FOCUS TESTING: v1.32.5.6 elevator stick input changes

[DOC]

READ THIS BEFORE TESTING AND ESPECIALLY BEFORE POSTING ON THE SUBJECT

Remember that any comparison testing requires that YOU NEED TO HAVE YOUR JOYSTICK PITCH INPUTS SET TO DEFAULT or you will be damaging the results of the change on the handling of the planes in question. Any "short sticking" you might have used in the past will exaggerate bad traits of the aircraft with the new data in this patch. This applies to testing of the "old" data you might do for comparison, ie: if you fly the "live" verson against this new "test" version for comparison, you must do so with default .cfml stick inputs in both cases (at least in terms of the pitch axis of joystick throw) ... if you fail to do this your results will not correctly indicate the changes for analytical purposes.

If you have been using custom elevator values (pitch axis of joystick .cfml) to cope with these aircraft in the past, from now on you will almost certainly need to remove that configuration and start back at default values ... and maybe even work from there to find a new value to best get the best performance from these aircraft.

The purpose of this change is to make it so that you do not NEED to have custom pitch scaling just to make these aircraft easier to handle like you did in the past. A new player could start out with the default scaling (as they tend to do in most cases) and not struggle like you did before you customized your .cfml files to cope with the handling of the aircraft in question.

If you understand the issue under test here then feel free to add your feedback after comparison testing this aspect of the new release.

[DOC]

bump

[GOPHUR]

Really? No takers?

[gutted]

Do i really have to post it twice?

http://discussions.playnet.com/showthread.php?t=346665

Tested out the 109F4 elevator, felt pretty good in my limited test.

Then i tried the 109G6, and immediately i noticed something was wrong. The elevator trim is out of whack. On a nuetral setting it feels like it has alot of "nose up" trim dialed in.

300kph

nose level

trim neutral

hands off...

nose climbs to 45 degrees.

Almost flopped the thing on takeoff when i gave it my normal "takeoff pressure" the first time LOL. I have to dial in a good amount of nose down trim just to keep it level at 350kph.. that cant be right, its supposed to be the other way around.

Overall it felt good though in the 109F4 & 190. Cant say for the others planes. The testing server FPS was so bad for me i logged out after those few tests.

Edited November 29, 2010 by gutted

[eigenman]

Just DL'd the beta. I never touched my JS pitch settings. Everything default. I'll load it up now and see what's up.

[Slpr]

tested the 109E4/F4/G6 and Fw190 so far. similar in all there is a huge improvement in low speed manouverability. especially in the 09's when they depature,

it is easier to get em straight again if they don't come around by their own. usually you flop and end up in a kinda wobble stall, not such much in the new version.

perhaps you flop over once but with a rudder kick that is easily corrected. the 09's gained amazing turn stability at high AoA's above 14. even the Fw190 now

feels more competitive in a turn fight but that we'll have to see. i already love those improvements after only one hour of testing. perhaps gutted is right and the

default trim of the 09G needs to be looked at, in default the gustav climbs at about 300 to 400kph.

all in one a big plus, the wobble monster seems tamed.

so long slpr

[eigenman]

Tested all 109s. Everything seems good except the G6 seems like it's overly sensitive to up trim pretty much like gutted said. 400kph and G6 just wants to pitch up more than the other 109s. Give it a notch of up trim and it goes up pretty hard. Much harder than all other 109s. Almost like it's CG is a bit further back.

Hard to say if changes are good till actually fighting with it. Just running maneuvers doesn't flesh it out.

I do agree with rote that it is easier to recover after flop.

Edited November 30, 2010 by eigenman

[rote7]

Great work doc, I like the changes so far. I was not really positive about the possibility to change the wobbly planes for the better, but it worked pretty good.

My short tests in the 109s came to the same results as all of the previous posters - with one exception. I think the 109F4 climbs to easy on takeoff, same problem as the G6 to a lesser degree.

More feedback when I had the time to test more.

