Optimizing research - MoO1

[Daemon]

Hey guys.

I'm a long-time MoO1 player and i have played countless games, yet i do not know what that 'optimize research' concept is supposed to mean. I'm using kyrub latest patch and i would appreciate someone explaining to me how come that maxing out a research field does not produce the best results, and how should RP allocation be handled.

Thanks a million.

P.S. I have tried to decipher this on my own but failed miserably.

P.P.S. I've got the bug again and have been playing around the clock on my android tablet and android phone using DosBox Turbo app. It works great with absolute position mouse tracking, so wanted to share that w/ you too.

[haphazard1]

Welcome to the RB MoO forum, Daemon! Always great to see new people posting here.

Someone like Catwalk would be a better choice to explain the whole optimized research thing -- he seems to be a big proponent of the idea. I am rather more casual about it -- I don't really enjoy breaking out a spreadsheet to squeeze out every possible RP, as has been noted in my game report a while back.

The general idea is that you get an additional research bonus on a small fraction (7.5% if I am recalling correctly) of the RP already invested in a field. So if you invest an initial slug of RP into a field, and then feed in a steadily growing trickle of RP after that, you get the overall optimal results for total RP invested. This is most useful early in the game when you have relatively small amounts of production to invest in research, as the optimizing process requires time to produce the best returns since you spread a small amount of spending over more turns to get the bonus on all your spending every turn.

Anyway, as I said someone like Catwalk can probably explain it better.

[Daemon]

Hey it's my 5th post . Thanks anyway.

I will take any explanation that makes my light bulb light up .

After reading it the 5th time, i think i'm getting the basic idea.

If i spend 100 RP initially, i get a bonus of 7.5 RP. Is that a 'per turn' bonus? Like if i reduce the RP spent to, say, 5, i will be getting 5+7.5 RP each turn for the rest of the research process? Is this bonus applied only to the initial amount of RP allocated to 'year 0' of that research? And does it ever 'wear out'?

So that would (ideally) translate in having to allocate as much RP as possible to each of the new research projects, then dropping it to some meager amount, finishing the research in another field to get a new project there as well, and after all 6 fields have new "optimized" projects, i get 6x7.5%=45% more RP each round if i just even them out?

Hmm...

If i get this right, the 1st opportunity to apply this trick is to max out 1 of the 6 tech lines as soon as i start a new game, and never get any research rolling until i got a reasonable amount to feed off afterwards, right?

[haphazard1]

Well, 5 posts is still fairly new around here.

OK, I found an example posted by Catwalk in my game thread a while back. The key is that research spending equal to 7.5% of the total already invested in the current tech gets tripled. Any spending beyond 7.5% of the total already invested does not get any bonus. So for max optimization you want to always be spending 7.5% of what you have so far put into a tech, meaning your spending needs to creep up steadily each turn.

Here is Catwalk's example:

In order to optimize research, you need to be at the optimal research level (7.5% of current total points invested) and then increase your research by 22.5% each turn. Example of early Planetology research:
- Open the field with 2 RP (you lose 10% on an unopened field)
- Invest 39 RP
- Invest 3 RP => 49 RP at EOT, 3 is still optimal
- Invest 3 RP => 58 RP at EOT, 4 is optimal
- Invest 4 RP => 70 RP at EOT, 5 is optimal
- Invest 5 RP => 85 RP at EOT, 6 is optimal
- 100 => 7
- 121 => 9
- 148 => 11
- 181 => 13
- 220 => 16
- 268 => 20
- 328 => 24
- 400 => 30
- 490 => 36
- 598 => 44
- 730 => 54
- 892 => 66
- 1090

This is quite doable throughout the game if you research one tech at a time. For multiple techs, the easiest thing is to leave it at one tick once it has reached 10% or whatever number you're comfortable with (low percentages have good payoffs unless you need it urgently). Then either multiply your research by 0.98 or simply ignore that last 2%.

I think the first turn he lists is off by one (3 spent with 39 total invested should produce 9, for 48 the next turn rather than 49). But otherwise the example shows the progression. You effectively get three times the research for the same production invested. But it takes time to scale up unless you can invest a quite sizable slug at the start, so if you need a tech urgently you may just want to pour on the spending.

