open All Channels
seplocked Science and Industry
blankseplocked Invention introduction guide
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: 1 [2] 3 4 5 6

Author Topic

Gaogan
Gallente
Solar Storm
Sev3rance
Posted - 2007.05.16 04:49:00 - [31]
 

How do the mentioned probability multipliers on the decrypters stack? I saw somewhere that the base chance of inventing ships is 15%, and after 6/6 failed ship jobs with test reports, it does seem to be much lower than 3/5 succeeded module jobs with no decrypter. The question is, what skill level is that 15% chance with ( I have 4/4/4 ) and how does that stack with the decrypters?

If they multiply doesn't that mean the -60% chance +9 runs decrypter would give you only a 6% success rate inventing ships? So wouldn't you be an idiot to use anything other than the +30% chance one?

Ittey
Whiskey Tango Foxtrot
Posted - 2007.05.16 06:30:00 - [32]
 

You have to factor in how much more money are you going to be able to make with those extra runs. If you are inventing hulks then you may be willing to take a 4% success rate over a 10% because when you get that success you get 10 runs instead of 5, or 1 with no decryptors.

And what success rate skill 4/4/4 or any other level is still unknown, but there are getting enough samples that it will get more and more accurate.

Qual
Gallente
Cornexant Research
Posted - 2007.05.16 12:59:00 - [33]
 

Noticed you where looking for info on negtive ME as well, I did that formula some time ago here.

ISD Zhuge Liang


ISD STAR
Posted - 2007.05.16 15:07:00 - [34]
 

Good stuff, sticked with an unholy amount of glue.

Gaogan
Gallente
Solar Storm
Sev3rance
Posted - 2007.05.16 23:57:00 - [35]
 

5 runs at 10% is better than 10 runs at 4% hands down. The question is, what are the actual odds you get after the decrypters are applied on top of your skills?

Aykido
Gallente
Lobster of Babel
Posted - 2007.05.17 01:12:00 - [36]
 

Edited by: Aykido on 17/05/2007 01:11:13
A statistical sample is around 1000 identical invention attempts, needs to be repeated with different skill levels, decryptors etc.

Lots of work

And still you'd have a small chance that you would get a wrong/skewed result.

Furthermore, the devs seem to have tweaked chances a few times since invention became possible. And it is possible that they implement new tweaking first on Sisi, or at least that they sometimes do.

So this is going to take a while.


Dominique Vasilkovsky
Gallente
BFG Tech
Posted - 2007.05.17 11:58:00 - [37]
 

Originally by: Gaogan
So wouldn't you be an idiot to use anything other than the +30% chance one?


Possibly.

There are two approches for ship invention:

High probability / high cost

Max run BPC
Ship as meta item
+30% 4-run decryptor

Each attempt cost a lot, and you have 20-25% chance to get a 5 run BPC


Low probability / low cost

1-run BPC
-60% 9-run decryptor

Each attempt is really cheap so you can do quite a few more attempts compared to the one above. You have 8-11% chance to get a 9-run BPC.

It's up to you which approach you want to follow but I have now invented 12 9-run ship BPCs in 111 attempts.

Btw, my record is 24 consecutive fails so your 6 fails is just a bad start.

Kdah
Posted - 2007.05.17 14:42:00 - [38]
 

Edited by: Kdah on 17/05/2007 14:41:02
Ittey, your formula for number of runs produced seems to be incorrect.

You say it's:

Runs = max(1;Round.Down(ActualRunsInputBPC/MaxRunsInputBPC*MaxRunsOutputBPC + DecryptorBonusRuns))

Shouldn't the DecryptorBonusRuns be outside the parenthesis?

When you give the example, you put it outside of the parenthesis:

Runs = max(1;Round.Down(1000/10*1000)) + 4

And, in that example you have the order of the variables wrong. It should be:

Runs = max(1;Round.Down(1000/1000*10)) + 4

Also, if what Dominique says above is true, then your Max() function is wrong. If the Max() were correct, then the least number of runs you could get on a BPC with no decryptor would be 1. Max(1,anything-below-1) = 1

And yet Dominique says that when she uses the 9-Bonus-Run decryptor with a minimum run BPC, she gets only 9 runs on the resulting invented T2 BPC. That would indicate that the Max() function isn't working.



Kdah
Posted - 2007.05.17 14:51:00 - [39]
 

Ok, I think I know what the problem is. The formula should read:

Runs = max(1;Round.Down(ActualRunsInputBPC/MaxRunsInputBPC*MaxRunsOutputBPC) + DecryptorBonusRuns)

That makes sure that you always end up with at least one run, but if you use a decryptor with bonus runs and a low-run BPC, you can end up with only the bonus runs on the resulting BPC.

Your example should then read:

Runs = max(1;Round.Down(1000/1000*10) + 4)

Qual
Gallente
Cornexant Research
Posted - 2007.05.17 21:26:00 - [40]
 

Originally by: Kdah
Ok, I think I know what the problem is. The formula should read:

Runs = max(1;Round.Down(ActualRunsInputBPC/MaxRunsInputBPC*MaxRunsOutputBPC) + DecryptorBonusRuns)

That makes sure that you always end up with at least one run, but if you use a decryptor with bonus runs and a low-run BPC, you can end up with only the bonus runs on the resulting BPC.

