MMPOL Fast Jets/Modern Air Combat

[Pei]

Iwas wondering the other night why nobody has made a massively multiplayer online fast jet sim. I'more of a prop fan myself but I'd certainly be interested. Anybody got any ideas why nobody has tried this? There seems to be plenty of interest in fast jet sims

[LtHans]

I would like to play something like that too.

The only reason I can see is that modern jets are somewhat less "skill" and more "button pusher" and so they're not as fun in mulitplay.  You just lock on and launch...then thats it.

Still, I would buy one.  I am torn on what era of the jet age that I would prefer though?

My favorite would probably be the transitional period in the cold war where missles and radar were just starting to be used, but not so perfect that they replace dogfighting.  You would still need to manuever around behind your target to launch a heat seeker, and be skilled enough to understand cryptic radar displays.

F8 Crusader
F106 Delta Dagger
F104 Starfighter
A4 Skyhawk
F100 Super Saber
BAC Lightning
Mirage 1
Mig 19
Mig 21
Su 22

[J_A_B]

The former kesmai studios was working on a MMP Vietnam jet sim awhile back, but this got canned when EA bought them out.


The first of many Kesmai projects to be killed by EA.

J_A_B

[Dinger]

well...
Fast Jets have this problem
you see, online, the faster you go the more the predictive motion engine has to predict, and the less accurate it can be; so the more lag you add.  But the more lag you add, the more odd the results.  Now, if you have missiles flying at mach 3, you're gonna get some weird results (i.e. apparent long misses scoring home).
Sure, these things can be overcome, but the interest threshhold can't:
WWII was the last really big conflict in which the opponents had closely matched planes.
Ever watch the *cough* "History" channel?
we have 10000 years of history recorded in some form or other.  Economic history, Social History, Political History, Intellectual History, Institutional History.
But the History Channel programs something like half of its time on the military history of six years in Europe.
THat's how much WW2 sells.

P.S., dude you forgot the bombers.  Do the hustle!

[streakeagle]

You neglected to mention that the other half of the time is pretty much split between Korea and Vietnam.

Anyone that thinks killing a MiG-17, MiG-19, or MiG-21 with a gunless F-4 is easier than WWII air combat needs to do some reading. Far from being "push-button", victory for either side required even more careful exploitation of your aircraft's strengths and the enemy's weaknesses than the furballs you usually see in AH. Of course the rules of engagement (such as visual identification before firing) held back American success as much as any failure in missile technology, pilot training, and aircraft design.

Even Korea was not the one-sided "10 to 1" victory that is so often quoted. At various times during the Korean War, MiG-15 performance, pilot skill, and tactics mixed with political constraints permitted them to bounce F-86s at will resulting in very unfavorable kill ratios that the Air Force always neglects to mention.

The arrival of reliable AIM-9s toward the end of Vietnam War signaled the onset of "push-button" air combat that had been predicted prematurely in the 1950s. The IAF over Lebanon and Syria and the USAF over Iraq finally proved that missiles could be used beyond visual range with great success and little or no risk to the firing aircraft.

IMHO, there is plenty of justification for jet combat in the MMOL market in terms of challenging air combat. But are there actually enough people that would pay for it? Third Wire's Project 1 and Targetware's Target Korea way someday help answer that question since they may be the box sim precursors to a jet age MMOL game.

[funkedup]

www.targetware.net

[Pei]

I've been following Target Korea since spring (and I'm signed up for the closed beta whenever they get round to expanding it..the F9F with Buckeyes skins are looking very nice: check them out). I was thinking of later periods (who knows the idea of a vietnam variant on the Targetware engine has been bandied about a few times by vadr et al). There certainly seem to be enough people playing Falcon 4 and the various Janes fast jet sims to constitute a market. TK's Project 1 will have an MP element but it sounds as if it's heading more towards a boxed game with some MP aspects and I'm thinking of more along the lines of an AH/Warbirds for fast jets: MP from the ground up with 100s online.
Dinger I take your point with regards to lag and speed effects. Perhaps we have to wait for total high speed broadband (hah).

[Skuzzy]

Dinger brings up a valid point.  The faster an object moves in a 3D game the more extrapolation the game engine must do between position updates.
However, this could be worked around, by reducing the effective changes in the position of the object and manipulating everything else around the object to make up for the difference in speed.
For instance, you could have the client side move the environment to compensate for the reduced movement of the plane you are flying.
This would allow for much less motion of the object while allowing high speed to be a perceived speed.  Thus extrapolation becomes less of a problem, as the environment is usually pretty static.
The only problem with this approach (i am thinking this through as i type it, so bear with me), is the remote planes.  You would still have extrapolation issues.
To get around this the remote planes would have to be considered as part of the environment.

Let's see if this works:
Client plane makes a high speed turn to the left.  There is another high speed plane in view.  We rotate the enviroment to the right, to give the perception of increased speed and move plane 2 with the environment.  Hmmm,..I think it will work.
Basically, we move the environment the equal speed of the client plane in the opposite direction.  This would allow us to reduce the amount of extrapolation by half and could make smooth play.
If that is not enough, you could move the environment by 2 times the actual change of the client.  Basically, you would have to test what would work, but I think this scheme may do the trick.

[moose]

eek. I think my brain just imploded reading that skuzzy.

[Skuzzy]

lol moose.

Another problem with what I propose is the amount of calculations needed to make it work.
What I have done is move some of the bandwidth issues to compensate for lag between updates to the local client CPU, which translates to a significant increase in the amount of floating/double matrix math.
You would need a very fast CPU/FPU unit to be able to do this scheme without visual performance degradation.

moose, here is a simple test to show what I am talking about.  Take a box and mark a dot on the inside of it.  Take a ball and hold it inside a box.  To make a 360 turn in one second, you need to turn the ball once per second.
Now, slow the turn of the ball to one rotation every 2 seconds, but this time rotate the box in the opposite direction once every 2 seconds.  You will note that a full turn from the point in the box to the next time around is 1 second.
Bascally accomplishing the same task as only rotating the ball once per second, but moving the ball half the speed.

[AKDejaVu]

There has been an MMPOL jet sim:

F16/Mig29 by IMOL.

AKDejaVu