Aircraft Design And Performance
Calfu Baransky
It was designed from the onset to conduct maneuver air combat. It is a maneuverable aircraft. Unfortunately, they did not teach very many Russian pilots how to do air combat maneuvering at the time so it was few elite pilots who knew how to do this. Most American pilots (and Western pilots in general) were trained to do WVR combat.
Considering they made wild claims about the MiG-23 even being capable of shooting 20mm NATO from their 23mm I will discount any claim from the constant peg program until it is proven by actual documentation and flight testing.
And early on there were many nations without any BVR capability for their F-16s so the MiG-23 series was quite formidable! When the MLD came out and was equipped with R-73 it maintained a fierce opposition to American gen4 fighters.
Therefore, you indicate there was no real competent evaluations conducted by Airforce and Pentagon analyst down to the bolt and nut. That no experience or real knowledge came to the pilots who the Mig23 daily, in any test that put the Mig23 far past what operational manuals dictate, some losing their lives. That nothing was learned as the pilots also served as top secret aggressor squadrons in other CONSTANT subprograms where only hand selected air crews of the Navy and Airforce would face. Even certain A-10 pilots were flown against the 23.
So, any data that was gathered in the CONSTANT PEG program was a failure and any actual documentation published of their evaluations is to be brushed off as made up, or propaganda all because someone in the program claimed something incorrect. That is fascinating.
Ha... I just had pulled (after some negotiations) the trigger on a copy of Dechamps, H. und K. Kutzbach: Prfung, Wertung und Weiterentwicklung von Flugmotoren. Still unsure if the (for us) relevant aricles in that volume are present readable. Fingers crossed. For $70 including shippment, I don't know what to expect. Most exapmles I found were obscenly pricy. Hate buying antiques without putting my fingers on it first, but what can you do? As I'll be on vacation the coming two weeks, shippment has to wait. And pulling the trigger on that one will have to wait accordingly.
Quick update: @Han replied to my PM and added some additional developers to the PM thread. I'm happy to say that they are interested in the book, so I will send them one as soon as I get my personal copies which is slated for the end of the month.
The book arrives! Needless to say I was not disappointed and devoured it in 2 days. I've never before come across such a comprehensive study of WW1 aeronautical engineering written by someone with this level of technical knowledge, as well as a keen eye for history (and familiar with both German and British/French sources), and strong contacts within the European vintage aviation community.
Most books on WW1 aviation tend to be histories and/or coverage of surviving data sources, but very few if any dig into the detail of what we can learn applying modern modelling techniques to the evidence we have available, probably because most are written by historians and not engineers.
I wont spoil it but will say that @Holtzauge covers a brief history of aeronautical engineering in the early C20th, every engineering aspect of consequence that might materially impact aircraft performance from types of drag, to wires, engines, wing shapes, aspect ratio, fuels etc etc, as well as what surviving data he's used to tune his models which take into account tons of variables. You will likely need some knowledge of engineering and half decent math skills to fully understand the content - it's deep!
You've then got the good stuff on performance data which throws up a couple of surprises to anyone who has flown Rise of Flight and then Flying Circus for many years, but perhaps not that shocking in some cases (Mr Jonsson has already shared some data with us on the poor N28).
If I were a sim developer this would be my handbook, and I'd be knocking his door down to get access to his models, plus ask for more types to be plugged in. It begs the question of what 1C could do to the flight models with such data.
EDIT: it's also interesting to see how much the developers have actually got right (if the models are accurate) without access to former SAAB employees. I get the sense that the sim can only get better.
This is a good point @US103_Rummell: I think Flying Circus (and Rise of Flight) are very good and enjoyable simulations and I enjoy flying the planes in both of them. And as you say, many things are done right and I would recommend anyone interested in WW1 aviation to get both titles.
That being said, there is always room for improvements in simulations. That goes both for the ones in Flying Circus and my C++ simulations. And when my personal batch of books arrive (scheduled for Tuesday next week) I will send a copy to the developer and take it from there.
