, merely that I have not experienced the problem recently myself. It may also be the case that I benefited from Xiniu playing mass Sandstinger on my team, who was quick to snipe those T3 red spellcasters.
I agree that Bioshards for Hero kills should scale to a stronger degree. Sacrificing your Hero has the opportunity cost of losing your Basic and potentially your Ultimate, but these sacrifices in the current meta feel worth it if you save even one T2/3 unit. For example, I would sacrifice Alder to save my Bramblethorn in my games and have done so... same for Trebuchets, Batterhorns, or my allied T3 units.
Reduced respawn times for Medium camps would help create separate points of contention at the same points of the game. I would recommend lightly tying Medium camps to Titan spawn times (maybe spawn at 3:30 and respawn every 4:00). This creates more decisions for the winning and losing teams... do we stop pushing this Titan we're ahead of so we can grab a Medium? Both of them? If we're losing, do we push for Titan anyways? Or do we safely contest two Mediums and sacrifice the Titan? This is my hope, in any case.
I do not have a strong feeling about having the Titan weaker or remaining at current strength. On the one hand, a weaker Titan allows for Bo1 in tournaments and would grant more opportunities for the losing team to battle back. On the other hand, keeping the current Titan strength means you can have Bo3+ and the added games will allow comebacks to occur through multiple games.
As for unit tiers and the current heavy teching meta: I believe it comes from the combination of four items:
- Tech units got universally buffed through lower Bioshards cost
- Upgrades were nerfed due to increased cost
- Teching became faster due to the boost in Bioshard gains on production
- Teching became faster due to players starting with two basic units and a small amount of Bioshards
As such, I think removing one or two of these concerns will solve the intention of making teching easier (and thus avoiding T1-or-Die) without keeping the Tech-or-Die that we have now. I recommend reducing the amount of Shards you're granted per production at Core and buffing upgrades. Upgrades affect units globally, and as such you get a higher return on investment for upgrades the more units you possess or the stronger the values of the individual unit. Teching very quickly means putting money into T2 instead of putting it towards upgrades, and thus there's the possibility of using a small number of unupgraded tech units or a mass of better upgraded weaker units.
Tier 1.5 units SHOULD NOT be more powerful. The Howling Commando and Sabre, for example, are fantastic units (though maybe Cinder Beetles could get some love... 7 Energy is rough). They excel at fighting T1 clumps. The reason they're poor in the current meta is that T2 and T3 units also smash T1 units, have more HP, have more survivability, can engage harder, do more damage, and most importantly THEY ALSO APPEAR AT THE FIRST TITAN FIGHT. If i can get two Bramblethorns or I can get six Howling Commandos for the first Titan fight, my choice is made for me.
I would like the side farm to be more impactful in terms of Bioshards. I would even be willing to try removing the Bioshards from the Core and making the map the only source of Bioshards (and boosting the side lane farming to compensate).
As I'm the one who came up with the "Replace Hydros ult with granting Plate to nearby allied units", I think I'll leave this alone ^_^
Apocalytes
I disagree with the suggested apocalyte changes because of how it implements complexity to the unit. Maybe in the worst of bad scenarios we would have to do something like that, but at the moment, it's not necessary. Apocalytes really embody the explosive nature of red very well. In terms of their offensive nature, they aren't much more oppressive than conduits I think, the difference is that they are much more survivable as a result of not needing to channel their casting. One solution could be to 'lock' the apocalyte for 1.8 seconds before the shot actually lands.
Personally I would wait and see how the meta develops after shard acquisition takes a hit. I always felt like there were many counters to apocalytes, like fast melee units. The problem is that these become unable to deal with 7 big melee units at the 8 minute mark, so they quickly became obsolete. Other spellcasters such as grove tenders and pre-nerf frostcallers also put some pressure on apocalytes. Another solution would be to reduce Apocalyte's casting range by 2 or something, but I am reluctant to support this because I don't want the whole colour to be put into a 0-14/15 range spectrum. Another suggestion would be to increase their shard cost by 10 or 20.
For the Changes about the Throne (Archive) hp or shield being buffed by other buildings:
While I agree, that rushing the Throne and ignoring all the other structures is maybe a bit to powerful and at some times frustrating for the enemy Team, I think it is a way to get the from the devs intended Game lenght (about 20-25 mins). Especially at the Start of the open alpha games would often last up to 45 minutes. The first big patch definitely helped that a bit, but imo there should be drastic changes made to other parts of the Game, if a Change is made to how the throne hp or shield interacts with other buildings (to reach the optimal game time). I think a System simmilar to Dota 2 or Heroes of the Storm would be best.
