Page 6 of 166 FirstFirst ... 4567811 ... LastLast
Results 76 to 90 of 2476

Thread: Buckyball Racing Club presents: The A* Challenge

  1. #76
    Originally Posted by Hanekura Shizuka View Post (Source)
    Well, okay... let's calculate it... some woolly thinking here, but let's assume that, in general, the plots are increased by 0.37, on average. 0.37 ly/jump*roughly 725 jumps (it was 27-29 jumps per thousand, as you'd normally expect, depending on distance from core) = 268.25 ly increase. That's a savings of 7 or 8 jumps, so... about 5 minutes saved.
    The stuff below is more or less what I said in my chat with mw_aurora :

    My last run was with a 37.23ly range 2-jump Asp, but between Sol and the edge of the core only about 3 of my jumps were greater than 36.5, so for that part of the route I'd have been just as well off with my previous ship.

    Once I was inside the core almost every jump was at least 37ly, but the final jump in each leg was always short, sometimes as low as 33ly, so I'm not sure that I gained anything there either. I did travel a lot faster though the core, but I think everyone does as long as they use the correct number sorcery.

    However I was possibly travelling too low on the way, I didn't have any time to plan (or sleep!) before my run, so I used my existing waypoint list 500ly below the plane. Since the chat with mw_aurora I've made my run back at 100ly above the plane, and a significant number of my jumps were 36.8ly - 37.3ly, (I'd left a present with CMDR Metta so my range was up to 37.35).

    That's without a DSS, I just went out with an ADS to make room for the cargo. I don't really want to do that again though.

    I definitely agree that an optimised route is the best way to improve the Asp times, especially where you transit through the brown dwarf layers just above and below the plane, I've had to replot at least once before Aucocks every single time (but never in Aucocks).

    I don't think that Alot's time is beatable in an Asp though - even if you had Cthulhu as a copilot and he made the stars just right!

    Also I had a good time in 1.2 and I know that 1.3 is faster (more consistent witchspace tunnel times). Esvandiary lost quite a bit of time due to obsolete 1.2 FSD technology, so if one of us was to just pip his time, I suspect that he could go out again and smash the current times...

    - - - Updated - - -

    No harm trying though! I just need another 50 million or so, maybe after my rally!

  2. #77
    Originally Posted by Hanekura Shizuka View Post (Source)
    Odd question: Are we doing quotes anymore?
    To quote my response when that was asked earlier in the week:

    Originally Posted by drakhyr View Post (Source)
    There were a number of people who hadn't provided them, and they felt like a bit of a speciality of the Buckyball Run itself. That said, they could make a return if there is sufficient demand...
    Well done on a great time Hanekura Shizuka, and good luck to the others preparing to set off!

  3. #78
    Originally Posted by Hanekura Shizuka View Post (Source)
    Well, okay... let's calculate it... some woolly thinking here, but let's assume that, in general, the plots are increased by 0.37, on average. 0.37 ly/jump*roughly 725 jumps (it was 27-29 jumps per thousand, as you'd normally expect, depending on distance from core) = 268.25 ly increase. That's a savings of 7 or 8 jumps, so... about 5 minutes saved.
    I don't think you can look at the whole run to calculate this, you have to look at individual plots (because of the plotting algorithm). If you try to plot just a couple of LY over your optimal range, you will add another jump. For example, if you magic number is 990, trying to plot 993 Ly could add a whole jump for that extra 3 Ly. Crossing the gaps between arms, you could add 2 jumps for a minimal distance increase.

    Saving 0.37Ly won't alter your optimal plot range enough or start knocking jumps off at that level. You may see the saving in the 'short' plot you have to do to get the right overall distance, but this will be one or two at best.

  4. #79
    Originally Posted by mw_aurora View Post (Source)
    I don't think you can look at the whole run to calculate this, you have to look at individual plots (because of the plotting algorithm). If you try to plot just a couple of LY over your optimal range, you will add another jump. For example, if you magic number is 990, trying to plot 993 Ly could add a whole jump for that extra 3 Ly. Crossing the gaps between arms, you could add 2 jumps for a minimal distance increase.

    Saving 0.37Ly won't alter your optimal plot range enough to start knocking jumps off at that individual plot level. You may see the saving in the 'short' plot you have to do to get the right overall distance, but this will be one or two at best.
    Okay... let's take a look at the two cases that any Bucky runner sees: the core, and the outer areas.

