PDA

View Full Version : [TEC] Anyone grok random groups?


tater
07-02-07, 10:56 AM
OK, I had a great patrol last night. Was sent from Surabaya to drop off some supplies to the philipines. Having accomplished that (I sank a lone, small frieghter on the way), I decided to patrol to Palau. I spent about a week there dodging the patrols around the islands. Set up an attack on a DD, fired 2 fish min depth, they both ran under. Got DCed for my trouble.

Kept driving around on the Palau-Davao route. Watch calls out a contact, it's an escort turning away. I assume they are heading for Palau, make flank north for a few minutes, then dive to PD making std. Watch hydrophone, I have a ship. I'm keeping the scope down because I have AI set up so they can actually see the scope.

I decide I need to take a peek. Up scope. I left it on high mag, and wow, a CA fills the scope. Look left, 2 BBs, a Kongo and Fuso, followed by a CVE, and more (cruisers of some kind) in line astern. Escorts are out on the flanks. I let Maya pass, and fire all 4 tubes at Kongo in a spread. 1 clean miss, 1 premature, 2 hits, but one is a dud. Not bad for manual TDC and hardcore torpedos. Down scope, turn to parallel their heading at flank to get the stern tubes in action. Up scope. Loads of smoke from BB!. PK is still on, and with a little correction I fire 4 more fish (I'm next to Kongo about 1000 yards off her starboard side). 2 prematures, 1 miss, 1 hit. I make a hole in the water and go deep. 3 DDs keep me busy while the TF steams away at 16 knots. Kongo sinks (damn "instant success" messege).

What does this have to do with random groups, you ask?

There is one TF instance that can be where I was (I made it) at that time..
The group indeed has Kongo, in fact 2xKongo with a 15% chance of occurence. Why was there only ONE Kongo? If you set 2xKongo at 15%, does that mean that it does two checks for individual Kongo at 15%? I assumed there was a 15% chance of 2 appearing, and a 0% chance of 1 appearing.

The CVE was also set to 2xTaiyo @ 30%, this instance had ONE.

So what's the deal with random units and "No=_" in the mission file?

Edit: I just noticed that the leader entry for that TF is 2xKongo. From my previous experiments I determined that the leader is ALWAYS there, regardless of % chance. So I assume that in the case of an NxShip entry where "Ship" is the leader, then 1xShip is a 100% occurence, and the total "N" amount works based on the % chance. OK, that makes sense for Kongo in my example above. What about the 2xTaiyo @30% though. I still only had one in the group, I'd have thought I'd see 2 or none...

tater

tater
07-02-07, 01:51 PM
OK, random groups work nothing like the way I would EVER have expected.

5xGENERIC tanker 50%

I assume most of you would have thought that this was a way of setting 5 tankers to appear, or to NOT appear in a group. 50% chance of 0 tankers, 50% chance of 5 together. No chance at all of anything but 0 or 5.

What this does in game? It's 5 separate 50% chances of a single tanker.

I had some random groups where I had a carrier division, but a low probability. I wanted a 10% chance that a given TF would have TWO Hiryu CVs. I wanted a ZERO % chance of 1 CV, if I saw one in this group, I wanted to see two. Turns out this is not possible.

So, you can make an entry like 10 X GENERIC cargo at 10%, and you might see none, or 1, or 2, or 3, ..., or 10.

I can still get the 2xCV groups I wanted, just make a new random TF, set them to 100%, and reduce the spawn probability of that TF to cover the % I wanted. Still, I doubt anyone would have read the Nxtype @ X% as being individual chances. Particularly since they have escorts in most TFs set up as a bunch of 80% chances of single generic DDs.

Useful stuff to know, know if I could only understand how the game decides to arrange them in formation...

Sailor Steve
07-02-07, 03:30 PM
"Once upon a time there was a Martian..."