reported F16 Look Down

[cmbaviator]

1 hour ago, skywalker22 said:

That's the current look-up and look-down issue of APG-68(V)5 radar. 

We did some test recently, you can check it out here, and it's still under investigation by ED team.

 

I thought it was a penalty in the look down like irl, i would thought that wouldn't be the case in the look uo as there is no noise in the sky as there is no terrain. So currently its better to be on the same level as the Foe ?

[skywalker22]

More or less, until ED makes a fix for the radar.

[Brody]

Spent some time over the weekend doing a basic mapping the DCS F-16 Viper maximum buggable ranges against a fighter-sized target.

Intention to open discussion about DCS F-16 FCR data with the community and invite others with similar data to share their own.

 

Topics particularly interesting: look-up penalty, low-altitude penalty, static ranges in large areas of radar testing.

 

Please take a look at the full test here (Google Sheets viewer link):

https://docs.google.com/spreadsheets/d/1pZQJnwurX1wkXZ1NUumATFFQFxkqaRzFYnDsC_GCyGg/edit?usp=sharing

Includes Conditions, Methodology, Data, and Graph representation of results (this to focus on)

 

DCS_F16_2D_Color_plot_ver2.ggb

[llOPPOTATOll]

There is a supposed “white paper” they are working on but I haven’t heard any update on that for a while. Is this still being done?

[Moonshine]

As for this topic, some data has been collected in the bug report here: 

 

[Tango3B]

2 hours ago, llOPPOTATOll said:

There is a supposed “white paper” they are working on but I haven’t heard any update on that for a while. Is this still being done?

This. I would also like to know if there is any ETA on this white paper...

[Comrade Doge]

Who knows maybe the Viper team has less people now, and the speed of doing things changed...

[Tango3B]

5 hours ago, Comrade Doge said:

Who knows maybe the Viper team has less people now, and the speed of doing things changed...

Yeah, maybe. And I would be totally fine with that, of course. These guys are surely working on tons of stuff behind the scenes and that takes a lot of human resources, I know that. A little heads up would be very much appreciated, though. But because several months have passed now and because we were actually told that ED was about to promptly publish the said white paper which would clarify most of these concerns and questions rightfully raised by part of the more active F-16 community you should not go radio silent on this matter but you should actually deliver or at least inform the said part of the community. Right? Right. Thanks.

[Shibiswag]

Not sure if this has been posted or not yet or if this is a bug or normal radar operation post 2.8
When at a alt higher than the hostile even over the water where no radar interference from ground clutter, the radar seems to be unable to detect the target until about 27-30NM.

I do have AWACS on to provide data link.

When below the target the radar can detect the target at roughly 40NM.

Please see track files and advise if this is a bug or is this is how the radar now is supposed to operate?

This was not happening pre 2.8 beta

f16abovetarget.trk f16belowtarget.trk

[Moonshine]

there is posts about it, quite many actually and quite a long time already:

 

plus this in the hornet section of the 2.8 OB update

Edited November 12, 2022 by Moonshine

[313 SQN JD]

Is the radar of the F-16C modelled in DCS actually that bad irl or is it underperforming? There seems to be a lot of turmoil about this subject on the DCS forums.

Effectively the ranges at which you can lock a contact are very short. I find it hard to believe this is the case irl and with me many others. This has been discussed on various forum posts but these topics get closed left and right because the titles suggests another issue.

This is not a specific bug rather a question if what we are getting is good enough.

As i am not the only one that feels this way i am pushing this topic again in the hopes we can get some fixes/answers in the near future.

Here are some topics where this is discussed in but are closed because the title suggests a different issue.

https://forum.dcs.world/topic/299832-f16-tms-up-intermittently-working-in-rws-end-of-april-2022-update/page/2/

https://forum.dcs.world/topic/299740-dcs-f-16cm-apg-68v5-look-down-issues-and-also-weird-aim-120c-behavior/

[BIGNEWY]

Hi, 

the data is correct based on the evidence we have, if you have better data that is for public use please DM me. 

thank you

[313 SQN JD]

Thank you for your quick reply.

I don't. But i would like to keep this thread open for discussion within the community because the findings and tests that i've read on other posts correspond with the findings of me and a group of players that i speak for.  Who knows what might come from that.

And if not, at the very least to possibly get tips and tricks to maximize the use of the radar in the sim.
 

