I read that the same way, you can accelerate the ship all you want as long as you don't accelerate to jump velocity before the calculations are finished...doing so simply throws you into a random point when you drop out of the void because the nose wasn't pointed in the proper direction.

Since canon states 3-15 seconds in the void I simply house rule a second for every light year traversed. Beyond that meager time frame it still takes the aforementioned 100 hours of acceleration just to reach jump velocity (five 20hr GST days) and as also mentioned all but two of the AD jump calculations can be made during that accel time (assuming two 'gators working constant ten hour alternating shifts). For those other two the ship can simply alternate between accelerating and decelerating at 1g while the astrogators finish the calculations for the corresponding 11 & 14LY jumps.

Of course there's also the other five 20hr GST days of deceleration on the other end, so it takes a full tenday to complete all but two of those interstellar voyages. As such I need to retract my statement that SF is marginally faster than Traveller, as Traveller mandates a seven 24hr day week for jumping so Traveller is the one that is marginally faster (by 32 hours).

I appreciate this discussion. My own SF universe has several changes from the standard, like I don't use the Rim races but have included their planets. Space travel is not somthing that has been a big part of it so will use all this to figure out how I want to do it, either using KH standard or changing it.

One question if the void trip takes 3-15 seconds but 100 hours to figure, how do you make a quick exit from a system as they descibed in lore about starship battles? Is it watching the attacker coming at you, then a few shots fired at each other and then just watching each other slowly separate?

I'm not familiar with the Dragon article but I'll interject here.

At ADF:4 a UPF frigate can outaccelerate any sathar warship save for another frigate, so depending upon what the other two ships were all the frigate has to do is accelerate at that rate of 4 until it is well out of gunnery range, after that it can "slow down" to a rate of 3 or 2 or whatever the enemy ships top out at for as long as said ships are willing to vainly take chase, and then drop down to 1G if/when they give up.

Obviously the 'gator was trying to smoke the jump in that shortened time and as mentioned failed & misjumped.

Or we can imagine that for a limited combat with just a few ships they were using advanced damage rules and the UPF frigate had taken some ADF hits making its situation so dire as to require a smoked jump.

Nothing in the universe is sitting still.  Whether you are accelerating to jump velocity or docked in a station, all are in motion at all times with variations in velocity  (acceleration) due to gravitational interactions.  Doing astrogation before departure or after makes no difference. 

The more variables you add to calculations and simulations the exponentially longer they become.  This lends to the required use of computers to reduce the calculation times from months to days.  Think of the modern examples of times to render digital animation for movies, time to simulate nuclear explosions, time to plot NASA space ship trajectories.   We build super computers and render farms (aka budget super computers) to do this.  Astrogation is going to be many more variables than r, theta, phi.  This is going to include time, mass, change in mass, mass of each other object passed, velocity vector of each object passed (r, thete, phi, v), velocity vectors of stellar wind, velocity vectors of galactic wind, and maybe a few more know properties of the universe that I'm not thinking of.  Then add in a half dozen or more variables about the void.  Summing all this up you have a minimum of 33 variables for just one ship and the two stars it is traveling between, and ignoring and planets in each system.  If it takes 10 seconds to plot an airplane course, r, theta, phi, v, mass, and change in mass (6 variables), then we can estimate that 33 variables would be something of the magnitude of 10^(33-6)=1e+27 seconds.  That works out to be 3.17e+19 years.  So, astrogators are using computers and they are powerful computers and the calculations take a long time. In fact the times guessed at in the books are generous.

I agree. We can currently plot the course of planets, stars and galaxies without having to calculate their positions each time we try to find them in the night sky. SF astrogators would have all this knowledge at the ready and should be able to figure out the jump calcualtions with a hand calculator in less than 100 hours much less with a level 3 computer with a jump program.

Another game system uses a different method. A recent one I read about has beacons along the common jump routes which allow starships to plot their course pretty quickly. Going to a new system takes longer.

I like the idea of a level 6 robotic brain number crunching a void jump, though. Its not cannon but interesting. 

The astrogator is the hardest working character (or characters, depending on the allowance of teams), so I have no problem with their down time on the other end of the trip. 
 

Beacons aren't stop signs on Alabama back roads. What you are describing is the "money in the mattress" theory. Banks get hacked and robbed so best to keep all my money in my mattress.

Some factors to think of would be the size of the beacons and they would need to be pretty big. They will need to house multiple subspace transmitters to handle the traffic. Basing this on Clarion having a ship dock every 100 to 200 minutes. If those ships are needing 80 to 100 hours of prejump communications that is alot of ships communicating with the beacons.

The transmitters are going to need power. How big a power plant varies depending on other demands like defences and shielding and whether maintenance robots are included.

This does not go with manned beacons but some sort of visitation access would be needed for updates and repairs robots could not make.