Japanese R&D

Post bug reports and ask for help with other issues here.

Moderators: wdolson, MOD_War-in-the-Pacific-Admirals-Edition

bk19@mweb.co.za
Posts: 258
Joined: Tue Jul 26, 2011 8:27 pm

RE: Japanese R&D - revised

Post by bk19@mweb.co.za »

ORIGINAL: pompack

It has also been my observation that the GFU is one of the most frequently upgraded (or at least revised) items in the TO&E with the GFU Mk. VII being the most advanced cited to date. It seems that the GFU has a tendency to become seperated during periods of high WitP stress and rapid map movement.

Just as well he is not a sailor, otherwise he would be talking in terms of GFU[port][mark] !!
Numdydar
Posts: 3283
Joined: Fri Feb 13, 2004 9:56 pm

RE: Japanese R&D - revised

Post by Numdydar »

ORIGINAL: kmitahj

ORIGINAL: Numdydar
Have you also considered that (as I've stated before) a larger factory repairs faster than a smaller factory? But the downside being that there is more to repair. There is an overlap (correlation) between the two that make larger and smaller meet somewhere in the middle. I did some testing a while back and found they end up being repaired fully around the same time... although it was a while ago and I'm foggy on it now. So while they're harder to repair, size does matter.

If there are more damaged factories, they will have a greater chance to have a one point repair simply due to the number of chances invilved as each damaged factory is rolled for seperately. The more rolls, the move likely a repair will happen. However, as the number of damaged factories approches zero, you have the same odds of getting that last point repaired as if it was a 0(1) factory. Because as far as the repair routine is concerned it is the same. I.E only one chance to repair. So the perception is that larger factories repair faster and they do to a point, but it slows down as the limit <-0.

Of course I thought I uderstood the acceleration process very well too, lol. So it would not surprise me if my understanding of the R&D factory repair system was totally wrong as well [:)]
This is not so. What matters for repair process is total size of the factory (sum of active and damaged devices) which is constant over whole repair process (unless you decide to expand in the middle - not good idea usually). In fact over time the chance of success is growing slowly because the second factor - number of days till plane arrival - is getting lower day by day.
And second as was said many times before because chance of repair depends on the FactorySize all factories repair from fully damaged to fully repaired state in about same time on average. No real reason to prefer small factories over 30 size. Of course in reality you will see large dispersion as usual for (kinda) stochastic process - some factories will finish repair and crank r&d points earlier some will lag behind expected average time. Thats live.
I think it is as it schould to be. Research must not be a low hanging fruit. One have to invest first and take his time before he got his reward [:)]
Note also that once the research process was demistifid thre is no more uncertainty in it. Using 30+ factories you can predict next month advance as well as final date of going into production basically up to a day. Only unpredictability left is now in the repair process. And unpredictability is good thing for research I dare to say.

I know that the highlighted item is definately not acccurate. I have 0(2) factories for aircraft in late 45/early 46 some of which have fully repaired while others have not repaired a single point. I also have factories that are under 10 that have gotten their factories fully repaired while larger factories are still waiting to be fully repaired over the same time frame for planes being available within the same six moth window. I will be glad to show you the data to support this if you want.

It makes no sense that the repair routine for R&D factories to care about undamaged factories as it should be checking only damaged ones for the roll. I would assume if the R&D acceleration routine has a check for damaged factories >0 that the repair routine whould have a similar check for damaged factories >0 too as there would no reason to even run the routine if therewere no damaged factories.

Of course given my track record in understanding how the game works is very poor, I could be just as wrong about this as R&D acceleration [:D]. But I would like to see some actual data to back up the claims that the total factory size matters versus just the damaged ones.
User avatar
viberpol
Posts: 858
Joined: Thu Oct 20, 2005 3:16 pm
Location: Global village, Poland, EU

RE: Japanese R&D - revised

Post by viberpol »

ORIGINAL: n01487477
ORIGINAL: Numydar
Actually if a plane is one month away, it can very much so get an advance as it happened to me [:)]. I had a plane scheduled to be available in 2/43. I was waiting until later in 1/43 to change it to the next plane in squence To my surprise, half way through Jan, they start showing up in my pool. I checked and all the R&D factories had swiitched to production [X(] I was NOT a happy camper lol. Since then I have paid a MUCH closer attention to when they switch [:)] Of course it helps that Michael has color coded them to be highlighed when they are within 45 days of conversion [:D]
Yeah - I was wrong in saying that. Of course planes can R&D mid-month or any time. But you need to have the points accumulating. Not just a one off factory still repairing.

