Rama Posted March 2, 2015 Posted March 2, 2015 I have to stick to the interpretation that seems more probable. More probable for you (whatver interpretation it is). For the last point, I'll stay to the statement: "you can't define model parameters from non-quantitative description, and you can't control the outcomes with non-quantitative description". You can build an opinion/belief on the feeling of it.... but no more. The model itself (aka the mixt of physical laws and empirical solutions you choose to implement), may be driven by the qualitative "feeling" of the outcomes (like for exemple if the outer wing stalls before the inner wing in a fast symetric coordinate turn at 50°/60° inclination... the, except if the dihedral angle is abnormal, it should'nt happen) but then: 1) My own belief is that it has been very well done on BoS, better (for me) than in any other flight sim game I tried (not talking about the performances correcness there, just about the feeling) 2) When you change a piece of a complex model, it generally has impact on all the outcomes of the model, impacts which are difficult, if not impossible to forecast.... so a modeller should not change a component of a model unless he's 100% that this component is highly eroneous. 3) All what I see as FM complains on this board, if leading to corrections, would be a matter of model parameters, not model components. ... so, the conclusion is that only quantitative data/charts can be helpfull to find discrepencies that will lead to model parameters adjustments. 1
6./ZG26_Gielow Posted March 2, 2015 Posted March 2, 2015 Handle the FW190 like you are disarming a nuclear bomb that could explode any second and you should do just fine. That is the secret !!!
CaK_Rumcajs Posted March 2, 2015 Posted March 2, 2015 I bet it's clear that hard numbers are hard numbers. If there is methodology how they were obtained and if they were confirmed then there is no way to argue. So Rama, how do you explain for example the unexpected good performance of the Yak fighter above 4000m? Well I know I don't have numbers so I shall not be surprised. And how do you explain the nearly total indifference of full rich mixture settings on high altitude performance that the planes equiped with manual mixture control sport? Are you gonna defend the state of the engine simulation? What about the roll rates of La fighters? When 1CGS goes to BOK (Kursk) are we gonna be told the La5FN was with its improved handling the best in roll dicipline out of WW2 fighters? Because there were improvements in handling so we should really notice the new quality comparing to La 5. Don't get me wrong. I love the Yak 1 as it is. I spare maybe 50% of my flight time to it and I enjoy flying the airplane. I also say the best fighter we currently have is the Bf 109F4 without any doubt. But there are things that are suspicious and they will probably be changed. Like the adjustable trims were redesigned. Everything was perfect but suddenly it's "better". So I say the FM is the best in the world till next version changes it. Then the new FM will take over the King of the hill title. If I were you I wouldn't spend so much time defending current state of things. FM is gonna be modified/tweaked/adjusted with coming versions. Maybe not soon but once changes arrive someone will find this old discussion and will point out what believers were saying. Simulation is always compared to reality. And then adjusted. There is no perfect mathematical model. And games can't use really advanced models anyways. PCs do not have the horsepower needed to run such models. So it boils down to some simplified model used that is fed by a set of constants and curves. But these are huge simplification of what the real world is. So huge it's even funny to read words like "this is the mathematics and physics behind". Where is the physics when we run our Yaks with 100% mixture at 6000m? Well I love the game. It's the best there is. And I say that to anyone who asks. That all with respect to people who have created the simulation and allowed us to enjoy it. And I also know it's pointless to defend FM/DM because of reasons I mentioned. The only constant is change.
