These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

Science & Industry

 
  • Topic is locked indefinitely.
 

PI materials lost

Author
Alexis Cato
Circle of Blue
#1 - 2012-01-02 23:05:22 UTC
Is it just me or is there something seriously wrong with PI still? I seem to lose cycles randomly on my planets when factories either fail to cycle properly or don't route their outputs as they should. For instance, if I have a setup with 6 P3 factories feeding into a P4 all the necessary inputs and sometimes one or more if the input factories will fail to deliver and the whole thing is left with incomplete materials to start the P4 cycle. It seems to happen at random times, but the end result is that if I drop in enough stuff to run a 24h cycle I will end up with 23 finished products and a P4 factory with 2/3 of its inputs filled.

At first I thought this happened only during downtime, but it seems to be entirely by chance at all times. Some days you can check a planet halfway through and everything is fine and when you return to it half a day later it will have messed up. Please, someone tell me I'm not just imagining things and that this is a known bug.
Steve Ronuken
Fuzzwork Enterprises
Vote Steve Ronuken for CSM
#2 - 2012-01-02 23:09:27 UTC
Feed /everything/ through a launchpad. Factories have no output storage, so if it tries to send to a full factory, you'll lose the materials.

launchpad -> P3 factory -> launchpad ->p4 factory -> launchpad

Should take care of it.

Woo! CSM XI!

Fuzzwork Enterprises

Twitter: @fuzzysteve on Twitter

Alexis Cato
Circle of Blue
#3 - 2012-01-02 23:53:23 UTC  |  Edited by: Alexis Cato
Steve Ronuken wrote:
Feed /everything/ through a launchpad. Factories have no output storage, so if it tries to send to a full factory, you'll lose the materials.

launchpad -> P3 factory -> launchpad ->p4 factory -> launchpad

Should take care of it.

It's just that feeding it from a launchpad to an empty P3 factory and from there to an empty P4 factory should have zero chance to fail. Since there is room for one cycle of input storage on a factory, even if the cycle from the P3s wont finish completely in synch, as soon as they all have the P4 cycle should start from the pent up inputs and clear the buffer for the next cycle. Basically, as long as I start the P3s roughly at the same time, and I do submit all the routes at once, there should never be anything in the input buffer on the P4s. My problem is that either the P3s don't send their stuff down the line or that some of them skip a cycle.

I understand that extra routing would probably solve the issue, but that is another hour to wait for the initial setup and should honestly be superfluous in my case anyway. The fact that it works in the two other identical factory rings I have on each planet and that not all planets are affected, and even the ones that are still only suffer it randomly, points to this being a bug.
Eladaris
Indefinite.
#4 - 2012-01-03 00:02:22 UTC
Alexis Cato wrote:
I understand that extra routing would probably solve the issue, but that is another hour to wait for the initial setup and should honestly be superfluous in my case anyway. The fact that it works in the two other identical factory rings I have on each planet and that not all planets are affected, and even the ones that are still only suffer it randomly, points to this being a bug.


In essence a bug that's been around since day one has a clearly defined work around that resolves the issue perfectly, but you don't want to do it because it adds some time to your setup the first time.

I'm ~fairly~ certain CCP will look at PI someday... but probably not SOON (tm)... You should probably just re-route things and call it a day.
Droxlyn
School of Applied Knowledge
Caldari State
#5 - 2012-01-03 01:58:35 UTC
Alexis Cato wrote:
Steve Ronuken wrote:
Feed /everything/ through a launchpad. Factories have no output storage, so if it tries to send to a full factory, you'll lose the materials.

launchpad -> P3 factory -> launchpad ->p4 factory -> launchpad

Should take care of it.

It's just that feeding it from a launchpad to an empty P3 factory and from there to an empty P4 factory should have zero chance to fail. Since there is room for one cycle of input storage on a factory, even if the cycle from the P3s wont finish completely in synch, as soon as they all have the P4 cycle should start from the pent up inputs and clear the buffer for the next cycle. Basically, as long as I start the P3s roughly at the same time, and I do submit all the routes at once, there should never be anything in the input buffer on the P4s. My problem is that either the P3s don't send their stuff down the line or that some of them skip a cycle.

I understand that extra routing would probably solve the issue, but that is another hour to wait for the initial setup and should honestly be superfluous in my case anyway. The fact that it works in the two other identical factory rings I have on each planet and that not all planets are affected, and even the ones that are still only suffer it randomly, points to this being a bug.


You can route from the "route" tab of the space port, so take the incoming route item, make a new route to the P4 factory. No time wasted. I just wish I could create a from-route in a similar manner so no materials have to be on the planet before starting. (Double click the incoming route and work from there.)
Humphreys Stig
Aideron Robotics
Aideron Robotics.
#6 - 2012-01-03 02:16:48 UTC
I've linked my thread on this issue.

https://forums.eveonline.com/default.aspx?g=posts&find=unread&t=49209

The response from the GM's is the usual everything is working as intended and we should talk to Game Dev's.

It appears that if you set up a task in PI without confirming an output destination, then you lose all the output till you realise the error. This will continue till either your cycles finish or there is no new materials produced.

And yes IMO the game mechanics should be changed to prevent this from happening.
Alexis Cato
Circle of Blue
#7 - 2012-01-03 10:34:35 UTC
Eladaris wrote:
In essence a bug that's been around since day one has a clearly defined work around that resolves the issue perfectly, but you don't want to do it because it adds some time to your setup the first time.

Not at all, I'm fine with changing my setup, I just wanted to know if this was a known bug with a fix in the works or not. Droxlyn's tip of routing before the materials cycle will suffice for now.