Just throwing this out there...redux on the m/c cost per MK.
I think a resource reduction would need to be combined with constant damage - which makes it the same as constant resources with increasing damage (for efficiency). Both reduced costs AND increasing damage would product too sharp an improvement.
Picking the resource cost is the problem with either constant or reducing costs. If your resource cost halves each Mark increase, you end up with (1, 2, 4, 
as your efficiency. So, you'd need something odd like 40/30/20/10 as your resource costs. Although it would produce a pleasant (1, 2, 3, 4) efficiency and a Max Damage of 468, 624, 936, 1872 million for each Mark.
You also end up with a MK V that is dirt cheap and can run forever 
No.. that wouldn't be very good. Decrease the rate at which the resources go up.. say like 1, 1.75, 2.5, 3.25, 4. So if MK I cost 100, MK II would be 175 (250,325,400 on up to V).
Oooh, reduced
increase in cost. Hmm. Let's see, the normal cost ratio is (1, 2, 4, 6,
, or
Cost_ratio | Mk 2+ = base_cost * 2 * Mk.
This can be re-written as
Cost_ratio | Mk 2+ = base_cost + 2 * X * (Mk - 1); where X is the increase, which defaults to base_cost for most ships.
Graphing out the efficiency ratios with this second equation [ F(x) = Mk / (1 + (Mk - 1) * x) ] produces a typical 1/x graph. All marks intersect at X = 1/3, f(x) = 1.5. That means, using the normal resource cost increase sequence (Mk II cost twice Mk I, Mk III twice Mk II), any step increase ratio of less than 1/3 base cost means the higher marks are less efficient. At x=1/3, all Marks 2+ are equally efficient, 1.5 times as good as the Mk I. At step increases less than 1/3, higher Marks are more efficient, but the cost increase are very small and the efficiency changes are very large... This implies that the Railpod probably shouldn't use the standard cost increase scaling.
If we do, as has been suggested above, a linear cost increase scale, we get
Cost_ratio | Mk 2+ = base_cost + X * (Mk - 1); where X is the increase.
On this efficiency chart, all marks intersect at X=1: Linear increase in cost matches linear increase in damage. Bit obvious, but a good starting point. Any increase ratio less than 1 causes higher marks to become more efficient, increase ratios greater than one are less efficient.
Again, however, there's an issue. Even as the X approaches zero, and efficiencies for higher marks go way up, the difference between higher marks, say Mk III and Mk IV, gets much smaller.
In the X = 0.75 example you suggested, costs would be (1, 1.75, 2.5, 3.25, 4) and efficiency would be (1, 1.14, 1.2, 1.23, 1.25). A good step from Mk I to Mk II, but little from Mk II to Mk III, and almost nothing from Mk III to Mk IV or Mk IV to Mk V.
This suggests that the only way to get a better efficiency increase is to reduce the increase in cost at each step. For example (1, 2, 2.75, 3.25, 3.5); a cost increase of 100% from Mk I to Mk II, but only 75% from Mk II to Mk III, and so on.
I looked at a bunch of different charts for various X values and found what I think may be a good one. (1, 1.8, 2.4, 2.8, 3.0). This gives efficiency ratios of (1, 1.08, 1.22, 1.41, 1.68).
The actual table would look like this:
(Costs round up to nearest even number for High Caps compatibility)
| Mark | Cost | Damage | Dam per Cost | Max Damage |
| 1 | 26 | 9,300 | 715 | 715,383,900 |
| 2 | 48 | 18,600 | 775 | 774,999,225 |
| 3 | 64 | 28,000 | 875 | 874,999,125 |
| 4 | 74 | 37,200 | 1005 | 1,005,404,400 |
| 5 | 78 | 46,800 | 1200 | 1,199,998,200 |
Mk I and Mk II are almost unchanged - slight improvements to Mk II, but only about 4%. Mk III is 60% better than it was, and 22% better than the Mk I. Mk IV is about twice as good as it was, and a file 41% better than the Mark I (rather than 33% worse). The Mk V is 2.5x better than it was, and 68% better than the Mk I.
The costs for Mk I and Mk II are basically unchanged. The Mk III is 35% cheaper, and the Mk IV is 50% cheaper. The Mark V is 60% cheaper. That's kind of significant - although the new Marks will still blow through Max resources in about 2-3 minutes using just MSDs. Add in enginneers and you can go broke in mere seconds, just like before.
Of course, the other suggestion that was just as good and a lot simpler was just to adjust the number of shots each Railpod can survive.
If Shots = Mark,
| Mark | Cost | Damage | Dam per Cost | Shots | Max Damage |
| 1 | 26 | 9,360 | 360 | 1 | 359,999,640 |
| 2 | 50 | 18,600 | 744 | 2 | 743,999,256 |
| 3 | 100 | 28,000 | 840 | 3 | 839,999,160 |
| 4 | 150 | 37,200 | 992 | 4 | 991,999,008 |
| 5 | 200 | 46,800 | 1170 | 5 | 1,169,998,830 |
Efficiency (1, 2.06, 2.33, 2.75, 3.25) of the new Mk I or (0.5, 1.03, 1.17, 1.38, 1.63) of the old.
This is a 50% nerf to the Mk I, but a significant increase to each of the migher Marks. Vyndicu had originally suggested having the Mk I keep 2 shots, in which case the Mk I would be unchanged.
Both of these ideas produce similar damage and efficiency charts, and both represent what I consider a good enough reason to upgrade to high Marks of Railpods.
