Jump to content

Recommended Posts

Fonck was very well known to be a an excellent shooter.

Even on the ground he could hit a coin with a gun, as often reported.

He used it in air combat by shooting either the engine or the crew.

Against 2 seaters, 1 small burst against the gunner, one small burst against the pilot.

  • Like 1

Share this post


Link to post
Share on other sites

Aug. 30, 2016 Update: A few new plane types added (Fokk. E.IV, the late Nieu. 16 with Vickers, an early "radiator ears" Alb. D.II, the BE12 and 12a types), the Halb. D.V further tuned, and other tweaks such as a modified general aircraft ai data ini file included. Please consult the relevant section of the included "Read Me" file for more detailed info. and installation instructions.

 

Happy flying,

Von S

  • Like 2

Share this post


Link to post
Share on other sites

Hi VonS

Thank you very much for the latest update! :good:
Your efforts are much appreciated! :kudos:

Here are a few more observations, please note that this is with Realism Pack v.8, since I didn't notice the update until now.

Albatros DII: Soon after take off the AI's goes into a steep climb to about 500m. This behaviour seem to be a common FE2 behavior though, and other aircraft behaves similar. I recall I've seen a tweak for this on the forum, but I can't find that thread again....

If I had climbed like that I would have stalled. Does the AI have different FM than the player?

Albatros DII: When in formation the AI is often flying slightly below my aircraft, so they are hard to keep in view.

In a few occations I have shot off the vertical stabilizer and the rudder on Nieuport 17, but they continue flying and manouvering as nothing has happened. :crazy:


S!

vonOben

  • Like 1

Share this post


Link to post
Share on other sites

Hi VonOben,

 

Thank you for the continued comments on the FM tweaks - it is always interesting to read the feedback.

 

The steep climbs that the AI sometimes engages in, upon takeoff, can be remedied through the following two entries under the AIdata section of the data ini files, if I remember correctly from previous posts on the FE2 forums:

 

PitchForAltitude=

 

TakeOffRotationAngle=

 

I would start first by tweaking the RotationAngle, without touching PitchForAltitude - to see if that improves the behavior.

 

The values on the early Alb. D.II for example, for PitchForAltitude and RotationAngle, are 0.00587 and 15.0

 

I recommend that you lower the RotationAngle to something between 5.0 and 10.0 - this may give an improvement (more gradual ascents rather than the extreme, vertical takeoffs that sometimes happen). The AI, theoretically, should be using the same FM as the player - but can sometimes exploit that FM more effectively, flying to the edge of its flight envelope before stalling, spinning, and so on.

 

In formation flying, the AI on several aircraft types are sometimes slightly below the leader - I have noticed this with some of the Eindeckers too - especially if I am fairly high up (around 2,500 meters altitude). I recommend that you reduce the MaxRollFormation entry to something around 3.0 (to reduce waltzing during formation flight...and this might help to stabilize the altitude at which the AI flies). The MaxRollFormation entry is located under AIdata in the data ini files (and overrides that entry in my general aircraft ai data file).

 

It's possible to keep flying if your rudder and vertical stabilizer are shot off - as long as your ailerons are still operating, and the elevator. I have managed to crash land aircraft with a lost rudder on a few occasions (such as the Fokk. E.IIa), and while difficult it is not impossible - I recommend that you control engine throttle if you lose a rudder, or the blip switch on rotaries (manipulation of the engine gives more control in those situations than if you forget to use the engine - try to think of the engine as your extra control surface in those cases...might not always work but it's worth a try).

 

I am assuming that since the Nieu. 17 is maneuverable, the AI is probably using engine power, ailerons, and elevator as best it can when the rudder is shot off - to try and continue flying. I have seen other aircraft that do not take advantage of this. For example, the Halb D.5 - when its rudder is damaged, or even half of its elevator - usually attempts a wide turn that ends up being a shallow, wide dive into the ground.

 

Happy flying,

Von S

Edited by VonS
  • Like 1

Share this post


Link to post
Share on other sites

Hi VonS

 

I'll test the RotationAngle as you suggested!

 

My concern regarding the Nieuport 17 with shot off stabilizer and rudder, was that the AI could maneuver it just as good as it was undamaged. A shot off stabilizer and rudder ought to have some impact on the maneuverability, also for the AI IMHO. But that is perhaps nothing to do about it?

 

I have tested your updated files a bit now in my Armchair Aces campaign and the AI shoot much but they are very bad hitting, so I went back to my own AIRCRAFTAIDATA.INI and AIRCRAFTOBJECT.INI files where the AI also shoot much but hits a bit better.The AI seem to perform slightly better with later planes with double machine guns compared with the Eindeckers, and they are fairly aggressive. But it might also be the effect of my ongoing edits...