Your example should then read:

Runs = max(1;Round.Down(1000/1000*10) + 4)



I guess I am to blame for that error. And you are of course right. Good catch!

Lady Rachel
Posted - 2007.05.22 13:58:00 - [41]
 

Edited by: Lady Rachel on 22/05/2007 13:57:17
someone has a list where is described to the several the date interfacewho invention can make?
example
esoteric data interface: all missile lounchers, ballistic control sistem, all railguns,.

tnx

Gaogan
Gallente
Solar Storm
Sev3rance
Posted - 2007.05.23 01:47:00 - [42]
 

How the hell are you supposed to make money inventing ships when you fail 90% of the attempts? Especially ones larger than frigs that take a lot of data cores? Isn't it bad enough that you only get a 1 run bpc out of it, and they take a lot of cores? Why should they also have a horrible chance of success?

Take occators, which are still fairly high price items. At 8 data cores per attempt, you are looking at an approximate cost of 30 million isk per attempt. If it takes 10 tries to get a 4 run bpc, that's a per run cost of 75 million isk, which is about what they sell for, which is at least 3 times what they should cost.


Dominique Vasilkovsky
Gallente
BFG Tech
Posted - 2007.05.23 08:58:00 - [43]
 

Originally by: Gaogan
How the hell are you supposed to make money inventing ships when you fail 90% of the attempts? Especially ones larger than frigs that take a lot of data cores? Isn't it bad enough that you only get a 1 run bpc out of it, and they take a lot of cores? Why should they also have a horrible chance of success?

Take occators, which are still fairly high price items. At 8 data cores per attempt, you are looking at an approximate cost of 30 million isk per attempt. If it takes 10 tries to get a 4 run bpc, that's a per run cost of 75 million isk, which is about what they sell for, which is at least 3 times what they should cost.



With the best chance decryptor you should have about 25% chance to get a 4-5 run bpc depending if you use max run BPC or not.

From personal experience of inventing Hulks, 111 runs of Covetor BPCs became 108 runs of Hulk BPCs using the max run decryptor. That means if you do enough inventions the Occators will cost 30-35M per run to invent.

Ittey
Whiskey Tango Foxtrot
Posted - 2007.05.24 22:01:00 - [44]
 

Originally by: Zhuge Liang
Good stuff, sticked with an unholy amount of glue.

Yay, my first sticky thread here.

Just a small update this time, fixed the typo in the Runs formula and added negative ME to the advanced page.

Originally by: Qual
Noticed you where looking for info on negtive ME as well, I did that formula some time ago here.

Added.

Originally by: Kdah
Your example should then read:

Runs = max(1;Round.Down(1000/1000*10) + 4)

Fixed, good catch indeed.

That is pretty much it, will be a while before I can make another edit so hope this will do.

Chrome Ozone
Posted - 2007.05.25 16:59:00 - [45]
 

Originally by: Kdah
Your example should then read:
Runs = max(1;Round.Down(1000/1000*10) + 4)

Fixed, good catch indeed.



Ok, I'm struggling to understand here. I took a 7 run ferox bpc, and with a +4 decryptor, got a 6 run nighthawk print. By the math from the guide it should look like

Runs = max(1;Round.Down(7/15*1)+4)= 4(after rounding)

I'm curious if anyone else has ever tried running a non-max bpc...something I did notice, if I switched the division, i.e. (15*1/7)+4 you actually get 6 after rounding down, BUT put in a max run so (15*1/15)+ 4, and you get 5. I'm throughly confused.


Frigid
Posted - 2007.05.28 15:55:00 - [46]
 

Good stuff!

One comment, you say in the skills section that the starship engineering skills are under Mechanic. Not so. All the datacore skills are science skills.

Phedra Kane
Posted - 2007.05.28 17:02:00 - [47]
 

so, on current economics, I think i've worked out that unless you want to invent very high value items, like T2 cruisers/BCs, then its not worth the effort. I was looking at inventing TII ship modules, retailing for 3-4m each. However, besides the cost of datacores, with the encryption skill going at 200mill, there is no chance of gaining a profit in any reasonable timescale.

Is this conclusion correct?

Anopheli
Pillowsoft
Total Comfort
Posted - 2007.05.28 18:18:00 - [48]
 

Originally by: Phedra Kane
so, on current economics, I think i've worked out that unless you want to invent very high value items, like T2 cruisers/BCs, then its not worth the effort. I was looking at inventing TII ship modules, retailing for 3-4m each. However, besides the cost of datacores, with the encryption skill going at 200mill, there is no chance of gaining a profit in any reasonable timescale.

Is this conclusion correct?


Not true. I've run up about a few hundred million profit in around a week. Modules have a much higher success rate than ships, and the key is not using decryptors on low value items, and seeking out things you can manufacture easily. Also avoiding modules that use Mechanical Engineering cores, as that's a good way to start flushing mountains of cash.

