PTR Changelog 2018-02-28: lab reaction time



  • IMO, this solution is too complex for the issue. A simpler solution would be the following: 1: OH boost takes 20 ticks to create (was 10 ticks, 40-70 total ticks to create T2 boosts) 2: X boost reactions with a T2 mineral take 50 ticks to create (was 10 ticks, 90-120 total ticks to create T3 boosts)

    This makes the changes simple and easy to understand for AIs, while still adding a reason not to use catalyzed boosts.



  • @tigga Fingers crossed they heal the right creep though...



  • @shibdib shard1 and shard2 were only added a few months ago, there aren't going to be huge stockpiles throughout the game. And these stockpiles will deplete over time.

    T3 boosts certainly have got harder, but that's only because the cost/benefit was out of whack before. Some of the T1 boosts have got a lot easier to build - e.g. T1 Tough boost. That's a big benefit to new players and may have very little impact on the boost consumption of the older players.



  • This post is deleted! 🤟🤦

  • Dev Team

    All boosts have different bonus progression, some are linear, some are not, this is why they have different reaction time progression as well. It has nothing to do with their popularity, and it explains why it's a bad idea to make all tier times uniform. Now each "total reaction chain time" strictly correlates to the bonus value within each effect type series. We got a solid spreadsheet proving that, these numbers are not random 😉



  • I think better solution will be in increasing number of ingredients to produce T3 boosts. Like a 5*T2 + 5X = T3, or something like that. Not one-to-one like now. It will raise usability of T2 boosts.



  • Dismantle is a massive outlier for not that big of a gain. Dismantle creeps already suffer from a number of downsides compared to attack creeps (namely, inability to fight creeps and dropping half of the energy used to build the walls on the ground to be picked up later to rebuild the walls), and they're higher damage output in an optimal situation isn't a reason to implement an incredibly long boost time specifically for them while attack, ranged attack, and heal are given the standard 95 tick boost times. Just because it "can only be used for attacks" doesn't mean that it should have 4x the cost of a similar mineral that can theoretically be used in other roles.



  • @fetid I think artch's view is that this would break existing code more than changing the lab job time would https://screeps.com/forum/topic/2038/introduce-boosts-one-teir-at-a-time-rather-than-all-at-rcl6/11


  • Dev Team

    @crusher48

    Just because it "can only be used for attacks" doesn't mean that it should have 4x the cost of a similar mineral that can theoretically be used in other roles.

    True. And this is why the cost is the same. It is time that differs, not cost.



  • I mean time is a cost here, because it appears that the new bottleneck with high-level boosts will be lab time, not minerals.



  • @artch While I'm not entirely in @Crusher48's camp on this, time is a form of cost itself.

    👆


  • @dissi said in PTR Changelog 2018-02-28: lab reaction time:

    attacker runs out of boosts

    Should read as "runs out of credits". So, again game seems to be a bit more versus new players.


  • Dev Team

    I believe increased time won't contribute to the market price to the same extent as the mineral cost does.



  • This post is deleted! ☝


  • I've been frying my brain trying to figure out a good way to implement labs/reactions/boosts/etc for a while now and that was BEFORE this change. Seems like its taking something already complicated and confusing and making it even more so. I haven't actually gotten code for reactions written yet but it seems like while this won't make the actual reaction logic any different, it will make the decision of which minerals to focus on even more complicated than it already is.


  • Dev Team

    Update:

    ZH (dismantle) series reaction time set to 20/40/160 ticks.

    New column "Time/benefit ratio" added to the table above. It shows diminishing returns on each tier, indicating that the time/benefit progression is not linear, but is in favor of lower tiers.

    The release date of this patch is set to March 5 along with other PTR patches.



  • @artch That seems pretty quick for the reaction time change, I am not even sure, that everyone affected by it has even read about it yet. The other 3 changes are all additions/fixes (and great! I was eagerly waiting for them). But reaction time change might force people to change their code.

    👍

  • Dev Team

    @mmmd This change isn't breaking. Your old code will continue to work, so changes are not mandatory. A lot of people won't even notice this change (in fact, only a small minority of players look at PTR patches at all), and the rest can start gradually optimizing their code while keeping old code functional.

    I can see why it may be inconvenient for someone since some parts of economy may become suboptimal, but it's not critical, and economy things are rarely optimal, you know 😉

    We really delayed this PTR release for too long, we can't postpone it any longer. As for splitting reaction time into another patch, small frequent releases are bad for marketing, so we'd like not to do that.



  • I still believe that dismantle is already a weak option, and having it have the highest reaction time in the game certainly won't help.



  • It will certainly affect me (I assemble T3 from its base minerals in one pass, except for the ones using G)

    I'm not sure this will help much to stimulate fights. It mostly encourages stockpiling and increases the power of large coalitions. (a small nimble group of players will run out of boosts faster).

    I would have preferred having unbalanced mineral respawn rates (perhaps by shard or general area, to encourage trading and inter-shard commerce) or have to some recipes use more then one mineral (ex: 4X + UH2O -> XUH2O), again perhaps with some random element over time to avoid the market balances out.

    If it's to reduce the DB load, which I suspect, I'd prefer a change that increases both reaction time and minerals used at the same rate (ex: 10 X + 10 UH2O -> 10 XUH2O with a 50 tick cooldown)