Comments on: Two-dimensional spatial hashing with space-filling curves http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/ Bryan O'Sullivan's blog Wed, 08 Feb 2012 06:41:38 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Mats Rauhala http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/comment-page-1/#comment-328578 Mats Rauhala Wed, 11 Jan 2012 06:39:12 +0000 http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/#comment-328578 Could you point on to some material about hilbert curves, especially about mapping 2d->1d, 1d->2d? Could you point on to some material about hilbert curves, especially about mapping 2d->1d, 1d->2d?

]]> By: taotree http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/comment-page-1/#comment-321964 taotree Sun, 04 Sep 2011 05:29:55 +0000 http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/#comment-321964 Thanks for the info and especially the code implementations. Ezra is correct, there is a typo. I used the code given and it results in incorrect values. I took Ezra's fix and it looks much better now. It would be nice if the author could correct the code in the post. Thanks for the info and especially the code implementations.

Ezra is correct, there is a typo. I used the code given and it results in incorrect values. I took Ezra’s fix and it looks much better now. It would be nice if the author could correct the code in the post.

]]> By: jberryman http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/comment-page-1/#comment-229804 jberryman Thu, 14 May 2009 14:51:49 +0000 http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/#comment-229804 Cool, thanks! I wonder if space-filling curves can be used in functional languages to represent a grid of mutable objects, like a checker board. Cool, thanks! I wonder if space-filling curves can be used in functional languages to represent a grid of mutable objects, like a checker board.

]]> By: Ezra http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/comment-page-1/#comment-45800 Ezra Sun, 13 May 2007 20:10:03 +0000 http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/#comment-45800 Hello-- Is there a typo in the lebesgue algorithm? It appears you're defining x' and y' and then ignoring them. Is the last line meant to be in x' .|. (y' `shiftL` 1) instead? It seems to work better that way. Hello–

Is there a typo in the lebesgue algorithm? It appears you’re defining x’ and y’ and then ignoring them. Is the last line meant to be

in x’ .|. (y’ `shiftL` 1)

instead? It seems to work better that way.

]]> By: DadHacker » Blog Archive » nice http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/comment-page-1/#comment-9272 DadHacker » Blog Archive » nice Mon, 15 Jan 2007 02:02:18 +0000 http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/#comment-9272 [...] A nice way to hash a 2-D point, while preserving some locality.  Link. [...] [...] A nice way to hash a 2-D point, while preserving some locality.  Link. [...]

]]> By: Malcolm Rowe http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/comment-page-1/#comment-8611 Malcolm Rowe Fri, 12 Jan 2007 15:41:03 +0000 http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/#comment-8611 Nice, I was aware of Hilbert curves, but I hadn't heard about Lebesgue curves before! See also: http://xkcd.com/c195.html Nice, I was aware of Hilbert curves, but I hadn’t heard about Lebesgue curves before!

See also: http://xkcd.com/c195.html

]]> By: Bryan O'Sullivan http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/comment-page-1/#comment-8472 Bryan O'Sullivan Thu, 11 Jan 2007 16:38:26 +0000 http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/#comment-8472 Ben - a number of years ago, I wrote a spatial database in Scheme, using Hilbert R-trees. The fast Hilbert distance algorithm was a useful little snippet that fit nicely into the constraints of a blog posting :-) Ben – a number of years ago, I wrote a spatial database in Scheme, using Hilbert R-trees. The fast Hilbert distance algorithm was a useful little snippet that fit nicely into the constraints of a blog posting :-)

]]> By: Patrick Mézard http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/comment-page-1/#comment-8445 Patrick Mézard Thu, 11 Jan 2007 12:59:43 +0000 http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/#comment-8445 For the record, I did performance comparison of Hilbert vs Z-order indexing for a cartographic tiling or Western Europe, mostly containing the road network and water/forest/building features. The indexed database was something like 15/20Gb of vector data, test machine looks like a bi-PIII-700Mhz with 1Gb of memory and SCSI disks (was 7 years ago). I did not find relevant differences despite what several papers/thesis claim (Moon,Faloutsos&Jagadish 1996, Lawder 2001). Maybe the code/tests I wrote were incorrect or biased in a manner which invalidated the expected locality gains, anyway nothing spectacular came from it. I was sorry about it, I thought it was a cool idea. Too bad it just did not work for me. For the record, I did performance comparison of Hilbert vs Z-order indexing for a cartographic tiling or Western Europe, mostly containing the road network and water/forest/building features. The indexed database was something like 15/20Gb of vector data, test machine looks like a bi-PIII-700Mhz with 1Gb of memory and SCSI disks (was 7 years ago).

I did not find relevant differences despite what several papers/thesis claim (Moon,Faloutsos&Jagadish 1996, Lawder 2001). Maybe the code/tests I wrote were incorrect or biased in a manner which invalidated the expected locality gains, anyway nothing spectacular came from it.

I was sorry about it, I thought it was a cool idea. Too bad it just did not work for me.

]]> By: Ben Collins-Sussman http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/comment-page-1/#comment-8444 Ben Collins-Sussman Thu, 11 Jan 2007 12:45:33 +0000 http://www.serpentine.com/blog/2007/01/11/two-dimensional-spatial-hashing-with-space-filling-curves/#comment-8444 Hey Bryan, why the interest in fractal curves? Are you using them in some application? Hey Bryan, why the interest in fractal curves? Are you using them in some application?

]]>