A Collection of Bugs

[Bullethead]

Howdy-

Long time, no talkie.  I've become a trucker and have only just arranged things where I can participate again.  Anyway, here's what I've noticed since we last spoke, all of which are still present in the current version:

1.  Amazing Ditching
It's practically impossible to die if you hit the ground at a shallow enough angle, regardless of your speed or how many trees you hit.  Once you hit the ground, you're safe.  You can belly land at 300+ and slam right into trees, and walk away.  However, if you hit those same trees while still airborne, even at 150, you're instantly dead.  The other day I also had a Pz4 in neutral and was coming down a mountain at like 100.  Then I hit a bump and rolled over several times, coming to rest upside down.  Used to be, that killed you but not any more.  And I note that people are exploiting this to land before you can kill them.  Instead of having to slow down for a landing, they just belly in at 300+ and are gone before you can close to fire.

2.  Convergence Problem
All guns are locked in the same elevation plane, which is that of the primary guns.  IOW, if you converge your primaries at 200 yards and your secondaries at 300 yards, all guns cross the LOS at 200 yards, although the 2ndaries ain't converged but still off to the sides.  When the secondaries converge at 300, they're well above the LOS.  At least, that's what using the .target command shows.

3.  Yak-9T Model Problems
A)  The outboard ends of the flaps flicker constantly when the flaps are retracted.  B)  The right main leg swings backwards as it retracts instead of just folding into the middle.  C)  The tailwheel retracts forward instead of backwards like it's supposed to.

4.  Frame Rate Hits
What happened there?  I've got a P4 3.4, 2 gigs of RAM, and a 256MB Radion 9800.  With 2.01, I had great FPS.  However, with 2.02, my FPS sucks so bad I've had to turn all the graphics settings way down.  As a result, ACM is just barely possible, but I can't see anything on the ground while far enough away to be tactically useful.  

5.  Dead Six Misses
This has been a problem all through AH2, betas included.  Something strange happens to bullets fired into the fuselage within a narrow cone of dead 6.  I can get 1-pass kills at 30-45^, but I can also get right behind the target at exactly my convergence range, jockey the throttle to maintain that distance, and fire all my ammo right into his fuselage to no effect.  Only a few stray shots hit the outer wings from time to time.  I even did this the other night with a guy I caught AFK so had a steady target, unlike the turning drones offline.  Most of my squaddies report similar experiences.  It's like bullets from dead 6 go into limbo.

[214thCavalier]

Some of the maps, and specifically some of the CT maps seem to have a huge fps hit.

[jpeg]

Regarding issue #5: I've been noticing that since AH2 came out, thought I just sucked bad. But I've noticed it more and more that I can hit better at angles rather than dead 6 so I try to avoid dead 6 shots these days.

[Bullethead]

Originally posted by jpeg
Regarding issue #5: I've been noticing that since AH2 came out, thought I just sucked bad. But I've noticed it more and more that I can hit better at angles rather than dead 6 so I try to avoid dead 6 shots these days.

I've seen sho-nuff "dead 6" bugs in other games.  It appeared several times in different versions of AW and, IIRC, once in WB.  Thus, I know there CAN be a real bug that negates hits from astern.  

I'm not certain that's the case now in AH2, but it's really starting to look that way to me, both from my own experience and what everybody in my squad's saying.  It's something that took a while to come up, because everybody was a bit ashamed to admit they were missing perfect shots, but once it got mentioned, everybody has described the same problem.  In any case, I think investigating to see if there's a "dead 6 bug" in AH2 would be a good place to start.  And if there isn't, I'd like to know what I need to do different to hit that way, because what I've been doing for the last 10.5 years sure ain't working now.

[TrueKill]

Originally posted by 214thCavalier
Some of the maps, and specifically some of the CT maps seem to have a huge fps hit.

y are u saying that only one map is a FR hog and thats the BoB map everyother map i get the max FR i can get

[Cobra412]

Bullet which birds are the convergence issues with?  I know HTC was going or did change the convergence and harmonization of the guns.  From what I understood is that previously the guns all stayed on the same elevation throughout their flight path.  

I've got an old photo taken from what I believe was an original manual of the 190 A8.  The file was posted by another user on this board awhile back.  It shows the convergence and harmonization for all the weapons onboard.  The MG131s were set to a convergence of 400m, the wing root MG151s were set to 550 and the inner and outer gondolas were set to 550.

With this setup the guns would all be in harmonization together above the LOS at approximately 225m and at approximately 40cm above the LOS.  The convergence spread at 225m and 40cm above the LOS would be as follows.  

