The Kerbal Dakar 2017 entry vehicles thread

[Azimech]

16 minutes ago, katateochi said:

I think default should be the ....er default. Just to make it so most folk's machines can hack it and enable us all to be on the same terrain. The high settings terrain is more challenging but tbh I think default is good enough.  It still has plenty of rover ending terrain, but also has areas where you can really floor it (which I think makes it more fun).

 

 

Actually I discovered something interesting.

1: terrain detail at high hardly has an effect on framerate if part count is low enough, the difference between default and high isn't that great.

2: Terrain detail at default actually makes it harder to drive: there are less potholes and such but because the angles are much steeper the effect of hitting them is many times harsher, the effect of riding down a hill with your bike ... now imagine the same angle and distance riding down the stairs. Going up is even worse. Easy to lose control because you're bouncing everywhere.

 

[DoctorDavinci]

4 minutes ago, Azimech said:

 

Actually I discovered something interesting.

1: terrain detail at high hardly has an effect on framerate if part count is low enough, the difference between default and high isn't that great.

2: Terrain detail at default actually makes it harder to drive: there are less potholes and such but because the angles are much steeper the effect of hitting them is many times harsher, the effect of riding down a hill with your bike ... now imagine the same angle and distance riding down the stairs. Going up is even worse. Easy to lose control because you're bouncing everywhere.

 

So what you're sayin is that default terrain detail seems to be more punishing than high?

Edited December 21, 2016 by DoctorDavinci

[Azimech]

3 minutes ago, DoctorDavinci said:

So what you're sayin is that default terrain detail seems to be more punishing than high?

Yes, it's generally somewhat flatter but especially in the mountains there are lots of very unforgiving and radical changes in terrain angle. Often when a hill begins, it's like hitting a wall.

[DoctorDavinci]

19 minutes ago, Azimech said:

Often when a hill begins, it's like hitting a wall.

Yup ... been experiencing that quite a bit lately 

Stage 2 and 3 are just insane on the suspension when goin at a clip

Edited December 21, 2016 by DoctorDavinci

[Thor Wotansen]

I find the same.  One has to adapt one's driving style to the sharp changes in terrain.

[Slam_Jones]

Decided to do my first timed and recorded sighting lap for Stage 1.  All went mostly well, except for one flip in which I lost a boost flap.  Luckily, the larger bottom flap is the one that really blocks the Junos thrust, so I was able to continue.  I imagine without at least one working boost flap, I can't really prevent myself from flying, and would be out, so protecting it has been important in this case.

The timing, in this case, is a little imprecise, but I set myself up so that at 1 minute mission time, I would be next to the start flag, so I could hit the engines and get rolling.  From what I can tell, it looks like I got within 50m of the Stage 1 flag at about 12:21 mission time, for a stage time of about 11:21.  No idea if this is fast or not, but regardless, my rover is still able to proceed, so that's good enough for now.  Assuming I can make it the whole way without losing anything vital...

Compared to the first vehicle I had built for this competition, the Outlaw is a lot more stable than quick, and as a result, should be able to finish the rally in working order!

[MedwedianPresident]

On 13.12.2016 at 10:54 PM, DarkOwl57 said:

Well, I'm probably just going to have to raise the ride height, but @Draconiator's Etherium Car is going to come to the challenge!!!

Here's my adaptation of his car. Mine is the AS-1A (don't mind the front wing damage; I just finished some testing when I took these on my phone)

Do you know how to take screenshots?

[Ger_space]

On 18.12.2016 at 9:09 PM, Slam_Jones said:

So I did a bit of work with kOS, and have a basic script figured out so far.

I also wanted to have it auto-correct to follow horizon while airborne, but I haven't quite gotten that part figured out.

Regardless, my brakes now automatically deploy in the air (I added the boost flaps to the Brakes group, as well as a red light).  Check it out:

Also still modifying the Spring/Damper settings to get as smooth a ride as possible.  It's a bit better now, but still far from perfect.

(Oh, and also trying to shield the boost flaps better so they don't get ripped off in a crash)

Here is a paper with the formulas for slope and aspect.

http://www.caee.utexas.edu/prof/maidment/giswr2011/docs/Slope.pdf

Good luck and make a post on the kOS reddit, when you have finished your script.