Numydar, actually, in some latest beta Michael introduced a clever switch to get rid of the problem of mid-month unwanted change from R&D factory into an active.

In the Industry window and R&D Air column you set "Prod" to "No", even if the frame accelerates one month and become "active", you have moreoptions.
You can still change it into the next researched (not active) frame or uncheck the production and produce it/or change like a normal active plane.

ORIGINAL: n01487477
It might be nice to see in the Aircraft pool screen how many R&D points are currently accumulated though...

You can see it now on the mouse-over in the Industry screen as a percentage. Michael has put it in, so I'm going to put it in Tracker for the next release...
Having said that I kinda liked the mystique of not knowing

Erm, what? Damian, can you elaborate? Mouse over in industry screen?[&:]
I always lacked knowledge in this matter and I didn't know it was possible to get the info!
Przy lackim orle, przy koniu Kiejstuta Archanioł Rusi na proporcach błysł
User avatar
n01487477
Posts: 4764
Joined: Tue Feb 21, 2006 12:00 am

RE: Japanese R&D - revised

Post by n01487477 »

ORIGINAL: viberpol

ORIGINAL: n01487477
ORIGINAL: Numydar
Actually if a plane is one month away, it can very much so get an advance as it happened to me [:)]. I had a plane scheduled to be available in 2/43. I was waiting until later in 1/43 to change it to the next plane in squence To my surprise, half way through Jan, they start showing up in my pool. I checked and all the R&D factories had swiitched to production [X(] I was NOT a happy camper lol. Since then I have paid a MUCH closer attention to when they switch [:)] Of course it helps that Michael has color coded them to be highlighed when they are within 45 days of conversion [:D]
Yeah - I was wrong in saying that. Of course planes can R&D mid-month or any time. But you need to have the points accumulating. Not just a one off factory still repairing.

Numydar, actually, in some latest beta Michael introduced a clever switch to get rid of the problem of mid-month unwanted change from R&D factory into an active.

In the Industry window and R&D Air column you set "Prod" to "No", even if the frame accelerates one month and become "active", you have moreoptions.
You can still change it into the next researched (not active) frame or uncheck the production and produce it/or change like a normal active plane.

ORIGINAL: n01487477
It might be nice to see in the Aircraft pool screen how many R&D points are currently accumulated though...

You can see it now on the mouse-over in the Industry screen as a percentage. Michael has put it in, so I'm going to put it in Tracker for the next release...
Having said that I kinda liked the mystique of not knowing

Erm, what? Damian, can you elaborate? Mouse over in industry screen?[&:]
I always lacked knowledge in this matter and I didn't know it was possible to get the info!
Not sure I'm fully in favour of it ... mouse over any plane and voila ...

Image
Attachments
R&D.jpg
R&D.jpg (254.77 KiB) Viewed 542 times
User avatar
viberpol
Posts: 858
Joined: Thu Oct 20, 2005 3:16 pm
Location: Global village, Poland, EU

RE: Japanese R&D - revised

Post by viberpol »

ORIGINAL: n01487477

Not sure I'm fully in favour of it ... mouse over any plane and voila ...

