[X] My Assistant

The equipment stats are all there in the page, so you could make a userscript that applies any scoring function you like, displays the numerical score next to each piece, and lets you sort the page by equipment score and sell everything below a particular cut-off value. You could display two or three numbers, reflecting the values to different builds, or in different theories of Hentaiverse, and sort by any one of them.

If one of the alternative scoring metrics ever gains wide acceptance, people can start listing "Equipment qs 650+" instead of "Exq+" in their thread titles, and have the shop helper script blat the values into the listings for them.

This is taking the approach that drop quality doesn't need to improve, and just making it easier to pick out the good pieces.

Looking at the other debates, I don't expect people to come to close agreement on a good scoring function.

This post has been edited by mechafujoshi: Mar 6 2013, 06:23

How is equip tier determined?
Wiki seems to imply that tier and quality is independent, how I feel that they are somewhat correlated.

How is equip tier determined?
Wiki seems to imply that tier and quality is independent, how I feel that they are somewhat correlated.

T3 has better chances of dropping with PABs thus give better quality in average.

Because of this items are completely unbalanced, more stats to roll means quality has to be higher to get decent stats.

Wrong. Assuming every stat has equal weight and is continuous, the number of stats has nothing to do with the quality, which is the average of all stat's bonuses. (The average of 10 dice roll should equal to the average of 100 dice rolls)

It's
1. The discrete (and with large spacing) nature of base rolls
2. the different weight among stats
that would probably affect the average.

How is equip tier determined?
Wiki seems to imply that tier and quality is independent, how I feel that they are somewhat correlated.

Probably because more T3 are produced in higher difficulty. You can check if that's true by recording the equips from shrining and plot the pattern.

Thus supporting my theory that average quality increase goes 100% towards raising expectations.

People spend most of their time whining. Q.E.D.

This post has been edited by varst: Mar 6 2013, 14:23

Wrong. Assuming every stat has equal weight and is continuous, the number of stats has nothing to do with the quality, which is the average of all stat's bonuses. (The average of 10 dice roll should equal to the average of 100 dice rolls)

It's
1. The discrete (and with large spacing) nature of base rolls
2. the different weight among stats
that would probably affect the average.

Even assuming every stat has equal weight and is continuous (for the purposes of determining HV quality), that does not mean 3 Superiors have equal stats. The one may, indeed, have above-average rolls over all its stats, the other may have awesome Legendary rolls on its most desirable stats and crap rolls on the others, and the 3rd may have Magnificent rolls on its less desirable stats and average rolls on its most desired stats.

From a very strict, technical POV, you're right, of course. But a Fine Phase that had awesome EDB/INT/WIS might be preferable to a Superior Phase of the same suffix with lower EDB/INT/WIS, but with mucho high DEX/AGI to compensate.

Which, if that is what you meant by different weightage, OK. Then I agree with you.

Wrong. Assuming every stat has equal weight and is continuous, the number of stats has nothing to do with the quality, which is the average of all stat's bonuses. (The average of 10 dice roll should equal to the average of 100 dice rolls)

You know what I mean....
Or are you trying to suggest its just as easy getting decent stats on a superior plate vs superior kevlar?

Even assuming every stat has equal weight and is continuous (for the purposes of determining HV quality), that does not mean 3 Superiors have equal stats.

Or are you trying to suggest its just as easy getting decent stats on a superior plate vs superior kevlar?

Both of you are refering to a 'decent stats'. So what's a decent stats? This changes across patches. Something that's good before may be nerfed, something that's bad may be buffed. Not to mention how people value stats differently. The current system is okay as it let the player's market to determine what's more valuable.

Both of you are refering to a 'decent stats'. So what's a decent stats? This changes across patches. Something that's good before may be nerfed, something that's bad may be buffed. Not to mention how people value stats differently. The current system is okay as it let the player's market to determine what's more valuable.

Nope, check my post again. Never once mentioned 'decent'.

Thing is, I used the word 'desired' and 'desirable', which as you rightly point out, is market-driven and based on actual usage/utility. Someone like MikukoAya desires longswords (God help the poor fellow) and does not desire estocs. The (most of the) rest of us Heavy Melees think he's barmy. But that's his right.

I think even MikukoAya will agree, however, that the whole point of a Legendary Ethereal Estoc of Slaughter, however, is:

1. 0 Burden & Interference
2. High ADB
3. High proc % and duration
4. Super awesome STR/END and it would be great if DEX/AGI was pretty all right too.