[Moonshine]

if the discussion is purely about the strenght of the radar it might even be correct as it is, however the limiting factor to finding and locking stuff at reasonable ranges is the excessive lookdown penalty and how this is applied to the radar in DCS. no point listing it all here, all data and evidence of how absolutely borked it is can be found here:

and here:

instead of saying "correct as is" dear Newy, and therefore cause even more discussions, maybe list the items that are currently under investigation. by far it isnt correct as is and the evidence is there too.

Edited December 14, 2022 by Moonshine

[313 SQN JD]

@Moonshine Thank you! 

We have definitely noticed this. Just now someone of our group sent me a screenshot, flying at 27K, target at 10, cold at 22K at 15nm and the target does not even appear on the FCR. 

Like i said i am not an expert, i also do not have the time to test this all myself but the general feeling of the radar as it is now is that its unreliable and very inconsistent, The other issues reported about lookdown etc might be one of those reasons. 

Im glad for all your work on these topics and thank you for sharing those so we can bundle this into this topic! 

[Cepheus76]

Could it be that in this particular case the target can't be picked up because of similar (ground)speeds, resulting in zero (or too small) dopplershift? Heatblur describes this type of blindspot in their Tomcat manual, page 225.

 

[Edit: Typo- and Grammar-Correction]

Edited December 14, 2022 by Cepheus76

[VarZat]

In my experience the radar has been working exceptionally well in the last months, im really happy with it.

[Moonshine]

before starting all over again, i encourage you to actually take a look and perform the same tests as performed in the comments linked here:

https://forum.dcs.world/topic/305171-f16-look-down/?do=findComment&comment=5012201

and here:

https://forum.dcs.world/topic/305171-f16-look-down/?do=findComment&comment=5016740

 

 

and here is a whole lot of data collection based off of such tests with all sorts of ranges, altitudes etc:

https://forum.dcs.world/topic/305171-f16-look-down/?do=findComment&comment=5048120

 

and please dont take it personal, yet this comment is ruining a lot of work for people who have been actively trying to get ED to fix this over the course of this year now. just saying "it works for me" is not achieving anything without bringing up valid data to not only support this claim (and i doubt you did any of these tests) but also disprove everything else that people have reported. 

Edited December 14, 2022 by Moonshine

[VarZat]

1 hour ago, Moonshine said:

before starting all over again, i encourage you to actually take a look and perform the same tests as performed in the comments linked here:

https://forum.dcs.world/topic/305171-f16-look-down/?do=findComment&comment=5012201

and here:

https://forum.dcs.world/topic/305171-f16-look-down/?do=findComment&comment=5016740

 

 

and here is a whole lot of data collection based off of such tests with all sorts of ranges, altitudes etc:

https://forum.dcs.world/topic/305171-f16-look-down/?do=findComment&comment=5048120

 

and please dont take it personal, yet this comment is ruining a lot of work for people who have been actively trying to get ED to fix this over the course of this year now. just saying "it works for me" is not achieving anything without bringing up valid data to not only support this claim (and i doubt you did any of these tests) but also disprove everything else that people have reported. 

 

Thanks for mentioning. I see now how that is not helping at all and is rather stupid.

[NineLine]

Just as a small update, we are currently refactoring the Hornet radar, and once complete, we'll revisit the Viper radar. Thanks!

[MARLAN_]

8 minutes ago, NineLine said:

Just as a small update, we are currently refactoring the Hornet radar, and once complete, we'll revisit the Viper radar. Thanks!

This is great news, excited to see improvements for both the Hornet and Viper

[WHOGX5]

4 hours ago, 313 SQN JD said:

Is the radar of the F-16C modelled in DCS actually that bad irl or is it underperforming? There seems to be a lot of turmoil about this subject on the DCS forums.

Effectively the ranges at which you can lock a contact are very short. I find it hard to believe this is the case irl and with me many others. This has been discussed on various forum posts but these topics get closed left and right because the titles suggests another issue.

This is not a specific bug rather a question if what we are getting is good enough.

As i am not the only one that feels this way i am pushing this topic again in the hopes we can get some fixes/answers in the near future.

Here are some topics where this is discussed in but are closed because the title suggests a different issue.

https://forum.dcs.world/topic/299832-f16-tms-up-intermittently-working-in-rws-end-of-april-2022-update/page/2/

https://forum.dcs.world/topic/299740-dcs-f-16cm-apg-68v5-look-down-issues-and-also-weird-aim-120c-behavior/

People in this thread mention look down penalties and doppler notches as the main issues but when we're talking specifically about maximum range of the radar in DCS compared to real life, there is another issue that has a much bigger impact.

