Turn rate has tanked with new update

[Hummingbird]

Also just to be clear, the testing we do is always using the ctrl + y infobar for accurate true speed, alt and G readings.

[Victory205]

You need to link or PM the F14 data that you are referencing. 

[fat creason]

1 hour ago, Hummingbird said:

Also just to be clear, the testing we do is always using the ctrl + y infobar for accurate true speed, alt and G readings.

 

Not that I am questioning your statements in any way, but I assume you're hand flying these tests. The only spots you can check on the chart while hand flying is along the Ps=0 line. If you are, are you reporting results after doing all of the following:
 

• hold constant speed +- 2 knots
• roughly hold +-100 ft/min on rate climb (hence the Ps=0 curve)
• maintain altitude within 200-300 ft of target altitude
• have the correct loadout/weight
• have unlimited fuel
• be at the correct altitude MSL
• have standard day conditions (15°C at sea level)
• have the correct flap setting
• stabilize this flight condition for at least 10 seconds before you take any G or turn rate measurement
• have it stabilized at least 4 (or more) times like that and take an average

 

Hand flying this test while adhering to the above requirements for a valid test is extremely difficult (I have to admit I'm not a good enough stick to fly this!). Just want to make sure that everyone is testing in the same manner so a worthwhile discussion can be had. I know there are issues and they're being worked on, but this unfortunately the last in line for the changes since it's the last thing tuned. 

Edited January 12, 2021 by fat creason

[Victory205]

Need all of that, with emphasis on weight, maneuver flaps operating, KIAS instead of Mach for target, and a stable turn for at least a full 360 or more. It is not valid to take a transitory snapshot. I hope your technique isn’t what I saw on the video on this thread.

 

To set this up, start twentfive knots slower than your test point, on altitude and at the correct weight and load out, plug in the burners and carefully roll into your bank as you apply G. Use less G initially than predicted, and increase as the ASI approaches the test velocity, increase G and bank for a level turn, using bank angle to stay level, rolling with very small rudder inputs. Freeze the aft stick and use the lift vector to stabilize at the correct bank for sustained level flight. When you are completely stable, then look at G.

 

If you use too little G or descend you will accelerate beyond target speed, and if you have excess G available, you can pull harder. If you don’t, because you are at an airspeed that already sustains 6.5 or so G’s, then you will have to reduce power to decelerate, and start the process over.

 

You are also burning close to 2000 ppm at 5000 MSL, so your gross is going to be changing rapidly. I have missions set up with reduced fuel that puts me at a charted gross weight, starting at the correct altitudes.

 

You guys keep saying the performance is off, but you have not provided the test target points and results that you obtained, on what aircraft, load out or weight, much less the charts themselves. All we have is a video, that’s done at sea level and never stabilized, an altitude for which there are not charts in the performance data publications that I have.

 

I’m going to look at the F14B tonight and see what gives.

[Hummingbird]

56 minutes ago, fat creason said:

 

Not that I am questioning your statements in any way, but I assume you're hand flying these tests. The only spots you can check on the chart while hand flying is along the Ps=0 line. If you are, are you reporting results after doing all of the following:
 

• hold constant speed +- 2 knots
• roughly hold +-100 ft/min on rate climb (hence the Ps=0 curve)
• maintain altitude within 200-300 ft of target altitude
• have the correct loadout/weight
• have unlimited fuel
• be at the correct altitude MSL
• have standard day conditions (15°C at sea level)
• have the correct flap setting
• stabilize this flight condition for at least 10 seconds before you take any G or turn rate measurement
• have it stabilized at least 4 (or more) times like that and take an average

 

Hand flying this test while adhering to the above requirements for a valid test is extremely difficult (I have to admit I'm not a good enough stick to fly this!). Just want to make sure that everyone is testing in the same manner so a worthwhile discussion can be had. I know there are issues and they're being worked on, but this unfortunately the last in line for the changes since it's the last thing tuned. 

 

 

Absolutely 100%, every single thing you can check mark.

 

Also using KTAS to check vs TMN.

 

That said regarding flap setting, everything is ofcourse in AUTO, and as I've noted earlier in a bug report the maneuver devices don't actually start deploying at 0.58 mach as they should, instead they stay in until speed drops to 0.51 mach. So that's an issue.

 

56 minutes ago, Victory205 said:

You need to link or PM the F14 data that you are referencing. 

