Re: Friday Facts #137 - The release scarecrow
Posted: Sun May 08, 2016 2:01 pm
deleted
www.factorio.com
https://forums.factorio.com/
First off, I like your attitude of trusting the devs. They haven't steered us wrong yet!Losash wrote:There are many things which I dislike according to it, starting with trains, which is most capable and powerfull transport in the game already, got another huge buff. Add information from another FFF, in which they said they plan to increase amount of slots in carts too (in the future). But I don't want to start a list. What's the purpose anyways. I've already said that I let developers to "ride a ship", and it will be okay.
You're not saying why are belts necessary. Or why are belts better than trains for connecting networks? Trans offer much bigger throughput, and they're much easier to set up on scales you talk about. With current inserters belts are just inferior in late game.Losash wrote:...Tev wrote:Errr no. Just no.Losash wrote:Morale: extreame throughput in extreame lategame in my personal oppinion, a guy who spent more than 20 hours to read articles, is impossible without belts even in 0.12.
But for 0.12? It's just bots > belts, period. That's why rapid is good addition.
I'm talking about that you need them to connect 2 bot networks.
...
And long story short, those bases can't be done without a relitive use of belts, neither in 0.12, nor in 0.13.
I don't think microsoft is to blame for this. Blame mojang, or the fact the minecraft wasn't exactly designed with modding functionality in mind. Minecraft 1.8, which is apparantly is a *lot* of work to update a mod to, was released September 2, 2014. The rumors of a microsoft purchase came a week or so after that, unless my google-fu missleads me.albatrosv13 wrote:Another thing to point is that people may stay at older versions. I saw that in minecraft when microsoft took over. Some people(including me) still stay at versions of pre-microsoft. Mainly because of some mods(modmakers didn't update their mods anymore).
I agree with this. Fact is that trains are still pretty much pointless if all you want to do is fire up a rocket at default game settings. But very fun, and should therefore be encouraged. Rapid inserters should make getting started with them slightly easier since you can build smaller mining outpost stations without getting bottlenecked at the belt-to-chest transition.Tinyboss wrote:As far as buffing trains, I don't think it's a problem. Yes, they're the most powerful transport in the game, but they're expensive (in terms of materials but more importantly player time) and fun, and there's no point in using them until you really need to move all those items.
It seems pretty off-mark to invoke objectivity when talking about the creation of a work of art.albatrosv13 wrote:That's because one of the developers really love transport tycoon. Talk about objectivity.Losash wrote:There are many things which I dislike according to it, starting with trains, which is most capable and powerfull transport in the game already, got another huge buff. Add information from another FFF, in which they said they plan to increase amount of slots in carts too (in the future).
Another thing to point is that people may stay at older versions. I saw that in minecraft when microsoft took over. Some people(including me) still stay at versions of pre-microsoft. Mainly because of some mods(modmakers didn't update their mods anymore).
You don't even need belts to connect separate networks. If the networks are one tile apart, inserter->chest is enough. If further apart, inserter->chest->wagon->inserter->chest.Losash wrote: Sigh... Did you even read my message? You didn't understand my points. I'm not talking about using belts on a primary basis. I'm talking about that you need them to connect 2 bot networks. I repeat once again, I know how bots work and I know that bots>belts, and believe it or not, it will stay like that in 0.13, even if any other argument falls off, they still produce less lag than belts.
Actually those bases can be done without any belt use ins 0.12. Unless there is some fundamental changes to how robots work, so will 0.13. Clearly you have not built one or you would know. But this is rather sidetracked.Losash wrote: I know what bots excel in and I know how to use them. If you would read my message you'd get that I'm talking about bases which do rocket/minute and more. If they are doing 1 single task over a relitevely short distance, they can throughput any amount of items like ores per minute, limit is only your PC, which will show slideshow. Unloading a train, for instance is something bots will always excel. And I insist, that logistic network of roboports on a train stations block should be independent from a global base one, and don't even bother replying if you don't understand why is it so, when you are talking about bases, operating dozens of thousands of ores per minute.
And long story short, those bases can't be done without a relitive use of belts, neither in 0.12, nor in 0.13.
It HAS to have "reset timer" (maybe few seconds to make it not worth in sparse flow), or else it'll lead to frustrations.As the rapid inserter currently waits until its hand is filled with items, it is not clearly better than fast inserter. When the bonus is 5, and it has 4 items in hand, it can wait minutes (or forever) for the last item, which might be undesirable
Well, fuck.As the rapid inserter currently waits until its hand is filled with items [..] When the bonus is 5, and it has 4 items in hand, it can wait minutes (or forever) for the last item[..]
