[Catalyst 7.5]

That's what I am using, just the basic 7.5 driver + ATIT tools. I don't think CCC is particularly a bad program; it's just that ATIT does everything that CCC does using less system resource. (~7mb vs ~3.5mb, respectively) And it's also quicker to use.

[Catalyst 7.5]

Actually I hated the Catalyst 7.4 because no matter what I did AA / AF and Vsync did not work when I set them in CCC. Last night I figured out a way to do this in registry via regedit.

 

Best of all, I figured out when I enable AF it sets it automatically to High Quality with Trilinear.. and this stupid Triliear filtering is what's giving me stutters. I managed to set it to Quality via regedit and LockOn is seriously much much smoother.

 

Wait, I thought trilinear filtering is only used if you click on the 'High Quality AF' option under anistropic filtering settings. When you leave the box unchecked, doesn't it use the regular bilinear filtering?

[Vista: 32 bit or 64 bit?]

There are still only a handful of applications and games that are optimized for 64bit processing, and I am pretty sure Lock On isn't one of them. In fact, in my limited experience, the CryEngine was the only game engine I felt the benefit of 64-bit processing. (Even then, FarCry was more GPU dependent than CPU)

 

According to the following link, Half-life 2 and Colin McRae were another games that were designed for 64bit processing.

http://www.start64.com/index.php?option=com_content&task=view&id=190&Itemid=116

 

So, if I were you, I would stay with 32bit until an occasion that calls for 64bit arises.

[Planning my new badass machine.]

AMD Barcelona (x4) is due out July, and Intel Penryn (45nm) is also due Q3/Q4. I won't be gettin them since they will most likely cost me an arm and a leg. But I do smell a price war around that period on current chips like C2D, C2Q, and X2. So I am gonna wait till late July/ early August. Until then my ghetto P4 3.2 stays.

 

AMD X2 5600/6000 is already inexpensive, going around 150~210USD, so I wouldn't be surprised if they drop to around 120USD within the next month or two, when Barca makes debut. I am thinking of getting the AMD X5600/AM2 mobo combo for under $250 when that does happen.

 

According to tomshardware.com's reviews, contrary to what many are led to believe, AMD x2 chips are perfectly comparable in performance to C2D, provided that you are not using your machine for heavy encoding purposes. Also I am not a big fan of C2Q, you lose out on many advantages of C2D like overclockability, reduced heat generation, and low power consumption.

[American IR Missiles]

Just out of curiosity, what task did you set the MIG-25 up for? Because if you have it on "CAP" or "Fighter Sweep", if it detects you it's still going to fly straight at you, and right into your AMRAAM too.

 

In any case, I wouldn't look too much into it. BVR combat in Lock On is pretty arcade-ish right now, and hopefully the entire thing just gets redone from the ground up in the future. In this respect, Falcon 4.0 is far more realistic, IMO.

 

I didn't set it up to do anything. It just flew around a bit and headed for that airfield. When my missile was about 8~10nm away, it seemed to have noticed it, as it tried to put the missile on its 3, as you can see from the screeshot.

 

Anyway, having played Falcon 4 for 7 years, I wouldn't say F4's BVR is realistic either. You can detect mid-range missile launches from miles and miles away even before the radar onboard the missile goes active. Making evasive maneuvers even easier to perform. In fact, you can fly the entire campaign without ever having to worry about getting shot down A-A. (SAMs on the other hand can be quite deadly.)

 

Watch these 2 movies, and see how the detection of incoming missile can be done at the instant it is fired, thus making identification and preparation for evasive maneuver that much easier. Also, I've read that HoJ is less realiable than a solid radar lock, but for some odd reason in Falcon 4, HoJ is deadlier than radar guidance. That's why the dude piloting in the second movie turns off the ECM near the end.

 

http://rapidshare.com/files/33527959/Missile_test_2.avi.html (only 4mb)

http://www.youtube.com/watch?v=8N-SEWKqWDA (direct feed)

[American IR Missiles]

D-Scythe, I shoot 20nm out and I keep getting the kill. I can do this all day long with pretty much the same result all the time. I will try it again tonight with a different heading, location, and distance between the respective aircrafts (I am gonna use a non-MiG25 also, to see if it's the bandit AI that's making the difference). Anyway here are the screenshots from my latest attempt.

 

Here is the how I set it up, MiG25, 20nm out, no waypoints like you said.

 

 

I fired as soon as I got the lock.

 

 

I dropped the lock right away as you can see.

