|
Dr. Stab posted:I hope they looked at how DSP handles stacking items of different qualities w/ proliferator spray. Because it works really smoothly there I was trying to say that but worded it poorly, and I may be remembering incorrectly because it's been a while. I think what happens is the display averages out what's in the container and when it spits something out the percentage of each item determines the chance of what you get. So if you've got 500 mcguffins and 100 are better quality, you've got a 20% chance of getting a good one sent out and then the numbers are recalculated based on what's left. That way nothing is ever lost.
|
#
?
Sep 8, 2023 21:51
|
|
|
|
| # ? May 12, 2024 06:19 |
|
|
Xerol posted:That seems like a quick way to lose your legendary personal roboport that you spent hundreds of hours crafting and recycling parts by accidentally stacking it with a common. It works the way I thought. Keeping track of each individual item within a stack would not be that hard. Just store 5 numbers instead of 1 per stack. But, this does raise the issue of how you do want different behaviours for different situations. I want my buildings to be separated by quality in my inventory, but if I'm loading stuff on a train, I want it to stack together. In DSP only intermediates have qualities, so they don't have this issue. Also, I just realized that it's not either accept randomness or make a guaranteed legendary with recycling loops. You can also filter by quality every step of the way, and effectively have a tiered factory, where high quality intermediates filter up to eventually make, eg, beacons, and low quality filters down to make, eg belts. And then you get all the fun overflow shenanigans to maintain uptime on every level regardless of what your rolls are.
|
|
#
?
Sep 8, 2023 21:54
|
|
|
I mean they are going to need to implement a cycle held quality hotkey anyway for the purposes of ghost building, and that will work just fine for handling what's in your inventory.
|
|
#
?
Sep 8, 2023 22:04
|
|
|
Gadzuko posted:I was trying to say that but worded it poorly, and I may be remembering incorrectly because it's been a while. I think what happens is the display averages out what's in the container and when it spits something out the percentage of each item determines the chance of what you get. So if you've got 500 mcguffins and 100 are better quality, you've got a 20% chance of getting a good one sent out and then the numbers are recalculated based on what's left. That way nothing is ever lost. Apparently it's factorio that does the silly average thing. When you stack damaged items, the health is averaged.
|
|
#
?
Sep 8, 2023 22:24
|
|
|
Dr. Stab posted:Apparently it's factorio that does the silly average thing. When you stack damaged items, the health is averaged. It used to be different (like pre-0.12) which was an inventory nightmare - each damaged item took up a different inventory slot, even if two were damaged the same amount. But in that case it's not really losing anything of importance - whether you have two turrets at 60% health or one at 50% and one at 70% in your inventory doesn't really matter, especially when a repair pack can get them both back up to full for the cost of a fraction of a plate. IIRC things with "charges" like ammo, repair packs, and science instead add up their partial amounts, so picking up e.g. 5 turrets with 2.5 ammo each you'd end up with 12.5. I think it's only damaged items that the averaging applies to.
|
|
#
?
Sep 8, 2023 22:35
|
|
|
I think the main issue I have with this is that adding recyclers to production chains doesn't actually make the builds more complex or interesting. It just adds more busywork since every recycler array is going to be identical, less the number of actual recycler buildings to handle however much you're producing. It's just adding complexity for complexity's sake.
|
|
#
?
Sep 8, 2023 23:48
|
|
|
Dr. Stab posted:It works the way I thought. Keeping track of each individual item within a stack would not be that hard. Just store 5 numbers instead of 1 per stack. Except you'll probably want to divert any higher quality belts (at least belt 1s) to see if you can make higher quality science beakers. Oh god this is going to be a mess and after playing modded too much and for too long I love it!
|
|
#
?
Sep 8, 2023 23:55
|
|
|
Pseudoscorpion posted:I think the main issue I have with this is that adding recyclers to production chains doesn't actually make the builds more complex or interesting. It just adds more busywork since every recycler array is going to be identical, less the number of actual recycler buildings to handle however much you're producing. It's just adding complexity for complexity's sake. I'm not sure what you mean. Obviously it will have a blueprint solution but the same is true of literally the entire game. I don't get what the problem is. It's a new tool to solve a new problem. E: it's also not something you put on every line, only when you are pushing quality which you won't always want. It's an actual decision you have to make Gadzuko fucked around with this message at 00:21 on Sep 9, 2023 |
|
#
?
