Jump to content


Photo
- - - - -

SE5a FM - review & fixes!


  • Please log in to reply
331 replies to this topic

#121 Josh_Echo

Josh_Echo
  • Posts: 3931

Posted 10 September 2011 - 18:30

Simply Incredible response and attention to detail - Brilliant and I can't wait to see the results !
Huge thanks for all this great work - I can't wait to try the new SE5a FM !

Huzzah!
  • 0

#122 MiG-77

MiG-77
  • Posts: 2651

Posted 10 September 2011 - 19:31

An.Petrovich post should be placed as sticky, so every one would know how RoF FMs are made ;)
  • 0

#123 Capt.OReilly

Capt.OReilly
  • Posts: 170

Posted 10 September 2011 - 19:39

Thank you for this depth of research & investigation - it's what sets RoF apart from some other sims out there.

I look forward to the new flight model!

:S!:
  • 0

#124 =Fifi=

=Fifi=
  • Posts: 10329

Posted 10 September 2011 - 20:01

Thank you for this hard work, and most of all for your head up!

My english isn't that good :oops: but i've understood that every FM factor is linked to others, and change one may change bad way others…right?

The result of your new SE5 FM, might give a totally new SE to fly, and some SE addicts could be a bit desapointed?…(slower roll rate etc)

Anyway, if the glass engine is fixed, that's good thing :) Hoping other changes aren't that big…
  • 0

#125 89

89
  • Posts: 593

Posted 10 September 2011 - 20:01

Thank you very, very,very much! SeVa was such an amazing aircraft and it is such a privilege to experience such an accurate computer model of it!
  • 0

#126 Greywing2

Greywing2
  • Posts: 433

Posted 10 September 2011 - 21:10

I knew Andrej is "tha man"….really, what a dedication…..I hope, that after DVII and Se5a we will see fixed N17 roll rate, Pup climb rate, Brisfits, etc…..
  • 0

#127 Gunsmith86

Gunsmith86
  • Posts: 842

Posted 10 September 2011 - 22:24

Thank you very much for all the hard work you have done!!!
  • 0

#128 catchov

catchov
  • Posts: 4166
  • LocationQld, Australia

Posted 11 September 2011 - 00:47

- is here some another S.E.Va with another Viper engine??? I don’t understand it all… No any weight loading or low engine compression can give the same result to the this source. On this picture the airplane is dramatically slow and it's a very pure climbing. May be somebody tell me - how is it possible?.. and… welcome to the club, who has read all until this line. :)

It is not possible surely Andrey ? It must be wrong. I think you have done all you can to get a good approximation of what will be the RoF Se5a's new performance. :)

And now, I have to salute you, for your outstanding contribution to RoF. :S!:

And I also take my hat off. 8-)
  • 0

#129 Rexmundi

Rexmundi
  • Posts: 160

Posted 11 September 2011 - 01:16

I've never seen such an open debate including statements of uncertainty with an FM programmer before, thanks for that.
  • 0

#130 piecost

piecost
  • Posts: 1318

Posted 11 September 2011 - 02:20

Thank you Andrey, the explanation is above and beyond what we should expect.

Well done for getting the RoF to within 5% of the quoted performance! I'm sure that the original planes were not that consistent. One document I have seen mentions that the performance measured from 2 examples of the same type of propeller should not be compared ! (due to differences in the stiffness of the wooden blades, warping due to changes in humidity and damage to the blades, I assume).

I found inconsistency in the drag polars for the SE5a:

Flight Model …..+544#p293139
post #133

A polar calculated from glide angle data in R&M 544 showed a different curve to R&M 603. I don't know if the alternative drag polar gives a closer match to Climb_T28118.jpg. But that does not matter and you have the SE5a sorted :)

Attached Files


  • 0

#131 J2_Adam

J2_Adam
  • Posts: 2453
  • LocationVancouver, BC

Posted 11 September 2011 - 03:08

Good one Andrey! Thanks for the close attention to detail. Nice job.
  • 0

#132 piecost

piecost
  • Posts: 1318

Posted 11 September 2011 - 12:27

Found another document, paragraph 5 has interesting information about the vibrations experienced when diving.

Stresses in an aeroplane when Diving Steeply

Reports & Memoranda No. 494

Attached Files


  • 0

#133 hq_Jorri

hq_Jorri
  • Posts: 14143

Posted 11 September 2011 - 13:14

My hat is off to you, An.Petrovich. The whole team does such great work, but you keep the heart of the simulator pounding: the flight models!

