DCS: F-14 Development Update - AIM-54 Phoenix Improvements & Overhaul - Guided Discussion

[IronMike]

12 minutes ago, WarthogOsl said:

Great about the lofting, but does that mean we're back to the situation where the missiles immediately do a 180 if the track drops?

Nope. Although someone in the test team reported they might still do that when grossly manually overlofted, but that was general missiles, not aim54 specific. But the fix for the looping stuff remains in place. The "reverted" refers to lofting.

[Rinz1er]

Any status on when we may get the target size switch functionality fixed?

[IronMike]

41 minutes ago, Rinz1er said:

Any status on when we may get the target size switch functionality fixed?

There is nothing to fix. ED would have to change how actives work in DCS. It works for the A as intended and only affects the C though.

It is also a bit over-rated in DCS, and a DCS-ism due to the lack of possibilities to model certain things in missiles. To not get into too much detail, irl it is not meant as a sneaky mode, but reflects when the missile is likely to detect a target on its own. If misjudged, active guidance may be inhibited, for example if using large on a fighter sized target (or normal or small on a bomber sized target). But this is something we most likely will never see in DCS, it is far beyond the scope of what a consumer sim should and can offer.

Edited October 26, 2022 by IronMike

[Callsign JoNay]

Missile loft is improved. Level shot, no loft assisting, missile climbed up to 75k.

Unfortunately it's still extremely susceptible to notching and does this unusual pull up maneuver, almost as if to avoid impact instead of maintaining a trajectory that would give it a chance to re-acquire a pitbull.

[WarthogOsl]

4 hours ago, Callsign JoNay said:

Missile loft is improved. Level shot, no loft assisting, missile climbed up to 75k.

Unfortunately it's still extremely susceptible to notching and does this unusual pull up maneuver, almost as if to avoid impact instead of maintaining a trajectory that would give it a chance to re-acquire a pitbull.

It's so weird...like the bandit has a force field around it and the missile bounces off.  It doesn't even look like the angle was right to get in the notch in your screen shot, either.  Happens to me all the time, though.

Last night I was only getting the missile to around 65k ft from a 40k launch with no assist.  Haven't tried assisting since the patch.

[lunaticfringe]

38 minutes ago, WarthogOsl said:

It's so weird...like the bandit has a force field around it and the missile bounces off.  It doesn't even look like the angle was right to get in the notch in your screen shot, either.  Happens to me all the time, though.

 

It does.  The bandit flight path looks to have rolled over on its back, turned to almost the near vertical dive, then pulled out. 

The (over-wide) notch exists in three dimensions; hit 90 degrees perpendicular to the radar in any direction and you're there. 

[IronMike]

44 minutes ago, WarthogOsl said:

It's so weird...like the bandit has a force field around it and the missile bounces off.  It doesn't even look like the angle was right to get in the notch in your screen shot, either.  Happens to me all the time, though.

Last night I was only getting the missile to around 65k ft from a 40k launch with no assist.  Haven't tried assisting since the patch.

the notch behaviour is not in our hands unfortunately.

if the loft went only to 65k feet from 40k launch altitude, I suspect it was fairly close range. the further away, the higher it will loft naturally. Manual loft assistance is not advised.

[WarthogOsl]

40 minutes ago, IronMike said:

the notch behaviour is not in our hands unfortunately.

if the loft went only to 65k feet from 40k launch altitude, I suspect it was fairly close range. the further away, the higher it will loft naturally. Manual loft assistance is not advised.

One of the three was short, but the two others I'd consider reasonably long range.  Both were from level flight (no assist).

1st: Range 58 miles, launch at 40,100ft , Mach .9.  Loft was to 65,200 feet.

2nd: Range 62 miles, 44,000ft, Mach .68.  Loft was to 68,000 feet.

To be fair, I haven't tried a 70+ mile launch since the patch.

