[Monster pack + extra, 1.6.4] Reactorcraft and a strange tick issue.

  • The FTB Forum is now read-only, and is here as an archive. To participate in our community discussions, please join our Discord! https://ftb.team/discord

Paul17041993

Active Member
Mar 17, 2013
90
17
33
Australia
www.vectrobe.net
Ok just recently I started screwing around with reactorcraft, worked out how it all works and can set things up, however, I seem to have encountered a severe tick bug, in that every now and then the entire server (client server) will lockup for about ~5 seconds, but then continue as normal, so something seems like its getting caught in a loop or large process of some sort.

tried to do a quick search about it to see if it was already known, but haven't been able to find anything, nor have I been able to determine the specific component at fault, didn't seem to be steam lines or the reactor itself, however I'm not sure if it could be neutrons that have gone into "space" without dying or the steam physics.

seeing as Reika is pretty active here already it seems like a good place to start to work out what might be happening, unfortunately I don't seem to have anything in logs to provide more info, just seems to be "cant keep up" and packet overburden (when the locks occur of course).

also tried thumping the ram allocation from 4GB to 6GB to see if the server had been hitting garbage collection, but that didn't seem to change anything. permgen is 512MB as per my normal settings.

2014-05-01_213025_zpsb28ed080.png
 

Paul17041993

Active Member
Mar 17, 2013
90
17
33
Australia
www.vectrobe.net
ok bit weird, it seems a little better then before despite not changing anything, however I believe the main source of this issue is just the shear amount of chunk updates (2 per frame) happening on this system from the neutron entities and the steam blocks, move to one side to load more chunks and the packet manager gets slammed with 13K+ packets to iterate, Ill have to check later in mcedit if there's entities built up in the outer chunks.

reactorcraft version 1.6 V19b fyi, one thing that I find a problem though is both steam and neutrons have no decay time, or at least its very very long, [if you haven't already fixed this] neutrons should be 1-2 seconds and steam ~5-10, otherwise the results can be disastrous...
 

Reika

RotaryCraft Dev
FTB Mod Dev
Sep 3, 2013
5,079
5,331
550
Toronto, Canada
sites.google.com
ok bit weird, it seems a little better then before despite not changing anything, however I believe the main source of this issue is just the shear amount of chunk updates (2 per frame) happening on this system from the neutron entities and the steam blocks, move to one side to load more chunks and the packet manager gets slammed with 13K+ packets to iterate, Ill have to check later in mcedit if there's entities built up in the outer chunks.

reactorcraft version 1.6 V19b fyi, one thing that I find a problem though is both steam and neutrons have no decay time, or at least its very very long, [if you haven't already fixed this] neutrons should be 1-2 seconds and steam ~5-10, otherwise the results can be disastrous...
Your version is very out of date, so I cannot comment on its performance.

However, neutrons do not cause chunk updates unless they trigger radiation.
 

Paul17041993

Active Member
Mar 17, 2013
90
17
33
Australia
www.vectrobe.net
version is the same that the latest version of the Monster pack has, more then likely Ill manually update reactorcraft to a newer version later on in the week.

and yea, definitely found the problem, and its almost exactly what I was expecting...
tensthousandsneutrons_zps58a269fc.png


nobody wants [particles] to live forever! was this fixed in a newer version?

side note; add 4.5K to that number, I hadn't selected them all as i thought I did, more then likely the opposite chunk has 20K and each adjacent has ~5-10K.
 

Reika

RotaryCraft Dev
FTB Mod Dev
Sep 3, 2013
5,079
5,331
550
Toronto, Canada
sites.google.com
version is the same that the latest version of the Monster pack has, more then likely Ill manually update reactorcraft to a newer version later on in the week.

and yea, definitely found the problem, and its almost exactly what I was expecting...


nobody wants [particles] to live forever! was this fixed in a newer version?

side note; add 4.5K to that number, I hadn't selected them all as i thought I did, more then likely the opposite chunk has 20K and each adjacent has ~5-10K.
Yes, that would be your issue, and yes, that was fixed as of v20 or so.
 

