Modelling the bad bits

[HagarTheHorrible]

How far does/should the sim go towards modelling the bad bits or imperfections in an aircraft's design ?

 

Someone mentioned in these forums that Russian canopies tended to yellow quite quickly and presumably reduce visibility.  Should there be some variation in canopies even for the same aircraft ?

 

I watched a video, on these forums that had a pilot criticize the cramped cockpit in the 109, possibly not in it's self a big problem, but he said that it limited how much he could move the joystick left and right, presumably reducing the theoretically possible roll rate of the aircraft, which if not given any thought when composing the FM's might lead to a higher performance that the pure mechanics might indicate.

 

As an aside, I was just reading about how good the Mig 3 was, if not for the roll it was asked to perform by the Soviets, interestingly the author claimed that one of the reasons older Western designs may have persisted in popularity was that they came fitted with reflector sights as standard but in comparison many, particularly earlier on, Soviet fighters only came with simple sights, if any at all, making them far less effective by comparison when it came to the business end of what fighter aircraft actually do, shoot down other aircraft.  This, amongst other reasons, was why the Germans gave a good account against a potentially better aircraft.

 

So what say you about including, or not, aircraft foibles ?  I'm sure, if you tried, you could think of many similar or small details that prevented or limited some weapon of war from reaching its full potential and therefore not performing to its best.  You only need to think of how the Germans struggled before Moscow because of deficiencies in lubrication for weapons and vehicles, a small detail, but vital if your life depends on it.

[Cybermat47]

I like the idea of imperfections in a plane. This should definitely be modeled in-game. Thanks for bringing this up!

Edited September 21, 2013 by Cybermat47

[eklund]

You cannot model badly trained russian pilots im sorry.

[Cybermat47]

You cannot model badly trained russian pilots im sorry.

Most Russian pilots by winter 1942 were competent, after getting experience hunting down the slow German transports in the past year.

[LukeFF]

Design defects? Sure. Manufacturing defects? No thanks. 

[csThor]

Most Russian pilots by winter 1942 were competent, after getting experience hunting down the slow German transports in the past year.

 

No, they weren't. Just a simple statistics from 6 months later shows just how many raw rookies the VVS had: When 17th Air Army attempted to attack the german airfields around Kharkov on July 5 1943 93% of the aircrew taking part had never ever flown a combat sortie before.

[Cybermat47]

93% of the aircrew taking part had never ever flown a combat sortie before.

Ouch! Poor guys...

[LukeFF]

Even at the end of the war in 1945, German pilots commented on how easier it still was to score victories against the Soviets as opposed to the Western Allies (Axel Urbanke, Green Hearts).

[Cybermat47]

(Axel Urbanke, Green Hearts).

 

I take it that's a book? Where can I find it?

[LukeFF]

http://www.amazon.com/Green-Hearts-Combat-Library-Eagles/dp/0966070607

 

It was originally published by Eagle Editions.

[Cybermat47]

http://www.amazon.com/Green-Hearts-Combat-Library-Eagles/dp/0966070607

 

It was originally published by Eagle Editions.

Looks good! I love the Dora. I've got a 1/72 Airfix model of it. Beautiful plane.

 

 

EDIT: 100th post! Woo-Hoo!

Edited September 21, 2013 by Cybermat47

[LukeFF]

Indeed, that book is really good and gives a lot of good insights into the struggles the day-fighter Luftwaffe force in the West faced in 1944-45.

[theOden]

You cannot model badly trained russian pilots im sorry.

Oh it will as soon as I get my Pe2 up and running

[Lord_Haw-Haw]

Oh it will as soon as I get my Pe2 up and running

  Oh you will have that when fighter pilots try to fly a bomber, or bomber pilots

try and fly fighters. Meat for the grinder!

[Panzerlang]

I don't know what it is with a lot of these WW2 books. Amazon UK want £60 (yes, that's sixty uk stoiling painds) for the hardcover.

 

Like, kiss my ass Amazon (or those supplying it to them).

Edited September 21, 2013 by JG3-Siggi

[II./JG27_Rich]

I like the idea of imperfections in a plane. This should definitely be modeled in-game. Thanks for bringing this up!

Every fighter had it's imperfections. I think if you have an aeroplane that is good in few areas for eg. climbing, diving and speed you may not have such a great turner. On the other hand there were problems on the 109 ..opviously...that could have been fixed imo and never were like the undercariage witch I think was the biggest weakness.