 

HB already have it, but I'll PM it to you.

 

According to the charts the Ps=0 at 5 kft should be:

M 0.30 = 2.30 G

M 0.35 = 2.90 G

M 0.40 = 3.45 G

M 0.45 = 4.15 G

M 0.50 = 4.80 G

M 0.55 = 5.10 G

M 0.60 = 5.70 G

M 0.65 = 5.85 G

M 0.70 = 6.40 G

M 0.80 = 7.00 G

M 0.90 = 6.80 G

 

This is with the 4 x AIM9 + 4 x AIM7 load out at 55,620 lbs.

 

Source: 

Pages XI-9-4, XI-9-5, XI-9-6, XI-9-29 & XI-9-57 in the F-14B performance manual.

 

So those are the load factors you are supposed to be able to hold, but I am consistenly ending up quite a bit below that, esp. near 0.5 to 0.6 mach, and the issue is worse at SL it seems. 

Edited January 13, 2021 by Hummingbird

[Hummingbird]

51 minutes ago, Victory205 said:

You are also burning close to 2000 ppm at 5000 MSL, so your gross is going to be changing rapidly. I have missions set up with reduced fuel that puts me at a charted gross weight, starting at the correct altitudes.

 

We always test with correct fuel load and unlimited fuel set to on, otherwise weight will ofcourse change with time and thus ruining the accuracy of the results.

 

If you're not using unlimited fuel option, then that would probably explain why you're getting the feeling the a/c is overperforming as it gets lighter and lighter during the time needed to settle into a stable Ps=0 turn, which can take quite some time and thus once you get a reading you're already much lighter than on the chart you're comparing with.

Edited January 13, 2021 by Hummingbird

[captain_dalan]

3 hours ago, Hummingbird said:

 

I really wish you would've read the whole thing, because as I said "the version that performs correctly performance wise", which the old one did at least in terms of STR in the subsonic region (to within 0.1 to 0.05 G infact), something we tested exhaustively, and do so for pretty much all the aircraft  - DCS F-15 for example is spot on the charts for the entire speed range: 

 

Link to verify it is legally and publically available (https://www.eflightmanuals.com/ITEM_EFM/SITEM_EFM.asp?cID=3778)

 

 

 

Dude, don't! Your posting privileges aren't worth risking over this. If some people want to dismiss your tests, let them. The tracks are there, the reports have been made and data doesn't lie. We have done our best. As is, ACM isn't especially recommended activity right now. It will probably take a while before it becomes. I've shelved the plane for the time being, except for WCS and Missile tests. As soon as i complete those, it will serve mostly as a formation flier and/or airshow bird.

The alternative is of course to revert to an older version. Some servers may have done it, as i can no longer find them and i run the latest patch. Anyways, ACM in A and B will have to wait. At least against competent competition. Cheers, stay safe and stay healthy!

[Victory205]

17 hours ago, Hummingbird said:

 

We always test with correct fuel load and unlimited fuel set to on, otherwise weight will ofcourse change with time and thus ruining the accuracy of the results.

 

If you're not using unlimited fuel option, then that would probably explain why you're getting the feeling the a/c is overperforming as it gets lighter and lighter during the time needed to settle into a stable Ps=0 turn, which can take quite some time and thus once you get a reading you're already much lighter than on the chart you're comparing with.

 

 

I start with fuel that is a two thousand pounds over target to allow for burn time to settle out, if I can’t get stable, I simply reset or work on something else for a few minutes.

 

Good news is that we’re on the same page.

 

I looked at the TMN vs KIAS and calculated it out for a standard day at 5000 MSL to ensure that we weren’t missing something in a temperature or position error, but it works out to within four knots vs chart. To make it straightforward, I used .5 .6 and .7 Mach on the chart and correlated that with KIAS (careful that the TAS doesn’t get switched on on the info bar). So your corresponding targets look as good as you can ascertain reading the G curves on the data at 4.8, 5.7 and 6.4 from your post above. Good work there, it’s tedious to plot that out.

 

So using the Open Beta release I got a steady state G of 4.7 against 4.8 at .5 TMN (305 KIAS); 5.7-5.8 G at .6 TMN (365 KIAS) vs 5.7, and 6.8 G vs 6.4 at .7 TMN (430 KIAS).