Either add Towers that have to be destroyed to destroy other things, or make it really hard to go through certain areas to rush something down without taking heavy losses.
And PS: Towers should be stronger imo. Especially the ones further back in the base.
Good post, I just think you(devs) should be very careful with multiple changes in a patch. Like you(Millea) say about titan scaling too much; it's also about the archive being so weak, so if you enforce both changes(harder to kill archive + weaker titan) we might have an undesireable outcome.
I agree with TokOwa that any changes to tier 3 balance probably should wait until we see the other effects of making earlier tiers more viable and prominent, or at least be very carefully examined. We could for example end up with an extremely dominant purifier meta.
I think emberfiends are enough for long range for red.
The problem I have with shard cost increase is that apocs are often not killed throughout the entire game. The range reduction (which I was more in favor of than the projectile) would serve to make them need to take risks in order to deal damage and make them vulnerable to being picked off (Making things like grove tenders microed well
Good suggestions, agree with almost everything. I think the suggestion for apocalyte is needlesly complex. Simply reducing the range it can be cast could be better.
Also don't think tier 1 lasting 10 minutes would be good. Tech rushing should be an option. The problem right now is that it is the only option. 10 minutes of tier 1 units would make for very boring gameplay.
My suggestion for reducing the dominance of tech options is slightly reducing the bioshards from camps and/or generator. Bioshard gains from grubs, towers and heroes should remain same. This would make split farming, killing towers and killing heroes more rewarding, while reducing the overall amount of bioshards you have.
I agree that 10 minutes is too long for tier 1 to be the only option, but then again that's not really been the case in prior patches. You could field a full squad of 8 Tier 1.5 troops by the first Titan or a couple of game-changing Tier 2 units, and that seems about right. It's been possible to make only 2 basic troops and jump straight to T1.5 for the entire game for a while now. That's why I think shard gain/costs should be reverted to close to the prior patch while attempting to make more attractive ways to build armies at T1.5 and T2 instead.
I don't think you want to have the ability to field a squad of Tier 2 by the first titan, because at that point it's hard to really justify Tier 1.5, but it should be possible to be bringing a small number of Tier 2 dudes if you rush that tech because that opens up real, divergent avenues of play and strategy.
Part of making this a reality would probably be adding a 2nd T1.5 option for each faction, much like Blue has. Then it's probably a matter of making sure there are enough avenues of gameplay at tier 2 to build more than 1 strategy for each faction around it. This might require more T2 units, since some factions are very limted in what they can do here, but I think it would do a better job of encouraging a diverse midgame than forever tweaking shard gains.
Ideally lower tier units should be useful throughout the game like marines or zerglings in StarCraft. I think delaying the access to higher tier units is not the way to solve the issue. Lower tier units should be viable even when high tier units are available. Maybe this can be solved if the amount of bioshard you gain throughout the game is reduced so you have the decision of having an upgraded lower tier army or a high tier army without many upgrades. Right now you can get an army of tier 2-3 units fully upgraded so there is not much downside.
About Apocalytes:
Here is the basic problem i see, they are burst dmg units implemented in a very binary way. You either hit that "skillshit" and kill enough units with it, or you do not. There isn't really interesting decision making/micro going on on either side. As the guy controlling the apocs you wanna do the most dmg possible/kill specific units and as the defending player you wanna move the units out of the circle. That's it. It's not very interesting and fun.
Let's look at another implementation of bursty attacks: The Dredgecrawler
It (in theory) gives you more counterplay options as the defender, you can either run away or split the units to minimize dmg. It's imo the superior "burst attack".
I think when designing ANY spell/unit you want to make sure that there is counterplay possible AND that you as the guy controlling the units/the spell can always get better at it mechanically.
I don't have an implementation idea atm, but the current apocs fail at both these things imo.
In general though i think burst dmg makes game feel volatile because one good use potentially can win/lose you the game, "storm in sc2/bw" is the superior design here because it's not as binary (not the best solution possible either though!)
I think that doesn't work with the current production slot system. If you have a low tier army, you're unable to spend shards upgrading it because all your production slots are army focused.
I am not sure if that wouldn't make burst damage even more powerful than it currently is, because with 3 apocs, you can kill a medium camp almost instantly.
If the resistance debuff was a lingering one though, this might work out well.