Rama Posted March 2, 2015 Posted March 2, 2015 So Rama, how do you explain .../... Maybe you missed it, but I don't try to explain anything concerning the models outcomes, nor do I participate to any FM discussion. I know it's useless. I just try to explain what can be usefull for the dev, and what isn't usefull for sure. So I say the FM is the best in the world till next version changes it. Then the new FM will take over the King of the hill title. Correct.... and you can say the same for physic theories. One theory is the best in the world until next one changes it abd become the King ot the Hill theory. If I were you I wouldn't spend so much time defending current state of things. Where did you read I'm defending the "current state of things". Can you quote please? FM is gonna be modified/tweaked/adjusted with coming versions. Of course, Never said the contrary.... or were did I, can you quote? someone will find this old discussion and will point out what believers were saying. Maybe.... and he will have hard time to find a belief in my words. I'm a skeptic by nature. But yes, A FM debates, to which I don't participate, are in big parts confrontations of beliefs, that's what I explained a few pages ago.... Simulation is always compared to reality. And then adjusted. There is no perfect mathematical model. Yes, and?.... did I said the contrary? can you quote? And games can't use really advanced models anyways. PCs do not have the horsepower needed to run such models. So it boils down to some simplified model used that is fed by a set of constants and curves. The FM models used in this game are really Advanced compared to the models used in previous sims, and less Advanced that some used in Professional sims, themthelves less Advanced that some used to "test" physical theories, etc.... My small garden is smaller than a big garden who's smaller than a park, who's smaller than a country.... and? But these are huge simplification of what the real world is. Physical laws are also approximation and huge simplification of what the real world is... They're still good enough to model it, and to allow all the technology we use today... So huge it's even funny to read words like "this is the mathematics and physics behind". What is funny? I don't catch it... may you explain so I can laugh too? 1
WWChunk Posted March 2, 2015 Posted March 2, 2015 Ok then I'm a little afraid that this great stuff might get lost in this big dicussion without reaching the right devs... shouldn't anybody who had contact with them previously send it, maybe to Han and Zak? They promised that they would look at it if is sent the right form, which they specified. Yes, if you have info that can help, please send it to the devs, or to one of us, so it gets where it needs to go. At the very least, they'll have additional info that could possibly help.
CaK_Rumcajs Posted March 2, 2015 Posted March 2, 2015 What is funny? I don't catch it... may you explain so I can laugh too? The models used are empirical data based plus some mathematical concept. This allows for good aproximation within boundaries of where there is data to match against. But there is no way to believe anything outside these boundaries. So OK there are performance curves and numbers. It's certainly possible to tweak input of the mathematical model to meet these with acceptable precission. But flight is way more complex than data available. And definitelly goes far beyond what is possible to simulate in real time. Examples of limitations of these models are found in all flight sims BOS including. Also you sound rude. You are a moderator. Would you mind to tone down a bit?
303_Kwiatek Posted March 2, 2015 Posted March 2, 2015 (edited) Yes, if you have info that can help, please send it to the devs, or to one of us, so it gets where it needs to go. At the very least, they'll have additional info that could possibly help. Data was send many times but it looks that devs are not such interested in changing some things. To remind : devs were informed about A-3 poor climb rate with test results and German charts. Han replayed but they minimalize the problem and adjusted very poor A-3 climb rate a litttle for calm down some fans but truly speaking they will never fixed it like they should. Result ---- A-3 still got too weak climb rate. Han wrote some time ago : 2. Fw 190 A-3 climb rate in game is less than it's noted in reports. - Actually, it's only 5% less than our reference, but we now have a new method of performance adjustment that will allow us to make more precise tuning of this parameter. Claim is not critical, but corrected allready in updates. I don't think they are really interested in historical accuracy so much ( there are too many examples for it unforunately) Edited March 2, 2015 by 303_Kwiatek
II/JG17_HerrMurf Posted March 2, 2015 Posted March 2, 2015 Anecdotally, I was caught for the first time in almost six mos by a Yak from behind (even if I start at a slight E disadvantage). I was very confused as I almost never get caught by them in a level footrace. Then, after considerable thought, I realized I was flying as a Jabo and had that damn bomb rack slung underneath with the associated drag. Modelled correctly, that? Probably. ~S
GOAT-ACEOFACES Posted March 2, 2015 Posted March 2, 2015 I find that hard to believe There is another possibility.. Chances are more likely that they found your testing to be in error or inconclusive, or your sources were off from what they implemented. Hope that helps I don't think they are really interested in historical accuracy so much.