Certainly, such a weapon with above stats, regardless of how the market regarded it in relation to other types of 2H weapons, would be far more saleable than an 'equivalent' Legendary Ethereal Estoc of Slaughter that:

1. Had the lowest roll possible for ADB in that class
2. Had 25% PA chance... for 2 rounds

Won't you say?

Similarly, a Balance anything is valued for its Accuracy and Crit rate. If I had a strange fetish for Battlecaster gear (ooh, kinky!), ceteris paribus I would pick the one with the highest mana conservation as long as its primary stats' not entirely hopeless.

It's easy to weigh when you only have a few stats to compare. If I was deciding between Legendary Power Slaughter and Legendary Plate Protection, then it's rather obvious that at a minimum, the Plate Protection's PMI had better be significantly higher than the Power Slaughter's.

But if you had, say, a Staff, which you valued for a particular few stats (maybe MDB + EDB), then you might get extra frustrated looking for an Exquisite which had just the correct combination you wanted, simply because there is a higher chance that all the other stats, less desirable, rolled better.

And let's be honest - for any given patch, the vast majority of players are going after the exact same stats for any given equipment prefix and suffix. Anyone sane who's going after Gossamer Heaven-Sent isn't going after it for its DEX PAB more than its Holy prof bonus.

I don't think that it's too much to ask that at a minimum, Proc, Prefix and Suffix stats are given higher priority when deciding what quality an item is. Maybe they are, I don't know, but my experience is generally that they're not.

How is equip tier determined?

http://ehwiki.org/wiki/Loot_Drop_Rolls

Difficulty; that's it.

1. Have you ever seen an legendary estoc with crude ADB roll? Or with less than 5 turns PA?
What I have seen from players' common complaints:
- Why's my legendary only has mag grade ADB?
- Why's my legendary with 3 PABs instead of 4?
i.e. people complain when their leg equips roll in the mag range, or when their ex equip roll in the superior range. Which I think is far from your description.

2. Believe or not, the existence of suffix/prefix already gives you better chance to get something relevant.
For example, assume a blank power armor (without suffix) of total quality score of 100, with ADB contributing to 10% of the total quality (i.e. 10). With slaughter bonus contributing another 10 score, the total portion of quality of ADB among the total quality will be 10+10/100+10 = 18%, which is larger than 10%.
i.e. Existence of suffix/prefix should already give you extra value for that particular stat boosted.

3. As for case of staff/power/plate you've mentioned, it really goes down to how many stats you would like to be good (which is already something about desired stats). People are more demanding on staff's stats (i.e. they want more stats to be good), so of course they're going to have difficult time finding a good one.

4. So assuming Tenb increase the portion of AGI before 0.7.4 and decrease them after 0.7.4, would people simply accept that as 'all right, it's a new patch, everything should change' or 'WHY WOULD YOU NERF OUR AGI EQUIPS?'

5. Prove your 'experience' with hard data.

This post has been edited by varst: Mar 6 2013, 18:46

that the whole point of a Legendary Ethereal Estoc of Slaughter, however, is:

4. Super awesome STR/END and it would be great if DEX/AGI was pretty all right too.

Two-hand weapons can only roll STR/DEX/AGI, not endurance. Only way a TH weapon can get endurance is through potencies. Most people want awesome STR/DEX on an estoc cause they add to damage. AGI is nice cause it increases your odds of good potencies, but otherwise it's pretty useless for heavy melee after .74.

This post has been edited by T_Starrk: Mar 6 2013, 19:41

Both of you are refering to a 'decent stats'. So what's a decent stats? This changes across patches. Something that's good before may be nerfed, something that's bad may be buffed. Not to mention how people value stats differently. The current system is okay as it let the player's market to determine what's more valuable.

Of course I am referring to decent stats... The point I was making was about it being more difficult to get decent stats on an item with more stats still stands.
More stats to roll = more chance of getting stats with low values or them not appearing, which makes for a large unbalance between different items, which is why I used the example of plate vs kevlar.

What I have seen from players' common complaints:
- Why's my legendary only has mag grade ADB?
- Why's my legendary with 3 PABs instead of 4?
i.e. people complain when their leg equips roll in the mag range, or when their ex equip roll in the superior range. Which I think is far from your description.

