| Pages: [1] 2 :: one page |
| Author |
Thread Statistics | Show CCP posts - 0 post(s) |
Endless Subversion
   |
Posted - 2007.03.28 00:27:00 -
[1]
I'm unsure how to find the rate at which ships accelerate. Ideally I'd like to be able to look at a ship's mass, max velocity and movement modules and be able to calculate it's acceleration rate. Anyone know the formula(s) that Eve uses?
|
InsanlyEvlPerson
Gallente
Night-Stalkers
|
Posted - 2007.03.28 01:59:00 -
[2]
y not just test it out liek the rest of us do...
|
Endless Subversion
|
Posted - 2007.03.28 02:39:00 -
[3]
What a great response....
Because trying it out means I have to actually buy the module and ship, when in reality im often looking at what something could do, potentially, if I had faction/officer mods.
In other words, I want to know how the game works so I can create ship setups that fulfill the role that I play in my gangs. To do this the best that I can, I need to know things like how fast my ship accelerates....
|
Megan Maynard
Minmatar
RONA Deepspace
|
Posted - 2007.03.28 03:44:00 -
[4]
The MWD have a Force value in Newtons. Take that and devide it by the mass of your ship, that will give you a acceleration. Now figure out the max speed of your ship with the MWD on and divide it by your acceleration. That is the amount of time required to reach full speed.
hope that helps. (PS this is a game so that value might be off, i wouldn't expect it to be very far off.)
|
Hoshi
Blackguard Brigade
|
Posted - 2007.03.28 12:14:00 -
[5]
Problem is acceleration in EVE is not linear. It looks more like the cap/shield recharge graph. You might be able to calculate avg acc.
----------------------------------------
A Guide to Scan Probing in Revelations |
Aramendel
Amarr
Queens of the Stone Age
Anarchy Empire
|
Posted - 2007.03.28 14:57:00 -
[6]
Edited by: Aramendel on 28/03/2007 14:57:15
There is a way to calculate a ships "agility" - problem is however that so far noone has any clue what this number actually means. Just that the lower the number the better the agility.
You can still use it to compare different ships/setups, though. For example, if a frig and BS get the same number both will need the same time to acellerate from 0% speed to 100% speed.
You need for this the agility mod of a ship (not visible ingame, but the battleclinic.com database has it), the ship mass (+ ab/MWD mass addition if you want agility when those are active), the mass reduction and inertia reduction from modules/rigs (-1% mass and -1% innertia have an identical effect on all aspects of ship agility) and the boni from skills and implants.
In short: agility mod * mass (before modules) * inertia reduction * mass reduction * skill reduction
For example, a dominix with 2 LH instabs (-20% inertia), 1 t2 nano (-12.5% mass), 2 t1 housing rigs (-15% mass) and 1 joint rig (-15% inertia):
0.155 (BS agility mod) * 105 (domi mass) * (1 - 0.2 (first instab)) * (1 - 0.2*0.87 (2nd instab + first stacking penality) * (1 - 0.15*0.57 (joint rig with second stacking penality)) * (1-0.15 (first housing rig)) * (1 - 0.15*0.87 (second housing rig with first stacking penality)) * (1 - 0.125*0.57 (t2 nano with second stacking penality)) --> 6.75 "somethings"
With max skills that would be 6.75 * 0.9 (SC) * 0.75 (EM) -> 4.56.
And to compare, a throax has without modules or skills 6.6 "somethings". So said dominix would be around 2% less agile (-> needs around 2% longer to accellerate from 0 to 100% (or from 0 to 50% or from 0.75%.. you get the picture))as an unfitted thorax.
|
Kazuo Ishiguro
|
Posted - 2007.03.28 16:00:00 -
[7]
Edited by: Kazuo Ishiguro on 28/03/2007 16:04:04
 Originally by: Hoshi
Problem is acceleration in EVE is not linear. It looks more like the cap/shield recharge graph. You might be able to calculate avg acc.
We need someone who gets a high, stable fps (100+) to fraps a slowed-down industrial accelerating from rest in a straight line. Then it would be a simple (through tedious) matter of going through the recording frame by frame (with virtualdub or something like that) and plotting a graph of velocity against time.
I don't get anything like enough fps to do this on my own, but if someone can provide me with a video of the required quality, I'm prepared to do the boring part (assuming that there isn't some easier way involving ocr software).
