Feedback Thread - F-14 Tomcat patch Jan 27th 2022

[lunaticfringe]

8 hours ago, UWBuRn said:

With all the respect for your competence and the work you're putting into it, it's hard to belive it's a guidance issue. Guidance was an issue also before, FM just seems to be off now.

Btw, any news on moving to the new API?

 

The stalling loft illustrated in this thread undermines your contention that it isn't guidance, because it's proof the guidance logic doesn't see the kinematic change in the weapon. 

[cgstever91]

not really more than one thing can be broken at once  

[UWBuRn]

2 hours ago, lunaticfringe said:

 

The stalling loft illustrated in this thread undermines your contention that it isn't guidance, because it's proof the guidance logic doesn't see the kinematic change in the weapon. 

I've been taking a couple of shots in a DCS Liberation mission with 54C without loft being involved that showed very poor performance - no real test, just looked for a while at the missile in F6 and it had 500-600 kts lass than usual for that kind of shot while approaching target.

Is guidance part of the problem? No doubt about it, but if that wasn't touched substantially then it's hard to believe it's the main culprit. Let's just say it's broken altogether... it just makes me raise an eyebrow on why those changes were rolled out in this state.

[IronMike]

2 hours ago, UWBuRn said:

I've been taking a couple of shots in a DCS Liberation mission with 54C without loft being involved that showed very poor performance - no real test, just looked for a while at the missile in F6 and it had 500-600 kts lass than usual for that kind of shot while approaching target.

Is guidance part of the problem? No doubt about it, but if that wasn't touched substantially then it's hard to believe it's the main culprit. Let's just say it's broken altogether... it just makes me raise an eyebrow on why those changes were rolled out in this state.

It was grossly over-performing before, and if you took a straight shot, and it was performing less than before, it is likely that it was performing as it should. Folks got used to the over-performance, so it is natural that eyebrows are being raised at the closer matching performance now. What causes the issues is neither the FM, nor CFD, it is the guidance. It has been rolled out in this state, because ED requested it so that the CFD data would get transferred to the new FM. We are now at a point and required to be at a point, where we can focus on guidance, rather than guessing FM states and tweaks to match real life shots. If you look at the FM data, it is, in fact, still over-performing ever so slightly, which is ok however, since the CFD has to be taken conservatively.

The guidance did not change, it is the same guidance as before. But now it shows much clearer how it affects the overall performance and it has been rolled out, because one part of moving it to the new API is done and we can move to the second part with much clearer test results - which also needs community feedback beyond in-house testing. The missile is not useless at all, it just gives you a bit of a shorter range for the time being.

 

[UWBuRn]

45 minutes ago, IronMike said:

It was grossly over-performing before, and if you took a straight shot, and it was performing less than before, it is likely that it was performing as it should. Folks got used to the over-performance, so it is natural that eyebrows are being raised at the closer matching performance now. What causes the issues is neither the FM, nor CFD, it is the guidance. It has been rolled out in this state, because ED requested it so that the CFD data would get transferred to the new FM. We are now at a point and required to be at a point, where we can focus on guidance, rather than guessing FM states and tweaks to match real life shots. If you look at the FM data, it is, in fact, still over-performing ever so slightly, which is ok however, since the CFD has to be taken conservatively.

The guidance did not change, it is the same guidance as before. But now it shows much clearer how it affects the overall performance and it has been rolled out, because one part of moving it to the new API is done and we can move to the second part with much clearer test results - which also needs community feedback beyond in-house testing. The missile is not useless at all, it just gives you a bit of a shorter range for the time being.

 

If, as you say, this was requested by ED, than it's more on their side, but the point still stands: from an user point of view i can't really get the reason why incomplete changes (FM adjustments without guidance) are rolled out, especially on an hot topic like missiles (see what happened on AIM-120s few patches ago).

Personally, flying mostly offline i care up to a certain point and for sure i'm not going to loose sleep about it, but yet i think some criticism is needed.

I say what i'm saying knowing that in the end everything will just be better than before, that you will work hard to fix the guidance just like a lot of issues before this. And please know that i appreciate it, i know maintaining this kind of product in the long run it' not easy.

[IronMike]

25 minutes ago, UWBuRn said:

If, as you say, this was requested by ED, than it's more on their side, but the point still stands: from an user point of view i can't really get the reason why incomplete changes (FM adjustments without guidance) are rolled out, especially on an hot topic like missiles (see what happened on AIM-120s few patches ago).

Personally, flying mostly offline i care up to a certain point and for sure i'm not going to loose sleep about it, but yet i think some criticism is needed.