[Catwalk]

In that example I open the field with 2 points, of which 1 point is lost. So after investing another 39 points you will be at 40 which is the 3-point threshold.

Whether to do this or not is primarily a matter of whether fiddling around with numbers appeals to you or not. If not, there is absolutely no reason to be doing this as it does require a fair bit of fiddling. If it does appeal to you, there are real benefits to be had.

The best way to do it (IMO) is to make a large investment when starting on a tech and then riding max bonus all the way to completion. How much to put in up front depends on how fast you need the tech:

2t: 81.6%
3t: 66.6%
4t: 54.4%
5t: 44.4%
6t: 36.25%
7t: 29.6%
8t: 24.2%
9t: 19.7%
10t: 16.1%

Note that this is how many turns it takes to reach the base cost of the tech, after which you need to start accumulating percentages and get lucky. Adding another 4t to the above numbers is a reasonable estimate. Note that you should scale down investments once you hit percentages, it's usually inefficient to keep researching full-bore after you get a 20% chance for example (unless you need it right now or you detest any kind of gambling).

So if you want a tech in around 14t, invest ~1/6 of the tech cost up front, switch to 7.5% of the invested amount the following turn and then increase your research by 22.5% each turn.

As for how to actually implement this method, it's tricky. Whenever I'm serious about it I use the editor that gives me the research values, it's very impractical to keep multiple fields running this way otherwise. kyrub's patch does make things a lot easier if you just want to be roughly on track, as the diode lights up when you reach the amount for max bonus. It is easiest to do early on with just a single field at a time, I recommend whipping out a table of numbers that you then simply follow each turn.

As for how it works, here is a simple explanation: The first 7.5% of your CURRENT research points invested in a field count triple. If you have invested 200 RP in a field, the first 15 BC invested will give you 45 RP. After that it's 1:1 so 20 BC will give you 50 RP.

That said, I do consider this a flawed concept for two reasons:
1) It shouldn't be this much hassle using the system properly. If the bonus was calculated based on the tech cost rather than the amount invested, it'd be so much easier to keep track of.
2) I don't like how the bonus cuts off sharply after 7.5%. Ideally, it should be something like x5 for the first 1%, x4 for the next 2%, x3 for the next 3% and x2 for the next 4%. That will give a combined bonus of x3 at 10% invested, close to what the current system gives. You wouldn't need to pay close attention to your exact spending, nor would you need in-depth knowledge about how the system worked. If you crash fund a research project you pay more for it, if you abort funding you pay more for it and if you keep a steady flow of research you profit. That's a simple mechanic that can easily be used by anyone.

[Daemon]

I think i finally got it. Thanks for the explanation! To summ it up: follow the light! :P

[Daemon]

I have to say that this has helped a lot. I've played a couple dozen games using the optimized research plan and it pays off.

On an unrelated subject, take a look at these poor Klackons, owning half of a medium galaxy with impenetrable (at the time) planets, full of waste (30-50m out of normal 80-100m ones :P). No waste elimination tech, no increased population tech, and no robotic controls ahahahaha!

[haphazard1]

Ouch! Some pretty devastating holes in that tech tree.

They did not have any of the waste reduction techs, but they had enhanced/improved/advanced eco restoration so they weren't totally screwed on clean up. The lack of any terraforming techs at all is painful though.

A race like this can really explode, though, if they can manage by research, trade, espionage, or conquest to gain a key missing tech or three. Give them just RW60%, IRC III, and Terra +40 (all seriously outdated compared to things they do have) and their net economy would probably roughly triple in about 8-10 turns.

[Daemon]

Yeah, i was playing Alkari with not the best planets around, and i had to completely ignore research and put everything into eradicating the Silicoids which had plenty terraforming tech, and were being conquered by the Klackons.

The Klackons finally got +100m but it was too late, fortunately .

[Waterd]

Here is something I dont understand on how the tech bonus works, once i did find a new technology, how does the next turn work? I never understood why its bad to just go full throttle into one area until one gets the technology and changes, never I see clarified how the mechanic works that makes it better to just spread the reserch evenly among the fields.