Rework of harvest boost



  • Harvest boost seems to be flawed due to the fact that harvesting 4x faster is minimally effective when a source can easily be harvested by 5 regular WORK parts currently. Theoretically, it does allow for intents cost to be saved, but the boost still doesn't seem worth it.

    To fix this, harvest boost should have it's effect changed to "Increase energy gained from harvesting by 50/70/100%. Does not affect amount lost from the source." This change means that harvest boosts can be used to double the energy gains from a source in exchange for using minerals, which could shake up economy tactics.



  • Doubling available energy income would certainly shake up economy tactics...



  • It could completely change the game. I'm trying to figure out what I think about it. I think once power comes out it's clear that energy will be the limiter for many players as I think getting 50 energy/tick/room is harder than getting 1 power/tick/room. Maybe that means that energy could be made more readily available?

    IMO it'd certainly make at least T1 harvest boosts very common. I could spend 150 T1 boosts and 100 energy with a payoff approaching 5 energy/tick. That's basically 1.5x energy income at very low cost. On the other hand maybe CPU is a limiter: it becomes much harder to oversize harvesters as you'd be paying in either boosts or efficiency. Right now I can just have 10W harvesters and double the number of sources I harvest. I'm pretty sure T2/T3 harvest boosts wouldn't be good enough. It just takes too much lab time.

    So I think the impact of this would be to bring harvesting closer to owned rooms and reduce the amount of area you need to have a good energy income. I don't think that's a good thing tbh. Whether it'd actually be much of an efficiency improvement would take a bit more thinking about.

    There'd be a huge jump in mineral income. In this case intents don't matter so much so T1 is a no-brainer. T3 is less clear cut, especially if the reaction time went up.

    If a change similar to this were to be made I think it should be balanced around the trade-off between lab time and remoting footprint. Boosts should be objectively worse than not boosting given plentiful sources, however if you only have a limited number of sources available to you boosts should be better. I'd have to do more sums, but my gut feeling is that 50/70/100 would be too powerful and a lower percentage would be better. Alternatively, it could be balanced with high reaction times.



  • I love the idea! Makes the harvesting boost way more interesting from a game design perspective.

    Although I think it should start with some smaller percentages, like 25/50/75% or 33/66/100% increase in yield and if that proves too low get buffed. 🙂



  • Some quick maths. Firstly, minerals including the cost to create the boosts and assuming all minerals are equal:

    • T3 would give 1.4x income
    • T2 would give 1.24x income
    • T1 would give 1.26x income

    Seems reasonable given the lab time required for T3. T2 sucks, but T2 sucks at everything so that's no surprise. Maybe T1 is too powerful here.

    Now sources. This seems a lot more complex to me. First off, I'm going to discout T2 and T3. I just don't think they're worth the mineral or lab time, and sustained T3 across all of a room's harvesters just isn't going to be possible. Maybe after power creeps, but not today. The complexity here comes from balancing of energy costs, mineral costs and CPU costs. Picking a few points of comparison:

    1. 2x5W T1 boosted vs 3x10W unboosted. Matched on income. Costs are close, IMO boosted is slightly cheaper (1500 energy + 300 T1 boosts vs 3750 energy), but boosted use 1.5x more CPU.
    2. 2x10W T1 boosted vs 3x10W unboosted. Matched on income. Boosted are now more expensive (2900 energy + 600 T1 boosts vs 3750 energy). Unboosted is 1.5x the CPU.
    3. 2x10W T1 boosted vs 2x10W unboosted. Matched on CPU, but boosted is clearly more expensive. The income more thank makes up for this.

    It's not super clear-cut to me. I feel that boosted is probably a bit better than unboosted and can work fewer sources for the same gain. I think that boosted creeps also become more powerful when there are two sources close to each other.

    So in both cases it seems T1 is just a little too good. I'd consider 35/70/100 as percentages instead. I think that'd make it situational and potentially powerful as in-room harvesters and harvesters when you could multi-source, but not quite so clear cut for isolated remotes. Minerals would still turn a profit with T1 boosts, but the trade-off against lab time would actually be closer and the profit wouldn't be huge.

    EDIT: Following further discussion on slack, it seems like such boosts should be expensive in terms of lab time: maybe as expensive as dismantle.



  • This is definitely an interesting proposal.

    There's also precedent with the upgrade and build boosts which both effectively create energy out of nothing.

    I'd also mirror @Jacudibu and suggest that it should start with lower impact and be increased if it proves too low.

    My personal opinion on boosts and economy is that boosts should be useful only in "niche" circumstances and not an omnipresent aspect. It should not be worth it to boost everything all the time. It should require your AI (or yourself) to decide in which cases it is warranted.

    The extra energy from the harvest boost would come in handy when you're building up a new room and can benefit from extra local energy. I can see quite a number of tactics coming in handy from a harvest boost that increases the effective local energy.

    Excellent proposal.



  • Great suggestion! I'm all in favor of it, I was just thinking about it a few days ago @davaned said in PTR Changelog 2018-08-28:

    Remember when the carry boost also made you require more move parts to get around? I'm all in favor of making logical changes to get it in a more balanced state. Honestly I think harvest would be another good one to look at, maybe it should make you mildly more efficient as well as faster.

    It might not be bad to keep it faster AND make it more efficient, considering we are going to have ways to increase the raw source amounts. Something like mining twice as fast and 25% increase in returns.

    You could also split the mineral/energy effect too. Getting bonus minerals might be a bit too strong comparatively, getting 100% increased return on mineral seems too good.