 

Here are some figures for comparison (all the missions were rather similar with fights with early Albatros D.II agains Nieuport 17):

 

TweakedFMandRealismPack8.5 unchanged:

 

1917-01-10: Fired / Hit / % / Kills
Weissenberger 800 / 20 /  2.5 / 1
Fahl                    60 /   0 /  0 /    0
Schenk             800 / 20 /  2.5 / 0

 

1917-01-13: Fired / Hit / % / Kills
Weissenberger 774 /  1 / 0,1 / 0
Hartmann         122 /  1 / 0,8 / 0
Hahn                644 / 22 / 3,4/ 1

 

Using my edited AIRCRAFTAIDATA.INI & AIRCRAFTOBJECT.INI (with VolumeStructurePoint=6000):

 

1917-01-16: Fired / Hit / % / Kills
Johnen     242 / 18 /  7,4 / 1
Harbort     550 / 98 /18    / 0
Kompch    712 / 39 /  5,5 / 2

 

1917-01-19: Fired / Hit / % / Kills
Heckman  800 / 89 / 11    / 2
Harbort     800 / 42 /   5.3 / 0
Kompch    274 / 68 /  25   / 0

 

Best regards

 

vonOben

Share this post


Link to post
Share on other sites

Hi VonOben,

 

Thank you for the feedback. It is strange that the maneuverability would not be impacted at all on the Nieu. 17 once the stabilizer and rudder are removed. I will try to reduplicate this on my system to see what happens - possibly some missing info. in the data ini files is the cause of this...conversely, I have noted that there are a few data ini files (for some plane types) that include data for a "stabilizer" where there is no stabilizer, only a free-floating rudder, such as the comma-shaped one on Eindeckers. My theory is that, in such cases, a shot off rudder would not impact greatly on flight control considering that some phantom stabilizer info. is still present in the data inis. This perhaps impacts most of all on the Nieu. types...this will be interesting to look into.

 

Your aircraft ai data ini results look very good - I have been successful at increasing firing times, pilots now usually get all of their ammo. out, but I will see if smaller firing angles will help improve accuracy further. My general aircraft ai data file is, by the way, tested in early and mid-war settings - it's possible that the frequent firing is more effective with double machine gun setups in late 1917 and 1918 scenarios.

 

Happy flying,

Von S

Edited by VonS
  • Like 1

Share this post


Link to post
Share on other sites

Maybe rudders are meant to be attached to stabilizers by the game (this is a "son of a jet sim" anyway). Maybe you can change the damage box to have the faux stabilizer more coincident with the rudder one. This way, when the rudder is lost, also is the stabilizer. You can also change the stabilizer DamageRating to DESTROYED.

Your work look very promising. Some day I'll try it.

Share this post


Link to post
Share on other sites

Thank you Ojcar for the kind words regarding the FM tweaks. I will look further into the faux stabilizer and will test it with the damage rating set to destroyed. Will report back once I try out some variations with the Nieu. types.

 

Happy flying,

Von S

  • Like 1

Share this post


Link to post
Share on other sites

Hi VonS

 

I've switched to Albatros D.III (160) in my campaign before I had the chance to test your RotationAngle suggestion, but the D.III doesn't have this problem.
Comparing the values: The D.II has a has a much bigger PitchForAltitude value than the D.III and The RotationAngle is the same, so that's probably the cause.

Albatros D.II Early
PitchForAltitude=0.00587
TakeOffRotationAngle=15.0

 

Albatros D.III (160)
PitchForAltitude=0.00040
TakeOffRotationAngle= not in ALBATROSD3_DATA.INI

My edited AIRCRAFTAIDATA.INI:
TakeOffRotationAngle=15.0

 

Regarding the AIRCRAFTAIDATA.INI I have much lower values for CannonFireAngle, between 3.0 (Ace) and 4.0 (Novice).
I also have much lower MaxCannonRange: 200 for all, if they fire at 1000m they are only wasting their ammo IMHO.
Ive also increased MinCannonRange: 70 for all to prevent the AI's getting too close and crashing into each other.

 

Some other observations: the SPAD 7 C1 (150) catches fire easily and seems slightly faster than the Albatros D.III (160).

 

Best regards

 

vonOben

  • Like 1

Share this post


Link to post
Share on other sites

The spad7 (180 hp)'s commands look very heavy, I am not very experienced, but for instance, I find it even less maneuvrable than the Spad 13

Share this post


Link to post
Share on other sites

Hi VonOben and Jeanba,

 

I've double-checked the data inis now and the PitchForAltitude command is the one to tweak for more gradual takeoffs - values such as indicated for the Alb. D.III should be good ones for a gradual ascent upon takeoff - something around 0.0005 to no more than about 0.001.

 

I will reduce the MaxCannonRange in my general aircraftaidata ini file and will retest. This will probably increase the number of good hits by the AI (thank you for sharing the data info. with me VonOben).

 

Glad to hear that the early SPAD 7 is burning well and catches fire frequently. This once again proves that the following two entries are working:

 

DamageTempDelta=60

OverheatDamageRate=0.7
 
(This translates to high post-damage rate for the early inline engines and increased risk of fire.)
 
