Aces High Bulletin Board
Help and Support Forums => Aces High Bug Reports => Topic started by: TDeacon on April 30, 2006, 08:18:10 PM
-
I did some offline testing today, motivated by recent experiences in the
MA. Based on my testing, there seems to be an inconsistency in the P51's
flap behavior, which appears to be a bug.
Details: P-51b: 50% fuel (123 gal); fuel burn multiplier minimum; 3K
alt, full ammo; no WEP; auto trim on; stall limiter off. BF-109-F4 same,
except 50% fuel was 53 gal. I turned on the "toggle screen info" which
puts text at the top of the screen, and used this as a guide in keeping
the plane's nose in a consistent position for each test. I tried to
stablize the turn rates before starting the measurements. I turned at
the maximum rate possible without getting the shaking effect. The value
t is for 5 complete circles, to allow averaging the results. Test
results in table below, with v being indicated velocity (I know this
distorts the data, but only in an absolute sense; the relative values are
correct and are what matter here), t being time in seconds to do 5
complete turns, r being calculated turn radius in yards, and rate being
calculated turn rate in degrees-per-second.
Formulas were r = (vt)(.01556) and rate = (v/r)(28.01127).
FLAPS - v(mph) - t (s) - r(yds) - rate(dps) - comment
*******************************************
P-51b
0/5 159 150 371 12.0
1/5 159 150 371 12.0
2/5 150 151 352 11.9
3/5 139 156 337 11.6
4/5 127 201 397 09.0 discontinuity in radius, r !!!
5/5 119 205 379 08.8
********************************************
Bf-109-F4
0/4 139 143 309 12.6
1/4 140 142 309 12.6
2/4 133 139 288 12.9
3/4 124 139 268 13.0
4/4 108 139 234 12.9
********************************************
Anyway, this kind of cripples the 51s in stallfights. After 2/5 or 3/5 flaps,
there appears to be no further tactical advantage. In contrast, The 109,
and I assume most other planes in the planeset which don't have this
bug, can achieve advantage by going to full flaps in a stallfight, if
needed. Of course this is a last resort tactic, but it is still quite important.
I wish I had the energy to do this for other US planes. I also wish I had
the energy to determine if the high-speed flap deployment the US planes
have means anything anymore. Low speed 1/4 flap seems fairly useless
from the above, for both planes.
Anyone else want to try to duplicate this?
********************************************
Disclaimer: Please do not dismiss this as a "whine". Admittedly, I
have been flying the P-51b a lot, due to the artwork. However, I
want all planes to be modeled reasonably. For example, I am glad
that the wierd instability problem which plagued the LW planes has
been fixed in this version. For another example, I am distressed
that the P-38s seem to have been adversely affected by this version,
even though I never use them. So there.
-
Sorry about the tables; the BB removed the white space.
-
Corsair looks like 109, as there is no discontinuity in radius, r, which decreases as one goes down the table. So far, only P-51 data shows
the jump in radius, so it still looks like a 51 bug.
As usual, t is for 5 x 360 degree circles, not just one.
FLAPS : v(mph) : t(s) : r(yds) : rate(dps)
*******************************************
F4U-1C
0/5 150 207 483 08.7
1/5 147 202 462 08.9
2/5 131 200 408 09.0
3/5 120 201 375 09.0
4/5 108 200 336 09.0
5/5 099 204 314 08.8
-
Originally posted by TDeacon
I turned on the "toggle screen info" which
puts text at the top of the screen
Pretty off topic, but how does one do this?
-
Not trying to dis ya here, but what makes you think its changes since patch 2? Do you have figures for pre patch 2 turn results?
-
No figures from before the patch. However, the sudden large turn radius jump at 4/5 (for the 51 only) is bad, and should be fixed regardless of when it was introduced.
Since I also have the previous version installed, I theoretically could obtain them. I imagine that HTC has some sort of automated procedure for making these kinds of measurements, however. I wish there was some way to know if they are able to check my measurements, and if they plan to fix it in the next patch...
-
Originally posted by TDeacon
Sorry about the tables; the BB removed the white space.
use the (code) (/code) tag
use "[" and "]" instead of "(" and ")" .
-
TDeacon,
I've been testing all the planes. I'll be posting my turn list in the Planes forum, but for now I can confirm most what you have here.
One of the interesting things you noticed was that both the 109 and P-51 gain no improvement with one notch of flaps. That seems to be true for the P-38 as well. That is true for all the 109s, all the 38's, and both 51's. I'm not sure how many other planes have this same effect, I've been limiting most of my testing to just No Flaps and Full Flaps.
I'm at work so don't have my data in front of me, but will be posting it soon.
Another side effect of the new drag model: It really helped the F4Us turning under full flaps (as mentioned in several threads), but slowed the 300 to 350 acceleration time way down. The F4U-1C takes 13 seconds longer from 300 to 350 than it did before 2.07 and lost 1 mph in top speed on wep, and about 10 mph w/o wep.
Thanks for publishing your formulas. I am using differnet ones, but I like yours better. My testing is using three orbits, not five, how will that change the factor used for Radius?
-
3/5 flap i don't know whether the planes be stalled or not, but 4/5-5/5flap, it turns suddenly to lose all control with deep stall and banking to a side quickly. it never happened in the previous version at least december last year when i quit this game temporarily at the time. p-51 without loading aux fuel was very stable stall fighter. but in low speed even with 30%fuel, it is very unstable plane. please don't ask whether it is bug or it is close to exact historical performance. Anyway it somehow changed.
-
Sorry for the late response; I was out of town. The full formulas were:
radius, r = ( (v) (t) (5280) ) / ( (2) (pi) (3600) (3) (5) )
rate = ( (v) (5280) (360) ) / ( (r) (3600) (3) (2) (pi) )
Therefore, in the first formula for "r", if you measure "t" over "n" turns, just substitute the value of "n" for the "5".
-
Originally posted by MOSQ
Thanks for publishing your formulas. I am using differnet ones, but I like yours better. My testing is using three orbits, not five, how will that change the factor used for Radius?
if I understand your question, the more valid "orbits" you do for your average, the more realistic/accurate your average will be........
example 1 or 2 orbits = not very accurate ( think emprical data here :D )
3 orbits = more accurate than doing 1 or 2 orbits
5 orbits = more accurate than doing 3 orbits
I prefer 6 to 10, 10 being the maximum for averaging, since all of this is time consuming anyways.......5 is a good # to though.......
to check to see if you are obtaining the best turnrate verses speed and or time , try speeding up 2 or 3 mph and test your turnrate by averaging , then compare results, then slow your speed down by 1 or 2 mph and repeat, if you gain in turn rate in either case, ie radius and time, then you pretty much spot on for the given data you are claiming to be accurate....( someone can prob give a lot better explanation than I have, fer instance Badboy? )