303_Kwiatek Posted March 2, 2015 Posted March 2, 2015 (edited) or your sources were off from what they implemented. Not my sources but rather historical sources. But you could right German data for German plane could be way different from Soviets data for captured German plane. Edited March 2, 2015 by 303_Kwiatek
GOAT-ACEOFACES Posted March 2, 2015 Posted March 2, 2015 Not my sources but rather historical sources.Sorry, I didn't mean to imply you got the ROC data from your personally owned Fw190.. I just assumed that went without saying and that everyone here knew we were referring to data collected during WWII testing. Please accept my apology if I confused you or anyone else. But you could right German data for German plane could be way different from Soviets data for captured German plane.Not just German data vs Soviet data.. As has been noted in this thread, some detective work to be applied to researching the data. Little things like a differents in fuel loads can have big effects in rates of climb Hope that helps
Rama Posted March 2, 2015 Posted March 2, 2015 The models.../... I still miss the funny part.... too bad for me, I'm gonna miss a laugh or a smile Also you sound rude. You are a moderator. Would you mind to tone down a bit? That's probably because I'm not paid enough.... More seriously, you can find the procedure to address your concern in post #255 of this very same thread 1
CaK_Rumcajs Posted March 2, 2015 Posted March 2, 2015 I still miss the funny part.... too bad for me, I'm gonna miss a laugh or a smile Reference to outputs of a simple mathematical model with words like "this is the math and physics behind". Because it's a simple model. Because it's fed by very limited data. Because it changes a lot. Do you remember early versions of BOS? Or even 1.008? Is it necessary to list examples of behaviours changed since release? What if next update changes FM? Will we still refer to 1.009 as "the physics and math behind"? Funny is that with all that science behind it changes a lot. 1
GOAT-ACEOFACES Posted March 2, 2015 Posted March 2, 2015 And games can't use really advanced models anyways. PCs do not have the horsepower needed to run such models.Depends on which model your talking about.. The PC is more than capable of running the typical six degrees of freedom model (6DOF) which AFAIK all PC flight sim use.. The first commercial PC flight sim to implement one was MICROPROSE's Pacific Air War 1942, which was back in and around 1994ish? At that time they used 'fixxed point math' because the PC's math co-processor were just not up to the task of doing floating point calcs in real time. There are some other 'models' out there.. For example, I think it was over ten years ago, a PC flight sim implemented a 'fluid dynamic' model.. Shoot, that is not the right name for it? But it was using fluid dynamic equations to calculate the airflow over the surface of the plane.. In this case the sim was limited to a stunt plane. As noted, that was over ten years ago, and they did even at that time state they were not fully implementing all aspects of the fluid dynamic equations.. Sadly the game did not do well, and I have not seen any since then try that again. Long story short, modern PCs are more than capable of running advanced versions of the 6DOF model
StG2_Manfred Posted March 2, 2015 Posted March 2, 2015 (edited) Depends on which model your talking about.. The PC is more than capable of running the typical six degrees of freedom model (6DOF) which AFAIK all PC flight sim use.. The first commercial PC flight sim to implement one was MICROPROSE's Pacific Air War 1942, which was back in and around 1994ish? At that time they used 'fixxed point math' because the PC's math co-processor were just not up to the task of doing floating point calcs in real time. There are some other 'models' out there.. For example, I think it was over ten years ago, a PC flight sim implemented a 'fluid dynamic' model.. Shoot, that is not the right name for it? But it was using fluid dynamic equations to calculate the airflow over the surface of the plane.. In this case the sim was limited to a stunt plane. As noted, that was over ten years ago, and they did even at that time state they were not fully implementing all aspects of the fluid dynamic equations.. Sadly the game did not do well, and I have not seen any since then try that again. Long story short, modern PCs are more than capable of running advanced versions of the 6DOF model Ridiculous! Show me a PC flight game where real air and it's flow is modelled in real time! Also, have you ever provided any data to claim or support a certain flight model behaviour or do you only explain others why they're wrong? Edited March 2, 2015 by StG2_Manfred
