Northstar98

Just to demonstrate that the problem isn't the Shrikes tracking the SON-9, here are 4 tracks tested from a player aircraft. In all 4 cases the Shrikes do acquire and track the SON-9 (albeit they have fairly dreadful accuracy, usually landing short). In these tests the SON-9 starts tracking the aircraft at around 12.5-13 nmi, the Shrike is fired around ~9 nmi. In all 4 I've used the WRCS Shrike mode, with the guidance sections set to loft attack, using a lofted profile. The problem I'm encountering is with the AI, where the AI seemingly makes no attempt to engage the SON-9, as they do with other radars, regardless of whether or not the SON-9 is in search or track. They simply overfly the SON-9, as if it isn't there and proceed to follow their waypoints. The same mission as above works with nearly every other radar that falls within the frequence range the Shrike's guidance sections can target. AGM-45A_SON-9_Mk50_WRCS_loft.trk AGM-45A_SON-9_Mk24-34_WRCS_loft.trk AGM-45A_SON-9_Mk24-5_WRCS_loft.trk AGM-45A_SON-9_Mk23_WRCS_loft.trk

This is the case in the tracks above - the SON-9 can be seen to be tracking the target. Typically, I observe the SON-9 begin tracking at around 12-13 nmi. Attacks with the Shrike by the AI, from a starting altitude of 18000 ft, typically fire at a range of 4-5 nmi. To be clear - this is not a Shrike failing to track problem - this is an AI refusing to engage problem. The AI simply overflies the target and lands, not even attempting to fire. There are some cases where an AI-fired Shrike doesn't track the targeted radar, but here the problem is that the AI doesn't make an attempt to engage at all. And, for what it's worth, the Shrike is indeed capable of tracking radars that are rotating:

Then the real problem here is that AI groups don't communicate and coordinate with each other (i.e. the cornerstone of any IADS functionality). Right SAM groups in DCS should only be battery-level units - it doesn't support having multiple batteries within the same group. It doesn't know which units should be part of one battery or another and treats the whole thing as if it's one - leading to a multitude of issues, some of which you've described. The actual thing we should be advocating for is the ability to set up communications and/or data links between AI groups and have them coordinate with each other.

Why should it be able to fire at more targets with another C2 (FDC) unit? Why would a NASAMS battery have 2 FDC units to begin with?

Hi everyone, Could we have an option to have the AI fire anti-ship missiles bearing only? Allowing us to define a bearing and range to an activation point. A way of accomplishing this could be something similar to the fire at point task, only the point would define where the missile should activate. There should also be the usual string of quantity settings (though those currently don't work as is for ship/submarine-fired anti-ship missiles). This would however probably require a rework to most anti-ship missiles in DCS, as quite a few are either only defined with either simple seekers (such as most of the Chinese asset pack weapons, as well as the AI RB 04E and RB 15F) or don't have seekers defined at all (such as the AGM-84A, Kh-22N, Kh-31A, Kh-35, Kormoran, P-270, P-500, P-700 and the RGM-84D (despite the AGM-84D having a seeker defined - the 2 missiles share the exact same seeker)). EDIT: It seems that even missiles with no seeker definition in the files will search for and choose another target if the first one is lost (for instance, via group deactivate), so, hypothetically, they should work as-is. This could also work for torpedoes, provided we get guided torpedoes implemented (the Yu-6 and Mark 46 should already have active/passive acoustic homing, as well as wire-guidance for the former). All this would also be useful to have in the combined arms interface.