All of the SPADs should be faster than the Albatros (both early and late varieties of Alby.). The 150HP SPAD 7 has a top speed of about 192 kph, as is historical for the type. The 180HP SPAD 7 has a top speed of about 205 kph. This again is in keeping with historical stats. The SPAD 13, depending on the power plant, ranges anywhere from about 205 kph to 218 kph for top speed. Only the SE5a and Italian Ansaldo are faster aircraft, top speeds in the 220s kph, and the higher-HP (late) variant of the Airco DH.4 (about 230 kph).
 
The Alby. types range in top speed from 168 kph (for the D.I) to about 186 kph (for the 180 HP variant of the D.Va). Austrian Alby. types are faster (different power plants), with late variants of their D.III ranging from 195 to about 203 kph in top speed. German variants of the Alb. D.III vary in top speed, depending on HP, from about 172 to 175 kph.
 
The SPAD 7 aileron input sensitivity, also rudder and elevator sensitivity, etc., are "approximations" of how it might behave - since they are only flown by the AI in game. I therefore had to go "backwards" from the SPAD 13 which is flyable in game - to get approximations of the total engine output, sensitivity, etc., for the SPAD 7 data inis. There is a SPAD 13 to 7 conversion file available on CombatAce but I did not tweak that since it merely disables one of the machine-guns on the 13, but still gives the larger surface area of the 13. From what I have been able to observe, the SPAD 7 variants do approach top speeds of about 190 to about 204 or 205 kph in game - for more accurate tweaks of the SPAD 7 I would have to be able to get into a "flyable" version of it. Perhaps some of our clever modders will be able to do something with this (I have experimented with getting a flyable SPAD 7 to work but had no success, by "cloning" a SPAD 7 folder and applying necessary files to the folder).
 
Currently I am testing the faux stabilizer situation on the Nieu. 17 and 24 types and hope to have some results soon.
 
Happy flying,
Von S

Share this post


Link to post
Share on other sites

Hi VonOben,

 

After further testing, what works best is if the DamageRating setting is changed from HEAVY to DESTROYED under the [VertTail] section of the data ini files (thank you Ojcar) for the Nieu 17, 17bis, and 24bis types. A shot off rudder will then make the AI more sluggish - it can still fly but not maneuver as effectively, making it easier to shoot down. Have now tested this in one-on-one dogfights between Nieu. 24 types (RAF) and Nieu. 17bis types (German-flown) in the Palestinian theater, spring of '18.

 

The earlier Nieu. types and the 27 and 28 don't appear to suffer from this problem but do let me know if peculiarities are present - the same advice as above should then apply to solve the problem.

 

Do not remove the [VertTail] component of the data inis since, even though a faux stabilizer, this will ruin the FM model for the planes that have this in place - have tested with the [VertTail] section missing from the Nieu. 24bis file and there is awful left/right yaw then present, making the plane very difficult to control (presence of the faux stabilizer data, from what I could experiment with, seems to bring the FM closer to historical behavior for the 24 type). Its presence in the other Nieu. type files is probably integral to the FM behavior too - adding some minor directional stability to the types, further compensated through rudder input.

 

Happy flying,

Von S

  • Like 1

Share this post


Link to post
Share on other sites

Hi VonOben and others following this thread,

 

Here is some more info. regarding the "faux stabilizer" situation, after further study. Looking through other Nieu. data ini files, and those of Fokkers, etc., I have found that the [VertTail] component of the files may correspond directly with the rudder itself - in some cases its component name is Rudder. Then again, in other cases, it carries a generic component name of Tail, and sometimes it is actually called Vertical Stabilizer. There is a separate section in the data inis for [Rudder], where the control surface data may be manipulated - as I often do when tweaking the FMs - but it is my conclusion from further study that the [VertTail] entry is best left in place, as indicated in my previous post, with its DamageRating set from Heavy to Destroyed. Below is a list of the DamageRating for the Nieu. types in unmodified form, under the [VertTail] section:

 

Nieu. 10 - Heavy

Nieu. 11 - Heavy

Nieu. 16 - Destroyed

Nieu. 17c.1 - Heavy

Nieu. 17bis - Heavy (modifying this to Destroyed makes the AI behave more realistically post-hit)

Nieu. 23 - Heavy

Nieu. 24bis - Heavy (modification to Destroyed gives greater realism as for the 17bis type)

Nieu. 27 - Destroyed (two-gun variant, Vickers and Lewis)

Nieu. 27 FAS - Destroyed (French Air Service, Vickers only)

Nieu. 28 - Destroyed

 

Happy flying,

Von S

  • Like 1

Share this post


Link to post
Share on other sites

Hi VonS

 

Some more observations for you:
 
I saw a Spad VII started burning, and the fire was about half way between the pilot and the vertical stabilizer, what can burn there?
The Sopwith 1 1/2 Strutter seems very fragile, a few hits and it starts burning (I'm using VolumeStructurePoint=6000).

 

Cheers

 

vonOben

  • Like 2

Share this post


Link to post
Share on other sites

Hi VonOben,

 

The Spad VII fire problem is probably the result of poor positioning of the ExhaustPosition coordinates. Those are located under the Engine section of the Data ini files. Have caught the same problem on the Fokk. Eindeckers and corrected that in the earlier FM update versions, but haven't flown the Spads enough to notice the problem. I have now retested flying Albatroses against Spads, and better coordinates are as follows:

 

ExhaustPosition=0.0,0.30,0.04

 

This works well for the 150 and 180 HP Spad 7 variants - it gives a good fire that begins near the front of the engine and engulfs the pilot seating area.

