This is a follow-up to the 'Dud Engines' thread in the general section.

It appears to be the case that most aircraft engine failures in WOFF are determined by the generic simulation.xml set of instructions (thank you, Von S, for pointing me in that direction), but some specific aircraft have their own identical (but with different values) set of instructions that override this. If this is the case It should be possible, but would be a lot of work for one person, to give every aircraft in WOFF its own set of engine failure instructions (rather than just a select few) rather than having to rely, as now, on the generic set in the simulation.xml file. It would not be quite so bad, as it is the engine that fails rather than the aircraft, and many aircraft use the same engine and would therefore share the same instructions.

Another thought - if the figure for 'hours' in these instructions represent the mean time for failure within the sim, and a figure of '60' for example represents a 1/60 chance of failure in a 1 hour flight (?) then it might be possible to use the Mean Time Before Overhaul found in most aircraft engine manuals of the period to stand in for a Mean Time Before Failure number that could be used as a starting point for the numbers used by the sim (?). Some of the data for this is in TheAerodrome forum thread that I linked to in the 'Dud Engines' thread, but some would have to be estimated from anecdotal evidence, and the outcomes would have to be tested to see how this would work in sim.

Does anyone think that this might work? If several people worked together it might cut down on the time and effort to find and add this data to the aircraft files, and test the results...

No, I don't think that would work - I can't see anything that I could change to make this happen. Scratch that idea

Not a bad idea but sounds like a lot of work. I don't think every aircraft type needs its own individual engine failure model, however, as the failure rules in the simulation.xml file does a pretty good job on its own and can even be tweaked to suit individual preferences.

I guess the first step would be to identify all the aircraft types in WOFF with customized engine failure rates (easy) and then compare that with a list of "unreliable" engine types used in the war to see if any aircraft using the default failure rules deserve custom failure rules. I bet you won't find many in WOFF, but if you do, they should be easy to fix.

OTOH, if you wanted to go hog wild, you could fashion individual failure rules based on other factors, such as squadron rating, with "poor" squadrons having higher failure rates than "good" or "elite" squadrons, etc. Now that might be a project!

I don't think it would be too difficult, because I think all you need to do is edit the "failure" section in an individual aircraft's .xdp file. However, copying that to each aircraft folder might be tedious.

Hallo @Bletchley,

"... to give every aircraft in WOFF its own set of engine failure instructions ..."
That could be a very good mod!
But maybe it is not necessary to do it for every aircraft. It is enough in my opinion to do it for every different engine (f.e. 5 different aircrafts with the same engine need only once the same new engine failure instructions).
So in my eyes it is senseful, like a prework, to check, how many different engines do we have in WOFF and which historical failure-data from these engines are available. Then you have a better overview about the real work and you can make your decision easier.

If it is too much work, maybe it is easier to change the simulation.xml for "similar engines /engine-failures" in different groups and in consequence in different mods. So you can activate the the mod1 "engine failure group 1" (for all corresponding aeroplanes) or activate mod 2 "engine failure group 2" (for all corresponding aeroplanes) and so on. On this way you can take into account the much/big different features of real different engines in different groups, but not the little differences between engine-type A1a and type A1b.

So you activate the "similar engines/engine-failures-group"-mod, which contains the aeroplane you fly at the moment.


Only a thought!


Greetings!

Glad that my suggestions regarding the simulation.xml file were helpful Bletchley -- I am, overall, getting good results with those values, from WoFF UE/PE, in BH2. Also a thumbs up to the other posters in this thread -- especially to Becker's idea (that might be a sensible middle ground, to tweak as per engine type, rather than to comb over every file for every aircraft type in WoFF).

In conjunction with changes to the simulation.xml file, it should be sufficient for good, broad representation of historical engine failure/unreliability rates -- for example, the early Hisso 200 hp, the 220 hp Hisso, etc., also early inline-six Mercedes engines vs. later (au) variants, and of course various rotary types, such as the Oberursels, Clergets, LeRhones, Siemens-Halske, and so on.

