PDA

View Full Version : Grinding: Very High = 90%



rayner
Nov 14, 2006, 04:55 AM
When going to grind a weapn I've been trying to figure out a success rate percentage because "Very High" seems pretty vague. So I made a spreadsheet of all my attempts and what grinders I used ( C+1, C+2, C+3... ect ) and after 100 attempts I had exactly 90 successful upgrades. But what this also means is that I had 10 broken weapons, and who really only grinds to LV2 on a C Rank weapon anyway. So here are some other observations for "Very High" chance success-rate.

If you take 10 C Rank weapons and you want to upgrade them all to Lv 4 with Lv 0 luck... because Male Humans never get a break. The 1st grind is free... so all are lv 1. Then take those same 10 and then upgrade them to lv 2... 1 will break with the 90% success rate. Cool now you have 9 lv 2 weapons. Take them back again to upgrade and 1 more will break... leaving you with 8 lv 3 weapons. Then for lv 0 luck you don't really want to go more than 4 grinds because you lose that "Very High" chance... You bring those 8 back and 1 more will break. Leaving you with 7 Grinded lv 4 weapons out of 10 or 70%.

Another way to loo at it is that if you want to go to lv 4 Grind with 90 successful attempts on a C Rank Weapon with lv 0 luck that means 90/3 ( 1 Free & 3 Very High attempts at lv 0 luck ) = 30 weapons at lv 4... and 10 broken weapons. 30/40 = 75%

So for those that say "Very High" is lower than 90%... that's not true, it's just that you're grinding your weapon more than once. Make a chart and when you get to 100 attempts I bet you will have 90 successful tries + or - 3. Oh yeah and for some reason all 26 of my C+2 Grinders worked? Weird...

Oh this was done on the Xbox 360 version.

<font size=-1>[ This Message was edited by: Rubicon on 2006-11-14 01:59 ]</font>

Itsuki
Nov 14, 2006, 05:04 AM
Perhaps a test is in order. I could do an offline test of things later. Online, I don't think anyone would have the money to spare.

ViciousXUSMC
Nov 14, 2006, 05:04 AM
sorry but this means nothing, very high is an average general area of succsess probably in the range of 80-99% its not a set % because the higher grinder you use the better the % even if they are all rated very high.

and just because 90/100 worked doesnt mean its 90% it means you had a 90% average for your test but its still non conclusive and can not be held as a truth.

rayner
Nov 14, 2006, 05:12 AM
On 2006-11-14 02:04, ViciousXUSMC wrote:
sorry but this means nothing, very high is an average general area of succsess probably in the range of 80-99% its not a set % because the higher grinder you use the better the % even if they are all rated very high.

and just because 90/100 worked doesnt mean its 90% it means you had a 90% average for your test but its still non conclusive and can not be held as a truth.



Yeah I guess 90/100 = "average" sure... I'm already on my second test, we'll see if I get another 90/100

If it were 900/1000 would that be conclusive?

Oh and here are the grinders I've used and the chance:

C+1 = 10/12
C+2 = 26/26
C+3 = 25/26
C+4 - 25/31
C+5 = 4/5

I basically use my C+1 for the 1 free chance.

<font size=-1>[ This Message was edited by: Rubicon on 2006-11-14 02:22 ]</font>

ViciousXUSMC
Nov 14, 2006, 05:21 AM
definitly would be more conclusive http://www.pso-world.com/images/phpbb/icons/smiles/icon_razz.gif still its only an average you cant assume its a fixed rate.

etlitch
Nov 14, 2006, 08:54 AM
Online / offline succes chances for grinding are different. +5 Grinder always awards 100% grind succes on offline or so.

AngelLight
Nov 14, 2006, 09:05 AM
The only way you're going to get solid empirical data...enough to get a grind chart out....is to have several people try it at different luck states and different times. Pretty much, you need several people doing the same thing you just did. Also, you're going to want as many controls as possible in this experiment to try and weed out "human error".

--A set number of people (arbitrary, but prolly 10+)
--No more then 2 of any race/gender combo in the case of 10 total people. This would be higher with more people.
--The same rank on weapons (all Cs, all Bs, etc.)
--The same grinder rank and level (all C+1, all B+5, etc.)

Then you can figure in your variables: Luck Score, Different Races, Different grinder levels, etc.

When you have all of that you can start putting together a chart and poof, you're an instant legend in PSU circles ^_^.

The biggest problem you may have is getting enough people on board to do that, since alot of people are still grinding up characters (or even grinding up a 2nd character or more).

Don't give up now though, I think if you were able to get a real chart out you'd be pretty much loved by all.



<font size=-1>[ This Message was edited by: AngelLight on 2006-11-14 06:06 ]</font>

Pivotal
Nov 14, 2006, 09:39 AM
If there is a day with +3 luck, just make a char of that race/sex and you can upgrade GRM weapons with max luck. I ground my Cs to 7 or 8 and my Bs to 4 or 5 earlier today. I'm not sure what the reqs to transfer to Neudaiz are, so I'm not sure about Yohmei weapons.

Here is a partially complete grinding table: http://www.photohost.org/gallery/data/500/2032grindchart.gif