Sign up Calendar Latest Topics

  Author   Comment  

Posts: 3,007
Reply with quote  #1 
I have just been repeating my tests for Jump energy costs based on some comments that HaydenBarca made about his Jump Base mod, and I have found that Jump costs WAY less energy now that it used to, by a factor of at least 50. A normal 50 km Jump now costs about 8 energy, when it used to cost 400-500. I have to think that was an accident. 

I was able to confirm that the change was introduced in Artemis 2.5, the release after 2.4, when efficiency was split into shipefficiency and warp/jumpefficiency. Despite the change, in 2.4 the cost of Jump over distance travelled mostly remained the same as 2.3, the version I previously tested. The only changes were to the Dreadnought, and other ship classes that got an improved efficiency from the new calculations. 

I'll also note that according to the energyCoeffJump value in artemis.ini, assuming that we've all been reading it correctly, the energy cost should be 45 energy for 50 km, or 63 energy at difficulty 5. That's about 1/10 of what it was, but 10 times what it is now. Perhaps Thom was trying to fix the calculation and moved the decimal point too far? 

Anyway, this bug has been around for a while, but I thought that since I found it, I should report it. Maybe it has even been discovered already, I have been unable to do any testing for the last year or two since I didn't have a computer. 


Posts: 166
Reply with quote  #2 
Could you specify, what ship are you using for testing this? And what difficulty/default settings are you using?


Posts: 3,007
Reply with quote  #3 
I was using the Light Cruiser, on difficulties 1, 5 and 10. I made jumps in intervals of 10 km all the way up to 50 km. I also tested the Ximni Light Cruiser and got similar results on both a 50 km jump and a Combat Jump, which cost nothing. (That was consistent with my tests of a 10 km jump earlier)

I made the test with a balanced drain on system power that kept Energy nearly constant. I did this by turning off systems until my power gain was about 1 every 10 or 15 seconds. Since the energy cost was so low, I simply waited until power was 999 before jumping.

I did not test any of the other classes, but I expect them to be consistent. I also tested a jump with 175% power to Jump Drive and as I expected, I did not see any difference in cost based on power level. My previous testing under Artemis 2.3 did not show any change in cost at 175% power level. (That is, no change in the cost of the jump itself. Again, system drain was set to be nearly zero)

I will also note that I increased the energyCoeffJump from 0.9 to 9.0, and while I got the expected increase in energy cost, it was not as big as I was expecting. Rather than getting an energy cost of about 80 for a 50 km jump, I got an energy cost of 20. So even the modifier is not behaving as I would expect.

EDIT: I have confirmed that the Scout, Dreadnought and Ximni Light Cruiser, Scout and Dreadnought all have this issue. Their energy cost is proportional to jumpefficiency as you would expect, but cost (at Difficulty 5) varies from 13 units for 50 km (for the TSN Dreadnought) to 3 units. (For the Ximni Scout) 

I'm currently collecting data on Artemis 2.4 to try and get an idea of what the cost is supposed to be for the Ximni. I will note that in my original testing the energy cost of a 50 km Jump was 450, the same as the cost of travelling 50 km at Warp 4 in a Light Cruiser. Since the Jump is instantaneous and the ship can built its energy back up before and after the Jump, I expect this is the intended cost. 


Posts: 3,007
Reply with quote  #4 
Just to confirm, apparently Combat Jump is not supposed to cost any energy. It didn't cost anything back in 2.4. Since your Jump Drive setting determines how fast Combat Jump charges, I assume the energy going into the system is stored and used to make the jump. 

Also, setting energyCoeffJump to 45.0 restores the energy cost to what it was in 2.4. I confirmed that this holds true for all difficulty levels and all ship classes. So apparently, energyCoeffJump now has 1/50 the value it had before, but all other modifiers have remained constant. 

I'm pretty sure I have figured out the formula used for Jump Drive, unless Thom has an additional factor I am not considering. Going from the energy costs I calculated for 2.4, though, it is as follows:

energyCoeffJump * (4 + difficulty coefficient * 4) * distance in km * jump efficiency

The difficulty coefficient used is the same one used for Warp, but the fact that it is multiplied by 4 and then increased by around 4 (4.1 in my tests) made it hard to figure out the formula. The curve is linear, though, increasing by 0.8 every difficulty level, whereas Warp increases by 0.2 every difficulty level. 

I'm not sure where the 1/50 modifier comes in now, whether it is factored into the energyCoeffJump value, or if the calculation of the difficulty coefficient now includes a divide by 50. Or maybe the distance is somehow being measured from 0-1 instead of 0-50. Hopefully that'll help Thom track it down. 


Posts: 311
Reply with quote  #5 
Nice to see some NUMBERS on Jump ships! 
on TeamSpeak as GrayBeard {the Grim}  ;-) 
on Twitter as   @GrayberdTheGrim 

Chief Engineer of the Fulminata, one of the Pirates Of BeechWood! 

Posts: 3,007
Reply with quote  #6 
Yeah, my testing gave very consistent results in 2.4, and when I updated 2.7 to a energyCoeffJump of 45. I actually collected a lot more data that this, but I'll go ahead and post the highlights: 

TSN Light Cruiser, 50 km Jump, Difficulty 1, 5 and 10: 
     287, 430, 611 units - 10 second countdown
TSN Scout, 50 km Jump, Difficulty 5
     431 units - 10 second countdown
TSN Dreadnought, 50 km Jump, Difficulty 5
     518 units - 12 second countdown
Ximni Light Cruiser, 50 km Jump, Difficulty 5
     215 units - 5 second countdown
Ximni Scout, 50 km Jump, Difficulty 5
     172 units - 4 second countdown
Ximni Dreadnought, 50 km Jump, Difficulty 5
     259 units - 6 second countdown

I found that 430 cost for 50 km in a TSN Light Cruiser on Difficulty 5 to be extremely consistent. It should also be obvious that the energy cost is in proportion to the countdown time, which confirms that the jumpefficiency modifier is working on both cost and time. 

Again, these numbers are from Artemis 2.4, and Artemis 2.5 and later the cost is about 1/50th of this. I think I've concluded that my last guess is what happened; in 2.5 apparently Thom changed the Jump slider so it goes from 0.0 to 1.0 instead of 0 to 50. He probably meant to add back in the multiplier somewhere, but forgot, and here we are.

Previous Topic | Next Topic

Quick Navigation:

Easily create a Forum Website with Website Toolbox.