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

[RustBelt]

11 hours ago, Gareth Barry said:

I guess let me be a bit more frank-

I wonder what range of conditions and altitudes the nasa model was calculated for? And if it is a wide range of altitudes, what are we missing, as in, why does the nasa model not gel? How is it that a current model without nozzle exit area and and pressure that changes with altitude, 'correct'? What are we missing? Or has the time come to humbly ask Heatblur to reconsider their model? I know that this is has been covered in this thread, but the thread is looooong and i'm lazy....

I only play single player and get plenty of kills with the missile- i know it isnt really a 'mach 5' missile, but the high speed missile at altitude that i thought the thing was famous for is currently a bit more like a long, slow train a comin'.

 

What you’re missing is this is a game with a, to us, black box game engine. For all you know, that nozzle number may be a modifier to an internal pre modeled nozzle that DOES have a set diameter. 
 

What you’re missing is the actual source code.

Edited September 15, 2024 by RustBelt

[LanceCriminal86]

As I recall the nozzle exit value came from ED while it was being sorted out.

[Gareth Barry]

If anything i have said has come across as arrogant then i sincerely apologise, that wasnt my intention.

Draconus, you are right that one we cannot just go off feelings to say that the model is wrong. What made me question slightly was the nozzle exit area issue, and people modifying to be in line with tavarish's calculations getting higher performance result at high altitude. HOWEVER Rustbelt makes an excellent point - without the source code i guess there is no way of knowing what the nozzle exit area is actually doing in the game. For all we know, they may have needed to set it to that value as a work -around, fudge fix for some other issue, but if the final result ends up being accurate then that is a successful work around.

Edited September 16, 2024 by Gareth Barry

[Katsu]

On 9/15/2024 at 12:07 PM, draconus said:

HB made the model to match and recreate real life test shots. There's not much more data on the performance so everything else is theory but it also matches that. You also don't know the model fully, not all is in lua files. So what are you looking for? What makes you think it's wrong? You have to base it on something more than feeling.

Draconus, respectfully, I don't think anyone here wants a "war" about this.

Believe that this kind of response of “what makes you believe that other than your thoughts” is an unhealthy and presumptuous way of debating something.

We are here trying to understand what Heatblur has done, and trying together with the detailed study that Tavarish kindly provided, why the missile “reacts” the way it does in-game.

If we just use data without calculations like "x is equal to x because my SME told me", we have already seen several cases of SME in podcasts or interviews speaking inaccurate information (in no way trying to diminish because these people are true Heroes who have dedicated their lives to protecting their country not to mention the many who have paid with their own lives). But back on topic, I want to remind you that we have a paper from ED showing how they used data to reproduce the 120 in and this allows us to understand the performance presented in game (And we know that this performance must also be different from the real missile because there are several factors that cannot be simulated in DCS) and also DCS is a game/simulator aimed at the general public for military matters there is the MCS,

And if we if we go as drastic as “my data against your data!!!” we have several sources on the internet from the navy itself saying that the Phoenix was a missile capable of reaching mach 5 (and we know that it wasn't, not in the conditions that its launch platform could reach).  

you can find many sources from "official" documents that contradict the game's current performance like this one: 

I think what we're looking for here is to understand the calculations and where the metrics used to make the missile the way it is in the game came from.

And whenever we try to debate this, we're treated like teenagers trying to improve a toy, when in fact I believe that everyone here is looking for something that, even if it can't be identical, can at least be simulated approximately correctly. 

We don't have children here, you're talking to engineers, programmers, maybe even military personnel and all kinds of professionals who have this as a hobby and who share this passion for knowing how things work and why. 

We're trying to get technical answers, not an arguments fight.

My best regards.

 

 

Edited September 18, 2024 by Katsu

[Despayre]

13 minutes ago, Katsu said:

 

And whenever we try to debate this, we're treated like teenagers trying to improve a toy, when in fact I believe that everyone here is looking for something that, even if it can't be identical, can at least be simulated approximately correctly. 

We don't have children here, you're talking to engineers, programmers, maybe even military personnel and all kinds of professionals who have this as a hobby and who share this passion for knowing how things work and why. 

We're trying to get technical answers, not an arguments fight.
 

I can see how you could think that, but don't forget, you're the ones trying to come up with the answers, when you have no idea what numbers you're missing in the source code you do not have access to. You can make all the assumptions, guesses, and even calculations you want, but at the end of the day, without the chunk of code you don't have access to, you're just guessing, and that's just... well, let's just say, it's extremely unlikely to be useful at all, making this a pointless exercise in totality that continues to go round and round without end. 

Until someone has access to ALL the code, it's pointless, and you CANNOT get the technical answers you're looking for, so that claim just doesn't fly (much like the Phoenix, haha).

[NEW is my Callsign.]