I approve of that. It makes it still possible to directly attack the archive in case you have an overwhelming lead plus maybe the titan in order to win the game before the enemy can reproduce his big units, but it rewards gradually increasing your lead by killing structures (and getting shards for them) in order to eventually strike at a weakened archive.
However, I feel like the archive should also be repairable by an engineer. As far as I can remember, this is currently not the case.
Going straight for the archive is a risk vs reward thing. You misjudge the situation and you usually lose most of your army; you make the right call and you get an immediate win. But if you fail, yet take away a huge chunk of HP from the archive, it is a lot easier if you go for it next time, because the archive is already damaged. However, an easier to kill the archive is supposed to be achieved by killing key structures that weaken it's shield, not by bypassing them.
So in order to have 2 distinct options, it is, at least in my opinion, important that one is an all or nothing with the option for the enemy to repair the archive in case you misjudged the situation, while the other one is a gradual advance towards a weakened archive.
I am not completely sold. Someone in here said this spell is binary - you either hit or not. I am not quite of that opinion. I think it more than that because you can his a different amount of units depending on your skill, possible combos and the enemies skill.
I have several suggestions for the Apocalyte:
1) Make the spell have a smaller radius when cast from afar and make the radius increase the closer to the apocalyte you cast it. That would allow the apocalyte to still be used against single stationary targets (conduit, purifier, trebuchet, sludge, rootet brambles). It would also reward good planning and execution of combos and add a risk vs reward to it. If a stun/freeze occurs and the apocs are already in position, they hit a lot more than if they were kept save. However, if the stun/freeze fails/doesn't hit enough, the apocs might be easy pickings. Additionally, one the stun/freeze is over, the apocs are still relatively close so if you misjudge a situation, you can get turned on.
2) Keep the radius (maybe slight tweak?), but change the damage to be max. in the center and less on the corner while the apoc is far away. the closer he gets, the bigger the max. damage center gets up until it does full damag ein the whole circle when it's relatively close. This is similar to suggestion one regarding risk vs reward, but it's a different approach, especially regarding the apocs spell upgrade.
3) Freeze the apoc in place while/after casting for some time. Optionally, this could also depend on the distance the spell was cast at. Longer distance means longer freeze. This change would make the apoc more vulnerable without messing with it's offensive capabilities. Also, if would only make the apoc vulnerable once it already cast and only when you actually use it. It would make for interesting split second choices for the enemy too (do I go for that apoc now, or is that maybe a trap or just too well guarded?)
I like this, but I think levels shouldn't only be worth more shards on kill, heroes should also scale a bit better with levels regarding damage/health.
I think with the titan rescale suggestion, games may be a bit longer and therefore camp will spawn more often. But I am not entirely against this. I take a neutral stance here.
This is actually the most important thing in this thread! Titans do scale too much. I also think they gain a little too much HP when being claimed. A later titan (3rd plus) always has some bigger impact when claimed even if it was really close. I think if should be easier so kill a claimed titan if you did lots of damage to the other one in order to represent that both teams have done an almost evenly good job at killing the titan. The titan should also scale a lot less. Later titans feel way too strong. I also wouldn't be against weakening titans overall damage, beginning at the first titan. The HP are relatively fine for the first titans, it's especially the damage that I think should be lowered.
Totally agree. going for a tech rush should equally as viable as going for lots of T1 units, not a must. Tech balance will probably require a lot more patches, but right now it just feels wrong. Starting with more expensive and longer tier upgrades seems like a good start.
How here we have to statements that I don't think necessarily belong together. Tech rushing is too good, yes, but that doesn't mean we need to stat the power creep. Making teching more expensive and take longer will already make T1 shards units (I don't like the term T1.5 unit as it implies a prerequisite in order to build them) more valuable. I think T1 shards units are relatively well balanced if it wasn't for fast tech being the strictly superior strategy. I suggest not touching T1 shards units in order to make teching worse, at least not for now. I think we should wait what other changes to how teching works bring before we get to this.
I cannot agree here. Critters are so weak, 1-2 T1 units (depending on what T1 unit you have) can kill critters themselves so there is not much risk involved and therefore, there shouldn't be too much rewards I think. I feel like farming critters with a single unit still has an impact big enough to justify doing it and even to stop the enemy from doing it. I like the way it works right now and would rather see it kept that way for the time being.
This is an interesting suggestion and I like the direction it is going. However, I want to add something. I suggestion making hydros ult last for a few seconds similar to how long it lasts now with a initial bigger heal and maybe 2-3 plate stacks when the ult is popped and then a smaller heal over time and 1 plate per tick (whatever a tick will be for this ult) to the units affected. I think that could make for a lot more interesting situations, since hydros could still be focused down to end the effect early, or he can move to get in range of other units and grant them the effect etc.
