Relative Research Costs DB V1.2
 

Something that confused me early on in my Dominions 3 campaigns was which is the best mage to use for doing that all important research. Since the logical approach of always using the mage with the highest Research Ability will quickly lead many nations into money problems due to crippling upkeep costs. So to make things easier for new and old players alike, here is a database of the 'Relative Research Costs' for each nations recruitable mages.

The main function of this database is to aid the player in working out which nations mages are the most cost effective at doing research. Although please note that the "most cost effective research mage" does not necessarily mean "the best research mage".

Besides the basic information of a mages purchase cost and upkeep costs, I have also included how many Research Points (RP's) have been accumulated by that mage after 15 and 30 turns, and therefore how much money each RP has cost you relative to the mages recruitment and upkeep cost. Note that after 15 turns a mage gains +1 to their Research Ability (RA) due to them gaining their first experience star. This has been included in my calculations since it has a noticable bearing on the figures. A further note is that a mage DOES NOT gain any extra experience from researching (as page 58 of the manual indicates), and a mage with the 'defend' order gains experience at the same rate as a mage with the 'research' order.

I have not included in my calculations any random magic paths that are not guaranteed (ie. anything below 100% is counted as 0). This may lead to some mages getting lower scores (for want of a better description) than they would if they managed to get a complete set of random picks. I will use Arcoscephale's 'Mystic' as an example of what I mean here -

The 'Mystic' unit has a base RA of 2, and then +1 RA for the native Astral path, and +1 RA for the guaranteed ?1 100% FWES path, giving a final RA total of 4. It also has three ?1 50% chances of getting additional magic picks, but I have not included these in my calculations since they are not guaranteed. But it goes without saying that a 'Mystic' who succeeds in getting all three 50% chances, for a total RA of 7, will be a more cost effective researcher than one who fails to get any. I may included these additional figures, and the percentage chance of mages getting them, in future versions of the database. All depends how many people would find the extra info useful :-)

I have only collected the data for Early Age so far, this is mainly because I want to test the water with regards to how useful people find this database before putting time into doing the other ages. If enough people give it the thumbs up, I shall compile likewise lists for MA, LA, Indy mages, and possibly summons (although upkeep costs are mostly irrelevant to the latter, so any calculations may be better based on gem costs)

I also haven't attempted to analyse the data here, or to make any suggestions as to which mages each nation should use for researching. Since that has the problem of requiring me to be a far better Dominions player than I actually am :) So I'm very happy to leave these jobs to the players in the community who know a lot more about these things than I do.

Finally, please feel free to point out any mistakes or omissions I have (most likely) made, and I am very open to any suggestions of what other research/cost related info that might be useful which I have not included.


V1.2 Edit
Done a bit of a tidy up, and added some input functions. The data can now be altered to reflect a given turn number, and the research bonus/penalty from the Magic Scales. The data available is still only for EA at the moment, but once I'm happy with the layout and functions, I will start on the other ages.

All feedback, comments and suggestions are as always very welcome :)

Re: Relative Reseerach Costs DB V1.0
 

This is going to be a very useful tool. However, if you are not already using it, you can find all the mages (recruitables, summonables etc) in the Dom3 DB along with a list of their randoms. It can save you a lot of time compiling this thing. It also has army lists for the various nations, so you can build the list more quickly than going through everything by hand.

Re: Relative Research Costs DB V1.0
 

Thanks Edi. I use your (invaluable) database on a regular basis to check up on a whole host of things. To be honest, using your database scares me more often than not simply due to the vast amount of data it contains!

Although I didn't use it to compile the above database, mainly because I wanted that 'compiled from scratch' feeling for the first draft. But I will certainly be turning to it for future versions. I don't think I could even contemplate doing the list of summons for this database without having your database to turn to for all the info I'd need.

Re: Relative Research Costs DB V1.0
 

A man after my own heart then, wanting to do everything from scratch just to make sure it gets done right. I can certainly understand that. Especially for the first draft. :)

Keep up the good work! :D

Re: Relative Research Costs DB V1.0
 

Calahan
It certainly seems a very good tool but it may be made much more better. It cries dynamics: user should be able to enter (at least) duration and magic scale level. Probably something else can be added.
Great work anyway. Thanks!

Re: Relative Research Costs DB V1.0
 

This is very cool and useful. My one non-essential suggestion would be to just have the title rows once and freeze panes. Maybe make the nation type a column. That way we could sort easily, for example to see who are the most cost effective. Great work!

Re: Relative Research Costs DB V1.0
 

Thanks for the suggestions so far.

Ano - I'm hoping to make several improvements as the database grows, and your idea of adding dynamics certainly sounds like a good direction to go in.

DC - Having a separate nations column is so obviously a good idea that I now consider myself a complete idiot for not doing it like that in the first place :doh:

Re: Relative Research Costs DB V1.0
 

Personally, I think that an approach that includes non-100% picks should be included.

You can do this in a few ways. The most straightforward analysis would be to use Expected values.

Re: Relative Research Costs DB V1.0
 

If by expected values, you mean long term averages, I was about to suggest the same.

So if a mage has 4 paths and a 10% pick, just call him 4.1 in the listing. I would have thought it worthwhile to just drop that, until you used the Mystic as an example. ;) They should average 1.5 random paths, and that is a massive difference in the "baseline" research level that you should get out of them.

Otherwise, awesome idea, this is a cool tool. 8 )

Re: Relative Research Costs DB V1.2
 

Good job there mate, it's definitely very useful already for working out who gets the research grants, any extra functionality will be a bonus.