26 minutes ago, Despayre said:

Until someone has access to ALL the code, it's pointless, and you CANNOT get the technical answers you're looking for, so that claim just doesn't fly (much like the Phoenix, haha).

Have you even read Tavarish’s claims? Serious question. 

[Despayre]

5 minutes ago, NEW is my Callsign. said:

Have you even read Tavarish’s claims? Serious question. 

For about a year now. He's put in a lot of work, and made some reasonable guesses and done a lot of math, but that doesn't change the fact that he cannot see what the DCS code does to the missile after it's handed off from the LUA code.

Edited September 17, 2024 by Despayre
grammar

[draconus]

5 hours ago, Katsu said:

...we're treated like teenagers trying to improve a toy...

You read more into it than I have written and I appreciate good science but let me remind you this quote I replied to.

On 9/15/2024 at 11:52 AM, Gareth Barry said:

...the high speed missile at altitude that i thought the thing was famous for is currently a bit more like a long, slow train a comin'.

[Gareth Barry]

The last time my love of 70s rock music caused such a problem, i was still teenager living with my parents

Edited September 17, 2024 by Gareth Barry

[tavarish palkovnik]

On 9/17/2024 at 3:47 AM, Despayre said:

but that doesn't change the fact that he cannot see what the DCS code does to the missile

 

Don't be so sure  

Indeed I can't see what exact code is implemented but with reversible method it can be easily figured out. Maybe you expect that motors in DCS are modeled in details, but they are not and like said it is understandable, who would make all these motors to be precise and perfect. So everything can be detected because of simplifications used in DCS and all other similar games. Give me several level (horizontal) flight envelopes velocity-time and distance-time for active time (when motor burns) let's say 1,2M launch velocity (or better 1M because 1,2M at 1km exceed limits) at 1, 5, 10 and 15km and let's see can we see or not what is code, what is motor thrust at these altitudes.

DCS simplified principles is Isp*m, always linear thrust, not involving details like chamber pressure, pressure at nozzle exit, nozzle expansion ratios, throat diameters etc etc etc and that's perfectly right for most of motors. Even 7% principle when increasing thrust as altitude gain is fair enough because it fits quite fine for most of motors...but not this one. Deliberately said ''one'' not ''ones'' and who was reading me know what it means. This motor simply can't be ''molded'' together with Sidewinder, Amraam, Sparrow and pallets of various other tactical motors because with its huge nozzle expansion ratio it is simply different and needs different approach...believe it or not 

We have saying to someone when ruin established facts, you are destroying snowman, and I know that is exactly what I'm doing  But not from wickedness but from engineering needs for precision 

Edited September 18, 2024 by tavarish palkovnik
1,2M -> 1M

[DoorMouse]

On 9/15/2024 at 5:52 AM, Gareth Barry said:

I guess let me be a bit more frank-

I wonder what range of conditions and altitudes the nasa model was calculated for? And if it is a wide range of altitudes, what are we missing, as in, why does the nasa model not gel? How is it that a current model without nozzle exit area and and pressure that changes with altitude, 'correct'? What are we missing? Or has the time come to humbly ask Heatblur to reconsider their model? I know that this is has been covered in this thread, but the thread is looooong and i'm lazy....

I only play single player and get plenty of kills with the missile- i know it isnt really a 'mach 5' missile, but the high speed missile at altitude that i thought the thing was famous for is currently a bit more like a long, slow train a comin'.

 

The NASA model is significantly modified and lighter. The paper they published details some of the changes they made 

[Kageseigi]

On 8/21/2024 at 3:15 PM, Katsu said:

Working as intended? 

 

That looks so bizarre. The poor Phoenix appears to be virtually useless in DCS. Did you have lock whenever you fired all of your missiles? Did you ever lose lock due to notching?

It looks like your second missile never went active... or at least didn't alter course at all when it got close. But why not? It didn't look like it bit on chaff or any maneuver... it just whiffed.

And did you really shoot down your own missile? How does that even happen? Something seems to be quite rotten.

It's a miracle that the Soviets never realized that all they had to do to defeat the Navy's premiere air defense was to fly in a big circle until they were out of missiles. It's like a cat chasing a laser pointer.

[Katsu]

12 hours ago, DoorMouse said:

The NASA model is significantly modified and lighter. The paper they published details some of the changes they made 

NASA has never launched a Phoenix in this test.

It's all based on public data and simulation:

Edited September 21, 2024 by Katsu

[Katsu]

11 hours ago, Kageseigi said:

That looks so bizarre. The poor Phoenix appears to be virtually useless in DCS. Did you have lock whenever you fired all of your missiles? Did you ever lose lock due to notching?

It looks like your second missile never went active... or at least didn't alter course at all when it got close. But why not? It didn't look like it bit on chaff or any maneuver... it just whiffed.

And did you really shoot down your own missile? How does that even happen? Something seems to be quite rotten.

