Fresh stuff from sukhoi.ru (Discussion)

[Peril]

Originally Posted by jimbop

I mustn't understand this properly... surely there were hundreds of pages of debate and records dredged up etc during late IL-2 development - is there still not broad consensus of what the model windows should be? If not I doubt there ever will be.

Jimbop, I can say 100% from experience that not everyone will agree on an end result from a FM build. Your last comment is correct, there will 'never' be a 100% agreement on which data to use. If I dig up data for an A6M that has it at 358mph @ 15000ft, and also data that shows a test of the same plane as 325mph on the day of testing, which is correct?

Just one example of the quandary.

Of course if you aim to only use calculated data, it's equally a problem. You can't win the data source argument, only do your best to ensure it's 'consistent' criteria data, and that all aircraft are treated equally and with parity. The coffee table books most often used as a source of data by newer sim users looking to help, are a constant source of aggravation for those who have studded the topic for many years and invested lots of money buying original source material. There will always be arguments over this plane verses that based on stories, books, and even opinions.

My only advise, and the way I tackle it, is to gather as much original data, then use this as your basis for determining performance. It should be original source data but more so you need to understand aerodynamics, research any differences in the data, to understand why any variations exist so you are better armed to make a judgement call of which data is more likely accurate. Yes, unfortunately it required lots of study and experience to be able to do a good job in building FMs, and many 'sim builders' don't have the background or the spare time to do justice to this area.

I remain hopeful that some of the hard core FM guys that stalk these places can offer their help and vast collections of data and experience to what is most often (understandably) an under financed (in time and money) aspect of any commercial simulation.

-------------------------------------------------------------------------------------------------------------------

If they ever release the core engine for us to build a game around, I'm first on board to build a Pacific Simulation. I have a few mates that may also be interested in working for nothing just to battle over the pacific.

IL2-BOP (Battle Over the Pacific)


----------------------------------------------------------------------------------------------------------------------

Hang in there BlackSix, you're doing a great job, I know how hard it is to manage a community of passionate people

[ACE-OF-ACES]

Originally Posted by Peril

Jimbop, I can say 100% from experience that not everyone will agree on an end result from a FM build. Your last comment is correct, there will 'never' be a 100% agreement on which data to use.

Agreed 100%

If by everyone you mean any Joe Blow flight simmer.. You know the kind of flight simmer who complains his P51 is too slow.. And when you asked what values he used, TAS or IAS.. His response is a blank stare and he says 'what is TAS and IAS'?

But..

IMHO if you limit the 'everyone' to 'everyone who is qualified' than I think you could obtain 100% agreement on 'most' items

Originally Posted by Peril

If I dig up data for an A6M that has it at 358mph @ 15000ft, and also data that shows a test of the same plane as 325mph on the day of testing, which is correct?

The test that provides enough information to re-produce the test.. Chances are the difference in speed is due to some difference in the plane configuration and or test day conditions.. As in maybe the data was not converted back to STD ATM conditions.. That is to say, most if not all of the data I have reviewed, where there was a different in speed, there was also a difference in the plane configuration.

Originally Posted by Peril

Of course if you aim to only use calculated data, it's equally a problem.

Agreed 100%

Originally Posted by Peril

You can't win the data source argument, only do your best to ensure it's 'consistent' criteria data, and that all aircraft are treated equally and with parity.

The goal should not be to 'win' over every Joe blow simmer.. for reasons I noted above. Because there is no winning that fight! The goal should be to win over a qualified group of people that the Joe blows trust

Originally Posted by Peril

The coffee table books most often used as a source of data by newer sim users looking to help, are a constant source of aggravation for those who have studded the topic for many years and invested lots of money buying original source material.

Agreed 100%

Nothing like Joe Blow simmer reading the coffee book table that says the P51's max speed is 426mph, and than he wonders whey he cant get no where near that value at sea level! Why? Because it was obtained at 26kft!

Originally Posted by Peril

There will always be arguments over this plane verses that based on stories, books, and even opinions.

Combat reports do cause a lot of problems.. Why? Well for one most if not all of them are one sided stories.. Most if not all of them DON'T provide enough information to recreate the scenario.. And most if not all of them did not have any recording devices running like they do in a standard test.

For example a P51 pilot writes in his report that he 'turned inside' a Bf109 and shot it down..

Problem begins when Joe Blow simmer reads that and thinks his simulated P51 should be able to out turn a Bf109..

Because Joe Blow simmer never stops to ask what altitude did this occur at, what was the P51s E state retaliative to the Bf109s E state prior to the scenario, what was the Bf109 pilots state? As in did the Bf109 pilot even see the P51 and try to avoid it? Was he wounded? etc.. etc..

Originally Posted by Peril

My only advise, and the way I tackle it, is to gather as much original data, then use this as your basis for determining performance. It should be original source data but more so you need to understand aerodynamics, research any differences in the data, to understand why any variations exist so you are better armed to make a judgement call of which data is more likely accurate.

Agreed 100%

Originally Posted by Peril

Yes, unfortunately it required lots of study and experience to be able to do a good job in building FMs, and many 'sim builders' don't have the background or the spare time to do justice to this area.

So true

Originally Posted by Peril

I remain hopeful that some of the hard core FM guys that stalk these places can offer their help and vast collections of data and experience to what is most often (understandably) an under financed (in time and money) aspect of any commercial simulation.

I am willing..

And able!

I have been analyzing IL-2 data for some 10 years now and have written several programs to extract the data from IL-2 via DeviceLink and have written dozens of MATLAB script files to process the data collected during testing

Problem with CoD at the moment is they have not implemented the DeviceLink interface, there is the C# script method that can be used to collected (get) data but as far as I can tell there is no C# script methods to send (set) data. Thus porting over the IL-2 3rd party auto pilot will be hard to do with no way to send commands..

Thus we would have to rely on real sim pilots to preform the test.. And the down side there is, the errors due to the sim pilot are bigger then the errors you are looking for..

That is to say real sim pilots are no test pilots!

At least based on the dozens upon dozens of track files I have reviewed from IL-2 over the past 10 or so years