And I believe these are fair complaints to make about the system, and I suggested a way to balance that

2. Believe or not, the existence of suffix/prefix already gives you better chance to get something relevant.

It still does not change the fact that things like this happen. Just because it boosts one stat does not mean it can make up for crap rolls on other stats. Which is why I suggested that the range for rolls on stats be tightened with increasing quality to give more balanced stats. The large range for low-mid quality equipment gives players the chance to find a superior item that they can use, as it allows that with a few good rolls on important stats the item is usable, but for mid-high equips this is a drawback as it allows for 1 or 2 bad rolls to make a mid-high quality item unusable (or undesirable if you prefer). Allowing stat range to tighten with increasing quality would further reduce and hopefully eliminate at legendary the chance of getting items with the important stats low or missing.
I imagine Tenboro would probably say that this would simply raise peoples expectations but I think it would be fair that increasing quality reduced the lower range of stats.

It seems like the easiest way to solve all of these issues would be to ask 10B to change the order that equip loot rolls happen - i.e. Slot, Tier, Suffix, Prefix/Ethereal/Elemental, Quality, and the # of PABs are all determined, THEN determine where the equip falls in terms of individual stats.

One way to fairly distribute the actual rolls would be to randomly order all of the stats to be determined, then roll for each stat within a pre-determined range, which is further narrowed by the average of all the stats that have already been rolled.
Another way to put it: Using the PXP/quality number, the average value across all the stats has to be within a certain range to still be Exq versus Mag. You use that number in conjunction with set ranges (i.e. a raised minimum roll for some/all stats at that quality level), or possibly have a sliding range that gets narrower based on what values have already been used (mostly low rolls before, minimum goes up, mostly high, maximum goes down) to determine what each stat should be.

I was imagining the equation would be along the lines of going from the average to something calculable, so from quality = (stat1 + stat2 + ... + statN)/N
to N*quality = stat1 + ... + statN
where once you've rolled statX, you subtract it from N*quality, and then redo the calculation, using the new adjusted average like (N*quality - stat1)/(N-1) = (stat2 + ... + statN)/(N-1), but I can't figure out how to describe the way to adjust the stat rolls to reflect the "new" average for the remaining stats.

... Gah. I just wasted a bunch of time doing random math searches, since I know what I want to explain is possible, but I don't remember enough statistics and probability to recall how to describe it. The easiest way to do it would be to use the exact quality that you already rolled, and then roll each stat individually, modifying the later one's stat ranges to ensure the final average will be equal to the PXP/quality total. I just got caught up in trying to create an equation that would represent the somewhat recursive nature of this issue.
The reason I recommend randomizing the order that stats are assigned, is to prevent having max attack/crit/mit guaranteeing that your PABs and elements/proficiencies will suck - it can still happen, but it's less likely.

Sorry if this was slightly incoherent and/or redundant, but I did leave this post multiple times, and I think that I still convey my idea. Also, I'm too lazy to retype this and streamline it.

1. Have you ever seen an legendary estoc with crude ADB roll? Or with less than 5 turns PA?
What I have seen from players' common complaints:
- Why's my legendary only has mag grade ADB? I have seen ones with exquisite ADB. I have seen plenty of legendary phase with not even superior max EDB and PABs
- Why's my legendary with 3 PABs instead of 4?
i.e. people complain when their leg equips roll in the mag range, or when their ex equip roll in the superior range. Which I think is far from your description.

2. Believe or not, the existence of suffix/prefix already gives you better chance to get something relevant.
For example, assume a blank power armor (without suffix) of total quality score of 100, with ADB contributing to 10% of the total quality (i.e. 10). With slaughter bonus contributing another 10 score, the total portion of quality of ADB among the total quality will be 10+10/100+10 = 18%, which is larger than 10%.
i.e. Existence of suffix/prefix should already give you extra value for that particular stat boosted.
Only for certain types. For phases it's irrelevant. So is plate I think (only 2 suffix now IIRC).

3. As for case of staff/power/plate you've mentioned, it really goes down to how many stats you would like to be good (which is already something about desired stats). People are more demanding on staff's stats (i.e. they want more stats to be good), so of course they're going to have difficult time finding a good one.

4. So assuming Tenb increase the portion of AGI before 0.7.4 and decrease them after 0.7.4, would people simply accept that as 'all right, it's a new patch, everything should change' or 'WHY WOULD YOU NERF OUR AGI EQUIPS?'

5. Prove your 'experience' with hard data.

This post has been edited by ChosenUno: Mar 7 2013, 01:50

Can we have some form of trophy merging?

For example: Boss class trophies, you can merge 10 of those to get 1 schoolgirl/legendary class trophy. Then 20 of legendary class trophy can be merged to form 1 god class trophy.

