PDA

View Full Version : ROLLRATE TESTING: 4.02m Table of Contents



TAGERT.
10-24-2005, 09:28 AM
Enjoy

SimHQ (http://www.simhq.com/simhq3/sims/boards/bbs/ultimatebb.php?ubb=get_topic;f=144;t=001337)

ploughman
10-24-2005, 10:02 AM
Thanks. I'm astonished at how far out those ingame roll rates you've compiled are compared to real world data.

Capt.LoneRanger
10-24-2005, 10:31 AM
Nice to read. Thanks for link & testing.

And if you find ANY sim out there closer to reality, please let me know. http://forums.ubi.com/images/smilies/16x16_smiley-wink.gif

Chadburn
10-24-2005, 11:18 AM
Originally posted by TAGERT.:
Enjoy


Thanks for the hard work and sharing these.

I think it would be interesting to recreate the NACA chart using the in-game data. Planes performing close to their real-life counterparts would be great, but if they're not, it would be nice for them to at least maintain some relative performance compared to each other in the game.

I've combined two of your charts to show what I mean. Not only are these two planes not performing to their historical counterparts, but more importantly, they are not maintaining their relative performance in-game against each other either. (Unfortunately, the NACA chart doesn't say which FW190 is being tested, but you get the idea.)

http://home.cogeco.ca/~jkinley/Roll_rates01.jpg

Capt.LoneRanger
10-24-2005, 11:52 AM
Now THAT is a very good point, Chadburn!

Kwiatos
10-24-2005, 12:46 PM
Plz add SPitfire to the chart http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

AND P-47...

That will be shame...

SlickStick
10-24-2005, 08:23 PM
Wow, that's some mighty fine chartin' Tagert.. It is going to be interesting to see how the late-war CWs do in this test. The 1943 Vb CW is getting robbed blind!!!!! http://forums.ubi.com/groupee_common/emoticons/icon_eek.gif

As I'm just stating the blatantly obvious from your data, it's supposed to out roll the FW up to near 220mph and it drops to an almost pitiful 100?/S at the speed, when it should be the fastest @ 150?/S, and just keeps falling.

Now, I've never really complained about the physics of this game, just taking it for what it was, adapting, and have also been one to not look a gift horse in the mouth, but that's got to be an oversight, no?

Some of the charts look like the curve is shifted left, too fast, too slow, as you've stated, while other roll rates are just strange when compared to their real life counterpart.

Great way to see it so plainly. One thing I'm curious about. Would someone else and their different joystick settings yield different in-game data than yours? Or is it based on specific settings? (I'm probably Slick-come-lately on this one.)

It still amazes me the things folks do around here to help. Hmm, sometimes the internet can be a good place. =S=

TAGERT.
10-24-2005, 09:15 PM
Originally posted by Chadburn:
Thanks for the hard work and sharing these.

I think it would be interesting to recreate the NACA chart using the in-game data. Planes performing close to their real-life counterparts would be great, but if they're not, it would be nice for them to at least maintain some relative performance compared to each other in the game.

I've combined two of your charts to show what I mean. Not only are these two planes not performing to their historical counterparts, but more importantly, they are not maintaining their relative performance in-game against each other either. (Unfortunately, the NACA chart doesn't say which FW190 is being tested, but you get the idea.)

http://home.cogeco.ca/~jkinley/Roll_rates01.jpg NICE!

Im glad I went out of my way to keep all the graphs the same size so folks could do this kind of stuff!

And from that overlay, the Fw190 is getting robbed at speeds above 270mph.. relitive to a P51B. Which was the *historical* zone where the 190 should have had an advantage.. Yet in game it does not. Also, that higher than expected RR at slow speeds is actually more of a negative than a positive imho, in that the 190 rolls too fast and makes it hard to fly and basically twitchie imho! So, on both counts the 190 is getting robbed imho!

And @ the rest of you, Thanks! And thanks for the positive feedback!

Max.Power
10-24-2005, 11:29 PM
The purity of this information brings a tear to my eye. Hehe.

