Jump to content

Searching for proof of underperforming AN/APG-73 radar


GumidekCZ

Recommended Posts

My question;

Is it normal and realistic to loose a tanker completely, when it's 10nm ahead same level at 20,000ft, just because it turns and changes direction? I understand PD notch etc etc, but that tanker has the RCS of an apartment block, surely it shouldn't just disappear?

Link to comment
Share on other sites

4 hours ago, equinox137 said:

Here is a perfect example of why the Hornet radar is worthless. 

Marianas Map, AWACS sees 2 Mig-29 contacts flying over open sea in clear weather at angels 10, 60 miles from me according to AWACS bogey dope.   I am at angels 33, my radar is on TWS 4bar, PRF is on medium, antenna pointed slightly down-angle, scanning 80 miles out, 40 degree sweep, time out is at 16. and the contacts are flanking at my 1 o'clock.  TDC Top caret says 40, bottom caret says -5 with a distance of 56.1 nm according to the datalink....and my radar CAN'T SEE the 2 contacts? Seriously?  There is no reason this radar shouldn't at least see these two contacts.

The technobabble of the last 5 pages notwithstanding, the end product is that this radar doesn't work.  I'm hoping to God that pilots IRL don't have to trust their lives to this absolute junk or that DCS has the APG-73 extremely under-performing in the simulation.  Truth is, I've had better luck not using this radar at all.

radar.jpg

 

 

So the key word you don't understand is "flanking". "Doppler"  Radars work "well" in HPRF when someone is "coming right atcha bruh", when someone is 90 deg to you they dont work or work less well. They also work less well when someone is "running"....

 

This has been a DCS educational moment. brought to you by the the letters HPRF, and MPRF, and by the the numbers 090 (from your actual bearing).

 

 

 

1 hour ago, norman99 said:

My question;

Is it normal and realistic to loose a tanker completely, when it's 10nm ahead same level at 20,000ft, just because it turns and changes direction? I understand PD notch etc etc, but that tanker has the RCS of an apartment block, surely it shouldn't just disappear?

HPRF or MPRF?

 

4 hours ago, equinox137 said:

 

The technobabble of the last 5 pages notwithstanding, the end product is that this radar doesn't work.  I'm hoping to God that pilots IRL don't have to trust their lives to this absolute junk or that DCS has the APG-73 extremely under-performing in the simulation.  Truth is, I've had better luck not using this radar at all.

So. The fact you don't seem to understand "technobabble" or "radar" or ultimately the consequences of the aforementioned "technobabble" is a "YOU" problem. That being said, ED's modeling of A/A radars of any sort is well... "lacking" it should be worse in many cases and better in other cases. If you want a vaguely decent radar model in DCS go look at the M2k, not ED...

 

 


Edited by Harlikwin
  • Like 3

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

3 hours ago, GumidekCZ said:

Question, which may be rised here ... ,can ED contact JANES, may be pay a little sum and get the information about all DCS radar preformances (including may be new and more accurate APG-65 detections for FC3 F-15C).
Or can somedoy do it for ED?

You're under the mistaken impression that JANE's is a worthwhile source of information for any of this.  The F-15C uses the APG-63 and actual radar test data has been provided to ED.  No JANE's needed.

You can even cross-check with the radar equation to see if the stated ranges make sense.


Edited by GGTharos

[sIGPIC][/sIGPIC]

Reminder: SAM = Speed Bump :D

I used to play flight sims like you, but then I took a slammer to the knee - Yoda

Link to comment
Share on other sites

2 hours ago, Harlikwin said:

 

So the key word you don't understand is "flanking". "Doppler"  Radars work "well" in HPRF when someone is "coming right atcha bruh", when someone is 90 deg to you they dont work or work less well. They also work less well when someone is "running"....

 

This has been a DCS educational moment. brought to you by the the letters HPRF, and MPRF, and by the the numbers 090 (from your actual bearing).

 

 

 

HPRF or MPRF?

 

So. The fact you don't seem to understand "technobabble" or "radar" or ultimately the consequences of the aforementioned "technobabble" is a "YOU" problem. That being said, ED's modeling of A/A radars of any sort is well... "lacking" it should be worse in many cases and better in other cases. If you want a vaguely decent radar model in DCS go look at the M2k, not ED...

 

 

 

Wow, Jesus mate, slow down. No need to be completely condescending.

We're all here because we enjoy DCS, learning and growing our knowledge, and maybe even through robust discussions, improving the product that DCS is.