I think we all agree it is in good hands :)

Also thanks to Piecost. You should do him a job offer :D
  • 0

#134 gavagai

gavagai
  • Posts: 15542

Posted 11 September 2011 - 13:51

I had a chance to thoroughly read Andre's post, and I am happy with his approach to the rate of climb. A little faster to 5000ft, a little slower to 10000ft; it seems like the right compromise.
  • 0

#135 FlyingShark

FlyingShark
  • Posts: 1941

Posted 11 September 2011 - 14:03

Thank you for your hard work, Andrey.

:S!:
  • 0

You can vote my post up by clicking the green arrow on the right.


#136 O_Taipan

O_Taipan
  • Posts: 2291

Posted 11 September 2011 - 14:18

Looking good! Thanks for this, making the best flight sim even better ;)
  • 0

#137 -Ginge-

-Ginge-
  • Posts: 200

Posted 11 September 2011 - 15:21

many thanks Andrey. I see the Se5a as one of the most important planes (not as important as SPADs of course, Fifi!), and I really appreciate the time and effort you are putting into this.

Salute
  • 0

#138 NickM

NickM
  • Posts: 1625

Posted 11 September 2011 - 16:07

:S!:
  • 0

#139 swissdragon

swissdragon
  • Posts: 144

Posted 11 September 2011 - 21:41

Wow, what a great mini review about FM's.

Thanks a lots for this highly interesting details Andrey "Sherlock Holmes"!
  • 0

#140 =IRFC=AirBiscuit

=IRFC=AirBiscuit
  • Posts: 2455
  • LocationNaples, FL USA

Posted 12 September 2011 - 14:52

This is the kind of communication I wish we had been getting all along, but I'm still super happy to see it coming now. It's convinced me to go buy the Spad 7..

I really hope the lot of earlier planes will be reviewed. 2/3 of the Nieuports are quite flawed, and the Albatros and Pfalz D.IIIa all share the same, generic placeholder FM in my opinion. It would be so wonderful (and critical to the combat experience) if these planes each received this same degree of refinement.
  • 0

=IRFC=Air Biscuit

http://quetoo.org


#141 An.Petrovich

An.Petrovich
  • Posts: 565

Posted 12 September 2011 - 17:02

Hi everybody,
and thanks a lot for your attention to FM!
I've made this report above special to explain you how FM works and tune (I know, it's always interesting for gamers; cause I've been a gamer myself some years ago), and moreover, I hope it'll cancel any speculations about FM fixes in RoF in the future. We know about some issues of FM for some airplanes in RoF, and will fix them only according to the true sources and docs, after deep analysis of flight performances these airplanes. Not because "enemy" better than "friendly", of course. :)

Don't know if you have this data …
Exactly. You can find it in my report.

I have some more written report on these tests and also some handwritten RAE reports on French built 150/180/200 geared Hisso's if required.
Thank you! Please, post these docs here.

The links to the second source of engine power curve and the source of general geometry of the airscrew is not available for me at least.
Hmmm… still now?

Put this fix in with the stutters fix for ATI cards and call it 1.021b and you will have some very happy customers!
We're exactly going to do this. :)

Since aerodynamics of the wing have been modified (according to new CL and CD curves), then the roll rate (aileron performance) is also changed as a consequence. Now the roll rate is worse then before.

I had wondered about this. The current SE5a is in the category of super-rollers with the N28, and Albatros D.II/Va.
Some good news more:
after beta-testers complained of a very bad roll rate - I've made the aerodynamic test of S.E.Va in RoF today, and really they were right: CL curve of ailerons on S.E.Va in RoF differed from needed curve, how it must be according on calculations (in Excel). So, this bug has already fixed, and now roll rate is ok!

An.Petrovich post should be placed as sticky, so every one would know how RoF FMs are made ;)
Good idea, may be we'll do it :)

My english isn't that good :oops:
the same to me, so, let's try to understand each other :D

i've understood that every FM factor is linked to others, and change one may change bad way others…right?
It's a physics of real things, not because coded such strange. Many factors of airplane performances linked to others, that's why FM in RoF looks like to this.

The result of your new SE5 FM, might give a totally new SE to fly, and some SE addicts could be a bit desapointed?…(slower roll rate etc)
I've already answered about roll rate, but in general you almost about right: some of S.E.Va performances have changed, and players will have to get used to it.

