FreeOrion

I've noticed that Robotic Interface Shields and regular shields stack.
Is that intentional?

No, the cross interference is supposed to be negating stuff.

Can you upload a savegame of that specific point, setting up to test will be a PITA anyway.

Sorry, forgot to mention I'm not playing the most bleeding edge version.
I was playing one of the earlier RC's
FreeOrion_v0.4.6-RC1_2016-09-09.3058052_Win32_Setup.exe

I'll try and get a save game up.
May take awhile.

Only other shields look to be deducted from the total applied from Robotic Interface.

If this ship did not have a Defense Grid equipped, it should have the same total shield strength (in the same scenario).
I assumed this was all intentional, allowing a more expensive ship with a little more versatility (it can leave a stack and maintain some shield strength).

Fleet of Nano-Robotic hull ships with just Robotic Interface shields.
Everything normal. 44 ships getting a shield bonus of 17.3
Same turn. New fleet of Self Gravitating (SG from now on) hull with Robotic interface AND Blackshield.
Robotic species crew. Everything normal as it should be here too.
No bonus from Robotic Interface.
Now I have moved the two fleets into the same system.
The Nano-Robotic ships are getting a capped +20 bonus from the 62 ships present.
Still everything normal
Continued in next post. can only attach 3 files.
Ships have Blackshield now and not Defense Grid because I went back and reloaded the game and built the SG ships with Blackshield instead. I wanted to see if I could exceed the +20 shield cap.
As you can see I couldn't, but that is the only good news here.

Same turn. SG fleet at same location as Nano-Robotic fleet.
Ships are getting the +15 from Blackshields, but also getting +5 from the network.
Not sure if this is normal or not. Probably is.
Turn 333. I've moved the SG fleet away.
All is as it should be
Turn 333 still. Things still normal

Turn 334. Ships back in same system
All is normal.
On turn 335 I gave orders to move the SG fleet to Starn and the Nano-Robotic fleet to the empty system
Turn 336 the SG fleet is on its way to Starn.
This is where things start to get weird.
The SG fleet still retains the +5 network bonus even though they are no longer in the same system as the Nano-Robotic fleet.
Sill turn 336. I have built more SG ships at Starn. No Nano-Robotic ships present, but theses ships are getting a network bonus

Turn 337. Previous SG fleet still retains its network bonus although not near Nano-Robotic fleet
Still turn 337. I've built more SG ships at Starn
Now things start getting really weird.
Ships are getting Blackshield bonus plus network bonus, but now there is some weird -3.74 from Unknown.
Other SG fleet at Starn. Its Unknown minus is only 3.52 for some reason even though there are the same number of ships in this fleet as the previous (19)

Still turn 337. More weirdness
Original Nano-Robotic fleet still has +20 network bonus even though we know from previous that 44 ships only produces a bonus of +17.3
Turn 338
All SG ships back at Starn.
Still getting network bonus, but weird Unknown minus gone.

Well, something is definitely odd here. I've never used the Robotic Interface Shields, so I probably don't understand correctly how they work, but what are those two different boni anyway? I see an entry for "Robotic Interface: Shields" and another, separate entry "Robotic crew network". What is the difference between the first and the second? Judging only by its name, it wouldn't be that odd that "Robotic crew network" gives a bonus all the time, because all these ships are robotic ships manned by a robotic crew (Etty)...

I have no idea.
This is the first time I tried playing around with them too.

Vezzra wrote:what are those two different boni anyway?

There is a value for the capacity of the part, this will always be 0 with the current entry.
The other is the sum for the 'network' effect, dependent of number of acceptable ships within 0 jumps (not just in the same system).

Nice work with the detailed report EricF.
The ships that are less than 1 jump from Starn will count towards each other, so normal behavior there (at least how I interpret the entry).
No idea about the unknown malus here.
The stack of 44 receiving 20 for network crew looks like an issue, ((43 * 7) ^ 0.5) = ~17.35

dbenage-cx wrote:

Vezzra wrote:what are those two different boni anyway?

There is a value for the capacity of the part, this will always be 0 with the current entry.
The other is the sum for the 'network' effect, dependent of number of acceptable ships within 0 jumps (not just in the same system).

Nice work with the detailed report EricF.
The ships that are less than 1 jump from Starn will count towards each other, so normal behavior there (at least how I interpret the entry).
No idea about the unknown malus here.
The stack of 44 receiving 20 for network crew looks like an issue, ((43 * 7) ^ 0.5) = ~17.35

Maybe the fleet heading to Starn is still "Close enough" to the Nano-Robotic fleet to give them the +20 instead of +17.3
But that still doesn't explain why the SG fleet originally does not have a +5 Network bonus, but later having it.
Unless that 'distance to' is not Zero like it's supposed to be...

Right, it's working as intended but not displaying as intended.

With the number of ships present in the blackshield pic, you would be getting the max (currently +20, I never rebalanced them down), the blackshield gives you +15 so the RIS gives you an extra 5, without the BS you'd get +20 from the RIS.

The interference label is meant to show you -15 and the RIS giving you +20, but it's not.

Regarding this as something to fix: the RIS and the current shield macros were done before we had effects priority and they're coded incredibly complexly in order to both give the correct result and ensure the AI gets the correct reports of stats (IIRC Morlic was involved in the macros). I am fairly sure that using priorities the code could be cleaned up substantially, made easier to read and work correctly with correct displays now that effect priorities are possible.

However, doing so is beyond my ability.

NB: If we need to rework RIS from scratch to give a similar, but different, effect in order for it to display correctly and play well with the other parts I'd be happy for that.

MatGB wrote: Regarding this as something to fix: the RIS and the current shield macros were done before we had effects priority and they're coded incredibly complexly in order to both give the correct result and ensure the AI gets the correct reports of stats (IIRC Morlic was involved in the macros).

Yes, that was my work.
I suppose nowadays we could use NoDefaultCapacityEffect to get rid of the subtracting of all the shields capacities. Then we order normal shield priority by strength so the strongest shield is used first and then use a stacking group to make sure no other shields are evaluated?

No spontaneous idea how to simplify RIS though.