GOAT-ACEOFACES Posted March 2, 2015 Posted March 2, 2015 (edited) Ridiculous! Show me a PC flight game where real air and it's flow is modelled in real time!Ridiculous? Did you miss the part where I said they did it over 10 years ago? A simple google of it would have saved you from making yourself look silly.. The flight sim was called Flight Unlimited Here is a link for the google challenged http://en.wikipedia.org/wiki/Flight_Unlimited Also, have you ever provided any data to claim or support a certain flight model behaviourYes many times or do you only explain others why they're wrong?You mean like I just explained to you why you are wrong? Hope that helps Edited March 2, 2015 by ACEOFACES
Brano Posted March 2, 2015 Posted March 2, 2015 Interesting to witness how yet another Fw-bomb dropped at forum created such a nice crater....9 pages wide
Rama Posted March 2, 2015 Posted March 2, 2015 Reference to outputs of a simple mathematical model with words like "this is the math and physics behind". Because it's a simple model. Because it's fed by very limited data. Because it changes a lot. Do you remember early versions of BOS? Or even 1.008? Is it necessary to list examples of behaviours changed since release? What if next update changes FM? Will we still refer to 1.009 as "the physics and math behind"? Funny is that with all that science behind it changes a lot. I allready answered to you that FMs are subject to changes, like any model output when you change the parameters (or sometimes a component, if you really need to).... this doesn't make the model "bad" or "obsolette", nor does it change it anything in the "physics and math behind" (and which is more complex than you seems to think... but it does'nt matter). The "science behind" doesn't change a bit when the outcomes of a model changes because you change the value of a parameter or even just the initial conditions. Yes, I do remember the early versions of BoS, and other versions, and I can tell you that the model itself hasn't changed much. So I understand now why I didn't found the funny part. It's because that what you find funny is your own missrepresentation of what a model is and how its outputs are tuned. I couldn't guess that.
StG2_Manfred Posted March 2, 2015 Posted March 2, 2015 Ridiculous? Did you miss the part where I said they did it over 10 years ago? A simple google of it would have saved you from making yourself look silly.. The flight sim was called Flight Unlimited Here is a link for the google challenged http://en.wikipedia.org/wiki/Flight_Unlimited Yes many times You mean like I just explained to you why you are wrong? Hope that helps lol! They can claim whatever they want. On this Wiki is stated "and he reacted by coding a simulated atmosphere....". Have you ever seen this 'coding' and can evaluate the worth of it? It was established 1995 on a Pentium 200 Mhz. CPU, correct? Unfortunately I did not study physics and therefore cannot go into any depth of FM/DM but I'm a computer professional the last 20 years and tell you something... - it took DeepBlue from IBM to just beat a human in playing chess... - here in BoS it takes only a handful of AI which just have to perform a couple of logically flight maneuvers and it brings the CPU to it's limits - and you want to tell there is the fluid of airflow (after Bernoulli) calculated ? Rumcajs is right with his statement, that the FM/DM has to be (drastically) simplified to be able to run on a recent PC, especially if there has to be more than the plane and an empty room. 1
GOAT-ACEOFACES Posted March 2, 2015 Posted March 2, 2015 I'm sorry Aces! You were right when you said there was a PC flight sim made over 10 years ago that simulated real air flow in real time. I should have known better than to think you would just make something like that up! Please accept my apologies and I retract my 'Ridiculous' statement! S! Fixed that for ya.. Oh, and apology accepted!