Related question: if you set your ship to orbit at v m/s around a target r metres away from you (and your ship has agility z), what transverse speed do you actually get? I've already begun collecting data on this, but it's hard to pin down a stable figure for transverse speed due to wobbles in the orbit and I'm not optimistic.
If we could work out what a ship's maximum transverse acceleration is, that would place limits on the above problem, since in that scenario the ship is attempting to accelerate towards the target at v^2/r m/s^2 . I suspect that this is again related to agility somehow, but it would probably be very tedious to investigate - each observation needs about a minute for the ship to settle into a stable orbit, and it's difficult to accurately gauge distances larger than 10km.
------
Top speed calculation spreadsheet - updated for the recent speed nerf :) Army of doom headcount: 26,045 |
Endless Subversion
|
Posted - 2007.03.28 16:17:00 -
[8]
Hmm. Thanks for the responses guys. I, too, get totally crap FPS, but if you ever do find someone to do that, I'd be really interested in the results.
|
Taipan Gedscho
Muzzletov Gewaltski Inc.
|
Posted - 2007.03.28 16:20:00 -
[9]
Originally by: Endless Subversion
I, too, get totally crap FPS,
lil off topic, but setting buffersize in prefs.ini to half your max memory in MB should help, as well as CTRL+ALT+SHIFT+T
(turn off turret effects) as CTRL+ALT+SHIFT+E (turn off the other effects, its called "all" effects, but still leaves turrets enabled. iceland english ftw)
Only you can save mmorpgs - Stop crying for nerfs today! |
Megan Maynard
Minmatar
RONA Deepspace
|
Posted - 2007.03.28 16:53:00 -
[10]
Yeah, my idea was wrong btw, working on a new one.
Your signature exceeds the maximum allowed filesize of 24000 bytes -Sahwoolo Etoophie ([email protected]) |
goodby4u
Logistic Technologies Incorporated
Free Trade Coalition
|
Posted - 2007.03.28 17:29:00 -
[11]
Originally by: Aramendel
Edited by: Aramendel on 28/03/2007 14:57:15
There is a way to calculate a ships "agility" - problem is however that so far noone has any clue what this number actually means. Just that the lower the number the better the agility.
You can still use it to compare different ships/setups, though. For example, if a frig and BS get the same number both will need the same time to acellerate from 0% speed to 100% speed.
You need for this the agility mod of a ship (not visible ingame, but the battleclinic.com database has it), the ship mass (+ ab/MWD mass addition if you want agility when those are active), the mass reduction and inertia reduction from modules/rigs (-1% mass and -1% innertia have an identical effect on all aspects of ship agility) and the boni from skills and implants.
In short: agility mod * mass (before modules) * inertia reduction * mass reduction * skill reduction
For example, a dominix with 2 LH instabs (-20% inertia), 1 t2 nano (-12.5% mass), 2 t1 housing rigs (-15% mass) and 1 joint rig (-15% inertia):
0.155 (BS agility mod) * 105 (domi mass) * (1 - 0.2 (first instab)) * (1 - 0.2*0.87 (2nd instab + first stacking penality) * (1 - 0.15*0.57 (joint rig with second stacking penality)) * (1-0.15 (first housing rig)) * (1 - 0.15*0.87 (second housing rig with first stacking penality)) * (1 - 0.125*0.57 (t2 nano with second stacking penality)) --> 6.75 "somethings"
With max skills that would be 6.75 * 0.9 (SC) * 0.75 (EM) -> 4.56.
And to compare, a throax has without modules or skills 6.6 "somethings". So said dominix would be around 2% less agile (-> needs around 2% longer to accellerate from 0 to 100% (or from 0 to 50% or from 0.75%.. you get the picture))as an unfitted thorax.
That pile of math is why i usually just test it out with t1 overdrives/istabs/nanos.
__________________________________________
Yes it is great being amarr.
I am minmatar,fly amarr,use gellente drones and am in caldari space. |
smallgreenblur
Minmatar
Wreckless Abandon
Triumvirate.
|
Posted - 2007.03.28 17:34:00 -
[12]
Originally by: Endless Subversion
What a great response....
Because trying it out means I have to actually buy the module and ship, when in reality im often looking at what something could do, potentially, if I had faction/officer mods.