[IronMike]

1 hour ago, WarthogOsl said:

One of the three was short, but the two others I'd consider reasonably long range.  Both were from level flight (no assist).

1st: Range 58 miles, launch at 40,100ft , Mach .9.  Loft was to 65,200 feet.

2nd: Range 62 miles, 44,000ft, Mach .68.  Loft was to 68,000 feet.

To be fair, I haven't tried a 70+ mile launch since the patch.

I did a 110nm shot and it climbed to 113k or 114 cant remember

[WarthogOsl]

4 hours ago, IronMike said:

I did a 110nm shot and it climbed to 113k or 114 cant remember

I know it might not be advised or realistic, but loft assist still seems to give a definite energy advantage.  Here's me firing two Phoenix's at two MiG-21's at equal range (~70nm).  The first shot isn't lofted and climbs to around 77k.  The second shot, taken immediately after, gets a loft assist and goes up to nearly 100k feet, and gets to the bandits' general altitude with around .5+ Mach more speed.  It actually manages to just about catch up with the first missile.

loftvsnonloft.acmi

[KlarSnow]

Yes a loft assist will improve the missiles performance. You just have to understand that by doing so you are risking the missile doing something unpredictable like flipping over backwards or not guiding properly. If thats worth you doing a pitch up to help out then sure go for it. If you get absolutely infuriated if the missile messes up once, then maybe stay away from it and just stick to the recommended employment which is straight and level. 

It should be mostly fixed to my understanding, but there is no guarantee that if you start employing it outside the recommendation that it wont start doing crazy backflips or start doing some of the strange things that happened before. Thats really all that Ironmike or anyone is saying by the recommended employment. If you want to get the best chance of things working as intended, employ straight and level. If you want potentially better kinematics and performance BUT at the risk of the missile doing something untoward an unknowable amount of the time, that’s what the loft assist can buy you.

At this point its kind of stuck between a rock and a hard place with what can be done to it for flight guidance. All employment options are options but you the user must understand what you are risking by going for a min max employment mindset. 9/10 times it will probly work fine, but 1/10 it may just yote itself into nowhere, and right now that is not something that can be controlled with any guarantee without going back to prepatch loft values.

[Exorcet]

 

On 10/26/2022 at 4:00 AM, IronMike said:

If misjudged, active guidance may be inhibited, for example if using large on a fighter sized target (or normal or small on a bomber sized target). But this is something we most likely will never see in DCS, it is far beyond the scope of what a consumer sim should and can offer.

 

 

I was with you until this point. I've always really disliked the way seeker target size is modeled in DCS because is so far off. I feel like the correct implementation with all the downsides should be modeled at some point.

8 hours ago, WarthogOsl said:

I know it might not be advised or realistic, but loft assist still seems to give a definite energy advantage.  Here's me firing two Phoenix's at two MiG-21's at equal range (~70nm).  The first shot isn't lofted and climbs to around 77k.  The second shot, taken immediately after, gets a loft assist and goes up to nearly 100k feet, and gets to the bandits' general altitude with around .5+ Mach more speed.  It actually manages to just about catch up with the first missile.

loftvsnonloft.acmi 500.69 kB · 1 download

For the sake of completeness, why not try it the other way around too; launch the lofted missile first. A small difference in launch time can have a surprising effect on which missile gets there first.

There is absolutely nothing unrealistic about loft assist though. You're allowing the missile to put all its thrust into gaining altitude and you're giving it a boost in doing so. Without the assist a few seconds of burn time are wasted on a pitch up maneuver and the missile will also never reach as high a speed as the assisted one. The problem is that the current loft profile is not optimal and almost ends up negating the assist. The missile wants to be as high as possible, but going straight up is one of the worst ways to achieve that. Going vertical means that the missile has to waste energy turning on to the target at peak altitude, and it also ignores the very important horizontal component of velocity. While you want the missile to go high, you also want it to close in on your target as fast as possible. Not taking air density into account, this is going to be a straight line from you to the target.