II./JG77_Manu* Posted March 2, 2015 Posted March 2, 2015 Fixed that for ya.. Oh, and apology accepted! You are making half of each and every topic in the whole forum to a ridiculous dickfest, where you want to undermine every one who has a different opinion then you. Using dubious and ridiculous accusations and explanations for your rightness. But even you just hit a new low. Writing your own stuff in a quote from another guy...the only one who is "making himself look silly" in here, is you, my dear. Btw..BTT Manfred is right, you are wrong. But it shows one more time that you have no idea what you are talking about, when you think that a PC can handle thermodynamics on a 300x300km map in realtime . 1
GOAT-ACEOFACES Posted March 2, 2015 Posted March 2, 2015 Apparently celestiale, you still have me confused with someone that values your opinion! Hope that helps
CaK_Rumcajs Posted March 2, 2015 Posted March 2, 2015 I allready answered to you that FMs are subject to changes, like any model output when you change the parameters (or sometimes a component, if you really need to).... this doesn't make the model "bad" or "obsolette", nor does it change it anything in the "physics and math behind" (and which is more complex than you seems to think... but it does'nt matter). The "science behind" doesn't change a bit when the outcomes of a model changes because you change the value of a parameter or even just the initial conditions. Yes, I do remember the early versions of BoS, and other versions, and I can tell you that the model itself hasn't changed much. So I understand now why I didn't found the funny part. It's because that what you find funny is your own missrepresentation of what a model is and how its outputs are tuned. I couldn't guess that. I also believe the model hasn't changed much. Just the data it takes as input. Regardless of size of the changes the final product has changed significantly. And is changing with each version. I'm an mechanical engineer specialized in theoretical mechanics, kinematics, dynamics and also dynamics of fluids. There is a very slim chance you could tell me something new about the models behind. 1
II./JG77_Manu* Posted March 2, 2015 Posted March 2, 2015 Apparently celestiale, you still have me confused with someone that values your opinion! Hope that helps I think here are plenty of people around who value my opinion. On the other side it's pretty obvious what nonsense is coming mostly from your posts. Like a cheap tabloid you are manipulating data, charts, and quotes from other people, twisting anyone's every word , to represent a certain agenda, and make yourself show right in every discussion. But i am pretty sure most people here in the forum already noticed that. 2
GOAT-ACEOFACES Posted March 2, 2015 Posted March 2, 2015 Apparently celestiale, you still have me confused with someone that values your opinion! Hope that helps
JtD Posted March 2, 2015 Posted March 2, 2015 Since this topic has completely moved off topic, my 2 cents on "anecdotal" evidence. I don't consider it as valuable as measured, but it should not be disregarded. It can on occasion be used to improve a flight model, but more importantly, it can be used to improve the understanding of measured data and as a plausibility check for models. If you have 18 out of 20 statements A>B and 3 out of 3 measurements B>A, you can be certain that there's a link missing so that all makes sense. Just going with B>A is unlikely to be correct. First side note: Several quantities today are still being most objectively evaluated by gathering "anecdotal" evidence. These quantities are usually subjective ones, like comfort or noise. The only direct way to "measure" it is a poll. This also applies to some aircraft qualities, for instance to control forces "light" or "heavy". Simple example, due to ergonomics, 15kg might be considered heavier than 20kg. If you now model both sets of controls with similar force limits, the model is not properly reflecting reality, even though you went with the numbers. Second side note: Last week I supervised a safety relevant test and it was evaluated by me looking at it and saying it was OK / not OK, based on my experience in the field. No hard data, nothing quantifiable. Still safety relevant and necessary for approval for use with the public. 2
StG2_Manfred Posted March 2, 2015 Posted March 2, 2015 Fixed that for ya.. Oh, and apology accepted! You have obviously no idea about computers, you embarrassed yourself
GOAT-ACEOFACES Posted March 2, 2015 Posted March 2, 2015 You have obviously no idea about computers, you embarrassed yourself Ah, I see where you are confused! Note I did not work on Flight Unlimited and implement the fluid dynamic code.. I am simply the messenger here! And as the messenger I simply pointed out that there was a flight simulation that implemented fluid dynamic equations to simulate air flow over the surface of the plane, and that it was done over ten years ago. The purpose of me bringing that up was in regards to the notion that the 6DOF modle was too much for modern PCs. So, if you want to accuses flight sim makers like 'Looking Glass Technoogies' and 1C/777 of lying, being biased, stupid, etc.. I would recommend you take it up with them, not me! Because I am just the messenger! And I simply reported what they said. And based off what they said, they implemented fluid dynamic equitations in thier flight model. Which means I was right and you were wrong when you said I was ridiculous for saying that. Hope that helps!
