PDA

View Full Version : Rewritting history



XyZspineZyX
09-05-2003, 05:09 PM
1C did an awesome job on 1.1 patch, now I can outclimb MW50 109s in my humble Yak1BS, built with pride by the soviet worker for the soviet worker.


G10 climbs to 1000m (from 300km/h, 270-280km/h sustained) in 49sec, which means a climb rate of 20.4m/s.
F4 does the same thing in 57sec, for an average climb rate of 17.5m/s.
Yak1b climbs to 1000m in 51sec, average climb 19.6m/s.
La7 -//- in 38sec, average climb 26.3m/s.

Also La7 (2 cannons variant) got an new improvement over already stellar turn performance. From 16sec per 360deg turn now we have 15sec turns (320km/h sustained above the sea). Awesome!

gtg now, I'm busy catching those 109s in my Jug.


<center> http://www.stormbirds.com/images/discussion-main.jpg </center>

XyZspineZyX
09-05-2003, 05:09 PM
1C did an awesome job on 1.1 patch, now I can outclimb MW50 109s in my humble Yak1BS, built with pride by the soviet worker for the soviet worker.


G10 climbs to 1000m (from 300km/h, 270-280km/h sustained) in 49sec, which means a climb rate of 20.4m/s.
F4 does the same thing in 57sec, for an average climb rate of 17.5m/s.
Yak1b climbs to 1000m in 51sec, average climb 19.6m/s.
La7 -//- in 38sec, average climb 26.3m/s.

Also La7 (2 cannons variant) got an new improvement over already stellar turn performance. From 16sec per 360deg turn now we have 15sec turns (320km/h sustained above the sea). Awesome!

gtg now, I'm busy catching those 109s in my Jug.


<center> http://www.stormbirds.com/images/discussion-main.jpg </center>

XyZspineZyX
09-05-2003, 11:36 PM
Did somebody try the same tests?


<center> http://www.stormbirds.com/images/discussion-main.jpg </center>

XyZspineZyX
09-05-2003, 11:40 PM
You are wrong about hte Jug. Period. Get off it. You have enough things to worry about with how the LW is getting screwed (again...imagine that... /i/smilies/16x16_smiley-indifferent.gif ).

XyZspineZyX
09-05-2003, 11:44 PM
Have you tried the same tests DDT? Like 16 sec per 360 deg turns in La7 were not bad enough, now we have 15 sec turns/i/smilies/16x16_smiley-sad.gif


<center> http://www.stormbirds.com/images/discussion-main.jpg </center>

Message Edited on 09/05/0305:47PM by Huckebein_FW

XyZspineZyX
09-06-2003, 12:59 AM
I haven't. It would not surprise me in the least however.

I did get the impression that the G-14 was sluggish, and I've been reading horror stories about all the 109s w/r/t RPM, power, thrust, overheat, and as such, climb, and top speed.

Modeling proper cooling flap drag is fine, but, it sounds like they pulled a few spark plugs out of it and replaced the parts (of the engine) with non-heat treated aluminum.

I don't get it.

Oleg - *if* you read this, can you please explain why the 1.0 109 FM was the way it was, and why these changes have been made? Was it not possible to do it this way originally? Did you feel the 1.0 was correct? If so, why change it? If not, why not change it prior to release? We know you aren't obligated to explain to us, and we know you won't reveal your source info. That's fine. However, we would really appreciate some of the info that went into your decisions and changes. It would reduce a lot of the screaming too. (granted, not all, nothing will ever eliminate it all.)

Kweassa, Oryx, Ugly Kid, et al, what is your take on this. The 1.0 109s climbed like everything I've ever heard about them, and actually behaved like fighters. Proper tactics and engagements were possible. Was it really that far off? Is it really any closer now?

Kweassa, you were a big proponent of increasing E retention and accleration in IL2 for very legit reasons. We seem to be losing that now. Or is it me? Any thoughts on it?

XyZspineZyX
09-06-2003, 01:26 AM
The La-7 easily out turns a Rata on the deck with full power or as required. The Rata can't turn inside it at all, that's plain silly for a plane in this weight class. But that's OK because the Rata can match and exceed it's acceleration for quite a bit, a poor La pilot can't break away. You could say they are an even match in a turn fight./i/smilies/16x16_robot-very-happy.gif

ZG77_Nagual
09-06-2003, 02:07 AM
I never try to turn with those things below 500 kph anyway.
They still dissolve nicely /i/smilies/16x16_smiley-happy.gif

http://pws.chartermi.net/~cmorey/pics/p47janes.jpg

XyZspineZyX
09-06-2003, 07:07 AM
It has to have become clear for some of you by now that regardless of Olegs claim to have inside information on the flyables, there's something very wrong with the way that information is translated into a 3d object flying around inside the game.