Maximum coverage from outer gondola to outer gondola would be approximately 2m left and 2m right of centerline giving you a maximum left to right coverage of almost 12ft.  

The inner gondolas would be just slightly inside the outer gondola maximum coverage at approximately 1.8m respectively off of centerline.

The wing root MG151s would be at a little over a half a meter spread off centerline at that same convergence setting.

The MG131s would be less than .1m left and right of centerline.

If you wanted to check this out even more then you could test this setup and see how it checks out.  I think HTC said they were turning of some kind of random bullet spread generator when they did the adjustments.  I forget the technical term used for it.  It basically made the bullets always fly true to their convergence and harmonization.  I've also got some info for how the P-51 was setup for it's guns.  I've also got one that shows the boresighting for the P-38J also.  They were all found at Zenos Warbirds.  They show the distance setup to the boresight board and all the measurements needed to gain a particular convergence.

[StarOfAfrica2]

I get dead 6 hits alot of times.  Now.  But thats after seeing my hit % climb from 2 or 3 up to 6 to 9%.  For awhile last tour I was over 9% on my hits.  Thats good for me.  Now lets say on a good day I managed a 10% hit rate.  That means 10% of the bullets I fired hit the target.  Taking a dead 6 shot gives the smallest hit bubble.  I'm sure it cuts the number of hits, but I dont know by how much.  But if I only land 10% of my bullets on a GOOD day, how many 6 shots can I land, even on the good days?  And consider also, thats in planes with matched guns (all MG's, no cannon).  Throw in different guns with different ROF and different projectile velocities................... ....

[GunnerCAF]

Originally posted by Cobra412
Bullet which birds are the convergence issues with?  I know HTC was going or did change the convergence and harmonization of the guns.  From what I understood is that previously the guns all stayed on the same elevation throughout their flight path.  

 

I did some testing with the current version and posted to our squad board.  This is what Bullethead is refering to.

Here is a quote from my post:

I just found something interesting.  The gun sight adjusts for trajectory based on the convergence setting.  If you set your convergence of all guns to 250, they will both converge, and elevate the guns to hit at that distance.  What if you set the primary and secondary at different distances?  I did a test.  Both guns will be elevated to the distance set for the primary.  On the target, they hit on a level line, at the distance set by the primary, but both will not converge at that distance.

 

My tests were off line, using the target.  I used the FW 190 D9 or A8 in this test.

Gunner

[Cobra412]

Gunner I hope I'm reading your post correctly.

If you set one set of guns to lets say 250 and the other to 350 they all hit the target level at exactly 250 but the guns which were set to 350 are not converged at that point.  Is this correct?

If this the case then there is one thing you can do to see if the bullets are actually rising and falling as they should and converging as they should.  In order to find where the bullet hits its peak rise above the LOS you'll have to bring the target in towards you as close as possible.  Set your guns to whatever range you typically fly with.  First fire the primaries and document where they hit both above or below the center of the target and left and right of the target. Slowly walk the target out and record the same thing at each distance to the target with that same set of guns.  Then repeat this same process with your secondaries.

You should see your hit marks steadily rise on the target through center mass then at a point they will start to fall again until they hit dead center at the convergence range you set in the beginning.  Through this process you should also start to see your left and right guns start to converge towards center mass.

If you have the distance information that will tell you how far left and right from center line that a gun is mounted and how high or low a gun is mounted compared to the LOS then you can set up a pretty accurate graph as to how the current trajectory of these weapons are acting.  At some point your primaries and secondaries will cross at the same harmonization point.  At this point though your total coverage area could be fairly wide.  Though they may all be hitting at the same hamonization point doesn't necessarily mean you'll be getting a good grouping of rounds.  

In your previous test it's very possible you may have found the harmonization point of the primaries and secondaries at a particular convergence range.  That harmonization point could have very well been your primaries convergence range.  But the only way you'll know for sure that the weapons trajectory is rising and falling as it should opposed to flying straight and level to it's convergence is to setup a test like I posted above.  If it's modeled correctly you will see a similiar pattern as I stated above.
 
Starting below your convergence range flowing up and through your convergence range and then dropping back down to your convergence range.  You should also see it converging in from the left and right also for wing mounted weapons.  For the most part nose mounted weapons should always have the exact same spread from left to right regardless of your convergence setting.  The only thing that will change is where they cross the LOS initially, hit their peak above the LOS and then fall back down to your convergence distance.  The farther out you set your convergence the bullet will reach it's initial rise through the LOS farther away from the airframe.  Basically it'll have a flatter trajectory.  If you want I'll shoot you an email with the boresight information I have for the 190 A8 R1.

[GunnerCAF]

Originally posted by Cobra412
Gunner I hope I'm reading your post correctly.