 

As the velocities increase, it becomes more difficult to stabilize since the aircraft wants to accelerate through the target speed, which is exactly what my bud described during air show performances. If you don’t get the G on, the aircraft will accel right through the corner velocity and head towards 459 plus. The F14B is a beast.

 

Things to look for are ensuring a stable G, altitude and airspeed, including ensuring that the wings stay in auto and the maneuver devices in auto. If the bank angle is slightly off, then the aircraft will descend or climb, and you’ll see an increase or decrease in G for a given AOA, exactly as the Ps figures show on the chart. 

 

It must be stable, you can’t just snapshot a transitory state while the aircraft is accelerating or decelerating, or climbing or descending.

 

This is very similar to what I found on the F14A, it was slightly underperforming at low speeds and over performing at higher velocities.

 

It’s all going to change very soon, but it was interesting to examine. Based on my objective data, which was obtained independently before I looked at the charted targets, there is no reason not to go fly the hell out of the sim aircraft as it sits today.

 

 

Edited January 13, 2021 by Victory205

[SgtPappy]

Good work, Hummingbird, Captain_Dalan and Victory205! 

[Hummingbird]

Will be running a full set of tests at SL, 5 kft and 10 kft this weekend and then report back the results.

 

Conditions will be as always: 

- Std. Atmosphere, 15 C @ SL 

- Zero wind

- Unlimited fuel on

- Load out 4xAIM9 + 4xAIM7, 50% fuel (55,620 lbs total weight)

 

 

[Victory205]

If you are going to spend the time, then a better approach would be to test at 5K and at least 15K to get some separation in performance. SL is a waste, there are not charts for that for obvious reasons and you’d be interpolating. 

 

Couple things to keep in mind. The Ps charts are developed by flight test maneuvers including acceleration profiles and turn derivatives including windup turns and spot checks, and are largely mathematically derived. They aren't accurate to tenths of G's through the entire spectrum. The whole idea of Ps comparison is to give a pilot general idea of performance comparison in order to better understand how to approach an engagement in a methodical manner. Most folks focus on the Ps=0 line, but the bleed rates are also very important for developing rules of thumb. In the end, it's a tool, not the burning bush for determining which aircraft wins a fight.

 

I’ve moved on to a different build and won’t be devoting any further time to the current release.

Edited January 15, 2021 by Victory205

[maxsin72]

10 hours ago, Hummingbird said:

Will be running a full set of tests at SL, 5 kft and 10 kft this weekend and then report back the results.

 

Conditions will be as always: 

- Std. Atmosphere, 15 C @ SL 

- Zero wind

- Unlimited fuel on

- Load out 4xAIM9 + 4xAIM7, 50% fuel (55,620 lbs total weight)

 

 

Very good job, i'm really interested to see your results when ready. 

Edited January 15, 2021 by maxsin72

[Hummingbird]

7 hours ago, Victory205 said:

If you are going to spend the time, then a better approach would be to test at 5K and at least 15K to get some separation in performance. SL is a waste, there are not charts for that for obvious reasons and you’d be interpolating.

 

Not really, it's easy to calculate what the performance would be when we have the 5 kft, 10 kft and 15 kft for reference. Only thing that would make it problematic was if the engines somehow didn't behave as predicted at SL vs 5 kft, but since we have performance vs alt charts for the GE engines, that isn't an issue and we know there's no unpredictable change to performance at SL, which would also have been very odd.

 

In other words the performance to be expected as SL and targets are:

 

Est. F-14 perf @ Sea Level, based on the real life 5 kft, 10 kft & 15 kft figures as well as engine perf charts

M 0.3 = 2.85-2.90 G

M 0.4 = 4.25-4.30 G

M 0.5 = 5.65-5.70 G

M 0.6 = 6.50-6.55 G

M 0.7 = 7.40-7.45 G

M 0.8 = 8.20-8.25 G

 

Edited January 15, 2021 by Hummingbird

[captain_dalan]

19 hours ago, Hummingbird said:

Will be running a full set of tests at SL, 5 kft and 10 kft this weekend and then report back the results.

 

Conditions will be as always: 

- Std. Atmosphere, 15 C @ SL 

- Zero wind

- Unlimited fuel on

- Load out 4xAIM9 + 4xAIM7, 50% fuel (55,620 lbs total weight)

 

 

Will you be testing both the A and the B?

