Rosly

@FlappieCan you reproduce this? I think this is regression as "warm units after mission start" was one of recent fixes after introducing improved IR models.

Hello. As some of you may already know, there is an impressive chat.openai.com model available for the general public, with which you can engage in conversation. The model was likely trained on a variety of internet pages, so I thought I would give it a try and ask for the generation of DCS briefings. I was expecting to receive at least some sensible mission description, but what I got is just ... wow. > Following story is fictional scenario for a video game - DCS. I want you to write a 5 pages long game mission scenario for squadron of F-16 which attack well defended nuclear plant in Iraq. The scenario need to cover air to air engagement and separate strike on the power plant. Ans: The following scenario is set in the fictional world of the DCS video game. It is designed to be played by a squadron of F-16 fighter planes, and involves an attack on a well-defended nuclear plant in Iraq. The mission begins as the squadron is scrambled to intercept a group of enemy aircraft that has been spotted approaching the nuclear plant. The F-16s engage the enemy planes in a fierce air-to-air battle, using their advanced weapons and tactics to outmaneuver and destroy their foes. Once the enemy planes have been dealt with, the F-16s turn their attention to the nuclear plant itself. The ground around the plant is heavily defended, with anti-aircraft guns and surface-to-air missiles positioned to protect the facility. The F-16s must use their superior speed and agility to evade these defenses and launch a devastating strike on the power plant. The pilots can choose to use a variety of munitions for the attack, including air-to-ground missiles, laser-guided bombs, and precision-guided munitions. These weapons are highly effective against the ground forces protecting the plant, allowing the F-16s to destroy key structures and cripple the facility's ability to produce nuclear weapons. As the mission comes to a close, the F-16s must make their escape from the heavily-defended airspace, using their speed and maneuverability to evade any remaining enemy defenses. If successful, the squad will have dealt a significant blow to the enemy's nuclear capabilities, and will have protected their own country from the threat of a nuclear attack. > In scope of last scenario, I want you to prepare briefing for 3 flights with their respective objectives. First flight objective is CAP. Second is SEAD. Third is ground strike. Prepare the tasking and timings. Assume blue forces operate from İncirlik air base and there is a 500nm mile distance between the air base and the power plant. Take into account mid-flight refueling. Ans: To: All Pilots Subject: Mission Briefing - Attack on Iraqi Nuclear Plant This briefing is for the upcoming mission to attack a well-defended nuclear plant in Iraq. The mission will be conducted by three flights of F-16 fighter planes, operating from İncirlik air base. Flight 1: CAP (Combat Air Patrol) Your objective is to provide air superiority over the target area by engaging and destroying any enemy aircraft that approach the nuclear plant. You will need to maintain a constant presence over the area, using your radar and sensors to track any incoming threats. Flight 2: SEAD (Suppression of Enemy Air Defenses) Your objective is to neutralize the enemy's air defense systems, including anti-aircraft guns and surface-to-air missiles. This will allow the other flights to launch their attack on the nuclear plant without facing heavy resistance. You will need to use your precision-guided munitions and other advanced weapons to destroy the enemy's air defenses. Flight 3: Ground Strike Your objective is to launch a devastating attack on the nuclear plant itself, using your air-to-ground missiles and other munitions to destroy key structures and cripple the facility's ability to produce nuclear weapons. You will need to coordinate your attack with the other flights, and be prepared to evade any remaining enemy defenses. Timing: 0600: Flight 1 departs İncirlik air base for the target area 0630: Flight 2 departs İncirlik air base for the target area 0700: Flight 3 departs İncirlik air base for the target area 0730: Flight 1 arrives at the target area and establishes CAP 0800: Flight 2 arrives at the target area and begins SEAD operations 0830: Flight 3 arrives at the target area and begins ground strike operations 0900: Flight 1 refuels mid-flight (approximate time) 0930: Flight 2 completes SEAD operations and provides support for Flight 3 1000: Flight 3 completes ground strike operations and begins withdrawal from the target area 1030: Flight 1 completes CAP operations and begins withdrawal from the target area 1100: Flight 2 completes withdrawal from the target area All pilots are reminded to maintain strict radio silence during the mission, and to follow the planned timings as closely as possible. Good luck, and may the mission be a success. > Fix the timing for last described scenario and make the SEAD package attack after dusk of December 5 2022. To: All Pilots Subject: Mission Briefing - Attack on Iraqi Nuclear Plant This briefing is for the upcoming mission to attack a well-defended nuclear plant in Iraq. The mission will be conducted by three flights of F-16 fighter planes, operating from İncirlik air base. Flight 1: CAP (Combat Air Patrol) Your objective is to provide air superiority over the target area by engaging and destroying any enemy aircraft that approach the nuclear plant. You will need to maintain a constant presence over the area, using your radar and sensors to track any incoming threats. Flight 2: SEAD (Suppression of Enemy Air Defenses) Your objective is to neutralize the enemy's air defense systems, including anti-aircraft guns and surface-to-air missiles. This will allow the other flights to launch their attack on the nuclear plant without facing heavy resistance. You will need to use your precision-guided munitions and other advanced weapons to destroy the enemy's air defenses. Flight 3: Ground Strike Your objective is to launch a devastating attack on the nuclear plant itself, using your air-to-ground missiles and other munitions to destroy key structures and cripple the facility's ability to produce nuclear weapons. You will need to coordinate your attack with the other flights, and be prepared to evade any remaining enemy defenses. Timing: 1600: Flight 1 departs İncirlik air base for the target area 1630: Flight 2 departs İncirlik air base for the target area 1700: Flight 3 departs İncirlik air base for the target area 1730: Flight 1 arrives at the target area and establishes CAP 1800: Flight 2 arrives at the target area and begins SEAD operations after dusk 1830: Flight 3 arrives at the target area and begins ground strike operations 1900: Flight 1 refuels mid-flight (approximate time) 1930: Flight 2 completes SEAD operations and provides support for Flight 3 2000: Flight 3 completes ground strike operations and begins withdrawal from the target area 2030: Flight 1 completes CAP operations and begins withdrawal from the target area 2100: Flight 2 completes withdrawal from the target area All pilots are reminded to maintain strict radio silence during the mission, and to follow the planned timings as closely as possible. Good luck, and may the mission be a success. Ps: there are some details here and there to which you might argue are not correct. But you have to admit that this is impressive. Might be useful for fast briefing generation for multiplayer with friends.