Sep 9, 2023 00:15
|
|
|
Re: stacking
|
|
#
?
Sep 9, 2023 00:24
|
|
|
A bit painful. Hopefully splitters can filter for quality.
|
|
#
?
Sep 9, 2023 01:06
|
|
|
New modder challenge: try to make a mod that adds quality to vanilla factorio as close to the FFF design as possible. The recipe tabs would become ungodly.
|
|
#
?
Sep 9, 2023 01:07
|
|
|
Rynoto posted:A bit painful. Hopefully splitters can filter for quality. they show splitters filtering for quality in the blog post
|
|
#
?
Sep 9, 2023 01:07
|
|
|
Manyorcas posted:New modder challenge: try to make a mod that adds quality to vanilla factorio as close to the FFF design as possible. The recipe tabs would become ungodly. I was looking to see if anyone had done this, but not yet. I do expect one to show up soon though, maybe even by the end of the weekend. In other news I'm still futzing around with various nonstandard mod combinations and thinking about doing a dangoreus + MSP game with cranked up tech costs. I just bailed on a 200 hour dangoreus + AAI + 10x cost run after taking that long to get blue science fully automated (95% of that time was just waiting for miners to clear space and stuffing as many iron chests as possible into every blank space to store extra coal/copper). I didn't realize you could change dangoreus ratios but according to an old changelog item you can: quote:- Ore ratios in some modes (perlin, pie, spiral) can be adjusted by changeing ore frequency So now I'm about to fire up a spreadsheet and figure out what ratios are needed for MSP (and then how I would need to change the default 3:2:2:1 ratio to match). Getting to uranium also poses a unique challenge as I believe that comes up relatively early in MSP (after blue but before purple/yellow science?).
|
|
#
?
Sep 9, 2023 01:15
|
|
|
yeah, really feel like that's going to make inventory/storage management a nightmare. Like even if you aren't hand crafting, you might end up in a situation where you need to pick stuff up, or just end up with poo poo in your inventory. I guess there's bots and a recycler if you want a process to just get rid of crap.
|
|
#
?
Sep 9, 2023 01:20
|
|
|
ymgve posted:they show splitters filtering for quality in the blog post Ah, missed that. At least organization will remain pretty painless. Oxyclean posted:yeah, really feel like that's going to make inventory/storage management a nightmare. Like even if you aren't hand crafting, you might end up in a situation where you need to pick stuff up, or just end up with poo poo in your inventory. From my own reading it seems like you can't even get a higher quality item without slotting in modules so handcrafting /early bases will remain exactly the same and you'll only have to faff with all that extra stuff if you want to.
|
|
#
?
Sep 9, 2023 01:23
|
|
|
Manyorcas posted:New modder challenge: try to make a mod that adds quality to vanilla factorio as close to the FFF design as possible. The recipe tabs would become ungodly. Py comes close to this with the alien life, which has recipes with a very low chance to produce a higher tier "creature" (implemented as a module) than your current tier. It's very painful but mostly because of how it's balanced. Rynoto posted:Ah, missed that. At least organization will remain pretty painless. And beyond being optional, it's also a strategic decision on WHEN to do it since there will be a point where you could start but probably won't have the resources. So even if you're the kind of person who can't ignore a mechanic, you can put it off until you need something to do (lol it's factorio you'll always have more things to do than time to do them) KillHour fucked around with this message at 01:34 on Sep 9, 2023 |
|
#
?