So, my proposal:
Class | Required | Reward |
Boss | 10 | Konata-class |
Konata | 5 | Yuki-class |
Yuki | 10 | Sapling/T-Shirt |
Sapling | 5 | Unicorn Horn |
Horn | 5 | Noodly |
Noodly | x | Special trophy! |

Of course I am referring to decent stats... The point I was making was about it being more difficult to get decent stats on an item with more stats still stands.
[...]

But in the end you're only complaining about the range of rolls. And the truth is tightening the ranges would change absolutely nothing about your whining, since you can define range only through available equips.
To be more simple, if the ranges were tighter, you would still be comparing your equip to others, to determine its value. And our individual buying/selling behaviours determine the price we give that value.
So in the end nothing would change.

It seems like the easiest way to solve all of these issues would be to ask 10B to change the order that equip loot rolls happen - i.e. Slot, Tier, Suffix, Prefix/Ethereal/Elemental, Quality, and the # of PABs are all determined, THEN determine where the equip falls in terms of individual stats.

The loot generating system is quite complex, but if I'm not mistaken, what varst said was that this is already happening. Again, you're wishing for a higher value for same quality loot, which cannot happen since we base value not only on quality but also on availability.

I imagine Tenboro would probably say that this would simply raise peoples expectations but I think it would be fair that increasing quality reduced the lower range of stats.

... And yet both of you are actually kinda proving his point.

But in the end you're only complaining about the range of rolls. And the truth is tightening the ranges would change absolutely nothing about your whining, since you can define range only through available equips.
To be more simple, if the ranges were tighter, you would still be comparing your equip to others, to determine its value. And our individual buying/selling behaviours determine the price we give that value.
So in the end nothing would change.
The loot generating system is quite complex, but if I'm not mistaken, what varst said was that this is already happening. Again, you're wishing for a higher value for same quality loot, which cannot happen since we base value not only on quality but also on availability.
... And yet both of you are actually kinda proving his point.

What I said about tightening the ranges was for items of higher quality so the lowest value a stat could go would lessen with increasing quality. I never mentioned anything about buying/selling or what price an item is because I do not trade.
But yes lots would change if the lowest possible value a stat could go lessened as quality improved because it would mean the likelihood of getting a exquisite cotton piece with no intelligence like I posted would be either far less likely to happen or no longer happen depending on how it was set.
I do not know much about statistics but one would imagine that reducing the lowest range a stat could go would not have as large an impact on trading value as you think, because it would also lessen the likelihood of getting items with the non-desirable stats at their lowest and all the desirable stats at their highest.
My suggestion is about evening out the rolls so we do not have as large a variation between the quality tiers like we do now because of which stats are important and which are not as important either now or in the future.
And the point I was making about the higher the stats an item has the harder it is to find a decent one still stands. I do not have a suggestion as to how to balance it however. I believe this is also one of the reasons light armor is not used as much because it is so hard to find one with the "desired" stats.

Tightening the range on rolls would just make equip drops more boring. Equip drops would be less unique which is a bad thing, imo. Quality and drops are fine how they are.

This post has been edited by T_Starrk: Mar 7 2013, 04:42

Tightening the range on rolls would just make equip drops more boring. Equip drops would be less unique which is a bad thing, imo. Quality and drops are fine how they are.

How about instead of tightening them he just feeds them to me

Can we have some form of trophy merging?

For example: Boss class trophies, you can merge 10 of those to get 1 schoolgirl/legendary class trophy. Then 20 of legendary class trophy can be merged to form 1 god class trophy.

So, my proposal:
Class | Required | Reward |
Boss | 10 | Konata-class |
Konata | 5 | Yuki-class |
Yuki | 10 | Sapling/T-Shirt |
Sapling | 5 | Unicorn Horn |
Horn | 5 | Noodly |
Noodly | x | Special trophy! |

Actually, I wouldn't mind if there was a separate merged trophy for each tier, that gives an improved quality bonus or something, just over the next tier of trophies that drop. I think that fewer might be better, but that's just me.
My proposal
Tier | Required | Reward
Boss | 5 | Bronze Trophy
Schoolgirl | 4 | Silver Trophy
Ultimate | 3 | Gold Trophy
God | 2 | Platinum Trophy
Platinum | 2 | Awesome Trophy

Each can create the next tier up by merging three together, or directly with one less than the amount needed to create the next tier up. (i.e. if 5 bosses make a bronze, then 12 make a Silver, 33 make a Gold, 96 make a Platinum, and 285 make an Awesome, which is 120 less than doing it one by one.)