 

[SasquatchM]

In a fit of insomnia last night this was created - no name as of yet.

 

[mrmcp1]

Who needs brakes (or juno engines)?

[katateochi]

10 hours ago, Azimech said:

Actually I discovered something interesting.

1: terrain detail at high hardly has an effect on framerate if part count is low enough, the difference between default and high isn't that great.

2: Terrain detail at default actually makes it harder to drive: there are less potholes and such but because the angles are much steeper the effect of hitting them is many times harsher, the effect of riding down a hill with your bike ... now imagine the same angle and distance riding down the stairs. Going up is even worse. Easy to lose control because you're bouncing everywhere.

I just did another quick comparison and I think you're right. high res has more little bumps that can flip you when going flat out, but the foot hills are not as extreme.  And yes, texture detail on high doesn't seem to impact my FPS. 

[mrmcp1]

Managed to finish the course with my Armadillo buggy

Only rolled once which surprised me. Definitely the most stable of my vehicles (full throttle down a 45 degree incline) 

Group photo showing the state they finished in

[DoctorDavinci]

So I've discovered that the Beaver is amazingly fast and stable on the flats but takin the high road tends to lead to RUD's if I'm not careful ... top speed is well over 130m/s even when going uphill but landing insane jumps is the problem ... gonna have to slow down a bit on the highland stages methinks

As for the electric support vehicle, the B4-R, top speed is around 46m/s on the flats and in the mid 30's going uphill ... and since she's carrying all the fuel needed to top up the Beaver, she is being a bit difficult when it comes to the high road

Loving the challenge so far and I have managed to finish all the stages intact with them ... now to figure out balancing the fuel between the two so that I can make the B4-R lighter and less prone to going boom (she'll get faster as she dumps fuel into the Beaver)

Bill's Big Bad Beaver Stage 1 time is sub 7 minutes ... speed is only useful when me gots flatish ground to drive over

B4-R Stage 1 time is sub 15 minutes ... may have to take the low road with this one in later stages

Edited December 21, 2016 by DoctorDavinci

[Slam_Jones]

22 minutes ago, DoctorDavinci said:

So I've discovered that the Beaver is amazingly fast and stable on the flats but takin the high road tends to lead to RUD's if I'm not careful ... top speed is well over 130m/s even when going uphill but landing insane jumps is the problem ... gonna have to slow down a bit on the highland stages methinks

As for the electric support vehicle, the B4-R, top speed is around 46m/s on the flats and in the mid 30's going uphill ... and since she's carrying all the fuel needed to top up the Beaver, she is being a bit difficult when it comes to the high road

Loving the challenge so far and I have managed to finish all the stages intact with them ... now to figure out balancing the fuel between the two so that I can make the B4-R lighter and less prone to going boom (she'll get faster as she dumps fuel into the Beaver)

Bill's Big Bad Beaver Stage 1 time is sub 7 minutes ... speed is only useful when me gots flatish ground to drive over

B4-R Stage 1 time is sub 15 minutes ... may have to take the low road with this one in later stages

 

Aye, fast is fun, but stable is best!  At least that's the philosophy I put into the Outlaw.

Besides, speed is only good if you can finish the entire rally.  And idk about you guys, but for me, more speed means a much rougher time.

[DrunkenKerbalnaut]

I'm finding that my lead foot is the root for many of my problems. Beginning to think it may be best if use slower wheels to curb my lack of self control haha. 

[Slam_Jones]

1 hour ago, DrunkenKerbalnaut said:

I'm finding that my lead foot is the root for many of my problems. Beginning to think it may be best if use slower wheels to curb my lack of self control haha. 

At first I was a little bummed that I had to tweakscale down the Junos a little bit to fit on the Outlaw... now I am glad for it, since it's not insanely overpowered.  Even with my fancy script (not fancy) it can be hard to control at higher speeds.  Though admittedly, all the script does is automate the boost flaps so I'm not flying, as well as setting SAS to Prograde unless it's going to slow to register Prograde in which case, it switches to Stability (otherwise it would throw an exception and shut off the script).