Chadburn
10-25-2005, 01:07 PM
Originally posted by Max.Power:
The purity of this information brings a tear to my eye. Hehe.

Yes...if only we could disect other aspects of the sim to this degree it would certainly bring to light what's good/bad/ugly in 4.02.

TAGERT.
10-25-2005, 09:57 PM
Originally posted by Chadburn:
Yes...if only we could disect other aspects of the sim to this degree it would certainly bring to light what's good/bad/ugly in 4.02. Guess you missed my climb rate, dive rate, top speed, etc testing I have done in the past? http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif Im currently adding turn rate and energy testing. But, I like to keep the testing to things we have real world data on.. the energy stuff is just for fun.

TX-Gunslinger
10-26-2005, 01:55 AM
Tagert,

Thanks for the unbiased analysis and hard work. I've had that NACA report for years and wished for charts like you are producing for the sim. I had hoped to produce outputs like these from UDP graph after it became available. I "felt" that the roll rate was not quite right for some of these planes, but not to the degree you've found.

One of the first things I did when we got the CW Spitfires back in AEP was to test it's roll rate. I expected it to "feel" somewhat like my Focke Wulf's due to the NACA chart, but it did'nt.

You correctly point out one of the most annoying behavior issues for me with Focke Wulfs, particularly Antons. The most significant manever advantage for an Anton is high speed roll rate since it can't turn. The roll rate advantage we have is at unusably slow speeds.

Comparing the relative shapes of the "in game curves" across these different aircraft makes one wonder what input parameters the IL2 engine utilizes from the aircraft files to produce this output. It is striking to me that not one of the curves you've measured is smooth. They all have the same "peaked" shape of an FW/CW Spit. The smooth curves of the P40/P51B/P63 from the real-world data, aren't reproduced in the output of the FM.

An even more telling anomaly, is in the shape of the 190 A4 peak. It departs from the general shape of the all other aircraft at the peak roll speeds, possibly representing a caustic in the model.

To be honest, the "caustic" could also be the result of a bad data point for whatever reason.

It would be very interesting to see more data points in that area (160-220 mph, but I can imagine the tediousness of such a drill so I won't even ask.)

This is some of the most revealing testing I've ever observed in this sim. Thank you.

Do you happen to have links to your other charts?

S~

Gunny

Chadburn
10-26-2005, 06:58 AM
Originally posted by TAGERT.:
Guess you missed my climb rate, dive rate, top speed, etc testing I have done in the past? http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif Im currently adding turn rate and energy testing. But, I like to keep the testing to things we have real world data on.. the energy stuff is just for fun.

Unfortunately I did miss them! But I'd really like to see them if they are available on-line somewhere.

People can talk all they want about how this or that "feels", or relate a dogfight experience to support an argument for modifying a plane's fm, but these charts are infinitely more informative.

Once again, thanks for taking the time to do this. As Gunslinger said, testing this game is tedious work, so it's greatly appreciated that you've put in the time and effort.

Udidtoo
10-26-2005, 07:49 AM
The work you put into this is appreciated Tagert. As were the previous one's as well.

OldMan___
10-26-2005, 09:05 AM
Originally posted by TX-Gunslinger:
Tagert,

Thanks for the unbiased analysis and hard work. I've had that NACA report for years and wished for charts like you are producing for the sim. I had hoped to produce outputs like these from UDP graph after it became available. I "felt" that the roll rate was not quite right for some of these planes, but not to the degree you've found.

One of the first things I did when we got the CW Spitfires back in AEP was to test it's roll rate. I expected it to "feel" somewhat like my Focke Wulf's due to the NACA chart, but it did'nt.

You correctly point out one of the most annoying behavior issues for me with Focke Wulfs, particularly Antons. The most significant manever advantage for an Anton is high speed roll rate since it can't turn. The roll rate advantage we have is at unusably slow speeds.