Unfortunately, "holier than thou" attitude posts like yours, really add nothing, and only derail these discussions. I'll be the first person to admit when my assumptions are wrong, as I'm here to learn, not prove how smart I am. However I do expect to be treated with courtesy and respect during this process. It's the least we can do for each other among what is already a small community.

Anyway, are you saying that no matter the size of a target, or it's position relative to the radar, when in the notch, contacts are always lost? There is nothing modern signal processing can do to overcome this, even in the most exterme circumstances? Such and an A380/Antonov 225 at 10 nm?

  • Like 4
Link to comment
Share on other sites

2 hours ago, norman99 said:

Anyway, are you saying that no matter the size of a target, or it's position relative to the radar, when in the notch, contacts are always lost? There is nothing modern signal processing can do to overcome this, even in the most exterme circumstances? Such and an A380/Antonov 225 at 10 nm?

From my admittedly limited understanding, yes. I didn't read that far back, so apologies if I'm saying something you know already. The doppler radars are filtering signals based on their closing speed. With this ground clutter is removed. So if the closing speed of a return is in a certain area (in DCS F18 this currently seems to be +-100knots, no idea if this is accurate, have no hints either way) it get's filtered out as ground clutter, at least in a lock down configuration.

Link to comment
Share on other sites

2 hours ago, Cathnan said:

From my admittedly limited understanding, yes. I didn't read that far back, so apologies if I'm saying something you know already

This depends on many things, it's not that simple. MPRF filters by both range and doppler, so to get rejected as clutter, not only does the contact need to match the relative closer of the ground, but its range as well. Plus, in real life, digital receivers can utilize dynamic notch thresholds, which won't necessarily notch out the main beam and zero doppler freqs completely, they can have the treshold set above the predicted ground return and allow the contact to not get filtered if the return exceeds the set treshold. This depends on SNR, so a tanker notching you at close range can still be picked up, even in a look down. 

 

To simulate this at some degree of fidelity, we'd need to have a proper SNR simulation with RCS that's more in line with real specs, changes with aspect and loadout and the SNR takes into account the RCS, the range and of course the ground returns as well. 

 

4 hours ago, norman99 said:

Wow, Jesus mate, slow down. No need to be completely condescending.

 

He wasn't replying to you with the part that you quoted. One of the easiest ways to deal with all aspect targets in modern radars is MRPF waveform, which filters by both range and doppler and has drastically better performance against all aspect targets at the cost of reduced detection range. PRF ambiguities aren't really simulated in DCS in detail though, but this part is. 

Link to comment
Share on other sites

9 hours ago, norman99 said:

Anyway, are you saying that no matter the size of a target, or it's position relative to the radar, when in the notch, contacts are always lost? There is nothing modern signal processing can do to overcome this, even in the most exterme circumstances? Such and an A380/Antonov 225 at 10 nm?

In DCS they will be always lost.  IRL it depends heavily on the radar and the techniques it uses, but very broadly speaking if a target drops into the notch that's a break-lock.  Modern radars have ways to recover the target if it's not able to stay in the notch long, but there's no magical way of continuing to track it if the filter drops that target.   It's now considered to be part of the radar clutter.

In an F-18 you have some options, such as continuing to track with the pod in A/A, although this should also be subject to clutter, just not in the same way (sort of random change of dropping the track, but we don't need to go into this here).

As for how radar work IRL - an aircraft approaching 90 deg will have a much larger RCS, something not modeled in DCS - no point in complaining about that though, it just is what it is and it's ok.

At the same time, while MPRF is less sensitive to notching (better clutter rejection) it is absolutely not immune to it and because of the way MPRF works (lower duty cycle and lower signal integration), the detection and tracking range is significantly reduced compared to HPRF.  The main difference here is that HPRF is great for head-on targets, very poor against receding (for detection/tracking at range...will work fine at the ranges required to operate tail-shots for sparrows etc) while MPRF gives you an overall consistent performance at all aspects with the caveat that it's got something on the order of 40% of HPRF range.

  • Thanks 1

[sIGPIC][/sIGPIC]

Reminder: SAM = Speed Bump :D

I used to play flight sims like you, but then I took a slammer to the knee - Yoda

Link to comment
Share on other sites

15 hours ago, norman99 said:

Wow, Jesus mate, slow down. No need to be completely condescending.

We're all here because we enjoy DCS, learning and growing our knowledge, and maybe even through robust discussions, improving the product that DCS is.