Hi everyone, At the moment, when a ship comes under attack the AI makes no attempt to manoeuvre defensively and accelerate, be it to try jinking to avoid fire from naval guns or trying to make a bombing run just that little bit harder, or to unmask defensive weapons. This not only makes ships easier to hit, but it also is something that's very easy to exploit and game to increase your chances of successfully engaging a ship, sometimes significantly so, simply by attacking from a direction where the ship's more effective defensive weapons are masked. It would be better if ships would accelerate and jink to avoid weapons fire and to manoeuvre so to unmask defensive weapons systems. The same could also apply when engaging surface targets with guns (and some AI ships already manoeuvre to engage targets with anti-ship missiles or torpedoes). Ideally, there would be some decision making logic for the AI, so that it can deal with multiple threats, taking into account the speed at which the ship can turn, the direction of the threat, the arcs of weapons and the threat's ETA. Perhaps how fast the AI reacts and how effectively they open up firing arcs could be determined by the skill level. Like aircraft, there should also be a setting in the advanced waypoint options/triggered actions to control whether the AI should react or not. As an example, I've got 2 tracks below where I have an OHP firing an RGM-84 at a single Tarantul III. In the first track, the incoming missiles[note 1] approach from its port quarter and the AK-630s successfully manage to defend the ship (despite the FCR for said AK-630s being masked). In the second track, I have the missiles approach from directly ahead - despite the ship definitely detecting the threat (when the Tarantul III switches from alarm state green to red, shutters close on 4 of the bridge windows) it makes no attempt to unmask its defenses. Seeing as the AK-176 in DCS, like almost every other naval gun, is only capable of engaging surface targets[note 2], this renders the ship as good as defenseless from attacks from this direction. Numerous ships also have blind zones where certain defensive weapons cannot engage, or would improve their chances of successfully defending themselves: The Grisha V has a close-in weapons system and a naval gun on the stern, which cannot engage directly forward. The ship would be able to better defend itself if it would turn to unmask these weapons (especially when the SA-N-4 only has a single target channel and only 2 missiles ready to fire before needing to reload, making it relatively easy to saturate). Conversely, the SA-N-4 system cannot engage targets approaching from astern. The Oliver Hazard Perry and Invincible have SAM systems that also cannot engage approaching from astern. The OHP also has a CIWS (and a naval gun) that cannot engage targets directly ahead. Incidentally the STIR is masked when firing SM-1MRs (the SAM the OHP should be firing) directly forward (though the Mk 92 FCS can also provide an illumination channel forward and the STIR doesn't even exist in DCS, when it absolutely should if the OHP has Mk 13 GMLS). The Krivak II could perform double the number of intercepts with its SA-N-4 systems, if it places the target on the ship's beam. It would also unmask the 2 AK-100 guns (if they would engage airborne targets). The Slava's close-in weapons systems have an aft blind zone and the SA-N-6's FCR is masked directly forward at low altitude, the SA-N-4 is also masked directly forward. Placing the threat on the ship's beam allows all 3 weapon systems to engage, maximising the chances of interception. Notes: While the behaviour has changed somewhat, AI ships still don't respect weapon release settings in terms of quantity, attack quantity or group attack. See this thread. Nearly every naval gun currently in DCS should have dual purpose capability, often firing dedicated rounds. The AK-176 can fire the ZS-62 projectile, which contains 400 g of AI-X-2 explosive, with an AR-51L radar proximity fuse that functions up to 8 m away from the target. The AK-100, AK-130, Mark 75 and Mark 45 guns all have proximity-fused rounds available. Unfortunately, with a single exception, they all are only capable of engaging surface targets, with a high-explosive, impact-fused round. See this thread. Tarantul_III_RGM-84_attack1.trk Tarantul_III_RGM-84_attack2.trk

Likely on ED - the CBU-52/B having blue stripes has been a problem for ages:

https://cat-uxo.com/explosive-hazards/aircraft-bombs/samp-type-25-aircraft-bomb

Yep, spot on. Absolutely agreed on every single point.

+1 I'd certainly like a KC-10. One thing I will say though, at the moment tankers in DCS only support a single refuelling type exclusively (which is why the KC-135RT w/ MPRS only supports probe and drogue, but not the boom). It's not the only limitation but this one in particular is relevant to the KC-10 as the KC-10 supports both flying boom and probe and drogue without modification. If this limitation isn't resolved, this will take away some of the utility of the KC-10 (unless you have 2 duplicate aircraft in the unit list and make missions with 2 tankers to support both refuelling methods).

As of DCS 2.9.8.1107 the AI will now use Mavericks, though the issues in the previous post still apply (though the bug with the Paveway III applies to all AI aircraft, not just the F-4E).

