A New Direction [BC API/RF API]

[ScottulusMaximus]

This, however, I like. The issue with blowing up machines I have is that it's kind of a newbie trap. Once you get over it, it's not a danger any longer, but "oh, I read the instructions" doesn't feel like an accomplishment (unlike figuring out your perfect layout). There isn't much gain from it in terms of gameplay apart from the feeling.

Risky overclocks, however, is something that's done knowing the consequences, so that's fun. You could have machines where you disable the aforementioned "circuit breaker" - you can run things higher, faster, better... but then they WILL blow up if you stress/overheat them too much. But now it's not a trap, it's a deliberate gamble.

That, just so that^^^

Melting a cable 200 blocks away because you decided to see if one more overclock would work and plunging your whole base into darkness then trying to figure out how you stuffed up would be just fantastic... The heat buildup/wear of machines is also a very underused mechanic I feel

 

[GreenZombie]

Wtf is this thread about?
Power systems?
Or how to talk about inflamatory topics?

Either way, it doesn't have much direction. Is there an actual proposal?

 

[Queue]

Wtf is this thread about?
Power systems?
Or how to talk about inflamatory topics?

Either way, it doesn't have much direction. Is there an actual proposal?

Good question. Since I have school, I can't really keep the thread on topic too much so instead I have just let free form discussion occur. I hoped that the initial thread would get all of the meaningless arguing out of people's systems, and thus leaving this thread clear and far more civil. That being said, however, the last thread was indeed far more civil and full of thoughtful discussion, be it directly related to the question or not, than I ever could have hoped.

As for whether this thread was a proposal, word count restrictions really crammed the amount of meta that I could fit into it, so I am sorry it was unclear. The thread is many things. It is a proposal for future methods of handling the oh-so-infamous modded MC "drama", through 2 threads. It is also a second forum for discussing the same topics as the first thread if anyone wanted to clarify their points, and it is a thread about power systems. The latter I did not specifically ask for in the OP but it seems to be a topic people want to discuss, and hey, we are on a forum right? I have no problem if this thread wants to form into what the people want it to be, all I care about is that people be kind and leave the random shit for the Shitstorm itself

 

[ratchet freak]

I point I haven't seen brought up is propagation speed of power,

for example the old RP2 blutricity took ages to cross long distances in response to power input changes even though it had decent throughput when saturated, nearly all of the other power networks treat power transfer as instant with a delay in the order of a few tick at most when going through a power buffer.

 

[Azzanine]

One thing I ask in these type thread that has yet ro be properly defined is, "What do you mean by depth and complexity?"

Define it. Is it having a bunch of steps to get to a goal? Is it crafting? What do you mean?

I would say nuance.
Some examples of nuanced energy transmission:
With classic MJ it was a on demand system back before cells MJ couldn't be stored, so you needed mechanism in place that leads you to not waste fuel. Also getting the power to go to the right machines took some thinking.

EU had a feature with the tiered voltage, the known quartet of Low, med, high and the often forgotten Extreme voltage. If you didn't want to have your factory go boom you had to step that voltage down with transformers.

RotaryCraft has probably the most nuance in it power transmission as what powers a machine is determined by 3 variables that differ between each machine, Torque, Rotational speed and wattage which is the result of those two. You not only have to pick the right engine but also figure out what ratio of gearbox will work best.

RF transmission as it stands is as simple as laying the conduit from the source to the machine with no real decisions to make other then where you are going to put the pipe. Make note I am speaking of just the transmission as all energy production and usage is the same in even in the case of RotaryCraft. Even RF has some neat power gen mods, BigReactors has some real potential for tinkering.

What people mean by "depth and complexity" is something that makes you think and gives you opportunity to make decisions.

 

[Reika]

Seeing as I do have nothing against RF as a power system on its own merit, I must say I like the idea of a voltage-driven RF system. The API does not support that natively, but it could very easily be faked. Whenever RF is added to a compliant machine, three pieces of information are supplied: Direction, Amount, and "simulate", i.e "is the power actually being added or just simulated", and the machine then returns the amount "successfully" added (or what would have been added). That amount variable is normally just added to an internal buffer, but there is nothing stopping a machine from returning zero if the amount is below some threshold.

Technically, this is a current limit, as it is specifying a minimum RF/t, but that is as close as the RF API is going to allow.

More refinement is needed to that core idea to ensure there are no deadlocks or other issues with things like conduit throughput, but I think it has potential. However, it does need to be handled carefully, or it could very easily become its own power system again.

 

[Pyure]

There is one in the planning stage currently.