If you set one set of guns to lets say 250 and the other to 350 they all hit the target level at exactly 250 but the guns which were set to 350 are not converged at that point.  Is this correct?

 

Cobra, that is correct.  I did this over a week ago, and can't remember what I set the guns at.  I set the cannons (secondarys in close, like 150, and the MG (primarys) out like 400.

My main reason for testing was to find why I couldn't hit anything a very close range.  After testing, I decided to bring my convergence settings in much closer then I have had in the past (long story) and off topic

I don't plan to use different convergence settings for cannons, so this feature or bug will not matter much for me.  I don't plan to spend any more time testing.

Though they may all be hitting at the same hamonization point doesn't necessarily mean you'll be getting a good grouping of rounds.

The FW cannons seem to have a lot of dispersion.  The A8's 4 wing mounted cannons spray all over so convergence at long ranges will do you little good.  This was one reason I reduced my convergence settings.

Here is a tip for anyone who want's to do some testing.  Use the DA terrain off line.  Start at one of the high fields.  This way you can test as soon as you clear the field.  Open the target (.target xxx), where xxx is the range you want the target at.  Turn to the target and use auto angle.  Trim to line up the target center.  Fire away.

If you do this, note the pattern at long distances.  This dispersion (grouping) pattern looks like a shotgun hit on a target.  This with the wrong convergence settings could explain some of the reasons why why direct 6 shots are not effective.   But I also think something isn't quite right with the direct 6 shots.

Gunner

[Cobra412]

Gunner I hear ya.  As far as the dispersion goes I have no idea how HTC calculates that.  IIRC when the subject was brought up before HTC turned off dispersion to do some testing for convergence and harmonization.  I'll have to find that old discussion before I quote much more than that.  

I don't remember what patch it was that they changed this but I'm almost positive they did do some changes in the trajectory of the bullets.  

They (HTC) may actually have test data on dispersion rates of particular guns already.  I'm sure there has to be many agencies that did this kind of testing on each and every gun.

[Bullethead]

If this the case then there is one thing you can do to see if the bullets are actually rising and falling as they should and converging as they should.

What I"m seeing is NOT normal trajectories due to different ranges.  Let me try to explain it better.  Mentally draw a + through the target centered on your line of sight.  That is where you want the bullets to converge at the range you have set for those guns.  Elevation of the guns controls whether they're above or below the horizontal line, and inward angle controls whether they hit left or right of the vertical line.  Obviously, if you have a long enough convergence range set, it's possible for bullets rise above the horizontal line and then fall back down to it at the convergence range.  But that's NOT what I'm seeing.

What I'm seeing is all guns hitting the vertical line at the range they're set for.  That's all well and good.  But the problem is, ALL guns hit the horizontal line at the same exact range, which is always the range that the primary guns are set for.  This is wrong, because as a result the 2ndary guns miss the horizontal line at their convergence range, if that's different enough from that of the primary guns.  Depending on the different convergence ranges, the type of guns involved, and their position on the plane, the 2ndaries will be either high or low at their convergence rance.  

That's the problem.  The only way to make the 2ndary guns hit the horizontal line at the range you set for them is to set all guns for the same convergence range.   While this is usually the best approach anyway, in some planes, especially those with both nose and wing guns, sometimes you want the 2ndaries set to a significantly different range than the primaries.

[Cobra412]

I'll take a look at it.  Unless we can turn off dispersion it won't be extremely easy to see the differences.  I'll use the A8 and the factory boresight settings since I have the diagrams on it.

[SlapShot]

Originally posted by Bullethead
2.  Convergence Problem
All guns are locked in the same elevation plane, which is that of the primary guns.  IOW, if you converge your primaries at 200 yards and your secondaries at 300 yards, all guns cross the LOS at 200 yards, although the 2ndaries ain't converged but still off to the sides.  When the secondaries converge at 300, they're well above the LOS.  At least, that's what using the .target command shows.

5.  Dead Six Misses
This has been a problem all through AH2, betas included.  Something strange happens to bullets fired into the fuselage within a narrow cone of dead 6.  I can get 1-pass kills at 30-45^, but I can also get right behind the target at exactly my convergence range, jockey the throttle to maintain that distance, and fire all my ammo right into his fuselage to no effect.  Only a few stray shots hit the outer wings from time to time.  I even did this the other night with a guy I caught AFK so had a steady target, unlike the turning drones offline.  Most of my squaddies report similar experiences.  It's like bullets from dead 6 go into limbo.

These 2 are huge and explain alot of my frustration as of late. I hope that you have really stumbled upon something that is truly wrong and it is something that is correctable.