Level Bombing: Tests and results with 1.08.18956

[Blackdog_kt]

Let me just transfer here what i was discussing on the ATAG forum, for the benefit of all sim pilots who fly the He-111 and the Ju-88.

I'm just describing a simple airstart mission that i used for testing, flying the Ju-88 in a bomb run with the autopilot in R22 mode (the level attitude hold mode). Bonus track file at the end of the post

I did some more tests just now. I made an airstart mission in the FMB with 35% fuel (i think that's what i set it to) and a full bomb load (both bays full of SC50s and four SC250s externally). Apparently, with 75% pitch (2100 RPM) and 80% throttle (about 1.3 Ata with the superchargers at high gear), it almost stabilizes.

The spawn is at 5km but until i can set trims and everything from the default values, i'm down to 4k or so. Then, engaging R22 while at a somewhat high speed (320-340km/h IAS) results in the aircraft stabilizing in a slight descent of about 1m/sec. This can be accounted for easily by periodically looking at the instruments and adjusting bombsight altitude every 20-30 meters of lost altitude. A ballpark value for ground speed at that altitude and IAS is 370-380km/h (i didn't use tables, just engaged automation and played around with it until the sight stabilized in the vertical axis).

The interesting thing is that i also found how to account for drift. In my first run i was a bit out of trim when i engaged the autopilot but instead of correcting it, i thought i might see how drift works and well, it works just fine

What you do is move the bombsight sideways (controls: bombsight left/right) in the direction of the drift, not opposite to it. The idea is that the crosshairs should point where the bombs will impact. If you adjust for drift in this way, at some point you will see that with automation engaged the sideways motion of the sight is negated.

So it's a two step process really: adjust speed value to get a static sight picture in the vertical axis (target doesn't drift above or below crosshairs), then adjust drift value to get a static picture in the horizontal axis(target doesn't drift left or right of the crosshairs). It's very easy to judge, because we have the sight lines, just see if a ground feature near the lines moves parallel to them (correct drift) or tends to cross them (needs adjustment).

After all these adjustments are done, your aiming point (the crosshairs center) will have of course drifted off target. You can then command some slight turns with the autopilot in R22 mode to align the crosshairs with what you want to hit. This is your final correction.

After all is done, you should go back to the pilot's seat and adjust for the dropping altitude every 20-30 meters lost, switch to bombsight view and confirm the sight still tracks correctly.

At some point between 30 and 40 degrees of bombsight angle, the bombs release.

In my tests i used a series release with a spread of 1 meter. This creates a much bigger spread due to the bombs drifting, which is more pronounced the higher the altitude you drop from. A spread of 1 meter worked well for the airfield i was targeting.

I also found out why some people report backwards exploding bombs. The bombs are correctly modeled with regards to their different densities. All bombs fall with the same acceleration in a vacuum, but through the air the denser/heavier bombs have an advantage because they can penetrate the air easier. This is the same reason that gliders carry water ballast: a lighter glider climbs better in a thermal, but a heavier glider can attain higher airspeeds. So, glider pilots load water ballast and they have a jettison valve to dump some of it if they need more climb rate at the expense of forward speed.

What happens in the sim is the same thing. I think it's the forward bomb bay that drops first, then the rear bay and then the externals (SC250s). Well, the SC250s overtake the lighter bombs and finally impact at about the same time as the SC50s from the forward bay, while the SC50s from the rear bay impact last. This creates the effect of a backwards moving string of explosions, but it's not a bug.

It would be a bug if we released a stick of the same type of bombs (eg, only the internal SC50s in the 88, or 8 SC250s in the 111) and we saw the explosions moving backwards. However, in my tests i observed that bombs of the same type (the SC50s in the bomb bays) which drop at similar speeds, correctly explode in the sequence of impact.

All in all, i'm very very pleased with my findings, huge thumbs up to 1C for their work in this.

You can see some of what i'm talking about in the attached track file.

Get the track file from the link below, just unzip into Documents\1C SoftClub\il-2 sturmovik cliffs of dover\records

Then, you can load the track from inside the sim and observe how the bombs drop and how the lighter ones lag behind the heavier ones.

[Jaws2002]

This is great. This fix alone is going to make a good number of players happy. The extremely limited visibility distance for the ground objects may be a problem. I did a few quick bomb r7ns yesterday with the Ju88 and I noticed you can't see any buildings from 2000m. It is going to be hard to find targets for level bombing this way.

[robtek]

The devs should make it possible for map builders to set markers (i.e. black dots) on targets wich should have a visibility range of about 15 to 20km to nullify the disadvantage of too late visible targets.

[Blackdog_kt]

To be honest with you, as long as i can navigate reasonably well i have no problem putting bombs on target. And i fly on mostly medium graphics settings too.

The thing we need is for mission design to move to strategic objectives. It's one thing to look for a tank column (10 dots in a street running among the woods) from 6km of altitude and a totally different thing to simply have to recognize a city and target its industrial area.

