R-27ER Experiment & Findings

[JunMcKill]

This is where we disagree Rage. You're making the statement that you want the game to be less realistic to make it more fair for some pilots. As certain levels of realism the balance swings back and forth. But ultimately if you have a step towards realism you have to take it and let the metagame sort itself out.

 

+1 Rep

[Ironhand]

The videos I posted on which Chizh is commenting are all single player not MP. The tracks play the same every time. I'm not sure he realises that.

He doesn't know which they're from. For that matter, I was under the impression they were MP. Regardless, why not PM him about your concerns and send him the tracks (assuming they still play correctly). If not, create new ones. Explain your methodology and conclusions. See what he has to say?

[IASGATG]

Did some pretty preliminary sparrow testing.

 

I started with the AIM-7E2 first as the AIM-7F (which is aerodynamically the same as the 7M). The 7F dlz chart looked a little extreme so i followed the 7E charts from the F-4 manual.

 

At 5kft M0.7 co-alt co-speed head to head, DLZ states RMax at 12nmi

Vanilla: hits the target at 42s, M0.49, the missile was subsonic 14s after launch.

My 3.0 missile mod: hits the target at 34s, M0.67, and was subsonic 21s after launch.

 

Even my mod seems to make the missile still too slow.

 

The 7F

Sea level M0.9, co-speed, etc.

Vanilla: ran out of time.

3.0 mod: at 51s the missile ran out of energy, was still 3.3nmi from the target, more over was subsonic 18s after launch. Considering the motor has a 15s burn time this isn't well modelled, but the vanilla will be even slower.

[Bushmanni]

Using single player to test missile guidance and physics prevents you from having a realistically behaving target who will take advantage of every flaw in the missile, realistic or not.

[OnlyforDCS]

Did some pretty preliminary sparrow testing.

 

I started with the AIM-7E2 first as the AIM-7F (which is aerodynamically the same as the 7M). The 7F dlz chart looked a little extreme so i followed the 7E charts from the F-4 manual.

 

At 5kft M0.7 co-alt co-speed head to head, DLZ states RMax at 12nmi

Vanilla: hits the target at 42s, M0.49, the missile was subsonic 14s after launch.

My 3.0 missile mod: hits the target at 34s, M0.67, and was subsonic 21s after launch.

 

Even my mod seems to make the missile still too slow.

 

The 7F

Sea level M0.9, co-speed, etc.

Vanilla: ran out of time.

3.0 mod: at 51s the missile ran out of energy, was still 3.3nmi from the target, more over was subsonic 18s after launch. Considering the motor has a 15s burn time this isn't well modelled, but the vanilla will be even slower.

 

No one doubts the veracity of your claims, and I believe I speak for many here when I say that I really appreciate the time, research and effort you put into your mod. However, ED have definitively spoken on the issue of range, if I remember correctly wasn't it you and ggtharos who had a discussion with czih himself? Please correct me if Im wrong but isn't the crux of the matter the difference in the interpretation of the meaning of DLZ data charts? I think I remember ggtharos also mentioning that in the current atmospheric modeling, giving the missiles "correct" ranges at low and medium altitudes would increase their high altitude performance to unrealisticly high ranges?

Edited March 17, 2017 by OnlyforDCS

[Schmidtfire]

Wow, I did not think that the AIM-7M would be that slow... This Is a theory I have had for some time, isn't everything feeling a little bit "draggy" in DCS? Planes, missiles etc. are performing just at or below their expected speeds, especially with pylons and bombs attached. The feeling of flight is great, Im not saying it's not on the money, but just a bit "draggy" if that makes sense.

This is my subjective opinion and I have no data or charts to back it up :music_whistling:

[OnlyforDCS]

Wow, I did not think that the AIM-7M would be that slow... This Is a theory I have had for some time, isn't everything feeling a little bit "draggy" in DCS? Planes, missiles etc. are performing just at or below their expected speeds, especially with pylons and bombs attached. The feeling of flight is great, Im not saying it's not on the money, but just a bit "draggy" if that makes sense.