In other words, I want to know how the game works so I can create ship setups that fulfill the role that I play in my gangs. To do this the best that I can, I need to know things like how fast my ship accelerates....
Test sever ftw.
sgb
|
Aramendel
Amarr
Queens of the Stone Age
Anarchy Empire
|
Posted - 2007.03.28 18:25:00 -
[13]
Originally by: goodby4u
That pile of math is why i usually just test it out with t1 overdrives/istabs/nanos.
It's actually rather easy when you have done it once or twice. It only looks complicated because of the stacking reduction mods.
|
Fluffernator
|
Posted - 2007.03.28 18:38:00 -
[14]
I am guessing the velocity function of a ship in eve resembles something like this:
V(t) = Vmax*(1-e^(k*t)) k is some constant that has to do with your ships mass, or agility.
Now the acceleration graph would just be the derivitive the velocity function.
|
Kazuo Ishiguro
|
Posted - 2007.03.29 10:00:00 -
[15]
Edited by: Kazuo Ishiguro on 29/03/2007 09:56:36
Bumping this in the hope that someone kind person able to provide the required fraps will notice the thread 
------
Top speed calculation spreadsheet - updated for the recent speed nerf :) Army of doom headcount: 26,045 |
Hoshi
Blackguard Brigade
|
Posted - 2007.03.29 12:15:00 -
[16]
Originally by: Kazuo Ishiguro
Edited by: Kazuo Ishiguro on 29/03/2007 09:56:36
Bumping this in the hope that someone kind person able to provide the required fraps will notice the thread
Cpu/Gpu speed is not so much the issue here, it's hard drive speed. You need a raid 0 setup to do it. Best I can do is 45 dps at 1024x786 full size and 80 dps at half size.
----------------------------------------
A Guide to Scan Probing in Revelations |
Kazuo Ishiguro
|
Posted - 2007.03.29 19:19:00 -
[17]
The resolution of the video doesn't matter as long as the speedometer is always legible. It'd still be possible to get results from 40 fps video, but they'd probably not be quite so accurate.
If you do record a video, please mention what skills/implants/rigs/modules etc you're using that affect mass or inertia or velocity.
------
Top speed calculation spreadsheet - updated for the recent speed nerf :) Army of doom headcount: 26,045 |
Kazuo Ishiguro
|
Posted - 2007.03.30 18:33:00 -
[18]
At considerable, painstaking length, I've extracted a set of acceleration data from a video that Hoshi captured (which I subsequently edited):
http://dl.eve-files.com/media/corp/KzIg/Acceleration_data.zip
The race is now on (no pun intended) to work out what sort of curve this is, and how it fits the parameters of the ship used:
Originally by: Hoshi
Ship: Typhoon
Fitting: 7x 1600mm Steel Plate I
Navigation 5, Spaceship Command 5, Evasive Maneuvering 5.
No implants
Top Speed: 187.5 m/s
Mass: 126.250.000 kg
Inertia Modifier: 0.104625
------
Top speed calculation spreadsheet - updated for the recent speed nerf :) [right][red]Army of doom headcount: 26,045[/r |
Endless Subversion
|
Posted - 2007.03.31 00:25:00 -
[19]
Nice job!
Makes my head hurt looking at that formula. Any chance you could throw that into a simple open office spreadsheet and link it in your sig or something.
|
Hoshi
Blackguard Brigade
|
Posted - 2007.03.31 02:11:00 -
[20]
We have already more or less come to the conclusion that the above formula is not 100% correct, for one thing it uses mass in the wrong way. BUT it has the right slope, we just need to fill on the semi constant values like mass and agility correctly.
----------------------------------------
A Guide to Scan Probing in Revelations |
Kazuo Ishiguro
|
Posted - 2007.03.31 09:39:00 -
[21]
I've updated the uploaded spreadsheet (use the link above to get it) so that people can play around with the model. I more or less agree with Hoshi - the pattern of residuals from the current model suggests that the formula is slightly off (there's a chart showing this in the new spreadsheet).
The next step is to repeat the entire process thus far with a different ship - I've suggeseted a slowed-down frigate accelerating to about 100m/s (that way, observations can be taken to the nearest 0.1m/s) and see if there's a formula that works well with higher agility multipliers and lower masses.