Sep 9, 2023 01:30
|
|
|
I wouldn't completely rely on this because I made a lot of assumptions that are decidedly not going to be true for some phases of the game. -No productivity accounted for -Assumes smelting iron and copper plates in steel furnaces with coal, but doesn't account for smelting steel or stone brick (steel, at least, is likely a significant amount not accounted for) -Doesn't count any coal for power -I probably missed some intermediates (I had to go back through everything and add the coal cost of plastic to like half of the list since almost everything past blue takes red chips to some extent) -Assumes every tech needs every pack, which is definitely not true even for the vanilla packs. A lot of techs seem to skip packs 13-16 (most nonmilitary techs past advanced oil), there's the purple-yellow split on lategame techs, and of course anything that doesn't need military. So it seems like having a ratio pretty close to the 3:2 iron-copper will work until you get into the super late game, when copper will become a bottleneck. Stone can almost certainly be cranked down a bit, maybe by half or more. Uranium still appears in "patches" in dangoreus so you can mine around them if you don't need it or feed the extra to the biters, so that probably doesn't need touched. The only possibly rough factor is the coal - in my previous games I was swimming in coal until coal liquefaction, but even after then never felt short on it. So that could probably also be turned down by as much as half. I also found you could run down your coal pretty fast by digging out some zones with burner miners. But there's also the problem of early-game just having way too much copper (or, more accurately, not enough iron). And with so many different science packs to manufacture, especially if I'm going to be running a higher tech cost multiplier, I'm not sure if I should turn up copper at all. (I should also note I'm playing on the completely random mixed ores setting.) Even though I will end up about 13% short on copper during the space science stage with the default settings, I'm still going to use a ton of iron in building the base, making ammo, etc. - basically I can almost always find a use for extra iron, unlike copper. So maybe I even just turn iron up? What I'm going to try first here is: -150% Iron -133% Copper -75% Coal -75% Stone These modify the default 3:2:2:1 ratio that dangoreus uses, so this will give me ratios of 9 iron : 5.33 copper : 3 coal : 1.5 stone, compared to the 9 : 6.8 : 1.4 : 0.6 I get from my calculations. This makes the endgame copper problem worse, but like I said I can probably find ways to use up extra iron. Might still be too much stone as well, but landfill eats a lot and I'm probably going to use a lot for concrete too. Gonna try these out and report back (probably after 100 hours of playing). e: And now I look through the settings and see that some packs got off their default crafting values, so I need to run through all of this again. Xerol fucked around with this message at 02:56 on Sep 9, 2023 |
|
#
?
Sep 9, 2023 02:52
|
|
|
Industrial Revolution 3 - Arc Furnace setup. I'm building an all new section before launching into the final stage (space!). I decided to go with a train-fed molten metal bus where molten metal is piped to casters at each build area. Casters can kick out basic materials like ingots, rods, plates or gears. Honestly the efficiency at this point seems like it's just too much, as a decent sized ore patch at train-distance will probably last the rest of the game. The pictured setup spits out the molten-metal equivalent to 3 red belts of copper ingots (or plates in vanilla as it's 1:1) plus a red belt equivalent of Nickel (actually a blue belt but I didn't add the smelters for the extra 15/sec). The input for this is a single red belt of copper ore. So 4 red belts out for 1 red belt in. I do have to add oxygen to the mix but that's simple as oxygen just comes from compressed air and filters. It also produces carbon dioxide gas which I'm simply venting (pollution) and the normal washing byproducts that I will have to ship off somewhere to deal with (sulfur, gravel and silica).  The reason for the big buffer on the input side of the nickel is that I am reading the buffer size of the pure nickel material to automatically switch between warm and cold ore washing to try to control material balance. If I have more than 10000 (less than 1 chest) of the nickel, then the pumps automatically swap to cold washing and drop the nickel production down to 10% from 50%. The buffer is probably bigger than it needs to be by quite a bit. E: I also didn't realize that cargo wagons effectively carry 4x as much ore as vanilla - they are twice as big and ore stacks to 100. I should have gone with 1-2 trains probably.
|
|
#
?
Sep 9, 2023 08:08
|
|
|
Quality sounds fun and will be interesting to see the most effective way to manage it (skimming uncommons off general production? Specific quality recycling loops?) But its gonna have a lot of irritating things to manage like clearing out your trains when the wrong quality item is in there, or needing to set up and manage a quality depot to send all your skimmed items to... interesting stuff.
|
|
#
?