6./ZG26_Gielow Posted March 2, 2015 Posted March 2, 2015 (edited) I am really happy to see the new standards for free speech on this forum. One year ago you guys would be banished for much less. I was really important expanding the boundaries and probably the only "ban ace" to be writing in this forum. Voltaire would be proud Edited March 2, 2015 by =[Coffin]=Gielow
StG2_Manfred Posted March 2, 2015 Posted March 2, 2015 Ah, I see where you are confused! Note I did not work on Flight Unlimited and implement the fluid dynamic code.. I am simply the messenger here! And as the messenger I simply pointed out that there was a flight simulation that implemented fluid dynamic equations to simulate air flow over the surface of the plane, and that it was done over ten years ago. The purpose of me bringing that up was in regards to the notion that the 6DOF modle was too much for modern PCs. So, if you want to accuses flight sim makers like 'Looking Glass Technoogies' and 1C/777 of lying, being biased, stupid, etc.. I would recommend you take it up with them, not me! Because I am just the messenger! And I simply reported what they said. And based off what they said, they implemented fluid dynamic equitations in thier flight model. Which means I was right and you were wrong when you said I was ridiculous for saying that. Hope that helps! You said: "But it was using fluid dynamic equations to calculate the airflow over the surface of the plane.." (....in ~1995). That was your post http://forum.il2sturmovik.com/topic/14806-so-whats-verdict-fw-190-now/?p=239069 I say: "There is even nowadays not a single molecule of air within a PC flight game, let alone the fluid of it" ...
Rama Posted March 2, 2015 Posted March 2, 2015 If you have 18 out of 20 statements A>B and 3 out of 3 measurements B>A, you can be certain that there's a link missing so that all makes sense. Just going with B>A is unlikely to be correct. Not really. 1) 20 statements is not really enough to be statistically meaningfull 2) A>B is equivalent to A-B>0, if you don't know the error margin of A-B (so the error margin of the 0 value of the statement), then your 3 B>A measurements may very well be totally correct, Inside the uncertainly of your statements. 3) Even if you had 20 ou of 20 statements A>B, if the real error margin (that you don't know) of the 0 value is X, then the result of the statistic is that X/squareroot(20)>ABS(B-A)... so if your 3 measurements B>A and B-A<X/4.5 then your 3 measurements are correct, considering the real comportment of A toward B ... so if you don't know X, you know nothing.... (and since 2 of 20 of your statements are different from the 18 other, you know at least that X is non-negligeable). I supposed that the statement distribution was gaussian, in order to make the explanation short... but if the distribution isn't gaussian, you can also make the same kind of reasoning And worse, if A and B are not perfectly defined, with all condition of A known and indentical to the conditions of B.... then your 20 Statements may be non-comparables (and same for the 3 measurements) Simple example, due to ergonomics, 15kg might be considered heavier than 20kg. If you now model both sets of controls with similar force limits, the model is not properly reflecting reality, even though you went with the numbers. Your example would be usefull if the user/player installation was reflecting the real ergonomics. If it's not, or if you can't have real force feedback in your devices, then it's useless, either for the user/player interface, and of course for the model (who can't model the user/player installation). So yes, a flight sim player is sitting in a chair, in front of a computer screen, and using gaming devices for input, he isn't in a real plane.... I agree with you that's something no model in the world can change. The only solution is to place your user/player in a real cockpit of a professionnal simulator. 1
GOAT-ACEOFACES Posted March 2, 2015 Posted March 2, 2015 (edited) You said: "But it was using fluid dynamic equations to calculate the airflow over the surface of the plane.." (....in ~1995). That was your post That is only part of my post.. Had you not taken what I said out of context.. You would have realized that I agree with you.. For example, here is a full quote of what I said and not a partial out-of-context quote that you 'picked' There are some other 'models' out there.. For example, I think it was over ten years ago, a PC flight sim implemented a 'fluid dynamic' model.. Shoot, that is not the right name for it? But it was using fluid dynamic equations to calculate the airflow over the surface of the plane.. In this case the sim was limited to a stunt plane. As noted, that was over ten years ago, and they did even at that time state they were not fully implementing all aspects of the fluid dynamic equations.. Sadly the game did not do well, and I have not seen any since then try that again. As you can see, I went as far as to point out that they did not fully implement all aspects of the fluid dynamic equations.. Which means they and I understand that trying to do so would bring a PC to it's knees.. Even today! Sadly, you were so excited in thinking you 'found' something I said to be in error, and in such a rush to say reply with ridiculous.. That you didn't even realize that we agree. I say: "There is even nowadays not a single molecule of air within a PC flight game, let alone the fluid of it" ... Are you saying there is no current commercial flight sims on the market that have implemented the fluid dynamic equations as their bases for a flight model? If that is what your saying, than I agree! In that I already stated, I know of no fight sim that has tried that since looking glass tried it over 20 years ago. Hope that helps! Edited March 2, 2015 by ACEOFACES