I'd dearly love to have a spreadsheet of the changes in performance from IL2 1.0 to FB 1.1f. As you all know, there has been rather a lot of variance in the figures as time has gone by.

And each and every iteration has been 'correct'.

What makes it especially frustrating to try and track these errors down is that Oleg refuses to reveal any sources, or at least, I've never seen him do it. Are we really supposed to believe that every plane in the game uses privately held data that can't be republished, ever? What would stop the public domain stuff being released?

These questions have all been asked of Oleg before and as far as I know there has never been a satisfactory answer.

The burden of proof is always, always, always on the user community for any changes to be made to the FMs. And they're supposed to shoot at changing a complete black box, with no idea what other information Oleg is putting in alongside the submitted data.

That is of course, assuming that the user submitted data gets included at all, and doesn't just get a one line "you is wrong" reply with absolutely no further explanation.

We can go around in circles about this as much as we like, but there is never going to be any kind of definative answer on aircraft performance until the data files in the game themselves are publicly viewable. There is too little trust in the online community for user submitted game tests to be taken seriously, you've all seen what happens when anyone submits a test here - usually one page of questioning the users methods, the next page questions their flying skills, and if you make it to 3 pages without a nationalistic flamewar having been started, it's a good day.

If people really want to see what's going on under the hood, well, Oleg won't tell you.

This is something that can only be determined from looking at the base data and playing with it til an impression is formed as to how the code interprets it.

XyZspineZyX
09-06-2003, 08:18 AM
clint-ruin wrote:
- If people really want to see what's going on under
- the hood, well, Oleg won't tell you.
-
- This is something that can only be determined from
- looking at the base data and playing with it til an
- impression is formed as to how the code interprets
- it.

Got a Beowulf cluster handy? /i/smilies/16x16_smiley-happy.gif


You are absolutely correct of course. I've been pointing this out since IL2 1.04. The Devout won't listen. They can't listen. Prepare to be flamed once they catch wind of your post. /i/smilies/16x16_smiley-indifferent.gif

XyZspineZyX
09-06-2003, 08:27 AM
now oleg is building a new myth!!!my poor 109

XyZspineZyX
09-06-2003, 08:41 AM
BlitzPig_DDT wrote:
- clint-ruin wrote:
-- If people really want to see what's going on under
-- the hood, well, Oleg won't tell you.
--
-- This is something that can only be determined from
-- looking at the base data and playing with it til an
-- impression is formed as to how the code interprets
-- it.
-
- Got a Beowulf cluster handy? /i/smilies/16x16_smiley-happy.gif

I think the difficulty of doing this has been overestimated slightly.

We already have the decrypted data.

It's loaded into memory to play the game.

I made a post about this a day or so ago - I'm a bit rusty with it, but I've played with "easy" hacking and hex editing a little.

This isn't the RC5 challenge - we have access in memory to the algorithm that's used to decrypt the data and the unencrypted data when it's loaded.

If anyone else has played with this I'd love to hear from you. This is something that more just takes time, I think, when you're working from scratch - experimentation and setting up various testable scenarios to see what data gets loaded into memory, then working out where in a file it's from, and what it does.

There was a gentleman who posted here a while ago with his experimentation with VC6 about which .dlls the program accesses and how they're linked together - that stuff is beyond me, since I'm more a network guru than a coder. I'd love to hear from him if he's reading this.

I think the .trk playback definitely helps here, since as far as I know it just records the .mis file and stick input and runs it all off the AI/FMs in combination with the recorded user input.

I'd love to see a Russian attempt at DCMA enforcement after the elcomsoft fiasco :>

- You are absolutely correct of course. I've been
- pointing this out since IL2 1.04. The Devout won't
- listen. They can't listen. Prepare to be flamed once
- they catch wind of your post.

Frankly I could give a crap - even if I did care what an Oleg worshipper thought of me, the level of flaming expertise here is about that of an AOL chatroom. You can just tell that the inability to type swear words really cramps some peoples style :>

XyZspineZyX
09-19-2003, 01:57 PM
What a joke.

XyZspineZyX
09-19-2003, 02:07 PM
there are still people who worry about the fm of fb? come on, nearly everyone knows by now that this whole fl_thing is not to be taken seriously /i/smilies/16x16_smiley-wink.gif

it changes everytime and that in a sometimes drastical way.

il2 or fb were never realistic, but thats ok - at least its just a game that has to fit to different marketing situations... /i/smilies/16x16_smiley-very-happy.gif /i/smilies/16x16_smiley-wink.gif

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



http://homepage.ntlworld.com/paul.bryant3/ETSigGermany.gif




under 30k?