 

I don't know what it is with a lot of these WW2 books. Amazon UK want £60 (yes, that's sixty uk stoiling painds) for the hardcover.

 

Like, kiss my ass Amazon (or those supplying it to them).

Ebay Siggi Ebay 

[II./JG27_Rich]

Just an eg. Siggi. Used and good condition.

 

 

http://www.ebay.com/itm/Messerschmitt-Roulette-The-Western-Desert-1941-42-by-Geoffrey-Morley-Mower-/111167529958?pt=US_Nonfiction_Book&hash=item19e219e3e6

[Finkeren]

In RoF it has been confirmed, that there are modelled slight variations in the performance of each plane, so two planes of the same type might behave a little different from each other based on random selection at the start of the mission. It's something that was never really talked about for some reason and was only confirmed I think last year.

[Cybermat47]

Thanks for the info Finkeren. This is really good news.

[crossfade]

In RoF it has been confirmed, that there are modelled slight variations in the performance of each plane, so two planes of the same type might behave a little different from each other based on random selection at the start of the mission. It's something that was never really talked about for some reason and was only confirmed I think last year.

 

Confirmed for Rof i take it i think this would be very nice to have in Bos.

[Crow]

This is an interesting topic. I'd say this is one of the most popular "features" of new flight sims in recent years and definitely a tough design decision. As I see it,  you have a few basic options:

 

For interior/exterior skins:

1. Create a perfect example of the aircraft, as if it just rolled off the factory floor.

2. Create a slightly worn aircraft as if it had seen some use. An aircraft that has been "broken-in," but not run hard.

3. Create an aircraft with the level of wear an average pilot might see.

 

For flight model/damage model:

1. Create a perfect, factory-new aircraft with no defects

2. Create an average aircraft on the front lines assuming proper maintenance

3. Create an average aircraft on the front lines assuming actual historical maintenance and supplies.

 

Then you get into the can of worms of whether or not you want to model random mechanical failures and the complexities of choosing between a purely random or a historically accurate failure rate.

 

Personally, I don't particularly care which method they use for the interior and exterior. Any of them can look good if done right, but I will say that developers tend to go overboard when trying to add wear to airplanes. The safe bet is to avoid wear as much as possible because it is difficult to do right and screwing it up can really ruin the look and usefulness of the aircraft. The danger is the developers will use modern examples which are very worn, when historical photos would probably be a better source of information.

 

For flight/damage models, I'd prefer #1 or #2, I don't care which as long as all aircraft in the game are modeled with the same standard. #3 could be interesting in a specific single-player mode like the Career mode, but it might be very frustrating and definitely wouldn't be useful in a competitive, multiplayer scenario. It would severely handicap one side depending what part of the war and what theater you are in and you would be hard pressed to find anyone that would enjoy that handicap. In an ideal world it would be a difficulty option, but the resources necessary to make two flight and damage models for each aircraft make this impractical.

 

As far as random failures, I think they could be interesting in single-player, but like historical maintenance they have no place in competitive multiplayer. That said, it's easy to see how such a feature could seriously frustrate pilot and make the game less fun to play. Part of what we play combat flight sims for is the experience and historical replication of aerial warfare--and those failures would be accurate in that context--but we still do it for fun and having a maintenance-induced engine failure mid-mission can really ruin your enjoyment. It would definitely need to be a difficulty option that could be turned off if it was included.

[BFsSmurfy]

_{I`m pretty sure I asked at Gamescom if inconsistent build quality in aircraft performance would be in the game, and the answer was NO, it would create a lot of work and leave a lot of people pissed if their Yak caught fire on take-off or their 109 blew it`s motor through no fault of the pilot. Ask yourself this, would it be a great feature if you`re flying out of your skin in a Lagg just getting the better of a good pilot in a 190 and your motor burst into flames, because Oleg at the factory had left a washer in the sump and that was accurately factored in?????? nah thought not!!!!}

[71st_AH_Hooves]

In RoF it has been confirmed, that there are modelled slight variations in the performance of each plane, so two planes of the same type might behave a little different from each other based on random selection at the start of the mission. It's something that was never really talked about for some reason and was only confirmed I think last year.

Hmm Ive never heard this confirmed anywhere..  Link??