Paul17041993

Active Member
Mar 17, 2013
90
17
33
Australia
www.vectrobe.net
ok awesome, got all your mods updated to v22 and everythings golden, were some mild ID conflicts, mainly ones that were out of the scope that the FTB team assigned, however had to leave tile#2 in expandedredstone on its default as both b1222 and b1225 were already used, corium fluid also oddly needed the still and flowing IDs swapped around for some reason, not that it'd be an issue as there hasn't been a meltdown anywhere at all.
:rolleyes:

now I just have to get the gist of using electricraft efficiently, been a long time since Ive done hardcore electronics...
would love for a transistor/potentiometer and a PWM motor to be added though, first could be redstone controlled I guess and the later would use a PWM clock gen and signal wire, otherwise just GUI controlled or even remote controlled if you wish.
voltmeter of some sort is a necessity though, atm I have no idea what V/A I'm getting, nor do I know the efficiencies of the motors and gens...
;)
 

Reika

RotaryCraft Dev
FTB Mod Dev
Sep 3, 2013
5,079
5,331
550
Toronto, Canada
sites.google.com
would love for a transistor/potentiometer
Those are two very different things. Also, they lend themselves more to integrated circuits, not high-power (multi-megawatt) usage. Also, the relay is essentially a transistor of sorts.

and a PWM motor to be added though
For what purpose?

first could be redstone controlled I guess and the later would use a PWM clock gen and signal wire, otherwise just GUI controlled or even remote controlled if you wish.
ElectriCraft has no GUIs, as it is extremely lightweight.

voltmeter of some sort is a necessity though, atm I have no idea what V/A I'm getting, nor do I know the efficiencies of the motors and gens...
;)
Use an Angular Transducer or WAILA.
 

Paul17041993

Active Member
Mar 17, 2013
90
17
33
Australia
www.vectrobe.net
Those are two very different things. Also, they lend themselves more to integrated circuits, not high-power (multi-megawatt) usage. Also, the relay is essentially a transistor of sorts.

yea I meant a kind of unit that could dynamically adjust the current flow, and yea I wouldn't expect a realistic transistor unless you wanted to go the mile and add capacitors, chokes and all that other jazz too, atm the relays are only 2-state so a similar setup involves a block of those coupled to individual or sequential resistors.

For what purpose?

PWM motors have the advantage that they keep a solid RPM based on the clock you feed it, independent of the input current/voltage, pretty much what your magnetostatic engine already does.

ElectriCraft has no GUIs, as it is extremely lightweight.

then yea, PWM clock gen (and cable, technically optional) or you could allow the engine control unit to set the RPM somehow

Use an Angular Transducer or WAILA.

hadn't actually thought of that, awesome.

oh, the other thing was transformers and maby ability to change the coil winds in the gen and/or motor, realistically these are the only ways to change the V:A ratio, and transformers at least mean you could increase the voltage for long distances to reduce power loss.
oV:eek:A is directly proportional to the wind ratio in the transformer, so would be very simple to implement code-wise, assuming of course the setup already involves the seperate V:A values.
oV = iV * ratio * effFact; oA = iA / ratio * effFact;
of course want to check the ratio is >0, otherwise your transformer is effectively just a type of choke. (and of course avoids the /0 problem)
 

Reika

RotaryCraft Dev
FTB Mod Dev
Sep 3, 2013
5,079
5,331
550
Toronto, Canada
sites.google.com
yea I meant a kind of unit that could dynamically adjust the current flow, and yea I wouldn't expect a realistic transistor unless you wanted to go the mile and add capacitors, chokes and all that other jazz too, atm the relays are only 2-state so a similar setup involves a block of those coupled to individual or sequential resistors.



