R&D question

[harrer]

Hi all,


If i get it right, the R&D has two possibilities to advance, each limited to 30/month but cumulative :

1. the R&D of the aircraft itself, with a factory full repared, and max 1 point / day (30 / month), so having a big lab isn't worth it because more than 30 capacity it's useless and will take a lot of time before it's fully repared and producing R&D points.

2. If you have more than 500 engines in stock, you got 1 research point / day.
Ok ... but for all planes to come using this engine ?
For example, if i have more than 500 Nakajima HA 35, will I win 1 research point/day for all following aircrafts ? :



Also, is it possible to advance the availability date of an engine?
( For example, A7M2 sam in september 1945, engine Ha 43 in september 1945, not worth it to research the plane if I cannot produce the engine)


Thanks in advance, couldn't find the answers in the forum.

[InfiniteMonkey]

ORIGINAL: harrer
If i get it right, the R&D has two possibilities to advance, each limited to 30/month but cumulative :

1. the R&D of the aircraft itself, with a factory full repaired, and max 1 point / day (30 / month), so having a big lab isn't worth it because more than 30 capacity it's useless and will take a lot of time before it's fully repared and producing R&D points.

Mostly true. Repair rate scales with factory size as long as size < days til aircraft arrives, so it is not so much the repair rate. Theoretically, a size 60 factory and a size 30 factory have the same repair date for research on the same aircraft. Otherwise, you are correct.

ORIGINAL: harrer
2. If you have more than 500 engines in stock, you got 1 research point / day.
Ok ... but for all planes to come using this engine ?
For example, if i have more than 500 Nakajima HA 35, will I win 1 research point/day for all following aircrafts ? :

Yes, but keep in mind that the extra point of research costs you an engine. If you are not producing engines at the rate that you are producing both aircraft and those extra research points, you will dip below 500 and stop getting the research bonus. This happens for each research factory, so you could get the bonus for only some factories in a turn. The game engine checks EACH TIME it checks for production of a point to see if the engine pool is > 500 and deducts engine(s) from the pool if the pool is > 500.

ORIGINAL: harrer
Also, is it possible to advance the availability date of an engine?
( For example, A7M2 sam in september 1945, engine Ha 43 in september 1945, not worth it to research the plane if I cannot produce the engine)

Yes. Similar to aircraft, but engine factories repair 1/day as long as supply is available at the base. There is not a random chance of repair with engines.

[harrer]

Hi InfiniteMonkey, thanks for your answers

ORIGINAL: InfiniteMonkey

ORIGINAL: harrer
If i get it right, the R&D has two possibilities to advance, each limited to 30/month but cumulative :

1. the R&D of the aircraft itself, with a factory full repaired, and max 1 point / day (30 / month), so having a big lab isn't worth it because more than 30 capacity it's useless and will take a lot of time before it's fully repared and producing R&D points.

Mostly true. Repair rate scales with factory size as long as size < days til aircraft arrives, so it is not so much the repair rate. Theoretically, a size 60 factory and a size 30 factory have the same repair date for research on the same aircraft. Otherwise, you are correct.

My point was more that a R&D lab with a size of 60 will take more time to repair than a R&D lab with a size of 30, so the 30 will start produce R&D points faster

ORIGINAL: InfiniteMonkey

ORIGINAL: harrer
2. If you have more than 500 engines in stock, you got 1 research point / day.
Ok ... but for all planes to come using this engine ?
For example, if i have more than 500 Nakajima HA 35, will I win 1 research point/day for all following aircrafts ? :

Yes, but keep in mind that the extra point of research costs you an engine. If you are not producing engines at the rate that you are producing both aircraft and those extra research points, you will dip below 500 and stop getting the research bonus. This happens for each research factory, so you could get the bonus for only some factories in a turn. The game engine checks EACH TIME it checks for production of a point to see if the engine pool is > 500 and deducts engine(s) from the pool if the pool is > 500.