 

 

There is our AIM120, still climbing, high above the bandit, 15nm out. Maybe it's lost, maybe not, we will see.

 

 

Still about 9 nm out, it changed its heading by 2 degrees, it went from traveling straight to a slight descent. It may have detected the MiG already.

 

 

 

3 seconds later, 6nm out, it's making a mad descent, dashing toward the crash course. As expected, it definitely sees the bandit now.

 

 

Poor bugger, my missiles got a lead on him. He's been traversing the plane towards the airport, putting my missile on his 3 O'clock. But it's seems to be of little use.

 

 

Here's the pic right before the impact. Like I said I think the missile's own radar gimbal is pretty big, or at least bigger than what I thought it to be, given the point at which it started maneuvering even without my own guidance.

[American IR Missiles]

Hopefully, these pictures can show you why I thought there is a discrepancy between the AIM120 in LO and AIM120 in F4. (like I said, I don't know which is a more accurate representation.)

 

You saw from my earlier post that that under near identical conditions (MiG25 20miles out), the AIM120 in LO successfully made a kill. As you see in the screenshots, AIM120 in F4 failed to do so. Here are the conclusions or possible explanations that I deduce;

 

1) AIM120's onboard radar gimbal is larger in LO than in F4,

2) LO's AIM120 is smarter at optimizing its flight path. (judging from how long range AIM120 shot in LO resulted in the missile's rapid climbing to maximize its travel distance) F4's AIM120 on the other hand stayed at a relatively flat trajectory despite having launched it at 20nm out.

3) F4's AI engages defensive earlier than LO's, ruling out the possibility of accidentally crossing AMRAAM's search gimbal.

 

I cannot pinpoint which is the correct answer for discrepancy, but I wouldn't be surprised if it's some combination of above 3 possibilities.

 

Anyway here are the screenshots of the last test I performed on F4. You can probably get a sense why I felt that the same weapon in two sims behave a bit differently.

 

p.s. I deleted more screenshots of the previous tests b/c of the upload limit issue, if any of you wants to see them again, I can send it to you via PM.

[American IR Missiles]

Anyway thanks guys for your attention. I didn't bring this up to cause a stir, or even dispute a point. The fact is I am very much clueless as to how the real AMRAAM behaves. All my perception and knowledge is solely based on how it was portrayed in different sims. Even then, like I said on my first post, my experience with LO is pretty limited, since I just bought this software 5 weeks ago.

 

I just brought it up because AIM120 in LO seems to behave a bit 'smarter' than the AIM120s in different versions F4. In F4, it was easy to tell whether missile has a target, or whether it's in its dormant search mode. Once AMRAAM is launched, F-16's FCC calculates the time to impact and starts a countdown. The HUD displays a letter A and a number of seconds next to it, right below the DLZ scale. Once the number reaches 0, the letter A changes to T and a new number is displayed next to it, indicating that the missile is now tracking a target and is in fact fully autonomous, at which point you can freely break the lock. But if you disengage before this happens, the PK is close to nil.

 

Of course, on the HUD of an F-15 I don't see any of these info, so the only way for me to check the point at which the missile goes autonomous is to devise up some tests where I purposefuly break radar locks prematurely, at different ranges and aspects, to see whether that missile gets lost, or tracks its targets. I was pointing out the scenarios when the missile successfully found its target without my aid, and conversely, when it was unable to find a target without my aircraft's radar lock.

 

"AMRAAM has a radar in the nose of the missile which can lock onto a target. Since this radar is much smaller than the radar in the F-16, it cannot track a target as far away as the F-16's radar. The F-16 therefore must find the target and guide the AIM120 to a point close enough for the AIM120's smaller radar to acquire it. When it reaches this point, the missile becomes autonomous and guides without further help from the F-16." -F4:AF manual pg. 114/716

[American IR Missiles]

In your tests, I think it's just a matter of the target flying a straight and predictable profile, RIGHT into the missile.

 

Modify your tests. Place the target as high off boresight as possible and shoot the AMRAAM. Make sure you break missile lock before the missile completes its turn into the bandit. Don't give the AI a waypoint so that it starts looking for an airbase and doesn't fly in a straight line.

 

I did what you said. I think the pictures are pretty self-explanatory. The seeker finds the target even when the bandit is not travelling toward the missile, without any outside guidance. It certainly doesn't look like the case of bandit accidentally coming across the AIM120's radar. Another intersting thing I saw is that the missile didn't climb. It didn't even speed up for long, the burner went out within a couple of seconds into its flight, if not less. It definitely seems to be tracking at a distance greater than 7~8nm from the target. And no matter how quickly I break the lock (which was within a half second in all of these tests), the missile either 1) turns directly into the bandit, or 2) starts its climb. I tried shooting it way off-angle, at the edge of the gimbal, but to no avail, it still behaves in the one of the two ways I described.

