Multiple Radar/AWG-9/Phoenix Issues since patch 2.5.6.52437

[Kula66]

I find PAL very reliable and apart from in guns mode, gives a clear indication of sweep. Being a 15 miles scan, it gives distinct edge over fighters with a 10 mile scan ... the 3 second pause before 54 launch is a pain though!

[OnlyforDCS]

... the 3 second pause before 54 launch is a pain though!

 

There is a switch for that, on the front missile arming panel, that lets you fire the 54 right of the rail, almost no delay (maybe a second) radar-hot. I forget what it's called but it's very, very useful for "close" range BvR.

Edited September 8, 2020 by OnlyforDCS

[TLTeo]

That would be the ACM switch and yes, it shortens the delay to 1s.

[Hummingbird]

It isn't? Works fine for me as long as I am a) in A2A mode and b) using anything but guns.

 

Well it's not good if it aint working with guns :)

[Kula66]

It is working, but can't show you the gun pipper and the radar scan indicator at the same time!

[Kula66]

That would be the ACM switch and yes, it shortens the delay to 1s.

 

As I understand, this makes the 54 come off the rail active ... but would you want that at 15 miles?

[Hummingbird]

It is working, but can't show you the gun pipper and the radar scan indicator at the same time!

 

I know, but it seems way slower to pick up targets now, and it loses them a lot too for some reason.

 

Infact the AWG-9 loses targets extremely often ingame, it's by far the worst performing radar atm.

[Kula66]

I fly mainly SP, and only fly a little MP, and always with Jester - but I haven't noticed it being too bad ... do you use a human RIO?

 

It is a generation behind the other radars ... is it correct? I'm not qualified to answer that.

[Harlikwin]

I fly mainly SP, and only fly a little MP, and always with Jester - but I haven't noticed it being too bad ... do you use a human RIO?

 

It is a generation behind the other radars ... is it correct? I'm not qualified to answer that.

 

So I haven't been using the F14 online much but I did take it up for a quick hop yesterday, and really I'm far from being an F14 expert so its possible that it might have been "me". My experience using jester was the following: At least online its nearly constant loosing locks, tracks vanish, and jester can't even keep fighters STT locked, you can lock fire, and about 50% of the way there he looses lock in STT. Maybe the enemy fighters notched it, IDK. In TWS he couldn't keep anything locked and was constantly loosing contacts and also could barely IFF contacts, which made prosecuting a phoneix engagement impossible without telling him to lock contacts in STT (not enemy since he couldn't iff), which see above.

 

As for how realistic that is. IDK, maybe the enemy fighters were entering the notch to break locks/tracks, which well should work .

Edited September 9, 2020 by Harlikwin

[Kula66]

Hi Hawlikwin, do you review your flights with Tacview? Definitely worth the time and money.

 

Also, are you flying SP against an Ace AI??? They fly almost perfect evaison manoeuvres ... do you the same issues against 'Trained'??

 

I'm no expert, but generally use TWS and with TWS-A, I find it very effective, even against multiple bandits, unless they go for vertical separation which it struggles with.

[Bearfoot]

I fly mainly SP, and only fly a little MP, and always with Jester - but I haven't noticed it being too bad ... do you use a human RIO?

 

It is a generation behind the other radars ... is it correct? I'm not qualified to answer that.

 

The vast majority of these problems occur (for me) only on MP (with Jester RIO). Even the problems that occur on both SP and MP seem, to me, occur with a much high prbability in MP. SP play has been mostly trouble free for me. I do not believe that SP play (alone) is a reliable or useful way to replicate, assess, investigate, or fix these bugs.

[IronMike]

The vast majority of these problems occur (for me) only on MP (with Jester RIO). Even the problems that occur on both SP and MP seem, to me, occur with a much high prbability in MP. SP play has been mostly trouble free for me. I do not believe that SP play (alone) is a reliable or useful way to replicate, assess, investigate, or fix these bugs.

 

 

 

 

Exactly why fixing them poses such a huge problem for us. For one we cannot debug in MP. For the other, we don't have the time to fly in MP so much that we could experience them often enough. Mostly we're trying to reproduce them in SP, which doesn't really work.