JtD Posted March 2, 2015 Posted March 2, 2015 Not really. 1) 20 statements is not really enough to be statistically meaningfull 2) A>B is equivalent to A-B>0, if you don't know the error margin of A-B (so the error margin of the 0 value of the statement), then your 3 B>A measurements may very well be totally correct, Inside the uncertainly of your statements. 3) Even if you had 20 ou of 20 statements A>B, if the real error margin (that you don't know) of the 0 value is X, then the result of the statistic is that squareroot(X/20)>ABS(B-A)... so if your 3 measurements B>A and B-A<squareroot(X/20), then your 3 measurements are correct, considering the real comportment of A toward B ... so if you don't know X, you know nothing.... (and since 2 of 20 of your statements are different from the 18 other, you know at least that X is non-negligeable). I supposed that the statement distribution was gaussian, in order to make the explanation short... but if the distribution isn't gaussian, you can also make the same kind of reasoning And worse, if A and B are not perfectly defined, with all condition of A known and indentical to the conditions of B.... then your 20 Statements may be non-comparables (and same for the 3 measurements) If your "not really" really was "not really", you wouldn't need to make a paragraph full of hypothesis' in order to show that there is a chance for both to be true. That chance was already covered by me by using the word "unlikely" as opposed to "impossible".
Rama Posted March 2, 2015 Posted March 2, 2015 If your "not really" really was "not really", you wouldn't need to make a paragraph full of hypothesis' in order to show that there is a chance for both to be true. That chance was already covered by me by using the word "unlikely" as opposed to "impossible". Nope, you don't understand. The paragraph is not full of hypothesis, There's an error margin (sorry, I should say "uncertainty") on everything, so X exist and is different from 0, and it's value is unknown; that's a fact that should be added to your statement. The rest is standard statistics. The "if" Inside are logical if. Moreover, the measurement being different from the statements is not "unlikely", since you don't know anything about the value of "X"... and using your presentation, that 2 statements out of 20 are of the same sign that the measurements, then the measurement are in fact very plausible (but with only 20 statements, even if all 20 were of same sign, as allready said, it would not be statistically robust). Also, the last sentence you quoted is important. In "anecdotal évidences", most of the time, A and B are not well defined, and hardly comparable. So instead of 18 statements A>B, you may very well have a set of A>B;C>D;E>F, etc... (which is the same as saying that X value is of the same order as A and B)
JtD Posted March 2, 2015 Posted March 2, 2015 Yes, I see. You're hell bent on disregarding anecdotal evidence: Nope, you don't understand.That's what Crump usually claims when he's lost, isn't it. It's also usually my point of exit in debates with him, just as much as it is here. Good luck!
Rama Posted March 2, 2015 Posted March 2, 2015 That's what Crump usually claims when he's lost Correct.... but I'm not lost. 2
361fundahl Posted March 2, 2015 Posted March 2, 2015 As long as they take into account dingle arm length and turbo encabulator volume then it's a good sim in my eyes.
Valok Posted March 2, 2015 Posted March 2, 2015 Very interesting topic we have going on here. At any rate, correct FM or wrong FM my verdict is that I'll stick to the 109 since honestly I just don't see the 190 as a true fighter (WOAH, EASY NOW! This is just my noobish and insignificant opinion, please calm down and lower the gun). It is just that the plane can't really "sustain" itself in a fight. You do some passes, maybe some curves and then you better (try to) start running because that Yak will be on your rear faster than a vaselined blob of mercury on a teflon-coated luge circuit. 2
Bearcat Posted March 2, 2015 Posted March 2, 2015 I am really happy to see the new standards for free speech on this forum. One year ago you guys would be banished for much less. I was really important expanding the boundaries and probably the only "ban ace" to be writing in this forum. Voltaire would be proud I don't know about all that .. but just keep it straight now .. and even a "ban ace" like you will have no problems..
Recommended Posts