View Full Version : Little redux q&a?
Andrath
07-03-2006, 01:22 PM
How exactly can I find out the percent of redux I have?..I've never had to do it before.
Stunseed
07-03-2006, 03:29 PM
www.xygon.net has a redux calculator. It isn't accurate, but it will give you a generalized idea.
Redux as we know it is gone. I'll leave Latrinsorm to post his findings/beliefs in a heavily articulated paragraphical form.
GuildRat
07-03-2006, 03:46 PM
www.xygon.net has a redux calculator. It isn't accurate, but it will give you a generalized idea.
Redux as we know it is gone. I'll leave Latrinsorm to post his findings/beliefs in a heavily articulated paragraphical form.
Good call. Lanistrom's been doing the crunching on redux, and is still a bit confused as there seems to be a randomization built into the basic formula.
StrayRogue
07-03-2006, 03:56 PM
Latrinstorm's Redux means shit, to be honest. I'd be much happier with a generalized % of damage reduced.
Latrinsorm
07-03-2006, 09:35 PM
Preface: Remarks.
The reason mine's better is because it actually says what's happening; namely, crit damage, raw damage, and crit divisor are reduced. DFRedux only accounts for raw damage, therefore it produces wildly inconsistent numbers for the same skillset. You can very easily see a 50% and a 70% DFRedux reading from the same skillset.
There's nothing left in the redux area that confuses me, just things I don't quite know yet (the relationship between RF and CDDF and the relationship between redux points and redux factor). I recognize that those are two pretty important topics, but it's not really confounding, just unknown. The thing that's been bugging me recently is Jinsem has this 10% raw damage reduction armor *and* redux, and let me tell you it's an incredible pain in the neck trying to figure out how/if they work together. If it wasn't for that Mark fellow on the officials I'd probably still be completely stumped.
Synopsis: The Cliffnotes Version
DFRedux bad.
Damage redux (redux factor) good.
1 - (actual total damage received / total unreduced damage) = redux factor
Body: GemStone IV and Redux (or, How I Learned to Stop Worrying and Love Crit Damage Reduction).
In GemStone III, there was a property known as Damage Factor Reduction, or DFRedux for short. This property reduced the damage factor of incident weaponry, rendering the attackee less vulnerable to DF-based assaults. How it worked was pretty simple. Training in a certain set of skills would either promote or penalize redux. A mostly accurate formula was obtained for converting the amount of redux-gaining skills to redux with penalties based on level and spells known. The results of this formula was given as a %, such as 60%. It applied in everyday GemStone as follows:
Every weapon has a specific, static DF versus every Armor Group (neglecting things like paladin spells and perfectly forged weapons. Handaxe vs. Plate, for instance, is .210. A DFRedux value of 60% meant that every DF was reduced by 60%. Therefore, any handaxe wielded against a plate-wearer with 60% redux would actually have a DF of .084; a marked improvement. This was the full extent of redux's effects.
.
With the advent of GemStone IV, this scheme was replaced by Damage Reduction. Some (not all) of the website documentation was changed to reflect this fact. However, nobody was sharp enough to pick up this subtle and inconsistent change in terminology, so the player community continued on assuming that DFRedux existed just like it had in GSIII. I think the first to publically realize that something was amiss was Porcell, who was doing pioneering work on the relative value of skills with regards to redux (coining the term "redux points" in the process). The method of determining redux involved finding the range of damage factors that could produce any given hit, then finding the overlapping regions of multiple hits. Eventually, this would produce a single damage factor (out to the thousandths place). In GSIV, this was no longer the case. DFRedux was quite wobbly, at times obnoxiously slow. In what seemed at the time as an unrelated problem, certain warriors began noticing that hits that shouldn't have been causing wounds were, in fact, causing wounds. The crit divisor (what determines how much raw damage is required to cause a particular crit) had also apparently become quite wobbly. The official explanation at the time was that certain crits had been applied as the negative of their actual value, though I've since grown skeptical of this.
This all came to a head when Nodyre inexplicably asked Jolena and Stunseed to produce hits from a storebought dagger on the torso of a storebought metal breastplate. I say inexplicable because under the DFRedux paradigm one wants hits that do a lot of raw damage, as this makes the range of possible DFs smaller. These hits were absolutely unbelievable under any GSIII understanding of combat. One even appeared to have 0(!!!!!) raw damage. At this point, the player base advanced and rejected a number of hypotheses, from natural damage padding to natural negative damage weighting to a nebulous, undefined "wonkiness" that occurred on low-DF hits (I advanced the last one, and I like saying "wonkiness". Wonkiness!).
Later on, I realized that the only rational explanation was that crit damage was also reduced in much the same fashion as raw damage. Given this remarkable new capacity for redux, I decided to rename the quantity "redux factor" and post it in the form of a decimel out to three places, much like a damage factor. The main purpose for this was to avoid confusion with the quantity "redux %", which I felt would be indistinguishable with the DFRedux idea. This is why the quantity is now referred to as "Damage Reduction"; it reduces all damage, not just raw.
Unfortunately, this also failed to explain a number of the hits that were seen. It so happens that crit divisors are also reduced on people with redux, though not to the extent that damage is. I've yet to determine the exact relationship between the two, but I'm quite certain that they are directly proportional throughout most if not all of the redux progression.
Given all this, calculating redux is still pretty simple. Here's what you do:
1) Absolutely ensure that no damage weighting or damage padding exists in the trials. Some creatures appear to have natural damage weighting. It's better to have another PC attack you anyway, as you can (hopefully) trust one to aim away from your vital areas.