    For the core: Once we're using magic numbers (and that's close to half of the trip), it's optimized plots anyway; that's basically what magic numbers are - optimized lengths of plots to keep the route plotter from having a nervous breakdown. So the increase in average should apply there straight-up. This condition applies for any case where we have knowledge of the magic number.

    As for the outer areas, where we don't have knowledge of the magic number... it gets a little odd. My feeling is that it should have an occasional effect, i.e. about one out of every 4 thousand-ly plots, it offers one less jump. Here's why.

    It's difficult to know what the magic number is outside of the core, because the distribution of stars is changing gradually, and the plotting algorithm will calculate it anyway. So, in some cases, we may have a plot where the number is close to ideal, such as this (hope you don't mind me borrowing your graphics, Alot):

    Name:  NfXPnP.jpg
Views: 70
Size:  50.6 KB

    On the other hand, we might have a case where our number is not so ideal:

    Name:  Wnphpp.jpg
Views: 81
Size:  26.3 KB

    Now, we are basically looking for =MAX[next jump] for most of these jumps - easily the first 22 jumps involved - such that the jump length is less than the limit. If we have a uniform distribution (and we can generally assume a uniform distribution over the span of a single jump, as it doesn't change that much in that short a space), then if we increase the limit, on average, our result will be increased by the amount we increase. With each individual jump, it doesn't guarantee a change - maybe it goes to the same one in some cases, maybe it can reach out and get a much further one in others - but the average should increase by that amount. So, on average, .37*22=8.14 ly. (There's a standard deviation involved here, so sometimes it'll be less, sometimes it'll be greater, but 8.14 should be the average; this increase will have a Normal distribution, thanks to the Central Limit Theorem.)

    So, suddenly, we're about 8.14 ly closer to our target star. Well, if we have very little clue as to what the 'magic number' is because of the change in star distribution, how often will that 8.14 ly increase lead to us having one less plot?

    Our 'guess' as to the magic number is uniformly distributed, because we really don't know. Granted, Alot has come up with ways to estimate, but those estimates are, by his own admission, 'half-decent approximation(s) I made up based on experience.' Thus, our guess in relation to the actual magic number for that time frame could be anywhere along the number plane.

    So. The probability that this increase in jump range leads to an immediate decrease in the number of jumps?

    8.14/37.54=.217, or 21.7%.

    That said, there's something else we haven't taken into account, isn't there? Not only have we had an increase in average jump range through each of the first 22 or so, but we've had an increase in jump range through the last few plots, the one that is going to have our route plotter thinking. So, if we assume 5 jumps with a potential increase, what's the probability that a shorter number of jumps could be achieved there, if the change in jump range didn't get us there? Well, it's the same math. The change in the expected maximum is 5*.37=1.85 ly. We then divide this by the jump length:

    1.85/37.54=.05, or 5%.

    The sum of these two probabilities gives us the probability that an increase in jump range will result in one less jump: 21.7+5%=26.7%. Thus, just a tiny bit over 1 in 4 thousand-ly plots will result in one less jump.

    This is a minimum probability, by the way, because we're not dividing by the right thing; rather, we're dividing by the only jump length we have, which is our maximum. What we really should be dividing by is the actual average jump range, rather than the theoretical maximum. If we have an actual average jump range of 35, for instance, our probability would be 28.5%. This probability would also be higher on the edges because there's more jumps - and, thus, more jumps to eliminate.

    One odd thought, and this may be something fun to ask of Frontier for the next update. When a plot is made, is it possible for them to add a small bit in the galmap showing the number of jumps in the plot, rather than having to go back into the cockpit to look? This, I think, would allow for 'optimal' jumping without taking too much time.

  5. #80
    Just completed run yesterday evening. Was one hell of a trip and made it in (13.31:32)
    .
    Was a bit of a trail run and could have easily taken an hour or so off due to faffing about trying to stream footage and the usual wife requests.
    Might try it again in a few months time.
    .
    Have emailed screenshots as requested.
    .

    Cmdr - Khal Rayark / Sirius Corp
    Ship Name- Nostradamus
    Ship - Anaconda
    Quote- "Bad to the core."