The new F-15 radar really lets you see the difference a good loft (AIM-120) makes. The F-15 can now fire AMRAAM probably 100 miles and still kill a fighter when taking advantage of speed and altitude. Part of it is the better aerodynamics of AMRAAM, but it's also because the AMRAAM lofts efficiently (until it doesn't, my testing also helped me appreciate the work that goes into modeling in DCS - The AIM-120 seems to be worse at maneuvering above 100k ft compared to the AIM-54, and when the latter is fixed this will probably be an important consideration in 120 vs 54 matchups, anyway it's nice to see the missiles have tradeoffs where as less detailed sims would just make the 54 a longer ranged AMRAAM). I really can't wait for the Phoenix to get the new API.

Edited October 30, 2022 by Exorcet

[Callsign JoNay]

21 hours ago, IronMike said:

the notch behaviour is not in our hands unfortunately.

So, we should post a bug report on ED's primary forums? It would probably mean more coming from HB, since you're the dev and have ED's ear, etc. The missile used to maintain course after getting notched and would have a good chance of re-acquiring the target. This new upward pull up behavior after the first moment of losing lock via notch is more detrimental to the AIM-54 PK% than the borked loft trajectory ED installed to fix the looping bug pre-2.8. 

[WarthogOsl]

4 minutes ago, Callsign JoNay said:

So, we should post a bug report on ED's primary forums? It would probably mean more coming from HB, since you're the dev and have ED's ear, etc. The missile used to maintain course after getting notched and would have a good chance of re-acquiring the target. This new upward pull up behavior after the first moment of losing lock via notch is more detrimental to the AIM-54 PK% than the borked loft trajectory ED installed to fix the looping bug pre-2.8. 

I don't think this behavior is new at all FWIW.

7 hours ago, Exorcet said:

 

 

I was with you until this point. I've always really disliked the way seeker target size is modeled in DCS because is so far off. I feel like the correct implementation with all the downsides should be modeled at some point.

For the sake of completeness, why not try it the other way around too; launch the lofted missile first. A small difference in launch time can have a surprising effect on which missile gets there first.

There is absolutely nothing unrealistic about loft assist though. You're allowing the missile to put all its thrust into gaining altitude and you're giving it a boost in doing so. Without the assist a few seconds of burn time are wasted on a pitch up maneuver and the missile will also never reach as high a speed as the assisted one. The problem is that the current loft profile is not optimal and almost ends up negating the assist. The missile wants to be as high as possible, but going straight up is one of the worst ways to achieve that. Going vertical means that the missile has to waste energy turning on to the target at peak altitude, and it also ignores the very important horizontal component of velocity. While you want the missile to go high, you also want it to close in on your target as fast as possible. Not taking air density into account, this is going to be a straight line from you to the target.

The new F-15 radar really lets you see the difference a good loft (AIM-120) makes. The F-15 can now fire AMRAAM probably 100 miles and still kill a fighter when taking advantage of speed and altitude. Part of it is the better aerodynamics of AMRAAM, but it's also because the AMRAAM lofts efficiently (until it doesn't, my testing also helped me appreciate the work that goes into modeling in DCS - The AIM-120 seems to be worse at maneuvering above 100k ft compared to the AIM-54, and when the latter is fixed this will probably be an important consideration in 120 vs 54 matchups, anyway it's nice to see the missiles have tradeoffs where as less detailed sims would just make the 54 a longer ranged AMRAAM). I really can't wait for the Phoenix to get the new API.

 

I think it's unrealistic in terms of the f-14 because I don't think Tomcats ever launched Phoenix's using a loft.  That said, I wonder if this lower loft profile was a fix for the missiles lofting too much on short range shots in the past. Perhaps that is compromising the medium-long range shots?

[Naquaii]

7 hours ago, Exorcet said:

 

 

I was with you until this point. I've always really disliked the way seeker target size is modeled in DCS because is so far off. I feel like the correct implementation with all the downsides should be modeled at some point.

