Factorio Forums

https://www.factorio.com/blog/post/fff-224

I realize this is exceedingly late in the day and could be labeled as rude necroposting, but I still think it needs to be said. Whatever would give a huge UPS boost on belts would do the trick in bringing players back into the belting fold.

Everything in factorio is practically infinite in terms of resources - except memory and cpu. Space, raw materials, electricity, they are theoretically infinite. However, we're all limited by budget by what our computers can do.

Whatever can utilize those real life resources best would do the trick in changing the nature of the game. A new belt type that perhaps behaves differently but is highly UPS efficient would be what I would have added.

Maybe some kind of high speed belt that can't be interfered with except at the input / output points.

A mod that brings stretched chests (eg 1x10) to the game would do the trick.

transportation from one end to the other end would be instantaneous. connections could be made via loaders.

But I don't know how ups intensive that would be compared to belts.

blazespinnaker wrote: ↑Fri Sep 25, 2020 12:23 amMaybe some kind of high speed belt that can't be interfered with except at the input / output points.

You mean a train?

But seriously, a single belt is highly optimized. The UPS drops occur when you put splitters everywhere.

blazespinnaker wrote: ↑Fri Sep 25, 2020 12:23 am I realize this is exceedingly late in the day and could be labeled as rude necroposting, but I still think it needs to be said. Whatever would give a huge UPS boost on belts would do the trick in bringing players back into the belting fold.

Blue belts already give you better UPS than transporting the same amount of items via bots.

blazespinnaker wrote: ↑Fri Sep 25, 2020 12:23 am https://www.factorio.com/blog/post/fff-224

I realize this is exceedingly late in the day and could be labeled as rude necroposting,

Have you tried to read these threads ?
viewtopic.php?f=38&t=56218
viewtopic.php?f=38&t=56519
viewtopic.php?f=5&t=56295
And these suggestions as additional bonus.

It's no necroposting if you have something new to add to the debate .

Now, I dare you to build a base big enough for the ups to drop significantly in two versions : as similar as possible (same trains, same power generation, ...) but one with belts only, and one with bots only. I'd be interested in the performance of each.

The problem late game with belts is that for massive throughput you need massive belt lanes. And then most often you need to figure out how to balance those massive lanes. Where as bots can inherently self balance if setup correctly and have almost unlimited throughput by just adding more bots and upgrading speed. Ups isn't really an issue for me lategame its figuring out the large space needed for using massive belt systems

Rjskeet wrote: ↑Wed Sep 30, 2020 10:30 pm The problem late game with belts is that for massive throughput you need massive belt lanes. And then most often you need to figure out how to balance those massive lanes.

Unless you use bots to balance the belts.

Rjskeet wrote: ↑Wed Sep 30, 2020 10:30 pm The problem late game with belts is that for massive throughput you need massive belt lanes. And then most often you need to figure out how to balance those massive lanes. Where as bots can inherently self balance if setup correctly and have almost unlimited throughput by just adding more bots and upgrading speed. Ups isn't really an issue for me lategame its figuring out the large space needed for using massive belt systems

This sounds like a job for a multi level belt. (nudges devs)

Bots have a limited throughput. The limit is every tile between the in- and output chest being plastered with roboports. Iirc there was a test on that in a fff, probably back when the issue last came up there. I don't remember details but it was surprisingly low.
Edit: Found it, #225 "The result is, that bots did 16.4k per minute while belts only 9.6. "

Indeed, bots shine as long as they don't have to recharge. If one has something to transfer during a short enough duration that arbitrarily as many bots as desired can do the transfer without having to recharge before it's finished - and they have enough roboports at range to recharge once they are done before the next transfer spike occurs, then bots are unbeatable. However, a little testing shows that the constraints are very limiting, and sustained high throughput is not as achievable as commonly thought.

mergele wrote: ↑Thu Oct 01, 2020 3:14 am Bots have a limited throughput. The limit is every tile between the in- and output chest being plastered with roboports. Iirc there was a test on that in a fff, probably back when the issue last came up there. I don't remember details but it was surprisingly low.
Edit: Found it, #225 "The result is, that bots did 16.4k per minute while belts only 9.6. "

AND all those Roboports will each be continuously consuming ~4 MW of power. If using solar then that also translates into a lot more space (~1170 tiles), or if using nuclear it translates into even more UPS cost.