2) Ensure that both the weapon and the armor have no peculiarities to them that will cloud your results. The 10% damage resistant armor I mentioned earlier is a prime example of this. I strongly recommend purchasing plain weaponry from an NPC store for any and all redux tests.
3) Get hit at least 5 times. Redux is somewhat randomized (just like everything else in GS), but 5 samples will give you a good idea of where you are. Endrolls should be at least 150, preferably above 200. Use a high DF weapon (no closed fists vs. plate).
4) On each hit, calculate the damage you should have received. You will need a DF table and a crit table to calculate this. Multiply (endroll-100) by the DF; this is your unreduced raw damage. Simply locate the crit message in your crit table; this is your unreduced crit damage. Sum them; this is your total unreduced damage.
5) Divide the damage you did recive (...and hits for XX points of damage!) by your total unreduced damage. Subtract this decimel from 1. This is your redux factor for each particular hit.
It is not currently possible to calculate directly from skills (via redux points) to redux factor. The exact impact of spells is unknown, but singling in spells does not preclude one from having appreciable redux.
I think the first to publically realize that something was amiss was Porcell, who was doing pioneering work on the relative value of skills with regards to redux (coining the term "redux points" in the process). The method of determining redux involved finding the range of damage factors that could produce any given hit, then finding the overlapping regions of multiple hits. Eventually, this would produce a single damage factor (out to the thousandths place). In GSIV, this was no longer the case.
And once I figured this out was the time I stopped giving a shit about redux.
In GS3 I could peg redux down to several decimal places. Once I started seeing swings that show 52.36% - 53.46% and the very next swnig is 53.86%-55.14%, that's when I stopped caring.
So you can do all the fancy math that you want, but it really doesn't matter. There's so many factors that don't seem to make any sense.
If you ever really figure something out and can say with certainty, I might read up on it. Until that happens, I'm totally skipping anything that discusses redux.
-OoK
Latrinsorm
07-04-2006, 12:05 AM
There's deviation with redux factor, but it's nowhere near as bad as the redux % numbers are. I had like 250 hits on my warrior and the standard deviation was .01something (compared to .04+ on redux values on a specific body part, let alone all told). I even made a graph showing the randomization:
http://img.photobucket.com/albums/v456/johnnyoldschool/urs2v5.gif
Not too shabby for a couple hundred. The last big spike on the left is -.042 and on the right is +.04. I don't know with certainty if any of this is 100% correct, but I do know with certainty that we'll never figure it out if we don't try.
edit: Addendum: I don't know of a single part of the redux factor hypothesis that either doesn't make sense or doesn't explain any empirical hit. We are not in a state of qualitative befuddlement.
It is not currently possible to calculate directly from skills (via redux points) to redux factor. The exact impact of spells is unknown, but singling in spells does not preclude one from having appreciable redux.
I'm pretty sure training in spell under 2x does nothing at all to your redux except take points away that could go into redux skills.
I'm not sure about this, but from a little testing I think it's right.
I think if you had a warrior with 1.99x spells and another with 0 spells and the exact same skills otherwise their redux would be the same.
Latrinsorm
07-04-2006, 11:23 AM
I'm pretty sure spells have a penalty. Satira (she's the best!) let me do some testing on one of her characters that happened to be level 52, and I got a really nice linear progression from redux points to redux factor (for a while, anyway). I also happen to have a data point from a level 54 character with 60 spells (I think from Nuadjha). The second character had 170.4 redux point and a redux factor of .148. The level 52 character would have had an RF of approximately .287. My research has also indicated that redux points "mean" more to a higher level character, so even if the 2 levels make an appreciable difference they'd favor the spell-having character.
That said, I also have a number of very high level hits that all show roughly the same redux factor, spells or no spells. Whether this means there's a cap on redux at ~.540 or the spellless character's results were tainted somehow, I don't know. Now that I think about it, it could also be that the old "1 spell per 10 levels" thing is still in effect, as that would very neatly eliminate any spell penalty on the spell-having characters.
In short, having a lot of spells will definitely penalize redux, but having a few might not.
Bobmuhthol
07-04-2006, 11:39 AM
It wasn't *that* long ago that Simu said 1 spell per 20 levels was the limit. And I think it was in GSIV, too.
Chiv had 60 spells at 40 and his redux was generally the same as other rangers with a similar number of redux points as he had. Not terribly scientific, but there it is.
Latrinsorm
07-04-2006, 04:26 PM
If you were using redux % or only had a couple samples per target, those findings wouldn't surprise me. The other rangers being lower level would also explain it. Obviously I'd like to see the data points for my records if you still have them handy.
Andrath
07-06-2006, 06:21 PM
ok, so im lost...is there a way to calculate it to the closest known predictions of DFRedux
Bobmuhthol
07-06-2006, 06:42 PM
DFRedux no longer exists.
Damage Reduction does.
Latrinsorm
07-06-2006, 10:29 PM
Redux cannot be calculated from skills and level alone. Depending on your level, I might be able to give you a rough guess.
Redux can be calculated from a hit very easily. Redux factor is defined as (1 - (actual damage taken/theoretical damage calculated)). To find out the theoretical damage calculated, all you need is a DF table and a crit table. Find the DF of the weapon that attacked you versus your armor group and multiply it by endroll-100. Look up the crit message that your hit generated. Add the two values.
Powered by vBulletin® Version 4.2.5 Copyright © 2024 vBulletin Solutions Inc. All rights reserved.