|
Post by FunkySwerve on Apr 16, 2010 5:27:51 GMT
I see there was no discussion of my methodology to find out the benefits of iteration. No, your failure to 'do the math' made it seem unworthwhile. Not all hits are created equal, as a high-ab hit is more likely to crit. This is likely at least one of the reasons why you think our calculations 'overvalue' higher iteration values as a penalty. Given the amount of effort we already invested in developing those figures, and the additional amount of work it would take to recall exactly how we arrived at them, it would take something MUCH more rigorous to make me reconsider them. Funky
|
|
|
Post by kaezar on Apr 16, 2010 10:45:34 GMT
Even rested, I still fail to see your point.
Not all hits are created equal, I see that. What I don't understand is why is that germane to the question. If a hit is a critical, say, 15% of the time, it'll be a hit in that proportion for both iteration 3, 4, etc.
I think I see your point with the hit percent changing depending on AB, AC and curse variation, and I will do a more through analysis.
And although it wasn't mentioned, I think for characters with lower BAB at 20, it may show variations, I'll include that on calculations.
But I have got to say, I am pretty sure the number won't show much variance with the values we are inputing. And I am not sure *at all* that they will show *any* change even with radically different numbers.
I don't have time to really think about it now, since I have to leave for work, but I'll include all that in the next sheet. Since I'll probably have to travel saturday, I won't have this ready until sunday or next week.
Take care Kaezar
P.S. On the 1/2 str mod to weapons smaller than the character, does that mean that any character that uses a weapon smaller than his size will be getting 1/2 strength mod? Including criticals?
Will that include even, say, weapon and shield, non halfling, non-gnome rogues? Because all rogue weapons but morning star and club (both not finesseable) are small, iirc.
That will be a big hit for those kind of characters that already got quite beaten up..
Take care K.
|
|
|
Post by jonuhey on Apr 16, 2010 11:12:35 GMT
______ Im not even sure about all other numbers in this thread, as I didnt read most of the posts, but this statement is quite clearly wrong. Imagine that both my fist monk (progression 3) and my ranger (progression 4) had 15% of critical chance. For you it would be the same, but no. The ranger gets 5 base attacks each round and the monk gets 6. Now after 100 rounds, the ranger had 500 attacks and the monk 600. If you apply the 15% now you would see that the ranger had 75 criticals, while the monk had 90 (applying directly the 15%, what doesnt really happens in game).
|
|
|
Post by MightyKhan on Apr 16, 2010 12:38:12 GMT
what maximum BAB values do the iterations use? if this is 20, I get: 7 : 1/8/15 6 : 1/7/13/19 5 : 1/6/11/16 4 : 1/5/9/13/17 3 : 1/4/7/10/13/16/19
if this is 16: 7 : 1/8/15 6 : 1/7/13 5 : 1/6/11/16 4 : 1/5/9/13 3 : 1/4/7/10/13/16
|
|
|
Post by FunkySwerve on Apr 16, 2010 13:00:55 GMT
Not all hits are created equal, I see that. What I don't understand is why is that germane to the question. If a hit is a critical, say, 15% of the time, it'll be a hit in that proportion for both iteration 3, 4, etc. It's germane because the question is NOT how many hits the character will land with weapon X, but how much damage they will do independent of external factors like conceal - that's the entire point of the sheet, comparing relative damage. Crits do more damage, and lower iteration weapons get more ab (and therefore a better chance at critting), not just more hits, because each iteration subtracts less. Funky
|
|
|
Post by MightyKhan on Apr 16, 2010 14:07:27 GMT
letting AB range from 90 to 100 in steps of 2 and AC from 85 to 95 in steps of 2 and using: 7 : 1/8/15 6 : 1/7/13/19 5 : 1/6/11/16 4 : 1/5/9/13/17 3 : 1/4/7/10/13/16/19 for attack progressions and including 1 haste attack i got: iteration 3 : 149 iteration 4 : 117 iteration 5 : 100 iteration 6 : 94 iteration 7 : 82
this was just hits (including autofail and autohit), no crits involved
|
|
|
Post by kaezar on Apr 16, 2010 14:08:02 GMT
______ Im not even sure about all other numbers in this thread, as I didnt read most of the posts, but this statement is quite clearly wrong. Imagine that both my fist monk (progression 3) and my ranger (progression 4) had 15% of critical chance. For you it would be the same, but no. The ranger gets 5 base attacks each round and the monk gets 6. Now after 100 rounds, the ranger had 500 attacks and the monk 600. If you apply the 15% now you would see that the ranger had 75 criticals, while the monk had 90 (applying directly the 15%, what doesnt really happens in game). All that is right. My point is that, if all the other factors are equal (keen, IC, OC, DC, WM, PC, etc), the damage increase will be on the same proportion. The gain in damage (I'm not denying there will be a gain in damage, that would be ridiculous), will be on the same proportion that the gain on attacks). If I understood correctly, Funky disagrees, because higher AB's will generate a greater proportion of criticals. Off the bat, I can't contradict it without being... err, I don't know the word for that. Precipitate, maybe? Work went faster than I thought it would, and I started working on making the spreadsheet wield results on that. Time will tell. Take care Kaezar
|