For the sake of completeness, why not try it the other way around too; launch the lofted missile first. A small difference in launch time can have a surprising effect on which missile gets there first.

There is absolutely nothing unrealistic about loft assist though. You're allowing the missile to put all its thrust into gaining altitude and you're giving it a boost in doing so. Without the assist a few seconds of burn time are wasted on a pitch up maneuver and the missile will also never reach as high a speed as the assisted one. The problem is that the current loft profile is not optimal and almost ends up negating the assist. The missile wants to be as high as possible, but going straight up is one of the worst ways to achieve that. Going vertical means that the missile has to waste energy turning on to the target at peak altitude, and it also ignores the very important horizontal component of velocity. While you want the missile to go high, you also want it to close in on your target as fast as possible. Not taking air density into account, this is going to be a straight line from you to the target.

The new F-15 radar really lets you see the difference a good loft (AIM-120) makes. The F-15 can now fire AMRAAM probably 100 miles and still kill a fighter when taking advantage of speed and altitude. Part of it is the better aerodynamics of AMRAAM, but it's also because the AMRAAM lofts efficiently (until it doesn't, my testing also helped me appreciate the work that goes into modeling in DCS - The AIM-120 seems to be worse at maneuvering above 100k ft compared to the AIM-54, and when the latter is fixed this will probably be an important consideration in 120 vs 54 matchups, anyway it's nice to see the missiles have tradeoffs where as less detailed sims would just make the 54 a longer ranged AMRAAM). I really can't wait for the Phoenix to get the new API.

 

 

21 minutes ago, WarthogOsl said:

I don't think this behavior is new at all FWIW.

I think it's unrealistic in terms of the f-14 because I don't think Tomcats ever launched Phoenix's using a loft.  That said, I wonder if this lower loft profile was a fix for the missiles lofting too much on short range shots in the past. Perhaps that is compromising the medium-long range shots?

The loft assist we have is absolutely unrealistic as have been mentioned multiple times previously. The missile wouldn't loft differently just because you pitched the aircraft up as the missile would still fly it's profile. It likely might not matter at all as the missile (at least the AIM-54A) is likely to just have come off the aircraft, level out and then pitch up.

Even if the AIM-54C had a slightly smarter launch sequence the only benefit would be that the missile would have a slightly easier time of reaching correct pitch and lose slightly less energy (not majorly so though as the rocket motor is on anyway), it would however absolutely not pitch more because of a loft assist.

[Exorcet]

1 hour ago, Naquaii said:

 

The loft assist we have is absolutely unrealistic as have been mentioned multiple times previously.

By loft assist, I mean the launch platform pitching up, not the missile changing its profile.

1 hour ago, Naquaii said:

Even if the AIM-54C had a slightly smarter launch sequence the only benefit would be that the missile would have a slightly easier time of reaching correct pitch and lose slightly less energy (not majorly so though as the rocket motor is on anyway), it would however absolutely not pitch more because of a loft assist.

Exactly, though from what I've been seeing in DCS, I don't think the benefit from cutting out the initial pitch correction is so small that it's trivial, but then again if the loft was more optimized the missile probably wouldn't be pitching up 50 degrees all the time which should help with the energy loss.

[Naquaii]

4 minutes ago, Exorcet said:

By loft assist, I mean the launch platform pitching up, not the missile changing its profile.

Exactly, though from what I've been seeing in DCS, I don't think the benefit from cutting out the initial pitch correction is so small that it's trivial, but then again if the loft was more optimized the missile probably wouldn't be pitching up 50 degrees all the time which should help with the energy loss.

I got what you meant. Loft assist just wasn't a thing for the AIM-54.

And yeah, I don't argue that it can't get results currently in DCS but it's not realistic.

Unfortunately there's no way currently for us to stop it.