Hi everyone, With the recent weapon editions to the F-4E, the list of available bombs on stations 2 and 8 has completed filled the usable space to the point where some options are inaccessible without first weapons restricting. On my end, everything past the GBU-10 Paveway II is unaccessible, which includes the GBU-12 Paveway II, GBU-24A/B Paveway III the GBU-8/B HOBOS and the Mk 84 AIR - these weapons cannot be selected without first restricting weapons. Before weapons restricting, note how I cannot access any weapons below the GBU-10: After weapons restricting (in this case I restricted out the BL775 entries and the SAMP Type 25 entries): If there was a scrollbar I wouldn't need to restrict weapons in order to access these weapons.

Unfortunately not I'm afraid - I did attempt to find something online but was completely unsuccessful.

Thank you - though just for clarity, many of the issues brought up in this thread have now been fixed, it's only really the GBU-8 HOBOS without Pave Spike and a couple of radars that the AI refuse to engage using the AGM-45A. EDIT: I also forgot to link the general AI bug mentioned, which refers to AI employment of the Paveway III (they seemingly calculate for a purely ballistic trajectory, as would be expected with say, an unguided bomb - the Paveway III series in DCS performs a bump-up which means, when dropped at low altitude, the bomb lands long), please see this thread.

Hi everyone, The F-4E, when flown by AI, is able to provide guidance to multiple AGM-12s independently - this is something that shouldn't be possible. There is only one control (available only to the pilot) that provides a means to send steering commands to an AGM-12 - if multiple AGM-12s are in-flight simultaneously, they should all receive the same commands, it shouldn't be possible to provide separate commands to each missile. This behaviour can be seen with player aircraft - if a player has multiple AGM-12s in-flight simultaneously, they all receive the same commands and perform the same manoeuvres - controlling each missile independently isn't possible. This is most obvious with the AGM-12A and B as there's a significant delay between subsequent firings compared to the C. There you can clearly see that missiles are being guided independently (usually when a subsequent missile is fired, you can see the AI steering that missile onto target while the previous missiles receive no/much smaller steering corrections, which can be seen in the 2 tracks below). AI_AGM-12A_multiple_independent_guidance.trk AI_AGM-12B_multiple_independent_guidance.trk

As of 2.9.8.1107, it looks like the AI now uses AGM-65s (all versions) without issue - great to see! It looks like this bug is now mostly resolved - the AI will now employ near enough every AG guided weapon as expected, though there are still some outstanding issues (one of them however is a general AI bug) though they should probably be their own thread (as the AI will fire them). However: The AI still seemingly requires Pave Spike to be equipped in order to employ the GBU-8/B HOBOS (see above). This isn't a problem that affects either of the Walleyes. The AI won't employ the AGM-45A Shrike against the SON-9 or AN/MPQ-64F1, regardless of guidance section set. I'm not sure whether or not the AN/MPQ-64F1 is correct as-is, but the SON-9 definitely isn't - it is explicitly mentioned as an intended target for the Mk 23, Mk 24 Mod 5 and 34 guidance sections. AI_F-4E_AGM-45A_Mk23_SON-9_NoFire.trk AI_F-4E_AGM-45A_Mk24-5_SON-9_NoFire.trk AI_F-4E_AGM-45A_Mk24-34_SON-9_NoFire.trk AI_F-4E_AGM-45A_Mk50_SON-9_NoFire.trk

Looks like this is still an issue as of 2.9.8.1107 The AIM-7E-2 uses the AIM-7E model Both the AIM-7E and AIM-7E-2 transform into an AIM-7M when launched (as far as artwork goes). Both these issues can be fixed with small edits to Weapons.lua found in CoreMods\aircraft\F-4E\Entry: To fix issue #1, change "HB_F-4E_AIM7E" to "HB_F-4E_EXT_AIM-7E2" on line 575. To fix issue #2, change "aim-7" to "HB_F-4E_AIM7E" on line 324 and to "HB_F-4E_EXT_AIM-7E2" on line 524. I've attached a modified Weapons.lua file that fixes both issues (tested with both player and AI aircraft), this will however break IC so be sure to back up your original. F-4E_AIM-7E_model.trk F-4E_AIM-7E-2_model.trk Weapons.lua

