Carbineer Archive

Thread: Speed Formula

TAfirehawk
Thu Apr 21, 2005 6:46 am
#14






Zarqon wrote:


Sorry, didn't find anything about the test was for all weapon speeds. (original post says: tested for 'a' weapon, not all weapons)


But if it is tested for many different weapon speeds, then my formula is clearly wrong of course.


The reason I wanted it checked is that it sounds really strange to me to have a polynomial like that to govern the speed. From a theoretical point of view I mean. Don't get me wrong, I know you formula is that accurate in your test range, I didn't doubt that, I just wanted a bit... nicer formula in lack of better words.

Message Edited by Zarqon on 04-21-2005 03:31 PM






Oh, well I posted this in 3 different threads....I should just link it


And from a math perspective a polynomial IS the nicest way to do it.....Gabe suggests a lookup table that is easier to program for SOE, but it is based entirely off a polynomial.




Iebas Feania
Former Correspondent of the Former Profession, CARBINEER

TAfirehawk
Thu Apr 21, 2005 7:36 am
#15






Zarqon wrote:

ahh yes, from a computational perspective the polynomial/look up table is the way to go.


I was hoping to find some analytically nicer expression though (a generating function for the polynomial). The exponential was the first thing I thought of, but that doesn't seem to be the case then. Ohh well, can't have everything I suppose.







I wanted it to be simple too, but we haven't even gotten any data on the Warmup/Cooldown Delays on Specials yet either.


And I don't want everything, just a couple small somethings out of the CU would be ok.....right now is quite depressing from what is happening on TC-EP3.




Iebas Feania
Former Correspondent of the Former Profession, CARBINEER

Arxy
Thu Apr 21, 2005 9:56 pm
#16

Great work, but im curious how the formula ended up a qubic function when the data ended at 140 (Not a criticism just curious) Only reason I ask is that if it is truely a qubic function then you can get a little extra speed even using CA's ect. So those that have some nice modded clothes/armor you can still benefit from wearing them since the actual "speed cap" is not a hard number now.

Great Work Again.


Arxy
TAfirehawk
Fri Apr 22, 2005 6:15 am
#17






Arxy wrote:
Great work, but im curious how the formula ended up a qubic function when the data ended at 140 (Not a criticism just curious) Only reason I ask is that if it is truely a qubic function then you can get a little extra speed even using CA's ect. So those that have some nice modded clothes/armor you can still benefit from wearing them since the actual "speed cap" is not a hard number now.

Great Work Again.


Arxy





Not sure what the question is, because there is supposed to be incrementally smaller gains at the Player Skill Mod increases.....



Iebas Feania
Former Correspondent of the Former Profession, CARBINEER

TheBobulus
Fri Apr 22, 2005 9:00 am
#18

I think he means how, based on that formula, after around 150, you'd start getting an incremental *larger* speed gains with each new point.



-----
Master Weaponsmith, Phoenix City Mall, Theed, Valcyn
I'm looking for WS tapes

TAfirehawk
Fri Apr 22, 2005 1:33 pm
#19

UPDATE


Added "estimated" data to smooth out curve to +200 Speed, but it may be possible to reach over +200.....


I also added the Excel file to my GoA website for those that want to look at the data



Iebas Feania
Former Correspondent of the Former Profession, CARBINEER

Arxy
Fri Apr 22, 2005 5:37 pm
#20

Great work TA, very interesting data, and you said this is a general equation for all weapons, therefore, i think you have done more for the ranged prof's in terms of hard info than any other corr. I didnt mean to ask an obscure question initially, but in a word i think you have proven (via the data on hand) that speed CAP persee will be a thing of the past.

In theory--the greater the number the faster you will be, but at some point you get increasing gains at a decreasing rate (economics 101, i think i slept through that class in the late 80's)

Message Edited by Arxy on 04-22-2005 05:39 PM

TAfirehawk
Fri Apr 22, 2005 6:37 pm
#21






Arxy wrote:
Great work TA, very interesting data, and you said this is a general equation for all weapons, therefore, i think you have done more for the ranged prof's in terms of hard info than any other corr. I didnt mean to ask an obscure question initially, but in a word i think you have proven (via the data on hand) that speed CAP persee will be a thing of the past.

In theory--the greater the number the faster you will be, but at some point you get increasing gains at a decreasing rate (economics 101, i think i slept through that class in the late 80's)

Message Edited by Arxy on 04-22-2005 05:39 PM






Actually it was just a few hours to get close and another couple to refine it to where we are now. Of course the equation looks ugly and totally off the wall to most, but those math folks out there will recognize it....in fact a few of us said a polynomial would be required for diminishing returns of the NEW Speed Formula But Gabe has shown a simple lookup table is the most logical way to program it, so don't assume SOE is using an equation near to this, but it is safe to say this equation approximates whatever SOE did VERY CLOSELY.


The real work on TC is being done by memnoch as he spends MANY hours in MANY different scenarios trying to get the Mods on Specials....but the Devs have added so many variables and tweaks each patch that it is proving very difficult.....if you see memnoch give him a pat on the back




Iebas Feania
Former Correspondent of the Former Profession, CARBINEER

TAfirehawk
Sat Apr 23, 2005 12:50 am
#22






TheBobulus wrote:
I think he means how, based on that formula, after around 150, you'd start getting an incremental *larger* speed gains with each new point.






No, it starts "gaining" over +175 to about +190 is where it starts, not +150......and that is probably due to my data stopping at +125 as I have no access to "special" items on TC-EP3 and Chilastra was on TC3 for just a couple hours.


The real question to ask is have the Devs looked at the Speed Mod when it gets over +150......which is EASY to achieve with FS Speed and SEA's. And I don't have a clue on THAT question....




Iebas Feania
Former Correspondent of the Former Profession, CARBINEER

CaixCatab
Sun Apr 24, 2005 5:26 am
#23

... I miss my five star button so much it hurts right now


"Great work" doesn't cover how impressed I am!



-This message was deviously brought to you by:
Caix Catab S Darok Catab _ Hoti Catab
[Avenger] [Dead-Eye] [Businessman]
[INCAP]
"All warfare is based on deception. Hence, when able to attack, we must seem unable; when using our forces, we must seem inactive; when we are near, we must make the enemy believe we are far away; when far away, we must make him believe we are near."
Catab Trading and Clothing
2565 -4844 outside Mos Eisley Tatooine



FrostbiteS
Sun Apr 24, 2005 3:44 pm
#24


Nice, I tried the same with swords on Test. Or rather the 2,5 second swingtime axe.

But my excel didnt go so advanced on the formula.

http://forums.station.sony.com/swg/board/message?board.id=swordsman&message.id=62349


For melee speed ranging from 0 to 155 on test center


Melee speed 50, gives a 25% lowering of actual swingtimer.

Melee speed 120 gives a 40% lowering of swingtimer

Melee speed 140 gives a 42% lowering of swingtimer


With FS and SAE, a swordsman will be able to reach a melee speed of 192 if it goes live. yikety!

(MBrawler MSword TKM Pike 4000 Fencer 0400 155) +25tape +12fs melee speed


So odd bumps would be interesting on the curve.



For formula, I would be interested in one Y =1 / (x+a) +b as that would have a limes value of b.

But I dont know how to do that in Excel.

Message Edited by FrostbiteS on 04-25-2005 01:04 AM

Page 2 of 2