I hope, time-permitting, to test some more "hardcore" engine failure nos. than present even in stock WoFF UE/PE (for the simulation.xml file), in the near future - and will post under the casual campaigns thread if I notice any results worth reporting.

Cheers all,
Von S

As I say this is basically already in there for the more obvious less reliable engines, such as the 200 Hisso etc in early SE5 and SE5a, SPADS, and the Fokker Eindecker IV for example with the doubled up Oberursel EIII 14 clyinder

Whether it really needs subtle differences for many of the other engines not sure it's worth it?

Note at some point, and I cannot remember quite when, WM updated the random number calculator code, so numbers from UE and B&HII may not produce the same effect.

Hi Polovski, thank you, yes it would probably be a lot of work for only marginal gain to add individualised engine failure rules to every aircraft in WOFF (in many cases, as you say, with little noticeable difference), particularly as it would need a fair bit of testing to make sure it works without breaking anything else.

Hi Buckeye Bob, Becker01, Von S - I quite like the alternative proposed by Becker01 and by Von S, producing a different simulation.xml for a number of broad categories based on a simple hierarchy of reliability. I guess this would need to be (at least initially) a JSGME based mod so that, as Becker01 says, they can be swapped in and out as players change from one aircraft to another. That would be much easier to do, less likely to break anything else (as it is just tweaking the numbers in an existing file) and easier to test and re-tweak, but would end up being more fiddly for the player. We would just need to get our heads together and come up with a reliability hierarchy of engines, based on whatever historical evidence that we can find, and divide that hierarchy up into broadly similar groups that can then be matched against the aircraft in WOFF to create groups of aircraft that can then be allocated to a particular tweaked simulation.xml file. After testing, the numbers could be tweaked further until we find a reasonable balance between the historical evidence and playability. If that worked, at a later stage (and if someone had the time and inclination) then as Buckeye Bob suggests, the relevant engine failure lines from those tweaked simulation.xml files could be added to those individual aircraft files that don't already have them to create a follow up mod that would dispense with all the JSGME swapping in and out

Simulation.xml contains a lot of other configurations (eg workshop settings, ai behavior, etc) that vary from one player to another. I suggest trying to get in touch with JJJ to add it to his multimod which only changes the numbers in existing files without completely replacing them like jsgme would do.

OK, thanks orbyxP, that is useful to know.

I would also second Orbyx's recommendation to contact JJJ - such a fine-tuned (and excellent) engine mod would work best as a set of toggles included in JJJ's fine program "MultiMod" - for example, there could be three or four different settings implemented, and that change the relevant failure-rates section of the simulation.xml file, without altering anything else -- something as follows, perhaps:

option 1 - poor reliability - early rotaries, early Hisso and Mercedes engines (such as for Eindeckers, the early Albies and the Se5 and 200 hp Spad 13)

option 2 - usually acceptable reliability - mid-war rotaries and somewhat more reliable types (such as for Tripehounds, Dr.1s, the Hisso-powered Se5a, the Alb. D.III series, and so on)

option 3 - fairly good reliability - later-war types and generally more sound engines (such as for the Viper Se5a, 235hp Spad 13, the Alb. D.V/Va series, etc.)

option 4 - worst reliability - run-down squadrons, home-defence flights, obsolete types still in use, depending on player's own aircraft and time of war, etc. (and suffering from wear and tear)

Categories above are crude but workable suggestions -- it might be possible also to organize such options along engine types and time of war, such as --- early rotaries, mid-war rotaries, late-war rotaries, early inlines, mid-war inlines, late-war inlines, etc. (a min. of six options to toggle in that case) -- although one should keep in mind always the law of diminishing returns -- i.e., might not be worth it to fine-tune such options excessively, since much is randomized via the AI anyway in WoFF -- thus my initial (cruder) suggestions for three or four options above.

Hopefully JJJ will see these posts and pitch in with his own thoughts. Looking forward to further developments in this area gents'.