I found inconsistency in the drag polars for the SE5a …
Great analysis, as usual! Very interesting, indeed.
Thank you for cooperation, and let's focus to the Nie.17 ;) right now, may be it helps us much?.. (but guys… don't ask me "when?" :D )

Also thanks to Piecost. You should do him a job offer :D
I agree to that. :) He really knows what about he's speaking.
  • 0

#142 SYN_Mike77

SYN_Mike77
  • Posts: 1161

Posted 12 September 2011 - 18:53

Put this fix in with the stutters fix for ATI cards and call it 1.021b and you will have some very happy customers!
We're exactly going to do this. :)

Great News! I'll be sending my marketing analysis bill shortly. Don't worry, I accept payment in field mods :lol:

Just Kidding, Can't wait.
  • 0

#143 piecost

piecost
  • Posts: 1318

Posted 12 September 2011 - 18:59

We have been very lucky since the SE5a has one of the richest sources of data in the R&M reports. I can find nearly nothing for French aeroplanes; perhaps a French forum member can do some research in the Musée de l'Air et de l'Espace archives…

However, I may have a lead on the Nieuport 17 roll rate (see attached)

I found details of an experiment using an aeroplane based camera to measure the roll rate of an SE5a:

Current ROF Airplanes Flight Model Data Topic..+851#p289271
post#50

The technique was perfected using a Nieuport, according to the attached documents. Unfortunately the graph does not give the type of plane and airspeed were not identified. So it is practically useless. However, there is a slim chance of finding some information about the development of the technique and some data for the unidentified Nieuport.
Can anyone identify the variant in the photos?

Attached Files


  • 0

#144 =Fifi=

=Fifi=
  • Posts: 10329

Posted 12 September 2011 - 20:09

I can find nearly nothing for French aeroplanes; perhaps a French forum member can do some research in the Musée de l'Air et de l'Espace archives…

:( That's really a shame…i made high hopes on you to find N17 datas…and i can't myself find anything serious (maybe my connection is too far away lol).

N17 has to be the next one on the revised list, PLEASE!!
Even the EIII/DH2 have better flying.

EDIT: SORRY, i didn't read AN post above! Great news! I'm so happy N17 will be next one :P
  • 0

#145 gavagai

gavagai
  • Posts: 15542

Posted 12 September 2011 - 20:18

Since aerodynamics of the wing have been modified (according to new CL and CD curves), then the roll rate (aileron performance) is also changed as a consequence. Now the roll rate is worse then before.

I had wondered about this. The current SE5a is in the category of super-rollers with the N28, and Albatros D.II/Va.
Some good news more:
after beta-testers complained of a very bad roll rate - I've made the aerodynamic test of S.E.Va in RoF today, and really they were right: CL curve of ailerons on S.E.Va in RoF differed from needed curve, how it must be according on calculations (in Excel). So, this bug has already fixed, and now roll rate is ok!

So, does this mean the roll rate is unchanged, or only a little bit worse than before?
  • 0

#146 =FB=VikS

=FB=VikS
  • Developer
  • Posts: 1246

Posted 12 September 2011 - 20:25

We have been very lucky since the SE5a has one of the richest sources of data in the R&M reports. I can find nearly nothing for French aeroplanes; perhaps a French forum member can do some research in the Musée de l'Air et de l'Espace archives…

However, I may have a lead on the Nieuport 17 roll rate (see attached)

btw - is there anything more exept we already have in Windsock Datafile special tables and Profile publications about N17?
  • 0

#147 SYN_Vander

SYN_Vander
  • Tester
  • Posts: 4710

Posted 12 September 2011 - 21:07


Can anyone identify the variant in the photos?

I'm pretty sure it's a Nieuport 27, look at the distinctive tail. (in the small pictures)

Image

Still, I think it will be a good benchmark? +/- 10 % ?
  • 0

#148 hq_Jorri

hq_Jorri
  • Posts: 14143

Posted 12 September 2011 - 21:14

It has rather different ailerons, though, by the looks of it..
  • 0

#149 MattM

MattM
  • Posts: 2595

Posted 12 September 2011 - 21:25

Should be N27 or N24 (non bis).

Regardless of the plane type in that test, the ROF N17 rolls about as fast as the plane on that chart (~90 degrees bank in ~3 seconds). So unless we know the speed that test was made at, i don't see how it's of any use (ignoring the fact that it's not a N17 in the first place, which had a different airfoil than the N27).

But CLmax of the airfoil would probably be enough to determine the ailerons effectiveness.

Also about the N17, i would say that the current maximum speed is a bit doubtful (even if it says the same in a Windsock Datafile). And a too low top speed could also be one reason for the (subjective) slow rollrate. As could an unusual high rollrate of the opposing planetypes.

