Cliffs of Dover

[paul_leonard]

Can you please also post your .mis file.  Then I can run with Tacview.

 

On 12/10/2025 at 6:20 AM, clod65 said:

Hi Paul. I have attached procmon logfile, of interest refer to timestamp 7.55.4 to 7.56.0  this is when the plane bf109 disengaged from the fight.

I didn't make a video recording at the sametime, but I did note down when it happened. I'll be checking hex values as well, example 217 seconds in hex is D9.

I'll keep at it.. Thanks for the reply

 

oh one file i missed from previous entry was core_update.sfs, you will see this at 7.55.4, the one that stands out though is bob_plane.sfs

Logfile1.zip 3.28 MB · 1 download

the procmon at least narrows it down on which files were accessed between 7.55 and 7.56

What if the timer code was already loaded in RAM?

[clod65]

 

Hi Paul,

mis file attached, the skill level is set to veteran and not ace.

Just an interesting fact, not sure if you have seen these on my channel or aware of it:-

ai spitfire crashes into the channel, bf109 continues to target the spitfire
the whole event from:- 11.35PM to 12.28AM, I only started recording from 11.45PM. once the bf109 ran out of ammo it stopped targeting.

https://www.youtube.com/watch?v=KKzzFjVAS_A&t=135s

 

And in reverse:-

 

ai bf109 crashes into the channel and same thing happens, the spitfire
continues to target the 109 i captured 14 minutes. im assuming like the previous video once the spitfire runs out of ammo it
would stop targeting. I did not continue the recording until the ammo ran out.

so now how do we get the same thing happening but in the air? 

1on1_acelvl.mis

 

On 12/10/2025 at 6:57 PM, paul_leonard said:

What if the timer code was already loaded in RAM?

good point

Edited December 11, 2025 by clod65

[Airborne2001]

I had no idea about what I've read from above, and it really is saddening. Of course, this issues at play here go all the way back to the 2000s...

[kraut1]

On 12/8/2025 at 3:25 PM, BENKOE said:

Maybe a helpfull tool for -Blitz-  VR-beta-testers ... 

The ZEISS Aerotopograph Stereoskop has been seen in some WorldWar II and post WorldWar II movies, such as "Triple Cross" and "Operation Crossbow", among others.

Thanks very much for your great support!
It will certainly use it for testing the new VR Beta test version that was released yesterday.
(released only for one very specific test)

Edited December 11, 2025 by kraut1

[Aapje]

CoD has a record discount now, with 80% off.

And I noticed that they re-added the VR tag on Steam a month ago, after removing it 8 months ago. A sign of greater confidence that VR is going to work out?

 

[kraut1]

7 hours ago, Aapje said:

CoD has a record discount now, with 80% off.

And I noticed that they re-added the VR tag on Steam a month ago, after removing it 8 months ago. A sign of greater confidence that VR is going to work out?

 

We are currently doing the test according to the defined procedure with the yesterday released VR Beta test version.
And we are receiving constantly from time to time new releases for tests and we post the results in the VR Beta Forum + discussion if required.
 

[Catch]

11 hours ago, Aapje said:

A sign of greater confidence that VR is going to work out?

Was there ever any doubt my friend. It's full steam ahead. 🙂

[kraut1]

I have flown my first CLOD VR test mission 1 or 2 weeks after release of Tobruk with VorpX (no stereoscopic view) in Summer 2020.

Alltogether ca. 30 combat missions with Blitz and tobruk.

Sometimes it was great but frequently centering of the view did not work or was too chaoticly, so I stopped using it because it was too unpredictable if it worked or not.

But at least sometimes I had some fun:

 

Edited December 12, 2025 by kraut1

[clod65]

Hi Paul,

I have recorded video of flight and ran procmon at the sametime, I have uploaded both files here:-

https://drive.google.com/drive/folders/13K7kaOfTB__a7NRbSSUekEcZ9JMdAqAx?usp=sharing

In procmon log file search for 'Tally_ho.ogg', (this is when you hear it in the video) not long after that the ai pilot kicks in.