Cheers all,
Von S

Thanks Von S, here is my first stab at that (mostly off the top off my head, and I haven't checked to see if the engine/aircraft type exactly matches that used in WOFF):

Type 1 : Generally very problematic : Oberursel U.III (Fokker D III; Fokker E IV), Clerget 9Z made under license in UK (Sopwith Strutter RNAS), HS 8b geared (SE5a; Spad XIII)

Type 2 : Bit iffy, many of the early rotaries : Oberursel U.0 (Fokker E I; Pfalz A I), Oberursel U II (Fokker D II; Fokker E II; Fokker E III; Pfalz E III), Oberursel UR II (Fokker Dr I; Fokker D VI; Fokker E V); Gnome Mono (DH2)

Type 3 : Generally OK, early-mid rotaries with cast iron pistons : Le Rhone 9C (Bristol Scout; Caudron G4; Nieuport 10c1; Morane Saulnier L; Nieuport 11 Bebe; Sopwith Pup), Le Rhone 9J (Nieuport 16, Nieuport 17; DH 5; Sopwith Strutter RFC), Clerget 9B (Nieuport 17 bis; Sopwith Camel early RFC; Sopwith Triplane), Clerget 9Z French built (Nieuport 12, Sopwith Strutter), Gnome Mono 9N (Nieuport 28C-1 USA)

Type 4 : Improved rotaries with aluminium alloy pistons : Le Rhone 9Ja (Nieuport 17), Le Rhone 9Jb (Nieuport 23, Nieuport 24; Nieuport 26; Nieuport 27; Sopwith Camel), Clerget 9Bf (Sopwith Camel later), Bentley BR1 (Sopwith Camel RNAS), Bentley BR2 (Sopwith Snipe)

Type 5 : Mostly early-mid war stationary : Beardmore (FE2b), RAF1a (BE2c), RAF4a (BE12; RE8), Mercedes D I (Aviatik B I), Mercedes D II (Aviatik B II; Halberstadt D II), Argus AS II (Halberstadt D III), HS 8a (SE5; Spad VII), Mercedes D III early (Aviatik C I; Roland C II; Albatros D I, Albatros D II)

Type 6 : Mostly later war stationaries : Benz Bz IV (DFW C V), Mercedes D III late (Hannover C LIII; Albatros D III; Albatros D V; Albatros D Va; Pfalz D IIIa; Fokker D VII), Viper (SE5a); BMW IIIa (Fokker D VII F); Renault 300 hp (Breguet 14A2), Mercedes D IVa (Gotha G IV; Rumpler C IV), Rolls Royce Falcon (Bristol F2B), Rolls Royce Eagle (DH4)

Any comments from anyone? Quite happy to split or merge categories, or swap engines around from one to another. Some aircraft will be in more than one category, as they had several engines (Sopwith Camel springs to mind there!) although I have tried to go for the main production engine rather than variants.

I think we already have two baselines - the early engine failure rules (Type 2?) and the current engine failure rules (Type 6?). Type 1 is probably covered by the engine failure rules already placed in some aircraft files. Type 3 to 5 would be on a gradient between Type 2 and 6. When we have an initial set of engine failure rules between 2 and 6 we could then test them by moving them in and out of the simulation file, and tweak as appropriate.

That is just one more, I think, than you suggested Von S., but you might want to merge a couple of those categories. It could also be left to the player to upgrade or downgrade reliability for brief periods (i.e for wear and tear, or because an engine improved as early snags were ironed out)

Thank you for those categories Bletchley -- I hope to do some testing over the next week or two and will comment regarding results. I will initially create six separate failure entries as per your six types listed, but will then merge one or two -- to streamline testing - and will comment if I notice any interesting differences between the different simulation.xml files. My assumption is that stock failure rates in BH&H2, also in UE/PE WoFF, are somewhere around types 4, 5 or 6 - with UE/PE closer to type 4, and BH&H2 closer to six. Types 1, 2 and 3 might not currently be represented in-sim, with the exception, as you say, of individual aircraft xdp files that may cover some of the extreme unreliability of type 1. I will post here again once I create some failure sets since I have to be careful not to plug in nos. that might crash WoFF during the runtime startup or throw up errors on the screen.