XyZspineZyX
09-19-2003, 04:27 PM
The title of the post assumes an idea that there is an intent upon the producer of this game to simulate history.

I think that is a bit of a stretch.

Propaganda is a better word to describe the result of the game as it effects the general understanding of history.

I have a question for anyone germane to this topic.

Browsing through the local super store, waiting for a prescription to be filled I happened upon a set of die cast toy airplanes.

One was a P-51, another was a BF109E, and can anyone guess what the third plane was?





JG14_Josf

XyZspineZyX
09-19-2003, 04:38 PM
Actually if I recall the main problems at the end of IL2 was that overheat had no effect on the La5fn and the flaps on the FW.

XyZspineZyX
09-19-2003, 05:39 PM
with 1.11 FB is a good Shootergame; funny flying and shooting with little Ufo´s and Hollywood muzzle flashes /i/smilies/16x16_smiley-very-happy.gif

I´d like to have the old IL2 only with the new Planes !

http://mitglied.lycos.de/kubanskiloewe/110missing.jpg

XyZspineZyX
09-19-2003, 05:48 PM
I'd go for the Zero or the Spit, but I don't know what country you're in

http://www.endlager.net/fis/pix/banners/fis_banner_07.gif


She turned me into a newt, but I got better.

XyZspineZyX
09-19-2003, 06:28 PM
Huckebein_FW wrote:
- 1C did an awesome job on 1.1 patch, now I can
- outclimb MW50 109s in my humble Yak1BS, built with
- pride by the soviet worker for the soviet worker.
-
-
- G10 climbs to 1000m (from 300km/h, 270-280km/h
- sustained) in 49sec, which means a climb rate of
- 20.4m/s.
- F4 does the same thing in 57sec, for an average
- climb rate of 17.5m/s.
- Yak1b climbs to 1000m in 51sec, average climb
- 19.6m/s.
- La7 -//- in 38sec, average climb 26.3m/s.
-
- Also La7 (2 cannons variant) got an new improvement
- over already stellar turn performance. From 16sec
- per 360deg turn now we have 15sec turns (320km/h
- sustained above the sea). Awesome!
-
- gtg now, I'm busy catching those 109s in my Jug.


Me is starting to think that this is the way its gonna be just look at the il2 patches changing the FM,s of the non VVS AC like yo yo's and now the same with FB.
What i cant understand is that after several patches the turn rates and climb etc of VVS planes gets better and luftwaffe worse in 1.1b if the luftwaffe AC had their performance increased in the needed areas and the vvs toned down a little this would now be a great balanced game but their u go oleg is oleg.

In just about everysingle patch the VVS AC overperform in just about all areas where as luftwaffe AC underperform in all areas, if oleg wants to show people that he aint biased like he says then he aint doing his self any favours like this.

XyZspineZyX
09-19-2003, 07:04 PM
BlitzPig_DDT wrote:
- Kweassa, Oryx, Ugly Kid, et al, what is your take on
- this. The 1.0 109s climbed like everything I've ever
- heard about them, and actually behaved like
- fighters. Proper tactics and engagements were
- possible. Was it really that far off? Is it really
- any closer now?

I liked G-10 or G-14 a lot in 1.0, now I won't touch them even with a long stick, go and figure /i/smilies/16x16_smiley-happy.gif . I have to say I've been spending more time with FW lately. I just noticed yesterday that Bfs need re-learning. If Il-2 compare speaks anywhere near true language putting P-39N1 and Bf-109G-10 side by side and looking at climb rate at altitude speaks a kind of language. I think problem is partly overheating, partly huge drag from the radiators and partly a shear lack of power. Particularly above let's say 7km it gets really bad, I think that they should not be inferior to G-6 and I think G-6 should not be distinctively inferior to the earlier models. I don't know but I think that those dumb germans did not put 400 HP more
to G-2 airframe only to boil eggs faster /i/smilies/16x16_smiley-happy.gif (On the other hand I do not have an original document or NACA report to proove otherwise /i/smilies/16x16_smiley-happy.gif )

I am partly annoyed by this but on DF servers there are plenty of others to choose from. The only pity is the day they must be flown in VEF or other online wars /i/smilies/16x16_smiley-happy.gif

In my opinion G-10 and G-14 have serious unhistorical shortcomings. As for the complete Bf-series I do not know whether the generally hard times are more because of opponents getting a dosis of steroids or Bfs getting tranquilizers, I think former allthough overheating is back with vengeance.