  6. #81
    Esvandiary, hope you'll forgive me but I just had to post your two videos over here ...

    https://forums.frontier.co.uk/showth...=1#post2607138

    Watching those just always makes me want to run again :-)

  7. #82
    Originally Posted by D8ncer View Post (Source)
    Just completed run yesterday evening. Was one hell of a trip and made it in (13.31:32)
    Well done D8ncer! That's a nice time for your first run.

  8. #83
    Originally Posted by D8ncer View Post (Source)
    Just completed run yesterday evening. Was one hell of a trip and made it in (13.31:32) . Was a bit of a trail run and could have easily taken an hour or so off due to faffing about trying to stream footage and the usual wife requests. Might try it again in a few months time. . Have emailed screenshots as requested. . Cmdr - Khal Rayark / Sirius Corp Ship Name- Nostradamus Ship - Anaconda Quote- "Bad to the core."
    Good job, Khal! An excellent first effort.

  9. #84
    Originally Posted by D8ncer View Post (Source)
    Just completed run yesterday evening. Was one hell of a trip and made it in (13.31:32)
    Congrats CMDR

  10. #85
    Originally Posted by D8ncer View Post (Source)
    Just completed run yesterday evening. Was one hell of a trip and made it in (13.31:32)
    Grats CMDR!

  11. #86
    Declaring intent to race:

    Commander Name: Jack Anderson
    Ship Name: A Termite Ploy
    Ship Type: Cobra MK3

    I want to try to complete this challenge with my usual loadout, rather than strip down mass for a better jump range. Obviously this isn't going to be a very fast run, but I'm hoping to beat the slowest time posted for a Cobra.

    Ship Modules:
    Name:  BRCAC Modules.jpg
Views: 71
Size:  382.4 KB

    Ship Hardpoints:
    Name:  BRCAC Hardpoints.jpg
Views: 61
Size:  263.4 KB

    Start Time 16:50:00 31 July 3301:
    Name:  BRCAC Start.jpg
Views: 78
Size:  337.2 KB

  12. #87
    Originally Posted by Frobac View Post (Source)
    I want to try to complete this challenge with my usual loadout, rather than strip down mass for a better jump range. Obviously this isn't going to be a very fast run, but I'm hoping to beat the slowest time posted for a Cobra.
    Good luck Cmdr!
    Beating the slowest time for a Cobra should be entirely achievable. If you're up for it, I wouldn't rule out beating the fastest time for a Cobra, either!

  13. #88
    Evening all! Announcing a "swift" (I hope) jaunt to the core before gamescom duties! o7

    CMDR Elan Solo
    On "Sheiky" Ground
    Anaconda

    I'll be starting round about lunchtime 01-Aug-2015 and ending as soon as I'm able to!

    If anyone is interested at watching my (no doubt disastrous) 1st attempt! I stream my gameplay at: http://www.twitch.tv/elan_

  14. #89
    Originally Posted by Elan View Post (Source)
    Evening all! Announcing a "swift" (I hope) jaunt to the core before gamescom duties! o7

    CMDR Elan Solo
    On "Sheiky" Ground
    Anaconda

    I'll be starting round about lunchtime 01-Aug-2015 and ending as soon as I'm able to!

    If anyone is interested at watching my (no doubt disastrous) 1st attempt! I stream my gameplay at: http://www.twitch.tv/elan_
    Good luck CMDR

  15. #90
    Originally Posted by Frobac View Post (Source)
    Declaring intent to race: Commander Name: Jack Anderson Ship Name: A Termite Ploy Ship Type: Cobra MK3 I want to try to complete this challenge with my usual loadout, rather than strip down mass for a better jump range. Obviously this isn't going to be a very fast run, but I'm hoping to beat the slowest time posted for a Cobra.
    Originally Posted by Elan View Post (Source)
    Evening all! Announcing a "swift" (I hope) jaunt to the core before gamescom duties! o7 CMDR Elan Solo On "Sheiky" Ground Anaconda I'll be starting round about lunchtime 01-Aug-2015 and ending as soon as I'm able to!
    Good luck, CMDRs!

    (I was about to say, "Travel safe," but realized that anyone who planned to travel safe wouldn't go anywhere near a Bucky race!)

Page 6 of 166 FirstFirst ... 4567811 ... LastLast