[AndyJWest]

Random performance variations in RoF? See Loft's interview on SimHQ:"There are small differences between each one; therefore, two “identical” aircraft engines perform slightly different even with the same settings. Most Rise of Flight users do not understand this random element in our engine, but it adds an immeasurable bit of realism to flying our planes"   http://www.simhq.com/_air14/air_529a.html

 

See also Jason's post in this thread: http://riseofflight.com/Forum/viewtopic.php?f=49&t=30859&hilit=random+performance+variation&start=20

 

And yes, it is subtle...

[hiro]

I like modelling known quirks, like p-38's drive compression issue or the p-39's spin, or zero's flammability or some of the other known issues of particular model and types . . . 

 

 

But I have read lots that same model and type planes they all had their personalities and quirks about them. It would be cool if the devs could model that, but keeping in the spirit of the plane type and making it so its not a NERF (bad modelling) or frustrating. 

Also the devs would have to take into consideration .  . .  common and constant engine problems that plagued WW I planes (due to manufacturing, new development of aviation etc) would have been ironed out by late 1930's manufacturing . . . in a general sense.

 

 

But they should be optional . . .  

 

I can see someone calling in the support line. "Hi when the game loads and has the pictures of Stalingrad, they look fine, but when I'm flying everything looks yellowish, like that 300 movie. The game and my windows is fine but just when I'm flying a plane its all yellow sky." 

 

 

 

 

 

 

 

 

 

You cannot model badly trained russian pilots im sorry.

 

actually you don't need to model badly trained pilots, the problem is getting them to fly the Russian planes. (including me). j/k :D I'll jump on the VVS side, but the 109 is one of my fave's

[71st_AH_Hooves]

Random performance variations in RoF? See Loft's interview on SimHQ:"There are small differences between each one; therefore, two “identical” aircraft engines perform slightly different even with the same settings. Most Rise of Flight users do not understand this random element in our engine, but it adds an immeasurable bit of realism to flying our planes"   http://www.simhq.com/_air14/air_529a.html

 

See also Jason's post in this thread: http://riseofflight.com/Forum/viewtopic.php?f=49&t=30859&hilit=random+performance+variation&start=20

 

And yes, it is subtle...

very cool ty for the link!

[HagarTheHorrible]

Sorry, I possibly didn't make myself clear.

 

Rather than random variability, although a plus or minus performance variability of a couple of % might be interesting! I was thinking more along the lines of real life constraints that maybe limited the aircraft more than pure performance figures might suggest.

 

Let us suggest, purely hypothetically, that the bubble canopy on some types of aircraft supposedly gave superb visibility. This is modelled in the sim and is seen as a considerable advantage in game making it a favourite ride. In real life however, while the bubble canopy certainly offered advantages , in reality construction was so poor that the curved Perspex distorted the view added to which it aged rapidly with stress fracture, yellowing and bloom. In reality, it was a very mixed blessing and not the be all and end all that, on the surface it might appear. As I said this is purely hypothetical, a real life example however might be the cockpit canopy of the HE 111. In theory it offers superb visibility, however in reality it was less than perfect, one reason being internal reflections at night that obliged pilots to stick their heads outside a window so they could see when required at critical moments like landing etc.

 

As somebody else has brought up, the P38, it could dive extremely quickly, the problem being in the early models that they wouldn't stop until they hit something hard. Let's imagine for a moment that the games physics engine doesn't allow for compressibility problems and as such sim pilots can dive the P38 without regard to this, giving a performance that real pilots, with real lives to lose would have been loath to attempt, making combat with this type atypical of what happened in actuality.

 

As suggested in my opening post what also if the supposed performance of an aircraft, that might be suggested in pure mathematical terms, was not what it was in real life because of other constraints. Joystick travel in a cramped cockpit for example or engine/airframe stress and wear that played a significant part in the real day to day considerations of pilots that for sim pilots who don't have to worry about the next mount or even the next 10 minutes, have no impact whatsoever, but significantly squew an outcome in a sim.

 

As we all know, there is often a less than insignificant gap between what a shiny product brochure might suggest is the best thing since sliced bread and what we actually have to live with and accept as reality. I dought WW2 aircraft were any less so. A case in point might be the surprise expressed about Soviet fighters after the Iron Curtain came down, by some Western pilots, about how rough/badly finished Soviet aircraft were . Technically they might have been brilliant and from film footage extremely impressive but it wasn't until they could be seen hands on,with critical eyes, that the full picture became evident, that they also had their less than obvious imperfections.