  • @davaned said in Rework of harvest boost:

    You could also split the mineral/energy effect too. Getting bonus minerals might be a bit too strong comparatively, getting 100% increased return on mineral seems too good.

    You don't get 100% increased return as you spend minerals to get minerals. Assuming it's +100% and speed isn't changed you only get a +40% increased return on minerals, and obviously you pay lab time and energy for that.


  • Dev Team

    The idea is definitely interesting, but it looks too dangerous for those who use harvest boosts currently and rely on the speed boost provided by UO/UHO2/XUHO2.

    My experience tells me that these boosts usages are primarily situational, but still, we need to investigate how actively they're currently using... if nobody uses them, we want to be absolutely sure about that.


  • Dev Team

    I've collected some statistics about using boosts and we can clearly see that there are people using it. That means the proposed change would break their code. So in order to proceed, we should find these players and kill them invite them into this discussion because, obviously, these will be the most affected players by any harvest boost change.



  • I use T1 harvest boosts for extracting minerals, just to speed up depletion a little (the sooner the regeneration cycle starts the better). So if the proposed change is just about energy it wouldn't effect me at all.



  • I use UO and XUHO2 quiet extensively and it has it's place in this game already. You should use it when you want to reduce work parts for harvesting (obviously) and when you do, it has good returns for what it does.

    I think this change of increase in minerals/energy to be too good.



  • One of the overlords used to be very fond of harvest boosts.

    I used to use the Harvest boost ALL THE TIME for harvesting minerals, but I stopped using it in favor of being able to blast full production in the labs 24/7. It's indispensable when you only have a few level 6 and 7 rooms, but once you have enough level 8 rooms, it's way less useful.
    The other useful T1 boost is the build / repair boost. 50% increase in wall building power for a T1 isn't bad. The problem is when you're using a ton of them 24/7, it gets expensive. Boosts are just too stupid expensive to use. You need freakin' 30 of them PER part. using 2 repair creeps instead of one is far cheaper, and adds 100%.
    The core problem with boosts is their usefulness is completely dependent on your inability to make creeps. Once you can make enough creeps that it doesn't matter, you can "just make another creep" to get that 100% boost.



  • So I've not really considered harvest boots until this thread, but having looked into it they turn out to be the most efficient way of saving parts of all the economy boosts. With UO which is really quick to product, you can build harvesters that have 1M1C2W instead of something like 3M1C6W, which is a 4-6 part saving on each harvester from merely 2 boosted parts.

    So if you're exploiting 3 2-source rooms at RCL6, you can get maybe 24-36 extra parts. Which is maybe half a transporter... so using these boost doesn't guarantee that you can open up another remote room, but maybe sometimes the numbers work out. I will have to look into these boosts more closely.



  • I don't yet use harvest boosts, but I have several of the aforementioned uses working their way to the top of my "to-do" list. My brother and I used to laugh about the uselessness of harvest boosts, but they've been slowly growing on me. I now think of them as one of the most interesting economy boosts. I've had several pleasant nights of not-sleep-brainstorming, trying to think up new ways to use them.

    Not saying the suggestion of giving extra resources is a horrible idea, but I do think they are quite interesting as-is, and have some untapped potential.



  • This post is deleted!


  • I used the T1 Harvest and Repair boosts all the time (I'm the Overlord deft-code was referring to.) But I stopped once I started getting serious about manufacturing boosts through gang manufacturing.

    So, the last patch, lab.unboostCreep(creep); was introduced so I thought I'd give it a try with a repair (rampart/wall builder) creep as I'd just put in a boost on demand system that goes right back to production afterwards. As such, I no longer needed to dedicate a lab to just the repair boost.

    unboostCreep will have huge benefit for this. There is a slight overhead at the beginning of the creep's life (Have to boost it.) and at the end (Have to get it to the lab to unboost it.) and there is loss as it drops the boosts on the ground instead of the nice safe lab (I even made sure to have the same boost in the lab, just to see if it would put it there instead.) But that's not a huge deal, as the creep could just pick it up after it fell off.

    So... we're talking 5% overhead with the boosting / unboosting, some average loss in collecting the dropped boosts (recycle creep might be spawning / picking something else up. So leave that at 20%.

    Even then, you're getting a 35% or so repair increase. Not earth shattering, but the T1 is pretty cheap... sure, 2 creeps would be "more", but this saves energy and CPU over the long term.

    The cooldown. As advertised, the cooldown is the aggregate reaction cooldown for the entire batch recovered. With T1, it's not so bad... My test creep (8 WORK parts with the LH boost.) invoked a 360 tick cooldown. A T3 50 body creep would not even be practical with this.

    The cooldown had no effect on production, as the lab was a reaction output. So it was able to go right back to work after unboosting. It just needed to have the cooldown be zero when the next creep needed to be unboosted. So long as the cooldown is under 1400 or so ticks (31 or fewer boosted parts for LH, et al) it can be worked continuously.

    Will have to throw more math at for long term viability... Math.throw(longTermT1Use()); but I think I like this unboost change. Could breath life into T1.



  • Oh heck. What was I thinking? It's the INPUT labs that don't get the cooldown, having the output lab get hit with that essentially prevents production from occurring in that lab for the cooldown duration. So that's essentially a deal killer.

    And I hadn't even gotten to the part where I figure out the actual mineral cost... which is probably also a deal breaker.

    All in all, what is the point of this feature? Boosts are still too expensive to use for anything but the one place you HAVE to use them, combat.