PTR Changelog 2018-09-12: unboost


  • Dev Team

    This post describes changes on the Public Test Realm.

    Commits on GitHub:

    Changes:

    • Introduced new method StructureLab.unboostCreep. It removes all boosts from a creep and drops 50% of the mineral resources spent on boosting. The difference between this method and StructureSpawn.recycleCreep is that unboosting always returns 15 units of each boosted body part's mineral regardless the creep's remaining time to live. You can use it to take boosts off an old creep and reapply 50% of them to a new creep.

    • StructureSpawn.renewCreep now shows a message in console if you try to call it on a creep with boosts. In future, it will throw an error. Use StructureLab.unboostCreep to remove boosts before you renew a creep.

    This feature is supported in private server ptr branch:

    npm install screeps@ptr
    
    👍


  • So I guess the impact of this is:

    • Repair/harvest/upgrade boosts are effectively half mineral/lab time cost. Pretty rare you'll not get them home.
    • Carry boosts for high TTL uses are half mineral/lab time costs. One-way carry or carry for power/convoys won't really benefit as you would recylce these anyway
    • Short range boosted attacks get a big buff and are nearly half mineral/lab time cost. Medium may get a bit, long range is unchanged.

    Is the large buff to upgrade/repair boosts intentional? They're already very strong boosts and were previously nerfed a bit by lab reaction time. Now they've been buffed to be significantly better than they were before the reaciton time nerf (IMO).

    ☝

  • SUN

    Nice @artch. Lowering the cost for boosting, increasing flexibility, and adding room for future optimizations. My one complaint is that a boost shouldn't break the renewCreep, it would mean you can't renew anything boosted in a lower level room. Might be nice to be able to force a renew.



  • I agree with @davaned, it shouldn't break renewCreep completely.



  • This post is deleted!


  • Why throw an error instead of a new error code?



  • @davaned Hi mate good to see you are still keeping up with the times re screeps. My understanding of the current mechanics is that if you renew a boosted creep, it will lose all it's boosts. They just disappear. But yeah, it shouldn't break it (I don't actually use renewCreep anymore)

    👋


  • @davaned OK, I mis-understood the point you were making, it's totally valid.



  • What is the goal of this change? The biggest impact IMO will be to reduce the steepness of the GCL levelling curve and increase the strength of ramparts.

    Now that power creeps are nearly out, with a daunting cliff face of power to climb, it's not unreasonable to make GCL a little easier to get, so that newer guys can catch up with older vets.

    I'm not sure about the effect this change will have on ramparts though - I predict it will make bunkers overall stronger and harder to crack. If this is a goal of the change then that's fine, but if it's not an intended impact then I'm a little worried about the change.

    I suspect this change is intended to make it more viable to use base economy boosts such as harvest/carry boosts. I think it would do this a little, but the impact on these would be overshadowed by the impact on GCL/ramparts boosts. That is, I think if right now a lot of people think it is sensible to use boosts for ramparts and GCL, while not many people think it's sensible to use harvest/carry boosts for their economy, then I think this calculation would still largely hold true even if GCL/repair/heal/carry boosts all become effectively cheaper.

    Attacking people further away becomes relatively more expensive, while defences become stronger, so there will effectively be a shrinking pool of targets to attack. I wonder if the net effect of this change would be tougher bases and perhaps less combat overall, due to the smaller target pool.

    But perhaps the intention of this change is different than I am inferring, and these effects are intended?

    ☝

  • Dev Team

    @wtfrank The main purpose is to introduce a new mechanic and the corresponding coding challenge that allows to increase effectiveness of boosts in general for players who managed to implement this. We're still debating on the effect to GCL and repair boosts. The military aspect is something we have to consider too, thanks for pointing this out.

    👍

  • Overlords

    I can't support anything that removes boost sinks from the game. Yes there's still a 50% loss but that's still not enough.

    The biggest issue this game (MMO wise) has right now is the difference in power from a new player to even a semi decent one month old player. To allow that older player to get returns on his boosts just widens that gap. (not even considering the older players as the gap between them and a newbro is so horribly broken at this point that this change is merely a drop in the bucket)


  • Dev Team

    @shibdib I never take this argument about "old / new players" as valid. It's a MMO after all. It means players have constant persistent progress, and the game mechanics should make it possible to have progress for years. If we try to equalize new and veteran players, then we eliminate incentives to grow and apply efforts in this game world.