Cheers all,
Von S

WOW @Bletchley,

type 1 ... 6, exactly that was my tought with "groups".

More later!

Greetings

That's great Von S and Becker01, I look forward to that, and any more comments or suggestions. I am off work with Covid, so I have an unusually large amount of time at the moment to devote to playing and tweaking WOFF - so if I can be of any help with testing just let me know. I think testing and tweaking is likely to be the most time consuming part of this.

Hallo @Bletchley,

now more ... .
At first let me say, that I haven't any knowledge about the engines and their failures. So I can't say anything about this.

I think more about the most senseful mod-process. @OrbX of course is right with the simulatipn.xml-argument and the combination with the MultiMod is a good idea.
On the other side I'm sure, you will develop and test the mod very often during the first phases. And for that time it's better IMHO, that you have all in your hand, independing from other mods. During these weeks you can develop the mod for JSGME (f.e. 6 groups = 6 sub-mods) with your simulation.xml-file in every group/sub-mod. Additionally you can write a short briefing for the users -or test-pilots maybe- the way to copy their own simulation.xml-file in the sub-mods (6x) and to copy the different engine-changes in these user-specific simulation.xml-files (6x). Yes, that's a little bit work for the users when they prepare your mod the first time, but the best way during this phase in my eyes (or they work with your simulation.xml-file simply! Then they must only activate the group/sub-mod, they want to have).
In the next phase after all you will say "Okay, that's it! This is the final version. I have no more ideas, no new improvements and all data are in." This final version is ready for combination with an other mod.

Only an idea. What do you think?

Greetings!

Thanks Becker01. I have been thinking about this, and although we can use the simulation.xml file for creating and testing the different categories there would be a problem with making the final mod in this way as a JSGME mod (quite apart from the effect that this would have on any other mods that make use of the simulation.xml file). Changing the numbers in the simulation.xml file will effect not only YOUR aircraft, and the aircraft in your flight, but ALL aircraft in the WOFF environment. So, you could be flying an aircraft with a most unreliable engine, but the engine failure rate of every other aircraft in the WOFF world would suddenly become the same. Aircraft with Mercedes D III or Rolls Royce engines would be aborting their missions and making emergency landings all around you at the same rate as your aircraft with the dodgy rotary. I am not sure that even JJJ's multimod would solve this problem (?). In the end, I think, the solution may have to be that every aircraft has its own engine failure rules for this to work. This would be a lot of work, but with just 4 or 5 categories these rules, once created and tested (the most time consuming part), could then be copied and pasted into the appropriate aircraft files (all those that don't already have their own rules). This would also solve the problem, I hope, of the mod interfering with other mods, as by doing it this way no lasting changes would have to be made to the simulation.xml file. When completed, the simulation.xml file would remain unchanged.

Hmm, okay: if the changing of the numbers in the simulation-file effects also other aeroplanes in the environment (and maybe other mods), then of course the second way via simulation-file is difficult ... .
The third way via MultiMod? I don't know, @JJJ will know it.
If not, then I see only the step back to way 1: the detailed, safe but long work to change it for every aircraft. In this case let me remember on the prework, I have mentioned above. The work-priority would be, to feel no selfmade-pressure. It takes as long as it takes.

Maybe we'll get a little bit more feedback, which can help.

Greetings

On the other hand, do you really notice what happens to the AI aircraft that are not part of your flight? When you are flying your attention is focussed on your own aircraft and those immediately around you. If AI squads several miles away are having a greater or lesser occurrence of engine failures you are not really going to notice (especially as the process is randomised for each individual mission). Although a mod that changes the aircraft files would be better, a mod that changes just the simulation.xml file would be almost as good in practice and would avoid a mountain of work. Also, in the process of developing and updating the game the developers might make changes to the aircraft files or simulation.xml file, and if that happened the engine failure rules (unless adopted by the developers) would have to be re-added each time. It would be easier to do that for one file (the simulation.xml file) than all the aircraft files...