These are the files accessed often:-

-core_bank01.fsb

-bob_bank01

Inbetween the above two files was one:-core_update02.sfs, then these:- core_update01.sfs, bob_buildings.sfs, bob_plane.sfs, bob_update05.sfs

The AI was set to ACE skill level and Engage Fighters. The combat time was 5 minutes, i.e 300 seconds.

Anyhoo this is where I'll start searching the files mentioned above and testing.

 

quick update, I have done the following 5 files:-

-core_bank01.fsb

-core_update02.sfs

-bob_bank01.fsb

-core_update01.sfs

-bob_plane.sfs

with hex editor i have changed any 300 found to 600.

and any hex 12c which is 300 to hex 258 = 600

I had 79 hits of decimal 300 in bob_plane.sfs, and surprisingly after changing all them to 600 the game still ran.

There has been no change it still remains 5 minutes of combat.

I will continue on with the rest of the files, I also have not reverted the changes. the decimal and hex changes are being done at the same time.

 

 

AI Overview

 

 

 

Actual in-air combat time for a single engagement during World War II was remarkably brief, often lasting only 

seconds to a few minutes. A pilot's entire operational tour might involve a total of only a few hundred "combat hours," many of which were spent in transit or patrolling. 

 

Factors Limiting Combat Duration

Several factors contributed to the brevity of actual fighting time:

• Ammunition Limits: WWII fighter planes carried limited ammunition. For example, a British Spitfire had only enough for about 15-20 seconds of sustained fire, while an American P-51 Mustang had roughly 20-25 seconds. Pilots were trained to fire in short bursts to conserve ammo and prevent barrel overheating, meaning a single "dogfight" typically consisted of only a few passes before a pilot needed to break off or run out of bullets.
• Fuel Capacity: Combat maneuvering and high-speed flying were very fuel-intensive. While aircraft had a total flight time of a few hours (e.g., a B-17 bomber mission could last 6 to 8 hours), the actual time spent engaging an enemy at high power settings was a fraction of that.
• Tactics: Many effective tactics involved "hit and run" passes, where fighters would make one or two quick attacks on bomber formations before disengaging and re-forming for another pass, rather than engaging in prolonged, circling dogfights like those sometimes depicted in movies.
• Operational Definitions: For administrative purposes, a pilot might log "combat hours" for the entire duration of a mission that entered enemy airspace, even if they never fired their weapons or even saw an enemy aircraft. 

 

Total Combat Exposure

Pilots rarely stayed in constant combat. They were often rotated through tours of duty: 

• Fighter Pilots: Tours were often measured in flight hours, such as a maximum of 200 hours in one tour for RAF Fighter Command.
• Bomber Crews: Tours were generally measured in number of missions, typically around 30 operational flights. 

Many, if not most, troops were in service and supply roles and did not see much combat. For those in frontline aircrew roles, the experience was intense, but the actual time spent actively shooting at an enemy during any single engagement was incredibly short. 

[AndyJWest]

Why do people insist on posting next-word-guesser-bot drivel on forums? 

[paul_leonard]

On 12/14/2025 at 5:24 AM, AndyJWest said:

Why do people insist on posting next-word-guesser-bot drivel on forums? 

I don't think clod65 is wrong to point out that while it is frustrating that there is clearly a timer in-game, the actual experience was that air combat was a short fast knife fight, and not protracted dogfight furballs that we all seemingly prefer, myself included, in the non-lethal virtual world.  That he quoted from an AI chat in this case doesn't make it wrong.  He did fail to point out that in the real world discretion was the better part of valour, and that attacks generally only occurred when one side had a significant advantage in the initial attack, and otherwise they just passed on an attack.  We don't like cowardice/discretion in our non-lethal game world either.  Is what it is.