Tactical bombers like the Stuka can maneuver and identify the area where the targets are expected before diving (plus they can correct their aim during the dive), fighter bombers as well.

The difficulty comes from having to maneuver a formation of twin-engined bombers all around the place, while looking for a few vehicles. And the reason that targets are handled like that is that until now we had no way to determine if objectives are completed without having some units to be destroyed. It's the old IL2 trick of hiding trucks inside buildings: the game had no way of knowing if the buildings are destroyed, so mission designers made the targets to be units and hid them inside the buildings.

Since the last beta however, some scripting methods have been added that let us determine objective completion in other ways, more suited to formation bombing. This means that instead of the target being "destroy these static vehicles", the target could now be "place that much explosives within a certain radius of this point".

This means we can have buildings as targets. And for such targets of a certain size, this also makes it very easy to bomb from altitude. We just have to navigate there, look through the bombsight and identify the target by other landmarks. The nearby city, the airfield next to it, the bend in the river, etc. This is more or less how the crews back then did it as well. I saw a US air force bomber's manual and it places a lot of emphasis on pre-briefed landmarks around the target to be used as a way to identify it.

I'm not saying i will complain if i can see an isolated factory building from 4km, of course not. But i still feel that even if the building pop-up is improved, we will have to do some planning before going into raids or be familiar with the target area.

Which to be honest with you, is half the fun when flying bombers. Selecting the correct waypoints to come in undetected, having a flight path over the target that makes for a favorable orientation (eg, attacking an airfield across its runway length for maximum damage), egressing along the route of least resistance, etc etc

For example, i was on ATAG a little while ago and i started for the Manston area. Just as i was closing in, a server message came up saying that both of the nearby targets were destroyed. I didn't want to turn back and the other available targets were tactical in nature, so i just decided to attack the Ramsgate runway.

I came in at 4km, aimed the bombs smack in the middle of the runway and went home with no incident whatsoever. I don't know if and what hits i scored because i was too high to look, but if half of them fell within the airfield and any RAF pilots were CAPing around the area at low level, it would have been a real laugh to watch

[CaptainDoggles]

I also found out why some people report backwards exploding bombs. The bombs are correctly modeled with regards to their different densities. All bombs fall with the same acceleration in a vacuum, but through the air the denser/heavier bombs have an advantage because they can penetrate the air easier. This is the same reason that gliders carry water ballast: a lighter glider climbs better in a thermal, but a heavier glider can attain higher airspeeds. So, glider pilots load water ballast and they have a jettison valve to dump some of it if they need more climb rate at the expense of forward speed.

What happens in the sim is the same thing. I think it's the forward bomb bay that drops first, then the rear bay and then the externals (SC250s). Well, the SC250s overtake the lighter bombs and finally impact at about the same time as the SC50s from the forward bay, while the SC50s from the rear bay impact last. This creates the effect of a backwards moving string of explosions, but it's not a bug.

That is friggin' awesome

[Blackdog_kt]

Oh yes it is. You can't imagine the look on my face when i was watching the track and realised what was happening.

"Hmmm, a bunch of almost concurrent explosions. Ooops, here come some late explosions further back. Maybe it's the bug i've been hearing about. Hmmm, the ones that popped first are the big explosions, they must be the 250s. OH MY GOD, THEY DROPPED LAST AND EXPLODED FIRST, THERE IS ONLY ONE EXPLANATION FOR THIS" /jaw meets floor

Just watch the attached track and enable manual view control. If you go to zoom in view and look down beside the wing (i also moved the camera sideways a bit to get a better view), you can see for yourself.

I think i'll do a second track tomorrow with a static camera on the airfield, so that i can observe more closely

[chantaje]

GREAT!!! , i just tried it and i hit a bus from 4 k. with a 111 thank you!!! now i can know where my bomb will hit

[Redroach]

Originally Posted by robtek

The devs should make it possible for map builders to set markers (i.e. black dots) on targets wich should have a visibility range of about 15 to 20km to nullify the disadvantage of too late visible targets.

I did lengthy tests and came up with medium-to-high settings for my, at best, medium rig, to ensure that everything runs smoothly while having it look as good as possible.
And I had never problems with identifying targets from altitude, except when the targets themselves are very small - for ex. I never figured out where exactly those ammo dumps/storage areas were on that one ATAG objective.
Well, that's what I thing Blackdog is up to: Handing the strategic, bigger objectives to the level bombers while leaving the small ones to tactical bombardment.

[chantaje]

redroach do you know what options affect the ground target visibility? i know that land detail render far away ground textures in more detail and it help navigation but have no clue about how th other options affects ground target, S!

[Kaiser]

Originally Posted by Redroach

And I had never problems with identifying targets from altitude, except when the targets themselves are very small - for ex. I never figured out where exactly those ammo dumps/storage areas were on that one ATAG objective.
.

Here about it I wrote to developers about need of implementation for game of a photo reconnaissance.

http://www.sukhoi.ru/forum/showthrea...=1#post1845441