Image

Ahh... Thanks!
That must be the q4 addition?
Changed Updated RD totals to reflect actual R&D available for a/c and engines [MEM]
I am stuck with P8 as my opponent didn't want to hand me the benefit of friendly river crossing... [;)]
Przy lackim orle, przy koniu Kiejstuta Archanioł Rusi na proporcach błysł
kmitahj
Posts: 100
Joined: Sun Apr 24, 2011 10:31 pm

RE: Japanese R&D - revised

Post by kmitahj »

ORIGINAL: Numdydar
Of course given my track record in understanding how the game works is very poor, I could be just as wrong about this as R&D acceleration [:D]. But I would like to see some actual data to back up the claims that the total factory size matters versus just the damaged ones.
Numdydar, looks like you fight your battles to the very end. Respect! [:)]

As for the proof for total FactorySize as factor in r&d repair source code is the only true authority and source code is Michael's kingdom. So if you ask him kindly and if he is not too busy squashing bugs or simply too tired with this very topic he will give you your desired definite answer. All I can say with 99.9(9)% certainty level is that in classic WitP everyday repair test check is based on the condition
Random(NumberOfDaysTillPlaneArrival) <= FactorySize
where factory size is sum of active and damaged devices. Use it or loose it as you wish. As I've said before part of having fun with the game is doing as you see fit which includes making mistakes and learning form them.
I know that the highlighted item is definately not acccurate. I have 0(2) factories for aircraft in late 45/early 46 some of which have fully repaired while others have not repaired a single point. I also have factories that are under 10 that have gotten their factories fully repaired while larger factories are still waiting to be fully repaired over the same time frame for planes being available within the same six moth window. I will be glad to show you the data to support this if you want.
I think you don't pay enough attention to "on average" phrase. Remember that r&d factory repair from fully damaged to fully repaired is stochastic process - chain of random events. Think of it as repeated coin flipping where we want to know how many flips is needed to achieve certain fxed number of "heads". Only in our case the coin is heavy biased against us (probability of single success is very low in all practical cases) and that bias is not constant (probability of success grows slowly with every flip). For such process you need to use statistics approach and certainly you need more then few test runs to draw definitive conclusions.
Beeing a true Bayesian Hero ( [:D]) maybe I would be able to calculate in a blink of an eye both expected average as well as three-sigma interval and then use them to evaluate them against your sample results. But as a poor, little chap who tries to stay away from fat statistics textbooks I could offer you only (I did in the other thread) an intution why the fact that probability of single success is in proportion to factory size results in average of total repair times which do not depend on that very factory size (with silent but obvious assumption that NumberOfDaysTillArrival >> FactorySize).
Now I may supplement that with second intuition: though averages of times of full repair does not depend on FactorySize their deviations (and thus confidence intervals) are getting smaller with the the size of factory. With 1(0) factory one lucky flip with our heavily biased coin is enough to get done, with 30(0) factory there is enough time for lucky and unlucky results to even up a bit. Surely deviation still will be there and running say ten 30(0) factories you will see some of them finish repair well before average time whereas others will lag far behind. Again: that's life!

Numdydar
Posts: 3283
Joined: Fri Feb 13, 2004 9:56 pm

RE: Japanese R&D - revised

Post by Numdydar »

Not just averages but different available dates, different sized factories, even whether the factories are located may even play a part in the repair process. So, there are too many unknows without access to the code for a player to really answer the question of what goes on with R&D repair. Given the stupidly complex routine of R&D acceleration, I would guess that there are just as much in the repair code too. It may even have checks that if the plane is too far out, that it prevents repair at all [:(].

Bottom line is neither of us truely knows for sure. We both have the observations we have made and based on them come to different results. I tend to think that we are both right in certain areas and both worng in others. Our interpetations of what we think are right does not appear to intersect so we end up with a null set. This does not mean that one set is correct and the other wrong, it is just that the two sets contain similar data just analysized differently so that there is no intection even though the same data set was used [;)] Typical scienctific process imho [:D]
User avatar
michaelm75au
Posts: 12467
Joined: Sat May 05, 2001 8:00 am
Location: Melbourne, Australia

RE: Japanese R&D - revised

Post by michaelm75au »

ORIGINAL: viberpol

ORIGINAL: n01487477

Not sure I'm fully in favour of it ... mouse over any plane and voila ...

