acceleration test results
 

i wanted to post some numbers for my latest ship acceleration times.

im still scratching my head because if i get anywhere close to the real time for the slowest speeds then the upper speeds are way too long and to get the upper speeds right the lower speeds are not as slow as they should be.

basically the game engine wont follow the real life acceleration "curve" properly so its more of a line than a curve.

if you want to see real speed numbers look here for accleration time tests done on the HMS Hood and other ships are shown in the chart too.
http://www.hmshood.com/ship/hoodspecs4.htm


i ran these tests on the battleship kongo from a 1 kt idle speed accelerating up to maximum speed and these are the numbers i got for three different settings i tried:


acceleration test setup #1:

from 1 kts - 5 kts = 42 seconds
from 1 kts - 10 kts = 3 minutes 12 seconds
from 1 kts - 15 kts = 7 minutes 43 seconds
from 1 kts - 20 kts = 15 minutes 15 seconds


acceleration test setup #2:

from 1 kts - 5 kts = 55 seconds
from 1 kts - 10 kts = 4 minutes 15 seconds
from 1 kts - 15 kts = 10 minutes 23 seconds
from 1 kts - 20 kts = 20 minutes 20 seconds


acceleration test setup #3:

from 1 kts - 5 kts = 1 minute 22 seconds
from 1 kts - 10 kts = 6 minute 25 seconds
from 1 kts - 15 kts = 15 minutes 35 seconds
from 1 kts - 20 kts = 30 minutes 30 seconds


actual real times for the revenge class battleship: (closest match to the kongo)

from 1 kts - 5 kts = 3 minutes
from 1 kts - 10 kts = 7 minutes
from 1 kts - 15 kts = 10 minutes
from 1 kts - 20 kts = 20 minutes

I've also found that the DEVs have developed their own 'curves' in many areas of the game. Propshaft rev counting is a prime example. You can move the midpoint up and down the line, but it effects the opposite end
:)

yep, looks like real times arent doable, least not to my satisfaction, and even if it is in line with reality it may have too much effect on the game aspect if ships go too far and instead of being dragsters and they now seem to be in slow motion. thats why i think im going to concentrate on setup #1 and move on to testing in game conditions.

I think for the most part in actual gaming the player isn't going to encounter too many sitting targets, so acceleration from zero or one knot probably isn't as much of a concern as acceleration from 6-10 knots for merchants and 10-15 knots for warships.

Another thing the game doesn't take into account is the Pearl Harbor syndrome: ships that are at anchor have a very good chance of being in a steam-down condition. It doesn't matter how soon they spot the torpedo - that ship isn't going anywhere in the next hour or so.

The work you're putting into this problem is very much appreciated.:rock:

Tried your Acceleration thing, with what looks like good results. Just before the torp gets tothe target I'd notice the ship-hyperdrive effect, but not this time though :up:

http://www.vanjast.com/SHPics/Hit01.jpg

This is hard to do right (I did some testing a while back), but it needs to happen. Unfortunately, each environmental mod will alter this in non-trivial ways, too. Turning circles are another issue.

Personally, I'd gladly do without the environmental stuff if the bargain got me a game that played better. Webster, don't lose sight of the fact that no matter how far from reality you end up settling for, it's a thousand times better than you found it. Perfection just isn't going to happen here. Great improvement, however, is nearly unavoidable.:up:

you mean except for those pesky superhuman destroyers that that this topic was designed to address? lol those DD park anywhere and you always seam to be near them :hmmm:

yep, the rest of the fleet probably will never be starting off from a stop but often in evasive manuvers the wrong setting has ended up with the kongo stuck at 4 or 5 kts in zig zags so it never gains speed.

i had something that was within a 10%margin of error on real life acceleration times but in game it stalled out and couldnt get going, the rest of the task force left it behind. if i did that to all ships i would have had a task force under attack speeding away doing evasive manuvers at a whopping 4 kts lol.

i'll probably release it this weekend, maybe sooner, i have settled on all my numbers that slow ships as much as possible yet still keep the AI looking the same as far as how ships are manuvering and zig zags and stuff.

how it looks and feels in the game shouldnt change which is more important to me than the actual times i record.


this is what it will look like in the game:

again the kongo with my latest numbers: (the ones i will be putting in the mod)


from 1 kts - 5 kts = 7 seconds
from 1 kts - 10 kts = 40 seconds
from 1 kts - 15 kts = 1 minute 35 seconds
from 1 kts - 20 kts = 3 minutes 5 seconds

and the numbers for v3 i released:

from 1 kts - 5 kts = 5 seconds
from 1 kts - 10 kts = 15 seconds
from 1 kts - 15 kts = 40 seconds
from 1 kts - 20 kts = 1 minutes 15 seconds

and as with stock numbers:

from 1 kts - 5 kts = 1-2 seconds (so fast i almost missed it)
from 1 kts - 10 kts = 5 seconds
from 1 kts - 15 kts = 10 seconds
from 1 kts - 20 kts = 18 seconds

Oops.:oops: That's what I get for lacking SH4 experience. The only time I saw that in SH3 was in the torpedo school scenario, in which a sitting merchant would evade by accelerating. Destroyers are a problem, as they actually can accelerate fairly quickly. Some of them had a power-to-weight ratio of as much as 15-1. I wish I had a real hard-fact answer, but the truth is that nobody seems to have actually timed their acceleration when they were doing all those trials.

:hmmm: i can see it now....


NHRA Today Presents The US Navy Quarter Finals Time Trials


trouble with timing on water is painting that damn finish line at the end of the drag strip :rotfl: