Carbineer Archive

Extrapolate it out past +200 Speed and see what it graphs.

I did the Speed Formula 3 weeks ago.....a 3rd Order Polynomial does the fit perfectly. It might look messy, but all you do with the Speed Formula is put it in a spreadsheet or webpage anyways.

Message Edited by TAfirehawk on 05-03-2005 12:00 PM

---

Zhorx wrote:

Tried some data for the new weapon speed and came up with:

actual speed=weapon speed*(0.5231+0.4769*2^(-speed/46))

Got a random deviation of at most 5e-3 meaning all numbers i found so far follow this furmula after rounding errors.

If you would care to see if this works or if it is faulting please post data for when it doesn't work and i will try to improve it.

---

Ya know as I was going to sleep last nite I had that exact image pop in my head. Nice advice...

I just ran the numbers through my spreadsheet for this 'new' formula and it breaks down just like I thought when you get over +100 Speed.

It is possible to get around +200 Rifle Speed on LIVE so a formula must have diminishing returns up through that point, which my formula does....of course the Devs could have screwed it up and not thought about such high player mods, just haven't been able to test it past +135 yet.

Message Edited by TAfirehawk on 05-03-2005 12:09 PM

Hawkke got 200 speed with pikeman and he noticed absolutly little change between 130-150 and absolutly no change between speed from 150 to 200. I'm really doubting the significance of speed after getting 120-130 speed, and I dont trust the DPS numbers listed on the weapons. I think speed is overrated now.

---

Bennyboy4308 wrote:
Hawkke got 200 speed with pikeman and he noticed absolutly little change between 130-150 and absolutly no change between speed from 150 to 200. I'm really doubting the significance of speed after getting 120-130 speed, and I dont trust the DPS numbers listed on the weapons. I think speed is overrated now.

---

My guess is he is hitting the weapon specific hard cap, which nobody with a RED NAME has talked about, but I think is still there.

And my Speed Formula flattens out past +125 a lot, so it gives an accurate representation except a hard cap, if it exists.

If you see Hawkke, send him my way (when Chilastra gets back up) so I can get some data from him to verify the formula.

---

TAfirehawk wrote:

Extrapolate it out past +200 Speed and see what it graphs.

I did the Speed Formula 3 weeks ago.....a 3rd Order Polynomial does the fit perfectly. It might look messy, but all you do with the Speed Formula is put it in a spreadsheet or webpage anyways.

Message Edited by TAfirehawk on 05-03-200512:00 PM

---

---

Zhorx wrote:

---

TAfirehawk wrote:

Extrapolate it out past +200 Speed and see what it graphs.

I did the Speed Formula 3 weeks ago.....a 3rd Order Polynomial does the fit perfectly. It might look messy, but all you do with the Speed Formula is put it in a spreadsheet or webpage anyways.

Message Edited by TAfirehawk on 05-03-200512:00 PM

---

Could i have the data you used for your model as well? Like what speed and from where...skills, buffs, SEAs...and the weapon speed before and after.

Here is what it looks like if i extrapolate it a bit. Seems both our look the same up until about 140 then you get a very small payoff (SEAS?).

The third degree formula will freak at a little over 200 with increasing payoff.

---

Message Edited by TAfirehawk on 05-03-2005 12:42 PM

---

TAfirehawk wrote:

---

Zhorx wrote:

---

TAfirehawk wrote:

Extrapolate it out past +200 Speed and see what it graphs.

I did the Speed Formula 3 weeks ago.....a 3rd Order Polynomial does the fit perfectly. It might look messy, but all you do with the Speed Formula is put it in a spreadsheet or webpage anyways.

Message Edited by TAfirehawk on 05-03-200512:00 PM

---

Could i have the data you used for your model as well? Like what speed and from where...skills, buffs, SEAs...and the weapon speed before and after.

Here is what it looks like if i extrapolate it a bit. Seems both our look the same up until about 140 then you get a very small payoff (SEAS?).

The third degree formula will freak at a little over 200 with increasing payoff.

---

I linked to my spreadsheet already....

And if you want the 3rd Order Polynomial to go over +200, just plug in more data points and it will flatten right out. Trust me, I have been over this weeks ago with many people, you can't approximate the Speed Formula any better than this....or equation just keeps giving speed improvements past +125 and that is WRONG.

Your graph doesn't match the data.

Message Edited by TAfirehawk on 05-03-200512:42 PM

---

---

TAfirehawk wrote:

And if you want the 3rd Order Polynomial to go over +200, just plug in more data points and it will flatten right out. Trust me, I have been over this weeks ago with many people, you can't approximate the Speed Formula any better than this....or equation just keeps giving speed improvements past +125 and that is WRONG.

Your graph doesn't match the data.

Message Edited by TAfirehawk on 05-03-2005 12:42 PM

---

---

Zhorx wrote:

Umm, you got your model by estimating a lot of results there when it flattens out. My formula fitted all your REAL numbers exactly. Plugging in numbers is by no means a good way of obtaining results. Check your formula against these results too if you please. Your will fail here when rounding (i checked).

http://forums.station.sony.com/swg/board/message?board.id=swordsman&message.id=62659

---

ROFLMAO, of course I estimated WHAT IT SHOULD DO, but there is little proof what happens beyond the Examine Window on LIVE right now.

Fails on rounding, don't think so.......you should check your data against playing on LIVE, not TC old data from 3 weeks ago.

Graph REAL data points that you can get and see where they fall on your curve. You havelittle diminishing returns within about +175 speed and we know from LIVE that the real world effect drops off much sooner.

The equation falls off from a lookup table most likely, so these are just approximations anyway.....

Actually I did test Rifle Speed which goes to +135 and you numbers are off, just checked on TC-EP3. We can argue all day, but the fact is that going much above +100 Speed in any prof is a waste of time. Please look at Ben's post again.

---

Zhorx wrote:

---

TAfirehawk wrote:

And if you want the 3rd Order Polynomial to go over +200, just plug in more data points and it will flatten right out. Trust me, I have been over this weeks ago with many people, you can't approximate the Speed Formula any better than this....or equation just keeps giving speed improvements past +125 and that is WRONG.

Your graph doesn't match the data.

Message Edited by TAfirehawk on 05-03-2005 12:42 PM

---

And to prove your estimated curve fit doesn't fit the numbers that guy had i took the liberty to compare my line to yours with his test data. Over 120 your fails because of the false estimations you did.

Mine is blue, yours is green and the data points are crosses, clicky!

---

Like I said in my last post, get some new data as that Swordsman stuff is over 3 weeks old.

And I flattened out the curve because actual play testing showed zero speed increase past +100 to +125 on TC and Ben's post about Hawkke proves that too.

My formula is what it should be and using current data, please verify that 3 week old Swordsman data.....

In the end neither of us has any valid data to develop an equation.....and hence why I hadn't posted my ESTIMATED THEORETICAL FORMULA on the forums anymore. Too much was changing on TC at the time and the hard cap I feared for weapon types seems to be in effect on LIVE, so the results are hard to get right now.

Message Edited by TAfirehawk on 05-03-2005 01:06 PM