So ... the goal is to produce and spare all the engines we can, got it ^^

ORIGINAL: InfiniteMonkey

ORIGINAL: harrer
Also, is it possible to advance the availability date of an engine?
( For example, A7M2 sam in september 1945, engine Ha 43 in september 1945, not worth it to research the plane if I cannot produce the engine)

Yes. Similar to aircraft, but engine factories repair 1/day as long as supply is available at the base. There is not a random chance of repair with engines.

So scen 1 we start with an engine factory : Aichi Ha 60 125 x (0) avail 42/10.
From the beginning, it will give me a monthly bonus of 30 (125/month limited to 30/month) so let say it will arrive nearly 3 months earlier ?

[GetAssista]

ORIGINAL: harrer
My point was more that a R&D lab with a size of 60 will take more time to repair than a R&D lab with a size of 30, so the 30 will start produce R&D points faster

No, factories repair at the same rate in relative terms. In your case 30 and 60 factories will start R&D at the same time on average.

ORIGINAL: harrer
So ... the goal is to produce and spare all the engines we can, got it ^^

Remember that some of the engines you will not need later in the game, and they still cost HI to produce, and supply to build factories. E.g. it is common mistake for JFBs to overproduce Na-35 for early-mid Zeros end end up with huuuge useless stock by the end of the game. Some of the other engines you don't need at all, like everything w/o numbers in the name, except Amakaze.

[InfiniteMonkey]

ORIGINAL: harrer
My point was more that a R&D lab with a size of 60 will take more time to repair than a R&D lab with a size of 30, so

Your point is wrong. The rate of repair of a research factory is based upon a random number compared to a probability. The probability that a factory will repair a point on a given day is number of devices divided by days until aircraft arrival date. Suppose I create two factories today. One is Size 300, one is size 30. Both of them are for an aircraft that will arrive in 1200 days. This turn, my chance to repair the size 300 factory is 300 in 1200, or 1 in 4. This same turn, my probability to repair the size 30 factory will be 1 in 40. Mathematically, the time to fully repair a fully damaged research factory is 61-63% of the time to arrival. The EXPECTED repair date for both the size 300 factory and the size 30 is the same. As long as size of factory < (days to arrival/.63), this holds true.

ORIGINAL: harrer
So scen 1 we start with an engine factory : Aichi Ha 60 125 x (0) avail 42/10.
From the beginning, it will give me a monthly bonus of 30 (125/month limited to 30/month) so let say it will arrive nearly 3 months earlier ?

I've done less testing on engines.... so I'm less sure of this:

You will get roughly 2.16 research points per day. (1 every day for the 40 and 80 size factories, approximately 1 every 6 days for the size 5 factory). This allows for acceleration approximately every 47 days (100/2.16 = 46.2962963)

Between 12/7/1941 and 10/1/1942, there are 298 days (turns). On approximately

1/23/1942 new arrival date 9/1/1942
3/11/1942 new arrival date 8/1/1942
4/27/1942 new arrival date 7/1/1942
6/13/1942 new arrival date 6/1/1942

[PaxMondo]

I would just add a caution about using the engine bonus ... it is very powerful, but also dangerous to your economy. You only want to do it for engine model(s)that you know you will need that ultimate capacity.

EX: Ha-45. Building this to 600 will generate 20 engines every day, which will really fuel your RnD for several good late war planes (George, Frank, Grace amoung others), so this can be a good candidate because you may have a build plan to build at least 480xFrank and 120xGrace in your final builds, so 600xHa-45 is going to get used.

Just saying, that you don't want to just start building engine factories indiscriminately. You'll end up with a lot of over built factories and short on supply when you need it ...

[Lokasenna]

ORIGINAL: harrer



So ... the goal is to produce and spare all the engines we can, got it ^^

With some minor exceptions, yes.... since engine R&D happens immediately (engine factories always repair 1 point per day, it is not a die roll) you can build up the stockpile quite quickly. I had over 1500 Turbojet and over 1600 Ha-43 at one point in my game. I massively overbuilt the Ha-43 for sure... I overestimated how much I would be using that engine, and underestimated others. Decisions made in 1942 when I didn't have any late game contextual experience (e.g., June 1944 "IJNAF fighter apocalypse" where about 20 units withdraw and don't come back).