There were two other little things I wanted to add to the script, but not sure about yet.  One was the slope as stated above, and the other is a 'deadmans switch,' would would constantly check if a Kerbal is present on board the rover.  Assuming the Kerbal gets ejected, it would (ideally) initiate a stop sequence by shutting down the boosters, closing boost flaps, enabling brakes, and rounding up to avoid getting too far from the Kerbal, before finally shutting off the script to conserve electricity.  Got a long ways to go on it, but we'll see how it goes

Edit: Oh and the last thing I wanted to add was a simple math equation going on in the background, which would show my current speed per tick (every 1/4 second as I have it set up now), as well as the average speed over the last X seconds (I'm thinking average of the most recent 10 seconds).  Oh, and lastly, add a small timer to display how long each airborne section lasts.  I can sort of tell by counting how many of each message show up, but having numbers to look at is much easier.

Edited December 21, 2016 by Slam_Jones

[DrunkenKerbalnaut]

1 minute ago, Slam_Jones said:

At first I was a little bummed that I had to tweakscale down the Junos a little bit to fit on the Outlaw... now I am glad for it, since it's not insanely overpowered.  Even with my fancy script (not fancy) it can be hard to control at higher speeds.  Though admittedly, all the script does is automate the boost flaps so I'm not flying, as well as setting SAS to Prograde unless it's going to slow to register Prograde in which case it, switch to Stability.

There were two other little things I wanted to add to the script, but not sure about yet.  One was the slope as stated above, and the other is a 'deadmans switch,' would would constantly check if a Kerbal is present on board the rover.  Assuming the Kerbal gets ejected, it would (ideally) initiate a stop sequence by shutting down the boosters, closing boost flaps, enabling brakes, and rounding up to avoid getting too far from the Kerbal, before finally shutting off the script to conserve electricity.  Got a long ways to go on it, but we'll see how it goes

Deadman switch: Filthy thief, that's my idea  . Had an idea for ground tracking - use radar elevation. Measure every 5ms or so to find the angle. Idk how heavy that would be on KOS, but it's a simple method to get your rise and run. 

I didn't know KOS used EC. Nice. 

[Draconiator]

Remember - Wreck ONCE when this thing actually starts, and you are OUT.  Thankful for this testing period.

Oh, and my kerbals are so excited, some of them are jetting out to camp at the Stage 4 end flag.  I fear I may need to fix a flat after that one...May send a small support vehicle too just in case I have it far from a flag.

[DrunkenKerbalnaut]

WHAT A REVELATION!

 

Well, not that. That's just Bill losing an argument about hardtop roofs. 

Back to the balloon tires, it is. I've lost maybe a minute per stage, but have gained tons of stability. Without SAS. I'll lose so much less time due to rollover recoveries. Also, in concurrence with @katateochi and @Azimech; higher terrain detail is generally a healthier breakfast for a hungry rover. Turns out I don't suck so bad  .

[katateochi]

59 minutes ago, DrunkenKerbalnaut said:

I'm finding that my lead foot is the root for many of my problems. Beginning to think it may be best if use slower wheels to curb my lack of self control haha. 

ah a fellow member of the lead foot league! You can't fight it, you just have to design around it (and brake at the right moments!) 

[DrunkenKerbalnaut]

1 minute ago, katateochi said:

ah a fellow member of the lead foot league! You can't fight it, you just have to design around it (and brake at the right moments!) 

 So it seems! I opted to design around it. I'll have more fun with the whole thing if I can just hold down W and blast some "Rusty Cage" or "Kickstart my Heart". 

[Thor Wotansen]

Them be some good kerballing jams right there.  Also anything Bon Jovi.

[Azimech]

51 minutes ago, DrunkenKerbalnaut said:

 So it seems! I opted to design around it. I'll have more fun with the whole thing if I can just hold down W and blast some "Rusty Cage" or "Kickstart my Heart". 

Ehm ... Highway Star by Deep Purple, cover by Type-O-Negative.

Fuel by Metallica.

[DrunkenKerbalnaut]

Just now, Azimech said:

Ehm ... Highway Star by Deep Purple, cover by Type-O-Negative.

Fuel by Metallica.

What a trio! I've given too many likes today, but any Metallica before ~97 is choice as f---. By the way, which rover sound would you recommend for my jeep? 

[Triop]

Allthough I doubt this will make it, it drives very nice around the KSC.

I call her "9 parts" because, well...

(it has 2 retractable solar panels on the back that you can't see on this picture.)