This is my subjective opinion and I have no data or charts to back it up :music_whistling:

 

According to the missile performance charts, this certainly seems true for most missiles in DCS, at least at low and medium altitude. (IAGSTG's tests all seem to confirm that)

 

As for the planes themselves I have no idea, but I from what I've seen most of the AFM/EFM/PFM modules perform very close to their real world counterparts. At least in clean configurations.

[GGTharos]

I don't discuss what what's discussed with Chizh much - it's really not anyone's intent to start some sort of 'us vs ED' thing and it's not a discussion for the forums.

 

It took years to learn and understand how this stuff works and what we understand is that this is a very difficult task.

 

Yes, if you tune for low altitude to be correct-ish high altitude will over-perform, but it isn't not insanely ugly or anything, and it affects all missiles so IMHO it isn't a big deal.

 

 

No one doubts the veracity of your claims, and I believe I speak for many here when I say that I really appreciate the time, research and effort you put into your mod. However, ED have definitively spoken on the issue of range, if I remember correctly wasn't it you and ggtharos who had a discussion with czih himself? Please correct me if Im wrong but isn't the crux of the matter the difference in the interpretation of the meaning of DLZ data charts? I think I remember ggtharos also mentioning that in the current atmospheric modeling, giving the missiles "correct" ranges at low and medium altitudes would increase their high altitude performance to unrealisticly high ranges?

[IASGATG]

No one doubts the veracity of your claims, and I believe I speak for many here when I say that I really appreciate the time, research and effort you put into your mod. However, ED have definitively spoken on the issue of range, if I remember correctly wasn't it you and ggtharos who had a discussion with czih himself? Please correct me if Im wrong but isn't the crux of the matter the difference in the interpretation of the meaning of DLZ data charts? I think I remember ggtharos also mentioning that in the current atmospheric modeling, giving the missiles "correct" ranges at low and medium altitudes would increase their high altitude performance to unrealisticly high ranges?

 

There are a few things going on.

 

There is a philosophical argument over what the data means. One side suggests that the data represents the maximum point the missile can reach at stall, regardless of any other fact. (Such as battery time, loss of hydraulic fluid, self-destruct timers, etc).

 

With this philosophy, if you have a DLZ chart that you trust and know is reliable, and that chart states that Rmax is 60km at 10km alt at M0.9 head to head etc; then you build the missile such that the missile stalls at that intercept point. You then have other speeds and other altitudes and build the curve such that it stalls at all of those intercept points.

 

The competing philosophy is that DLZ charts only tell part of the picture, and typically aren't stall criteria (typically called Raero). You instead build a curve off of other data and then compare that curve to the DLZ's. If it's close then it's a good curve.

 

Now what do I mean by close? The missile could still be doing Mach 2, which seems like it has a lot of energy left in it, so how can that be Rmax? Well if the self-destruct timer goes off in the next second, then it is Rmax. If the missile is still Mach 1+ then it is still maneuverable and able to make make minor intercept jinks in the final seconds of flight.

 

How does this all tie back? Well the R-27ER is coded to fit the R-27ER DLZ chart from (I believe) a Flanker pilot's handbook. This is a strongly reliable source and so I don't doubt the validity of the data. However, because it is a stall criteria at each point, the missile doesn't actually fit the tail chase curves on the following page as the missile doesn't have the necessary energy.

 

Moving onto the AIM-7. From memory, Chizh has said they based the AIM-7M off of the DLZ chart from a USAF SMC of the AIM-7F. This is fine, from a aerodynamics point of view they are the same. However the chart they use states that Rmax for a 24nmi at 40kft with a head to head shot of M1.4 for target and fighter.

 

Ignoring how we feel about this and assume it's right, why might this not be be a stall intercept point? Well the chart on the same page states that the Rmax for a Sea Level shot at M0.9 is 20nmi. It's impossible for the missile to do both. The reason why the Rmax is 24nmi at 40kft, is that it's being fired from an F-4 Phantom, and the max range it can lock the RCS stated is 24nmi, which is stated.

 

In the same SMC, it states that the Raero for the AIM-7F is 57nmi. The problem is we don't know from this page what the shoot criteria are for this Raero shot. It does also state that the AIM-7E2 has an Raero of 27nmi. In the F-4 Phantom manual, it states the same thing and provides the shoot criteria, 40kft at M1.4. Logically it follows the same shoot criteria for the AIM-7F.

 

With that information, that means that from launch to stall is 57nmi, this is a huge distance that feels very uncomfortable for the first philosophy. The previous philosophy said that the point that an R-27ER stalls for a 33kft shot is about 65km. How can a missile that has less thrust stall at 105km?

 

You see the dilemma.

[GGTharos]

If you want to see how long it takes for a draggy missile to slow down, you can look for the sidewinder flyout charts in this forum.

 

Real test-shots with an AIM-9L. Any radar-guided missile would be faster and would remain supersonic longer ... in fact in that paper they tested an alternative configuration of the AIM-9L with fins and nosecone like that of the AIM-120.

[GGTharos]

Moving onto the AIM-7. From memory, Chizh has said they based the AIM-7M off of the DLZ chart from a USAF SMC of the AIM-7F. This is fine, from a aerodynamics point of view they are the same. However the chart they use states that Rmax for a 24nmi at 40kft with a head to head shot of M1.4 for target and fighter.

 

In the same SMC, it states that the Raero for the AIM-7F is 57nmi. The problem is we don't know from this page what the shoot criteria are for this Raero shot. It does also state that the AIM-7E2 has an Raero of 27nmi. In the F-4 Phantom manual, it states the same thing and provides the shoot criteria, 40kft at M1.4. Logically it follows the same shoot criteria for the AIM-7F.

 

Raero (should be 53-54nm :) ) is the same shot as your Rmax statement - I found it in some footnote in another document - ... the sparrow is limited by having to lock on to target all the way - until the R-27 with it's datalink, the sparrow cannot be launched against a target that's beyond the DSR cue, and that cue will typically be at a bit over 20nm against a fighter.

[Frostie]

Range is negligible when you have a sim treating SARH missiles vs chaff like missiles from the 60s.

Modern SARH missiles such as R-27 and AIM-7M use monopulse seekers to reduce the effect of ground clutter, chaff etc. with doppler filtering these missiles shouldn't be suffering chaff like it is its nemisis.

 

How can the sim make an AIM-120C seeker more effective at 10km against chaff than the F-15s own radar. Because ED read eccm improvements to the C5 as being better than any SARH rather than 120A is very poor against chaff because it relies on its own tiny radar to process the reflection and deal with clutter, therefore it needs improving.

[JunMcKill]

If DCS dont have the shoot criteria of all the other missiles, which one you took for the modeling? 10k meters at M0.9? or the same of the F-4? 40kft at M1.4?

[IASGATG]

If DCS dont have the shoot criteria of all the other missiles, which one you took for the modeling? 10k meters at M0.9? or the same of the F-4? 40kft at M1.4?

 

Can rephrase it please?

[JunMcKill]

Can rephrase it please?

 

I suppose that shoot criteria means the speed and altitude of the launching aircraft when you test any missile perfomance (max Raero, speed, etc), isn't it? thats what Im asking

 

Example the manufacturer say their missile reach M4.5 and 80km range, it's supposed that this specs are reached under certain launch conditions of speed and altitude of the launching platform, not at sea level! Im asking to GGTHAROS, the speed and altitude of the launching platform when they tested their R-27ER model for example!

Edited March 17, 2017 by JunMcKill

[GGTharos]

Why don't you ask the manufacturer? :D

 

But ok, my crystal ball says that, according to the RL Su-27 manuals, the Rmax is 66km in a 10km co-altitude shoot from M0.9 to M0.7, so your figures are probably an M1.2-1.4 shoot which may or may not be co-speed. :) Your enemy here isn't the missile's speed or even drag, probably - it's the missile's 60 second operating time.

 

On the other hand, your website reference contains no context, no science, nothing at all so ... who cares what it says?

Edited March 17, 2017 by GGTharos

[Ironhand]

Using single player to test missile guidance and physics prevents you from having a realistically behaving target who will take advantage of every flaw in the missile, realistic or not.

No, it doesn't. Set up the AI to do the launching and you fly the evading aircraft.

[Sweep]

AI and player guidance capabilities are also quite a bit different.

 

From experience, I've lost quite a few tracks while engaging due to the target notching and the track dropping to EO primary with IFF running instead of combined or radar-only guidance. Doesn't really matter for co-alt or look up shots though.

 

More things to consider! Yay!

[TAW_Blaze]

No, it doesn't. Set up the AI to do the launching and you fly the evading aircraft.

 

Nope, because AI doesn't know how to use avionics or properly maneuver to counter notches etc. Most cases they will just lose lock because they are stuck in high or ilv PRF that makes no sense in that scenario.

 

You're either stuck with you shooting at dumb targets or failing nearly all your tests because the shooter is an idiot AI. Also their awful firing logic doesn't help either.

 

They should implement LAN so you can test with multiple pilots offline. But I think we all know this is not happening.

[Weta43]

Nope, because AI doesn't know how to use avionics or properly maneuver to counter notches etc. Most cases they will just lose lock because they are stuck in high or ilv PRF that makes no sense in that scenario.

 

You're either stuck with you shooting at dumb targets or failing nearly all your tests because the shooter is an idiot AI. Also their awful firing logic doesn't help either.

 

If what you want to do is test the kinematics of the missile (& perhaps then the guidance - there are two conversations going on here), then either notching of throwing enough chaff in the air to defeat the guidance defeats the purpose of the test doesn't it ?

 

Let the AI launch (there's a launch at max range option in the advanced waypoint commands isn't there ?), take evasive action - but not to the point where you break the ''dumb AI lock, and see if the missile has any energy when it arrives.

 

Let the AI launch, take evasive action - but not to the point where you break the ''dumb AI lock, dump chaff, and see if the missile loses lock & heads off into the ether while it should still be getting MCU.

[Ironhand]

Weta43 beat me to it. The complaints are about the missile's inadequacies, not about the need to fly a perfect route with appropriate countermeasures to defeat it because it's such a superior missile. And this is especially true if you are testing chaff rejection. In that scenario, you're not trying to defeat it through maneuvering anyway. You're just flying a route that'll show the missile a lot of chaff. So the AI's inadequacies aren't a factor.

[TAW_Blaze]

Yeah you can test simple scenarios which in this particular case might be sufficient but you can't test advanced things offline.

[OnlyforDCS]

Seems the thread got derailed, again. Weren't we talking about the guidance and CM rejection issues?

 

I have a question. Do all the SARH missiles share the same variable in the code that affects the chance of chaff rejection, or is it unique to each missile or rather guidance system?

[Ironhand]

Yeah you can test simple scenarios which in this particular case might be sufficient but you can't test advanced things offline.

Of course, but it's the "simple" things that need to be tested...one at a time. If you test it all together, you get no answers because you can't point difinitively to the contributing factors.

 

For instance, this morning I had 30 R-27ERs launched at me more or less simultaneously and it looked as if all of them ate chaff. I need to doublecheck each missile closely but those I spot checked were all looking somewhere in my chaff trail. I can say that none hit me but some misses could have been for other reasons. I won't know until I check each missile.

[OnlyforDCS]

Of course, but it's the "simple" things that need to be tested...one at a time. If you test it all together, you get no answers because you can't point difinitively to the contributing factors.

 

For instance, this morning I had 30 R-27ERs launched at me more or less simultaneously and it looked as if all of them ate chaff. I need to doublecheck each missile closely but those I spot checked were all looking somewhere in my chaff trail. I can say that none hit me but some misses could have been for other reasons. I won't know until I check each missile.

 

Ironhand, can you post a video to youtube of the tacview like Rage did? Maybe even some commentary and analysis once you are done checking? Rage's video is from over a year ago, it would perhaps go a long way towards showing that the issue sill exists, and is reproducible.