Doesn't take away from your overall point, but shouldn't they display as an S, like other search radars? Ditto for the SA-10 S-300 "Tin Shield" - i.e. the exact same radar but on a 40V6M mast. Especially when, IRL, no version of the Tin Shield is associated with any version of the SA-5 - it's first a general-purpose EWR (which is what you'll mostly see it used as IRL) and second an alternate acquisition radar for the SA-10 (or at least some versions of it), with the primary acquisition radar being the Big Bird.

Yeah, out of all the Pave Tack footage you can find online, it's easily the funniest

New digital inertial navigation and attack system (storing 99 waypoints) (AN/ARN-101) CCIP capability, with both immediate and delayed release, using the exact same gunsight we have now. AN/AVQ-26 Pave Tack (first generation FLIR pod, a bit fancier than Pave Spike but not by a whole lot) - here's a video. TISEO - practically the same as TCS on the Tomcat - here's a video. GBU-15 - you can kind of think of this as more of a USAF Walleye II ERDL. It has Mk 84 and BLU-109 warheads available but I couldn't see BLU-109 versions listed in a 1990 -1. The DMAS Phantom can also equip the AN/AXQ-14, which supports man-in-the-loop guidance via data link and lock-on after launch capability. It also supports cooperative engagement capability, but that's probably something for ED to do, as their current aircraft don't support it, even though they should. LORAN navigation.

Hmm, I've had a play around with it - I can't see anything wrong with how you've set the tasking up. I was able to get success with the A-10C as an AFAC in my own mission (if you change the file extension to .miz from .trk you'll get the mission file), just using the FAC - Engage Group task, under "Start Enroute Task". The MQ-1A and MQ-9 pretty much always seem to give "No Mark", even though LOS should exist, they should be in-range and the laser shouldn't be masked by anything. I was able to briefly get a "marked by laser" message from the MQ-1, but I haven't been successful in replicating this. Personally, it looks like that there might be an issue with how the MQ-1 and 9s designator is defined - an identical tasking yields expected results with other units and as said above, the targets should be in-range, LOS to them should exist and the designator isn’t otherwise masked by anything. HMMWV_FAC_Mark.trk MQ-9_AFAC_NoMark.trk A-10C_AFAC_NoMark.trk MQ-1A_AFAC_NoMark.trk

If you go into your Saved Games\DCS\Missions you should be able to find your .miz file - to attach it, you simply drag and drop the file into the box that appears below where you enter text.

No to the former and yes to the latter, though that comes with some asterisks. The AI can use GBU-8, but for some reason only if the aircraft is equipped with Pave Spike. This shouldn't be the case, the GBU-8 is TV guided, not laser guided. It's also fire and forget and target acquisition has to be done via the GBU-8. The AI can independently guide multiple Bullpups simultaneously. I've no idea how it does this - it certainly isn't something players can do. If a player fires multiple Bullpups they all receive the same steering commands, it isn't possible to guide them independently. I wanted to test this with the AJS 37 to see if it was a general AI problem with regard to MCLOS missiles, but the AI doesn't want to use the RB 05As for some reason, for players though, only the last missile fired receives guidance. I'd recommend against using the Paveway III at low level for the time being, due to this bug, which causes them to drop too late, often leading to the bomb sailing over the target.

Something else for the to-do list (albeit minor) - the AA1 panel for the AN/APX-76 interrogator currently isn't present (blanked off). Of course, without wider IFF functionality (though RAZBAM have implemented a system that does everything required for their aircraft, which Aerges' F1 also uses) I'm not expecting functionality, but the equivalent for the F-4E (which is similarly non-functional outside of LotATC) - the AN/APX-80 (which includes the AN/APX-76), exists. The panel is rather simple and is described in HB's own manual. There have been mentions of it being planned but low priority on Discord.

Added San Carlos (not to be confused with Port San Carlos).