 

On 12/14/2025 at 4:50 AM, clod65 said:

AI Overview

 

 

 

Actual in-air combat time for a single engagement during World War II was remarkably brief, often lasting only 

seconds to a few minutes. A pilot's entire operational tour might involve a total of only a few hundred "combat hours," many of which were spent in transit or patrolling. 

 

Factors Limiting Combat Duration

Several factors contributed to the brevity of actual fighting time:

• Ammunition Limits: WWII fighter planes carried limited ammunition. For example, a British Spitfire had only enough for about 15-20 seconds of sustained fire, while an American P-51 Mustang had roughly 20-25 seconds. Pilots were trained to fire in short bursts to conserve ammo and prevent barrel overheating, meaning a single "dogfight" typically consisted of only a few passes before a pilot needed to break off or run out of bullets.
• Fuel Capacity: Combat maneuvering and high-speed flying were very fuel-intensive. While aircraft had a total flight time of a few hours (e.g., a B-17 bomber mission could last 6 to 8 hours), the actual time spent engaging an enemy at high power settings was a fraction of that.
• Tactics: Many effective tactics involved "hit and run" passes, where fighters would make one or two quick attacks on bomber formations before disengaging and re-forming for another pass, rather than engaging in prolonged, circling dogfights like those sometimes depicted in movies.
• Operational Definitions: For administrative purposes, a pilot might log "combat hours" for the entire duration of a mission that entered enemy airspace, even if they never fired their weapons or even saw an enemy aircraft. 

 

Total Combat Exposure

Pilots rarely stayed in constant combat. They were often rotated through tours of duty: 

• Fighter Pilots: Tours were often measured in flight hours, such as a maximum of 200 hours in one tour for RAF Fighter Command.
• Bomber Crews: Tours were generally measured in number of missions, typically around 30 operational flights. 

Many, if not most, troops were in service and supply roles and did not see much combat. For those in frontline aircrew roles, the experience was intense, but the actual time spent actively shooting at an enemy during any single engagement was incredibly short. 

True, but imagine playing this, or any other air combat sim for hundreds of hours only to get a few minutes of air combat.  Flying around in a fighter looking at the scenery.... oh wait, that's Microsoft Flight Simulator in a warbird.  I do enjoy that though, when I'm not looking for a dogfight.

But many players, myself included, want an air combat to play.  Most don't have the mission builder knowledge to create complex large scale air combats, nor are there many large scale missions available,  and so are relegated to playing simple small scale missions.  So such small scale missions quickly exposed the weaknesses and flaws in the AI, most frustratingly to many the seemingly unrealistic behaviour where the AI combatant would break off after a clearly fixed period of time, even when they were at the advantage.  To add insult to injury, once the AI disengaged, it would fly home, generally at the low and slow landing waypoint setting, and refuse to re-engage if the player went after them... the so-called drone.  You know all of this.

I have done my own research and concluded it is likely a timer that is triggered, runs and then sends the AI aircraft home after a fixed period of time.  I have also noticed that the landing waypoint becomes the next waypoint (skipping any other intermediary waypoints) once combat is commenced.  I've watch the AI cockpit to determine fuel remaining and ammo remaining, they do not seem to have any relation to the droning commencement.  

So I laud your work to try and figure out if you can brute force a change.  It is a real frustration to many in the community and if you are successful a real breakthrough.

Paul

 

 

clod65:

I reviewed your files noted above. May I ask what version of the sim are you running?

[clod65]

Paul

The version I'm running is:    GAME VERSION 1.11.20362  (October 18, 2012)

I too would like combat to last longer, there's two sessions now that i was in combat for 15 minutes, but when i run the game again

and making no changes, it's back to 5 minutes. I've plucked at this doing this and that and, even put this in GPT for shits and giggles:-

  

i want to change the combat time of 5 minutes to longer in the original il-2 sturmovik cliffs of dover

Thanks for clarifying original IL-2 Sturmovik: Cliffs of Dover (not Blitz) — that matters.