Sep 9, 2023 08:54
|
|
|
The fact that they don't stack actually diminishes a lot of that. As soon as you put quality modules anywhere, you HAVE to filter everything and implement recycling for every stream if it overflows, or it's guaranteed to logjam eventually. And that will also apply to anything where any quality > 0 component becomes an input, it also will require filtering and recycling for every tier of output. I tend to view that decision as a rare mistake by Wube, though it does have some technical and design advantages, I think the downsides are worse.
|
|
#
?
Sep 9, 2023 18:17
|
|
|
K8.0 posted:The fact that they don't stack actually diminishes a lot of that. As soon as you put quality modules anywhere, you HAVE to filter everything and implement recycling for every stream if it overflows, or it's guaranteed to logjam eventually. And that will also apply to anything where any quality > 0 component becomes an input, it also will require filtering and recycling for every tier of output. I tend to view that decision as a rare mistake by Wube, though it does have some technical and design advantages, I think the downsides are worse. Unless I am reading it incorrectly, this entire mechanic is completely ignorable if you don't like it, as if you never build one of those quality modules and slot it in an assembler somewhere, you'll never see any quality items at all.
|
|
#
?
Sep 9, 2023 19:01
|
|
|
Sounds like they'll stack transparently in assemblers and only differentiate in containers.
|
|
#
?
Sep 9, 2023 19:23
|
|
|
K8.0 posted:The fact that they don't stack actually diminishes a lot of that. As soon as you put quality modules anywhere, you HAVE to filter everything and implement recycling for every stream if it overflows, or it's guaranteed to logjam eventually. And that will also apply to anything where any quality > 0 component becomes an input, it also will require filtering and recycling for every tier of output. I tend to view that decision as a rare mistake by Wube, though it does have some technical and design advantages, I think the downsides are worse. Is it? At least in the basic miner -> belt -> assembler -> other belt/chest -> assembler etc paradigm, assuming you can mix quality in the assembler this ought not to matter at all. There doesn't seem to really be a downside to mixing quality on the belt, aside from not actually meaningfully benefitting from the existence of quality, and the extremely low ratios of the quality progression mean that benefitting from quality is a deliberate decision where you are intentionally setting up resource lines to filter for and stockpile higher quality ingredients at low rates of return instead of expanding.
|
|
#
?
Sep 9, 2023 19:31
|
|
|
If they don't stack in assemblers, then you can never have multiple quality tiers on a dead-end belt or it will deadlock when there's say, 1 uncommon gear in the assembler and only common gears in reach on the belt. And if they don't stack in inserters you'd randomly suffer significant throughput hits. Not stacking causes severe problems basically everywhere if you don't segregate qualities at every step, and kovarex said they don't stack. Maybe he meant that they only don't stack in your inventory, but honestly that would seem weird.
|
|
#
?
Sep 9, 2023 20:54
|
|
|
Just got to the spidertron in Space Ex and I'm very disappointed that you can't put 4 tesla guns in it for maximum lightning
|
|
#
?
Sep 9, 2023 21:04
|
|
God I fucking love Diablo 3 gold, it even paid for this shitty title
|
K8.0 posted:If they don't stack in assemblers, then you can never have multiple quality tiers on a dead-end belt or it will deadlock when there's say, 1 uncommon gear in the assembler and only common gears in reach on the belt. They do not stack in inserters. Watch the bottom right inserter, the one that's collecting rare-or-better chips from the splitter and putting them in a chest. At one point, it grabs 1 purple quality and lets 2 blue quality sit on the belt. https://cdn.factorio.com/assets/blog-sync/fff-375-quality-recycling.mp4
|
|
#
?
Sep 9, 2023 22:31
|
|
|
Looking at the example images I think they don't stack in assemblers, either, but rather you set the assembler to use certain quality levels and it only draws materials of that level. Hence why in the above video there's four assemblers with various quality levels set as recipes. Which, yeah, is going to be a nightmare when it comes to jamming, though I can think of ways of designing around it. I suspect this is going to be something that either gets ignored for most of the game or, as the dev blog notes, quality modules just added to the final product assembler. Having a factory devoted to increasing quality at every step is going to require some extra special design. I'm not sure how I feel about it, though it's a fair point that if you want you can just ignore it and never use quality modules. Bremen fucked around with this message at 23:16 on Sep 9, 2023 |
|
#
?