 

I wanted to test at a greater range using this off-bore approach, but that meant that I would have to shoot outside DLZ and the missile will never have any energy to reach. (I tried it, and as expected it just doesn't have the energy to maneuver.)

 

p.s. I exceeded my upload limit for jpg, so I had to delete some screenshot from the previous posts.

[American IR Missiles]

Why would it go blind?

The target's flying right into its scanbox.

 

No, I am saying, the missiles' scanbox seems to be active even from a great distance from the bandit. I always thought that AIM120s onboard radar goes active after it gets reasonably close to the target. (7~8nm like D-Scythe said) And to guide it close till its radar activates, you need to keep your lock on the bandit. At least that was how it was portrayed in F4.

[American IR Missiles]

OK, I got a good news and a bad news.

 

Good news is that I just got my first miss on AIM120C, using the same testing method. I dropped the lock and the missile got lost and flew into emptiness. I guess AIM120C in LO isn't totally autonomous after all.

 

Bad news is that I was 25 nm out when I fired. It seems that the seeker on the missile goes active pretty early its flight and it's definitely not when it's just 7~8nm away from the bandit. It seems to go active at much greater distance than that (more like 20nm, judging the distance to the bandit from the MFD on the second screenshot).

 

Anyway, here are the screenies.

[American IR Missiles]

Did you by any chance watch in say, the outside view to see when the missile begins to actually maneuver to intercept the MiG? After the rocket burns out, you can watch the g on the missile in external view.

 

 

Actually, the missiles just go into a climb for 7~10 seconds just after launch, the motor burns out soon after. Once it gets around 7km within the target, it starts maneuvering.

 

I just re-did the test, only this time I fired at 22nm out, and of course, I dropped the lock right away. The missile still hit. I want to see my missile going blind and miss the target. :mad:

[American IR Missiles]

Are we playing the same game? The AIM-120 in Lock On activates its seeker when it gets within 7-8 nm of the target. If you break lock before then, chances are your AIM-120 would just loft into the sky.

 

 

See, you are assuming that the seeker activates after some distance. I found that to be not true; in this game, it guides to the target even without my own radar on the bandit.

 

Here is an easy test you can do, when you do a quick start on F-15, you can see a Mig25 on your radar, 2 O'clock, about 15nm out. He's coming in head-on aspect, so the distance will close rather quickly, so lock him up and fire the 120C ASAP. And as soon as the missile leaves, unlock him and disengage altogether. You will find the missile still hits regardless.

 

This is an easy test to replicate, I just did it twice, both times resulting in successful hits. Here are the screenshots.

[American IR Missiles]

I recently got LOFC and only flew for about 30 hours, mainly with Su-27. So my knowledge/perception is probably not 100% reliable. But from what I can see, 27ET seems very effective, so much so that it left me with a suspicion that it's probably not a good representation of the real thing. Mid-range IR missile with high PK? Sounds a little too good to be true. On the other hand, 27ER is pretty much a garbage compared to AIM120.

 

The real AIM120 is not fully autonomous, you still have to guide it with your AN/APG, until the missile's own radar kicks in. F4 did a good job portraying it. In F4, you can see the amraam going blind if you stop bugging your target, lose the radar lock or leave the TWS gimbal before the missile goes 'pitbull'.

 

ED, on the other hand, seems to have made AIM120 fully autonomous in LO. It's either that or the missile just goes pitbull really quickly, like right off the hardpoint, effectively turning AIM120 into a true fire-and-forget weapon. So I don't get why some of you say that AIM120 is under-modelled. If anything, it's vastly over-modelled/simplified in delivery.

 

Now, onto the famous sidewider... AIM-9M in F4 has even shorter range than that in LO, with almost 0 PK outside 2nm (3.7km), provided you are shooting the target in the correct aspect. So I can't tell which representation is "more" correct, though I get a sense that the range on the real thing probably sucks. Also, AIM-9M can't be fired using helmet-mounted target queueing system, that was only implemented with AIM-9X.

[F-117A Nighthawks Retired!]

Avionics for B2 is a beast to conquer. That and the usual 40+ hour sorties. Fighter pilots have to go through 2 extra years of training to get into a B2, according to Modern Marvels by History Channel.