[WIP] SJ Radar Mod version 2

[Hitman]

In SH1 -and I think that was a very well documented game- the PPI was orange and the A-scope green. :hmm:

[Nisgeis]

Originally Posted by Hitman

In SH1 -and I think that was a very well documented game- the PPI was orange and the A-scope green. :hmm:

Yep, that seems to be the way the evidence is pointing.

Thanks M.Sarsfield for the info, can't wait to see the pics.

[Nisgeis]

The range dials, though still in meters only are ready for testing. Would anyone like to be involved in the testing? I just got four hits at 6,500 yards whilst just testing this.

This will just be a test of the digital range readouts in all subs. It shouldn't matter whether you are running version version 1.4 or version 1.5, or any of the big mods. But we'll see about that when we test it.

The pixel shader work isn't ready yet, as it doesn't work with 1.5 the same way as with 1.4, so the PPI scope isn't ready yet, but in the mean time if you are interested in testing the digital readouts with range rings and bearing lines, you can use the previous version here:

http://www.subsim.com/radioroom/show...=radar+bearing

If you are interested in testing, please post here. Thanks.

[akdavis]

Glad to help test Nisgeis. Also, I was pondering last night whether there might be a way to cause at least large-caliber shell splashes (i.e. "Particles.P01_&Big_Shell_water_explosion") to return a radar contact, if not all calibers.

[Nisgeis]

Originally Posted by akdavis

Glad to help test Nisgeis. Also, I was pondering last night whether there might be a way to cause at least large-caliber shell splashes (i.e. "Particles.P01_&Big_Shell_water_explosion") to return a radar contact, if not all calibers.

I haven't looked at what causes an object to give a radar echo, it may be a property of the object. It was removed from sampans in one of the patches, so may be possible. Either it's hard coded into the radar routine, or there's a sim controller in the .dat file for the ship - it's on my to do list, to try to see if I can get ship contacts longer.

[akdavis]

Well, I have tested with stock 1.5 so far. Took me a few tries to understand the dial readings above 10,000m, but quite effective once I figured it out! Deck gun hits on a large target at 10,000m and over the horizon at 13,000m using radar for range (visual bearing on superstructure). Hits on a moving target at 7,000m (bearing obtained by visual). Interestingly enough, hits are not consistent, which seems to indicate the guns are not entirely gyro-stabilized as they might seem (but the variation is not apparent until you fire at these extreme ranges). Seems that there actually is a tiny bit of pitching up and down with the guns. Gun bearing seems to stay constant. Obviously shooting truly BVR (with true bearing unknown) will be much more challenging.

Noticed two things so far:

1. Correct bearing seems to be on the counter-clockwise side of the contact (i.e. rotate clockwise through the target until you lose the return, then rotate back until you first pick up the return, that is the accurate bearing (AI seems to return bearing on the edge of the return first reached rotating clockwise).

2. In the Gato, the range dial on the A-scope is reversed from the range dial on the PPI display (which has the correct range).



Here, PPI dial shows correct range of 9659m, but A-scope has this number reversed (9569m).

More obvious here:

A-scope reads 5109, PPI reads 9015 (once again, PPI is correct).

Porpoise in training school missions does not have this, but I have not tested any other subs yet.

[akdavis]

Okay, tested the other classes. A-scope range dial has reversed value on:

Gato
Gar
Tambor
Balao

Okay on:

S-18
S-42
Porpoise
Salmon
Sargo

[Nisgeis]

Originally Posted by akdavis

Well, I have tested with stock 1.5 so far. Took me a few tries to understand the dial readings above 10,000m, but quite effective once I figured it out!

What method are you using? Guestimating with the range rings how many thousand meters away they are and then fine tuning with the readout, or using the A-Scope to get the thousands? Inetersting, as I only thought it would be useful below 10 km.

Originally Posted by akdavis

Deck gun hits on a large target at 10,000m and over the horizon at 13,000m using radar for range (visual bearing on superstructure). Hits on a moving target at 7,000m (bearing obtained by visual).