Comparing the relative shapes of the "in game curves" across these different aircraft makes one wonder what input parameters the IL2 engine utilizes from the aircraft files to produce this output. It is striking to me that not one of the curves you've measured is smooth. They all have the same "peaked" shape of an FW/CW Spit. The smooth curves of the P40/P51B/P63 from the real-world data, aren't reproduced in the output of the FM.

An even more telling anomaly, is in the shape of the 190 A4 peak. It departs from the general shape of the all other aircraft at the peak roll speeds, possibly representing a caustic in the model.

To be honest, the "caustic" could also be the result of a bad data point for whatever reason.

It would be very interesting to see more data points in that area (160-220 mph, but I can imagine the tediousness of such a drill so I won't even ask.)

This is some of the most revealing testing I've ever observed in this sim. Thank you.

Do you happen to have links to your other charts?

S~

Gunny

The strange FW chart shape may be direct effect of the samet hign that makes it be the best roller of all (don 't know what it is).


Anyway THAT is the point where the FW is more UNDERMODELLED (even the super roll in very low speeds is undermodelling). FW simply cannot use they real world tatics and are fated to survive solely based ona good elevator authority.



A tune up in rol rates, specially in planes with special considerations upon it (like FW190 , spits vs cliped spits etc...) should be adressed when possible.

mynameisroland
10-26-2005, 10:00 AM
Thanks Tagert for the great work , would it be possible for you to post links to the other charts you have compiled Id be very interested in viewing them.

BaldieJr
10-26-2005, 10:35 AM
Under what circumstances are the rolls carried out? Counter/Clockwise? With/Without rudder input?

Its not like we can go back in time and grab the pilots to perform the test (I'm certain a trained test pilot will get different results compared to say, some guy).

Nor can we force G effects or stick-forces into the equation.

So I argue reality.

MEGILE
10-26-2005, 10:42 AM
The Focke Wulf's undermodelled roll rate is one thing.. but damm the Spit CW... http://forums.ubi.com/images/smilies/16x16_smiley-surprised.gif

Kwiatos
10-26-2005, 11:41 AM
Yea most planes are overmoddeled at low speeds and are undermoddeled at high speeds. The biggest hurt plane is Spitfire CW and normal wing. Fw 190 still have good roll rate at high speed even when is undermoddeled. See P-47 which should be good roller at high speed and in game is very weak.

http://www.simhq.com/simhq3/sims/boards/bbs/ultimatebb....c;f=144;t=001337;p=1 (http://www.simhq.com/simhq3/sims/boards/bbs/ultimatebb.php?ubb=get_topic;f=144;t=001337;p=1)

TX-Gunslinger
10-26-2005, 12:44 PM
Originally posted by Megile:
The Focke Wulf's undermodelled roll rate is one thing.. but damm the Spit CW... http://forums.ubi.com/images/smilies/16x16_smiley-surprised.gif

Ok, as far as roll rate alone, the dip in the curve is much worse than the FW over a significant speed regime.

From a relative balance perspective, you still have a signficant turn rate, climb rate, E-retention and looping advantage
over the Fock Wulf. Why do you think that your CW Spit porked roll rate is any more of a priority or travesty than the FW?

The way I see it is that yes, the CW is porked too, but still retains superiority in many other areas of manuever. Imagine trying to make best use of roll rate on a high wingloaded aircraft with poor turn and vertical manueverability.

Anyone who routinely flys FW at speeds below 210 mph (bout 330 kmh), is suicidal. Why? Because below that speed acceleration is very bad. Simply stated, the speed regime where the roll rate is the best is also the speed regime where the FW has the poorest acceleration, so if you go down in that regime almost every enemy A/C has superior acceleration and the only way you can match it is by giving up your altitude.

I would argue that the cumulative effect is greater on the FW.

But if I was a CW Spit flyer, I'd be upset.



S~

Gun

msalama
10-26-2005, 12:59 PM
Tagert, I salute you for your excellent work. This kind of critique/research is _EXACTLY_ what we need - and which has been sorely lacking so far.

But one small question still: would it be possible to compile all the data, when the testing is finished, into one big document/presentation/paper/whatever?