In ROF, the direction of bank also makes a difference. Rolling to the left is much "quicker" than rolling to the right (it's probably vise versa, not sure). This is probably also true for other planetypes, but it's especially noticable in the N17.
  • 0

#150 =Fifi=

=Fifi=
  • Posts: 10329

Posted 12 September 2011 - 22:00

Also about the N17, i would say that the current maximum speed is a bit doubtful (even if it says the same in a Windsock Datafile). And a too low top speed could also be one reason for the (subjective) slow rollrate. As could an unusual high rollrate of the opposing planetypes.

Shouldn't be faster than N11?
It's subjective, but i always feel faster in N11…in any rate.
N17 slow speed could be the answer to the bad roll rate, i agree with you.
What puzzle me is the N17 huge top wing ailerons, and i was expecting a quick roll rate due to it!

Can't find any official data on the web. Very desapointing :(
  • 0

#151 SC/JG_Oesau

SC/JG_Oesau
  • Posts: 2024

Posted 12 September 2011 - 22:03

We have been very lucky since the SE5a has one of the richest sources of data in the R&M reports. I can find nearly nothing for French aeroplanes; perhaps a French forum member can do some research in the Musée de l'Air et de l'Espace archives…

However, I may have a lead on the Nieuport 17 roll rate (see attached)

btw - is there anything more exept we already have in Windsock Datafile special tables and Profile publications about N17?

Three is a the Crowood Nieuport aircraft of WWI book, however it's performance figures are inline with the windsock data.
  • 0

#152 piecost

piecost
  • Posts: 1318

Posted 12 September 2011 - 22:14

The only information I have is an article about flying a radial powered replica. This leads to lots of open questions about how representative it is of the original aeroplane.

Data Topic for Airplanes Performance.
post#25

It is well worth a read.
  • 0

#153 gavagai

gavagai
  • Posts: 15542

Posted 12 September 2011 - 22:46

Shouldn't be faster than N11?
It's subjective, but i always feel faster in N11…in any rate.

Not subjective at all, fifi. The N11 is faster than the N17.
  • 0

#154 =Fifi=

=Fifi=
  • Posts: 10329

Posted 12 September 2011 - 22:49

Shouldn't be faster than N11?
It's subjective, but i always feel faster in N11…in any rate.

Not subjective at all, fifi. The N11 is faster than the N17.

So, should be very first thing to adjust/correct and see if something changes in the roll rate :)
  • 0

#155 gavagai

gavagai
  • Posts: 15542

Posted 12 September 2011 - 23:01

I don't see why it would make much difference. The N11 rolls faster at 150kmh than the N17 rolls at 150kmh.
  • 0

#156 Josh_Echo

Josh_Echo
  • Posts: 3931

Posted 12 September 2011 - 23:05

Moreover, I'm pretty sure that our Nieuport 11 rolls faster at its worst rolling speed than Niueport 17 rolls at her best rolling speed.
  • 0

#157 =Fifi=

=Fifi=
  • Posts: 10329

Posted 12 September 2011 - 23:44

Of course, good point…so i don't know what can be wrong in N17 (except the speed) :?

Hope devs will find out.
AN.Petrovitch is my only hope :lol:
  • 0

#158 gavagai

gavagai
  • Posts: 15542

Posted 13 September 2011 - 00:16

I hope the F.2B's climb is given some attention, too.

Attached File  bristol climb.PNG   93.47KB   277 downloads

As Mig-77 has reported:

Falcon II Bristol:

Climb:
1000ft 00:00:50,62
2000ft 00:01:44,37
3000ft 00:02:38,62
4000ft 00:03:41,46
5000ft 00:04:42,73
6000ft 00:05:51,34
7000ft 00:07:00,39
8000ft 00:08:21,65
9000ft 00:09:53,93
10000ft 00:11:32,70

Falcon III Bristol

Climb:
1000ft 00:00:44,59
2000ft 00:01:35,29
3000ft 00:02:23,04
4000ft 00:03:14,90
5000ft 00:04:11,32
6000ft 00:05:09,76
7000ft 00:06:10,04
8000ft 00:07:20,39
9000ft 00:08:34,51
10000ft 00:09:52,21

Thing I hate about the Bristol Fighter

The store page data seems reasonable, but the in-game F.2Bs do not match it.
  • 0

#159 HotTom

HotTom
  • Posts: 8177

Posted 13 September 2011 - 00:28

Gav, your chart doesn't say which engine. The FII is very close.
  • 0

#160 Josh_Echo

Josh_Echo
  • Posts: 3931

Posted 13 September 2011 - 00:55

Tom, that chart shows the F.III engine but with a time-to-climb that's worse than our F.II engine. The error according to that chart is about 16% for our F.III.
  • 0


0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users