I hope we could collaborate somehow and share ideas about what are the performance numbers to target. I am anyway open for such a collaboration and would gladly share my input if it could be of help to tune the flight models.
I intend to post some stuff here later on about my flight models and performance results, and how they compare to what we have in-game, but I have been holding back intentionally since I want the developers to get their copy of the book before I do.
I think this would be the most productive approach since otherwise, they will just get a lot of numbers with no context. Once they have the book though, it will be more clear why I arrive at the results I do. If you look in the book on pages 57-94 about wing profiles and wing planforms, these are really key to my models and why I get the results I do.
The first thing I would like to ask them about is their take on the high altitude modeling, where my results are much more conservative. For example, my turn time estimates (See pages 187-188) are in some cases nearly double those we have in-game at 5 km altitude.
Only 25% through it, but it's a book I can really settle down with and enjoy taking it all in. Feels like I'm reading history but seriously upgrading my aeronautical engineering knowledge at the same time.
Might suggest that when we get to the point of comparing your models to in-game testing, there is a very clearly stated test methodology that, naturally, aligns with the rules used in the computer model.
About the testing procedure @US103_Baer: Many here have extensive experience flying the in-game aircraft and I have seen videos with some nice flying keeping both altitude and speed in turns. But yes, I agree that losing a bit altitude or allowing the speed to drop does affect the result so some skilled flying is needed to get the decimal points right. My method when I have been doing it myself is to literally fly in circles trying to keep speed and altitude correct and record a track. Then post analyze the track trying to figure out which portions are the best and taking the average time of those. But when it comes to the best method on how to do this in-game, I will gladly defer this to you guys who have a lot more in-game stick time and not as rusty as myself.
Testing the instantaneous turns in-game is a lot trickier: But I have seen some nice flying done by @=IRFC=Artun to test this on the Fokker Dr.I and I believe he is compiling in-game data on other aircraft as we speak. The book only gives the time to complete two turns (pages 192-194) but I have given Artun data on 360 degree turns as well for some of the in-game aircraft and can post more instantaneous turn data for 360 deg turns here later on.
As general observation, I think we will find that that the in-game aircraft retain speed better in instantaneous turns (some, maybe not all!) and get a better return on zooms from dives than my models predict. In addition, many of the Entente scouts do better in turns in my model at higher altitudes than in-game which I also believe is correct for the reasons I outline in the book.
I'm yet to finish the book cover to cover, though I've paged through most of it and so far it's been by far the most interesting reference work on WW1 aircraft performance I've ever had the pleasure to read. The title of the book does not lie. Without spoiling anything: I had an absolute aha moment when reading the chapter on requirement specifications and landing speed. The fact that sustained turn was never a product of any military or functional requirement makes a lot of sense.
That said, it exposes quite a few flaws in the current FMs -- except, maybe, unsurprisingly (and not yet 100% certainly) -- in the newly released Snipe and soon to be released Siemens-Schuckert. But since the Snipe is not part of the book and the Siemens-Schuckert isn't out yet, I can't definitively comment on it. But yes, let's restate the obvious: almost all the FMs are over 10 years old, while the Snipe and Siemens-Schuckert's FMs are brand new and developed in a new, modern engine, and so is the data coming from this book (from the C++ simulations). And to continue stating the bleeding obvious: we absolutely need engine variants on most if not all of the planes.
But this is just the sort of thing I was hoping for a fruitful collaboration with the developers on: I have both data and my simulation engine. They probably have a lot of data on their end and one thing that we could both benefit from I think is to get a better understanding of what the Mercedes D.IIIa was truly capable off since this would improve both our flight models.
About a WW1 sequel: I will probably put that on the back burner for now: I will most likely instead go for the low hanging fruit with the HUGE market potential: The Bf 109E versus the Spitfire Mk 1 during the BoB period: What could possible go wrong? Bound to sell like hotcakes!
c80f0f1006