Oil/res. shrinkage during overland movement

[VSWG]

I did a series of tests to determine how many oil/res. points are lost when moving from one base to another. The formula is actually pretty simple:

Supply Path Value (SPV) / 5 = % of oil/res. lost. (rounded up)

To obtain the SPV see p. 188 of the manual:

When tracing a supply path for movement of ground units or overland movement of supplies, a
supply value is generated. This is determined by tracing a path from the base transferring
supplies or the ground unit moving to the destination. This supply value determines whether the
move is legal, and how much of the supplies are used up during the move. The supply value of a
move is calculated by subtracting the following from 100 (900 if a ground unit move):

• 2 for each hex moved along a rail/highway
• 5 for each hex moved along a road
• 25 for each hex moved along a trail
• 50 for each hex moved cross country

Example: Base 1 and Base 2 are connected by 5 railroad hexes. The SPV is 10, and 2% of all oil/res. points moving from base 1 to base 2 will be lost to shrinkage.

However, every turn there's a chance that oil/res. points will not move at all, and this chance increases with SPV. A SPV below 20 almost never causes oil/res. movement to fail, but with SPVs larger than 50 oil/res. movement will fail frequently. A SPV larger than 70 is a guarantee that oil/res. will not appear in sufficent numbers at a base that needs raw materiel. See post #2 for screenshots of some test runs with different SPVs.

Shrinkage during port-to-port movement: A different set of tests showed that adjacent bases separated by ocean hexsides must have at least level 3 ports for port-to-port movement to occur. Shrinkage during port-to-port movement is always 20% of the oil/res. involved. Port sizes larger than 3 have no effect on the amount of shrinkage; it will be 20% between level 9 ports, too.

How many oil/res. will move from base to base? The AI will attempt to draw oil/res. from other bases as soon as the stockpiles drop below 2xHI. The base will then attempt to draw enough oil/res. to keep the HI running for 16 days +25, but the AI doesn't take shrinkage into account, so fewer oil/res. points will actually arrive. The amount of oil/res. still present at the base is multiplied by 4 and substracted from this number.

Example 1: A base with 1000 HI and 0 oil/res. will attempt to draw 16x1000+25 = 16,025 oil/res.
Example 2: A base with 1000 HI and 1500 oil/res. will attempt to draw 16x1000-(4x1500)+25 = 10,025 oil/res.

I don't know why raw materiel already present is multiplied by 4, nor where the 25 comes from.

Bases with less than 1000 oil/res. will never give them away. However, as soon as the stockpiles are above 1000, all oil/res. points can move to other bases.

Example: A base creating 100 res. points every turn will give them away only every 11th turn, but then all res. points can move to other bases.

[VSWG]

Test Scenario:

I used CHS 2.08 for all my tests. To create a clinically clean environment for my test runs I removed all ships, LCUs, air groups and industry from Australia, New Zealand, New Guinea, New Britain and New Caledonia. Nothing but the bases remained. Thus I was able to run several tests at the same time. The standard test involves 2000 res./oil produced automatically (daily oil/res., not produced by resource or oil centers) at one base, and 2000 HI at a second base. This ensured that the 1000 oil/res. threshold is ignored, and the AI has to move oil/res. every turn to keep the HI running. On day 1 2000 oil/res. points are already present at the producing base.

Tests 1 and 2 deal with railoads and extremly small SPVs. They simulate conditions that are similar to these on Japan. Shrinkage is negligible, and not once did the overland movement fail altogether:





Test 3 is the same as 2, but with 3000 daily oil/res. and 3000 HI to check if increasing/decreasing the numbers involved would effect the formula. It didn't.

Test 4: A small road network. Here I had to separate the numbers for oil and resources for the first time, as the checks for overland movement are made separately. Oil failed to move twice (on the 15th and 22nd, the numbers are in a (barely noticeable) red), and res. once (on the 12th). The HI at Noumea couldn't work on the 15th:



Test 5: Two trail hexes cause significant shrinkage (10%), but the movement failed only 4 times:



Test 6: Three trail hexes, and the HI base never produced a single fuel/supply point:



Test 7: Same results for an extremly long rail line:

[USSAmerica]

Very nice work!

So it appears that for port - port oil/resorce movement, there is never a check performed. It will move whenever the source has more than 1000 oil or resources. Of course, assuming all the other requirements are met. Is this correct?

[VSWG]

ORIGINAL: USS America

Very nice work!

Thanks!

So it appears that for port - port oil/resorce movement, there is never a check performed. It will move whenever the source has more than 1000 oil or resources. Of course, assuming all the other requirements are met. Is this correct?

Yes, I think so. Port-to-port movement was the first thing I tested, and these early tests were rather inefficient (few oil/res. movements in a couple of turns). So this data isn't really on a firm footing, but I haven't seen port-to-port movement fail if more than 1000 oil/res. are present at the producing base so far. I'll add another port-to-port movement test to the to-do list.

[USSAmerica]

VSWG, I think this topic would be a great addition to the "Must Read Threads" thread.
 
(That would be easier for my lazy self than taking the time to copy all your hard work into a local file on my computer [;)])

[JoePirulo]

I agree with you USS America. This is a must read thread... otherways you´ll think that with a couple of tankers from Palembang to Singapore your production life is solved... [8|]

[Feinder]

(* remembers poor George Costanza *)
 
-F-

[VSWG]

ORIGINAL: Feinder

(* remembers poor George Costanza *)

-F-

[:D][:D][:D]

I called it "spoilage" at first, but some people confused it with spoilage at bases due to small port/airfield sizes. I'm open for suggestions, though. [;)]

Added to the Must Read List. [:)]

[Yamato hugger]

Some people have entirely too much time on their hands.

(Have to admit, it is nice work)

[Terminus]

Agreed...

"It shrinks? Like a frightened turtle!"