Page 4 of 6

Re: Version 0.15.11

Posted: Wed May 17, 2017 11:12 am
by bartekltg
I was sure I send it earlier:
kovarex wrote: It shouldn't make the efficient nuclear setups impossible, it should force you to build heat pipes shorter and branch faster.
The original idea of the whole heatpipe system was to force some kind of fractal designs, which is probably not going to happen, but this should be at least closer.
-The best reactor setup is a line 2xn, We have only one direction to build it;) No fractals.
-It seems that to make make fractals more attractive, the setup have to require large area spanned by pipes. Now we just make steam as fast as possible and push it outside. To make big fractals, energy transfer by pipes have to be more attractive than to transport steam, so I would build turbine next to exchanger (maybe hot steam loosing temperature?). Now we went deeper into opposite situation.
-The -1 degree penalty made branching less attractive. Without it if pipe branch, the temperature drop on each tile if divided by 2 (assuming equal load). With additional constant drop is less.
kovarex wrote: The point was to not make 52 tiles long heat pipe viable, so it is not a bug. You just need to update your designs.
I tough it was broken because unconnected pipes keep temperature gradient, but malventano explained it was side effect ot the new "-1degree" mechanic. Sorry. My later posts are about why I think linear resistance is still better, but this is only a opinion, not a complain;-)

My reactor still works (I'm using heat pipes to store energy (500MJ/tile, quite OP:)), so I need less heat exchangers per reactor than people who store in steam), and in sandbox I can draw 390MW from one double line, without too many problems one can draw all power from arbitrarily long 2xn setup.

I don't mind less efficiency (nuclear power still is the best energy source), just that flat drop itches my inner physicist :) Why long pipes can't be discouraged by increasing heat resistance? So it drops more than current 0.065C/(tile*MW). Effect is similar, we can draw too much power from the end of long line because temperature drop, and there is no effect of diminishing return with pipe branching.

Also, as always, great work!

Re: Version 0.15.11

Posted: Wed May 17, 2017 11:29 am
by gendalf
When you try to rename a blueprint and a drone delivers something to you the textbox resets, it's really annoying.

The blueprints shouldn't be an item, you get stuck with a blueprint in your mouse selected when your inventory suddenly gets full.

Re: Version 0.15.11

Posted: Wed May 17, 2017 12:46 pm
by Ranakastrasz
Oh, excellent. The range decay on heat pipes adds a second factor to reactors, forcing tradeoffs.

Re: Version 0.15.11

Posted: Wed May 17, 2017 3:54 pm
by Taipion
Is there actually any intention to fix the Kovarex Enrichment Process messing up the statistics?

Re: Version 0.15.11

Posted: Wed May 17, 2017 4:03 pm
by Nexarius
Ferlonas wrote:
Achievements are no longer unlocked by playing multiplayer game in which the player spent less than 50% of time online.
I have my doubts about this. I have a headless server running (several weeks now) that is not paused when there are no players on it. Since I've been pretty busy and didn't have a lot of time to play (neither did the others), updating to 0.15.11 will essentially mean I'm not eligible for any achievements any more. I don't know if that's feasible.

Regards,
Ferlonas
I guess it should be that you have to be more than 50% of the time online while ignoring the time without anyone online.

Re: Version 0.15.11

Posted: Wed May 17, 2017 4:12 pm
by Taipion
Nexarius wrote:
Ferlonas wrote:
Achievements are no longer unlocked by playing multiplayer game in which the player spent less than 50% of time online.
I have my doubts about this. I have a headless server running (several weeks now) that is not paused when there are no players on it. Since I've been pretty busy and didn't have a lot of time to play (neither did the others), updating to 0.15.11 will essentially mean I'm not eligible for any achievements any more. I don't know if that's feasible.

Regards,
Ferlonas
I guess it should be that you have to be more than 50% of the time online while ignoring the time without anyone online.
If you think of it like this, it is getting too complex, and difficult to track, and won't actually do what you want.

Instead, they should just add an option, so that each player for themself can decide if they want to get achievements from multiplayer games, without any time measurement.
After all, achievements are easy to cheat if you really want it, so I'd say this is a decent solution.