 

 

 

One of the issues with TWS, since we dont grab the readily available information in DCS, but emulate the signals while pretending to not know the source (like a radar would), is latency. For example, if there is rubber banding, and the contact is moving backwards, that just doesn't work, the radar will not know what to do with a target that is moving backwards, drop the track, and when it sees the target moving forward, create a new track. That's just one of the issues we're busting our heads over. But stuff like this explains why some of the issues are apparent in MP and not in SP. Either way, we are on it and hope to find a solution soon. A fake recreated radar - which simply knows what is available as info in DCS and then recreates what would be seen on the scope, rather than how it would be seen - does not deal with these issues.

 

I hope we can find something soon, but some sort of inaccuracy, due to latency, rubber banding, warping etc will always remain online.

 

Why Jester is bugging out rather online than offline, we still haven't fully figured out yet. Both issues (in summary) are quite complex to solve, but we're not giving up, of course.

[Hummingbird]

Exactly why fixing them poses such a huge problem for us. For one we cannot debug in MP. For the other, we don't have the time to fly in MP so much that we could experience them often enough. Mostly we're trying to reproduce them in SP, which doesn't really work.

 

 

 

One of the issues with TWS, since we dont grab the readily available information in DCS, but emulate the signals while pretending to not know the source (like a radar would), is latency. For example, if there is rubber banding, and the contact is moving backwards, that just doesn't work, the radar will not know what to do with a target that is moving backwards, drop the track, and when it sees the target moving forward, create a new track. That's just one of the issues we're busting our heads over. But stuff like this explains why some of the issues are apparent in MP and not in SP. Either way, we are on it and hope to find a solution soon. A fake recreated radar - which simply knows what is available as info in DCS and then recreates what would be seen on the scope, rather than how it would be seen - does not deal with these issues.

 

I hope we can find something soon, but some sort of inaccuracy, due to latency, rubber banding, warping etc will always remain online.

 

Why Jester is bugging out rather online than offline, we still haven't fully figured out yet. Both issues (in summary) are quite complex to solve, but we're not giving up, of course.

 

What I find odd is that atm the AWG-9 is the only radar ingame which seems to suffer from these very frequent losses of tracks, esp. in TWS where it's really REALLY bad (I've basically given up on long range TWS shots at this point) What's more is that it USED to work, hence why I believe it's rarely a server or latency issue.

Edited September 10, 2020 by Hummingbird

[UWBuRn]

...

 

IronMike, thanks for the feedback. :) Can you comment on the issues i'm experiencing in SP as described at page 8 of this thread?

[Spurts]

What I find odd is that atm the AWG-9 is the only radar ingame which seems to suffer from these very frequent losses of tracks, esp. in TWS where it's really REALLY bad (I've basically given up on long range TWS shots at this point) What's more is that it USED to work, hence why I believe it's rarely a server or latency issue.

 

 

 

Probably because it is the only radar that takes the binary detect/nodetect flag and reverse engineers it into a doppler return with the appropriate characteristics and then lets the AWG-9 decide what to do in the way it really would (range gates, doppler gates, etc). In other words, the AWG-9 is the only one not "cheating".

[Erdem]

What I don't get is the radar was much more stable online at launch. Its performance rapidly detoriated over the year. Last year I could hop on Blue Flag on a busy day and guide 2 phoenixes at two different contacts all the way and actually kill them both. I even got the occasional triple kill. When launched at 30-35nm the missile was very scary.

 

Right now I feel lucky if I get a kill every 5-6 phoenixes. Ghost contacts, contacts that keep popping in and out going in the same direction.

 

I know this is not helping but I'm a bit frustrated.

[Bearfoot]

What I don't get is the radar was much more stable online at launch. Its performance rapidly detoriated over the year. Last year I could hop on Blue Flag on a busy day and guide 2 phoenixes at two different contacts all the way and actually kill them both. I even got the occasional triple kill. When launched at 30-35nm the missile was very scary.

 

Right now I feel lucky if I get a kill every 5-6 phoenixes. Ghost contacts, contacts that keep popping in and out going in the same direction.

 

I know this is not helping but I'm a bit frustrated.

 

At launch TWS-A was not implemented. I think many of these issues have to do with TWS-A and Jester's interaction with it.

[Hummingbird]

Probably because it is the only radar that takes the binary detect/nodetect flag and reverse engineers it into a doppler return with the appropriate characteristics and then lets the AWG-9 decide what to do in the way it really would (range gates, doppler gates, etc). In other words, the AWG-9 is the only one not "cheating".