More info sir?

 

[RedBoss]

I would say nuance.
What people mean by "depth and complexity" is something that makes you think and gives you opportunity to make decisions.

I hear what you and others are saying. But even with MJ, once I got a boiler, there was no need to worry about anything. I suppose in my mind, there's no real differences in my mind in the "simple" RF mods and things like BC or Railcraft. You just plop down engine spam to defeat any "on demand" system. Any fuel issues are defeated by fuel creation abundance. Any wire system limits can be defeated by mining and crafting. Depth? Yes depth is there if you choose to swim, but I walk over that chasm with a wooden plank called brute force.

Semi-Pro guys, I actually read the RotaryCraft section of the wiki. I get it now.

The people that want depth want more than just fuel+pipes=goal. But to be honest, Railcraft, BC, TE, are all fairly similar in that department. I used pipe wires and gates one time for my biofuel setup. I hated losing lava turning them off and on, and I had a RC tank fuel of biomass, so I just left them on. From there, anytime I used MJ I'd just acquire so much fuel that I could leave the engines running. Early game I'd just unplug the machines that were draining energy. Problem solved. Boilers effectively destroy the need to even worry about it.

So I get the depth & complexity argument now. I just don't agree that, in terms of energy generation at least, that certain mods offered anything more than optional depth or complexity. This is good thing! I just see all the arguing and down playing certain mods for not being complex. In truth, once you have a certain level of experience with the mods, and you crave more complexity, none of the "farms=fuel" mods will satisfy you. You'll have to use mods like RotaryCraft or Atomic Science to get there.

 

[Padfoote]

More info sir?

That's about all there is currently. More info will be made available by the guys writing the mod when work begins.

 

[GreenZombie]

Seeing as I do have nothing against RF as a power system on its own merit, I must say I like the idea of a voltage-driven RF system. The API does not support that natively, but it could very easily be faked. Whenever RF is added to a compliant machine, three pieces of information are supplied: Direction, Amount, and "simulate", i.e "is the power actually being added or just simulated", and the machine then returns the amount "successfully" added (or what would have been added). That amount variable is normally just added to an internal buffer, but there is nothing stopping a machine from returning zero if the amount is below some threshold.

Technically, this is a current limit, as it is specifying a minimum RF/t, but that is as close as the RF API is going to allow.

More refinement is needed to that core idea to ensure there are no deadlocks or other issues with things like conduit throughput, but I think it has potential. However, it does need to be handled carefully, or it could very easily become its own power system again.

You need to explain for non modders, how exactly this would pan out in the many-mods in one pack ecosytem. i.e. Say that I have a mod(s) that provides machines that run off RF, and a different mod(s) providing RF power.

"RF as Current" would be implemented as new power generators? New Power consumers? Or both? And if both, will the existing power producers and consumers interoperate with the RF-Current based devices?

 

[Eliav24]

A Minecraft world has effectively infinite resources. You can always brute force your way through the game. If a mod offers any way to automate, you can always get effectively infinite of whatever is capable of automating. I still think that there is a fundamental difference between mods based on power versus ones based on energy, which is why it's easy to get "infinite" of things.

More to the point of the thread, RF reduces the complexity of most tech-based mod environment, It's impact on depth it is very dependent on the various mods and individual playstyles. The standards allow more mods to easily interact with each other (Which is very depth friendly for making complex inter-mod systems) and create optimal shortcuts (which is very depth destructive for each mod's specific balance).

If you want to keep the depth of each mod, you'd have to sacrifice it's ability to interact with other mods. You can't have the cake and eat it too- you have to balance how much to eat now and how much to leave for later.

If a mod is based with a specific power balance, it'd have to limit either it's unique power generation or it's specific machines (Though not necessarily closing it completely, just making the transition have it's own pros and cons)

 

[zilvarwolf]

Depth vs Complexity and Reward vs Punishment aren't the same thing. I might be reading too much into some of the posts I've seen here, but complex is not the same thing as punishing, and to conflate the two is probably a disservice to someone out there.

Mekanism's ore-quadrupling is (somewhat) complex, but if I don't keep up with it, I am not punished with an exploding machine or a useless chunk of rock. I'm rewarded for getting it right and staging the pieces of it to let it run smoothly. The better I stage it, the better the reward is (ie, it runs faster/longer without running out of essential material).