Unfortunately, "holier than thou" attitude posts like yours, really add nothing, and only derail these discussions. I'll be the first person to admit when my assumptions are wrong, as I'm here to learn, not prove how smart I am. However I do expect to be treated with courtesy and respect during this process. It's the least we can do for each other among what is already a small community.

Anyway, are you saying that no matter the size of a target, or it's position relative to the radar, when in the notch, contacts are always lost? There is nothing modern signal processing can do to overcome this, even in the most exterme circumstances? Such and an A380/Antonov 225 at 10 nm?

wasn't replying to you. Unless HPRF or MPRF was somehow rude. 


Edited by Harlikwin

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

18 hours ago, Harlikwin said:

 

So the key word you don't understand is "flanking". "Doppler"  Radars work "well" in HPRF when someone is "coming right atcha bruh", when someone is 90 deg to you they dont work or work less well. They also work less well when someone is "running"....

 

This has been a DCS educational moment. brought to you by the the letters HPRF, and MPRF, and by the the numbers 090 (from your actual bearing).

 

 

 

HPRF or MPRF?

 

So. The fact you don't seem to understand "technobabble" or "radar" or ultimately the consequences of the aforementioned "technobabble" is a "YOU" problem. That being said, ED's modeling of A/A radars of any sort is well... "lacking" it should be worse in many cases and better in other cases. If you want a vaguely decent radar model in DCS go look at the M2k, not ED...

 

 

 

Does the mirror tell you that you're the greatest everyday, or just on Sundays?

  • Like 2
Link to comment
Share on other sites

oHZZUJU.mp4

20 minutes ago, equinox137 said:

Does the mirror tell you that you're the greatest everyday, or just on Sundays?

oHZZUJU.mp4

 

grMQszU.gif


Edited by Harlikwin

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

Thanks for the technical explanation guys, I really appreciate it.

So in a nutshell, for a large target like a tanker, at close range and same level, IRL there’s enough advanced signal processing available that if the aircraft turns through the notch relatively quickly, it probably won’t get “dropped” so to speak. At least from what the pilot sees. Maybe one or two sweeps of the scan don’t pick it up correctly, but the trackfile would remain in memory, and be quickly updated when the target becomes visible again.

Now for DCS, the above would require an extremely advanced radar simulation, accurate and dynamic RCSs etc. These are most likely beyond what ED is willing to simulate, and therefore the current implementation and and limitations are reasonably accurate and acceptable.

  • Like 1
Link to comment
Share on other sites

43 minutes ago, norman99 said:

Now for DCS, the above would require an extremely advanced radar simulation, accurate and dynamic RCSs etc. These are most likely beyond what ED is willing to simulate, and therefore the current implementation and and limitations are reasonably accurate and acceptable.

Nothing terribly advanced about it.  It's already in-game in one way or another.  The closer you get shorter 'radar memory' is, since the target can move a lot more in terms of angles up close (where eg at 40nm even though he makes a 90 here or there, as far as moving the radar antenna goes it'll be a fraction of a degree)

[sIGPIC][/sIGPIC]

Reminder: SAM = Speed Bump :D

I used to play flight sims like you, but then I took a slammer to the knee - Yoda

Link to comment
Share on other sites

21 hours ago, norman99 said:

Thanks for the technical explanation guys, I really appreciate it.

So in a nutshell, for a large target like a tanker, at close range and same level, IRL there’s enough advanced signal processing available that if the aircraft turns through the notch relatively quickly, it probably won’t get “dropped” so to speak. At least from what the pilot sees. Maybe one or two sweeps of the scan don’t pick it up correctly, but the trackfile would remain in memory, and be quickly updated when the target becomes visible again.

Now for DCS, the above would require an extremely advanced radar simulation, accurate and dynamic RCSs etc. These are most likely beyond what ED is willing to simulate, and therefore the current implementation and and limitations are reasonably accurate and acceptable.

Honestly alot of this is not hard to implement. And in fact for example the Mirage 2000 radar has alot of these features already implemented as does the JF17, the issue is higeldy pigeldy level of "fidelity" within DCS itself. And ED's unwillingness to do the heavy lifts for things like AA radar, or to force some level of standardization of modeling amongst themselves or 3rd parties, which then leads to community frustration esp within the MP community. 

  • Like 9

New hotness: I7 9700k 4.8ghz, 32gb ddr4, 2080ti, :joystick: TM Warthog. TrackIR, HP Reverb (formermly CV1)

Old-N-busted: i7 4720HQ ~3.5GHZ, +32GB DDR3 + Nvidia GTX980m (4GB VRAM) :joystick: TM Warthog. TrackIR, Rift CV1 (yes really).

Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...