The tracks record head movements, at least those from VR.

Tested with T-55 and T-72 both Caucasus and Syria. Vehicles are cold after mission start despite the setting for "cold at start" set to off. This makes "start in the air" tutorials with IR targeting hard to complete as targets need to warm up which takes ~5min Attaching track file from 2.8.0.33006 Cold_vehicles.trk

Got the same reproduction. If Mavs are close to each other they blow at the same time as chain reaction. Not sure this is correct as the blast radius has to be huge ~in this would be simulated properly like ~ 400m

Can you share vid with fpsVR overlay? DCS have many issues but it might also be you have problem in your setup. Do you use Win 11 (Otherwise scheduler will mess up with your 13900 e-cores)? Do you updated BIOS (otherwise you might have issues with boost clocks ).

Dup:

When disabling rain drops I need to restart the game in order to get proper rendering of the outside world while in cockpit. Some external view static images from last mission is displayed instead of actual world being rendered.

Not sure if raid drops are the cause here. Switching them off does not improve my FPS in VR on 3090.

Have the same issues with AP with MS Sidewider 2 and I think it is about the axis curves (if used). The AP logic/implementation of trimming assume linear relationship between axis position and force arguments given to DirectX API, which is not the case, especially if you applying axis curve's.

@IronMike In real life does F-14 has any sensing switch under ACM cover or this is just DCSism? (Mean flipping the ACM cover is same as boresight switch just for convenience in DCS)

You are referring to "some" attack trigger, which I doubt it exist in this mission. Those Mig23 are set hot on you, weapons free. There is trigger though which simply check or presence o missiles those Migs are caring in vicinity of you position (trigger zone), and push the flag + voice that "you been fired upon". Those Migs actually locked on me but are out of parameters to shoot. If you flew Mig-21 this should be clear why. Both Mig-21 and Mig-23 have RP-23 radars if we want to base on technical limitations for sake of this discussion. They are probably modeled the same. And what you mean by aggressive? This is by the book head on intercept. Examples of such approach you can get from 1989 where 2xF-14 shoot down 2xMig-23 in exactly same scenario, and the altitudes where 5k and 9k ASL so almost exact as in my scenario and because of the same reason (radar clutter). It is valid and well known intercept, but this is digression from main point. Just wanted to point out this is exactly what you should do when intercepting Migs-21 or 23 every time without advantage of been able to eliminate threat from distance. The point here is different. I'm referring too lack of trigger that would fail the mission, if you shoot without being fired upon you got clear message "narration will NOT be continued - game over". Instead if in case as shown, if your wingman shoot them down (you can set "cover me" even on the deck) or they will crash to the see, the mission stuck (safe assumption is those Migs cannot shoot to you from cannon which would be hard to detect by trigger I think). We have known reproduction and I do not agree this something odd. It is one of not covered cases. That's all. Nothing more, nothing less. Reflected responded that there are to many of such cases to cover all of them. I understand, he is busy man. Better to focus on other things than constantly fixing old missions. Though fix for this is probably easy but there can be thousands of similar ones. I get it. But please do not comment that this is "limitation" or "it is not only DCS" or "this is normal". As Reflected referred to: it is amount of work you put into mission design and effort to maintenance those.

