Page 5 of 5

Re: UPS wars: smelting challenge (design competition)

Posted: Wed Jan 24, 2018 3:04 pm
by quyxkh
Here,try it with this one:
quyxkh.clocked.70pctcover.2x5-57.trains.zip
(1.95 MiB) Downloaded 206 times
Here's the output I get running my own benchmarking script/mod:
benchmarking log
I cut the outpost down to match output rates, there's no way to scale the other map up to 234K/min because crafting speed 11 vs 12 and 54 smelters vs 57 per lane, although I think you could squeeze another two lanes in for almost no penalty, all the miner-only beacons look useless to me.

Re: UPS wars: smelting challenge (design competition)

Posted: Wed Jan 24, 2018 4:02 pm
by mulark
https://docs.google.com/spreadsheets/d/ ... sp=sharing

Results on my machine ^

Results gathered using: https://github.com/mulark/factorio_benc ... _script.sh Just change the pattern to match each persons respective map.

It does seem to be somewhat more efficient, though I wonder what percentage of that is due to trains/locomotives versus things like the blueprint.

Re: UPS wars: smelting challenge (design competition)

Posted: Wed Jan 24, 2018 4:34 pm
by quyxkh
It's absolutely the trains thing, the extra 1/12 swing per plate introduced by the outpost buffers (or avoided in DaveMcW's outposts) is a large fraction of the total production cost.

But I think that's the point of tbterra's requirement: more-efficient production that requires a more expensive delivery setup is just shoving costs around, not eliminating them, so production + delivery cost is the figure of merit. What the buffered outposts lose in production efficiency they more than regain in the simpler (usable, even), more efficient delivery they allow.

Re: UPS wars: smelting challenge (design competition)

Posted: Sat Jan 27, 2018 12:09 am
by mulark
I was wondering how belts compared now in .16 so I made a little test.

https://docs.google.com/spreadsheets/d/ ... sp=sharing

Not too bad i'd say.