I say what i'm saying knowing that in the end everything will just be better than before, that you will work hard to fix the guidance just like a lot of issues before this. And please know that i appreciate it, i know maintaining this kind of product in the long run it' not easy.

This isn't a question of whose blame or on whose side it is. Please keep in mind - the Open Beta exists for precisely this reason. We need broughter community feedback on several issues, and we need to push certain things along which cannot be tested only in a confined branch, also for practicality reasons. This is exactly why a stable and an open beta version of DCS exist, so that we are able to push steps of a multi-step process. We strongly recommend to use the stable branch, especially when only flying Singleplayer, if such issues want to be avoided, which we understand completely. But we needed a broughter testbed on this, we need the community feedback, and we are half way there. Let us not forget please that by using the open beta, everyone subscribes to these kind of processes voluntarily. 

Thank you for your kind understanding and for your encouraging words!

[DoorMouse]

4 hours ago, IronMike said:

The missile is not useless at all, it just gives you a bit of a shorter range for the time being.

To be honest, it is really quite bafflingly inconsistent right now. I'll refrain from calling it useless, but it's maybe slightly better than useless.

Im compiling some examples, but it is overtaken by an aim7 or even the firing aircraft against targets inside 10 miles sometimes... You will see it often wobble back and forth like it can't decide what path it wants to take to the target- this is primarily during PSTT launches. Something is definitely wrong with the guidance. 

TWS is definitely incredibly reduced but 30-40 mile kills are still possible, but I Don't see how it could be performing any of the test shots the HB had listed as benchmarks. I can't say with 100% confidence, but I'm pretty sure the Phoenix should out range the aim120 a/b/c5?

My concern is that guidance is out of your hands and that means we are all at the whim of a third party who has many irons in the fire already to fix it.... Who knows when ED gets around to it. It's also the case that you guys have had moving this over to the new API as a priority on the public roadmap for a long time, so I presume there are still things ED has yet to provide you with in order to do it... Or it's just simply not possible yet.

Could you set any expectations as to when we may see a fix on this... While it's not "useless" it is absolutely hobbled. 

Edited January 28, 2022 by DoorMouse

[tae.]

Great to see more realistic missile behaviour being implemented in DCS, it is always welcome.

I ran ~40 tests last night and came to some interesting conclusions that I am not sure are helpful or not.

The first conclusion is that the new flight model does not necessarily equate to a reduction in range or effectiveness, particularly regarding shots taken at higher altitudes. The longer the range the missile is fired at, the more time it spends in extremely thin air up at angels 70+, allowing it to conserve huge amounts of energy. Furthermore, the steeper dive caused by the extended loft allows it to retain even more energy.

My second conclusion, related to the first, is that this results in an "effectiveness plateau" for mid-range shots (30-60nm), where shooting at a longer range may have actually made an equally deadly or in some edge cases MORE deadly shot. There is a certain point in the range that changes depending on launch + target aircraft flight parameters that I will refer to as the "burn through point". Passing this burn through point (usually between 25-40nm at higher altitudes) is where the missile effectiveness starts to break away from this plateau and steeply increase again as the flight parameters reach a point where the rocket motor is burning for the majority of the missile trajectory, offsetting the difficulties encountered by longer range shots.

Please refer to my mid_range_mk47 and long_range_mk47 tacviews, both missiles were shot under similar parameters, one at 40nm and one at 60nm - as per my previous point these missiles perform *very* similarly despite the big difference in range - I think primarily due to the time that the 60nm shot can spend higher up, and the steeper dive angle. Note the drastic reduction in loft angle on the 40nm shot as well.

When using the Mk60 motor, you can exacerbate this behaviour to extreme degrees - see 75nm_mk60 tacview as an example. Pay close attention to how well the missile retains energy at 87400 feet (its peak altitude) and how much energy is retained on the dive due to how steep the angle is (assisted by gravity). Making similar shots at a lesser, medium range all the way down to the "burn through point" I mentioned, I found typically results in very similar missile performance, sometimes even worse missile performance, due to the less effective loft trajectories at lower ranges.