[DOC]

Perhaps it's easier handling near stall means you can pull it off the ground sooner and hold onto it. It doesn't have more power or less mass so it's climb rate is the same as before, but it feels different at the stick because the elevator authority is changed. Bf199G models were much trickier to dogfight in than previous (particularly F model) 109's and did have a different CG, real LW pilots always remarked they'd rather have kept their F models for dogfighting in because the Gustav wanted to depart more easily.

PS: it's really important that you NEVER exaggerate what you are reporting, even though many people will feel that this lends strength to what they still want to see changed further. The reason is simple.

We DO read what you report and we do act on it whenever we can or it seems appropriate to do so. If you exaggerate a report, we WILL try to replicate what you are reporting, and if we do not find it, we will have to assume we overlooked something and will continue to expand time and resources towards determining what it is and why we didn't find it the first time we tested your observations. If you exaggerated the report in the first place, we won't know that. You may cause other work that needs to be done to be delayed or even not done at all because we are spending that time trying to figure out something that was not happening exactly as reported in the first place because the reporter exaggerated his report due to his strong desire to see that aspect given the attention he thinks it deserves.

You have to be really objective guys, in reporting, and ignore any bias or opinion or "what you think should be done" because this will find it's way into our testing and evaluation but unlike you, we won't know that it wasn't a straight up objective analysis and we will waste resources we need to give you other important stuff ... just trying to figure out why we can't find what the exaggerated report claimed.

I'm not directing this at any individual or even any report so far submitted, so don't get all defensive. It is however a real issue with open test reporting and it costs us a lot of time we could devote to other stuff we want you to have and that you want us to give you. What I write here is not coming from this thread or this feature alone, it comes from thousands of hours testing in closed and open beta tests we have run and the interaction with those who want to help us by "testing stuff" for us.

We're still a small team and every hour we spend on this means you don't get that. We want you to better understand the testing process and how to be a part of it to bring you a better product faster. It is compeltely within your power to cause us to be less able to do that if our attention is directed to places it doesn't need to be.

The testing mindset and process:

abandon all expectations/test/compare/evaluate/thinkaboutit/thinksomemore/then post

Never test something to see if it is what you think it should be. That produces an opinion but not an objective analysis. It also might cause your results to actually not be accurate, and it's very important that you be accurate if the end product is expected to be accurate at all.

[gutted]

So... have you or haven't you checked out the G6 trim? It's pretty jacked up in that build.

I can't tell because your posts are always convoluted

Edited December 3, 2010 by gutted

[DOC]

have you flown the .7 build of the open test beta ?

did you directly compare the old G6 with the new one ?

did you reset your stick configuration to default as outlined in the testing article ?

it's ok to say something is the wrong colour but not if you're wearing coloured lenses

(trim on the beta build you tested was identical to the live build as it was not changed)

[gutted]

My stick settings dont need updating. I never removed any throw on it.

No i haven't tried the latest build, the FPS on the test server was too awful to further test, so i removed it. Maybe the G6 just needs to be re-trimmed for level cruise. 300-350kph, hands off, trims neutral.. the nose climbs high. No other plane i've flown (not even the G6 on the live server) does that.

Edited December 3, 2010 by gutted

[DOC]

so your real meaning was that the Bf109G-6 has the wrong trim values in the live version (and every version previous since it has never been altered) rather than as a result of changes in this beta test

it's important to be clear and not confuse testing criteria, it costs us time we don't have to spend on things that aren't within the criteria of the work we're trying to get finalized for this patch, without undue delay

this is part of the process we have to adhere to, which means reports that adhere to the same testing process during an open beta helps us get stuff to the players faster

[gutted]

bah, edited my post.

Im really trying not to sound like a douche

[gutted]

this is part of the process we have to adhere to' date=' which means reports that adhere to the same testing process during an open beta helps us get stuff to the players faster[/quote']

In all honesty Doc. My first post i reported what i found, and gave the steps to reproduce it. That's how testing is done am i right?