 

The 1 1/2 Strutter was gradually replaced by the Pup, then by the Tripe. I have read some reports that the Pup was somewhat fragile, with wings sometimes folding after heavy dives. The 1 1/2 Strutter was similarly fragile I would think historically, but fairly maneuverable, much like the Pup. The Strutter should be somewhere between the early Nieuports and later Nieuports strength-wise.

 

Happy flying,

Von S

  • Like 1

Share this post


Link to post
Share on other sites

Hi VonS

I've edited the Spad 7 as you suggested:
ExhaustPosition=0.0,0.30,0.04

A bit strange that the exhaust position controls where the fire show up IMHO, but I haven't seen fire in strange places after the edit.

Another question please:
I don't have the OberurselUR3 sound file, do you know where I can download it?

S!

vonOben

Share this post


Link to post
Share on other sites

Hi VonOben,

 

Yes it's strange that the exhaust coordinates control the fire position too, but it is at least consistent across all of the data ini files, which simplifies moving the fire to appropriate areas on the aircraft.

 

For the Oberursel UR.III, I use the "alternative" UR.II sound that is included with the big engine sounds pack available here (have also pm-ed you the alternative sound):

 

http://combatace.com/files/file/13385-realistic-engine-sounds/

 

Happy flying,

Von S

Edited by VonS
  • Like 1

Share this post


Link to post
Share on other sites
 

See results below of editing the Dr.I's aircraft fire effect location to the same values you show for the SPAD VII's exhaust position - result is perhaps a little macabre (after all, it is almost Hallowe'en), but otherwise quite satisfactory!

 

Dr.I Engine Afire (2)

Dr.I Engine Afire (1)

Dr.I Engine Afire (3)

  

  • Like 1

Share this post


Link to post
Share on other sites

Hi VonS

 

Thanks for the information, much appreciated.

I had the engine sounds downloaded already, so I have the OberurselUR3 in my install now!

Perhaps you could include this information in the readme file, or include the OberurselUR3 sound file in the pack in future upgrades?

 

I've had strange experience with the Spad VII lately in a few campaign missions:
Example: A Spad 7 C1 (180) was badly damaged and smoking. Despite of that it flew in level flight with a slight climb of about 0,5m /sec at 200 km/h.

Is it a similar "über" problem that was in RB3D, with some planes gaining performance when damaged?

 

S!

 

vonOben

  • Like 1

Share this post


Link to post
Share on other sites

Hi VonOben,

 

I am currently tinkering further with the general aircraftaidata ini file - trying to increase hit accuracy from the AI - I will upload the changes with an 8.6 version update of the FM/realism package and will also include a link to the engine sound files, as I did previously for the custom pilot skin installs (thank you for the good advice).

 

A Spad 7 (180HP) speeding away after damage could be the result of a variety of factors - less weight if some pieces fell off, or less drag perhaps, or the engine revving at maximum rpm if throttle control has been lost - and therefore accelerating away at near top speed, and with a slight climb (the top speed on that engined-up version of the Spad 7 should be somewhere around 205 kph historically - and I have attempted to imitate that in its data ini as best as possible given that it is an AI-only plane).

 

I've had a similar experience after damaging a Morane-Saulnier H in an early Palestinian campaign, with an Eindecker, to find that it would then climb away (although it only suffered from minor tail damage in that scenario).

 

From what I've been able to observe, there is no "über" problem in FE2 - although I will be on the lookout for such peculiarities. I think that such a problem in RB3D was finally solved with the "ReDux" FM modifications for multiplayer in that game - although I always preferred the Greybeard FM tweaks over all the other FM modifications for that classic game.

 

Happy flying,

Von S

Edited by VonS
  • Like 1

Share this post


Link to post
Share on other sites

Hi VonS

 

Thanks for the information. All is ok then.

 

Regarding that the exhaust coordinates control the fire position:
That is ok for most planes, but it looks a bit strange with fire from the exhast pipes on the BE2...
Nothing to do about it I guess?

 

S!

 

vonOben

  • Like 1

Share this post


Link to post
Share on other sites

Hi Von Oben,

 

That is a good point regarding the fire from exhaust pipes on the Be2 types. It's possible that the Be12 types also have this same "problem." One solution would be to lower the fire position to where the engine is approximately - but this would create a new problem with normal exhaust fumes then coming from the engine area instead of the exhaust pipes. I will look into it once I tweak the general aircraft ai ini file some more. Perhaps the trade off is worth it - to have a fire in the appropriate position, since the exhaust fumes are barely noticeable anyway.

 

Happy flying,

Von S

Edited by VonS
  • Like 1

Share this post


Link to post
Share on other sites

Hi VonS

 

Have you figured out a way to reduce the number of collisions between friendly and enemy AI's?