[Crow]

If what we're talking about is something like not modeling the P-38's compression problems or not modeling the A6M's flammability, then I'm not sure I would called the FMs or DMs accurate anymore. I think any product calling itself a "sim" needs these kinds of features. I don't really think that's even a point of debate. The aircraft should have historically accurate FMs and DMs.

 

The more important debate is what does "historically accurate" mean. And for that, see my post above.

Edited September 23, 2013 by Crow

[Recon]

Random performance variations in RoF? See Loft's interview on SimHQ:"There are small differences between each one; therefore, two “identical” aircraft engines perform slightly different even with the same settings. Most Rise of Flight users do not understand this random element in our engine, but it adds an immeasurable bit of realism to flying our planes"  [/size] http://www.simhq.com/_air14/air_529a.html

 

See also Jason's post in this thread: http://riseofflight.com/Forum/viewtopic.php?f=49&t=30859&hilit=random+performance+variation&start=20

 

And yes, it is subtle...

In software development we call those bugs lol - now it's called 'random performance variations'

Sorry, can't resist

[Crump]

One of the largest issues with these games is player expectation.

 

They see a graph of aircraft performance under standard conditions that represents a very limited over boosted condition.  They then expect their aircraft to perform as such under all conditions as the norm.

[HeavyCavalrySgt]

You cannot model badly trained russian pilots im sorry.

Why can't training and experience be modeled?

[HeavyCavalrySgt]

I was just reading, for example, about some P-47 drivers.  Apparently that engine would run with multiple cylinder knocked off the engine and no oil left, if you ran it at full throttle.  Pull back the throttle and the engine would seize immediately.

 

One of the gentlemen (on his first combat mission in the P-47) hit a wire and sheared all the cylinders off the top of the engine and folded the cowling back, then flew 40 minutes home with no oil pressure.  The experienced pilots knew the engine would get him home, and were coaching him to leave the power in.

 

I would love to see quirks like that modeled too.

[LukeFF]

In software development we call those bugs lol - now it's called 'random performance variations'

Sorry, can't resist

 

[HagarTheHorrible]

Ok, I've got another poser.

 

I was watching an excellent documentary about the Spitfire yesterday, posted by TheBlackPenguin.  It was made in 1976 and included well known pilots and the actual designers talking about the aircraft.  At one point in the documentary Douglas Bader and Stanford Tuck describe the problems of the carburettor in the Spitfire verses the direct fuel injection in the 109.  This obviously, as we all know, meant that the best course of action for a 109 when attacked by a Spit or Hurricane was to stuff it's nose down to escape, the Merlin engined Spit needing to roll onto it's back to follow the 109 down, thus giving the 109 an initial advantage.

 

So far all good and well.

 

DB and ST then went on to say that all was not lost however because the Spitfire pilot could catch and cut across the 109's when the German pilots pulled out of the dive.  They surmised that some poor sod had ripped the wings of his 109 and word had then got around the Luftwaffe that high speed pull outs were a bit dodgy and probably a risky thing.  This meant, what might have been an advantage in the 109, was null and void because it also had a serious disadvantage.

 

That rumour, however, might have been utterly misguided.  We, 60 years on, might know differently, that the wings weren't likely to fall off when pulling out of a fast dive too quickly.  We might also, with nothing to risk, be far more prepared to take it to the wire than a real pilot with everything to lose, gaming the odds, " will the wings fall off or won't they ?"

 

In essence, 60 years on from the events and with lots of additional information at our hands and nothing to lose, sim pilots fly differently and use different tactics than those used all those years ago, thus, in effect, gaming the game. 

 

So, what do the developers do ? 

 

Let's have a theoretical debate, given the above scenario as suggested by DB and ST.

 

Do the developers model the 109 along the lines of best knowledge, modelling it as closely as possible to how it technically was or do they model in a series of flaws that model the perceived weaknesses or real life considerations that real pilots had to contend with ?

 

On one side of the argument you have an aircraft that is as faithful to the real thing as it is possible to make.  This possibly might mean that a sim pilot can do things that would have been considered dangerous, foolish or just not done in order to exploit a perceived advantage.  On the other hand, do you model in a flawed, less than perfect representation that more closely recreates the real world considerations and limitations of real pilots and consequently real life tactics as used at the time ?