[Slpr]

frames have gotten way better with .7 no change in ac behaviour to .6 tho.

so long slpr

[GOPHUR]

I'm liking these changes. This looks ready to go live.

[beazil]

I know this isn't what you want DOC, but is there any chance of us getting to try this on the live server now that it's intermission?

I know it needs testing. I know I should just round up a bunch of folks and hit the beta server - but it's sometimes like herding cats. (I also realize it's a pew pew fest right now during intermission).

So far, I can't say I've noticed a big change - BUT maneuvering changes lots when someone is shooting at you.

I hope to be on later tonight. Probably ~ 7pm EST if someone wants to hop on the beta server and play to really test out the 109's more. I'll check the two axis and allied regular TS servers to see if anyone is interested - or just pm me here.

Ty.

[dkamerad]

109G6

Spawn in offline

Verify axis mapped to elevator is at default

Engine on

Max/Max Wep On

Bring plane to 300kph at level flight

Release stick

AaKg3omAe4g

*note, I know the first video looks like crap, it's the same res and encoding as the second but youtube is taking it's jolly good time on it.

Result 1.32.4 : G6 needs back pressure on the stick to keep level up to 300kph. On release the G6 trends downwards and crashes

-ys1t0RKlFM

Result 1.32.5.7 : G6 wants to climb, on release continues to climb

Videos are short because I've only got the demo version of Fraps, and take place starting at 200kph IAS.

This effect is also present in the F4 and F2, whose results are very similar but less pronounced.

1.32.4 : Release stick, crash into ground

1.32.5.7 : Release stick, maintains level flight and slowly (but more slowly than the G6) begin to climb.

The F2, F4, and G6 all require forward stick pressure to maintain level flight at and leading up to cruise settings (max 1.0 wep off ~6km)

I didn't test the bell because I've only flied it once or twice and have no feel for it.

If the intent is to have all planes trimmed for level flight at max speed, then whatever the change was has borked these aircraft, probably the others changed too but the effect is less pronounced at low speeds due engine power differences.

[DOC]

I will probably go and retrim all of these aircraft but there won't be time to do that before they go live, unless we wait until after Christmas and that's almost certainly not going to happen. Our deadlines are far too critical to wait that long, we have tons of work in the pipe and we have to clear the branch to get it out.

Until then, I believe it is a fairly minor hassle to have to drop in a few clicks of trim.

Please note neutral trim should not be not at max. speed, it should be at cruise setting.

The changes have not "borked" the aircraft, and they perform properly in terms of their data. What it might have done is alter slightly the default trim for the elevator at it's hands off position. The week or so that will take to rectify, purely for hands off trim without pilot correction (additional trim) will have to be rolled into the next dev. cycle, this one we're in now is pretty much done and overdue to be birthed.

As I said, until then, I believe it is a fairly minor hassle to have to drop in a few clicks of trim to get the neutral state you are used to. You aren't losing any manoeurability or climb or speed but you have gained a wider range of available stick travel within that range and thus a higher level of fidelity. Fine control should be noticably easier to maintain.

[gutted]

Understood doc.

But it's only a minor hassle for those with trim wheels. My CH throttle doesn't have them.. so i use a hat switch on my stick to control it (hold for progressive, tap for incremental). When i want neutral trim i push a button that resets all trims back to neutral.

So hurry up!

[DOC]

My stick has trim wheels, but I use the keymapper for incremental clicks. (keyboard)

Works fine and in a a hundred hours of aerobatics testing none of these changes has been anything but a joy compared to before.

PS: they still aren't Spitfires

[khabanos]

I did some testing and it turned out the entire trim range has changed not only the neutral setting. The difference is dramatic in some aircraft like the F2.