To conclude my ramblings: It is not for me to say whether this behaviour is desired, or realistic, these are simply observations that I have made. In my opinion, the missile can still be coerced into performing extremely well against a target flying purely hot if fired under the right conditions. From what I read about the missile basically being "useless" compared to my tests I was scratching my head a little bit. The missile seems to suffer badly at lower altitudes but there are still some crazy shots that can be made. I think that the part that appears to be hard to grasp is that unlike other missiles in DCS where the effectiveness increases linearly as you improve shot parameters (reduction in range, increase in speed and altitude), this linearity is not necessarily true for the Phoenix in it's current state, at certain intermediary ranges.

edit: forgot to add that I did one last test for a laugh - target changed to MiG-31, sped up the F-14 to Mach 1.7 and raised it's altitude a little..and was able to achieve a 95nm kill with a Mk60. The shot had some very deadly energy in the terminal phase too, so this could probably be stretched even further. I've attached the tacview of that one too. Not very realistic to replicate outside of testing, but just goes to show some absolutely wild shots are technically still possible.

75nm_mk60.acmi mid_range_mk47.acmi long_range_mk47.acmi

95nm_mk60.acmi

Edited January 28, 2022 by tae.

[Krippz]

27 minutes ago, DoorMouse said:

To be honest, it is really quite bafflingly inconsistent right now. I'll refrain from calling it useless, but it's maybe slightly better than useless.

Im compiling some examples, but it is overtaken by an aim7 or even the firing aircraft against targets inside 10 miles sometimes... You will see it often wobble back and forth like it can't decide what path it wants to take to the target- this is primarily during PSTT launches. Something is definitely wrong with the guidance. 

TWS is definitely incredibly reduced but 30-40 mile kills are still possible, but I Don't see how it could be performing any of the test shots the HB had listed as benchmarks. I can't say with 100% confidence, but I'm pretty sure the Phoenix should out range the aim120 a/b/c5?

My concern is that guidance is out of your hands and that means we are all at the whim of a third party who has many irons in the fire already to fix it.... Who knows when ED gets around to it. It's also the case that you guys have had moving this over to the new API as a priority on the public roadmap for a long time, so I presume there are still things ED has yet to provide you with in order to do it... Or it's just simply not possible yet.

Could you set any expectations as to when we may see a fix on this... While it's not "useless" it is absolutely hobbled. 

 

They said in the thread that the guidance issue is getting looked at. I'm sure it's a high priority to get fixed; just have to play the waiting game. 

[near_blind]

12 minutes ago, DoorMouse said:

I can't say with 100% confidence, but I'm pretty sure the Phoenix should out range the aim120 a/b/c5?

Context is key. Up high where you are reducing drag as a factor until the end game, allowing the missile to loft up and leverage its greater weight on the downward portion? Yes, I would expect it to out range an AMRAAM.

Down low where it to bull through the soupmosphere with its greater mass and diameter? Speaking of cuff without testing, I expect the AIM-120 to get where it's going faster and maintain its speed better doing it, which probably means it's going to go further.

With this FM the AIM-54 is going to suffer flying low and or flat, the lower and flatter, the worse the pain. The Phoenix seems to prefer falling to gliding.

 

That's not to say there aren't indisputable guidance issues that need to be solved. 

[IronMike]

58 minutes ago, DoorMouse said:

Who knows when ED gets around to it. It's also the case that you guys have had moving this over to the new API as a priority on the public roadmap for a long time, so I presume there are still things ED has yet to provide you with in order to do it... Or it's just simply not possible yet.

Could you set any expectations as to when we may see a fix on this... While it's not "useless" it is absolutely hobbled. 

 

I can put your mind at rest, ED already has provided further fixes for this very issue as we speak and we are both working hard to get it resolved asap. It has absolutely top priority for us to get it done, preferably ahead of the next patch.

The complete move to the new API guidance will take a bit longer though, because ED is not done on their side of things yet and understandably so does not share half-finished stuff. Keep in mind: they intend on making missiles better across the bord, and the aim54 is only 1 variable in the larger equation. In the meantime we are confident that we can get it back to a much more usable state, and certainly will not stop working on it until that has been achieved, with the complete move to the new API or without. ED is on board and ED is extremely helpful with these issues and we are very grateful for that.

Some may call these remarks "corporate speak", but I can assure it isn't. It is simply an honest relay of what is happening in the background, without getting into too much unnecessary detail.

Edited January 28, 2022 by IronMike

[tae.]

2 minutes ago, IronMike said:

I can put your mind at rest, ED already has provided further fixes for this very issue as we speak and we are both working hard to get it resolved asap. It has absolutely top priority for us to get it done, preferably ahead of the next patch.

The complete move to the new API guidance will take a bit longer though, because ED is not done on their side of things yet and understandably so does not share half-finished stuff. Keep in mind: they intend on making missiles better across the bord, and the aim54 is only 1 variable in the larger equation. In the meantime we are confident that we can get it back to a much more usable state, and certainly will not stop working on it until that has been achieved, with the complete move to the new API or without. ED is on board and ED is extremely helpful with these issues and we are very grateful for that.