Ahh... Thanks!
That must be the q4 addition?
Changed Updated RD totals to reflect actual R&D available for a/c and engines [MEM]
I am stuck with P8 as my opponent didn't want to hand me the benefit of friendly river crossing... [;)]

It represents the % to the next month jump. Just to satisfy players that R&D is moving along.[;)]
Michael
User avatar
n01487477
Posts: 4764
Joined: Tue Feb 21, 2006 12:00 am

RE: Japanese R&D - revised

Post by n01487477 »

ORIGINAL: michaelm

ORIGINAL: viberpol

ORIGINAL: n01487477

Not sure I'm fully in favour of it ... mouse over any plane and voila ...

Ahh... Thanks!
That must be the q4 addition?
Changed Updated RD totals to reflect actual R&D available for a/c and engines [MEM]
I am stuck with P8 as my opponent didn't want to hand me the benefit of friendly river crossing... [;)]

It represents the % to the next month jump. Just to satisfy players that R&D is moving along.[;)]
I just put it in Tracker too ... screen shot in the Tracker thread.

Also, just wanted to bring up a query about Allied R&D. I know that Allied planes have R&D and I'm not against it, however because the game sees they're all repaired the R&D brings many Allied planes into production at least a month early ...
User avatar
michaelm75au
Posts: 12467
Joined: Sat May 05, 2001 8:00 am
Location: Melbourne, Australia

RE: Japanese R&D - revised

Post by michaelm75au »

Allied planes don't generate R&D points. It is only a Japanese thing.
Michael
User avatar
n01487477
Posts: 4764
Joined: Tue Feb 21, 2006 12:00 am

RE: Japanese R&D - revised

Post by n01487477 »

ORIGINAL: michaelm

Allied planes don't generate R&D points. It is only a Japanese thing.
Well then Sydney we have a problem[&:]. This acceleration has been reported by Allied players since tracker & AE was released. I didn't query it at the time as the game was new & now that we're reporting R&D it is time to bring it up again... cause I can now see allied R&D happening. Maybe this is just an AI issue rather than a player issue ... in which case it's nothing, but as players have already reported it ...

So, this is happening with factories NOT on map R&D status and with (I think).

Image
Attachments
RDAllies.jpg
RDAllies.jpg (492.27 KiB) Viewed 542 times
User avatar
michaelm75au
Posts: 12467
Joined: Sat May 05, 2001 8:00 am
Location: Melbourne, Australia

RE: Japanese R&D - revised

Post by michaelm75au »

Yes, the AI has it's own factory producing (and R&D) code. But it doesn't do Allied R&D either.

If there is a loophole in the code somewhere, I'll try to track it down for the weekend.


Yes, the AI can do R&D on Allied planes, but not the human player. It is a mistake.
Michael
User avatar
Andrew Brown
Posts: 4088
Joined: Tue Sep 05, 2000 8:00 am
Location: Hex 82,170
Contact:

RE: Japanese R&D - revised

Post by Andrew Brown »

ORIGINAL: inqistor

Thinking of it...


Japanese planes, which have research factories set at start, should have availability dates set few months AFTER their historical arrival, to include this research which will be done, if player do NOT change anything.

As this is now, every model will arrive at its historical date, or earlier, even when player do not invest as much R&D, as Japan during war.

You may be right. I, for one, have never tested it, or calculated what the in-game arrival dates end up being. It would need someone with more time than me to do the maths and work out what the adjusted arrival dates for Japanese aircraft should be (assuming that they should be adjusted at all).

Andrew
Information about my WitP map, and CHS, can be found on my WitP website

Image
User avatar
n01487477
Posts: 4764
Joined: Tue Feb 21, 2006 12:00 am

RE: Japanese R&D - revised

Post by n01487477 »

ORIGINAL: Andrew Brown

ORIGINAL: inqistor

Thinking of it...


Japanese planes, which have research factories set at start, should have availability dates set few months AFTER their historical arrival, to include this research which will be done, if player do NOT change anything.

As this is now, every model will arrive at its historical date, or earlier, even when player do not invest as much R&D, as Japan during war.