 

If "cheating" is what it takes to simulate realistic performance, then so be it, I'm all up for that. It's much better than a system that doesn't work at all because it's thrown into a digital world where things that could never happen in real life (latency etc..) occur regularly, resulting in unrealistic performance.

[Golo]

If "cheating" is what it takes to simulate realistic performance, then so be it, I'm all up for that. It's much better than a system that doesn't work at all because it's thrown into a digital world where things that could never happen in real life (latency etc..) occur regularly, resulting in unrealistic performance.

 

 

I agree, HB did amazing thing with the radar modeling, but if its not working due to crappy net code/ping/whatever, then we cant use that radar to full potential as we should be able to, hence it needs some kind of workaround to display what we should be seeing and to be able to guide weapons correctly.

[IronMike]

I strongly disagree. It would be a tremendous step back. We want DCS to become more realistic, not less. And maybe our radar is a tad too early for DCS, but in general: latency etc affects everything online, even if it was a cheat, even if less.

 

The thing is, TWS in general was not as reliable, as DCS made it out to be with the simpler implementations most ppl are still used from FC3, etc... It wasn't a mode that was necessarily used for fighting. Okie for example would rather flip the ACM cover and just fire one or two phoenixes into the bandit's direction and break off, etc...

 

So we owe it to realism to also have these quirks impact our virtual radar work. But, as for the other issues, we're looking into it and we will find a solution, maybe later than sooner, but we will. I am confident in that.

 

Making it a fake TWS now, would mean we'd have to revert everything and toss over board years of work. That's not gonna happen.

 

It's also not as "useless" as some make it out to be, imo. Yes, at the moment we have some issues, and thanks to a customer feedback we suspect now it might be connected to datalink mixing up contacts, etc... But taking these instances out of the equation, TWS is quite close to how it is supposed to be in terms of reliability.

[QuiGon]

I strongly disagree. It would be a tremendous step back. We want DCS to become more realistic, not less.

Thanks you! I find the radar track issues in multiplayer pretty annoying too, but not annoying enough that I would want to get rid of the realistic radar simulation of the Tomcat! It makes the radar so much more immersive and fun to use!

[VC]

Definitely want realism and in-depth simulation over "alternative solutions". The radar is very usable with Jester if you understand the limitations. As QuiGon said, this makes it more immersive to use.

 

Things which are actually bugs have been acknowledged and I am happy to wait for that fix work to be done properly, not rushed.

 

I don't know if this is already on your radar IronMike (pun intended), but cranking in TWS-A seems to directly cause track loss, more likely the further you turn. Would you count this under expected/realistic unreliability of the set, or is this an actual bug?

[Hardcard]

Exactly why fixing them poses such a huge problem for us. For one we cannot debug in MP. For the other, we don't have the time to fly in MP so much that we could experience them often enough. Mostly we're trying to reproduce them in SP, which doesn't really work.

 

That's terrible news, then.

 

Sounds like you need a team of MP testers.

If I were you, I'd ask the community for volunteers, I'm sure many would answer the call (I can ask people who regularly fly tomcat in MP).

 

I would gladly perform tests and gather data for you, but I don't own the tomcat module (since I won't pay such a high price for a module that doesn't work as it's supposed to in MP)

[IronMike]

That's terrible news, then.

 

Sounds like you need a team of MP testers.

If I were you, I'd ask the community for volunteers, I'm sure many would answer the call (I can ask people who regularly fly tomcat in MP).

 

I would gladly perform tests and gather data for you, but I don't own the tomcat module (since I won't pay such a high price for a module that doesn't work as it's supposed to in MP)

 

 

 

 

Lack of testers is not an issue, we can and did whip up larger MP test sessions when needed. But if the developer cannot run the debugger, or at the least experience the issue over and over for himself, then it becomes just so much more difficult to find the real culprit. We know whats happening, we've also seen it ourselves, here and there. But as mentioned not on the level that is benefitial for the developer in questions.

 

However, fear not. We will find a way, and we will find a fix, and we won't stop looking for one, until we do.

[Kula66]

Any update on a fix or progress for this. I never really saw this in SP, but now flying more MP and it’s still a big issue ... jester goes click mad and ignores all radar control from the front seat. Really frustrating.