Go Back   Star Trek Online > Information and Discussion > The Academy
Login

Reply
 
Thread Tools Display Modes
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 31
02-16-2011, 11:45 AM
The NX also seems to have the 0.17 impulse modifier, I haven't gotten the calculations done on it, but that modifier gave me close to the shown values on cursory testing.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 32
02-16-2011, 11:53 AM
Quote:
Originally Posted by Taran_Tatsuuchi View Post
The NX also seems to have the 0.17 impulse modifier, I haven't gotten the calculations done on it, but that modifier gave me close to the shown values on cursory testing.
In theory it's supposed to have the same 0.20 that the escorts and the miranda have, as it's a 'light escort". The 'light cruisers' are also the only non-escorts that can mount dual cannons, presumably so you can get used to the playstyle if you're going that way.

It would seem then, that the remaining problems are:

How, exactly, does the tooltip speed vary with ship type? Is it related to the impulse modifiers or tied to (a) particular shiptype(s)?

What are the actual base speed values of all the engines, since they round off the decimals once they go above 10 flight speed?

Edit: It also wouldn't hurt to verify that the skill part of the formula still works right, since there's now 152 skillpoints available in a shiptype instead of 100. Prior testing I've done already revealed the effect of more points in that had a linear effect that didn't seem to be capped.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 33
02-16-2011, 12:08 PM
Unfortunately I'm not quite high enough to test the 'excess' skill available.

But the NX is defiantly 0.17 impulse mod, I've updated my post and its variation only goes up to +-0.05

edit: I've gone through all my test data so far, the escort is the only type that I've got to test that had variations in the tenths, everything else has variations in the hundredths.

The escort is also the only one where the combat engine's tooltip has a lower speed than the other types.

It would seem that the, and I'm taking a guess here, the impulse modifiers are in fact changing the flight speed of the engines, as the Nova and Miranda have the values listed on stowiki, the NX with the higher 0.17 impulse value also has a slightly higher flight speed on the engines.

Why are the numbers for the escort flight speed not following the pattern of having the combat engines with the highest flight speed stat...



Edit: I think I need to take a nap.. -_-;
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 34
02-16-2011, 12:58 PM
I'm only just able to follow the maths discussed in this thread. I'm anxiously hoping/waiting to see if you guys can come up with a new definitive engine speed formula. Thanks for all the effort!

/subscribed.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 35
02-16-2011, 01:46 PM
Quote:
Originally Posted by PatricianVetinari
I'm only just able to follow the maths discussed in this thread. I'm anxiously hoping/waiting to see if you guys can come up with a new definitive engine speed formula. Thanks for all the effort!

/subscribed.
Heh well the one here: http://theenginescannaetakeit.wordpr...chanics/#speed

Seems to be dead on right, once you actually get everything right as far I can tell.

Base + (FSpd + (EPow - 50) x IMod x EMod) x (1 + Stat / TMod)

Base is the base speed of the ship, 5 in all known cases.
Fspd is the flight speed stat of the engine.
Epow is your engine power setting,
Imod is your ship's impulse modifier,
emod is the engine's impulse modifier (0.6 for combat, 1.0 for normal, 1.35 for hyper),
stat is your total skill bonus in the ship type you're flying (EG, Spaceship command, + escort captain, + .5 of Heavy Escort Captain, + .5 of Tactical Escort captain, + Advanced escort captain for an advanced escort),
TMod is a Tier Modifier... confirmed by our testing to still be 192 for tier 1 and 2 ships, previously 400 for tier 3, 4, and 5 ships, edit: and it appears that it probably still is. Note that my previous figure of 300 was a misrememberance.

The burble we're finding, and the thing that tripped me up so bad, is fspd. In theory it's just the flight speed off the engine's tooltip, but there's two burbles that are complicating it up.

The first is that it rounds off the number it displays, which throws off the formula since it actually uses the non-rounded version to determine the speed. Particularly bad is that it rounds it off to whole numbers if the flight speed is 10 or more. As my testing accidentally demonstrated, the power system also does this, and the sub 1 point bonus I was getting from my efficient borg bridge officer was enough to create a noticable, if small, skew in the data. The flight speed actually being substantially rounded off (in the case of mark X whites, it's apparently .25 off on all three tpyes) REALLY throws off the numbers.

The second, and far more dastardly, is that the flight speed of a given engine ISN'T a constant. Different ships are getting different values on the same engines, and it's not clear what exactly is modifying them. It isn't simply the impulse modifier, because it's actually changing the tooltip speeds so much that the hyper and combat engines actually switch places as the fastest and slowest.

That last point is the big one: the first one you can get around by using the data to back-calculate the actual, unrounded flight speed. The second...we have no idea about any of it. Is it some kind of factor that's somehow linked to the impulse modifier, and thus the same for all ships with the same modifier?? Is it something that's only tied to particular ship types, a sort of unadvertised penalty/bonus that no one really noticed much before? EG: iIs it that escorts get a bonus, or do cruisers get a penalty? If it's particular ship types, does the carrier for example have an even lower tooltip value than a cruiser would? is a bird of prey's higher than an escort's?

We don't know any of those things. What it's based on, what factors it's using to determine it... the formula is valid if you pull the tooltip values for the ship type that you're testing(my screwup was using my cruiser captain's tooltips to try to calculate my bird of prey's speed. I didn't realize they were different!)

As of now though, we have very little idea of the extent or variance in the base engine speeds with different ship types... which is a problem, since that's what determines where the crossover point from combats to hypers is.