|
|
Post by kaezar on Apr 16, 2010 14:18:30 GMT
letting AB range from 90 to 100 in steps of 2 and AC from 85 to 95 in steps of 2 and using: 7 : 1/8/15 6 : 1/7/13/19 5 : 1/6/11/16 4 : 1/5/9/13/17 3 : 1/4/7/10/13/16/19 for attack progressions and including 1 haste attack i got: iteration 3 : 149.0 iteration 4 : 117.0 iteration 5 : 100.0 iteration 6 : 94.0 iteration 7 : 82.0 Khan, there is a cap at 6 attacks with your main hand, so you only get 6 attacks at iteration 3. Not sure if you meant this, but you count from the top down, so a character with BAB 20 at lvl 20 using an iteration 3 weapon would get attacks at 20 17 14 11 8 5. Note that if the BAB by 20 is lower, some of the iterations can be lost. Any attack who bonus drop to 0 doesn't happen. So if the same character with the same weapon had BAB 15 by level 20, he would have only attacks at 15 12 9 6 3, losing the attack that would have gone at 0. Note that the spells Tenser's Transformation and Divine MIght restore those extra attack. As they don't give back the BAB lost they can generate some extremely low bonus attacks. Staff and Bane Knight suffer specially from this. Staff 12 BAB by 20, BK has 14. So a staffy would end up with attacks (with iter 3 weapon) at 12 9 6 3. When he activates Tenser that would be in effect 12 9 6 3 0 -3. That happens because the attack bonus they get from tenser is included on the +20 cap. Take care Kaezar
|
|
|
Post by gandoron on Apr 16, 2010 15:23:20 GMT
Kaezar, Tenser attacks are in the "bonus attack" bucket. I've tested this, so they would follow the Haste attack with a -5 progression so if your staffie had 15/12/9/6/3/15 (with the last attack being haste) then tensers would give you 15/12/9/6/3/15/10/5
-G
|
|
|
Post by kaezar on Apr 16, 2010 17:12:51 GMT
Kaezar, Tenser attacks are in the "bonus attack" bucket. I've tested this, so they would follow the Haste attack with a -5 progression so if your staffie had 15/12/9/6/3/15 (with the last attack being haste) then tensers would give you 15/12/9/6/3/15/10/5 -G That would mean that the extra scripted attacks would not be following the modified iterations. Is that a fact? Could you point me at the reference, please? I went in the testing chamber with my rogue and was able to confirm what gand said, the "extra" attacks tenser gives come behind the haste attacks. But there are 3 of them, not two. With iteration 5, that is, with 3 it would be more. A staffy, example use, has by level 20 a bab of 12 (15 wiz -> 7 BAB, 5 WM -> 5, superior wf doesn't count for this). With 12 BAB a staffy has 3 attacks 12, 7, 2. With haste, it goes 12 7 2 12. The tenser, any tenser, gives back the last attack, going back to 4 base, 1 haste, but the recovered attack follows haste, going then 12 7 2 12 7. Dual wielding affects the base and the offhand atacks, but not the haste stack, so if the staffy went dual wielding a light weapon, he would go 10 5 0 12 7 10 5. If tenser was level 55, that would also add attacks, folloing the haste stack. So 10 5 0 12 7 2 -3 10 5 I am then still needing to know if the haste attacks stack will follow the iteration or will be iteration 5, always. Also, what happens with the offhand attacks? The follow the iteration of the weapon or the haste iteration? I'd say the weapon, but don't want to presume.One other question that came from doing the spreadsheet. Funky, do you agree that the important factor is the number of criticals caused at a given AB/AC/Curse, for each category (normal, IC/Keen, IC+Keen, etc) or do you also want to see the effects caused by severity (multiplier) of the criticals? I am working on number of criticals at the moment, I want to know if I have to proceed to damage comparisons. By the way, the table you showed does not mention power and improved power criticals, which has the potential to change things if you are right. Take care Kaezar [EDIT] Adde the underlined text. Take care K. [/EDIT]
|
|
|
Post by atomicsomething on Apr 16, 2010 19:22:50 GMT
Also, what happens with the offhand attacks? The follow the iteration of the weapon or the haste iteration? I'd say the weapon, but don't want to presume. With Monk AB progression, the offhand attacks dont follow the monk progression but the standard one (same for scripted attacks). I've always assumed it would be the same here.
|
|
|
Post by FunkySwerve on Apr 16, 2010 19:51:42 GMT
Funky, do you agree that the important factor is the number of criticals caused at a given AB/AC/Curse, for each category (normal, IC/Keen, IC+Keen, etc) or do you also want to see the effects caused by severity (multiplier) of the criticals? As I thought I already made clear, I have no interest in rehashing work done years ago - I'm quite certain you're wasting your time. Whatever you do, you're doing to satisfy your own curiosity, so the answer to this question is really dependent on whether you think our critical calcs are wrong too - my understanding was that you had issue with the iterations, not the crits, but only you can say. As I said, I don't recall precisely how we arrived at those figures, but given the rigor we exercised in arriving at them, I'm quite confident in them. Funky
|
|
|
Post by kaezar on Apr 16, 2010 23:42:58 GMT
Funky, my problem is only how the valuable the iterations are. Not on how much the criticals will do when/if they happen. I am just asking if *you* think that value is germane to the question of iteration.