In real life, the chance of a radar pulse returning to its sender is probabilistic. As an example (using completely made up numbers), every sweep of the radar maybe has a 80% probability to get a return head on at 30nm, while it only has a 10% probability at 70 miles. This is why real life pilots are always saying that STT is much more reliable than TWS, because in real life you get a hell of a lot more radar pulses on target every second in STT than you do in TWS, hence a bigger chance of actually getting a return. This means that even if you only manage to get a single return from a contact at 70 miles every tenth sweep with the radar, maybe 1-2 minutes in-between each return, you could still be able to get a stable STT lock simply because of the sheer volume of pulses you're sending out every second in STT. To use incorrect terminology, you could say that STT has a "longer range" than RWS or TWS, which is why it's so weird that STT has a "shorter range" than RWS and TWS in the DCS F-16C.

The radar model in DCS is deterministic. If you're within these and these parameters, you'll be visible on radar 100% of the time and if not, you'll be visible 0% of the time. This is why the "maximum range" is so short in DCS and also why a lot of people use TWS all the time, because with a deterministic radar model the benefits of STT are much, much smaller.

So, as long as a future refactor of the F-16's radar doesn't implement a probabilistic radar model, you won't see "maximum ranges" anywhere near to real life. Any kind of improvements to the F-16's radar are welcome though.

[skywalker22]

1 hour ago, WHOGX5 said:

People in this thread mention look down penalties and doppler notches as the main issues but when we're talking specifically about maximum range of the radar in DCS compared to real life, there is another issue that has a much bigger impact.

In real life, the chance of a radar pulse returning to its sender is probabilistic. As an example (using completely made up numbers), every sweep of the radar maybe has a 80% probability to get a return head on at 30nm, while it only has a 10% probability at 70 miles. This is why real life pilots are always saying that STT is much more reliable than TWS, because in real life you get a hell of a lot more radar pulses on target every second in STT than you do in TWS, hence a bigger chance of actually getting a return. This means that even if you only manage to get a single return from a contact at 70 miles every tenth sweep with the radar, maybe 1-2 minutes in-between each return, you could still be able to get a stable STT lock simply because of the sheer volume of pulses you're sending out every second in STT. To use incorrect terminology, you could say that STT has a "longer range" than RWS or TWS, which is why it's so weird that STT has a "shorter range" than RWS and TWS in the DCS F-16C.

The radar model in DCS is deterministic. If you're within these and these parameters, you'll be visible on radar 100% of the time and if not, you'll be visible 0% of the time. This is why the "maximum range" is so short in DCS and also why a lot of people use TWS all the time, because with a deterministic radar model the benefits of STT are much, much smaller.

So, as long as a future refactor of the F-16's radar doesn't implement a probabilistic radar model, you won't see "maximum ranges" anywhere near to real life. Any kind of improvements to the F-16's radar are welcome though.

I was thinking (but not expressing here on forum) this exactly same thing a while ago.

Amazing point @WHOGX5, I would say this is the major problem with all ED's radars, and with many other avionics, not just on F-16, but on other planes as well - simply all seems to be in binary way, 0-1. Like there is no analog data, the data between 0 and 1. It is, or is it not - as you said.

[MARLAN_]

Probability of detection is definitely a huge thing we are lacking in DCS.

As another example of how it drastically changed how things should be: 

A dragging (cold) contact will have a lower probability of detection per hit, however because of the much lower closure rate compared to a hot target you actually have a higher probability of detection at any specific range, because you get more hits. This is talked about specially in a 1983 paper by Ringel on the APG-66.

This is in contrast to DCS where a cold contact has a lower detection range.

[TheBigTatanka]

Another thing to mention with cold contacts, which friends who fly 4th gen fighters have explained to me, is that the whole issue of "closure speed is near zero, so it falls into a Doppler gate" isn't a problem in real life, because in real life, 4th gen radars are also looking at their own jet's closure with the ground. If you are flying at 400 knots ground speed, and your radar is painting (and filtering out) ground returns that are closing at 400 knots, then that range hit (aircraft )with zero knots closure looks mighty suspicious to the Doppler side of the radar. Radar systems are highly dependent on knowing your own aircraft's ground speed to make the filtering work -- not sure if ED has considered that.

Also explained to me was that the look down penalty isn't a thing; the altitude problem is when your own jet is down below a certain altitude (let's say 5K as a nice general number), then you get more false returns. The issue isn't that you can't detect down low, or that there's a look down penalty, but rather that you get more false returns because of the extra noise. Razbam radars do this now.

Apparently, the look down penalty that ED has modeled is specific to an early version of the flanker/fulcrum slotback radar. I'm guessing ED has that documentation, and applied that to all 4th gen aircraft.