Blimey, that's extreme shooting - are you using one of the playable BB mods?

Originally Posted by akdavis

Noticed two things so far:

1. Correct bearing seems to be on the counter-clockwise side of the contact (i.e. rotate clockwise through the target until you lose the return, then rotate back until you first pick up the return, that is the accurate bearing (AI seems to return bearing on the edge of the return first reached rotating clockwise).

Very observant! This is an issue I have been struggling with and I don't think there is a way round it. It makes no practical sense, but may be fine operationally. The hydrophone also has an arc of detection, but the hydrophone has the arc either side of the contact, with the correct bearing being the middle of the two bearings where the contact is lost. The radar has an arc that abruptly ends right on the leading edge. It's as if the radar beam is lobed out to the side, but that doesn't make any sense, as if that were the case, you wouldn't pick up the contact before the sweep got to it and lose it right on it. I can only ssume this is a bit of stock game weirdness.

Out of interest, how are you measuring bearing? The stock camera doesn't allow you to look at the radar antenna bearing indicator properly and it's also not accurate in itself - the bearings are only marked in 5 degree increments and even these aren't accurate - sometimes off by 2 degrees. I am working on a high res accurate bearing indicator, but with this I have noticed that targets directly infront and directly behind are accurate, but ones to the sides are off consistently by half a degree.

The bearing indicator problem may be a game measuring error, as the antenna apparently goes between 0 and 360 degrees, which is 361 degrees in a cirlce, or due to the measurement errors, it's possible it's a 3d model placement problem in the interior, the indicator being position a touch too high/low, and therefore being accurate at 0 and 180, but innacurate at the sides.

Originally Posted by akdavis

2. In the Gato, the range dial on the A-scope is reversed from the range dial on the PPI display (which has the correct range).

That's a bonehead mistake by me. I tested it all before packing it up, but must have packed one of the wrong files in. Thanks for spotting and I PMed you a new link to the (hopefully) corrected file.

Would anyone else like to test this?

[akdavis]

Originally Posted by Nisgeis

What method are you using? Guestimating with the range rings how many thousand meters away they are and then fine tuning with the readout, or using the A-Scope to get the thousands? Inetersting, as I only thought it would be useful below 10 km.

I haven't played a version with range rings yet. Above 10k, I use the A-scope to determine thousands (i.e. 13k etc.) and the range dial for hundreds (i.e. 13150m). At 10k or greater, the left number on the dial remains at 9 while the other three continue to rotate through the numbers, i.e. dial reading as such:

500m = 0500
1000m = 1000
5000m = 5000
9000m = 9000
10000m = 9000
10500m = 9500
11000m = 9000
11750m = 9750

Originally Posted by Nisgeis

Blimey, that's extreme shooting - are you using one of the playable BB mods?

No, but with the stock 3" gun, you can get 5 shells in the air at once at 13,000m. Had about a 20% hit rate on a very large target (CV).

However this was with known and observable bearings. I tried a practice shoot against large tankers at various unknown bearings and ranges from 4-13k at night with rain and heavy fog. This was much more difficult because even if you have the correct bearing, you won't know until you achieve a hit, and the slight pitching/rolling in the gun means you could fire many times on the correct bearing with the correct range and still miss. Add a slow reload time like RFB, and it is a real bitch.

Originally Posted by Nisgeis

Very observant! This is an issue I have been struggling with and I don't think there is a way round it. It makes no practical sense, but may be fine operationally. The hydrophone also has an arc of detection, but the hydrophone has the arc either side of the contact, with the correct bearing being the middle of the two bearings where the contact is lost. The radar has an arc that abruptly ends right on the leading edge. It's as if the radar beam is lobed out to the side, but that doesn't make any sense, as if that were the case, you wouldn't pick up the contact before the sweep got to it and lose it right on it. I can only ssume this is a bit of stock game weirdness.

Out of interest, how are you measuring bearing? The stock camera doesn't allow you to look at the radar antenna bearing indicator properly and it's also not accurate in itself - the bearings are only marked in 5 degree increments and even these aren't accurate - sometimes off by 2 degrees. I am working on a high res accurate bearing indicator, but with this I have noticed that targets directly infront and directly behind are accurate, but ones to the sides are off consistently by half a degree.