 

To sum up.  Wings fall off plane when pulling out of fast dive.  Rumour gets around other pilots and because they don't want to test the theory they limit the full potential of the aircraft they are flying.  This leads to a potential advantage being nullified.  Later, much later, testing proves this rumour to be wrong.  Do you use what they thought then or what future generations know now to model the aircraft, with all the potential ramifications and rewriting of history that that might entail ?

[6./ZG26_Emil]

Thats where Silent Hunter (3 with mods? ) had a good idea and had a +/- to the crush depth. Max Speed before structural failure should have the same.

[HagarTheHorrible]

I'm really surprised that there isn't more discussion around this, it goes right to the philosophical heart of combat flight sims.

 

Lets provide another scenario.

 

A fighter plane, it matters not as to which one, has a button, this button boosts it's performance for short periods, lets say for arguments sake 5-10 minutes. In real life this button was used exceedingly sparingly because it was extremely damaging to the engine and if used for the full ten minutes between overhauls would totally junk the engine such that the pilot would need to look for somewhere to force land or bail out. Using this button, even for short periods would also reduce the reliability of the engine, with consequenses of capture over enemy lines and also with spare parts in short supply lead to serviceability problems, again making the possibility of engine failure over enemy territory that much higher. Being caught meant the possibility of several years in a POW camp at best and torture and firing squad at worst.

 

In a flight sim however abusing the aircraft has no concequenses over the long term because pilots only live for the now. You find that, once combat has been joined, the button is pressed all the time because the only thing that matters is winning the fight, not the 20 minute flight home over enemy territory or even what aircraft you will use for the next flight because each time you go to fly you get a new aircraft and if not you turn around to your CO and throw a hissing fit and refuse to fly until the plane you want us available again.

 

What you end up with, because of the very short term nature of flight sim combat is very atypical situations or outcomes, gaming the game because you don't need to worry about five minutes down the line.

 

So do developers model for technical perfection that lead to ahistorical outcomes or do they model to prevent simmers from gaming the game even if it isn't strictly accurate in technical terms.

 

To put it another way, is it better to have technical perfect aircraft, as in a study sim, or do you concentrate on combat outcomes to the detriment of technical perfection?

 

Which is more important, the aircraft or the combat ?

[Crow]

Hagar, it sounds like you're playing on the wrong servers. Find yourself a good dead-is-dead campaign and a lot of the "playing for the moment" behavior disappears.

[HeavyCavalrySgt]

In a flight sim however abusing the aircraft has no concequenses over the long term because pilots only live for the now. You find that, once combat has been joined, the button is pressed all the time because the only thing that matters is winning the fight, not the 20 minute flight home over enemy territory or even what aircraft you will use for the next flight because each time you go to fly you get a new aircraft and if not you turn around to your CO and throw a hissing fit and refuse to fly until the plane you want us available again.

 

What you end up with, because of the very short term nature of flight sim combat is very atypical situations or outcomes, gaming the game because you don't need to worry about five minutes down the line.

 

 

You might enjoy playing a sim like Rise of Flight where you are down to two available aircraft to assign to a protection mission going way over the lines because your planes have been taking a beating.

[Stig]

Some very interesting points you are raising here, Hagar.

 

1. Known, documented 'foibles' should of cause be modelled in FM/ DM (if possble).

 

2. Percieved or imagined problems by RL pilots would probably be more difficult, as RL pilots didn't always have the same perception of a particular plane. I have seen many heated discussions in forums about what one RL pilot said contrary to another.

 

3. I doubt that devs can model anything that will stop anyone gaming the game.

 

Myself, I enjoy flying the planes as realisticly as possible, so even if I am able to exploit something like the 'button' in your example, I wouldn't do it. I also value RTB over getting kills. 

[Alesia]

I was just reading, for example, about some P-47 drivers. Apparently that engine would run with multiple cylinder knocked off the engine and no oil left, if you ran it at full throttle. Pull back the throttle and the engine would seize immediately.

 

One of the gentlemen (on his first combat mission in the P-47) hit a wire and sheared all the cylinders off the top of the engine and folded the cowling back, then flew 40 minutes home with no oil pressure. The experienced pilots knew the engine would get him home, and were coaching him to leave the power in.

 

I would love to see quirks like that modeled too.

Huh, never heard anything like this.