Some may call these remarks "corporate speak", but I can assure it isn't. It is simply an honest relay of what is happening in the background, without getting into too much unnecessary detail.

 

The openness and communication is not lost on us and is greatly appreciated. As various people in the thread have pointed out it is not like the missile is useless in it's current state, in fact far from it, and as you also mentioned previously I think a lot of the issue is tempering expectations of community members coming from the previous performance and capabilities of the missile.

Looking forward to improvements and fixes for the remaining issues.

[DoorMouse]

@IronMike That's good to hear and appreciate the response. 

[captain_dalan]

Short summary, only take mk60's in battle for the time being? 

ADDENDUM: how does one disable loft to perform test shots? 

[near_blind]

1 minute ago, captain_dalan said:

ADDENDUM: how does one disable loft to perform test shots? 

ACM cover goes brrrrrrt

[Strider21]

@IronMike I appreciate the time you guys are taking to look into this issue. This is obvious but just to be clear I have no idea how a real AIM-54C should perform and am glad you guys are happy that it matches the CFD data. That being said in its current state, at least by my testing, it isn't just a reduction in range but there seems to massive drop in percentage kill. In a previous comment you said something along the lines that the performance is reduced but is still usable. My experience is different.

I ran 25 shots of the AIM-54C (TACVIEW attached). All shots were in SP with Jester with an airstarted F-14B. Conditions were the NTTR map, with default ISA weather conditions, and the target was a ACE AI Su-30 with A2A loadout but no jammer. The target size switch for all shots was NORMAL. For the first two shots I included a crank post-firing but stopped but doing that so that there was no thought the radar was losing track. The shots were taken in TWS-AUTO. Some scripting was used to spawn the SU-30. In all shots the target roughly started co-alt (25K) but tended to dive in defense.  The first 10 shots the firing aircraft was a ~25K ft ~0.98 MACH at ~30 NM range to target. The second 15 shots the firing aircraft was a ~25K ft ~0.98 MACH at 25 NM to the target. The results were as follows:

30 NM  0/10 (0% Pk)

25 NM 0/15 (0% Pk) 

Having run this test pre-patch I was roughly seeing 50% Pk when the AIM-54C was fired at 35NM. I appreciated that data sets of 10 and 15 are not statistically significant but this is a limitation these tests. From these tests IMHO it is hard not to conclude that at least against the AI Su-30 the AIM-54C is not effective at 25 NM or 35 NM when co-alt at 25K. Is this real-life performance? Of course nobody knows. Is this a realistic setup and adversary for the F-14B? perhaps not. I would love to hear any critique on this setup or what I can do to improve Pk of the AIM-54C is this scenario. 

The TACVIEW file was unfortunately 6 MB (over the forum limit) but is a link to a google drive upload. https://drive.google.com/file/d/1bXMP9mw5-gkLJugt2p4Ap5X1Pao4Za8j/view?usp=sharing

Edited January 28, 2022 by Strider21

[captain_dalan]

5 minutes ago, near_blind said:

ACM cover goes brrrrrrt

But doesn't that mean an active launch? 

[IronMike]

12 minutes ago, Strider21 said:

@IronMike I appreciate the time you guys are taking to look into this issue. This is obvious but just to be clear I have no idea how a real AIM-54C should perform and am glad you guys are happy that it matches the CFD data. That being said in its current state, at least by my testing, it isn't just a reduction in range but there seems to massive drop in percentage kill. In a previous comment you said something along the lines that the performance is reduced but is still usable. My experience is different.

I ran 25 shots of the AIM-54C (TACVIEW attached). All shots were in SP with Jester with an airstarted F-14B. Conditions were the NTTR map, with default ISA weather conditions, and the target was a ACE AI Su-30 with A2A loadout but no jammer. The target size switch for all shots was NORMAL. For the first two shots I included a crank post-firing but stopped but doing that so that there was no thought the radar was losing track. The shots were taken in TWS-AUTO. Some scripting was used to spawn the SU-30. In all shots the target roughly started co-alt (25K) but tended to dive in defense.  The first 10 shots the firing aircraft was a ~25K ft ~0.98 MACH at ~30 NM range to target. The second 15 shots the firing aircraft was a ~25K ft ~0.98 MACH at 25 NM to the target. The results were as follows:

30 NM  0/10 (0% Pk)

25 NM 0/15 (0% Pk) 

Having run this test pre-patch I was roughly seeing 50% Pk when the AIM-54C was fired at 35NM. I appreciated that data sets of 10 and 15 are not statistically significant but this is a limitation these tests. From these tests IMHO it is hard not to conclude that at least against the AI Su-30 the AIM-54C is not effective at 25 NM or 35 NM when co-alt at 25K. Is this real-life performance? Of course nobody knows. Is this a realistic setup and adversary for the F-14B? perhaps not. I would love to hear any critique on this setup or what I can do to improve Pk of the AIM-54C is this scenario. 

