Posts made by wtfrank

I would certainly find the game less fiddly if container sizes were increased. There are all sorts of problems I have to solve because the container size doesn't map well to the size of cpu-efficient tranport creeps, and it means I have quite strict timing requirements to collect energy from containers. These are difficult and interesting albeit fiddly problems to solve. If we got made containers much bigger, it would remove these difficulties and make harvesting gathering an easier task.

It would open up some new solutions to energy retrieval such as boosted transporters or even trains, but it would generally simplify the problem of retrieving harvested minerals.

Would the game be better if this happened? I'm not sure.

I'm don't see how this in any significant way changes risk / reward. The biggest change is to how you size and programme your miners.

FWIW I've always used the https://screeps.com/a/#!/map page as my "portal". I rarely look at the front page.

1. view a highway portal room on shard2
2. click a portal to shard1 and view the linked room.
3. mouse-over a room on the minimap (minimap is correctly rendering shard1 rooms)
4. observe that the link to the room on the minimap goes to a room on shard2, not shard1.

This screenshot shows in the top right that we're viewing a room on shard1, the bottom left room in the minimap is moused-over, and you can you see in the bottom left of the screenshot that a shard2 room is linked, not a shard1 room, even though the minimap is rendering the correct shard1 rooms.

@deft-code said in Market for CPU Bucket:

Highly efficient code bases have some of cool-stuff tricks. Since they consume less cores they have a great cool-stuff/core ratio

I think ultimately every bot caps out at 300 CPU, any improvements in efficiency in the bot work out to more or less server load based merely on how effectively the 0.2 CPU intent cost maps onto underlying server costs.

I would pay for bucket when i've got some crisis going on. This is an interesting idea and I support it.

It would be useful if the console had a mode so that it would only show the console output that was generated on the same shard as the room that you're currently viewing.

(Also it would be useful if the console could be optionally present in the map view)

I have had unboosting built into my bot all year. A recent development is a container next to labs to catch the boosts so that I don't have to find a porter creep to get there ASAP after the unboost and pick up the boosts as they tick down. I expect to be typically unboosting 1 or 2 creeps per lifetime so the energy cost of the container is justified if it saves a few 10s of boosts per creep lifetime.

To me, unboosting is a way of getting hold of complicated boosts for "free"-ish. If you only produce into 8 labs then you have 2 labs you can unboost into at no cost. Even if those two labs are on cooldown (which sometimes happens for my bot), paying the lab cooldown cost of building the boosts without having to pay for the minerals seems like a good swap to me.

There may be people with perfect mineral set-up, but I find that there are some minerals that I obtain more than my empire can use, and there are some minerals that my empire is to an extent gated by. Particularly if the boost contains one of these gated minerals, unboosting seems like a massive win for me - it either leads to me producing more of the bottleneck boosts, or it leads to me obtaining more credits.

I haven't run into the issue with a level 7 room that you've experienced, but at RCL8 where you're only running 1 upgrader I have close to no issue with the cooldown time, even using T3 boosts for upgraders.

@smokeman said in Rant on mineral costs of boosts.:

No, I wasn't even going there. But I could if you want, but for completely different reasons.

OK well the reason I was asked was because of this:

Why should a +100% carry boost cost the exact same amount of Keanium as a +300% carry boost? How does that provide an incentive to learn about boosts and write boost code?

So I don't understand what the issue is with the cost in Keanium. To get the +300% carry boost you have to add 3 more minerals and perform 3 more compounding steps, with much longer cooldowns (since the change a year ago). So I don't understand the issue you're raising as it's much easier, faster and less complicated to make the T1 boost than it is to make the T2 or T3 boosts. I don't think it's a useful way to measure the boost in "number of Keaniums per tier".

Is the issue you're saying that you feel like the return on the boost at tier 1 is nothing?

Are you complaining about how there is a "market failure" within the T1/T2 boosts, where there are few listed orders, consequently the prices aren't very competitive?

Also you're considering DISRUPT source where you find 5 sources to disrupt, on cooldown, and you're claiming that disrupt source would drain 5 ops/tick.