Tiered crafting is not inherently deep or complex either. In many cases, it's just an exercise in inventory management (really? WidgetX's don't stack? /sigh)

I don't think that punishment and tiered crafting are the best things that minecraft is capable of. I believe that's been proved many times over. Gates are deep (provide for a great many options) and complex (capable of interacting in a lot of ways with a lot of things). Ore multiplication is increasingly complex, in that it requires the interaction and coordination of an increasingly large number of moving parts, but not very deep. MFR, as a cohesive whole, is increasingly complex, but probably not very deep (well, outside the biofuel generator).

There's a lot of room in the middle of this crossroads between Squid Power (lol) and OMG-I-blinked-KABOOM. Figuring out what each person means when they talk about Depth, Complexity, Reward, and Punishment might be a good place to start trying to find it.

 

[GreenZombie]

It is worth pointing out that the MJ system of constant power draw was inherently flawed due to the way Minecraft manages performance by unloading chunks.

Unless you went big with chunkloaders - which come with their own problems, you could end up with only parts of the BC based system being loaded at any one time - which could cause a system to shut down because it runs out of fuel as the refineries are not loaded.

Chunk loading refineries just means that you run out of oil reserves while offline (if playing on a server). The tendency of RCs boilers to blow up if deprived of water means they would detonate if you piped water in from a remote location where the pump might be unloaded.

The consequences of perdition were multiple failure modes that could not be designed around except for chunk loaders. And chunkloaders have negative consequences wrt fuel longevity and lag on servers.

 

[CaptPanda]

I'm trying to get my head round this RF current, is this going to be where the RF gets voided and the machine speed is based on RF/t? Because that seems like a pretty neat implementation.

 

[Reika]

I'm trying to get my head round this RF current, is this going to be where the RF gets voided and the machine speed is based on RF/t? Because that seems like a pretty neat implementation.

That is an easy and effective way to do it. You know what? I may even make the magnetostatics do this.

 

[Golrith]

I'm trying to get my head round this RF current, is this going to be where the RF gets voided and the machine speed is based on RF/t? Because that seems like a pretty neat implementation.

TE4 machines already do this in a way. They go at a speed in relation to their internal RF buffer. Low RF/t, slower machines. Some machines also do not start running until they have a minimum RF reserve.

 

[CaptPanda]

TE4 machines already do this in a way. They go at a speed in relation to their internal RF buffer. Low RF/t, slower machines. Some machines also do not start running until they have a minimum RF reserve.

I dunno if that's so much the faster you build up the buffer, the sooner it gets started. Pretty sure it's the same processing speed, but if you have a tragically slow system, it still takes forever.

 

[Celestialphoenix]

It is worth pointing out that the MJ system of constant power draw was inherently flawed due to the way Minecraft manages performance by unloading chunks.

Unless you went big with chunkloaders - which come with their own problems, you could end up with only parts of the BC based system being loaded at any one time - which could cause a system to shut down because it runs out of fuel as the refineries are not loaded.

Chunk loading refineries just means that you run out of oil reserves while offline (if playing on a server). The tendency of RCs boilers to blow up if deprived of water means they would detonate if you piped water in from a remote location where the pump might be unloaded.

The consequences of perdition were multiple failure modes that could not be designed around except for chunk loaders. And chunkloaders have negative consequences wrt fuel longevity and lag on servers.

Chunk Loading itself causes relatively little lag (think about how many chunks a player loads compared to the chunks force-loaded/size of someones base)
-Its normally the activities running continuously in the chunks which cause the associated lag, and if thats running 24/7 with many people doing it then it kinda stacks up...

Though it would be nice to see a series of relayed/networked chunk loaders- kept offline most of the time, but when one is loaded it brings the full network online, so any random passer-bys won't inadvertently cause problems.

(either set by player ID, or a frequency/name- with a list of coordinates and dimension IDs stored in the 'master loader'. An admin configurable limit to the maximum load per network/player might come in useful)​

 

[keybounce]

...

I give you my crown, and retire from "King of the wall of text", and down-grade myself to only the Jack of the wall of text.

(And yes, I read it all; took two days.)

Now, if I actually knew enough (or anything) about RF/MJ/EU, I could comment; all I know is that I was planning to add forestry to my 164 world because one player is an ex-bee keeper, another player likes bees, and I personally wanted extra trees (and can get them with just butterflies, never touching a bee). Said plan has been sidelined by mods being improved/updated/new world gen / testing and getting things to work together / rinse and repeat / oh, look, magic farm / ag skies / jampacked / 1.7.10 beta testing / real life / etc.

(Still haven't used forestry at all)

 

[trajing]

@Reika you seem interested in the idea, should I see if @Strikingwolf and @SatanicSanta will add you to the convo?