Fire Delay?

[ndkid (OP)]

I'm not posting this in a bug thread in part because I'm not entirely sure the problem wasn't me.

So, there was a group of Precursors sitting in a system. I built a jumpgate into it. Once it was complete, I ordered my main fleet through it and to a local planet in the system. ETA: 22 days. I clicked "30 days", figuring that if they met the enemy before they got to the planet, Aurora would stop moving time forward and let me know.

And so it did... with the enemy at 0-distance. That, in itself, didn't completely dismay me... I had set my shields to start charging when I entered the system, and I was *far* into the system when Aurora stopped time, so I wasn't caught with my pants down.

So, question 1: why didn't sim-ing stop when the enemy contacts were at the edge of my sensor range (1M km with this fleet), rather than at 0km? My fleet moves at 10k/s; intelligence tells me the enemy ships move at around 4000k, so it's not a matter of them being able to close that 1M k distance in a single tick. I've never seen a cloak used before, so maybe it was that... would there be a special message when enemies appear out of nowhere?

So, here I am, out-tonned but probably not outgunned. I set up all my ships lasers to target the enemy contacts, giggling to myself as I imagine the sheer volume of damage that many large-caliber lasers will do at point-blank. And I click the 5s button.

I see one of my two cruisers go up under a barrage of missle fire. Acceptable loses, I think, as I scroll down to see what my counter-volley looked like.

All of my ships are suffering firing delay. Even though we are so far from our entry jumppoint that people have probably been able to have a full night's sleep since the jump, my ships got picked off one-by-one for a good 6-8 turns from fire delay, despite their 34% across-the-board crew grade.

Question 2: Is there something other than jumping that causes fire delay? Because, if not, I think this looks like a bug.

I've seen variants of this before; if I have an enemy at long range, and a 1M res 0 sensor, and I jump forward 5 minutes or something, I will often come back from the time jump at the point that the missles are about to smack me, rather than when the missles first enter my active sensor range. Is it a known problem that if you attempt to move forward large periods of time in potentially hostile situations, the simming probably won't stop as early as you might expect?

[Andrew]

You almost certainly have untrained crews and the inexperienced crew penalty turned on so you crews are being delayed in carrying out the orders.
The lack of warning sometimes happens particularly if you already had the enemy detected when you started. Do not use 30 day turns when you expect trouble (the same thing can happen if you do a 1 hour update at close range with enemy ships closing right through your fire envelope)

[sloanjh]

I'm not posting this in a bug thread in part because I'm not entirely sure the problem wasn't me.

Good instinct  On the other hand, the right place for a "did I mess up"/"new user" thread (like this one) is The Academy - see the "Where should I post?" FAQ viewtopic.php?f=100&t=2043

So, there was a group of Precursors sitting in a system. I built a jumpgate into it. Once it was complete, I ordered my main fleet through it and to a local planet in the system. ETA: 22 days. I clicked "30 days", figuring that if they met the enemy before they got to the planet, Aurora would stop moving time forward and let me know.

And so it did... with the enemy at 0-distance. That, in itself, didn't completely dismay me... I had set my shields to start charging when I entered the system, and I was *far* into the system when Aurora stopped time, so I wasn't caught with my pants down.

So, question 1: why didn't sim-ing stop when the enemy contacts were at the edge of my sensor range (1M km with this fleet), rather than at 0km? My fleet moves at 10k/s; intelligence tells me the enemy ships move at around 4000k, so it's not a matter of them being able to close that 1M k distance in a single tick. I've never seen a cloak used before, so maybe it was that... would there be a special message when enemies appear out of nowhere?

Because it would take a week of clock time to run the subpulses.  See the part 12 tutorial "Aurora's Time Increment System" viewtopic.php?f=101&t=2060 for more information.  There are also a ton of long discussions in various threads while all this stuff was being worked out - searching for "timestep" or "increment" will probably take you to a lot of them.

The quick answer, though, is that Aurora will typically divide a move increment into ~50 (I think) subpulses.  So if you use a 30 day increment, each subpulse will be 1/2 day.  So you moved real close to the bad guys and then they moved on top of you (probably), all in the "same" subpulse (or vice-versa).  To put it another day, a single tick is probably ~1/2 day (or longer) when you use a 30 day increment.

There are two things to do in this case:

1)  As Andrew suggests, take smaller steps when you expect an encounter (for example when scouting a new system).
2)  Before you take each step, save a copy of your DB as discussed in this viewtopic.php?f=100&t=2124 FAQ thread.  That way you can "replay" the approach to the encounter by going back to the old DB, using the same movement orders, but using fairly small (e.g. 1 hour) increments when you're within an hour or so of when the encounter will take place.

So, here I am, out-tonned but probably not outgunned. I set up all my ships lasers to target the enemy contacts, giggling to myself as I imagine the sheer volume of damage that many large-caliber lasers will do at point-blank. And I click the 5s button.

I see one of my two cruisers go up under a barrage of missle fire. Acceptable loses, I think, as I scroll down to see what my counter-volley looked like.

All of my ships are suffering firing delay. Even though we are so far from our entry jumppoint that people have probably been able to have a full night's sleep since the jump, my ships got picked off one-by-one for a good 6-8 turns from fire delay, despite their 34% across-the-board crew grade.

Question 2: Is there something other than jumping that causes fire delay? Because, if not, I think this looks like a bug.

As Andrew said, "yes".  This is discussed in the following FAQ thread: viewtopic.php?f=100&t=2247

I've seen variants of this before; if I have an enemy at long range, and a 1M res 0 sensor, and I jump forward 5 minutes or something, I will often come back from the time jump at the point that the missles are about to smack me, rather than when the missles first enter my active sensor range. Is it a known problem that if you attempt to move forward large periods of time in potentially hostile situations, the simming probably won't stop as early as you might expect?

At this point I would call it a feature rather than a problem, but yes.  The reason I say feature is that Steve's done a ton of work to try to reduce the problem, and at this point it's fairly managable if you use the techniques outlined above.

One more comment.  I never use 30 day increments, and if I were to do so, I would only do it for a conventional start early in the game.  1-5 day updates are what the game is tuned for, so you can expect "coarse-grained" behavior if you try to do 30 day updates.

John

John

[ndkid (OP)]

Huh. every ship in the fleet had a crew grade of 34% and a TF score of 75%-100%, so I was rather surprised that the series of unfortunate events fell out the way it did. But if that's WAD, that's WAD.

[Charlie Beeler]

If I recall correctly (I've been wrong before), for the interupt events too work correctly automated turn must be active.  Without it surprises like enemy ships at range 0km occur.

Firing delay:  Just because you've been insystem for x amount of time has nothing to do with whether a firing delay, much less how long of a delay, is applied.  That is triggered by issuing a firing order.  Since the task force training rate was only 75% there is still a delay in firing orders.