To make it useful, we will increase the maximum rapid inserter capacity (instead of inserter stack size bonus) from 4 to 11, so it can hold up to 12 items of the same kind.
Fatmice wrote:...What should have been done, in my opinion, is instead of changing inserter stack bonus as a technology applicable only to the rapid inserter type, the technology should be changed to an inserter upgrade. This technology is researched once and will give access to inserter upgrade slots and an inserter stack bonus module that can be built. Each inserter stack bonus module increases the number of items picked up by an inserter by one and energy consumption by 20%. The stack bonus and energy consumption increase are additive. All inserters will have a number of slots, depending on their type, for inserter stack bonus upgrades that can be inserted by the player. In this fashion, the player has direct control over which inserter should operate with stack bonus. What separates the rapid inserter from the other inserters is the queue behavior and item compression, i.e. they wait until their buffer is full before executing a move and items placed on belts are fully compressed. The other inserters will behave as they do now in 0.12.x where they move up to the stack bonus but does not wait for a full queue nor will they place items on the belt in such a way that compression is easily achieved...
And how many more inserters to keep up with beacon boosted setups now that stack bonus do not apply to the non-rapids? I can only guess that more than the savings afforded by the rapid inserter. Of course this needs to be tested with megabase builds to see any real gains assuming all of the megabase is not affected much and can simply be loaded up under 0.13 without drastic change, outside of inserter replacements where needed, for as close to a one to one comparison as possible.ssilk wrote: - Use of CPU
instead of 14 inserters to fill a belt, we need only 4 (or 5?), a reduction of 60% and so a reduction of some CPU usage. Sounds like a bad argument, but both - the loader and rapid inserter - would be able to reduce the CPU load.
Yes, I remember this change. While I don't disagree with the reversion, I find that it lacks imagination. This was a good opportunity to make inserter modular, which will make stack bonus transparent and not something you research and forget about.ssilk wrote: - so o so: Inserter code must be rewritten, good time to do more changes
Due to the changes on circuit network, the inserters need some rework. But I admit, it's a bad argument. But the next is much better!
- Fixes the "one item logic"
Before 0.6 or so we did't have inserter stack size bonus. We could do circuit logic like "if there are exactly 2 items in the box". What should I say: The logic didn't work anymore, when the stack size bonus was introduced. No, even more bad: You just understood, how the circuit network works, and you are very proud of building the first working circuit logic. And then you research the SSB and you don't even notice, that your fine logic doesn't work anymore. This is in my eyes a real fuckoff, which is now fixed with this change.
Yeah, they are simpler due to fixed insert but they are also now heavily bottlenecked by the same token. Tileable beacon builds will look quite strange indeed.ssilk wrote:
- Fixes stupid furnace logic stuff
There are hundreds of posts about going around this "limitation" of furnaces, mostly things like: How to built reliable multiple smelters setups (copper, iron and steel), that can switch between the needed stuff. And similar. I think those construction could be very useful.
All those constructions are now simply possible, but need some circuit logic instead. Cool! Yeah! That enables a lot of more complex recipes for furnaces. Don't let the smelter decide what to smelt, instead let the inserters/combinators decide what to fill in the furnace. Really, really nice idea! I like that.
The initial introduction of stack-bonus was just that, initial. The initial reason for the stack-bonus was also for throughput and computational saving. This is now a revisit. I really wish there was a bit more imagination put into it instead of this rehash. Afterall, if all inserters are now "smart" why aren't they also modular when it comes to their functionality?ssilk wrote: - The stack size bonus silently changes the existing structures
As already mentioned: In v0.6 or so the stack size bonus was introduced. That felt a bit unlogical to me. I can remember back, I had two thoughts:
a) How should a player understand, what an inserter stack size bonus is? The game doesn't explain that by itself! So, if I'm a new player and if I research SSB and look to my setup, that uses only stack-size bonus of 1 I don't see any difference and think: Fuck, why did I research this, it does nothing or I don't understand, what it does.
With this change it becomes logical and self-explaining again.
b) More important was this: how should this stack size bonus work? Is that just a software update to the inserters? But why didn't they work like so from beginning? where are the stored items? Magic? It makes no sense!
The rapid inserter makes that very clear: there is a device, that is different, than the others. If the graphical representation changes also - to let the player see, that where the itemes are stacked - it will be perfect! Depending on how that graphic is done, the stack size bonus could be really just a software update to the rapid inserters. Otherwise (mechanical changes) see above: How does this work without operating the changes to each inserter?