Engineer repairing the archive is almost certainly out of the question because that would make the unit close to 100% mandatory I think.
Agree here. Having to have 3 engis per team is not a good place to be.
3? Does repairing stack to repaid quicker? if not, one would still be enough, right?
But I don't think it's a problem to have units that are mandatory or at least almost always something you want to pick up, else we'd probably have to rework sentinels.
But it doesn't have to be that way, the archive could simply regenerate HP, maybe based on the amount of key structures still alive, with no HP regen anymore once all key structures are gone and quicker regen speed teh more structures are still alive.
I personally think sentinels are tremendously oppressive and should be removed. I'd put large sums of money on the game being vastly better without them, for this and other reasons. I think Sentinels were an error and I hope that gets recognized.
i'd really like to see how the game plays without sentinels available. I think you might be right and it could make the game better, but I am not entirely sure.
I feel like sentinels were fine before the game turned into a tech time attack. Usually getting the detection ward upgrade was quite adequate against stealth units. I could obviously be wrong about this as I don't have a thought out opinion at the moment.
As with recovering archive. I don't like the suggestion, it is way too complex. The archive used to regenerate HP. Why did it get changed? Because people would invest in trying to destroy it and if they failed, it would just regenerate, and all the investment was for nothing. Not regenerating is a way of rewarding multiple ways of progressing the game. It could be one strong attack, but it could also be multiple smaller attacks.
The game is at the point in development where playing from behind is really horrible. Before, we had respawning Warp Spires and regenerating Archive. This is no longer the case. On the other hand, winning the game from the start feels a lot more rewarding than it has ever been. I don't want us to lose this feeling.
I think the compromise between defender and attacker has to be in the form of the shield. Don't mess with the hp or create some complex building dependent rule, just increase the shield amount. Shield generators is also a potential solution but we had that before and it brought along its own problems.
As players get better sentinels make it almost impossible to have interesting skirmishes around the map and will always encourage big fights in the middle. When you have perfect information (or nearly perfect) there's no reason to ever get caught out of position and i don't see how that ever gets fixed in a world where sents are easily available and effectively required on every team.
I agree with a distinction between HP and Shields being an interesting mechanic for archives. I think the nearby towers providing bonus shields is the best way to codify this probably.
if vision is available it's important to counter that vision by destroying the unit which provides it.
You can see that happen in all kinds of games. Don't see the problem, even though the map might be too small for it to have a net positive impact with the current vision given by a sentinel.
You could reduce the vision, but imo the real problem is the map itself
The issue is that Wards already allow this kind of counterplay. Sentinels on the other hand are so cheap and easy to fly to safety that there is no real way to keep them off the map, and there aren't enough places for armies to hide/move to that you can't keep tabs on everything with just a few sentinels.
They also can trivially guard entries to your own base that have been destroyed, easily diffusing flank attacks and render stealth units nearly useless to boot. The make engineers worse (who cares if my side towers go down if you can't flank me anyway?) and encourage big dumb army balls even more.
I agree that the map is a big part of the problem, but the game mode is really what's making sentinels so oppressive. In a MOBA-style world, vision is absolutely crucial and when a unit provides virtual maphack it's nonnegotiable in every circumstance while dumbing down the game.
Can't that be solved by simply increasing the population limit for sentinel? If it requires 5 or more population, then you probably won't be making more than one and making more will hurt your army size, which can be an interesting strategic choice. You don't need your whole population limit for most popular compositions right now but if that's fixed, then ti can be an interesting choice. Or just cap it at 1.
If you cap it at one, you're just making it so everyone on the team ABSOLUTELY has to take it, rather than maybe being able to justify just 2 people with sentinels. If you up the population cap, you're still forcing at least 1 person to take it (in case they have stealth) because 5 pop and mobility is a smaller price to pay than 100 shards and immobility on wards, but it won't feel good to have to do so.
Or you could just develop the ward minigame and have hero/unit abilities that grant vision actually do something unique and powerful. I don't know why people feel any sort of allegiance to Sentinels - they make the map less interactive, rather than more. Down with sents!
I mean you could say that vision should be harder to get, like with splitting up cheap units and such.
I would actually agree with this probably, even though i also like the sentinel because it takes up a slot and the enemy has to react to it. Maybe the counterplay atm is too hard, that could be the case.