Let's just pray that I don't get RSI from flicking through thousands of frames in virtualdub 
------
Top speed calculation spreadsheet - updated for the recent speed nerf Army of doom headcount: 26,045 |
Hoshi
Blackguard Brigade
|
Posted - 2007.04.01 20:15:00 -
[22]
Here comes my suggestion for a formula.
V(t) = Vmax*(1-e^-(t / (A*M)))
V(t) = Velocity after time t (in seconds)
Vmax = Max Velocity
t = Time in seconds
A = Inertia Modifier (this is the name it has now on sisi when it's displayed in the ship info, also called agility sometimes)
M = Total mass / 1.000.000 (if a ship weights 120.000.000 kg M would be equal to 120).
Simple, easy and fits perfectly with the data.
----------------------------------------
A Guide to Scan Probing in Revelations |
Kazuo Ishiguro
|
Posted - 2007.04.01 21:43:00 -
[23]
Note that in the case above, Hoshi's formula is exactly the same as mine. However, his returns correct acceleration times for other ships (timed using a stopwatch), so I think that'll do for the time being.
------
Top speed calculation spreadsheet - updated for the recent speed nerf Army of doom headcount: 26,045 |
Endless Subversion
|
Posted - 2007.04.01 23:41:00 -
[24]
Don't stop guys! (over my head) Make this into something the eve community can use!!
|
Kazuo Ishiguro
|
Posted - 2007.04.01 23:44:00 -
[25]
Differentiating V(t) gives V'(t) = Vmax/AM * e^(-t/AM) - this is always maximal at t = 0, so for any given Vmax, the greatest possible acceleration a ship can achieve is Vmax/AM.
A ship moving at speed V in a circular orbit of radius r accelerates towards the focus at V^2/r, so the tightest orbit a ship can sustain when travelling at Vm/s might be given thus:
V^2/r = V/AM
=> r = VAM
This agrees with what one would expect - lower agility grants a ship the ability to turn faster. This is especially interesting, because if it turns out to be true, we'll have a meaningful definition of agility. Of course, I still need to test this.
------
Top speed calculation spreadsheet - updated for the recent speed nerf Army of doom headcount: 26,045 |
Hoshi
Blackguard Brigade
|
Posted - 2007.04.01 23:56:00 -
[26]
You warp at 80% speed.
0.8 = 1 * (1-e^-(t/A*M) ->
0.8 = 1 - e^-(t/A*M) ->
0.2 = e^-(t/A*M) ->
t/AM = -ln(0.2)
t = -ln(0.2)A*M
----------------------------------------
A Guide to Scan Probing in Revelations |
Matyae
Elegance
|
Posted - 2007.05.20 02:42:00 -
[27]
Edited by: Matyae on 20/05/2007 02:41:36
Another parameter can "cap" the acceleration, and make it less than it should be. It's the (current orbit speed) / (max ship speed) ratio.
I made some tests orbiting things and calculated the effective acceleration.
Effective acceleration : Aeff = 2 Veff sin(Vrot/2)
With Vrot = angular rotation speed in rad/s while orbiting, and Veff = current ship orbiting speed.
Vrot = 2Pi/T with T = period.
T = 2 Pi R/Veff with R = radius = orbiting distance.
So Aeff = 2 Veff sin(Veff/(2R))
Now put on a graph X = Veff/Vmax = speed ratio, and Y = Aeff/Amax = accel ratio.
Where Amax = Vmax / I and I = inertia = A M / 10^6
and Vmax = max straight line ship speed
Graph
Looks like a circle, eh ? 
And indeed it is. Aeff is limited by following relationship :
(Aeff/Amax)¦ + (Veff/Vmax)¦ = 1.
So, for a given orbiting speed Veff, corresponding orbit distance R can be calculated with this formula :
R = V / (2 Arcsin(Aeff / (2Veff)))
Where Aeff = SQRT((Vmax¦-Veff¦)/I¦)
and I = inertia = A M / 10^6
If you're interested, all of this is applied in a small spreadsheet I made. Enter your ship & skills characteristics, your (real) orbit distance and it'll tell you your orbit speed. Et vice-versa.
Maybe not really user-friendly, I originally made it for me and a few friends. But oh well...