-------------------------------------
http://people.freenet.de/hausberg/schimpf.gif

XyZspineZyX
09-19-2003, 07:28 PM
My post was written right after patch and yes I was mad.
Now we know that from Bf-109 only half have erroneous performance and those are F4, G6, G10 (I did not test Emil). G6/AS and G14 also perform below their original performance, instead of a 24m/s they barely reach 22m/s, which is G2 and F4 original performance. Keep in mind that early MW50 109s had over 300hp more compared with G2, for only 100kg added weight.

On the other hand we have all major '43 VVS fighter climbing with 19.5m/s instead of 16.5m/s their historical performance at sea level (here you'll have Yak1b, Yak9, Lagg3'43, La5).

Also instantaneous turn and speed loss if completely off for 109s. I made those speed loss tests right above the sea, turning as hard as possible, starting at 500km/h, measuring the time until speed dropped to 350km/h. Some results:

G2 8 sec
G10 10 sec
Yak3 11 sec
P39 17 sec
La7 20 sec

Now, which values are correct? Calculating speed loss it is difficult but doable. You might be surprised but there are little differences in speed loss, which is a decceleration, between fighters produced in the same period of war, even if there are significant difference in acceleration performance. If somebody is interested I can post a calculation of speed loss for a particular plane.

Basically at 500km/h pulling 7G you'll have a speed loss of around 15km/h to sustained speed at 350km/s and 3G (for most late war fighters). To make a long story short the correct values are those for P39 and La7, those for 109 and Yak3 are erroneous (time for speed dropping to 350km/h should be almost double). Speed loss is a very important factor in dogfights, it should be corrected.


<center> http://www.stormbirds.com/images/discussion-main.jpg </center>

XyZspineZyX
09-19-2003, 07:38 PM
I live in California.

The third plane was a Mig 3



JG14_Josf

XyZspineZyX
09-19-2003, 08:14 PM
I tried a slightly different method that seems more accurate to me. It makes sure the automated propeller pitch is fully in action.

I get to the deck (Crimea map to get at sea level), wait till the plane gets 400km/h (TAS) then climb to 1000m.

I get :

32s for the Bf109G10 with 110% MW50 on
37s for the Yak1b with 110%

I use the no cockpit view in order to tell when I reach 1000m (when viewing the track)...

Can you tell us what were the initial conditions of your test ?

I also tried to get those 15s sustained turnrate for 360? that soemone (can't remember who) told a Yak could do. I only get about 20s with a Yak3. If I try to get anything shorter, I loose my speed real quick, not sustained anyway... And turn time increases after a short while.

I would like a track to enlight me and show me how to handle this UFO I heard about, but cannot fly...

Regards

Tym

XyZspineZyX
09-19-2003, 09:32 PM
im not a real life pilot

i just play this computer game

after flying it for nearly half a year i can say with authority that it has changed dramatically since the patch

i can say from game playing experience that German planes seem restricted in performance compared to Russian planes

XyZspineZyX
09-19-2003, 09:36 PM
This the most intelligent thread I've read, all I can say is that I agree entirely.

Tedious unoriginal philoshophical statement the pith of which is lost in repetition.

XyZspineZyX
09-19-2003, 09:50 PM
Huckebein_FW wrote:
- Also instantaneous turn and speed loss if completely
- off for 109s. I made those speed loss tests right
- above the sea, turning as hard as possible, starting
- at 500km/h, measuring the time until speed dropped
- to 350km/h. Some results:
-
- G2 8 sec
- G10 10 sec
- Yak3 11 sec
- P39 17 sec
- La7 20 sec
-

I think it is quite difficult to draw a conclusion. At 500 km/h as hard as possible is namely relative. You should relate this to the turn rate (average or so) that you obtain. My impression from the game is that La-7 does not have a very effective elevator in 500 km/h thus maybe not turning that tightly and maybe not bleeding that much. If indeed the turn rate is also superior you might be up into something.