[Lowpe]

ORIGINAL: InfiniteMonkey

ORIGINAL: harrer
My point was more that a R&D lab with a size of 60 will take more time to repair than a R&D lab with a size of 30, so

Your point is wrong. The rate of repair of a research factory is based upon a random number compared to a probability. The probability that a factory will repair a point on a given day is number of devices divided by days until aircraft arrival date. Suppose I create two factories today. One is Size 300, one is size 30. Both of them are for an aircraft that will arrive in 1200 days. This turn, my chance to repair the size 300 factory is 300 in 1200, or 1 in 4. This same turn, my probability to repair the size 30 factory will be 1 in 40. Mathematically, the time to fully repair a fully damaged research factory is 61-63% of the time to arrival. The EXPECTED repair date for both the size 300 factory and the size 30 is the same. As long as size of factory < (days to arrival/.63), this holds true.

Ok, I have tried this in an actual game. I have many size 60 r&d facilities, and in no cases have they repaired faster, or even at the same speed as a sized 30 facility.

So the n is low, but not that low as to make it worthless.

In no cases has a factory repaired more than 2 points (with engine bonus). There is very old info stating there is a small chance for greater than size 30 facilities to repair more than 1 point (more than 2 points with engine).

[Lokasenna]

ORIGINAL: Lowpe

ORIGINAL: InfiniteMonkey

ORIGINAL: harrer
My point was more that a R&D lab with a size of 60 will take more time to repair than a R&D lab with a size of 30, so

Your point is wrong. The rate of repair of a research factory is based upon a random number compared to a probability. The probability that a factory will repair a point on a given day is number of devices divided by days until aircraft arrival date. Suppose I create two factories today. One is Size 300, one is size 30. Both of them are for an aircraft that will arrive in 1200 days. This turn, my chance to repair the size 300 factory is 300 in 1200, or 1 in 4. This same turn, my probability to repair the size 30 factory will be 1 in 40. Mathematically, the time to fully repair a fully damaged research factory is 61-63% of the time to arrival. The EXPECTED repair date for both the size 300 factory and the size 30 is the same. As long as size of factory < (days to arrival/.63), this holds true.

Ok, I have tried this in an actual game. I have many size 60 r&d facilities, and in no cases have they repaired faster, or even at the same speed as a sized 30 facility.

So the n is low, but not that low as to make it worthless.

In no cases has a factory repaired more than 2 points (with engine bonus). There is very old info stating there is a small chance for greater than size 30 facilities to repair more than 1 point (more than 2 points with engine).

You mean...research more than 1 point? Not repair?


I'm inclined to believe the formula rather than a small set of personal observations, especially since my own personal non-rigorous observation of the size 0(55) Ki-84a Frank factory that you get in stock DOES repair individual points more quickly the the 0(30) size factories that I set up.

[InfiniteMonkey]

ORIGINAL: Lowpe

ORIGINAL: InfiniteMonkey

ORIGINAL: harrer
My point was more that a R&D lab with a size of 60 will take more time to repair than a R&D lab with a size of 30, so

Your point is wrong. The rate of repair of a research factory is based upon a random number compared to a probability. The probability that a factory will repair a point on a given day is number of devices divided by days until aircraft arrival date. Suppose I create two factories today. One is Size 300, one is size 30. Both of them are for an aircraft that will arrive in 1200 days. This turn, my chance to repair the size 300 factory is 300 in 1200, or 1 in 4. This same turn, my probability to repair the size 30 factory will be 1 in 40. Mathematically, the time to fully repair a fully damaged research factory is 61-63% of the time to arrival. The EXPECTED repair date for both the size 300 factory and the size 30 is the same. As long as size of factory < (days to arrival/.63), this holds true.

Ok, I have tried this in an actual game. I have many size 60 r&d facilities, and in no cases have they repaired faster, or even at the same speed as a sized 30 facility.

