Lokasenna
Posts: 9297
Joined: 3/3/2012
From: Iowan in MD/DC
Status: offline
|
quote:
ORIGINAL: InfiniteMonkey
1. You may have heard that the optimal size for an R&D factory is 30. This is because you only get one extra research point from engines PER FACTORY. It is therefore better to have 2 x 30 research factories than to have 1 x 60.
2. The optimal size of the research factory is related to the time til the aircraft arrives. Contrary to popular belief, building every research factory to size 30 IS NOT OPTIMAL FOR ALL SITUATIONS! For an aircraft like the Ki-84 Ia Frank (a JFB favorite for research), DO NOT TRY TO BUILD UP TO a 30 size research factory. In fact, I recommend changing the 0x (56) Frank factory at start to SOMETHING ELSE. Remember: Factories only contribute toward research progress when they are fully repaired. That 0x(56) factory will not repair fully (and will therefore contribute nothing to research) for YEARS. With a 3/44 arrival date, you have about a 1 in 27 chance to repair each day. If each repair takes the expected time to repair, the factory will begin contributing to the research of the Ki-84 sometime around Halloween day in 1943………….. Only 6 months before it is due to arrive anyway. The optimal way to accelerate planes like the Frank is to research with MANY factories til you have approximately 100 production repaired. You increment the size of factories ONLY when they are fully repaired and only by 1. Over time, some factories will repair more than others. Keep the big ones and reassign the smaller ones. Let's say I research with 20 factories until I have 100 fully repaired production units. Each month, with those in operation I get 100 research points and accelerate the Frank by one month. How long will it take me to begin accelerating the Frank? Around mid-April 1942, I will begin accelerating the Frank at a rate of about 1 month per month. I will get the Frank starting around 3/43 instead of 3/44.
1. 30 is the optimal size - for R&D. Many of my factories have also remained at 30 simply because it's easier that way and I don't want to produce too many planes, but it depends on the model. Size 30 is optimal because it is the smallest expenditure of supply to get there and you will only get 1 point of research no matter the size of the factory (barring engine bonus).
2. Your initial impressions are not wrong, however there is more nuance in these decisions. For example, changing that 0(55) Frank-a factory is going to waste a bunch of your supply. Why would you ever do that? Apparently you run a bunch of turns to test things. Run a bunch of turns and watch the 0(55) Frank factory and compare to a 0(30) Frank factory. They will repair at the same proportional rate - e.g., over many runs of the game (not many turns, many games) they will be fully repaired at the roughly same time. This is because of the algorithm for repairing factories. Now, you're going to want a lot of Franks. Why would you get rid of a size 55 Frank factory for something else? You should set Frank R&D at the start - its standard arrival date is early enough that the factories will begin to repair immediately, albeit slowly at first.
If you change that Frank factory, you're burning many thousands of supplies that you don't need to burn.
By the by, my Frank-a arrived in 6/43 and I converted 5 of my factories to Frank-r at some point during the R&D process, before Frank-a arrived. I used 7 factories, which is fewer than many JFBs on this forum use for that plane. My Frank-r then arrived in 12/43. I didn't even get the Ki-100-I until 7/43, and I did the move-through-the-line thing on it that you bring up later in your post.
Finally, smaller factories do repair slower than bigger ones. A 0(2) Frank factory is going to repair 1 point so much slower than 0(30) that the factory starting at 0(30) will have more than 2 points repaired before the 0(2) is repaired and you can do as you recommend, incrementing the factory up by 1. And then that factory that started at 0(30) is going to repair more than 1 point before that 2(1) factory repairs to 3(0), and so on. You can do it that way if you want to, but it's not going to work out very well for you. I did a few googles to find some threads, but sanderz's post #5 here is roughly correct:
http://www.matrixgames.com/forums/tm.asp?m=3754188
|
Printable Version
RE: Japan Questions - 
: 
New Messages
No New Messages
Hot Topic w/ New Messages
Hot Topic w/o New Messages
Locked w/ New Messages
Locked w/o New Messages
Post New Thread