Re: Version 0.15.11

Posted: Wed May 17, 2017 5:18 pm
by Aeternus
kovarex wrote: The original idea of the whole heatpipe system was to force some kind of fractal designs, which is probably not going to happen, but this should be at least closer.
Recommendation if this is the goal: Limit the neighbour bonus for reactors to 200% instead of current 300%. That makes a 2x2 reactor cluster maximally efficient, and allows players to focus around a quad reactor blob with short heatpipes branching in all 4 directions. The 300% neighbour bonus limits reactor designs for maximum fuel efficiency to straight columns to get the bonus from 3 neighbours, as long as possible (but typically no longer then 2x8 or so, at 16 reactors you run into heat capacity issues in burst operation). This allows various designs with 4 reactors, and if you need more power, you just build more of those then.
Yes, it can be done now already, but the OCDly efficiency fanatics among us want to squeeze out every last joule ;)

Re: Version 0.15.11

Posted: Wed May 17, 2017 5:20 pm
by malventano
Taipion wrote:Is there actually any intention to fix the Kovarex Enrichment Process messing up the statistics?
The 'statistics' actually work, but the ratio/chance thing itself is a bit unrealistic since you can possibly get two or zero items from a cycle, which does not happen anywhere else in the game. I wish the devs would consider changing it to a true 'or' mechanic, at least for when the probabilities total to 100%, as that would allow for interesting chance-based recipes that actually work as expected. Also, 10 uranium ore should always give you *something* when processed. Having cycles possibly output two or zero units breaks the ability to count outputs per cycle as part of circuit network monitoring.

Put simply - dice rolls can't give you zero or all results. Neither should uranium processing.

*edit* to make this clearer, it needs to not be a raw probability per item, but actually a discrete probability distribution.

Re: Version 0.15.11

Posted: Wed May 17, 2017 5:24 pm
by malventano
malventano wrote:
Arek87 wrote:
FactorioBot wrote:
  • The zoom level at which the map switches from 'map view' to 'world view' was increased.
WHY ? :( Map view lost its advantage
I can see why they did this. If you are zoomed out fully in world view and want to view ore levels, etc, switching to map view fully zoomed in gives the same level of zoom but the advantage of viewing those stats, etc.
The more I try to play with the new zoom change, the more it appears that the devs may have undermined the whole reason for implementing the remote viewing in the first place. You really should be able to at least see the radar vision area in 'grainy' world view, but currently, you are limited to way too tight of an area. Heck, you can view a larger area than that in the 'normal' (non-map) view! Seems backwards...

Re: Version 0.15.11

Posted: Wed May 17, 2017 6:00 pm
by Mendel
eX_ploit wrote:So, after this change I rearranged my 4x4 nuclear design to use as little heat pipes as possible, gotta go with the times.
I like that design. Any chance for a blueprint string?

Re: Version 0.15.11

Posted: Wed May 17, 2017 6:02 pm
by malventano
Mendel wrote:
eX_ploit wrote:So, after this change I rearranged my 4x4 nuclear design to use as little heat pipes as possible, gotta go with the times.
I like that design. Any chance for a blueprint string?
There's a similar one (that I've verified works to 480MW) here.

Re: Version 0.15.11

Posted: Wed May 17, 2017 6:35 pm
by Kelderek
I keep coming back to the discrepancy between the neighbor bonus for reactors and the shorter range of heat pipes. These two ideas conflict with each other. Do you devs have in mind an upper limit for how many reactors can be grouped together before you have to create a separate cluster?

I kind of like the suggestion of limiting the neighbor bonus to 200%, that would lead to fewer heat exchangers needed to hit full capacity, but I also worry that it may lead to just a boring straight line of reactors.

Re: Version 0.15.11

Posted: Wed May 17, 2017 6:42 pm
by malventano
Kelderek wrote:I keep coming back to the discrepancy between the neighbor bonus for reactors and the shorter range of heat pipes. These two ideas conflict with each other. Do you devs have in mind an upper limit for how many reactors can be grouped together before you have to create a separate cluster?

I kind of like the suggestion of limiting the neighbor bonus to 200%, that would lead to fewer heat exchangers needed to hit full capacity, but I also worry that it may lead to just a boring straight line of reactors.
You can string >160MW worth of exchanges directly off of the reactor if you just go in a straight line away from the reactor - on a single string of heat pipes. Realize that the temperature falloff rate per pipe decreases as you get further out because you are branching off heat (exchangers) along the way out.

Re: Version 0.15.11

Posted: Wed May 17, 2017 6:50 pm
by Distelzombie
In my nine builds I made in the last two days I came to the conclusion that about 8 reactors is the maximum. Maybe more, but it is very complicated to get everything right. And I have cerain requirements they have to comply to.
Im happy I got a 8 reactor setup to work that even looks good, but I dont think more than 10 would be possible. (If you want it take a look in my signature)

Re: Version 0.15.11

Posted: Wed May 17, 2017 8:01 pm
by Floaf
malventano wrote:
Taipion wrote:Is there actually any intention to fix the Kovarex Enrichment Process messing up the statistics?
The 'statistics' actually work, but the ratio/chance thing itself is a bit unrealistic since you can possibly get two or zero items from a cycle, which does not happen anywhere else in the game. I wish the devs would consider changing it to a true 'or' mechanic, at least for when the probabilities total to 100%, as that would allow for interesting chance-based recipes that actually work as expected. Also, 10 uranium ore should always give you *something* when processed. Having cycles possibly output two or zero units breaks the ability to count outputs per cycle as part of circuit network monitoring.