I need to correct one important point first so you don’t waste time:

⚠️ In ORIGINAL Cliffs of Dover, the combat time is NOT meant to be edited with a hex editor

In stock (original) CloD:

Combat / mission duration is defined in the mission file (.mis)

.mis files are plain text

The .sfs files are packed resources, not where mission timers are normally controlled

So if you hex-edit .sfs, changes often:

Don’t work

Reset

Or break the mission

✅ CORRECT way (original CloD)
1️⃣ Locate the mission file

Path (typical):

IL-2 Sturmovik Cliffs of Dover\Missions\

Look for:

.mis

2️⃣ Open the .mis file with Notepad++

(Do NOT use a hex editor for this step)

3️⃣ Find the combat time limit

Search for keywords like:

Battle
Combat
Timeout
TimeLimit
End

Common patterns in original CloD missions include lines similar to:

Battle 300

or

TimeLimit 300

or

Timeout 300

300 = seconds = 5 minutes

4️⃣ Change it to a longer time

Examples:

15 minutes → 900

20 minutes → 1200

30 minutes → 1800

Example edit:

TimeLimit 300

⬇️

TimeLimit 900

Save the file and test.

❌ When hex editing would be needed

Only if:

The mission is packed

Or you are editing a compiled campaign resource

Or the .mis is embedded in an .sfs

In that rare case, the value is stored as a float:

5 minutes (300 seconds) float
00 00 74 43

15 minutes (900 seconds) float
00 80 60 44

But again — this is NOT the normal or recommended method in original CloD.

✅ Summary (important)

✔ Original CloD → edit .mis (text), not hex
❌ Hex editing .sfs is usually wrong for mission timers
✔ Timer value = seconds

If you want, tell me:

The mission name

Whether it’s single mission or campaign

Whether the .mis is visible or packed

I’ll tell you exactly which line controls the 5-minute combat limit in your mission.

 

END

I will get back to you, I'm still on this, so might be quiet for little while.

 

Try the following if you may, which I have also done, belly land the plane you are in, jump into another plane and observe the opponent ai plane

I suspect that one of the sessions i did one of the ai planes in my team crash landed, the plane did not explode, and the opponent ai plane

kept in combat loop. 

 

I have also done this in the .mis file as per chat gpt:-

Battle 900
Combat 900
Timeout 900
TimeLimit 900

 

results:-

planes disappeared from the map putting the above values, no change still 5 minutes
you can still play the mission though.
[BoB_RAF_F_FatCat_Early.01_Way]
[BoB_LW_JG54_III.02_Way]

[CustomChiefs]
planes are in the map, no change still 5 minutes

[Stationary]
planes are in the map, no change still 5 minutes

Try this again but with more planes in the air. make sure not to get mission end message
have tried each line at a time from the above, no planes are seen in the map in fmb.
Also put just a 1 and no planes seen in map. so would seem anything you put under here, the planes will not be seen in fmb.
ok still no change still 5 minutes
[BoB_RAF_F_FatCat_Early.01_Way]
[BoB_LW_JG54_III.02_Way]

put the above 4 lines above [PARTS]
no change. still 5 minutes.

[BuildingsLinks]
no change. still 5 minutes.

[Buildings]
no change. still 5 minutes.

[BuildingsLinks]
no change. still 5 minutes.

[splines]
no change. still 5 minutes.

[AirGroups]
errors

[MAIN]
no change. still 5 minutes.

[GlobalWind_0]
no change. still 5 minutes.

 

 

I have also tried this in numerous files:-

5 minutes (300 seconds) float
00 00 74 43

changed them to:-

15 minutes (900 seconds) float
00 80 60 44

 

 

there's ton on info above apologies, this is what stuck out for me:-

Try the following if you may, which I have also done, belly land the plane you are in, jump into another plane and observe the opponent ai plane

I suspect that one of the sessions i did one of the ai planes in my team crash landed, the plane did not explode, and the opponent ai plane

kept in combat loop. 

 

Has anyone tried to contact Oleg about this?  I tried to contact him here:-

https://www.youtube.com/@OlegMaddox

 

I know this may not be him.