You may be right. I, for one, have never tested it, or calculated what the in-game arrival dates end up being. It would need someone with more time than me to do the maths and work out what the adjusted arrival dates for Japanese aircraft should be (assuming that they should be adjusted at all).

Andrew
I don't think there is any need, the system is back to what it should be. If a player did nothing, I doubt there would be any planes that they would get one months advancement. The R&D is just scattered to the wind in Scen 1. The AI does it differently - it cheats and I can live with that.

ORIGINAL: michaelm
Yes, the AI has it's own factory producing (and R&D) code. But it doesn't do Allied R&D either.
If there is a loophole in the code somewhere, I'll try to track it down for the weekend.
Yes, the AI can do R&D on Allied planes, but not the human player. It is a mistake.
OK - I'll do a test as Allied Human V Japanese AI to double check anyway. Thanks - I just needed to see if I needed R&D numbers for the Allies.

Cheers
User avatar
michaelm75au
Posts: 12467
Joined: Sat May 05, 2001 8:00 am
Location: Melbourne, Australia

RE: Japanese R&D - revised

Post by michaelm75au »

ORIGINAL: michaelm

Yes, the AI has it's own factory producing (and R&D) code. But it doesn't do Allied R&D either.

If there is a loophole in the code somewhere, I'll try to track it down for the weekend.


Yes, the AI can do R&D on Allied planes, but not the human player. It is a mistake.
Found the error - not checking for Japanese aircraft during one of the AI production points.
Michael
User avatar
n01487477
Posts: 4764
Joined: Tue Feb 21, 2006 12:00 am

RE: Japanese R&D - revised

Post by n01487477 »

ORIGINAL: michaelm

ORIGINAL: michaelm

Yes, the AI has it's own factory producing (and R&D) code. But it doesn't do Allied R&D either.

If there is a loophole in the code somewhere, I'll try to track it down for the weekend.

Yes, the AI can do R&D on Allied planes, but not the human player. It is a mistake.
Found the error - not checking for Japanese aircraft during one of the AI production points.
Just so I understand correctly - the AI should be able to do R&D on Allied planes ? But the human Allied player not ... right ? So why would this error be wrong?

Or both shouldn't be able to ?

Sorry confused [&:]
User avatar
michaelm75au
Posts: 12467
Joined: Sat May 05, 2001 8:00 am
Location: Melbourne, Australia

RE: Japanese R&D - revised

Post by michaelm75au »

Allied planes aren't suppose to be researched according to original code.

Something that was added for the AI, increased the R&D for planes without regard to side.
It should have not have done so for Allied.
Michael
User avatar
inqistor
Posts: 1813
Joined: Wed May 12, 2010 1:19 pm

RE: Japanese R&D - revised

Post by inqistor »

ORIGINAL: Andrew Brown

ORIGINAL: inqistor

Thinking of it...


Japanese planes, which have research factories set at start, should have availability dates set few months AFTER their historical arrival, to include this research which will be done, if player do NOT change anything.

As this is now, every model will arrive at its historical date, or earlier, even when player do not invest as much R&D, as Japan during war.

You may be right. I, for one, have never tested it, or calculated what the in-game arrival dates end up being. It would need someone with more time than me to do the maths and work out what the adjusted arrival dates for Japanese aircraft should be (assuming that they should be adjusted at all).
Calculation is pretty simple. We already know, that fully repaired factory will research up to 30 points per month. The only problem is estimation WHEN they will be repaired on-average. Then just multiply factory size by number of months remaining for arrival.

There is quite a lot of small factories, but I can see several larger. The first candidate is Ki-84a, which have size 55 factory, so it can be accelerated even by few months. Also first TONY seems to be safe bet.

I am actually wondering of TOJO. Its initial production in September 1942 amounted to only 8 planes, so that in-game-terms would suggest research acceleration at the end of month, with larger factory.

Image
Attachments
01Sep.2221.38.jpg
01Sep.2221.38.jpg (427.14 KiB) Viewed 543 times
Post Reply

Return to “Tech Support”