I'd have to agree with NegimaSonic and Awesomest. The highest value doesn't help so much because you won't do that damage until the last level. At the same time though the average doesn't help because it seems like you're incompetently recording numbers, "My sword doesn't do that low damage, it's doing X more (here at level 17, as opposed to 12 or 14)!" Providing the range lets you know "Ok, it'll start out here, but get to there at the end" and you know what to expect.

Also CrashFu, your name is familiar from another game, but I can't recall what.

Also agreed on range of Min - Max, that gives the best info without having too many numbers per stratum

Min-Max also lets you work out the average yourself as well.

Average damage doesn't work so well in Tier 2, which has an even number of floors in each stratum.

Min and Max would be better.

Are the damage numbers recorded at Lv1 or Lv10?

The damage numbers should be on level 10 weapons. I'm guessing a 5* Level 1 weapon starts off just above where a 4* Level 10 weapon ends, need more data to confirm though. So Level 10 damage numbers are needed to demonstrate the full potential of the weapon.

Remember the reason why we want damage numbers - to make informed decisions about weapon performance. When deciding upon a weapon there are many attribute to consider including aesthetics, swing style, weapon speed, mobility, and you also have damage potential. The latter is something we don't know a lot about except that the bars cannot be trusted.

In terms of what numbers are useful to track on the wiki, here are charts to give you an idea of the scaling of damage numbers...

Khorovod 4* Chart
Divine Avenger 5* vs Construct/Undead Chart

The values progress linearly within a Tier until the weapon reaches its intended strata. In the case of Khorovod, it is 4* and thus doesn't receive additional damage in the last strata. In the same manner, the damage increase of the DA has a higher slope in strata 5 than 6. This is probably due to a combination of the weapon nerfing based on depth and the increasing defense of monsters.

Edit: It just occurred to me, if we wanted to minimize the number of numbers, we could just have min-max per tier then mention which tier the weapon's damage maxes out. So...
Khorovod Hit 1: T1 43 - 64 T2 104-115 T3 169-229 Rated through Strata 5
DA vs Const/Undead Hit 1: T1 48-66 T2 128-144 T3 229-328 Rated through Strata 6

Half the numbers, but still lets you know when it maxes out.

Could it not be that players say "Not to trust the damage bars" because they're assuming they work linearly?

For instance, could it not be that a 3/4 full bar does e^(3/4) damage more than an empty bar? Though this would probably require some sort of base damage.

Once damage data is collected I would be more than happy to analyze it and see if there is any identifiable trends in the damage bars.

Do note, that this would require less data collection. We would keep the damage type the same, and solely from one source at first to make it simple. Attack speed is irrelevant for damage per hit purposes. Thus we'd only need the values of damage of say, a 1/4 full, 1/2 full, 3/4 full and full bar weapons on ONE kind of creature within the same depth. It might be interesting as well to see the damage from 'almost' equal damage bars of the same type on the same target. If this value differs by much then we really have a complicated situation but intuition tells me this shouldn't be the case.

Furthermore, as has been stated other places, the damage also depends on the depth, but identifiable trends should be able to be found if we keep the depth the same. And once we have a model for damage bar to damage on a fixed depth we can transfer that to other depths fairly easily.

@GeminiDreams
Once damage data is collected I would be more than happy to analyze it and see if there is any identifiable trends in the damage bars.

I'm willing to bet good CE that the data will reveal the bars track BS levels. >D

I mean, yeah, a faust is stronger than a wrench wand. The bars hold true for that. But if you put two equal barred weapons head to head you'll get sometimes vastly different damage results. It's possible that the bars are a DPS rating though which would make sense why a dread venom striker would be even remotely close to a leviathan blade despite them doing vastly different amounts of damage per hit.

Definitely Champions Online. I was the guy with the lion with the growth complex, exploiting all the bugs to make a character that could get to 4 times the size of the tallest average enemy. >D

I agree that the numbers should be per strata rather than across the whole dungeon stretch. The data you've got now seems to hold this as important fact. I thought it'd be nice if there were a graph too the way culture presented it, though I'd want some way to automate the graph output process rather than forcing someone to draw graphs by hand and trusting that it would remain consistent between two people (or even two graphs from the same person).

Editing the wiki is as easy as editing your post on this forum. :) And it has a preview function just like here, so you can see if you'd make the universe implode before you actually do so.

This makes me believe a previous poster's idea that the bars work as a dps indicator instead of a damage per hit indicator.

If this is the case, however, the data we're collecting thus far will not reveal the link since we're recording damage per hit. Has anyone tried to record much data on damage per second? I realize this would be more difficult than damage per hit however.

Edit:

I have a hard time believing that the bars are inconsistent because it would be so easy to show them correctly and I don't see what Spiral Knights gains from possibly displaying them incorrectly. Imo they are not inconsistent we just don't know in what way we have to look at them for them to be consistent.

Min-Max.

Also, fancy work.

Now, if we could somehow do this for shields.

Then perhaps it takes into consideration their charge attack damage or dps too?

All I'm saying is that I don't believe that the creators would purposely design the bars as inconsistent. Though it might be a software glitch or just something they did wrong maybe.

Here's what I want to see:

max Normal swing damage.

While In dungeon:
max NSD*multiplier=current NSD

edit: when damage is capped in a weird way (like 5* weapons in t2 will do the same damage regardless of level) then I'd like to see "max base damage for t2" or something instead. while in such a dungeon.

Could it not be that players say "Not to trust the damage bars" because they're assuming they work linearly?

Like I said in a different thread, I don't believe the bars work that way. I think it's just that the devs got lazy and used a static bar image on the weapons tooltip which they're supposed to update every time they decide to change the weapon stats (but haven't), instead of using a bar whose size scales with the number that represents the swords power.

Why do I think this? K.I.S.S rule.

Except the KISS rule would be a good advocate for programmatically generating bars.

It's less time consuming on the back end of things to write one routine to make a bar than it is to draw custom bars for every item. There are 52 swords alone, and generating those bars by hand would take anywhere from 30 seconds (assuming filters and templates are already set up) to 2 minutes (drawing the bars and their dividers by hand, and adding icons for damage types) per weapon. I'd rather spend half an hour writing a routine to draw bars so that all you have to do is change numbers for the weapon's stats and it is reflected instantly.