10 hours ago, Victory205 said:

If you are going to spend the time, then a better approach would be to test at 5K and at least 15K to get some separation in performance. SL is a waste, there are not charts for that for obvious reasons and you’d be interpolating. 

 

 

 

XI-9-4 to XI-9-6 provide  6 checkpoints without need for any calculation.

EDIT: besides, the difference seams to rise as the atmosphere density goes up - altitude goes down. Makes me think it's somehow drag induced. It's highest at sea level, 0.3-0.5 g and then drops slowly as you go further up high. 

Edited January 16, 2021 by captain_dalan

[Hummingbird]

Yes, the difference is largest at SL. Also we absolutely need to test at SL, we'd be doing everyone a great disservice if we didn't, as performance at SL ofcourse also has to be realistic.

 

For example it wouldn't make any sense if the F-14 suddenly lost any advantage in STR over the F-15 at the speeds where it holds it at 5 kft. Infact at SL the difference should be slightly larger in terms of sustainable load factor at the speeds where F-14 holds the STR advantage, with it gradually decreasing as altitude increases and available G's to both decrease.

 

4 hours ago, captain_dalan said:

Will you be testing both the A and the B?

 

I will try to make the A as well, I'm a bit limited on time though, and these tests take several hours to do due to how meticulous we have to be when carrying them out. So I might do the B this weekend and then the A next weekend.

Edited January 16, 2021 by Hummingbird

[Hummingbird]

4 hours ago, captain_dalan said:

XI-9-4 to XI-9-6 provide  6 checkpoints without need for any calculation.

 

Yup, those are easy benchmarks to check.

Edited January 16, 2021 by Hummingbird

[captain_dalan]

9 hours ago, Hummingbird said:

 

 

I will try to make the A as well, I'm a bit limited on time though, and these tests take several hours to do due to how meticulous we have to be when carrying them out. So I might do the B this weekend and then the A next weekend.

 

Roger that! Clear skies and happy flying!

[Hummingbird]

First round of results, these are for 5 kft, and they are interesting:

 

ISA, 55,620 lbs, 5,000 ft:

 

F-14B DCS vs RL

TMN 0.35 / KTAS 227 = ~2.70 vs 2.80  (-0.10 G)
TMN 0.40 / KTAS 260 = ~3.30 vs 3.40  (-0.10 G)
TMN 0.45 / KTAS 292 =  3.95 vs 4.20 (-0.25 G)
TMN 0.50 / KTAS 325 = 4.50 vs 4.80 (-0.30 G)
TMN 0.55 / KTAS 355 = 5.00 vs 5.40 (-0.40 G)
TMN 0.60 / KTAS 390 = 5.45 vs 5.65 (-0.20 G)
TMN 0.65 / KTAS 422 = 5.90 vs 5.90 (-0.00 G)
TMN 0.70 / KTAS 455 = 6.40 vs 6.40 (-0.00 G)
TMN 0.75 / KTAS 487 = 7.10 vs 6.80 (+0.30 G)
TMN 0.80 / KTAS  520 = 7.90 vs 7.00 (+0.90 G)
 

So we got some bad underperforming right at the speeds where the F-14 should shine, and some pretty extreme overperforming where the F-14 should actually start to peter off in performance (above M 0.75).

 

Now this actually took 4 hours to do, due to wanting the most accurate results, so fell abit short on time to complete a full test at other altitudes. Will try to complete the tests at 10 kft & 15 kft for next weekend, and then I'll wait with SL till last as that's the easiest one to do.

Edited January 18, 2021 by Hummingbird

[ustio]

so if i read that correctly, from 0.4 - 0.6 is underperforming, while from 0.75 and beyond that is overperforming.?

Edited January 18, 2021 by ustio

[Hummingbird]

5 minutes ago, ustio said:

so if i read that correctly, from 0.4 - 0.6 is underperforming, while from 0.75 and beyond that is overperforming.?

 

 

Yes, basically.

[draconus]

3 hours ago, Hummingbird said:

Now this actually took 4 hours to do, due to wanting the most accurate results, so fell abit short on time to complete a full test at other altitudes. Will try to complete the tests at 10 kft & 15 kft for next weekend, and then I'll wait with SL till last as that's the easiest one to do.