Sep 9, 2023 23:14
|
|
|
Solumin posted:They do not stack in inserters. Watch the bottom right inserter, the one that's collecting rare-or-better chips from the splitter and putting them in a chest. At one point, it grabs 1 purple quality and lets 2 blue quality sit on the belt. Good spot and a good example. Any belt or chest of mixed qualities is going to significantly hurt inserter throughput for any recipe where that matters. The only practical way to address it will be filtering between every output and input, and recyclers to create flow if one of the five outputs jams (which would subsequently jam all five outputs, as you know if you've ever done mixed belts). I'm actually quite disappointed by how it works. I don't think there will be much creativity involved, since I can only see a few options that don't brutally gently caress your factory. And a LOT of busywork for anything other than quality modules for only the final step in a production chain.
|
|
#
?
Sep 9, 2023 23:24
|
|
|
Another issue occurs to me - since it looks like assemblers are set to use uncommon, or rare, or whatever, for the entire recipe, leveling up all the intermediate products is going to be an enormous pain with recipes that use some products with several intermediate steps and then, oh, some iron plates or whatever. It doesn't matter if you have a bunch of purple advanced circuits if the recipe also requires some iron plates and you barely have any of those - you just end up with enormous amounts of purple circuits piling up that you can't use.
|
|
#
?
Sep 9, 2023 23:33
|
|
|
Since speed modules have a quality hit associated with them, I doubt you'll ever be capped on inserter throughout anywhere you're feeding potentially-quality items in.
|
|
#
?
Sep 10, 2023 01:30
|
|
|
I trust their playtesting process to work out stuff like everything discussed here.
|
|
#
?
Sep 10, 2023 01:44
|
|
|
M_Gargantua posted:I trust their playtesting process to work out stuff like everything discussed here. That's what I keep coming back to: I can trust wube to make this fun.
|
|
#
?
Sep 10, 2023 01:47
|
|
|
And you can trust modders to offer you whatever gameplay your complaints desire
|
|
#
?
Sep 10, 2023 01:55
|
|
|
you're never going to unintentionally have mixed quality belts of components because quality is only relevant when you slot quality modules. the thing to do with quality, in my mind, is set up a specific subfactory for making high quality stuff focusing on things like military and power generation and have everything else just use standard poo poo
|
|
#
?
Sep 10, 2023 02:34
|
|
|
Legendary Nuclear Reactor
|
|
#
?
Sep 10, 2023 02:45
|
|
|
M_Gargantua posted:Legendary Nuclear Reactor Specifically mentioned on the FFF: produces more heat from fuel!
|
|
#
?
Sep 10, 2023 02:53
|
|
|
M_Gargantua posted:And you can trust modders to offer you whatever gameplay your complaints desire I'm very excited to see what modders do with all this new stuff. I tried space ex but it just isn't really my thing, something like k2 that's more of a vanilla+ with all the space stuff will be great
|
|
#
?
Sep 10, 2023 02:56
|
|
|
Alkydere posted:Specifically mentioned on the FFF: produces more heat from fuel! Does that mean a fuel cell lasts longer, or that you need more heat exchangers and turbines to make full use of it?
|
|
#
?
Sep 10, 2023 03:53
|
|
|
I’m really excited to play with quality. Something you can completely ignore, use for specific high impact items (spiders, armor, weapons) or go insane and start producing quality science packs. The inventory/hotbar management is going to be a nightmare.
|
|
#
?
Sep 10, 2023 04:08
|
|
|
|
| # ? May 12, 2024 06:19 |
|
|
I am looking for someone to make a calculator with the ratios of how many extra raw resources it takes to make legendary equipment + MW + calculations on the recycling rates etc. Figure out what if anything in the main factory actually benefits from it vs just building bigger like normal. I like to think that most things won't really need quality, and that quality does add quite a bit of resource and energy inefficiency in exchange for its benefits.
|
|
#
?
Sep 10, 2023 04:14
|
|