It would certainly be possible to use it like that and spend it's whole time moving to the next source to hit the disrupt cooldown, but in what I consider a realistic usage scenario, in the next column, you're draining 2 ops/tick and keep 2 sources in a base room on lockdown.

So the realistic comparison is 1 or 2 ops/tick vs 1.25 ops/tick for disrupt terminal. It's in the same ballpark. It could be a little cheaper maybe, or disrupt terminal could be a bit more expensive, but asking for for disrupt source to be reduced by 80% of the cost would put the ability ridiculously far out of whack on a cost/benefit ratio.

If you're building a bunker, one fairly obvious way to site the bunker, room permitting, is so that the bunker is adjacent to the controller or a source, or if you're very lucky the controller and both sources. So that you get max tower coverage protecting those important sites, and so those sites are less harrassable. You wouldn't necessarily have the source inside the bunker, because of how constrained you are for space, but you would have it touching the edge of the bunker ramparts. It's quite possible that the source could be perfectly safely harvested for the entire time that the bunker holds.

@davaned said in Disrupt_Source cost is extreme for its value:

Now look at disrupt source. What does it offer? If you're in range 3 of the source, you're almost certainly in range of the creep mining it. Killing that creep does the same effect in a more permanent fashion. Disrupt source isn't letting you project power in a room assault situation, and it's impact is negligible at best unless combined with a disrupt terminal.

If you have a base layout that encloses the sources, disrupt terminal isn't necessarily very effective at shutting the base down as the internal sources can still feed 15,000 or 30,000 energy into the room per attack squad lifetime. If you can reach the sources with disrupt source then it could be the difference between this base standing and falling.

This is the spreadsheet I made ages ago. https://docs.google.com/spreadsheets/d/1TXuMGjLIEJnFyO13UhElMKqaG6OsBvMIjq1nQRDmOwU/edit?usp=sharing

Disrupt source is massively cheaper than disrupt tower, in ops/tick. Even the cheapest offensive power consumes ops 6x as fast as a creep can generate them. None of the offensive powers are self-sustainable. A 5000 range self-sustaining offensive creep might be rather a menace to creep society.

I don't see how disrupt source shuts you off from using the rest of a maxed power creeps' abilities. Could you explain that a bit better?

When I looked at the cost of the offensive abilities pre-launch, all of them had a higher op drain than the maximum a single creep can create, while all or most of the defensive abilities had a lower op drain. Consider the screeps attack/defence balance, and the balance around minerals. Defence is probably a bit easier than attacking. And you can't mine enough minerals to do everything you might want to do with minerals, you have to make tradeoffs.

This design of power creep offensive abilities doesn't seem out of whack with the rest of screeps. You can burst offensive ops, but not run them flat out.

You shouldn't compare it to "other abilities", you should compare it to "other offensive abilities". How does it compare to the other offensive abilities that you may pick instead of this one?

I just attempted to activate safe mode (on the s+ server), a few hundred ticks after my room had come out of the initial spawn safemode. There was no safemode cooldown, and no significant downgrade timer. The manual safemode attempts did not succeed. However a safemode generated by my bot shortly afterwards did succeed.

Each tick, a creep's memory is serialised to JSON, stored in the server's database, and then at the start of the next tick it is loaded back from the database and an object is created.

I'm not sure what the technical term is, but JSON can only serialise "basic objects". So while it can serialise basic data members (number, string, array, object) it can't serialise functions. This is why you get the error "myObj.run" is not a function.

Another example, if you store a RoomPosition object in memory, what you get back the following tick is an object with a "roomName", "x", and "y" members, but the object is not "instanceof RoomPosition".

If you want to use a class with the creep's memory, you would have to assign the class to each creep every tick. Or find a different way of organising the creep's memory.

console.log("Result of isRoomAvailable:", Game.map.isRoomAvailable("E52N1"))

[09:29:56][shard3]Result of isRoomAvailable: true

FWIW I extend the type via Declaration merging, and it at least prevents me extending a the prototype without having declared it up front.

Although it's "fair" to return unused fees, it discourages people pricing their orders "to sell" - so the spread will be wider, and less transactions will take place, as there's no penalty for having a sell order at such a high price that it never sells, or a buy order at such a low price that no-one sells into it.