Right now maybe but if more scouting units are added to the game, then sentinel won't be a must.
I don't have time to write a long post, but I like the sentinels, but they are perhaps a little too imba currently. You talk about "ward minigame" but we already have "sentinel minigame". Since they are so "imba", you also want to focus on killing your oppontents sentinels, which is not really easy, and makes for interesting gameplay.
Like in DotA, I think GoA should have some interesting scout/vision component, and the current hero ward system is not interesting in itself.
Killing opposing sentinels does almost nothing if 2+ people on a team have sents (which they should). You already figured out where the enemy is, can just slide another sent over to the location you need to watch, and will have a new sentinel back in place in two minutes. Furthermore being a moba-style map with moba-style objectives you don't really need to know where they are 100% of the time - knowing where they are every 15-30 secs is enough to very accurately predict where they are the rest of the time since there are only so many important objectives in their sphere of influence at once.
Furthermore, it's not exactly easy to pick off sentinels in important places. No one is doing a particularly good job microing them right now because it's only vaguely important to keep them alive since there's so little opportunity cost to them dying. Even then watching camps/backdoors to bases/important passages near the middle is pretty much trivial from relatively safe locations with only a few sents.
I agree with the progression thing, yet I think the normal way of progressing through the game is supposed to be making ones way through the enemies base, killing one building after another until you reach the archive and destroy it. So if there is a path to the archive open, then, in the spirit of game progression, by all means, make HP non-recoverable in that scenario. It's a low risk low reward way of slowly getting closer to the win, the enemy has multiple situations in which they can defend due to this taking more than just one or maybe 2 attacks and you have a nice path to retreat because the buildings in the back are already gone. Thats fair, don't make the archive regen in that case, totally with you there.
But if a team decides to go straight for the archive without destroying building, this is an allin situation. I think it should not be rewarded to fail at this. It's a high reward and needs a high risk to compensate for that, else it can be just done again next time with even less effort because of the archive already being almost dead regarding HP. You make the right call, you win, you make the wrong call, you don't get anything. I think thats fair because theres always the safer route to go for. So for building skipping, I thing if you fail, the archive should have some way to get back up to full HP.
Ok I really gotta ask this: Why this complex mechanics aversion? I have seen this somewhere else in the forums as well, though I don't remember if it was you or someone else there.
I get that complex mechanics are harder to learn, but they also make people more invested in a game because they can obtain vast knowledge about how things work.
I think complexity is totally fine if the learning curve doesn't get steep until there it a basic understanding of the game. "The Archive has to be killed to win the game. Killing this and this structure makes killing the archive easier" would be all that people need to know when they learn the game, they wouldn't go straight for the archive anyway in their first games. The other stuff can be added as small tipps while loading the game or something, but it's nothing one has to know in order to learn the game, so I don't see any problem there.
I personally like complex mechanics as they often make is easier to tweak certain aspects without influencing other aspects too much.
For the Sentinel thing:
Maybe making the sentinel a ground unit with flying vision would be interesting. It could'n escape as easy, yet it would still see up cliffs, and not only ranged units could do something against it.
Edit: If you want a less complex solution, why not make the archive have 1 HP and only use the shield?
It was probably me who had complexity aversion. It's just part of the game design that it should be easy to get into. I will admit that Complexity has its benefits. I also take great joy in breaching into complex games. But it is a double-edged sword. For some kinds of gamers it is awesome, for others, it becomes tedious. I don't want new players to have to read a 5-page manual to understand what is going on in the game. That's generally been the spirit of the game's design, and I think it has reaped a lot of benefits from this approach.
Your edit hits the nail on the head. My point isn't that solutions are limited to shield-hp. I am okay with Shield Generators for example, but as I said, we had this before and it brought with itself its own issues (I think the issue was that there was always a clear choice of what to kill, but this was solely a result of numbercrunch and not intuitive). The point is that if there is a simpler solution that achieves the same thing, it is most likely more preferred. With my complexity aversion comment I was mostly targetting the suggestion of variable HP regen based on number of alive structures.
I understand your position with regards to not being able to end the game quickly, but I think it's a very delicate problem. Games where ending the game is difficult can very easily turn into a situation where the losing team becomes a zombie for 10 minutes and are just waiting for the game to end. Hard to end games also benefit teams that are teching hard.
Personally I really liked that I could end the game by 10 minutes. I don't like that big melee rush was the only way of doing it though. I think there are a lot of people on both sides of the argument and it is not clear yet which direction is the best way to go.