I've increased MinCannonRange to 70 for all AI levels, and I think that made some improvement but it still happens too often.

Sometimes the losses in a mission is 50% caused by collisions, especially when there are many planes in the air.

Cheers

 

vonOben

  • Like 2

Share this post


Link to post
Share on other sites

Hi Von Oben,

 

I've been modifying the general aircraft ai data ini file further and haven't noticed too many mid-air collisions, maybe one in an entire mission. Today I had two Sopwith Schneiders crash into me and my wingman in E.I Fokkers once we had already landed at our base - the Schneiders were aggressive and followed us home - we were out of ammo. at that point. Otherwise  such collisions happen rarely, although I'm largely still testing things in early and mid-war missions.

 

Below are included my latest modifications to the general aircraft ai data file (will upload this with a ver. 8.6 update of the FM/realism pack should it prove better than my previous mods. of this file). It works well for one on one or one on two custom missions, also for single missions - campaigns may require more modifications of the file.

 

Happy flying,

Von S

 

Data ini code follows for the general aircraft ai data file:

 

[AIData]

AileronDeltaRoll=1.0
AileronRollRate=-0.8
ElevatorDeltaPitch=1.2
ElevatorPitchRate=-0.04
ThrottleDeltaSpeed=0.06
ThrottleVelocity=-0.06
MaxPitchForAltitude=25.0
MinPitchForAltitude=-15.0
PitchForThrottle=0.1
PitchForAltitude=0.00087
PitchForVerticalVelocity=-0.024
RollForHeading=10.0
RollForHeadingRate=-0.04
MinRollHeading=5.0
MaxRollForHeading=30.0
PitchForRoll=0.23
FormationSpeedForPosition=0.8
FormationSpeedForRate=0.5
GunnerFireChance=80
GunnerFireTime=5.0
GunnerAimOffset=0.080
GunnerAimAccuracy=65
MaxRollCombat=70.0
MaxPitchCombat=55.0
MaxRollFormation=3.0
FormationRollForPosition=0.01
FormationRollForVelocity=0.2
RollForGunAttack=1.0
RollForGunAttackRate=-0.1
RudderForGunAttack=2.8
RudderForHeading=0.7
RudderForYawRate=-0.3
RudderForSideslip=2.5
TakeOffRotationAngle=10.0
 
[DogfightNovice]
SafeAltitude=40
DefensiveAngle=35
ChanceDefensiveTurn=70
ChanceBreakTurn=80
ChanceHardTurn=70
ChanceTurnDirection=90
ChanceContinue=45
ChanceCheckNewTarget=30
ChanceUseVertical=40
RollForGunAttack=2.5
RollForGunAttackRate=-0.15
MaxRollForGunAttack=70
CannonFireAngle=20.0
MaxCannonRange=400
OptimalCannonRange=80
MinCannonRange=20
CannonBurstLengthShort=6.0
CannonBurstLengthLong=10.0
 
[DogfightGreen]
SafeAltitude=20
DefensiveAngle=45
ChanceDefensiveTurn=80
ChanceBreakTurn=90
ChanceHardTurn=80
ChanceTurnDirection=85
ChanceContinue=55
ChanceCheckNewTarget=35
ChanceUseVertical=50
RollForGunAttack=3.0
RollForGunAttackRate=-0.2
MaxRollForGunAttack=90
CannonFireAngle=20.0
MaxCannonRange=300
OptimalCannonRange=70
MinCannonRange=20
CannonBurstLengthShort=5.0
CannonBurstLengthLong=8.0
 
[DogfightRegular]
SafeAltitude=60
DefensiveAngle=60
ChanceDefensiveTurn=90
ChanceBreakTurn=90
ChanceHardTurn=90
ChanceTurnDirection=70
ChanceContinue=65
ChanceCheckNewTarget=45
ChanceUseVertical=65
RollForGunAttack=3.5
RollForGunAttackRate=-0.2
MaxRollForGunAttack=120
CannonFireAngle=15.0
MaxCannonRange=200
OptimalCannonRange=50
MinCannonRange=15
CannonBurstLengthShort=4.0
CannonBurstLengthLong=6.0
 
[DogfightVeteran]
SafeAltitude=100
DefensiveAngle=75
ChanceDefensiveTurn=100
ChanceBreakTurn=100
ChanceHardTurn=100
ChanceTurnDirection=55
ChanceContinue=75
ChanceCheckNewTarget=60
ChanceUseVertical=80
RollForGunAttack=4.0
RollForGunAttackRate=-0.25
MaxRollForGunAttack=150
CannonFireAngle=10.0
MaxCannonRange=150
OptimalCannonRange=30
MinCannonRange=10
CannonBurstLengthShort=3.0
CannonBurstLengthLong=5.0
 
[DogfightAce]
SafeAltitude=80
DefensiveAngle=90
ChanceDefensiveTurn=100
ChanceBreakTurn=100
ChanceHardTurn=100
ChanceTurnDirection=40
ChanceContinue=90
ChanceCheckNewTarget=70
ChanceUseVertical=90
RollForGunAttack=5.0
RollForGunAttackRate=-0.25
MaxRollForGunAttack=180
CannonFireAngle=10.0
MaxCannonRange=100
OptimalCannonRange=20
MinCannonRange=5
CannonBurstLengthShort=2.0
CannonBurstLengthLong=4.0
  • Like 1