And just to clarify, I'm trying hard as hell *not* to bias my answer. Whatever I may think and feel, I'm trying to organize the spreadsheet in a way *it* will give the answers.
However, before I proceed I'd like the answer to 2 questions.
First is simple: are you willing to discuss your results? If you are not, I'll just stop discussing this and wasting your and my time trying to explain the process. I will still work on the table, but for my own satisfaction.
Second is just a question about the mechanics of iteration. As people have pointed out on the above posts the extra attacks and the haste attack work outside of the mechanics of iteratioin. A Monk that is dualing weapons gets Monk progression on its main hands and normal iteration 5 progression on its off hands.
If said monk would use tenser to gain one or more extra attack, the extra attacks, they would also follow iteration 5. This is how it stands now. Tested on the test chamber with both rogue and monk, both doing normal, hasted and tensered attacks, dualing and not.
Question is, is this change going to modify this? If yes, how?
Assuming the answer to the first question is yes, the thought that went behind my sheet is bellow. I will need the answer to question 2 to be able to finish the table, however, as the attack values of all the extra attacks depend on them.
Take care Kaezar
I now arrived to a point where I can say I'll arrive at an answer, whatever it is. However, those will imply in to making a table with 300 cells *for each AB value of each iteration and each possible extra attack* Each table will find out the chance of critical for a given ab, depending on the options chosen at the entry.
Those tables will contain 1 to 20 in lines, and all the possible options of critical-influencing feats/spells abilities for columns, including some not considered on the original table.
That is a column for normal, a column for IC/Keen, Another for OC, DC, WM5, WM5 +OC, WM5+DC, WM7 and WM7+DC. Some columns will not influence critical chances differently. WM5 + OC gives the same result as WM5 + DC, for example. But its existence creates the possibility of pluggng in damage later if desired.
Each table will put value 1 in a cell if a confirmation roll with a given number rolled is a sucess and 0 if it is not (that is, 1 if it is a critical and 0 if not).
By adding the cells, you will know how many of the possible rolls will be criticals, creating a probability of critical. We'll then have a probablity of critical for a given ab. On each table.
By choosing the options at the entry, you'll be selecting the column used, and if power or improved power critical is influencing the result.
I can eventually even plug in how much a set amount of damage agreed upon is influenced, damage-wise, by the results. But that is for later, maybe.
For now, those critical probabilities will be input on the both tables of results. I am not decided on format yet, but I'm tending toward X/Y/Z for the table of absolute values, where, X is the number of attacks made, Y is the number of attacks that hit and Z is the number of criticals that will be scored with that number of attacks. so a number of 1200/650/50 would be read as 1200 attacks were made, 650 of them hit and 50 of the attacks that hit were criticals.
Just for your information, the spreadsheet now considers the number of attacks that actually happen according to the BAB the character has at level 20. So a character with a BAB of 14 by level 20 will get only 5 attacks at iteration 3, for example.
Take care Kaezar
Take care Kaezar
|
|
|
Post by FunkySwerve on Apr 17, 2010 1:42:11 GMT
First is simple: are you willing to discuss your results? If you are not, I'll just stop discussing this and wasting your and my time trying to explain the process. I will still work on the table, but for my own satisfaction. No, as I don't remember the precise way we arrived at them. I'm more than happy to discuss yours, but I simply don't remember all the things we considered years ago. I may be able to point out differences in your approach as opposed to ours, but I may not. I can tell you that we relied on math, not spreadsheets, and that acaos did the heavy calcs once we determined what should be relevant, and that we did a number of sets using different data for purposes of comparison. Acaos may remember more, but he may not. I don't know, as acaos will be doing the engine hack, most likely, and hasn't done it yet. I see no reason why it WOULD change, but then I haven't attempted to look at the code, so I have no idea what changes might be necessary - all I know is that it's a messy hack. Funky
|
|
|
Post by kaezar on Apr 17, 2010 2:11:03 GMT
Good enough, then.
I'll publish the results when I'm done. It'll probably be only sunday at the earliest, as I am traveling at 7 am and only come back at night.
I'll assume nothing will change. The way the sheet is organized now, I'll only have to update about 25 cells if it changes
Take care Kaezar
|
|