The bearing indicator problem may be a game measuring error, as the antenna apparently goes between 0 and 360 degrees, which is 361 degrees in a cirlce, or due to the measurement errors, it's possible it's a 3d model placement problem in the interior, the indicator being position a touch too high/low, and therefore being accurate at 0 and 180, but innacurate at the sides.

With stock, it is very difficult. You are too far away from th display, the contacts are too large and the degree markings to sparse. I found getting a rough bearing (within 1-2 degrees) using RFB much easier. The camera allows you to get close, there are more bearing markings and the contacts are much smaller. Although you still have the contacts giving range returns over a large arc, if you place the update line over the RFB size "blips," you will have a close to correct bearing.

RFB+radar mod:

Here you can see there are actually two long range contacts fairly close together. The sweep line is merged with the closer, more southerly contact, giving both range and a fairly accurate bearing. However, it should be noted that return "arc" of the closer contact is wide enough to mask the second contact, so you can get an accurate bearing off the second contact by moving the sweep line on top of it, but the range returned is still for the closer contact until you move the sweep line significantly closer to 90 degrees, at which point you will get a range return on the second contact, but the bearing of the sweep line will be off the contact.

I have also now tested it with PE3 alone, RFB alone and PE3 + RFB. PE3 works okay, but the cameras make it a bit more difficult to use the range dials. RFB or PE3 + RFB works like a charm, as noted above.

I'll download the new file and test shortly.

[Nisgeis]

akdavis, you seem to be an advanced radar user. . I need to make up a little mini manual of how to use the scopes, like how to differentiate targets on the same bearing with different ranges and all the stuff like that in the proper radar manual.

Stock game's bearing resolution on the SJ radar sets is 9 degrees, which is ahistorical. Actual bearing resolution was 5 degrees and makes it easier to distinguish targets on the A-Scope. The PPI stock scope is too good, with the contacts being dots rather than arcs.

Using the antenna bearing indicator, which is loacted above the A-Scope and a high res texture marked accurately, you can get bearing to within half a degree of the leading edge. As part of my testing, I set up a mission in heavy fog, rain and at night with no moon - it's what Radar Training Part 3 used to be like before I moved the submarine closer and have made 3 hits out of 4 with a three degree from 6,000 yards. If I take the time to plot a track and get the correct course and speed, I'm fairly confident I'll get some hits.

I've also used the antenna bearing indicator above the A-Scope to plot tracks from radar depth and that works very well too.

[akdavis]

Well, I must admit I didn't know about the antenna bearing indicator above the A-scope, but unfortunately it looks like RFB makes it pretty much unusable:



But without RFB, the indicator appears too low res and inaccurate to be much more useful than the PPI.

[Q3ark]

If memory serves, this was a problem with the re-texturing work done by Oakgroove. Trigger Maru uses the same work, and Ducimus fixed it with the latest version of TMO

[Hitman]

A bit of feedback here:

In the version I have tested, the numbers for range didn't show up when using the 1.4 version of the game. In SH5 they worked well, though.

I therefore did some testing in 1.5 (No other mods installed except your radar mod) and first impression is that the contact dots are a bit too large. :hmm: It becomes difficult differentiating contacts on close bearings.

Also -this happens already in the stock game- many times the A-Scope doesn't show a clear return while the PPI shows a clear contact, making it difficult to use the scale for determining distance when over 10.000 metres.

I will play a mission with a single ship and try to plot it before actually trying to plot a convoy -which is right now much more complicated- and will report.

FYI I'm using a plot in paper, not the game map, and do it with Kim Ronhoff's ruler for radar plotting. Will try to put a photo of the resulting plot when I have it

[akdavis]

Nisgeis, I can't remember if you are on the RFB team or not. Should I report the Antenna Bearing Indicator problem in the RFB thread?

[Nisgeis]

I'm not a part of the RFB team and yes probably best to report it there. Thanks!