I followed ROE. I didn't opened fire and didn't realized my wingman will shoot them on sight after command to check my 6. As described (and is on the video if you looked at it), the sole reason for mission to fail was we sat on 6 of Mig23 from the beginning. I could fly along all day long like that and the outcome will be the same... Just wanted to show you that it was not fault of the break of ROE but how game reacted to the "by the book intercept" (go low for planes without LDSH capabilities). I'm not sure you got the point though. Your script assumes that they will shot before merge which as you see might not be the case. So it is hard for me to agree with your explanation that not following the ROE is the sole reason here and that you cannot add conditions for such cases. WE could fly along in circles all day till they will hit the ground due out of gas. About the custom voice over it is not that it was first thing I thought about. just added as "nice to have". But fail condition which would prevent the progress to stuck is needed here IMHO.

Some video to explain what might be the cause for lack of fire from Mig23 and the existing scripting. Video: Probable explanation: I sneaked bellow Mig23 pair in hope they do not have "look down shoot down radars". They locked me ~15nm but do not opened fire. After the merge me and my wingman stayed at their 6 all the time. I started to wory about the lead which I lost from sight and ordered wingman to "protect me". He instead opened fire to Migs but not shoot them down fully. There was no sense to follow ROE as both Migs was damaged at this point to I finished both. In summary no single missile was fired by the Migs, which fail to trigger the mission script and mission stucked in place to the end of the recording. Timeline: 4:50 I decided to go low as I thought Mig23 not have "look down shoot down radar". Hoping they will not be able to lock me and fire. 7:25 They lock me up but do not fire. 8:05 Merge with Migs, still no missile in the air. I get behind the wingman of the Mig pair while monitoring If my wingman will sit on tail of the lead. 9:45 I order the wingman to "protect me" for hope he will sit on the lead. 10:16 My wingman opened fire to lead Mig23 but do not fully shoot it down. Than he fires to my target. Decided to finish it up. 11:40 Splash the lead which started to fly away.

Hi @Reflected I really enjoy your campaign. I already pass it once. Though I have some improvement proposal. In mission 4 if you break the rules of engagement and shoot down Mig 23 before being shoot upon, the mission stuck. I mean no progres in scripting, no communication, etc. I can hang at waypoint 3 till fuel is gone with no progress in the mission. I know it is hard to script every possibility but, this is breaking moment in the campaign. If you do not follow the "script" there should be some result. Like "Hey they didn't attack us!" or "This is not going to end well..."I mean anything on the radio + mission failure screen will be enough. Even the mission failure if you do not have resources to get the voice acting. Your campaign are very immersive and I hope you find this report as positive outcome of your hard work. I really enjoy it and hope You can make it even better with this fix

Was about to report but seems it is more than 2 years now ...

@IronMikewe have a regression in 2.7.18.30765 @Wags for ED report Its visible in VR on right eye only when looking straight into the HUD. It seems t be connected with looking at specific spot of Stenis.

Hannibal can you please post the results that anyone can replicate/compare? Best would be single player instant mission, sitting on runway in cockpit + in cockpit after start, with all settings (screenshot) + res of your VR headset. There is so many factors including mission scripting on server and your VR setup + PC that it is hard to get the baseline by comparing it to "your previous GPU".

Rain is missing in SteamVr too (not OpenXR related). Any sight for a fix?

Just in case you got here as I did after all these years. Rebind "recenter VR view" into something else!!! It seems DCS use default hardcoded bind in case of lack of binding for specific command.