Speedsheet
|
Endless Subversion
|
Posted - 2007.05.29 17:45:00 -
[28]
Edited by: Endless Subversion on 29/05/2007 18:36:05
Edited by: Endless Subversion on 29/05/2007 18:19:39
Edited by: Endless Subversion on 29/05/2007 18:06:35
Spreadsheet looks great, I just wish I could get it working.
The rotational stuff will help me a ton when I finish the training for my next ship, but for now I just need some straight line acceleration information.
Two Issues:
(1)The spreadsheet references the ships you added, but when I add more ships and their data to the ships tab the front page of the spreadsheet doesn't load the values. I don't know enough about the spreadsheet language to decipher it really, my spreadsheets are pretty basic ;p
(2)Even your 3 base examples don't work properly take, for example, the crow. Although the info loads in the 'ship specs' section it doesn't load properly for the final pre-mwd velocity. The answer is zero, despite the cells it's supposed to be multiplying not coming out to zero.
*after a bit more testing, everything in the formula =B37*C43*C47*B60*(1+B14/100)works fine except for the b37. If i set the cell to =b37 it displays the value in the b37, unless it's one of my ships, because then b37 doesn't display a value, despite my following the same format as your ships. However if set the cell to =b37*c43 it shows 0 instead of the expected value. This is despite the fact that b37*c43 ISNT zero... I really havn't the foggiest as to what is going on.*
I'd really like to be able to use that spreadsheet, as it looks pretty useful, I just need to figure out how it does what it does and why it isn't doing it.
|
Kazuo Ishiguro
|
Posted - 2007.05.29 19:45:00 -
[29]
Matyae, I wonder whether you could clarify a bit? Your explanation would benefit greatly from a few diagrams. Anyway, one question at a time:
Originally by: Matyae
Another parameter can "cap" the acceleration, and make it less than it should be. It's the (current orbit speed) / (max ship speed) ratio.
By max. ship speed, I suppose you mean whatever speed you've set?
Quote:
I made some tests orbiting things and calculated the effective acceleration.
Effective acceleration : Aeff = 2 Veff sin(Vrot/2)
So, if I follow correctly, Aeff = transverse acceleration, Veff = current orbit speed (not the speed you set, but the one your ship is actually orbiting at)
Quote:
With Vrot = angular rotation speed in rad/s while orbiting, and Veff = current ship orbiting speed.
Vrot = 2Pi/T with T = period.
T = 2 Pi R/Veff with R = radius = orbiting distance.
So Aeff = 2 Veff sin(Veff/(2R))
This bit is perfectly clear and needs no further explanation.
Quote:
Now put on a graph X = Veff/Vmax = speed ratio, and Y = Aeff/Amax = accel ratio.
With Amax = Vmax / I and I = inertia = A M / 10^6
and Vmax = max straight line ship speed
Graph obtained from some Crow tests
Looks like a circle, eh ?
And indeed it is. When orbiting, Aeff is limited by the following relationship :
(Aeff/Amax)¦ + (Veff/Vmax)¦ = 1.
So Amax is as I've found above for straight line motion, and Aeff is as you've explained above. How did you collect the values of Aeff that you used to plot your graph? Would you consider sharing your data?
Quote:
So, for a given orbiting speed Veff, corresponding orbit distance R can be calculated with this formula :
R = V / (2 Arcsin(Aeff / (2Veff)))
With Aeff = SQRT((Vmax¦-Veff¦)/I¦)
and I = inertia = A M / 10^6
It actually works.
If you're interested, all of this is applied in a small spreadsheet I made. Enter your ship & skills characteristics, your (real) orbit distance and it'll tell you your orbit speed. Et vice-versa. Quite handy. :)
Maybe not really user-friendly, it was originally made only for a few friends, and is not even finished. But oh well...
Speedsheet
The most useful thing to do with these formulae would be to find Reff and Veff from Vmax and R, or vice versa. I'll see what I can come up with. Good work!
------
Spreadsheets: Top speed calculation - Halo Implant tanking |
J'Mkarr Soban
Amarr
Viziam
|
Posted - 2007.05.29 19:46:00 -
[30]
This is a good thread! Thanks for all the work.
What I'd like to know is how I can increase my orbit speed whilst keeping the same radius. I was going to post a specific thread on it, but here seems a better place to put it.
|
| |
|
| Pages: [1] 2 :: one page |
| First page | Previous page | Next page | Last page |