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Similar Content

    • By VonS
      Recently I had an opportunity to re-install FlightGear (stable ver. 2020.3.19) on my 2013 Mac Pro and was pleasantly surprised - with no crashes, decent FPS for the most part, and various other improvements that were incorporated over the years. I had dabbled briefly in FlightGear a couple of times before (around 2013, and again around 2020) - but eventually removed the flight simulator from my hard drive on both of those occasions.
      Since my particular setup(s) are with dual AMD video cards, I thought I would post some (illustrative) tips below, as well as representative pics of the sim when at best, or very good, settings. I will this time around keep FlightGear on my hard drive - makes for a good, free flight simulator.
      For command line entries and tips that should be plugged into the "Additional Settings" option of FlightGear, see the info. immediately below. Make sure to remove the info. included in brackets if copying the commands into the relevant settings section.
      -----
      --prop:/sim/gui/current-style=0 (better GUI style for AMD cards)
      --prop:input/mice/mouse/mode/button[2]/binding/value=2 (gets rid of horrible mouse-as-yoke feature that sometimes automatically turns on)
      --prop:/sim/rendering/photoscenery/enabled=true (gives more photo-realistic scenery whenever possible; to be fully implemented in later FG versions)
      --prop:/sim/rendering/hdr/envmap/update-continuously=false (disables continuous siphoning and updating of terrains, to enable seconds command below)
      --prop:/sim/rendering/hdr/envmap/update-rate=5 (terrain refresh rate set in seconds; good values are 5 or 10, for minimal stuttering)
      --prop:/sim/rendering/max-paged-lod=1300 (stock max paged LOD no. is "200"; recommended is not to exceed about "1900" on AMD)
      --prop:/sim/rendering/plod-minimum-expiry-time-secs=60 (stock min expiry time is "180" secs.)
      --prop:/sim/rendering/multithreading-mode=DrawThreadPerContext (should improve multithreading capacity, at least slightly, and add a few extra FPS)
      --prop:/sim/rendering/database-pager/threads=12 (stock thread rendering no. is "4," with more threads improving ave. FPS; set as per your CPU threads no.)
      --prop:/sim/rendering/vsync-enable=false (force disables vsync to improve FPS)
      --prop:/sim/rendering/multi-sample-buffers=true (this is anti-aliasing; set to false to disable and remove entry immediately below it)
      --prop:/sim/rendering/multi-samples=2 (anti-aliasing value; lowest is 2, also good is 4; avoid 3)
      --compositor=Compositor/HDR/hdr (modern rendering pipeline for better graphics details and shaders; rarely breaks scenery; still mostly experimental on AMD vid. cards)
      --units-meters
      --disable-splash-screen
      --disable-horizon-effect
      --enable-distance-attenuation
      --enable-specular-highlight
      --enable-clouds3d
      --fog-nicest ("nicest" improves look of fog with more subtleties while "fastest" makes fog appear and disappear more quickly; no real impact on AMD FPS)
      --shading-flat ("smooth" apparently improves look or depth of shaders; preference is for "flat" since I have not noticed difference in quality on AMD)
      --texture-filtering=2 (this is anisotropic value; lowest is 1; also good is 4 or 2; avoid 3)
      --bpp=32 (can also use 24 but have not noticed an FPS improvement with 24; would say 32 is better for FPS overall)
      --terrain-engine=pagedLOD
      --lod-levels=3 2 5 3 1 (also good is 4 3 6 4 2 for best visual quality on AMD but still mostly FPS-friendly; recommended is 3 2 5 3 1 for balance between visual quality and solid FPS)
      --lod-res=2 (default is 1; also good is 3; avoid any other values besides 2)
      --lod-texturing=raster (better than "bluemarble" as far as I have been able to test, with smoother loading of textures)
      --lod-range-mult=3 (default is 2; also good is 1; avoid all other values besides 3)
      --enable-texture-cache
      -----
      On a broader note, it's important to tweak the rendering, shaders, and LOD range settings in FlightGear - to get the best experience on your AMD video rigs (be it a dual or single AMD setup).
      The frequencies, by the way, on my dual FireProD700s have been OC-ed, via MSI Afterburner, from a clock/memory of 850/1370 to 1024/1380 MHz.
      Be sensible with the rendering options - particularly with the maximum number of scenery and aircraft tiles - anything above a value of 1900 or so is both useless and an FPS hit. (The scenery/aircraft tiles no. may also be set via the relevant command line entry indicated above in this post.)
      Take note as well of cloud density and visibility values. Anything beyond a visibility of 35 km or so is questionable since it does not widen further the cloud carpet but is, once again, an FPS hit.