If we could create a fund page just so that we can fix this , wouldn't that be a good idea

let him know that we love this sim

Edited December 16, 2025 by clod65

[AndyJWest]

Oleg Maddox hasn't been involved in air combat sim development for many years.

[paul_leonard]

Huh.  I have original version, and both Steam versions.  I was unaware that original .mis files had a timer in them.  I am very familiar with editing .mis files with Notepad++.

over the holidays I will take a look at your informative posts above.

 

[Hiuuz]

8 hours ago, clod65 said:

Has anyone tried to contact Oleg about this?  I tried to contact him here:-

https://www.youtube.com/@OlegMaddox

 

I know this may not be him.

If we could create a fund page just so that we can fix this , wouldn't that be a good idea

let him know that we love this sim

Or try here
https://www.facebook.com/oleg.maddox

[clod65]

Thank you Hiuuz, I I have sent message.

[paul_leonard]

clod65:

I think your advice from ChatGPT is a little bit wonky.  I've now looked a dozens of old mission files from the 2011-2012 vintage and none have any kind of line references to your 4 keywords.  I also looked at older il-2 1946 mission files and nothing there.  The generative AI engines have a real habit of confusing things and mixing things up (usually when I know better and challenge them, they respond with something bright and bubbly like "oh, you are quite right".... and then happily move on to try and feed me something else).  While I have been able to do a number of neat things using the latest round of so-called AI (let's keep in mind it is a marketing term), I suspect this particular thread will lead you nowhere.

[Mysticpuma]

Oleg won't be able to help due to contractual reasons. Buzzsaw can elaborate more if required but Oleg is a dead end.

[clod65]

Paul:

i know, at this point i will try anything for new ideas, even if they are dead ends

Mysticpuma:

Thanks, it was worth a try, even if it is the old clod from 2012?

Edited December 17, 2025 by LukeFF
Language

[paul_leonard]

Don’t despair.  This is a big frustration for many and you are trying something.  If it works you are a hero.  Keep testing please.  I will support.

Paul

[clod65]

Thanks for the encouragement

 

chatgpt may not be all that wrong i suspect the file in question is in dll, mainly partBob.dll, that's the one ill be looking at.

 

so this is what i'e done so far

First i started line at a time, from the beginning and found that the game wouldn't start
i bypassed these lines until i found where the game would run then started disabling code 10, 20, 100 then a thousand
lines at a time.

These are all filled with 90 which apparently disables code, then I restored the original partBob.dll and filled 90s from 250 to a000 (a000) is 10000 lines
the game runs and I test using the ] bracket at maximum speed, the plane disengages about 20 second mark which equates to 5 minutes.
so far I have gone 1/4 of the way, I'm sure i'll hit a snag further down the line and will have to do few lines at a time.  

partBob.dll

[Hiuuz]

I wonder about you get any response from Oleg.

[AndyJWest]

51 minutes ago, Hiuuz said:

I wonder about you get any response from Oleg.

What sort of response would you expect from someone who has had nothing to do with CloD (or any other air combat sim) for well over a decade?

[clod65]

i will let you know of course if i do get one..

 

I'm changing tactic, Paul pointed out few messages ago about timer residing in memory, so i'm using cheat engine to scan for opcodes in memory

type this in chatgpt:-  cheat sheet of common timer-related opcodes in x86/x64 binaries.

 

Notes:- 

* `B8 / B9 / 68` → Load timer constants
* `DEC / SUB / CMP / JNZ / JMP` → Loop-based timers
* `CALL` → OS or library delays (`Sleep`, `usleep`)
* `XOR / MOV 0` → Reset counters
* Hex values for timer constants are **little-endian**, e.g., `300` seconds → `2C 01 00 00`.
 

[AndyJWest]

'Timer-related' opcodes? Yeah, sure. Unfortunately, 'more-or-less-everything-else related' too. Ask ChatGPT 'which are the most common opcodes in x86/x64 binaries?', and you'll get a very similar result.