Nuclear power is a hell of a thing lol I have a 4.8gw blueprint that takes up a fair chunk of space. But its super tiny compared to the amount of solar that I would need to replace it

Rjskeet wrote: ↑Thu Oct 01, 2020 9:54 pm Nuclear power is a hell of a thing lol I have a 4.8gw blueprint that takes up a fair chunk of space. But its super tiny compared to the amount of solar that I would need to replace it

Are you telling us that you literally cannot find the space to place a solar plant?

TheRangerLOL wrote: ↑Thu Oct 01, 2020 11:10 pm

Rjskeet wrote: ↑Thu Oct 01, 2020 9:54 pm Nuclear power is a hell of a thing lol I have a 4.8gw blueprint that takes up a fair chunk of space. But its super tiny compared to the amount of solar that I would need to replace it

Are you telling us that you literally cannot find the space to place a solar plant?

I wouldn't be shocked if it all that space is used up by roboports.

starlinvf wrote: ↑Thu Oct 01, 2020 11:20 pm

TheRangerLOL wrote: ↑Thu Oct 01, 2020 11:10 pm

Rjskeet wrote: ↑Thu Oct 01, 2020 9:54 pm Nuclear power is a hell of a thing lol I have a 4.8gw blueprint that takes up a fair chunk of space. But its super tiny compared to the amount of solar that I would need to replace it

Are you telling us that you literally cannot find the space to place a solar plant?

I wouldn't be shocked if it all that space is used up by roboports.

See he gets it lol

I don't get it...

DaveMcW wrote: ↑Fri Sep 25, 2020 1:35 am

blazespinnaker wrote: ↑Fri Sep 25, 2020 12:23 amMaybe some kind of high speed belt that can't be interfered with except at the input / output points.

You mean a train?

But seriously, a single belt is highly optimized. The UPS drops occur when you put splitters everywhere.

Yeah, maybe that's the spark, belts that feed directly into assemblers.

At what kind of factory size UPS gets a problem with belts?

I've built a belt-based mega-factory with 1000 SPM, and I have no issues with UPS. And some of the parts look really great - my electronics sub-factory for example is real eye-candy with all those colorful belts filled with those different circuits!

My next target is to build a new factory with 2000 SPM - I'll see if I run in performance problems there, or if my computer can still manage the load.

Note - I just checked how much of my hardware ressources my 1000 SPM base actually needs. One of my four cores runs at around 40 to 50% of its maximum performance, and I need 4 GB of RAM. The other cores are at around 20 to 30%. So I should be able to run a similar built 2000 SPM base I guess.

blazespinnaker wrote: ↑Fri Oct 02, 2020 7:15 am

DaveMcW wrote: ↑Fri Sep 25, 2020 1:35 am

blazespinnaker wrote: ↑Fri Sep 25, 2020 12:23 amMaybe some kind of high speed belt that can't be interfered with except at the input / output points.

You mean a train?

But seriously, a single belt is highly optimized. The UPS drops occur when you put splitters everywhere.

Yeah, maybe that's the spark, belts that feed directly into assemblers.

Rjskeet wrote: ↑Wed Sep 30, 2020 10:30 pm The problem late game with belts is that for massive throughput you need massive belt lanes. And then most often you need to figure out how to balance those massive lanes.

That depends on how you design your base. In 0.18 I built a 1500 spm (all 7 sciences) belt based base but I never needed a belt balancer larger than 4 belts. (I was using 1-4 trains. Hence I used 4 belt balancers to ensure that each cargo wagon was evenly loaded and that each belt at the unload stations able to draw from all 4 cargo wagons. If you are using 2-8 trains then you might want 8 belt balancers, but you could still cope with 4 belt balancers).

(Note that was on a 10 year old i7-860 cpu, so on a more modern system that sort of design should scale to over 3000 spm).

Rjskeet wrote: ↑Wed Sep 30, 2020 10:30 pm Where as bots can inherently self balance if setup correctly and have almost unlimited throughput by just adding more bots and upgrading speed. Ups isn't really an issue for me lategame its figuring out the large space needed for using massive belt systems

Large bot bases will run into issues unless you split up the base into multiple logistic networks.

To me the popularity of bot bases is that to new players they are a new toy, they look like an easy solution, and a new player doesn't recognise their scalability challenges. (For small isolated networks they are very scalable, and it is easy to design something high throughput. But as the bot network gets larger, then charging issues, space taken up by roboports, and UPS considerations makes belts more attractive).