I don't think the difference is that much even with fixed insert. Most belt driven builds are operating below the insert throughput anyway and people who "upgrade" their inserter just because are of the mind set that newer is better. That is not necessarily the game's fault or design but more to do with their own logic.ssilk wrote: - More diversity
In the middle-game you need only lots an lots of fast inserters, many smart inserters and some long arms. The basic inserter and burner inserters aren't needed anymore. Some players systematically replaces basic inserters with fast. That reveals a weak point of Factorio: There are too many inserters and their differences are too equal.
This looks now different! So I try to think about what inserters we will have in 0.13:
- burner inserter - they make sense in the early game and make the energy production more reliable against full black outs, also useful for no-electric power setups (gun turrets)...
- basic inserter - no question, that is a must
- filter inserter - that might be the replacement for fast/smart inserter
- long handed inserter - nothing changes here...
- rapid inserter - well...
I would hope that filtering and the NOT-operator is part of their smartness, a toggle or a selection box that could be ticked once the item filtering has been applied, and not a separate inserter. But I will have to hold out and see given their lack of imagination with this wonderful opportunity to overhaul inserter functionality.ssilk wrote: I see the filter function very critical, because there are other (in my eyes more clever) ways to filter bulk items, than just with inserters (*). But when really introducing the filter inserter I hope they implement a NOT-filter.
I see also not the point between basic and filter inserter. This problem (every basic inserter can just be updated with filter inserter) keeps the same and it makes no sense to have a slow filter inserter. I don't have a good solution for that, maybe the frame-idea (to have only one basic type of inserter and the diversity is done by puting different modules etc. into it) makes sense, but that doesn't add much to the gameplay and increases the complexity a lot and reduces also game-handling-speed.
All high-througput factories use fast inserter. This is inevitable. Then again, I do not see why the game or the developer must explicitly require diversity for all other use case. People should just build however they like it and use whatever inserter they fancy, even if that faster inserter is not really an improvement over the basic inserter for many use case.ssilk wrote: - Limitation increases creativity (Why they're buffed)
As already mentioned: Blue inserters are the thing! They are used for everything! In the end I asked myself: Why again do we have basic inserters? The buff is needed to reset the situation and bring in the diversity of the inserters again. But see above.
This is purely your opinion of course. I personally see the change as blasé and off the target. It looks to me more like they are feeling the crunch time of June and so haven't iterated over this change enough yet. What do you know even the title of the FF gave off that sentiment.ssilk wrote: - Gameplay
All changes should aim to add more gameplay and reduce complexity. What should I say: 100% full hit into the center! If this above has been suggested in the suggestion board I would have said something like "Interesting, and it fixes a lot of problems, I don't see also no disadvantages in gameplay, only postive effects! That is a good suggestion, but it needs to be more specific. But I doubt that the devs will do such a big changes". So I'm in fact quite glad, that they do this step, knowing, that many players will not like it.
Less because you would use the rapids in the beacon setup. Chest to chest the rapid would transfer even more items than the stack upgrade allows now.Fatmice wrote:And how many more inserters to keep up with beacon boosted setups now that stack bonus do not apply to the non-rapids? I can only guess that more than the savings afforded by the rapid inserter. Of course this needs to be tested with megabase builds to see any real gains assuming all of the megabase is not affected much and can simply be loaded up under 0.13 without drastic change, outside of inserter replacements where needed, for as close to a one to one comparison as possible.ssilk wrote: - Use of CPU
instead of 14 inserters to fill a belt, we need only 4 (or 5?), a reduction of 60% and so a reduction of some CPU usage. Sounds like a bad argument, but both - the loader and rapid inserter - would be able to reduce the CPU load.
Hm. Not everything. See, there are objective, measurable reasons, why this change is better, than yet; the main point is, that it reduces complexity by reducing similar working inserters. Or take the smart inserter, which will become worthless with 0.13. And it removes a very high (but hidden!) complexity (remove the SSB), which is not needed it any case and sometimes also contra-productive. And it combines that with more specialization, which increases the gameplay by increasing the complexity only a bit.Fatmice wrote:This is purely your opinion of course.ssilk wrote: - Gameplay
All changes should aim to add more gameplay and reduce complexity. What should I say: 100% full hit into the center! If this above has been suggested in the suggestion board I would have said something like "Interesting, and it fixes a lot of problems, I don't see also no disadvantages in gameplay, only postive effects! That is a good suggestion, but it needs to be more specific. But I doubt that the devs will do such a big changes". So I'm in fact quite glad, that they do this step, knowing, that many players will not like it.
Well, that is of course your opinion. There might be some truth in it. But I don't think so cause I cannot know it. Instead I assume, that the devs will do always their very best. I can sleep much better with this thinking.I personally see the change as blasé and off the target. It looks to me more like they are feeling the crunch time of June and so haven't iterated over this change enough yet. What do you know even the title of the FF gave off that sentiment.