So the n is low, but not that low as to make it worthless.

The RNG is involved. The information is a calculation based upon mathematical analysis and is an expected outcome, not a firm date. The EXPECTED dates that a factory will FULLY repair are the same if they both start repairing from fully damaged on the same day and are for the same aircraft. When I say EXPECTED, I am intending that the term be understood in the statistical sense. The EXPECTED date is the estimate for the fully repaired date based upon statistical analysis. You can argue, but the tests, mathematical analysis, etc. were done in the following posts and I will not rehash them here:

tm.asp?m=3022096&mpage=1&key=
tm.asp?m=2908046&mpage=2&key=
tm.asp?m=2253970&mpage=2&key=

The calculated date is the most common date that the factory will fully repair, but lucky or unlucky rolls can vary that date considerably.

ORIGINAL: harrer
In no cases has a factory repaired more than 2 points (with engine bonus). There is very old info stating there is a small chance for greater than size 30 facilities to repair more than 1 point (more than 2 points with engine).

I assume by this that none of your factories have repaired beyond 2 x (58) or 2 x (28). If that is true, you definitely do not have enough data to draw a statistically significant result.

Finally, engines have nothing to do with repair of aircraft research factories, they can affect the number of research points that are produced by a FULLY repaired research factory. The only things affecting repair of a research factory are

1) Factory is set to repair
2) Supply sufficient to repair in base (minimum 10k at base iirc)
3) If 1 and 2 are met, a random die roll from 1 to the number of days til arrival for the aircraft is generated and added to the number of devices (repaired plus damaged). If the number exceeds the number of days to arrival, the factory repairs. Example: 1000 days til arrival, 600 size factory. On a random result between 1 and 400, no repair. On a 401-1000, the factory repairs. When the size of the factory is equal to the number of days til arrival, the factory repairs every day.

[Lowpe]

I was talking about two distinct items and obviously wasn't clear.

1. Larger than size 30 facilities researching more than 1 point per day or more than 2 points per day with engine bonus.

and

2. The repair rate of facilities of different size. For me size 30 or size 60.

In the second case, I have/had 16 size 60 facilities. In not one case did one of these facilities repair at a percentage to completion rate faster than a size 30.

The smaller size 30 factories repaired as little as 1 month and as much as 3 months before the size 60.

I understand my sample is low.

[Xargun]

ORIGINAL: Lowpe

I was talking about two distinct items and obviously wasn't clear.

1. Larger than size 30 facilities researching more than 1 point per day or more than 2 points per day with engine bonus.

As for aircraft, I know the engine bonus is applied PER factory - and has nothing to do with size. So if you have 5 factories researching the Sam, at best you can earn 5 bonus research points per turn. Also, I believe the factory in question has to produce a research point in order for the Bonus to be applied. So if you are not earning research points with your factory having 500+ engines won't help at all.

[Lowpe]

ORIGINAL: Xargun

ORIGINAL: Lowpe

I was talking about two distinct items and obviously wasn't clear.

1. Larger than size 30 facilities researching more than 1 point per day or more than 2 points per day with engine bonus.

As for aircraft, I know the engine bonus is applied PER factory - and has nothing to do with size. So if you have 5 factories researching the Sam, at best you can earn 5 bonus research points per turn. Also, I believe the factory in question has to produce a research point in order for the Bonus to be applied. So if you are not earning research points with your factory having 500+ engines won't help at all.

Still not being clear.

There is a rumor, and it may even have come from a developer at one point years back, that if you build a larger than size 30 r&d facility there existed a chance it would repair 2 points per day rather than one. Forget about the engine bonus since that seems to be a point of confusion.

So for example: One fully repaired size 150 r&d facility has a chance to advance 2 points per day rather than 1.

I believe the rumor goes that you need to build close to a size 300 r&d facility to routinely see the extra point of advancement (it might even advance 3 points).