    We're going to launch a whole new game concept (Screeps Arena) that will be built around session matches and personal PVP rating, and game mechanics will be equal for any player regardless of his time played, but in this MMO game the time played should be directly (not exactly linearly, but still directly) converted to in-game power projection, otherwise we'll start losng veteran players.

    👍

  • Dev Team

    Update: unboosting now requires cooldown time equal to the total sum of the reactions needed to produce all the compounds being returned.

    It makes this mechanic more difficult to code properly (different creeps must go to different labs, probably even in different rooms) which is a good thing for such a powerful feature.


  • Dev Team

    This change is deployed on npm in ptr branch starting from 3.1.0-beta.5.



  • @artch said in PTR Changelog 2018-09-12: unboost:

    It's a MMO after all. It means players have constant persistent progress, and the game mechanics should make it possible to have progress for years. If we try to equalize new and veteran players, then we eliminate incentives to grow and apply efforts in this game world.

    I respectfully disagree. There are many games that make the gap between players who played the game for years and those who are pretty new to it very small, and the games are no worse for it. EVE Online is one of those examples, and probably the best. The key is making the progression less linear and more parallel/alternative. Older players just get the benefit of being better at more places, while a newbro specializing in one area can get good in it rather quickly.

    On the side note, I am not sure Screeps would work like that easily. But there surely could be things done to ease on the gap w/o making old players bored.


  • Dev Team

    @orlet It is good that you mentioned EVE, because it is a perfect example of what I'm talking about. Yes, it's not linear (same as GCL and PL in Screeps), but you keep progressing even if you are a vet. I played EVE for two years, and with every month my character got stronger even in one single field (perfecting all essential skills to L5 is not easy). A small gap between a new player and an old player, in EVE, really? What a rookie ship could do vs a Battlecruiser/Battleship? Or a T1 BC/BS vs a T2 BC/BS?



  • @artch said in PTR Changelog 2018-09-12: unboost:

    Update: unboosting now requires cooldown time equal to the total sum of the reactions needed to produce all the compounds being returned.

    It makes this mechanic more difficult to code properly (different creeps must go to different labs, probably even in different rooms) which is a good thing for such a powerful feature.

    Assuming it is done efficiently, I guess this change means that whereas before we'd save lab time and minerals by unboosting, now we just save minerals. I think that's an interesting change.

    • For people who aren't reaction time limited, it's no change to the original unboost. Just harder to implement.
    • For people who are reaction time limited sending far away creeps home to unboost is much less worthwhile: sending them home means you save minerals but costs you lab time (assuming the same total time on target is covered by multiple waves).
    • Furthermore, if people are currently reaction time limited they're not going to majorly start using new types of boosts as unboost doesn't reduce the critical path. They'll unboost current economic boosts, and maybe those extra minerals can go via the market to free up some lab time, but it's not a clear shift.

    For me this new change makes unboost pretty useless. I'll use it on my repair creeps, but that's it. I may be a special case though given I spend more than the average player on attacks.



  • @Pundemonium also pointed out on slack that for the extreme case (T3 boosted 40:10 dismantler) it's actually not unreasonable to destroy then rebuild the lab to work around the 30k+ tick cooldown if you're lab cooldown limited.



  • This feature should help me be more efficient with my labs. Currently my two central labs are never in cooldown. It just doesn't see worth the code to shift the compounds around to keep all 10 labs busy.

    However, now I can use the two central labs for unboosting since they're never in cooldown otherwise. In theory that should net me 20% created boosts.

    👍


  • @artch said in PTR Changelog 2018-09-12: unboost:

    A small gap between a new player and an old player, in EVE, really?

    Yes, but for EVE "old" player is 5-8 years. You can essentially master a ship class in a year or two for initial + core skills, then less for each subsequent ship, unless you're going capital/supercapital. Once you have all skills for specific ship to 5, there is absolutely no matter how much SP you have invested in other ships, and two players, say a 2y/o vs 10 y/o would be perfectly equal and down to players actual skill rather than how much sp they have. And the gap is even less so once industry parts are concerned, except that old player will likely have accummulated a vast collection of their own blueprints, while new one would have to rely on purchasable pre-researched copies.

    Or a T1 BC/BS vs a T2 BC/BS?

    There are perfectly valid situations for where a T1 BC could outmatch a T2 one, and same goes for T1 BS vs T2 one. Fitting, for one, matters. I've taken down T2 faction fit frigates with a T1 one with T2 fit.

    I could go on and on on this, but don't want to derail the thread any more than we already have.