Sisyphus: completed (maybe)

[Lotharsson]

fstuurman inspired me to revisit Sisyphus and keep tweaking, and now I've got to 251.0:

Sisyphus 251.jpg (8.69 KiB) Viewed 1329 times

...but mine don't seem to have as much oomph as fstuurman's do (they don't get as high once they come off the track), so they probably won't get much higher...

[fstuurman]

Very good!
Does the car lose some triangles on the way or did you manage to design it without? I could not persuade the designer to leave out the bottom triangles. The 251.1 car drops them on start. Which is probably why it is difficult to get progress because if they do not break off the cars scrape the floor.

I notice your front wheel is much smaller than mine. The center of mass is probably lower which makes it stick to the track. On the other hand, a smaller wheel gives less speed. May be I need to try smaller wheels to get more stick.

[Lotharsson]

Does the car lose some triangles on the way or did you manage to design it without?

It drops a couple on the bottom at the start too. So do (IIRC) all of my 250.7 and 250.8 cars.

I notice your front wheel is much smaller than mine. The center of mass is probably lower which makes it stick to the track. On the other hand, a smaller wheel gives less speed. May be I need to try smaller wheels to get more stick.

That's what I found stunning about your cars - I had not imagined that such a large front wheel compared to the rear wheel could get that high because I thought the centre of mass would start dragging the front wheel off the track too early. Maybe I need to try a larger front wheel

I do have a 250.6 with significantly larger wheels than yours, but the rear is a little larger than the front, and I haven't yet been able to improve its score.

[Lotharsson]

One more quantum leap:

Sisyphus 251.1.jpg (9.1 KiB) Viewed 1297 times

Two more such leaps to go

This car just barely lifts off at 251.0 but does get the extra 0.1 point before being terminated immediately. Due to the ... er, somewhat aggressive ... termination criteria I'm quite skeptical that any closely-related descendant will be able to score higher.

[Lotharsson]

I think it might be past time for this thread's title to be changed too

[fstuurman]

One more quantum leap:

Yes, a true quantum leap. Congratulations! It proves that different basic designs are possible, and convinces me that 251.2 must be in reach.

Unfortunately, my computer bluescreened after a day work (stupid ipod). couldn't check the result. New internet explorer 9 is also horribly slow. Saturates my quad core. Download new firefox & again up and running since this morning. Found another 251.1, even lighter than my other one.

[Lotharsson]

I modify my earlier assessment - I suspect 251.2 is reachable with something like my current design. I saw a car reach 251.1 appearing to still be in contact with the track, and then have the origin cross into the red section just out of contact - so close!

[Lotharsson]

Change the title again - I have a 251.2! It is at 251.1 when the origin crosses into the red part of the track, and goes a little higher and gains another 0.1 somehow. It gets terminated when the top of the back wheel is just about to cross into the red.

Fingers crossed for that last 0.1!

[Diseased]

I worked several days on the assumptions
- mass should be as low as possible
- center point should be as low as possible
- mass should be pressing on front wheel, preventing it from lifting

I've reached similar assumptions, except for the low mass.

I've been wondering about the mass issue, so here's my low-mass car (+ a tiny bit of evolution), hahaha:

Code: Select all

eNqz3/H1UH+MxjcHDgYwsH98ncu2YPF1+50zQWCW/U2xdm0mHk84/26EQJLvpY9w9Suh4jD5g2D5T3D++7NqU08C6e9CG2xA6vcw7Z0qCeQfu79pD1AeZIYDt8pmt3bBP/Y3zb2PLX8/DyjE7CCp+DvERVfC/kbg61a5HUH/gcD+z5tdJjNdz9pfWGfx7syZM0B1TA7cWWJK/Z/47R8Hqa7lAYsx2P+/LzHdNMDY/qtXsjFbWhpIr4PA3cuR22uy7ffd3CIFVAMWE7kT9aZ72hn7IwciOCBuYXQQ3Fp3T1fjnv37o5ksN2fOZDhrHcjAcDoVKJfKcPo0A8MMhtMM/2YwMMwuzQXjVKAUSA0uDHQjAK8Klic=

As goofy as my little exercise may be, when I was designing it, I found that reducing the size of the wheels undoubtedly made a direct improvement on the score. When they were about half as large as the maximum size the chassis could barely manage 160, but as I incrementally reduced only the wheel size the score went up consistently and pretty dramatically, capping at something like 240.1 when it started bumping the ground. I didn't even want to make a high-scoring car, I just wanted to make it as tiny/wide as possible.

Whatever sort of way the wheels affect the performance of the cars on Sisyphus, it definitely isn't obvious.

[fstuurman]

Change the title again - I have a 251.2!

You leave me speachless.

Whatever sort of way the wheels affect the performance of the cars on Sisyphus, it definitely isn't obvious.

I can agree with that. Worked several days tweaking the size of the front wheel. Made it a fraction smaller. I have one now at 251.1, but just barely and it does not appear to be in the league of my previous record holder. Now working on same size front & back wheels.

I removed internet explorer 9. Back to 8. Don't know if it is my configuration or something else. Anyway, with ie8, i can run 4-6 simulations together at top speed. with ie9 the top speed is lower and cannot run another at the same time without reducing to a crawl. with the new firefox 4 top speed is somewhat lower that ie8 but acceptable, however it tops with 2 simulations at the same time.