More levels of robot speed research

[immibis]

The top-end robot speed isn't insanely fast yet, and the robots (particularly construction robots) can take a while to travel across a large base. So why not add a few more robot speed upgrades?

[sillyfly]

Sounds reasonable.
While we're at it - why not add research to make bots' battery last longer (so they don't have to recharge so often on long trips), and one that make the roboports charge them faster?

Anyway - I think (do I am not sure) all of these can be very easily achieved through mods.

[sillyfly]

Your wish is my challenge
I made a mod that adds 5 more techs to speed up bots (can be easily changed by a variable in the data.lua file), each tech costs double the previous one, and has the same effect as the last in-game tech (i.e. - gives all bots +65% speed boost).

For the other things I have suggested - it seems it is not very simple to do, so I haven't.

Enjoy

[Garm]

Frankly i think robot speed is sufficient - when you limit the controlled area and don't cover the entire base with one.

Adding more research sounds to me like throwing resources to fix logistical challenge. whats the point of developing complex bases if it could be avoided by paying some extra resources?


As a mod, however, its much more plausible: everyone can choose their poison.

[keyboardhack]

Adding more research sounds to me like throwing resources to fix logistical challenge. whats the point of developing complex bases if it could be avoided by paying some extra resources?

We are already doing it with : Shields, Belts, inserters, Furnaces, Electric poles, Assemblers, Guns, Research and so on. We already throw resources at our problems to make it easier and less of a hazzle to deal with the logistical challenge that it would cause if we didn't have those upgrades.
The point is that if it's easier to develop larger bases, then we will be able to make larger and a lot more complex bases without having to deal with the small problems that we should be able to deal with, like robots speed.

[Nova]

A Robot MK 2 would be much nicer in my opinion. It could cost one normal robot, one speed module 1 and one processing unit. (Just as an example.)

There's already a mod out there which adds this new robots and even a roboport MK 2, but I didn't like the costs and values of the things, so I did change them for myself. The robots are now "only" 2 times as fast and cost more energy per meter and per second, not less like in the original version of the mod.

All things are insane. The combat shotgun, the destroyer capsules, the modules at the end... All this makes tactics not needed. Just use a bunch and you are fine. Either we change that or we do it for all things, robots too.

[Garm]

I would prefer the change however. Just because something isn't perfect doesn't make it ok to make it worse.

Edit: ive been building large vanilla bases for year and a half. Robot speed was never an issue (charging and zones were)

We are already doing it with : Shields, Belts, inserters, Furnaces, Electric poles, Assemblers, Guns, Research and so on. We already throw resources at our problems to make it easier and less of a hazzle to deal with the logistical challenge that it would cause if we didn't have those upgrades.
The point is that if it's easier to develop larger bases, then we will be able to make larger and a lot more complex bases without having to deal with the small problems that we should be able to deal with, like robots speed.

I beg to differ.

belts and inserters have a lot more than just more expensive = better, My belt lines always use a mix of all three for better compression, speed modification, and throughput fine control

Furnaces serve two specific functions (with the exception of cheapest ones but they are used for boilers) one doesnt need power (remote bases) other doesnt need coal (belt efficiency)

Electric poles - there is a reason why i carry all 4 when building - even basic\advanced have their use (sometimes having small area coverage is essential)

Inserters all serve different purpose too

shields and assemblers are mostly to cover early-mid-late game I agree, while gun turrets are plain broken still and waiting for combat rebalance

Research is literally resource sink - initial drive to build, nothing else. you wont get far in terms of enjoyment adding more and more researches, you add enjoyment by adding new mechanics.

[SHiRKiT]

The issue here is covering a large area of terrain. Again, I think layered (i.e. colored) regions for robots is THE best solution one can come up with. Different logistics networks occupying the same space is just perfect. Then, one could restrict also the construction robots to follow the same principle. This has been discussed before.

[User_Name]

The top-end robot speed isn't insanely fast yet, and the robots (particularly construction robots) can take a while to travel across a large base. So why not add a few more robot speed upgrades?

Because you want to brute-force bad layout to work fast by using research button.
We have DyTech for that.

[ssilk]

The issue here is covering a large area of terrain. Again, I think layered (i.e. colored) regions for robots is THE best solution one can come up with. Different logistics networks occupying the same space is just perfect. Then, one could restrict also the construction robots to follow the same principle. This has been discussed before.

Wouldn't it not be more useful, to have something, like a mobile roboport?

Place blueprint outside of construction area. The next "flying roboport" requests the needed materials, fuel and construction bots, if full flies (very fast!) to the area and builds all the stuff. Same for deconstruction.

[Nova]

The idea is good, but it should not be fast. It's a flying roboport, it has to be slow. Not too slow, but slow. To be honest, I like that idea. It would make it easier to create big solar fields.