      Pic 1 - Sensible/Best Rendering Choices for AMD Video (in FlightGear)
      Next we look at the shader options that also require careful thinking and tweaking. Take note that "landmass," "urban," also "water" - give different visuals if they are set to the maximum level (of five). I personally prefer how the landscape/terrains look with those three settings at a value of four - with crisper graphics - but tweak according to taste. Also worth noting is that I always run the LOD value on my rig(s) at "-1," via the excellent little program "ATISetLod" that is available under the top post of this thread.

      Pic 2 - Sensible/Best Shader Options for AMD Video (for crisp and fairly realistic graphics)
      Also important is to tweak the LOD range settings to get a good balance between visual quality and decent FPS, with no stuttering or crashes. Focus in particular on the maximum distances for the detailed, rough and bare scenery ranges - I decided eventually on cutoffs of 3, 17.5 and nearly 45 km. Other cutoffs worth considering are 2.5, 15, and 40 (or so) km, as well as 3.5, 20, and 50 km. Anything beyond the latter values will, again, most likely not make much of a difference with the visuals but will contribute a noticeable FPS hit. Take note also of the "high detail" and "AI/MP interior" values. I'm getting good results with values of 250 - 260 pixels for those options; also good are values of 300 (or so) pixels. Those two values may be tweaked to taste, for the most part - but, again, be sensible since they may impact on FPS.

      Pic 3 - Sensible/Best LOD Range Values for AMD Video
      Last, let's not forget to tweak our custom FlightGear profile (in our AMD settings panel) as best we can, to minimize stuttering, pausing, or crashes. Take note in particular how the shader cache is set to "On," not to "AMD Optimized," and how, even though I have enabled CrossFire mode, frame pacing is set to "Off" (which eradicates stutters on my rig in FlightGear - had tried with frame pacing "On" and was not pleased with the results). If you have only a single AMD video card, experiment with leaving the CrossFire option on (set to "AFR compatible"), or turning it off entirely. (NOTE: I have not noticed any FPS downgrade with anti-aliasing set to 4EQ and edge-detect, anisotropic value at 8, and tessellation at 16. Tweak to taste obviously.)

      Pic 4 - Good AMD Settings for a Custom FlightGear Profile
      Recommended also, if having any instability with FlightGear on AMD video, is to stick with the "Pro" variant DLLs (drivers) for AMD, instead of the consumer/Adrenalin ones. I particularly like the ver. 19.x.x Pro series of drivers, as well as the venerable ver. 17.x.x ones (the latter of which I have installed on the 2013 Mac Pro). To find links to the Pro 19.Q3 DLLs (the last version of the Pro DLLs to support CrossFire, by the way), see this page.
      Below follow several representative pics with the settings illustrated above applied - FlightGear provides a nice selection of aircraft with usually very good monoplane FMs, especially for tricycle-gear (small) civilian aircraft such as Cessnas, also for airliners like the (classic) Boeing 707. For multi-wing contraptions on the other hand and more realistic/historical FMs for biplanes, triplanes, and the odd multiplane, the gold standard(s) remain heavily-modded First Eagles 2 and RoF, as well as recent iterations (UE edition onwards) of the excellent WoFF series.






      Happy flying,
      Von S 
    • By VonS
      Hi All,
      Have been doing a bit of tinkering lately with my FlightGear install (see this thread for more info.) - and results are pleasing enough in what is a free, open-source flight sim. Managed eventually to get accurate photo-scenery working in FG, which is a noticeable improvement over the stock (dated) scenery that has been available in that sim for years on end (see pic. below for photo-scenery at work in FG, in this case in the Mesopotamian/s. Iraqi area; the Alb. D.I is a re-skin and FM-overhaul of a model done by Lester Boffo).

      On a hunch, I then converted a bunch of the photo-tiles from FG (stock format is DDS) to BMP format - so that they become visible in the FE2 terrains folder - and it does indeed work - but would require careful, aesthetic choices regarding what tiles to replace in the terrains folder, such as farms, cities, deserts, etc. - for best cohesion and not too many jarring disconnects between tiles.
      Nice to see that stock trees and buildings are populating over the satellite/ortho-terrains too, in FE2. Photo-scenery is simply an improved "satellite" carpet that covers the stock scenery and terrain framework in FlightGear - and by extension works the same way in the ThirdWire sims but requires manual conversion to BMP format and manual placement in the terrains folder (perhaps the DDS format would work too, but I didn't bother tinkering with the various settings/text files this time around in the terrains folder).
      The other good thing about FlightGear is that it is fully under the GNU GPL (General Public License; i.e., "copy-left" license) - so there are no copyright problems that I know of if one borrows such ortho-scenery for FE2 too (haven't bothered to test with SF2). And if working on such a terrain swap as a personal project - that is of course even simpler.
      A few representative pics. below with just a few ortho-scenery tiles placed over Stephen1918's upgrade of the Mesopotamian theater for FE2. Results look pretty good, particularly when flying higher up, and I haven't noticed any FPS drop with the photo-scenery tiles loaded. Some old tiles are still in place in the pics., for comparative purposes. Any empty areas are where I didn't bother to load tiles.