The TACVIEW file was unfortunately 6 MB (over the forum limit) but is a link to a google drive upload. https://drive.google.com/file/d/1bXMP9mw5-gkLJugt2p4Ap5X1Pao4Za8j/view?usp=sharing

 

Does the AI have chaff? This is a separate issue, but the AI will chaff as soon as you launch, in your tests make sure the AI has zero chaff.

Try moving in closer for the ace AI (which aircraft does not matter really). The C currently borks itself by lofting too high and basically almost stalling out at the peak of its loft, which is another issue we are currently looking into. Try the A60 instead for the time being. There are also slight discrepancies with the C's motor, which we are looking into as well. Bottom line: the C is probably the worst to use right now for anything but close to WVR.

 

[dundun92]

11 minutes ago, Strider21 said:

...

So all the missiles in the tacview had just enough energy to hit the target, what happened is that the AI chaffed them from well outside the notch like usual. . Its probably best to run the test against a human opponent that doesnt cheat with super effective chaff and notching. Otherwise, I think most of those would have been hits.

[Strider21]

1 minute ago, IronMike said:

Does the AI have chaff? This is a separate issue, but the AI will chaff as soon as you launch, in your tests make sure the AI has zero chaff.

Try moving in closer for the ace AI (which aircraft does not matter really). The C currently borks itself by lofting too high and basically almost stalling out at the peak of its loft, which is another issue we are currently looking into. Try the A60 instead for the time being. There are also slight discrepancies with the C's motor, which we are looking into as well. Bottom line: the C is probably the worst to use right now for anything but close to WVR.

 

Appreciate the comment. In all these test the AI had chaff. In previous test pre-patch, with the same conditions (AI with CHAFF) Pk was about 50%

[IronMike]

1 minute ago, Strider21 said:

Appreciate the comment. In all these test the AI had chaff. In previous test pre-patch, with the same conditions (AI with CHAFF) Pk was about 50%

Which likely should have been close to 100% except for the AI chaffing their ass out of abnormal situations...

[Strider21]

5 minutes ago, dundun92 said:

So all the missiles in the tacview had just enough energy to hit the target, what happened is that the AI chaffed them from well outside the notch like usual. . Its probably best to run the test against a human opponent that doesnt cheat with super effective chaff and notching. Otherwise, I think most of those would have been hits.

No I understand they had the kinematic energy. The reason I use the ACE AI Su-30 with chaff is this is what I have found to highest threat AI aircraft and represents what I would see in a DCS mission. Running the scenario without chaff and getting higher Pk won't really then represent what I will see in a mission. Exact same setup pre-patch netted 50% Pk.

 

In realistic "game" scenario against a Su-30 I would argue the 54C is not effective at 25NM or 30NM. Maybe its better at 20NM but I haven't run the test. 

Edited January 28, 2022 by Strider21

[dundun92]

Just now, Strider21 said:

No I understand they had the kinematic energy. I reason I use the ACE AI Su-30 with chaff is this is what I have found to highest threat AI aircraft and represents what I would see in a DCS mission. Running the scenario without chaff and getting higher Pk won't really then represent what I will see in a mission. 

For sure, im just pointing out why its happening, this is fundamentally an issue with both missiles being affected by chaff way too much, and the AI chaff being super effective for some reason

[IronMike]

1 minute ago, Strider21 said:

No I understand they had the kinematic energy. I reason I use the ACE AI Su-30 with chaff is this is what I have found to highest threat AI aircraft and represents what I would see in a DCS mission. Running the scenario without chaff and getting higher Pk won't really then represent what I will see in a mission. 

You are mixing two bugs though, then. 1) missile performance being hindered by guidance issues. 2) The AI chaffing prior to the missile going active, and thus rendering the missile useless. Which inhibits testing your point 1). We know these things are borked both, but if you mix them, you will not get a glance at what the current missile performance is, is my point. I hope that makes more sense like that.

[near_blind]

18 minutes ago, captain_dalan said:

But doesn't that mean an active launch? 

If you're just trying to test straight line kinematics got cfd validation, does it matter if the missle is active or not? Otherwise loft isn't really "configurable" for TWS/PD-STT shots.