Sir Bart
Vendetta Underground
Rule of Three
Posted - 2007.05.28 18:30:00 - [49]
 

Why do you need to create a max run BPC? I thought BPO's could be used... surely a BPO doesn't give worse results than a max run BPC?

-Bart

Hafthor
Gallente
The Fated
E.Y
Posted - 2007.05.28 23:01:00 - [50]
 

Any chance of adding RP's needed per Datacore info into this? I've been scouring the forums for this info but failed miserably.

Sir Bart
Vendetta Underground
Rule of Three
Posted - 2007.05.29 05:37:00 - [51]
 

Originally by: Hafthor
Any chance of adding RP's needed per Datacore info into this? I've been scouring the forums for this info but failed miserably.


100 rp for ship ones and 50 rp for the rest.

Vanye Inovske
Two Brothers Mining Corp.
Friend or Enemy
Posted - 2007.05.31 03:37:00 - [52]
 

Originally by: Sir Bart
Originally by: Hafthor
Any chance of adding RP's needed per Datacore info into this? I've been scouring the forums for this info but failed miserably.


100 rp for ship ones and 50 rp for the rest.

Graviton Physics is 100/core as well. There may be others, I have no idea.

Chruker
Posted - 2007.06.04 11:26:00 - [53]
 

Originally by: Lady Rachel
Edited by: Lady Rachel on 22/05/2007 13:57:17
someone has a list where is described to the several the date interfacewho invention can make?
example
esoteric data interface: all missile lounchers, ballistic control sistem, all railguns,.

tnx


You can see it in my item database. Ex. http://games.chruker.dk/eve_online/item.php?typeID=25555

Mycroft Kencyrath
Gallente
Standard Ore
Posted - 2007.06.04 22:33:00 - [54]
 

Edited by: Mycroft Kencyrath on 04/06/2007 22:35:34
Edited by: Mycroft Kencyrath on 04/06/2007 22:35:00
Very Nice. Very Happy




"Are you an angel? Aw, I'm just kidding. That's the worst line I've ever used. Hope some poor kid doesn't start using it."
-Atton Rand

Magwua
Amarr
Rayden Industries
Posted - 2007.06.05 23:14:00 - [55]
 

Originally by: Sir Bart
Why do you need to create a max run BPC? I thought BPO's could be used... surely a BPO doesn't give worse results than a max run BPC?

-Bart

You cannot use a BPO to invent t2 goods, you must use a t1 bpc.

Dr Bernard
Posted - 2007.06.12 20:20:00 - [56]
 

Edited by: Dr Bernard on 12/06/2007 20:26:09
Great work.

Just 1 clarification. The MaxRunsOutputBPC you mention for ships is 1 and modules 10. Are there additional stats for other bp types ?

If it is the case that it is always 1 for ships then if I invent without decryptors, I only need to create 1 run inpt bpcs ? If I use a decryptor then I should always use max runs on the input bpc otherwise I will essentially lose 1 run in the output ?

Thx

Risar Surtr
Pagan Heavy Arms
Posted - 2007.06.13 02:32:00 - [57]
 

Edited by: Risar Surtr on 13/06/2007 02:32:21
Originally by: Dr Bernard
Edited by: Dr Bernard on 12/06/2007 20:26:09
Great work.

Just 1 clarification. The MaxRunsOutputBPC you mention for ships is 1 and modules 10. Are there additional stats for other bp types ?

If it is the case that it is always 1 for ships then if I invent without decryptors, I only need to create 1 run inpt bpcs ? If I use a decryptor then I should always use max runs on the input bpc otherwise I will essentially lose 1 run in the output ?

Thx


I'm sorta confused about this myself honestly. Is the max number of runs on a "invented" T2 bpc really so small ? It seems like in order to "invent" a T2 frigate or HAC (for example) that your going to have to spend 30-40m just to get a result, and a 1 run result at that ? After the cost of supplies to build the silly thing your not going to see much (if any) profit it would seem. Its almost like just doing bulk T1 manufacturing is more profitable for the time spent.

Its almost like invention isnt meant to be a way to make isk, but more of a way to create the things you need for your own use.

luain
Posted - 2007.06.19 15:35:00 - [58]
 

2nd attempt ship invention used 1 Vexor Ship, 1x 5 run Vexor BPC, 1x Test Reports Decryptor and 8x each Mech Eng,& Gallente Starship Eng Data Cores and got a 1 run Ishtar BPC....Was told T1 item did not improve results so repeated attempt 3 times without Vexor Ship...Result No Success...then tried twice more using Vexor Ship as optional item...no success....Question does the T1 Item have any effect on result?

Trak Cranker
Finn Inc
Posted - 2007.06.21 09:28:00 - [59]
 

It would be very strange if it didn't.

As you can see the exact formula is still up in the air, but I can hardly imagine that the metalevel of the T1 item does not play into it.

Tamori Kachimasu
Posted - 2007.06.21 13:28:00 - [60]
 

After some tries, I have following results (twice so much) :

Type : Frigate
Input : 30run BPC
Basetype item : T1 Frigate
Decryptor : None
Result : _1_ run BPC.



Pages: 1 [2] 3 4 5 6

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only