It's a miracle that the Soviets never realized that all they had to do to defeat the Navy's premiere air defense was to fly in a big circle until they were out of missiles. It's like a cat chasing a laser pointer.

As explained, this is a limitation of DCS.

DCS doesn't actually simulate chaff detection based on factors like VC,RCS,Doppler Shift etc..., but rather “throws a dice” to see if your missile will hit the chaff or not and the chance of this “dice” hitting is based on a resistance number given to the missile, the more chaff you throw, the more chance of the “dice” hitting the chaff. 

So at 30,000 or 0 feet the chance doesn't change.

Unfortunately this is how the game works today, and in this case the only thing that heatblur could do to perhaps reduce the “problem” is to increase the resistance numbers of the missile to make it more realistic, but as has been seen on all these pages here there is no willingness to accept suggestions. 

[FWind]

1小时前，Katsu说：

NASA has never launched a Phoenix in this test.

It's all based on public data and simulation:

 

[Gareth Barry]

So, the Nasa calculations/simulations achieved only mach 4.7 ish from a mach 2,0 launch at 48 000ft....shucks, don't know if that's even achievable in the game. At first glance, mach 4.7 seems like a lot, but on second thoughts, fromm a mach 2.0 launch, at that altitude... dunno if the tomcat even has the performance to get there. If one was going to try to replicate this, with 1 singe phoenix and nothing else, maybe no gun ammo, only enough fuel to reach mach 2 at 48 000 ft, then run out of fuel, deadstick landing....which weapon pylon would have the least drag for the tomcat? One of the side pylons?

Edited September 21, 2024 by Gareth Barry

[DoorMouse]

4 hours ago, FWind said:

@Gareth Barry @Katsu

The test configuration was significantly lightened - Yes they never fired it... and their data is about as good as anyone here can ask for. 

As much as I personally would love the AIM-54 in game to be better. I think the Kinematics are probably pretty close... The Guidance is a major culprit of the poor performance, id wager. hey have a couple charts for performance. One is 45kft Mach 1.2 45 degree launch. The other is 45,000ft Mach 2 launch. Fire up the mission editor and test it, post tacviews.  I'll give that a shot later. null

171793main_fs-093-dfrc.pdf

Edited September 21, 2024 by DoorMouse

[KlarSnow]

If you go back to the start of this thread 2 years ago, we already tested and compared the current missile to these fly out graphs. Spoiler they are extremely close. As in almost perfectly matching.

Additionally you do have to modify the missile lua a bit to perform the test properly, since by default the missile will try and maintain the attitude you fire it at. If you disable all guidance and autopilot in the missile via editing the lua, so it will fly ballistically instead of holding an attitude or AOA, then it will in fact match the performance in the flyout graphs posted.

[Gareth Barry]

Just to clarify, i am not questioning Nasa at all- others might be, im not sure, but i am not in any position to do so, in fact the thought of someone like me questioning Nasa i find somewhat funny, considering how surface level my knowledge is at best. My last message post was simply reiterating that the nasa model shows that it really isnt mach 4.5+ missile in any practical sense. Beyond that, i am not in any position to have an opinion one way or the other, as both sides to my mind make good arguments and both sides are far smarter than me. Perhaps, Katsu and Tavarish, maybe for the purposes of moving the discussion forward, what do you see as the limitations, perhaps in terms of context, of the nasa test? With that, i bow out, and will go back to spending my time trying to fly the f14 better, because i still suck at it in so many ways. Still, ive certainly learnt something regarding rocket motor thrust and performance from this thread, so i thank everyone for that.

Edited September 21, 2024 by Gareth Barry

[Katsu]

@FWind It's great that you've posted this test again, so that I can point out some differences based on the data you've provided.

From my performance chart taken from the game "vanilla" how is it now no modifications made by me, I was able to simulate a shot very similar to the one provided by the NASA study unfortunately I just didn't get the same total flight time because I needed a target to be able to adjust the loft, (with a very similar climb profile as you can see comparing the altitude charts)

The missile can barely get above mach 4 while the missile should reach approximately 4.8 mach based on NASA tests.

 

Please take note that this NASA simulation its calculated with 1000lb missile as you can see on "launch condition":

Here we have the results from Nasa: 


Please note that NASA indicates that the missile has a window above mach 4 of 41 seconds, what we have today replicating the same launch is a window of around 2 to 3 seconds.
which in my opinion seems to be a very significant “lack” of performance, since all that less speed will interfere with the missile's travel time, leading to various consequences. 