PWM motors have the advantage that they keep a solid RPM based on the clock you feed it, independent of the input current/voltage, pretty much what your magnetostatic engine already does.



then yea, PWM clock gen (and cable, technically optional) or you could allow the engine control unit to set the RPM somehow



hadn't actually thought of that, awesome.

oh, the other thing was transformers and maby ability to change the coil winds in the gen and/or motor, realistically these are the only ways to change the V:A ratio, and transformers at least mean you could increase the voltage for long distances to reduce power loss.
oV:eek:A is directly proportional to the wind ratio in the transformer, so would be very simple to implement code-wise, assuming of course the setup already involves the seperate V:A values.
oV = iV * ratio * effFact; oA = iA / ratio * effFact;
of course want to check the ratio is >0, otherwise your transformer is effectively just a type of choke. (and of course avoids the /0 problem)
Given that the whole point of ElectriCraft is merely as an addon to RC to make power distribution easier, these are not really necessary, and in the case of transformers would be actively problematic.
 

Paul17041993

Active Member
Mar 17, 2013
90
17
33
Australia
www.vectrobe.net
Given that the whole point of ElectriCraft is merely as an addon to RC to make power distribution easier, these are not really necessary, and in the case of transformers would be actively problematic.

yea, just suggesting potential (semi-realistic and simple) additions that can be done in the future, main idea is they can allow flexibility without needing excessive gearbox setups etc.

almost considering making these additions myself in an add-on mod, assuming you would allow such to be done, though at the same time if I actually had the free time to do so I'd rather work on my own mods and get them at least something worth releasing, currently got academy game programming to do (physics atm) so time is very short and i may never get back to said mods, or at least not until mc1.8 comes out and/or 1.7 forge/MCP is actually something stable and not so obfuscated...
 

Paul17041993

Active Member
Mar 17, 2013
90
17
33
Australia
www.vectrobe.net
The environment temperature is applied to the fluid inside. Flammable fluids, for example, explode or ignite.

makes perfect sense that the contained fluid is affected by the environment and vice-versa, being an open-container and all, however, lubricant in an ambient temperature below zero creating a negative temperature so intense it freezes anything around it moreso then they already are, the logic there just gets a bit loopy...

not that this bothers me much mind you, I just encountered this when I was trying to use a reservoir instead of a standard BC tank, then had to quickly stop as the water for the steam engine had suddenly frozen over and refused to un-freeze, took me a while to realize it was the reservoir as I ended up making a ring around it. dont think there's anything exploitable about instantaneous infinite ice either, especially since IC2 removed fluid cooling in mc1.3.

however, it does seem to be useful in making quick houses...
2014-05-08_181027_zps96bc0aad.png


another issue now seems to be that the angular transducer clears the chat every time its used, not exactly severe but kinda annoying, been liking the all your mods so far though, its kinda like gregtech only not quite as excessive...
 

Paul17041993

Active Member
Mar 17, 2013
90
17
33
Australia
www.vectrobe.net
half pissed off a friend cause I destroyed his runic armor and harness (they don't use damage) with a bedrock sword, you don't use some hacky break enforce code do you? (if you're just doing standard apply-damage-to-player then please ignore this question and Ill report a thaumcraft bug)

other thing, you seemed to have a busted config check for electricraft platinum ore gen in v22, (simple '=' instead of '==' wouldn't surprise me, we do that all the time), however was this the "config entries being localized" bug that was fixed in v23? I don't have much time on my hands atm so I cant check myself, pack takes ~5-10 mins to load alone...

you got a bugboard anywhere? ;)
 

Padfoote

Brick Thrower
Forum Moderator
Dec 11, 2013
5,140
5,898
563
half pissed off a friend cause I destroyed his runic armor and harness (they don't use damage) with a bedrock sword, you don't use some hacky break enforce code do you? (if you're just doing standard apply-damage-to-player then please ignore this question and Ill report a thaumcraft bug)

That's a Thaumcraft thing. Reika's said something about what causes it in the past, but I can't remember exactly what it was. It also happens with MPS, as a warning.
 
  • Like
Reactions: Paul17041993