With possible update next week (as we heard they already made some changes) I doubt it is worth doing now, don't you think? And Victory205 already got similar results before you tested: "This is very similar to what I found on the F14A, it was slightly underperforming at low speeds and over performing at higher velocities."

Edited January 18, 2021 by draconus

[fat creason]

On 1/18/2021 at 5:56 AM, Hummingbird said:

Now this actually took 4 hours to do, due to wanting the most accurate results, so fell abit short on time to complete a full test at other altitudes.

 

Now does everyone believe me when I say that working on this type of stuff takes forever? 4 hours to test only up to Mach 0.8 at ONE altitude. Then one has to re-test AGAIN after making changes, analyze test results, make changes, then test AGAIN. Changes can take a long time. These changes are still WIP, do not assume they will show up in the next patch unless I specifically say the changes are ready.

Edited February 12, 2021 by Cobra847

[Hummingbird]

32 minutes ago, fat creason said:

 

Now does everyone believe me when I say that working on this type of stuff takes forever? 4 hours to test only up to Mach 0.8 at ONE altitude. Then one has to re-test AGAIN after making changes, analyze test results, make changes, then test AGAIN. When I can only devote a couple hours on a weekend to this, changes can take a long time. These changes are still WIP so don't expect them in the next patch.

 

Oh I always knew that, I was however under the impression that either:

a) You had a scripted autopilot which could test this stuff super accurate or b) that you had a bunch of test pilots who's finest job was to test this after every patch

 

The FM is by far the most important of any module (IMHO atleast), so hence I'd always place it as priority no.1 to get just right.

 

Also please take this as constructive criticism, not a hetz, as like I've said fsince the launch of this module, you did a fantastic job with the F-14 FM, it's one of, if not THE most believable flying module I've ever had the pleasure of flying. In short I'm a fan who desperately wants to help, not a mindless critic.

Edited January 18, 2021 by Hummingbird

[fat creason]

Just now, Hummingbird said:

 

Oh I always knew that, I was however under the impression that either:

a) You had a scripted autopilot which could test this stuff super accurate or b) that you had a bunch of test pilots who's finest job was to test this after every patch

 

The FM is by far the most important of any module (IMHO atleast), so hence I'd always place it as priority no.1 to get just right.

 

Also please take this as constructive criticism, not a hetz, as like I've said from the beginning you did a fantastic job with the F-14 FM, it's one of, if not THE most believable flying module I've ever had the pleasure of flying. In short I'm fan who desperately wants to help, not a mindless critic.

 

I agree with you and no offense taken. Up to this point a large majority of the time spent on the FM has been working on the handling aspects and not as much on the performance, since handling in this case is somewhat subjective (no handling data/tests available) and we wanted to focus on that with our SME. Performance numbers are a bit more objective and don't require as much SME input. ED does not provide any FM tools so any type of automated testing needs to be developed by each 3rd party individually. Testing via test pilots is also possible, but everyone is on a volunteer basis and testing these specific maneuvers takes large amounts of time and skill, so it's easy to have test coverage gaps.

[Hummingbird]

3 minutes ago, fat creason said:

 

I agree with you and no offense taken. Up to this point a large majority of the time spent on the FM has been working on the handling aspects and not as much on the performance, since handling in this case is somewhat subjective (no handling data/tests available) and we wanted to focus on that with our SME. Performance numbers are a bit more objective and don't require as much SME input. ED does not provide any FM tools so any type of automated testing needs to be developed by each 3rd party individually. Testing via test pilots is also possible, but everyone is on a volunteer basis and testing these specific maneuvers takes large amounts of time and skill, so it's easy to have test coverage gaps.

 

Understandable, and I know about the last bit all too well having dedicated large amounts of my own time to test fly most modules in DCS after every major patch (biggest hobby as I never got to chase the test pilot dream IRL, my hearing simply wasn't good enough), it is anything but easy and takes a lot of time to be able to finish these tests accurately and consistently, even with a lot of experience. Hence why I was sure you probably had a bot to do this for you, as the amount of time you'd have to spend on it otherwise didn't seem reasonable in my mind considering all the adjustments & changes that goes into fine tuning an FM, and then the cascading effect just one minute adjustment can have.

 

That said, just knowing you are dedicated to getting it spot on is actually enough to provide many of us with extra patience, and we know you can do it, because you were spot on or very close to it in the subsonic region for a long time.