Edited October 30, 2022 by Naquaii

[Exorcet]

38 minutes ago, Naquaii said:

I got what you meant. Loft assist just wasn't a thing for the AIM-54.

And yeah, I don't argue that it can't get results currently in DCS but it's not realistic.

Unfortunately there's no way currently for us to stop it.

 

Ah I see then the Phoenix should just level out regardless then? I missed where that was mentioned previously in the thread.

[Callsign JoNay]

3 hours ago, WarthogOsl said:

I don't think this behavior is new at all FWIW.

I disagree, I think it's pretty new. Here's a video I made comparing a June 2022 tacview to a Tacview I recorded today in 2.8.

I think this, more than any kind of nerf to the kinetics/loft of the missile, is the biggest factor hampering it's current PK%.

 

[Naquaii]

19 minutes ago, Exorcet said:

Ah I see then the Phoenix should just level out regardless then? I missed where that was mentioned previously in the thread.

I'm quite positive the AIM-54A would be like that at least. Harder to tell about the AIM-54C. But like we discussed the C would at most take the "free" pitch from the aircraft and still pitch to the same angle regardless. Not like what we currently have in DCS where it pitches more because of it.

It would be nice if we could get rid of that but that's likely not on our side.

[WarthogOsl]

4 hours ago, Callsign JoNay said:

I disagree, I think it's pretty new. Here's a video I made comparing a June 2022 tacview to a Tacview I recorded today in 2.8.

I think this, more than any kind of nerf to the kinetics/loft of the missile, is the biggest factor hampering it's current PK%.

 

FWIW, I made a post about this in early September, but it's something I had seen well before posting about it.  

 

[Rinz1er]

On 10/26/2022 at 4:00 AM, IronMike said:

There is nothing to fix. ED would have to change how actives work in DCS. It works for the A as intended and only affects the C though.

It is also a bit over-rated in DCS, and a DCS-ism due to the lack of possibilities to model certain things in missiles. To not get into too much detail, irl it is not meant as a sneaky mode, but reflects when the missile is likely to detect a target on its own. If misjudged, active guidance may be inhibited, for example if using large on a fighter sized target (or normal or small on a bomber sized target). But this is something we most likely will never see in DCS, it is far beyond the scope of what a consumer sim should and can offer.

 

So this is something to be fixed then. It just is on ED's side and not HB's, right? Like we used to have this functionality for all of the AIM54's and now it broke for some of them. I can make a bug report in the weapons section for them to fix this.

[IronMike]

12 hours ago, Rinz1er said:

So this is something to be fixed then. It just is on ED's side and not HB's, right? Like we used to have this functionality for all of the AIM54's and now it broke for some of them. I can make a bug report in the weapons section for them to fix this.

Thanks for the report, like I replied, it is not a bug as such, but ED reacted to it and wrote to us, so we will look if there is a solution, but no promise.

[skywalker22]

On this track file, Ace level AI uses version C Mk47 and Mk60, and all are lofting even on pretty "short" range to their target (me). This should not be happening!

 

Tacview-20221101-152512-DCS.zip.acmi

I guess this is just a not so good AI, which cannot use all the modes available.

[lunaticfringe]

58 minutes ago, skywalker22 said:

On this track file, Ace level AI uses version C Mk47 and Mk60, and all are lofting even on pretty "short" range to their target (me). This should not be happening!

 

Tacview-20221101-152512-DCS.zip.acmi 493.52 kB · 0 downloads

I guess this is just a not so good AI, which cannot use all the modes available.

Loft isn't selectable.  If it's above 20nm and not HOJ, the missile is going to loft, no matter what. 

The first four Phoenix shots were at 43.27, 41.90, 23.52, and 24.97 nm respectively.  The first Phoenix not to loft was shot at ~16.4 nm.   

Now, one question: what do you have as your "AA Missile Attack Range" setting? 

Edited November 1, 2022 by lunaticfringe