Edit: aaaand I just remembered that my tier 2 BoP pilot was getting the same tooltip speeds on the mark 4s as my tier 5 escort pilot was. whoo for sleepy.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 36
02-16-2011, 03:30 PM
This thread has helped me immensely on deciding impulse engine types, kudos to all.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 37
02-16-2011, 05:41 PM
The rounding is really irritating. Especially because the game doesn't just round but at least with weapons and power levels consistently seems to round down, regardless of what the fractional amount is. It's like the display just drops everything but the whole number, so if your weapon power is 124.99, it will display as 124, but if you look at your weapon tool tips it will be identical to what you'd see if you were at 125 (of course that is rounded too, which is why it's exactly the same when there is still probably tiny difference).

Anyway, great work on figuring this stuff out.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 38
02-17-2011, 01:16 AM
Quote:
Originally Posted by Nagorak View Post
The rounding is really irritating. Especially because the game doesn't just round but at least with weapons and power levels consistently seems to round down, regardless of what the fractional amount is. It's like the display just drops everything but the whole number, so if your weapon power is 124.99, it will display as 124, but if you look at your weapon tool tips it will be identical to what you'd see if you were at 125 (of course that is rounded too, which is why it's exactly the same when there is still probably tiny difference).

Anyway, great work on figuring this stuff out.
Tell me about it. The impulse engine speeds in the tooltips seem to round as you'd expect, the problem is, because of the way the formula works the small error from the rounding creates equally large errors in the results. From what I can see though, it creates a strictly *linear* error: it's the same amount of difference at all points in the curve...at least when you don't have any skill bonuses going in.

That means you can actually figure out what the actual fspd value is just by adding whatever your linear variation is to whatever the tooltip says.

Catch with that...is that the effect of the skill part of the formula is non-linear, so the fspd error gets lost in the data unless you factor the skill part of it out somehow.

Don't you just love algebra? :|
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 39
02-17-2011, 03:56 AM
I see that rounding of the fspeed caused the larger errors, but why on the escorts do the speeds for the engine types reverse?

Combat should have the largest fspeed with hyper the lowest...

Using the calculated values to determine the fspeed shows that the escorts are getting better fspeeds from hyper than combat hence skewing the efficiency line where it becomes more attractive to use hypers.

I just ran the math on my akira reversing the fspeeds swapping the hyper and the combat's calculated speeds...

Akira
tip_____10.2______10.7________11.2...............t ip_____11.2______10.7_____10.2
type____Com_______Imp_______Hyp..............type_ ___Com______Imp_______Hyp
25_____12.19_____10.68______9.44..............25__ ___13.20_____10.70______8.45
30_____12.79_____11.70_____10.79..............30__ ___13.80_____11.70______9.80
35_____13.39_____12.69_____12.14..............35__ ___14.40_____12.70_____11.15
40_____13.99_____13.68_____13.48..............40__ ___15.00_____13.70_____12.50
45_____14.58_____14.70_____14.83..............45__ ___15.60_____14.70_____13.85
50_____15.20_____15.69_____16.18..............50__ ___16.20_____15.70_____15.20
55_____15.79_____16.68_____17.45..............55__ ___16.80_____16.70_____16.55
60_____16.38_____17.69_____18.89..............60__ ___17.40_____17.70_____17.90
65_____16.99_____18.69_____20.24..............65__ ___18.00_____18.70_____19.25
70_____17.58_____19.70_____21.59..............70__ ___18.60_____19.70_____20.60
75_____18.20_____20.69_____22.94..............75__ ___19.20_____20.70_____21.95
80_____18.79_____21.68_____24.29..............80__ ___19.80_____21.70_____23.30
85_____19.38_____22.70_____25.64..............85__ ___20.40_____22.70_____24.65
90_____20.00_____23.69_____26.99..............90__ ___21.00_____23.70_____26.00
95_____20.59_____24.69_____28.34..............95__ ___21.60_____24.70_____27.35
100____21.20_____25.69_____29.70..............100_ ___22.20_____25.70_____28.70

_____________________________________56_____16.92_ ____16.90_____16.82
_____________________________________57_____17.04_ ____17.10_____17.09

With the fseeds swapped the akira's efficiency point shifts to the expected 55~60 rage, 57 to be exact...

I'm of the opinion the fspeeds for the escorts have gotten switched around somewhere along the line, the formula seems to still be accurate not accounting for skills.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 40
02-17-2011, 04:21 AM
Quote:
Originally Posted by Taran_Tatsuuchi View Post
I see that rounding of the fspeed caused the larger errors, but why on the escorts do the speeds for the engine types reverse?

Combat should have the largest fspeed with hyper the lowest...

Using the calculated values to determine the fspeed shows that the escorts are getting better fspeeds from hyper than combat hence skewing the efficiency line where it becomes more attractive to use hypers.
This is precisely what I've been on about this whole time, and I'm thinking it has to be a deliberate effect, probably something to do with how escorts are so speed dependent for defense.

Trick being they aren't just flipped, they're appreciably higher on an escort than on a cruiser, AS WELL AS being flipped. Some cursory calculations I did with mark 4s yesterday suggested that they COULD be being modified by the impulse modifier, except that the speeds are flipped. I don't know exactly how their item info database lookup works, but it seems like it'd take a specific piece of code to flip them like that, and probably isn't just a bug.

It likely has to do with escorts being really speed dependent for defense, but limited in their ability to set a lot of engine power due to needing it for weapons and to shore up their already fragile shields.

Additionally, I just did a full-up test with my adv escort, with all skills plugged in, against the formula with skill section plugged in... there's a significant deviation... but just eyeballing the graphs (easier said than done on this calculator's crappy screen) it looks like it's probably linear. Interesting thing is, the formula appears to be OVER estimating, not under.

I'm still crunching numbers on that though.
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off


All times are GMT -7. The time now is 03:08 PM.