HI apparently not getting resources

[erstad]

OK, here's some data and save files that I think may show a problem with resources not automoving where needed in the home islands. Since the game doesn't report what did/didn't produce, it's hard to prove what did or didn't produce but I'm seeing intermittent downward spikes in the rough vicinity of 5000, which is consistent with a major center not producing.

First, here's the witp staff HI tracking, the green curve goes with the axis on the right and shows the change in HI. There are occasional downward spikes of great magnitude. (Note the spikes are more similar in magnitude than they appear on the graph; the quasi-log scale used by Witp Staff makes (for example) a +100 to -4900 spike look far bigger than a -1100 to -6100 spike even thought the magnitude is the same.)

EDIT: I just noticed this turn does cross a month boundary. I vaguely recall some kind of post that the pilot training used something on a monthly basis. Was that HI, and if so what order of magnitude does training consume? A number of the spikes could be on month end boundaries so perhaps that's part of the explanation, but not all, there's at least one that is clearly not. But i would have to investigate to make sure I didn't do a ton on factory expansion or conversion or whatever on that turn.

[erstad]

Here's my save before sending the turn to my opponent. I'm not going to post my HI balance, but call it "X"

[erstad]

Here's the turn that came back from my opponent

[erstad]

Here's the turn after execution. My HI has gone from "X" to about "X - 10,000". I had been running a slight (maybe 1000) deficit on surrounding turns according to staff, this turn sequence was markedly different and there were no major changes to HI consumers (maybe a couple of factory points repaired, plus the usual randoms).

It's possible I'm crazy, but I think the evidence suggests the spikes are major centers in Japan (possibly Tokyo) not producing, and I assume the likely culprit is they failed to pull in the needed resources even though they are present in the HI (I've slipped slightly from game start, but still have something like a 25 day buffer).

This is 2 day turns.

BTW, I'm less certain of this, but I think part of the reason I had been running a small deficit is that I'm not sure the inland HI centers in manchuria are getting the fuel they need. If you guys have or insert some debug code to dump which centers produced and which didn't, you might take a look at those too.

I think I heard that some of the resource type movement may occur on a multi day cycle pattern, if previous saves are needed, I have them back to 12/7 (Well, excluding the one I accidentally overwrote)

[erstad]

If you read this thread before you might want to read my edit to the first post.

[Swenslim]

Problem with Tokyo, in 3-4 turn intervals it runs out of resources. So you must ship 90% all resources to Tokyo or Yokosuka.

And set few 30 000 convoys between Shimonseki and Tokyo.

[erstad]

ORIGINAL: Swenslim

Problem with Tokyo, in 3-4 turn intervals it runs out of resources. So you must ship 90% all resources to Tokyo or Yokosuka.

And set few 30 000 convoys between Shimonseki and Tokyo.

Maybe. I was pretty sure before I remembered the pilot thing, now I'm not so sure. Remember, there's no way to see the resources that are in Tokyo during the production phase. You can only see what's there afterwards. Have you actually seen Tokyo go away based on the HI production numbers (should be about a 5500 dip).

I can say for sure I don't see it on 3-4 turn intervals, from the WitpStaff graph in the initial post there are certainly some 4 digit dips but nowhere near as frequent as 3-4 turns.

I am still mighty suspicious of whether the fuel is getting to the inland bases. Someone (I forget who) posted what looked like a well researched analysis that showed his HI consistently coming up short by the value of those centers.

[Swenslim]

My Witp Tracker shows for example - major alert - Tokyo HI \ LI insufficient 19 000 resources......

[erstad]

ORIGINAL: Swenslim

My Witp Tracker shows for example - major alert - Tokyo HI \ LI insufficient 19 000 resources......

Right, but remember that WitpTracker is also looking at the turn after production has occurred.

So it doesn't prove the case either way, because there will be an Oil/Resource movement phase before the next production phase. Do you see a 5500 point HI drop (versus your normal HI movement)?

[n01487477]

ORIGINAL: erstad

ORIGINAL: Swenslim

My Witp Tracker shows for example - major alert - Tokyo HI \ LI insufficient 19 000 resources......

Right, but remember that WitpTracker is also looking at the turn after production has occurred.

So it doesn't prove the case either way, because there will be an Oil/Resource movement phase before the next production phase. Do you see a 5500 point HI drop (versus your normal HI movement)?

This is totally correct regarding Tracker & Staff ... and I've seen and been involved in other threads about this problem.
I have a feeling that it is being looked at tm.asp?m=2248064 ... and from previous postings prior to release, there was no way to completely test the complexity of the economy - and is being done now post release. But I'm no Dev.
I can confirm, once again the pilot-training uses HI at the beginning of the month, and am going to include that in the Tracker-Chart soon.
I still think there are accounting issues not yet revealed regarding HI production and use and until I get a handle on that, I can't give more than you guys talking about fluctuations in HI, post turn & Base HI production.

[HistoryGuy]

This thread makes me feel a BIT better as I wondered if I were doing something wrong.  I have taken to using human controlled resource convoys to Osaka, Nagoya, Tokyo, and Nagasaki just to help the auto mode a bit............seems to be working.