Put simply - dice rolls can't give you zero or all results. Neither should uranium processing.

*edit* to make this clearer, it needs to not be a raw probability per item, but actually a discrete probability distribution.
He is probably refering to that when they do produce 1+ glowing uranium net they consume a lot of glowing uranium and output a lot of the same so in production statistics the numbers are actually like 40x the net produced amount and that makes it quite hard to see if you are slowly gaining or loosing uranium.

Re: Version 0.15.11

Posted: Wed May 17, 2017 8:15 pm
by PhasmaNL
Aeternus wrote:Okay, the worldmap zooming change I do not like. It was really handy to be able to, provided that you had radar coverage in the area, be able to oversee a huge part of your factory from the world map. Now with the narrower zoom requirements that handy functionality is gone. I really don't understand why that had to be changed - if some people can't handle the wide area view, make it a configurable option perhaps? [..]
Agreed. I don't like this change. I get Factorio isn't meant to be a "god mode" game. You play a character. But I felt there were enough limitations even with the previous more generous zoom, that it wasn't an issue. It's not like you would/could play the majority of the game through map view, so there was no need to nerf it this much.

Re: Version 0.15.11

Posted: Wed May 17, 2017 10:24 pm
by jarcionek
The zoom level at which the map switches from 'map view' to 'world view' was increased.
Please revert this or at least make it configurable (or maybe a research?). After this change it is now possible to see more of the 'world view' while not in the map...

I really loved the idea of placing bigger and bigger blueprints - in the end the game is about automation and this change looks like a step towards micro-management.

Re: Version 0.15.11

Posted: Thu May 18, 2017 12:46 am
by Distelzombie
Also let us use shift+mouse wheel to have a more incremental zoom.

Re: Version 0.15.11

Posted: Thu May 18, 2017 3:11 am
by Taipion
Floaf wrote:
malventano wrote:
Taipion wrote:Is there actually any intention to fix the Kovarex Enrichment Process messing up the statistics?
The 'statistics' actually work, but the ratio/chance thing itself is a bit unrealistic since you can possibly get two or zero items from a cycle, which does not happen anywhere else in the game. I wish the devs would consider changing it to a true 'or' mechanic, at least for when the probabilities total to 100%, as that would allow for interesting chance-based recipes that actually work as expected. Also, 10 uranium ore should always give you *something* when processed. Having cycles possibly output two or zero units breaks the ability to count outputs per cycle as part of circuit network monitoring.

Put simply - dice rolls can't give you zero or all results. Neither should uranium processing.

*edit* to make this clearer, it needs to not be a raw probability per item, but actually a discrete probability distribution.
He is probably refering to that when they do produce 1+ glowing uranium net they consume a lot of glowing uranium and output a lot of the same so in production statistics the numbers are actually like 40x the net produced amount and that makes it quite hard to see if you are slowly gaining or loosing uranium.
That is correct, the Koverex enrichment messes up your statistics in that it adds ALL the products, that is, the one U235 that you actually produced AND ...ALL the 40 others to you "production"; whereas it adds 40 U235 to "consumed" where you did consume actually ZERO.

Also statistics are broken on another part in that they don't actually show correct values, easy to track by my 12 reactors, if they have not been running for more than an hour, I look at the production statistics for time = 1 hour, and activate them for one round, the 12 reactors consume 11 fuel according to statistics.
= 12 fuel consumed, statistics showing 11 fuel total consumed for last hour
But the devs are ignoring this one and I got fishy excuses why this displaying is actually correct, but I tell you one thing: If you consumed 12 in the last hour, and "last 1 hour" shows "consumed" "total" is 11, then it's broken, it's a bug, no matter how you put it.

Re: Version 0.15.11

Posted: Thu May 18, 2017 5:10 am
by porcupine
jarcionek wrote:
The zoom level at which the map switches from 'map view' to 'world view' was increased.
Please revert this or at least make it configurable (or maybe a research?). After this change it is now possible to see more of the 'world view' while not in the map...

I really loved the idea of placing bigger and bigger blueprints - in the end the game is about automation and this change looks like a step towards micro-management.
On a 4k monitor, it's now painfully close in map view. Noticeable to the point where you're zoomed in far closer then you'd naturally be zoomed in (IE: far closer then you'd be zoomed in, when playing without being in map mode).

[edit]I should clarify, on a *large* 4k monitor (IE: 43")[/edit]