Difference between revisions of "Talk:Leon"
Jump to navigation
Jump to search
Darkcanuck (talk | contribs) (→Quick Question: fast math.pow?) |
(Taylor Series) |
||
| (2 intermediate revisions by 2 users not shown) | |||
| Line 5: | Line 5: | ||
*Well, there were a lot of melee battles, but they tended to be 'everybody vs. Leon' rather than a free-for-all. Candidate for best films of Reno, Oldman, and Portman, all in one, imo. -- [[Martin Alan Pedersen|Martin]] | *Well, there were a lot of melee battles, but they tended to be 'everybody vs. Leon' rather than a free-for-all. Candidate for best films of Reno, Oldman, and Portman, all in one, imo. -- [[Martin Alan Pedersen|Martin]] | ||
* Wow, it would take a really really good bot to be named after "the professional". -- [[User:Darkcanuck|Darkcanuck]] | * Wow, it would take a really really good bot to be named after "the professional". -- [[User:Darkcanuck|Darkcanuck]] | ||
| + | :: When your tank can get close enough to use a knife, you can call it thus... -- [[Martin Alan Pedersen|Martin]] | ||
==Quick Question== | ==Quick Question== | ||
| Line 10: | Line 11: | ||
: Some day I'll integrate that. Do you find Leon slow? What I really need is a fast replacement for Math.pow() -- it's used heavily in my neural networks and reinforcement learner. But in any case Leon needs a complete overhaul for version 2.0 anyway... --[[User:Darkcanuck|Darkcanuck]] 15:39, 3 June 2009 (UTC) | : Some day I'll integrate that. Do you find Leon slow? What I really need is a fast replacement for Math.pow() -- it's used heavily in my neural networks and reinforcement learner. But in any case Leon needs a complete overhaul for version 2.0 anyway... --[[User:Darkcanuck|Darkcanuck]] 15:39, 3 June 2009 (UTC) | ||
| + | |||
| + | : Er, I meant Math.exp()... --[[User:Darkcanuck|Darkcanuck]] 15:54, 3 June 2009 (UTC) | ||
| + | |||
| + | :: Use the Taylor series 1 + x/1! + x^2 / 2! + x^3 / 3!... to calculate e^x, the natural exponent. You can Tailor (ha) the accuracy to whatever you want depending on speed. --[[User:Miked0801|Miked0801]] 23:21, 1 February 2010 (UTC) | ||
Latest revision as of 00:21, 2 February 2010
When I saw the name, Bladerunner naturally popped into my mind. Ironically Leon was great at 1-on-1 but didn't do so well against two opponents. -- Martin
- True, but I think that things would have been different if no guns were involved. :) -- Darkcanuck
The first thing I thought about when reading the name, was the film about the simple Italian soul Leon who was a unbelievable good hitman. How taking care of the neighbour girl fits into Robocode, I haven't figured out yet. -- GrubbmGait
- Well, there were a lot of melee battles, but they tended to be 'everybody vs. Leon' rather than a free-for-all. Candidate for best films of Reno, Oldman, and Portman, all in one, imo. -- Martin
- Wow, it would take a really really good bot to be named after "the professional". -- Darkcanuck
- When your tank can get close enough to use a knife, you can call it thus... -- Martin
Quick Question
Did you try Rednaxela's FastMath class? It may save you a couple of ticks. --Jacob Litewski 15:22, 3 June 2009 (UTC)
- Some day I'll integrate that. Do you find Leon slow? What I really need is a fast replacement for Math.pow() -- it's used heavily in my neural networks and reinforcement learner. But in any case Leon needs a complete overhaul for version 2.0 anyway... --Darkcanuck 15:39, 3 June 2009 (UTC)
- Er, I meant Math.exp()... --Darkcanuck 15:54, 3 June 2009 (UTC)
- Use the Taylor series 1 + x/1! + x^2 / 2! + x^3 / 3!... to calculate e^x, the natural exponent. You can Tailor (ha) the accuracy to whatever you want depending on speed. --Miked0801 23:21, 1 February 2010 (UTC)