The speeds were measured at 200m level flight:

[font=Lucida Console]AoA/speed                 live version      1.32.5.7
neutral trim    109 e1   -1.8 442 km/h     -1.3 359 km/h  max positive              3.9 174 km/h      7.5 138 km/hmax negative             -4.8              -6.0
                109 e4   -1.9 479 km/h     -1.5 394 km/h                          3.9 177 km/h      7.1 142 km/h                         -4.9              -6.0
                109 f2   -1.6 445 km/h     -0.6 307 km/h                          5.5 169 km/h     12.7 109 km/h                         -5.0              -8.0
                109 f4   -1.6 451 km/h     -0.7 323 km/h                          5.3 160 km/h     11.7 115 km/h                         -5.1              -7.1
                109 g6   -1.2 403 km/h     -0.3 300 km/h                          8.0 148 km/h     13.1 118 km/h                         -6.0              -8.1
                110 c4   -2.1 428 km/h     -2.1 428 km/h                         -0.1 255 km/h     -0.1 255 km/h                         -7.7              -7.7
                110 fb   -2.0 453 km/h     -2.0 449 km/h                          0.1 275 km/h      0.2 270 km/h                         -8.1              -8.1
                190      -1.0 518 km/h     -0.6 435 km/h                          6.2 180 km/h      7.2 167 km/h                         -7.9              -8.6
                ju 87    -1.5 300 km/h     -1.5 300 km/h                          0.1 210 km/h      0.1 210 km/h                         -2.4              -2.4
                he 111   -0.6 325 km/h     -0.6 325 km/h                          1.9 201 km/h      1.9 201 km/h                         -2.1              -2.1
                ju 52    -1.7 350 km/h     -1.7 350 km/h                          0.0 204 km/h      0.0 204 km/h                         -2.8              -2.8[/font]           

The F2's trim is now twice as sensitive as it was before resulting in a loss of precision and smoothness which is very noticable with the small trim wheel I use on the x52.

[DOC]

Trim has to be more sensitive than it was or you end up with the same stick to control surface movement ratio that caused the lack of finesse in regular stick travel outside of the usable range. This is where the "flop" being so hard to manage had it's roots. It's a trade-off that cannot be avoided. By the same token, it isn't more sensitive than the other aircraft, as it was less sensitive previously.

Aircraft performance is not affected, they all do the same top speed/climb/turn profiles they did previously. Achieving the same results may feel different but after flying aerobatics with a few testers it is reccomended that you compare how they can be thrown around to how the older examples could be.

No way you'd go back to the old ones. These are much more forgiving in a dogfight. Perhaps if you recalibrate your set-ups after some extesive testing you'll find that you have a much more dogfighable aircraft. Leaving everything as it was and saying "it's not the same" is not a very good approach to getting the best out of the airframe. Yes changing old habits is hard, and tough on the old vets, but this kind of improvement will generate more new German pilots where this has always been difficult in the past due to the touchy nature of the Bf109 in a dogfight (tight turn) scenario.

[DOC]

The aircraft will be retrimmed (it's a very small change) but that will probably have to wait until after the next release. That work cannot make the deadline unless it is special cased and even then it will be difficult.

You do have, as things stand right now:-

1. More control (joystick movement) of the elevator

2. This required the actual movement limits of the elevator be decreased, as there was movement that extended beyond the control limits of your joystick travel, ie: you were permitted to overcontrol the elevator and depart the aircraft through too much AoA on that control surface before you had run out of stick travel

3. This has nothing to do with historical limts as historically the aircraft was not controlled by a computer joystick and none of the actual physical limtations of that historical reality is modeled on any of the aircraft

4. As of this release, the Bf109s require a few extra clicks of down elevator to trim out level at max. cruise, this will be removed as soon as the next data update can be compiled

5. It does not affect their maximum performance ability, but it will affect the habits you have developed over time and some adaption might be nessesary, that depends on the individual pilot. Most who tested them with me found the better handling at the limit easily exceeded in preferance any "adapting" they had to do to enjoy it.

6. With the next data update the aircraft will all trim perfectly level at max. cruise without adding trim yourself.

This work is now finalized and no further changes are planned or intended outside of the retrim next data patch we can release them in.