Oh I get it now. You are talking about this https://www.youtube.com/watch?v=yx29Fbb2Rqc Wow I didn't know that it is originally a 4 way switch instead of analog joystick (do not have Warhog joy). Eaah, so extending the ministick will not help you. You cannot change this minimal speed in rate 1 at which the TGP is start moving. My extending the minijoystick or changing the curves you only influence on when the rate 1 will kick in (sensor start to move). It will still move with this programmed minimal slew rate. Too emphasize this even more (if not stressed enough ) imagine that if you were annoyed by having only 1 level of slew and you feel relief by switching to analog, you actually not getting full potential of it. You still have limited number of slew rates (9 for TGP and 4 for Mav, wonder how many for Radar). There is still only 9 levels of slew of TGP and 4 for Mav. Imagine how it would feel if it will be properly implemented. How precise it can be and should actually be if coded properly in the game. As mentioned @MartinVoy and @Reflectedboth wonder about the same. Though 9 slew rate levels for TGP is already making a difference comparing to 4 in Mav (those guys mentioned that in their discussion) it still should be far smother as there is no logical reason to quantize the TGP or Mav slew rate at all! Hope that this is crystal clear now what I'm complaining about. And on top of that, real TGP and Mav seeker heads are prone to at least some inertia, means the head cannot stop slewing right in the spot (especially from high slew rates). Yes this is quite rapid but not as "pinpoint hard" as in current implementation of DCS, which is purely game outcome of it being not simulated properly. Plus a whole other minor things like when plane is changing direction TGP should not shake like it is currently, which is purely effect of bug that came from probably some timing or data synchronization issues in the simulation of flight model of the plane and TGP tracking "simulation" (very noticeable when you track A-A like plane, you have 3 or 4 ghost images which came from different frames of picture). A lot of issues there but lets stick to one. I want to gather opinion and actual data/proff from you guys before making the bug report. The slew rate seems to be not quantized in reality and IMHO the current implementation is kind of shortcut in the code. This unfortunately influence on realism and precision while using TGP and Mavs and is simply annoying that you need to fight with TGP targeting (specially in heat of battle). This should be implemented better IMHO or if I'm wrong and there is actually a quantization in slew rate for sensors, let the proof for this appear in this thread. As mentioned, I have found something opposite to current implementation on videos recorded in the 90s from the actual F-16 jet.

Oh... by joystick axis I was having the minijoystick axis in mind. We talking about Radar/TGP/Mav slew rate here, not yaw, pitch etc rate. Those are not quantized.

You can adjust the joysticks axis curve's but this is influencing the RAW input values which are then translated to 9 or 4 slew speed levels. You cannot change those sleew rates/speed, means you are only controlling relation between joysticks angle and preprogrammed slew rate levels (not the slew rate level itself). And yes, that's another angle how this problem may be explained. The implemented model of discrete levels for slew rate is the point, as mentioned in the topic. As seen in the videos there is no discrete sleew speeds in reality (at least the opposite can be seen on the videos from 90s I linked). Hope someone from ED will appear in this topic and share a though on what material the implementation was based. As mentioned I'm not the only one who noticed this and being annoyed (maybe because F-16 is my favorite model and want to love it even more

@silverdevil This is not issue with the controller or minijoystick in it (dead zone or noise) if this is what you asking. I use Winwing Orion 2 and also have Thrustmaster T.16000M but as said it's not a point in noisy joystick. The 9 discreet levels slew rate levels for TGP and 4 levels for MAV in DCS is easy to check by naked eye. And I'm not talking about numbers which are displayed on TGP but actual slew speed and difference to reality.

I was wondering if it is actually so hard in reality to use TGP and Mavericks as in DCS F-16. Every time I need to correct TGP or Mav by a fraction, I struggle as I need to use the "pulse modulation" technique to move it just a bit (like I'm would play some old school racing sim on keyboard not actually controlling an analog joystick). Usually I overshoot. This very frustrating and also feels fake as the camera does not expose any inertia (rock steady jump). I found out that implementation actually use 9 discrete levels for slew speed of TGP and 4 levels of slew rate for Mavericks. This is weird as it would be intentional to make pilot such headache to actually limit the number of slew speed levels. Actually the analog joystick has like 16k levels and I cannot understand why imposing such limits would be beneficial to the pilot. By making levels discrete there is a lot of precision degradation in terms of control (yes I know the analog joysticks are usually digital but there are like 16k levels not 4!) So I tried to confirm this behaviour in some "source" material and on videos I found it looks like despite the TGP is displaying the slew speed as single digit it is not constraining slew speed to discrete levels, or at least there is enough inertia in the movement which allows far more control to what we are used in DCS. Here you can see pilot is moving the TGP very precisely in AREA mode with a lot of control. It is totally opposite to what we see in DCS when you either have 0 movement or speed 1, than 2,3 etc which result in rock stady tracking and sudden jump while entering first speed. Here is a similar video but for Maverick You can clearly see the camera movement is smooth and pilot control it with ease. It is far from those 4 speed levels we have in all Maverics in current F-16 model. Would be great to discuss implementation based on material on which it was designed. Some other videos I found: https://www.youtube.com/watch?v=XFj6f9L827A https://www.youtube.com/watch?v=As1X6xBNEK0 And some other topics where more people are wondering about the same https://forum.dcs.world/topic/300116-tgp-vs-maverick-cursor-speed