- Now, which values are correct? Calculating speed
- loss it is difficult but doable. You might be
- surprised but there are little differences in speed
- loss, which is a decceleration, between fighters
- produced in the same period of war, even if there
- are significant difference in acceleration
- performance. If somebody is interested I can post a
- calculation of speed loss for a particular plane.
-
That's pretty broad statement, I think. For example below Spitfire vs. 109

http://people.freenet.de/hausberg/untitled.tif

at 250 mph 5 g turn Spit needs about 16? dive to sustain it and 109 would need 25?. If I am not completely wrong the bleed in this case on level turn (thus not sustained) is therefore 9.81 m/s^2 multiplied by sin of this angle...Spit would momentarily bleed 2.7 m/s^2 and Bf 4.1 m/s^2. This is an example of contemporary fighters and differences are quite noticeable. Of course charts were made by brits /i/smilies/16x16_smiley-wink.gif





-------------------------------------
http://people.freenet.de/hausberg/schimpf.gif

XyZspineZyX
09-19-2003, 10:31 PM
Ugly_Kid wrote:
-
- Huckebein_FW wrote:
-- Also instantaneous turn and speed loss if completely
-- off for 109s. I made those speed loss tests right
-- above the sea, turning as hard as possible, starting
-- at 500km/h, measuring the time until speed dropped
-- to 350km/h. Some results:
--
-- G2 8 sec
-- G10 10 sec
-- Yak3 11 sec
-- P39 17 sec
-- La7 20 sec
--
-
- I think it is quite difficult to draw a conclusion.
- At 500 km/h as hard as possible is namely relative.
- You should relate this to the turn rate (average or
- so) that you obtain. My impression from the game is
- that La-7 does not have a very effective elevator in
- 500 km/h thus maybe not turning that tightly and
- maybe not bleeding that much. If indeed the turn
- rate is also superior you might be up into
- something.


Speed loss is one of those performance characteristics that depend on excess_thrust/weight ratio (like climb and acceleration at any nose pitch angle). Of course Bf-109 that had the best excess_thrust/weight ration in any year of war should have the smallest speed loss compared to other planes. But the differences are not that big, for the speed interval we are talking is around 5 sec (maybe more for P-47). In any way we have here another situation where Bf-109 was the best in reality but is the worst in FB. It is an error though because Yak3 suffers from it too, unlike the other models of Yak family. Most probably is related to the way induced drag is calculated, induced drag is much larger in hard turns than parasitic drag.

Also it's true that La-7, like all russian planes suffered from excessive heaviness of controls at high speeds, but at 500km/h they still are very much manageable. Hard turns begun at this speed does not suffer significantly. Control forces are best reflected in dive limit speed. All late war russian planes had the limit at 650km/h, Spitfire at 720km/h, Bf109E at 750km/h and P51 and P47 at 800km/h.


<center> http://www.stormbirds.com/images/discussion-main.jpg </center>

XyZspineZyX
09-19-2003, 10:43 PM
Yes among the otherthings. You can see from that Spit vs. Bf chart that Spit would have a better angle of straight climb and this results to this and that and so on. (I don't know how correct that chart is)

Anyway the interesting thing is the control heaviness in the game not in the reality. You can look at the IL2 Compare and you will see that La-7 is allegedly superior to both G-2 and Yak-3 all over the speed range in 0m. Yak-3 on the other hand is superior only in the upper end of the speed range...Then on the otherhand P-39 is totally inferior over the complete speed range. Now if this data truly compares to the data from FM it raises an interesting question, doesn't it? But as I wrote you should also relate your measurements to the actual load factor (and thus to lift factor) that you think you're flying and this is a bit difficult without g-meter, isn't it? I would initially think that mass plays also a role in the equation. If being in the situation where drag is greater than thrust (~loosing speed) the deccelerating force should be divided by mass, thus you might want to try what FW will do...



-------------------------------------
http://people.freenet.de/hausberg/schimpf.gif

XyZspineZyX
09-19-2003, 11:13 PM
clint-ruin wrote:
- We already have the decrypted data.
-
- It's loaded into memory to play the game.
-
- I made a post about this a day or so ago - I'm a bit
- rusty with it, but I've played with "easy" hacking
- and hex editing a little.
-
- This isn't the RC5 challenge - we have access in
- memory to the algorithm that's used to decrypt the
- data and the unencrypted data when it's loaded.
-
- If anyone else has played with this I'd love to hear
- from you. This is something that more just takes
- time, I think, when you're working from scratch -
- experimentation and setting up various testable
- scenarios to see what data gets loaded into memory,
- then working out where in a file it's from, and what
- it does.
-
- There was a gentleman who posted here a while ago
- with his experimentation with VC6 about which .dlls
- the program accesses and how they're linked together
- - that stuff is beyond me, since I'm more a network
- guru than a coder. I'd love to hear from him if
- he's reading this.
-
- I think the .trk playback definitely helps here,
- since as far as I know it just records the .mis file
- and stick input and runs it all off the AI/FMs in
- combination with the recorded user input.


I am working on it.

XyZspineZyX
09-20-2003, 12:20 AM
Huckie u is wrong for sure check this LOL:
http://forums.ubi.com/messages/message_view-topic.asp?name=Olegmaddoxreadyroom&id=zuntd&tpage=4