Now I'm going to present two case studies of my own where I changed the code of the missile and adjusted it according to the thrust of 18318N and burn time of 23.8 sec that @tavarish palkovnikprovided:

This first one: I didn't modify the loft profile of the standard used by the missile, even though I still achieved a speed of  4.6 mach, which is much closer to the NASA test: 


On the second one:, I modified the loft profile so that the missile achieved a higher peak speed of 4.8 mach , which gave me practically identical performance as NASA test, at least in terms of kinetic energy:
  

For these reasons that I presented in my previous posts, plus those that I'm showing here now together with the data provided from NASA test and Tavarish Calculations, I believe that today we have a simulation that tends to be incorrect in some kinectics aspects of the missile.

Remembering that at no point do I want to point fingers or say that the work hasn't been done well, but that it can be improved, not only in the fact of the guidance and loft profile as I showed in my tests that just a small adjustment of the standard profile that is currently configured made me gain peak speed in the missile, we are not far away, it is necessary to recognize all the work that Heatblur has done to date, but the fact that they are excellent developers does not exempt them from the fact that at some point they can also make mistakes, just like you and me, we are all human.

What I'm looking for is to try to help so that we can have the best possible digital recreation of this iconic airplane.  

Edited September 22, 2024 by Katsu

[Kageseigi]

13 hours ago, Katsu said:

As explained, this is a limitation of DCS.

DCS doesn't actually simulate chaff detection based on factors like VC,RCS,Doppler Shift etc..., but rather “throws a dice” to see if your missile will hit the chaff or not and the chance of this “dice” hitting is based on a resistance number given to the missile, the more chaff you throw, the more chance of the “dice” hitting the chaff. 

So at 30,000 or 0 feet the chance doesn't change.

Unfortunately this is how the game works today, and in this case the only thing that heatblur could do to perhaps reduce the “problem” is to increase the resistance numbers of the missile to make it more realistic, but as has been seen on all these pages here there is no willingness to accept suggestions. 

I only saw one of your four missiles appear to go for chaff. The other three seemed to "self-destruct" for three other different reasons. I wish chaff resistance was the biggest problem the Phoenix had. It seems to be more about supporting/tracking related to notching... almost as if the missile can be successfully notched if the enemy pilot simply turns his head to sneeze.

[tavarish palkovnik]

To help a little bit more, to understand how this motor is special and why it should be taken as uniqe case

 

 

Three methods how to calculate pressure at nozzle exit. First formula is the most precise but very inconvenient to use and it gives pressure 25237 Pa. Second method is approximate method but still fine and enough precise and it gives 25634 Pa. Thrid one is for those not like formulas, and it gives 26163 Pa. All three results are with average chamber pressure 45 bar and heat ratio 1,2

 

 

About this last image latter today...

[tavarish palkovnik]

So, in this last image ''red'' case is Phoenix with its nozzle expansion ratio measuring 18,5 and ''blue'' dots are various tactical motors, there are various Sparrows, Sidewinders, Falcons, than Amraam, Genie, pallet of Rs, 23, 60, 27, 27ER, 33 also some air-to-ground like Harm, Hellfire etc. Maverick is missing because its motor is also extreme with expansion ratio measuring just 2, extreme but on other side from Phoenix, so much extreme that third method is not even applicable.

But what image is showing? It's showing that all these motors are more or less grouped together, alright R-60 and Hellfire pop out a bit but still everything is under control and understandable why. And all these motors work just fine down, and also up with some slight gain getting with altitude. Only Phoenix is with huge difference in thrust at sea level and up there. That's why it is special case, it is intentionally designed to give maximum where tacticians placed it. Other motors are not so much customized but made as universal, fine at sea level, better when higher but that all is measured in just few percentages...except with dual thrust motors (R-33, AIM-7F, AIM-120B and AGM-88) where this percentage is rising significantly when motor works in sustain stage.

I got impression that when common internet write about this altitude gain only external diameter of nozzle is in focus. And that is wrong, without nozzle expansion ratio and chamber pressure all that story is useless. AIR-2 Genie has huge nozzle exit, 296mm is exit diameter what is bigger then Phoenix's bell of 238mm but expansion ratio is just 6,7 and gain with altitude no matter of huge exit area is just slight, nominal sea level thrust of 155kN will be improved just for few percent 

Maybe is time now to left this, it is up to you guys now to fight for it, to get Phoenix in game in more realistic shape. It will ask for different approach but special case looks for special approach. If I can help with something else just shoot 

 

Edited October 5, 2024 by tavarish palkovnik

[killerwhale]

The Phoenix missile both the A and C version can hardly hit any BVR Target nowadays. And I am shooting from within a reasonable range, less than 25. 

[RustBelt]

Less than 25 is too short and practically not BVR by the time the timeline runs out. 

You would be better served to TWS and run them into 15NM and shoot acm up. Or shoot STT out in the 30-50 NM+ range.

It works best as either a long stick, or a close in surprise jab. But trips on it’s own feet in the 20-30NM ranges.  
 

And always treat AWG-9 TWS as a party trick first, and a legit shooting choice second.