[BillBrown]

Read this thread, michaelm explains all. tm.asp?m=2825013&mpage=3&key=factory&#2908252

[Lowpe]

Thanks Bill.

per Micahel

To get 3 RD, you need to have:
(a) a fully repaired factory large enough to produce 20+ planes a day to get 2 RD points (1/10 of the daily output)
(b) +1 for no damaged devices.
That would require the factory to be about 570(0) to 600(0) depending on die roll.

[Lokasenna]

ORIGINAL: Lowpe

ORIGINAL: Xargun

ORIGINAL: Lowpe

I was talking about two distinct items and obviously wasn't clear.

1. Larger than size 30 facilities researching more than 1 point per day or more than 2 points per day with engine bonus.

As for aircraft, I know the engine bonus is applied PER factory - and has nothing to do with size. So if you have 5 factories researching the Sam, at best you can earn 5 bonus research points per turn. Also, I believe the factory in question has to produce a research point in order for the Bonus to be applied. So if you are not earning research points with your factory having 500+ engines won't help at all.

Still not being clear.

There is a rumor, and it may even have come from a developer at one point years back, that if you build a larger than size 30 r&d facility there existed a chance it would repair 2 points per day rather than one. Forget about the engine bonus since that seems to be a point of confusion.

So for example: One fully repaired size 150 r&d facility has a chance to advance 2 points per day rather than 1.

I believe the rumor goes that you need to build close to a size 300 r&d facility to routinely see the extra point of advancement (it might even advance 3 points).

You need to stop saying repair when you mean research, especially in these posts where you are also talking about factory repair rates.

[rustysi]

For example, if i have more than 500 Nakajima HA 35, will I win 1 research point/day for all following aircrafts ? :

In this case that's true as all those A/C use the Ha-35. If its a stock scenario that's probably scenario 2. OTOH in stock scenario 1 the Tojo uses a different engine, the Ha-34, so that A/C would need 500+ of those engines to receive the bonus. This difference is a dev decision, and no I have no idea why they choose to do it.

Basically I just wanted to point out that there may be some differences between scenarios and mods, so be aware.

[InfiniteMonkey]

ORIGINAL: Lowpe

Thanks Bill.

per Micahel

To get 3 RD, you need to have:
(a) a fully repaired factory large enough to produce 20+ planes a day to get 2 RD points (1/10 of the daily output)
(b) +1 for no damaged devices.
That would require the factory to be about 570(0) to 600(0) depending on die roll.

This is very easy to test/prove. Open the scenario editor and make some BIG research factories that are fully repaired. Then, run the turn and observe the progress towards research on an aircraft by mousing over the aircraft name in the production screen list. (filter to RD Air only or sort by size descending to make finding them easier). Note in the picture below that on 12/8/41 (1 game turn run) the 900 size factory has generated 3% research points on the Irving. It is the only repaired research factory and no engine bonus applies as the pool of engines is not > 500.

[ElvisDaKing]

I have a question

July 1942, I plan to start from scratch R&D program for Ki-100 I Tony.

I will have 3 R&D lanes available
I was wondering if it s faster to start the R&D with Ki-100 I Tony, or to go through the whole 'Tony' lane starting with Ki-61a ?

SC 1
PDU ON
I will not produce any Kawasaki Ha-60


My point is that Ki-100 I Tony is available on March 1945, so it will take very long time to build the R&D Ki-100 I Tony from 0(30) to 30(0)

But if I start with Ki-61 Ia Tony available February 1943, it is very likely that my 3 lanes will be fully functionnal [30(0)] by end of 1942, and then jumping from Ki-61 Ia to Ki-61 Ib (09/43) , Ki-61 Id (04/44), Ki-61 II KAI (08/44)...

So what should be fastest option to get the Ki-100 Tony ?

[GetAssista]

ORIGINAL: ElvisDaKing
I will have 3 R&D lanes available
I was wondering if it s faster to start the R&D with Ki-100 I Tony, or to go through the whole 'Tony' lane starting with Ki-61a ?

Depends on what do you mean by "go through". If you will skip through 61 models not researching them then 2nd option is obviously much faster. If you will research every consequtive model to arrival and only then switch, then times will be comparable I guess