[ssilk]

Perhaps a definition of "fast" for this: Faster than construction robot. Slower then the car. Somewhere inbetween.
Then it could be used for creating rail tracks and small outposts.
The speed is needed, because you have only one in the beginning (this is quite expensive) and it will construct only that part, that it covers, so it has only that many items in it's stacks. Then it needs to return. Or you bring the needed item more in place.

So the speed of building big solar fields is depending from the distance to the item-source. Quadratic.

Let's say you have only one roboport. The covered distance is 100x100. The flying roboport has also 100x100. If it was stationed at the roboport, he needs to fly 150 tiles to build the next part of the solar field. And back to refill.

Code: Select all

|<----- about 300x300 tiles -->|

XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX
XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX
XXX 1 XXXX XXXX 2 XXX XXXX 3 XXX
XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX
XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX

XXXXXXXXXX            XXXXXXXXXX
XXXXXXXXXX            XXXXXXXXXX
XXXX 4 XXX      R     XXXX 5 XXX
XXXXXXXXXX     CCC    XXXXXXXXXX
XXXXXXXXXX            XXXXXXXXXX

XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX
XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX
XXXX 6 XXX XXXX 7 XXX XXXX 8 XXX
XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX
XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX

R = Roboport
C = Logistic chests
One field is 100x100 tiles

Understood: The Flying Roboport needs to fly from roboport R to the first "field" (1) and back, then to field 2 and back, then to field 3 and back. And so on.
For easier calculation - it needs to fly about 8 x 300 tiles. A bit more, cause of diagonals, let's say it is 2500 tiles altogether.

Now let's think how long this last for the best case. Assume, the chests are really full, there are more than enough items and we have also plenty of logistic bots to fill the flying roboport.

- Refill takes 10 seconds.
- Construction takes 35, cause the construction bots need to fly 100 tiles in maximum (50 forth and 50 back, if it is at the border). That takes in the worst case 100 tiles / 1.2 tiles/sec = 83 secs and in the best case (with full robot speed) 100 tiles / 2.88 tiles/sec = 35 secs

So in the best case, without traveling time, the construction will take 8 * 45 secs = 360 secs = 6 minutes.

Lets's say, it should last not longer than 10 minutes, so we have 4 minutes of traveling time. How fast should it be?
2500 tiles / 4*60 secs = 2500 tiles / 240 secs = 10,41 tiles/sec
That is not very fast, in realworld about 37 km/hour. I think it should be faster. About 16.

There are more ideas, which come into my mind:

- maybe it can load for more than one construction area? It looks for a good route through all open constructions and loads as much as possible. Or it is reloaded during construction/travel? Some kind of transport robots? That would reduce the construction time drastically.

- This flying roboport searches for orders (deconstruction, construction in that order), which is not covered by roboports and fly to it (after 10 seconds) to construct it. I think this area they are searching for is limited somehow. It should work more or less like the roboport but on factor 10-20 more scale. Therefore this speed is really needed.

- And where do they dock and refill? I think to this idea: https://forums.factorio.com/forum/vie ... f=6&t=7150
Pneumatic tubes (or some other quite fast transport system for single items). Some kind of "Master roboport", which contains a delivery station for pneumatic tubes and the flying roboports docks on the top and is filled more or less directly.

[Garm]

How about instead of flying roboport we would have a train roboport?

Frankly I believe that every new step should either give us new opportunities or/and use established base structure. This is one large gripe we had with robots as soon as they were introduced: They were too detached from the rest of the factory design (except the production part) resulting in making factories obsolete.


Even now I find myself rushing towards roboport research because i know it will change the way I build my bases too much so making stuff before hand would be irrelevant.


I am afraid having simply flying roboports would cause similar effect.

[immibis]

Character Logistics Slots is another research that could have more levels for the end-game.

[SHiRKiT]

I agree with you ssilk, that would be a great idea for construction robots. But I still think that layered logistics systems is the way to go for the logistics ones.

[DevilXD]

There was a mod called "PocketBots", which included a power armor with a roboport mounted to it (author wanted to make it as a power armor module, but there isn't an API for that)...

It would be great, if instead of having flying or train roboport, we could actually BE a roboport Just put a roboport module in your power armor and go

[ssilk]

I used that mod a lot and I patched it a bit for my uses, more reach for example, but I was never really satisfied with this, cause it is only about 50% faster to build the stuff like so. There are several limits:

- You nearly never have the right items with you
- if you have all items, you need exactly double as much as expected. Always.
- You need to run to that "center" of your construction. That takes time. I would like to place just a blueprint anywhere and the robots will do the rest
- you need to wait, until all construction bots come back.

And some more issues. I'm sure, Kovarex is aware of all this issues, but for me - provided the awaited personal roboport works more or less similar to the current, it is still not fast enough.

Speed can be obtained by
- Building with much more speed. That is more or less this suggestion and the personal roboport.
- Building in parallel and/or half-automatic (place blueprint and forget). This is what I want.

[immibis]

This thread is not about personal roboports.