      Anyway - I hope you find this post interesting - it's perhaps the "cheapest" way of upgrading tiles across the various theaters available for FE2.
      If I find enough free time I might tinker further with this side-project; will post comments if the results prove particularly successful.
      Cheers & good flying,

    • By VonS

       
      Apr. 17, 2023 Update - Please find, available at the link below, my previously released mods. for RoF United Ed. - this time as a convenient, consolidated package that has been updated with the latest tweaks. Do read the included "Read Me" files, where available, for more info., particularly the "READ_ME_ROF_General_Info" file that is located in the root folder of the package when unzipped. This package is provided "as is," with no active support beyond the included explanatory files. The "Consolidated Mods. New" Package contains the latest AI mod. (ver. 1.4) for RoF United, as well as mods. previously offered separately, such as the "M-S Type H Campaigning Tweak," the "PWCG modulations" files, etc.
      Apr. 29, 2023 Update - Included below the link to the "Consolidated Mods. New" package is Ver. 3 of the PWCG "modulations" files (to be installed manually in the PWCG ver. 16.3.1; see directions included in the larger, consolidated package). Ver. 3 of the "modulations" files increases wind and turbulence settings, as well as provides greater chance of encountering more capable AI when flying PWCG campaigns, especially if used in conjunction with my newer AI mods. for RoF (such as ver. 1.36 or 1.4, included in the big consolidated package).
      Important Note: if using the Ver. 3 PWCG "modulations" files, instead of the Ver. 2 ones included in the big consolidated package for RoF -- make sure to change settings, if you choose to change any, from within the modulations, i.e., the RofCampaignSpecific, files themselves -- otherwise, tampering with relevant settings from within the PWCG advanced menu options will automatically delete extra entries for the AI that I've included in the Ver. 3 modulations. This note does not apply to the Ver. 2 modulations since those don't contain any extra AI settings.
      Feb. 6, 2024 Update - RoF AI mods. vers. 1.36 and 1.4 have been upgraded to vers. 1.37 and 1.41, respectively. The new versions have tweaked the AI's ability to follow commands more quickly and/or properly. Ver. 1.41 also contains further tweaks to safe alts. for the AI, to minimize unnecessary crashes and very low-level dogfighting. The entire consolidated package has been renamed from "Consolidated Mods New" to "Consolidated Mods New V.2" (i.e., Version 2).
       
      ROF_ConsolidatedModsNewV2_VonS.zip
      ForRoFpwcgModulationsVer3_VonS.zip
      Happy flying,
      Von S 
       
       
    • By VonS
      Hello Fellow Fliers of First Eagles 2,
      Version 11.0 of the FM/Realism Package for FE2 contains approximately 220 WWI aircraft types and sub-types, for your (virtual) flying enjoyment. For detailed installation instructions, please check over the main "Read Me First" file that is included, as well as several older (pdf) files and that cover aircraft performance info., tips, etc., across all previous versions of the FM update package.
      A great many thanks go out to Peter01 and Ojcar, also TexMurphy, for making most of those flight model files to begin with. Also a great thanks goes out to Stephen1918, MontyCZ and Laton for providing lots of beautiful planes and skins that go with those great flight models, to NBell for the many hitbox improvements provided for the planes, also to VonOben, Mike Dora and Crawford for many helpful suggestions, and to the A-Team by SkunkWorks for allowing me to tweak a few of the FMs for their models too. And of course a very big thanks goes out to Geezer for several fantastic, high-quality models that were produced for FE2.
      What I've done is tinkered with the data files further. Modifications in all cases, to a greater or lesser extent, include data under the following sections:

      (a) MissionData
      (b) FlightControl
      (c) AIData
      (d) Sound
      (e) AircraftData
      (f) Engine
      (g) Crew
      (h) Internal Guns
      (i) Control Surfaces
      (j) Landing Gears
      (k) Fuselage, Nose, Tail, Vertical Tail, LeftStab, RightStab, Rudder, Left/Right/Inner/Middle/Outer/Tip Wing components
       
      Enjoy and happy flying!,
      Von S
      NOTES: For several of Geezer's latest aircraft and my FMs for those aircraft (Nieuports, Pfalzes, etc.), which are not included in this package, please see the relevant post located towards the bottom of page 25 of the "New Aircraft" thread for FE2, on CombatAce. Also see relevant posts on pages 26, 27, and 28 of that same thread for Geezer's "early beta" collection that contains the Breguet XIV, Junkers J.I, and other updates, including performance info. for some of those aircraft types.
      DISCLAIMER: All Von S mods., for FE2, SF2, also WOFFue/pe/bh&h2, as well as for RoF (United Ed.), are subject to the CombatAce "freeware" terms of agreement. Mods. may be shared with others, included on other media devices, also modded further, providing that original documentation and/or credit is included, and providing that the mods. remain free to use. Von S mods. shall not be sold, resold, etc., and Von S takes no responsibility for injuries or fixations that may result from flying heavily tweaked FMs or from attempts to enjoy real flying without aid or instruction from a qualified flight instructor.
       
×

Important Information

By using this site, you agree to our Terms of Use, Privacy Policy, and We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue..