| 00:35.01 | *** join/#htc-linux donn (n=donn@p5085F5E4.dip.t-dialin.net) |
| 00:35.04 | donn | hiho :) |
| 00:35.05 | donn | i am back |
| 00:35.17 | donn | and engaged to have success to "get" access to the sd card memory :) |
| 00:35.27 | donn | but I think, nobody is here, or? |
| 00:35.28 | donn | :-P |
| 01:05.23 | donn | @cr2: http://forum.xda-developers.com/showthread.php?p=1303494#post1303494 what to think about this? |
| 01:10.42 | donn | do you mean, it could help? its for my smartphone |
| 01:22.20 | *** join/#htc-linux donn (n=donn@p5085F5E4.dip.t-dialin.net) |
| 02:41.50 | *** join/#htc-linux tsdogs_ (n=twostupi@62.123.180.130) |
| 04:04.43 | *** join/#htc-linux scorpio16v (n=scorpio1@ACB3147A.ipt.aol.com) |
| 04:47.08 | *** join/#htc-linux psokolovsky_ (n=psokolov@empty-crossing.volia.net) |
| 06:49.15 | *** join/#htc-linux tsdogs__ (n=twostupi@62.123.180.130) |
| 07:55.20 | cr2_ | tsdogs: good morning. |
| 08:00.59 | *** join/#htc-linux pH5 (n=ph5@p5485ED01.dip.t-dialin.net) |
| 08:05.57 | cr2_ | Kevin2: 'powermon' hangs on jornada820 |
| 08:12.38 | *** join/#htc-linux kiozen (n=oeichler@rgnb-d9bde541.pool.mediaWays.net) |
| 08:17.22 | kiozen | cr2: hi, read the log from last night. |
| 08:18.19 | kiozen | if you need a canvas the one of QLandkarte might help. |
| 08:18.38 | kiozen | all map rendering is done in CGarminMap |
| 08:19.08 | *** join/#htc-linux dion (n=dion@inhex.net) |
| 08:19.16 | kiozen | the base class of CGarminMap is IMap |
| 08:19.44 | cr2_ | kiozen: the raster canvas is needed. with some builtin scaling logic depending on the zoom level |
| 08:20.12 | cr2_ | the best one i know is http://www.gdal.org/gdal_tutorial.html |
| 08:20.19 | kiozen | if I did it right (I won't promise ;) ) you could make your own raster render built on top of IMap |
| 08:20.25 | cr2_ | for reading the data. |
| 08:20.36 | kiozen | zoom is done in the map object |
| 08:21.32 | cr2_ | ok, but zoom must have some limits |
| 08:22.19 | kiozen | everything concerning the map is done by the object drived from IMap |
| 08:22.32 | cr2_ | and it is also necessary to account for the "precompiled" zooms. |
| 08:22.39 | cr2_ | ok, i'll have a look. |
| 08:23.09 | kiozen | the first idea was to write a multipurpose map viewer |
| 08:23.22 | cr2_ | ozi works on a PDA with 16MB RAM, i'd like to have the same :) |
| 08:23.24 | cr2_ | ok. |
| 08:23.57 | kiozen | I abandoned that idea to make live easier for me :) |
| 08:24.03 | kiozen | but now ... |
| 08:24.44 | cr2_ | kiozen: the main requirements are 1. very low memory overhead 2. use the prerendered zooms already available. |
| 08:25.07 | cr2_ | i think most of the tools i know do not do it. |
| 08:25.42 | kiozen | if you concentrate on the while developing it must be possible |
| 08:25.58 | kiozen | GUI programmers often exchange memory for speed :) |
| 08:26.06 | kiozen | but we can change :) |
| 08:26.27 | cr2_ | that#s why i talk about a machine with 16MB RAM :) |
| 08:26.55 | cr2_ | kiozen: c++ and java programmers :) |
| 08:27.20 | kiozen | c++ can be nice to memeory, too |
| 08:27.31 | kiozen | java never ;) |
| 08:27.35 | cr2_ | lol |
| 08:28.01 | kiozen | I like constructors and destructors |
| 08:28.09 | kiozen | it's allways a pain in C |
| 08:28.37 | cr2_ | that's true. |
| 08:28.38 | kiozen | I can avoid virtual functions - if I have :) |
| 08:29.38 | kiozen | I think I plunge into these raster maps in July |
| 08:29.58 | kiozen | will be on vacation end of next week for two weeks |
| 08:30.11 | cr2_ | it seems that i need to fix 'gdalwarp', because GRASS is not designed for the normal people ;) |
| 08:30.48 | kiozen | yes, LSD API, somewhat.. |
| 08:31.33 | cr2_ | kiozen: i stay with my POV: if you don't want to pull your hair out, work only with tiled Mercator |
| 08:32.05 | kiozen | yes, I have read, I agree with you. |
| 08:32.11 | cr2_ | google has started with something not very clever, but them changed to tiled Mercator. |
| 08:32.15 | cr2_ | yahoo too. |
| 08:32.26 | cr2_ | they must have some brains :) |
| 08:32.39 | kiozen | everything else makes you drown in mathematics |
| 08:33.01 | kiozen | it's just the distortion |
| 08:33.10 | cr2_ | yes, and completly irrelevant for the navigation. |
| 08:33.23 | kiozen | I tried to work around with mercator dynamic. |
| 08:33.29 | cr2_ | you can do all these manipulations on the desktop machine. |
| 08:33.34 | kiozen | but that was no good idea either |
| 08:33.50 | cr2_ | for vector it may be aceeptable. |
| 08:33.56 | cr2_ | but for the raster- never. |
| 08:34.23 | kiozen | qlandkarte should make the scale bar dynamic |
| 08:34.27 | kiozen | like google |
| 08:34.30 | cr2_ | ozi is making it more or less right, but not completely to an end. |
| 08:34.40 | kiozen | but for that cs2cs has to go into proj4 |
| 08:35.16 | kiozen | thus you would prefere ozi as format, last night it sounded like tiff |
| 08:35.46 | cr2_ | i will prefer geotiff matching ozi. |
| 08:35.56 | kiozen | :) |
| 08:36.11 | cr2_ | but for that i need to fix buggy gdalwarp |
| 08:36.36 | cr2_ | because the only way i know to produce the right geotiff is to use GRASS. |
| 08:37.35 | cr2_ | and it#s a task for a true linux hardliner with a phd in geodesy :) |
| 08:38.07 | kiozen | ok go for it. tomorrow its done? |
| 08:38.09 | kiozen | ;) |
| 08:38.23 | cr2_ | and gdalwarp just segfaults on such operation ;) |
| 08:38.48 | kiozen | thus it did not pass the unit tests? |
| 08:38.53 | kiozen | lol |
| 08:38.55 | cr2_ | the funny thing here is that GRASS actually uses the same gdal API as gdalwarp. |
| 08:39.14 | cr2_ | RasterIO |
| 08:40.04 | cr2_ | btw, gdalwarp works on windows. |
| 08:41.15 | cr2_ | so i'll propose to use a desktop computer to produce a mercator 8bit tiled geotiff. |
| 08:41.38 | cr2_ | with "overviews". |
| 08:42.09 | *** join/#htc-linux rob_w (n=bob@BACac76.bac.pppool.de) |
| 08:42.10 | cr2_ | i.e. prerendered 2^zoom |
| 08:42.29 | kiozen | bay. landesvermessungsamt uses tiff, too |
| 08:42.37 | kiozen | but if it is geotiff? |
| 08:43.13 | cr2_ | if you have *.tfw files and use DFDN proj -then you can easily create geotiff. |
| 08:44.54 | kiozen | they even have a pda viewer |
| 08:45.02 | cr2_ | kiozen: do you have a sample file ? |
| 08:45.14 | cr2_ | i think 'learning by doing' is the best :) |
| 08:45.25 | kiozen | not yet, let's see if they have one on the page |
| 08:47.11 | kiozen | http://www.geodaten.bayern.de/ |
| 08:47.16 | kiozen | service and download |
| 08:47.56 | cr2_ | Nähere Informationen erhalten |
| 08:47.56 | cr2_ | <PROTECTED> |
| 08:48.00 | cr2_ | lol |
| 08:48.32 | kiozen | tiff and tfw everything there |
| 08:48.44 | cr2_ | ok, that's good. |
| 08:49.03 | cr2_ | now we only need the projection, but i think it's standard. |
| 08:49.50 | kiozen | I guess I will order the Topo50 for Bavaria |
| 08:50.14 | cr2_ | Testdaten der digitalen Ortskarte (DOK) 1:10 000 |
| 08:50.49 | kiozen | thoses maps let Garmin look bad |
| 08:51.05 | kiozen | on the other hand side the vector format would explode |
| 08:51.18 | kiozen | Moscow is a good example for that |
| 08:51.19 | cr2_ | hehe. think about the rendering speed ;) |
| 08:51.47 | cr2_ | and the moscow map didn't include topography. |
| 08:52.01 | cr2_ | altitude lines. |
| 08:52.13 | kiozen | that surely would drive it into polyline limits :) |
| 08:52.32 | cr2_ | don#t think so. |
| 08:52.58 | kiozen | there is one 2^16 if i remember right |
| 08:53.21 | cr2_ | per tile. |
| 08:53.35 | kiozen | Moscow is just one tile :) |
| 08:53.55 | cr2_ | unlikely. |
| 08:54.10 | cr2_ | the 2^16 limit is in the roadmap |
| 08:54.15 | cr2_ | per "county" |
| 08:54.30 | cr2_ | but in moscow you have houses as polygons. |
| 08:54.41 | kiozen | wow, they draw every tree on those topo maps |
| 08:54.51 | kiozen | so what's with Kyrill ? :) |
| 08:55.10 | cr2_ | hehe. not that many trees in Bavaria :) |
| 08:55.28 | cr2_ | yeah, we had some troubles here. |
| 08:55.50 | kiozen | sorry, son comes back from a week vacation, have to quit, bye |
| 08:55.51 | cr2_ | a lot of broken trees are still in the forests around. |
| 08:55.55 | cr2_ | bye |
| 08:59.28 | *** join/#htc-linux RoEn (n=roen@p54A67A45.dip.t-dialin.net) |
| 09:03.53 | cr2_ | hm. it's called DHDN, i always forget it ;) |
| 10:37.22 | *** join/#htc-linux |dion| (n=dion@inhex.net) |
| 10:40.59 | *** part/#htc-linux scorpio16v (n=scorpio1@ACB3147A.ipt.aol.com) |
| 10:47.47 | *** join/#htc-linux donn (n=donn@p5085E573.dip.t-dialin.net) |
| 10:47.52 | donn | hi@all :) |
| 10:57.18 | *** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com) |
| 11:07.14 | *** join/#htc-linux ellisway (n=ellis@host-87-74-241-174.bulldogdsl.com) |
| 11:27.43 | goxboxlive | cr2_, whats special with the gpsd at xanadux cvs? |
| 12:00.36 | cr2_ | goxboxlive: it's the last GPL release, and i have GPL patches for it. magnetic declination and sirfmon. maybe some others. |
| 12:01.17 | goxboxlive | ok, so i should rather use that source instead of the one provided in OE? |
| 12:02.00 | cr2_ | no, not yet. |
| 12:02.27 | goxboxlive | ok |
| 12:02.30 | cr2_ | i also need to commit my qt4 patches from yesterday. |
| 12:02.40 | cr2_ | there are also some qt3 bugs. |
| 12:03.29 | *** join/#htc-linux bd2 (n=cbou@89.113.85.104) |
| 12:10.14 | cr2_ | goxboxlive: committed the roadmap_editor qt4 fixes. |
| 12:10.27 | cr2_ | have you tried it on qtopia ? |
| 12:10.36 | goxboxlive | cr2_, ok good, how do i build it? |
| 12:11.36 | goxboxlive | no i havent |
| 12:11.37 | goxboxlive | do i build as the Makefile in OPieII |
| 12:11.37 | goxboxlive | like BabelO said we shoul? |
| 12:11.37 | goxboxlive | +d |
| 12:21.48 | cr2_ | don#t know. |
| 12:21.52 | cr2_ | i use |
| 12:22.26 | cr2_ | make DESKTOP=QT4 SHAPEFILES=NO SSD=YES QTDIR=/usr/local/TroolTec/4.3.0 |
| 12:25.52 | goxboxlive | and then you install it to your sd card and run it form OPie2? |
| 12:28.48 | *** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 12:36.13 | cr2_ | goxboxlive: no. only for the desktop. |
| 12:36.37 | cr2_ | BabelO: opie2 compile instructions needed :) |
| 12:56.48 | BabelO | hi cr2_ |
| 12:58.33 | donn | hi cr2 :) |
| 12:58.41 | donn | and, did you get the link from me :-P |
| 13:00.54 | BabelO | cr2_: did you check out new tree ? |
| 13:13.32 | donn | does somebody knows the standard ip address of "lin Prophet"? |
| 13:13.35 | donn | 10.226...?? |
| 13:17.03 | donn | okay, 10.226.6.6 :) i found it :) |
| 13:17.06 | BabelO | donn: uncompress tar.gz image on your computer and look in /etc/network/interfaces |
| 13:17.16 | donn | i found it... thanks .) |
| 13:17.29 | donn | @Babel0, perhaps you can help me, I have a small question |
| 13:17.40 | donn | for a few days I try to have access on the SD Cards |
| 13:17.46 | donn | well, I found a distribution, don |
| 13:18.09 | donn | 'don't know the name, and in the default.txt is the paramter root=/dev/mmcblk0 |
| 13:18.16 | donn | but the distribution doesn't work |
| 13:18.34 | donn | in my distribution, thats running well, there is a problem, that the root is set to /dev/ram0 |
| 13:18.48 | donn | so I changed it to mmcblk0 |
| 13:19.13 | donn | do you know, how I can check where the root partition is set to when I am connected via ssh? |
| 13:19.39 | BabelO | look with dmesg |
| 13:19.58 | BabelO | or only use df command |
| 13:20.20 | donn | hm, /dev/root.old |
| 13:20.21 | donn | ? |
| 13:21.48 | BabelO | check in dmesg, your partition number should be something like /dev/mmcblk0p1 if it is the first or p2 for second |
| 13:23.03 | donn | <5>Kernel command line: root=/dev/mmcblk0p2 init=/linuxrc |
| 13:23.17 | donn | hm, thats strange |
| 13:23.24 | donn | I have only 1 mb free space |
| 13:23.32 | donn | the same, when its load into "ram0" |
| 13:24.36 | donn | i split my sd card into two partitions.. one is "ext2" and the other is "fat" |
| 13:29.19 | donn | i try to mount each of the devies, because there are many displaced |
| 13:29.31 | donn | displayed, but nothing works |
| 13:29.37 | donn | do you have an idea? |
| 13:56.43 | *** join/#htc-linux sadeness_ (i=LamersIn@pppoe-92-48.elsv-v.ru) |
| 13:57.24 | *** join/#htc-linux hlbot (n=adm@iclem.net) |
| 13:59.33 | *** join/#htc-linux asylumed (n=insanity@dsl-245-191-92.telkomadsl.co.za) |
| 14:05.48 | cr2_ | donn: you don't have a working SD controller, so there is no /dev/mmcblk* for you. |
| 14:06.16 | cr2_ | tsdogs: are you alive ? :) |
| 14:09.19 | donn | @yeah, but the xda Atom works.. so i thought, it has the same controller |
| 14:10.20 | donn | @cr2: when do you have 10 minutes time to finish this operation? |
| 14:30.18 | Kevin2 | cr2_: Jornada works otherwise with CVS haret? |
| 14:36.20 | *** join/#htc-linux kiozen (n=oeichler@rgnb-d9bde541.pool.mediaWays.net) |
| 14:38.36 | kiozen | cr2: Topo10/50 projection: Gauss Krüger (12°), bezogen auf Bessel-Ellipsoid, Potsdam Datum |
| 14:38.47 | kiozen | no mercator :( |
| 14:47.57 | cr2_ | Kevin2: yes. |
| 14:48.13 | cr2_ | donn: have you got the latest haret version ? |
| 14:49.31 | cr2_ | kiozen: it's not a problem to convert DHDN into mercator. i've checked gdalwarp in gdal-1.4.1 and it does not segfault anymore, so i can create the datasets that i want. |
| 14:50.24 | cr2_ | Kevin2: i'll check with Athena a bit later. |
| 14:50.25 | kiozen | ack, I start to install grass.. |
| 14:50.42 | cr2_ | kiozen: have you already got qgis ? |
| 14:51.15 | cr2_ | GRASS is really for the people who know very well what they are doing :) |
| 14:51.34 | kiozen | not yet |
| 14:52.03 | kiozen | doesn't qgis work with grass? |
| 14:52.17 | cr2_ | ok. qgis will let you to check the raster datasets very nicely. and overlay any vector datasets and gpx tracks. |
| 14:52.44 | cr2_ | grass is used as a plugin in qgis. |
| 14:53.01 | cr2_ | qgis can do a lot even without grass. |
| 14:53.36 | kiozen | ah, started to download qgis :) |
| 14:55.13 | cr2_ | i've finally learned how to use 'r.thin' and 'r.to.vect' in grass for vectorization of roads. amazing :) |
| 14:55.48 | cr2_ | that is the first and most important application in my mind. |
| 14:56.20 | cr2_ | but if you'll search in google, you'll find very obscure applications, but not that one. |
| 14:56.41 | cr2_ | the same for graph/network functions. |
| 14:57.54 | cr2_ | the manpage for v.net.path deserves a prize for the most obscure manpage i've seen in my life. |
| 15:05.09 | kiozen | I hate automake, I hate SuSE's Qt installation... |
| 15:06.15 | *** join/#htc-linux freshr (n=freshr@89.113.75.13) |
| 15:06.33 | *** part/#htc-linux freshr (n=freshr@89.113.75.13) |
| 15:08.32 | cr2_ | kiozen: i have qt4 in /usr/local/TrollTech/4.3.0 |
| 15:09.19 | kiozen | me /usr/local/qt4, but includes are in /usr/include and libs in /usr/lib64, arrrg! |
| 15:09.32 | kiozen | but I have got it, need gdal |
| 15:09.40 | kiozen | compiling.... |
| 15:10.06 | kiozen | cr2: compile without grass support, that's ok? |
| 15:12.08 | cr2_ | suse grass is very old. |
| 15:12.24 | cr2_ | if you want grass you should compile the CVS version. |
| 15:12.26 | kiozen | do I need grass at all? |
| 15:12.31 | cr2_ | no. |
| 15:12.38 | kiozen | ok, skipped |
| 15:12.59 | cr2_ | gdal+ogr. this is the core. |
| 15:14.50 | kiozen | dual core is fun to compile |
| 15:14.57 | kiozen | the only task you really need it |
| 15:18.49 | kiozen | cr2: you know a linkto geos? |
| 15:20.01 | kiozen | never mind, got it |
| 15:51.11 | kiozen | cr2 |
| 15:51.14 | kiozen | ../core/.libs/libqgis_core.so: undefined reference to `geos::geom::Coordinate::Coordinate(double, double, double)' |
| 15:51.22 | kiozen | does that ring a bell? |
| 15:53.54 | cr2_ | which geos lib do you have ? |
| 15:54.07 | cr2_ | and what qgis version do you compile ? |
| 15:54.11 | kiozen | 3.0.0rc4 |
| 15:54.22 | kiozen | qgis-0.8.0 |
| 15:54.54 | cr2_ | <PROTECTED> |
| 15:55.00 | cr2_ | take 2* |
| 15:55.04 | kiozen | ok, i try |
| 16:00.04 | Kevin2 | cr2_: Looks like a bug. Can you "cvs up" and try again? |
| 16:00.42 | Kevin2 | cr2_: Also, can you try the new command: loadfunc coredll Sleep |
| 16:01.26 | Kevin2 | (The new command is loadfunc - a sample command would be "loadfunc coredll Sleep".) |
| 16:04.35 | cr2_ | did not help. |
| 16:05.41 | cr2_ | it hangs "more" that with the "help" before. i need to do a soft reset. |
| 16:05.51 | cr2_ | trying loadfunc |
| 16:06.51 | cr2_ | Unknown keyword: `loadfunc' ? |
| 16:07.38 | cr2_ | this is HaRET pre-0.4.9-20070610_015955 |
| 16:14.46 | cr2_ | Kevin2: sorry, it was some mistake. |
| 16:14.50 | cr2_ | powermon works |
| 16:14.53 | cr2_ | Function 'Sleep' in library 'coredll' at 01FA94EC |
| 16:16.33 | *** join/#htc-linux scorpio16v (n=scorpio1@ACB7510F.ipt.aol.com) |
| 16:16.34 | Kevin2 | cr2_: Okay. |
| 16:18.13 | cr2_ | Kevin2: do you disable MBGNT ? |
| 16:18.24 | Kevin2 | What is that? |
| 16:18.33 | cr2_ | then i'll try to boot the barebones kernel. |
| 16:18.45 | cr2_ | hmm. busmaster DMA grant |
| 16:18.53 | cr2_ | is also present on PXA. |
| 16:19.22 | Kevin2 | On StrongArm? There is no shutdown code there at all. See src/mach/arch-sa.cpp |
| 16:19.25 | cr2_ | http://wiki.xda-developers.com/index.php?pagename=BlueangelGPIO |
| 16:19.30 | cr2_ | gpio13 |
| 16:19.38 | cr2_ | ok. |
| 16:20.07 | cr2_ | where should i put the device-specific (not cpu-specific) shutdown code ? |
| 16:20.32 | Kevin2 | cr2_: Clone src/mach/universal and then fill in the hardwareShutdown stuff. |
| 16:20.39 | cr2_ | ok. |
| 16:20.52 | Kevin2 | src/mach/mach-universal.cpp |
| 16:20.57 | cr2_ | yes. |
| 16:21.19 | cr2_ | i need to stop the external OHCI, which has busmaster DMA access. |
| 16:21.58 | Kevin2 | Yeah - I'd make src/mach/mach-jornada.cpp. Just be sure to remove the jornada line from machlist.txt |
| 16:22.39 | cr2_ | i think blueangel will fail to boot too with the camera switched on. |
| 16:23.24 | cr2_ | just never tried it. |
| 16:23.35 | Kevin2 | Wont turning off dma in the core prevent the DMA REQ line from being used? |
| 16:23.35 | cr2_ | bbl. |
| 16:24.26 | cr2_ | don#t know, but on sa11x0 the SDRAM control is overtaken by the companion chip completely. |
| 16:24.36 | cr2_ | EDO RAM, actually :) |
| 17:11.04 | *** join/#htc-linux goxboxlive (n=goxboxli@195.159.97.196) |
| 17:11.13 | goxboxlive | BabelO hi |
| 17:11.24 | goxboxlive | Wich new tree are you talking about? |
| 17:19.49 | *** join/#htc-linux ZioPRoTo (n=ZioPRoTo@213.155.204.219) |
| 17:19.56 | goxboxlive | tsdogs are you here? |
| 17:20.17 | ZioPRoTo | Hello |
| 17:20.27 | goxboxlive | ZioPRoTo hi |
| 17:20.53 | ZioPRoTo | I'm been browsing handhelds.org |
| 17:21.18 | ZioPRoTo | But I cant find fresh images for HTC Blue Angel online :( |
| 17:21.49 | ZioPRoTo | On GNULinux.biz they have very new stuff for HTC Universal... but stuff for Blue Angel is one year old :( |
| 17:32.50 | *** join/#htc-linux ellisway (n=ellis@host-87-74-241-174.bulldogdsl.com) |
| 17:33.06 | goxboxlive | ZioPRoTo I know, but as far as i know there are someone who has started to make a new one now. |
| 17:33.58 | goxboxlive | ZioPRoTo But if you know how to compile your own kernel and modules can can use the universal image and aply the BA modules to it instead of the Universal ones. |
| 17:34.11 | goxboxlive | also remember to change the kernel zImage. |
| 17:34.58 | goxboxlive | There might be other thinhgs top change also, but you'll notice that when you boot it up. I think the BT and IRDA port is difftrent |
| 18:07.28 | *** part/#htc-linux scorpio16v (n=scorpio1@ACB7510F.ipt.aol.com) |
| 18:13.46 | *** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com) |
| 18:32.52 | cr2_ | goxboxlive: does roadmap compile for you ? |
| 18:35.59 | goxboxlive | cr2_, i havent been home, so i havent tried it yet. |
| 18:36.12 | *** join/#htc-linux kiozen (n=oeichler@rgnb-d9bde541.pool.mediaWays.net) |
| 18:36.51 | cr2_ | goxboxlive: oslofjorden. should i move NSWE ? |
| 18:37.49 | cr2_ | hm. nothing in the west. |
| 18:38.57 | goxboxlive | cr2_, ?? NSWE? I dont know what it is. Do what you think is best |
| 18:39.42 | cr2_ | north,south,east west. |
| 18:39.58 | cr2_ | it seems i don't have anything for Norway. |
| 18:42.19 | cr2_ | hehe. This product was discontinued as of 28th January 2004. |
| 18:47.16 | kiozen | hi, cr2: got qgis, guess I need gdalwarpgdalwarp |
| 18:47.24 | kiozen | ops |
| 18:48.37 | kiozen | ok I have got already |
| 18:49.27 | cr2_ | it's part of gdal. |
| 18:49.42 | ZioPRoTo | goxboxlive: thanks for your answer :) |
| 18:52.32 | cr2_ | kiozen: do you have a sample map ? |
| 18:52.40 | kiozen | yep |
| 18:52.56 | kiozen | just wanted to ask how to convert to geotiff :) |
| 18:53.17 | kiozen | ordered topo10 :) |
| 18:54.47 | cr2_ | kiozen: tell me the sample name, and where i can grab it. |
| 18:55.30 | kiozen | http://www.geodaten.bayern.de/bvv_web/downloads/DOK_zusammen_Aufl_200.zip |
| 18:55.38 | cr2_ | since yo have .tfw the only thing that is needed is the right coordinate system. |
| 18:56.43 | cr2_ | ok, so it's a1414col.tif for the image, and a1414col.tfw for the "world file". |
| 18:57.07 | kiozen | yes and Kundeninfo for the projection system |
| 18:57.08 | cr2_ | looking at .tfw |
| 18:57.26 | cr2_ | .5m ? hehe. |
| 18:57.54 | kiozen | what do the numbers mean |
| 18:58.14 | cr2_ | 4431000, 5331999 |
| 18:58.22 | goxboxlive | qtopia-core-4.2.0 builds in OE. I'll try to make it closer to our version |
| 18:58.23 | cr2_ | i'll tell you. |
| 18:58.40 | cr2_ | these are 6 numbers for the affine transform. |
| 18:58.49 | kiozen | 44* seems to be the meter offset |
| 18:58.49 | cr2_ | the rotation is 0. |
| 18:59.02 | cr2_ | the 4431000, 5331999 |
| 18:59.08 | kiozen | yes |
| 18:59.15 | cr2_ | are the upper left corner coordinates. |
| 18:59.28 | cr2_ | where do you have gdal installed ? |
| 18:59.39 | kiozen | on hdd ? |
| 19:00.07 | cr2_ | yes. i have it in /usr/local |
| 19:00.20 | cr2_ | usr/local/share/gdal/pcs.csv |
| 19:00.24 | kiozen | me /usr :) I hate local |
| 19:00.48 | cr2_ | ok. |
| 19:00.58 | cr2_ | search for DHDN |
| 19:00.59 | kiozen | found file |
| 19:01.15 | kiozen | as file? or in qgis |
| 19:01.24 | cr2_ | in this file |
| 19:01.36 | cr2_ | 31464,"DHDN / 3-degree Gauss zone 4",9001,4314,16264,9807,1,1,8801,0,9102,8802,12,9102,8805,1,9201,8806,4500000,9001,8807,0,9001,, |
| 19:01.45 | kiozen | ok |
| 19:01.50 | cr2_ | ---4500000--- |
| 19:02.16 | cr2_ | so your coordinate system is epsg:31464 |
| 19:02.24 | cr2_ | the first number |
| 19:02.35 | kiozen | aha |
| 19:02.44 | cr2_ | you can use cs2cs to check it. |
| 19:03.20 | kiozen | ? how |
| 19:04.51 | cr2_ | In qgis: Einstellungen->Projekt Einstellungen->Projektion->Suchen |
| 19:05.18 | cr2_ | EPSG ID : 31464 Suchen |
| 19:05.36 | cr2_ | sorry. Finden |
| 19:05.45 | cr2_ | QGIS SRSID: 1821 |
| 19:05.45 | cr2_ | PostGIS SRID:31464 |
| 19:05.45 | cr2_ | +proj=tmerc +lat_0=0 +lon_0=12 +k=1.000000 +x_0=4500000 +y_0=0 +ellps=bessel +datum=potsdam +units=m +no_defs |
| 19:05.59 | cr2_ | that looks good. |
| 19:06.47 | cr2_ | this is the 'proj4' name(string) for your EPSG projection. |
| 19:07.22 | kiozen | ok |
| 19:08.36 | cr2_ | so |
| 19:08.40 | cr2_ | cs2cs -f "%.9f" +proj=tmerc +lat_0=0 +lon_0=12 +k=1.000000 +x_0=4500000 +y_0=0 +ellps=bessel +datum=potsdam +units=m +no_defs +to +proj=longlat +ellps=WGS84 +no_defs |
| 19:08.50 | cr2_ | type your coordinates |
| 19:09.11 | cr2_ | 4431000 5331999 |
| 19:09.13 | cr2_ | get |
| 19:09.22 | cr2_ | 11.071795091 48.122270219 |
| 19:09.23 | kiozen | 11.071795091 48.122270219 4.420538137 |
| 19:09.26 | cr2_ | long / lat |
| 19:09.38 | kiozen | and the third? |
| 19:09.42 | cr2_ | z=4.4 m over the ellipsoid. |
| 19:09.47 | kiozen | ah |
| 19:10.00 | cr2_ | actually we always have 3 coordinates. |
| 19:10.10 | cr2_ | and transform with 3. |
| 19:10.14 | kiozen | cross check with qlandkarte :) |
| 19:10.24 | cr2_ | but if you omit the third, it's assumed to be 0 |
| 19:10.33 | cr2_ | :) |
| 19:11.24 | kiozen | there is a trail missing, Garmin sucks :) |
| 19:12.04 | cr2_ | lol. |
| 19:12.13 | kiozen | <PROTECTED> |
| 19:12.40 | cr2_ | never was in the west :) |
| 19:13.16 | kiozen | huh? |
| 19:14.16 | cr2_ | Nymphenburg was the most western point. |
| 19:14.24 | cr2_ | where i've been :) |
| 19:15.12 | kiozen | ah, but Nymphenburg is ok as western reference point :) |
| 19:16.01 | cr2_ | i lived 5 years in München, but in the east part. |
| 19:16.14 | kiozen | I lived 22 year in Munich south east :) |
| 19:16.27 | kiozen | I am born there |
| 19:16.34 | cr2_ | ok. |
| 19:16.42 | kiozen | riemerling |
| 19:17.08 | cr2_ | i see. i lived in Arabellapark. |
| 19:17.28 | kiozen | watching maniacs jump ;) |
| 19:17.43 | cr2_ | the world is small :) |
| 19:18.12 | kiozen | oh yes, there are just 7 links to each human |
| 19:18.19 | cr2_ | yeah. |
| 19:18.54 | cr2_ | ok, so you can replace the longlat in cs2cs to 'merc' |
| 19:19.04 | *** join/#htc-linux dion (n=dion@inhex.net) |
| 19:19.12 | cr2_ | +proj=merc |
| 19:19.50 | cr2_ | 1232506.591726166 6095389.532984379 4.420538137 |
| 19:20.13 | cr2_ | so there are the mercator coordinates for this point. |
| 19:20.28 | cr2_ | fyi. |
| 19:20.38 | rob_w | cr2_, do you know how to determine the page size and erase block size when creating a jffs2 image ? |
| 19:20.46 | kiozen | ok, so how do i do this for every point? |
| 19:21.04 | cr2_ | rob_w: not anymore. too far in the past. |
| 19:21.13 | rob_w | crap |
| 19:21.20 | cr2_ | kiozen: no. it was just the information. |
| 19:21.40 | cr2_ | gdalwarp is much more user-friendly. |
| 19:21.52 | cr2_ | in the most simple case: |
| 19:23.18 | cr2_ | gdalwarp -t_srs epsg:31464 a1414col.tif XXX.gtif |
| 19:23.28 | cr2_ | XXX.gtif is the geotiff. |
| 19:23.45 | cr2_ | you can check it with 'gdalinfo XXX.gtif' |
| 19:24.15 | cr2_ | it's a normal tiff file with embedded metadata. according to the geotiff convention. |
| 19:24.35 | cr2_ | you can see it with 'qgis XXX.gtif' |
| 19:24.36 | rob_w | is it recommended to pad a jffs2 into 1024x1024xX ?? wise |
| 19:24.41 | kiozen | A target coordinate system was specified, but there is no source coordinate |
| 19:24.42 | kiozen | system. Consider using -s_srs option to provide a source coordinate system |
| 19:24.44 | cr2_ | and overlay the gpx tracks. |
| 19:24.46 | rob_w | i padded currently to 9000000 |
| 19:24.58 | kiozen | is that ok |
| 19:25.06 | cr2_ | kiozen: sorry. |
| 19:25.27 | cr2_ | t_srs is the target . |
| 19:25.33 | cr2_ | s_srs is the source. |
| 19:25.42 | cr2_ | mercator is epsg:54004 |
| 19:25.47 | cr2_ | so it will be |
| 19:26.09 | cr2_ | gdalwarp -s_srs epsg:31464 -t_srs epsg:54004 a1414col.tif XXX.gtif |
| 19:26.58 | kiozen | ok |
| 19:27.42 | cr2_ | you can use |
| 19:28.03 | cr2_ | gdalwarp -s_srs epsg:31464 -t_srs epsg:4326 a1414col.tif XXX-WGS.gtif |
| 19:28.25 | cr2_ | to have the lat/long projection. |
| 19:29.12 | kiozen | hm I would have expected the gridlines to be exactly horiz/vert now |
| 19:29.32 | cr2_ | no. |
| 19:29.44 | kiozen | where is my fault? |
| 19:30.25 | cr2_ | the Gauß-Krüger projection grid (Transverse Mercator) is vertical only at the center. |
| 19:30.30 | cr2_ | at x=4500000 |
| 19:30.59 | cr2_ | there is no fault. |
| 19:31.21 | kiozen | but if I scale to mercator should move the points to the correct alignment? |
| 19:31.35 | cr2_ | yes. |
| 19:32.21 | *** join/#htc-linux pancake (n=xxx@55.pool85-52-227.static.orange.es) |
| 19:32.55 | kiozen | thus I would expect the grid to be aligned now |
| 19:35.07 | *** join/#htc-linux ellisway (n=ellis@host-87-74-241-174.bulldogdsl.com) |
| 19:35.19 | cr2_ | i'll check myself. |
| 19:36.00 | kiozen | I would expect parabolic tile borders, too |
| 19:38.57 | pancake | hi cr2_ any advances on the hermes front? |
| 19:39.18 | cr2_ | kiozen: the map is too small to see it. the grid is tilted, that's right. |
| 19:39.46 | cr2_ | pancake: the NAND works. now we need think about its partitioning. |
| 19:40.17 | pancake | the sd does not? |
| 19:41.03 | cr2_ | no. too difficult yet :) |
| 19:41.36 | cr2_ | kiozen: i've overlaid the OSM vector layer from April that i have. |
| 19:41.53 | kiozen | link |
| 19:41.56 | pancake | cr2_: we can use the extended rom for linux? |
| 19:42.26 | cr2_ | two streets and the highway are where they should be. but a bit coarse. |
| 19:42.48 | cr2_ | pancake: do you know the block numbers ? |
| 19:43.13 | pancake | no, but i can look for them |
| 19:43.35 | cr2_ | kiozen: i've converted .osm to .shp and imported it into grass. |
| 19:44.12 | cr2_ | pancake: we need to define the partitioning. look at the MTD struct in htchermes.c |
| 19:44.17 | kiozen | can you send it via email? |
| 19:44.35 | cr2_ | the picture ? or the shapefile ? |
| 19:45.11 | cr2_ | kiozen: to overlay the shapefile, you don#t need grass. |
| 19:45.11 | kiozen | the one to overlay in qgis :) |
| 19:45.16 | cr2_ | ok. |
| 19:45.17 | pancake | cr2_: does the partitioning is not published in /proc/mtd ? |
| 19:45.42 | cr2_ | pancake: there is no "dos-style" partition on the NAND. |
| 19:46.15 | cr2_ | kiozen: looking for it, don't want to reexport it back from grass. |
| 19:46.31 | kiozen | ok |
| 19:46.34 | pancake | but does exist a wce-supported partitioning |
| 19:46.48 | pancake | but linux does not handles it yet |
| 19:46.51 | pancake | right? |
| 19:47.28 | cr2_ | pancake: there is no partitioning at all. |
| 19:47.50 | cr2_ | just some blocks are used by one software, the others by the other. |
| 19:48.02 | pancake | ok i see |
| 19:48.05 | cr2_ | we need to put these borders into linux driver. |
| 19:48.11 | regulate | hi |
| 19:49.43 | cr2_ | kiozen: qgis is great. 'save as shapefile' |
| 19:50.10 | kiozen | yes it looks pretty neat to me |
| 19:50.18 | kiozen | once you compiled it ;) |
| 19:50.58 | cr2_ | lol. found a bug. the projection is saved as longlat. |
| 19:51.03 | cr2_ | checking again. |
| 19:52.20 | cr2_ | hm. yes, it's in longlat. maybe a feature, not a bug. |
| 19:52.59 | cr2_ | does not matter actually. |
| 19:53.05 | cr2_ | as long it is correct. |
| 19:55.51 | cr2_ | kiozen: sending. |
| 19:56.40 | kiozen | you took the gmx one? |
| 19:56.48 | cr2_ | hmm. 51MB ;) |
| 19:56.59 | cr2_ | will be rejected. |
| 19:57.16 | kiozen | ah ok that will take some time :) |
| 19:57.21 | cr2_ | i'll cut out the munich area. it's all europe with coastlines now. |
| 19:57.26 | cr2_ | full osm dump. |
| 19:57.40 | kiozen | pooooh |
| 19:58.00 | cr2_ | ogr2ogr |
| 19:58.27 | cr2_ | ogr2ogr -spat |
| 19:58.33 | cr2_ | min/max ? |
| 19:59.14 | cr2_ | 10-12, 47-49 |
| 20:00.01 | kiozen | that's still quite much |
| 20:00.07 | cr2_ | i'll check. |
| 20:00.36 | cr2_ | 6MB uncompressed. |
| 20:00.52 | kiozen | ok that should fit |
| 20:01.20 | cr2_ | 2.8MB |
| 20:01.23 | cr2_ | let's try. |
| 20:01.33 | kiozen | that must fit |
| 20:01.47 | kiozen | else I will cancel my pro account ;) |
| 20:01.55 | cr2_ | ok. |
| 20:02.24 | *** join/#htc-linux rob__w (n=bob@BACbf1f.bac.pppool.de) |
| 20:03.46 | cr2_ | status - sent. |
| 20:04.09 | kiozen | got it |
| 20:04.13 | cr2_ | ok. |
| 20:04.51 | cr2_ | unpack, and 'qgis kiozen1.shp' |
| 20:05.01 | cr2_ | hmm. |
| 20:05.15 | cr2_ | mercator is a user projection for some reason. |
| 20:05.33 | cr2_ | you will need to put the definition in the respective window. |
| 20:05.34 | kiozen | the railway doesn't look too bad |
| 20:06.01 | cr2_ | which one ? |
| 20:07.13 | kiozen | the black line right thru the center of Türkenfeld |
| 20:08.47 | cr2_ | ah, it's the railroad. |
| 20:09.15 | kiozen | qlandkarte makes it more obvious :) |
| 20:09.22 | cr2_ | but the street at the right side was not good. |
| 20:09.34 | cr2_ | ok, i#ve ignored the OSM tags. |
| 20:09.43 | kiozen | well if you code maps with 1 per mil |
| 20:09.43 | cr2_ | just imported the geometry. |
| 20:10.09 | kiozen | of alcohol of cource :) |
| 20:10.30 | cr2_ | it must have been some generalized garmin tracks :) |
| 20:10.57 | kiozen | like Topo USA, 22 bits of resolution |
| 20:11.16 | cr2_ | ;) |
| 20:11.17 | kiozen | you don't nee more for chessboard roads |
| 20:11.28 | cr2_ | true. |
| 20:11.51 | kiozen | and it does not load with qlandkarte :( |
| 20:11.59 | kiozen | that's the main problem |
| 20:12.17 | kiozen | but it will not with sendmap either :) |
| 20:12.27 | cr2_ | :) |
| 20:12.38 | kiozen | anyway looks like this conversion works |
| 20:12.56 | kiozen | I will cross check once I get this topo 10 |
| 20:13.08 | cr2_ | ok. |
| 20:13.29 | kiozen | but we have to do this easier for the casual user ;) |
| 20:14.50 | kiozen | and I have to save the log to remember |
| 20:14.50 | cr2_ | :) |
| 20:14.51 | kiozen | next I need a spec for that geo tiff |
| 20:14.51 | cr2_ | the casual user may put the GCPs |
| 20:14.53 | cr2_ | btw, get a trial version of ozi explorer. |
| 20:15.02 | kiozen | and I get scared by the size of the tile |
| 20:15.13 | kiozen | what for? |
| 20:15.26 | cr2_ | it's a defacto standard for raster nav. |
| 20:15.37 | cr2_ | don't be scared by the size. |
| 20:15.38 | pancake | cr2_: http://wiki.xda-developers.com/index.php?pagename=HermesMemoryMap |
| 20:15.45 | cr2_ | it's uncompressed. |
| 20:15.46 | pancake | NAND Flash partitioning |
| 20:15.52 | pancake | here is the flash partition table |
| 20:16.01 | kiozen | ok |
| 20:16.07 | cr2_ | pancake: hmm. i know this table, but... |
| 20:16.10 | pancake | what does the 'fake address' mean? |
| 20:16.21 | cr2_ | kiozen: i'll paste the "real" string you should use. |
| 20:17.01 | cr2_ | gdalwarp -of gtiff -ot byte -co tiled=yes -co blockxsize=128 -co blockysize=128 -co compress=deflate -co predictor=1 XXX.gtif YYY.gtif |
| 20:17.11 | cr2_ | for 128x128 tiles. |
| 20:17.23 | cr2_ | compare the sizes of XXX and YYY |
| 20:17.56 | cr2_ | i didn't check if you have 8bit palette in the original file. |
| 20:18.05 | cr2_ | imho, it's a must. |
| 20:18.14 | pancake | looks like substracting 0x50000000 from the fake address we get the 'real' address? |
| 20:18.32 | cr2_ | pancake: i've written 'fake', because these are internal SPL numbers. |
| 20:18.50 | cr2_ | how do we check it ? |
| 20:18.52 | pancake | ok, i'll later get a full dump of the nand |
| 20:18.57 | pancake | and will look for it |
| 20:19.03 | pancake | now is time to eat something :) |
| 20:19.08 | cr2_ | can you use 'pdocread' or something like that on the full NAND ? |
| 20:19.12 | cr2_ | ok. |
| 20:19.18 | pancake | yup i'll |
| 20:19.21 | pancake | thanks |
| 20:19.57 | kiozen | ok I stop hyperventilating |
| 20:21.32 | kiozen | is there a linux version of ozi explorer? |
| 20:21.54 | cr2_ | kiozen: when you have this tiled image, you can use RasteIO() from libgdal to read it. |
| 20:22.03 | cr2_ | no, but it works in wine. |
| 20:22.21 | cr2_ | you'd be interested in the GUI design. |
| 20:22.33 | cr2_ | everything else is not very useful :) |
| 20:23.21 | kiozen | installing |
| 20:23.28 | cr2_ | hehe. and try to georeference your image. psokolovsky thinks it's easy on ozi ;) |
| 20:24.16 | cr2_ | there is some hack you need to do. it's somewhere on the ozi homepage. |
| 20:25.03 | psokolovsky_ | cr2_: no hacks, it's all trivial |
| 20:25.32 | cr2_ | http://www.oziexplorer3.com/support/oziexplorer/other/linux.html |
| 20:25.35 | psokolovsky_ | cr2_: and you guys appear to be wanting to outsly yourselves with all those "parabolic tiles" ;-E |
| 20:26.10 | psokolovsky_ | Map is flat and rectangular. Take a ruler and prove, if you didn't noticed that before ;-E |
| 20:26.23 | cr2_ | psokolovsky_: it's .5m topo maps we are talking about. with single trees, as kiozen said. |
| 20:26.39 | cr2_ | lol. |
| 20:27.10 | kiozen | single very likely non existing trees, to be precise |
| 20:27.53 | kiozen | so what's about the GUI? It has the same charme like the one from MapSource |
| 20:28.18 | *** join/#htc-linux law|ibook (n=_law_@213.173.86.202) |
| 20:29.55 | cr2_ | kiozen: it's very popular. like microsoft office ;) |
| 20:30.17 | cr2_ | in certain circles. |
| 20:30.39 | kiozen | is it a good product? I don't know, but we sell quite a lot and people don't complain.. |
| 20:31.16 | kiozen | I like mine more, even if I am pretty alone with that :) |
| 20:31.59 | cr2_ | it's good for a non-professional. |
| 20:32.16 | cr2_ | and runs even on wince2.0 |
| 20:32.57 | kiozen | well agreed the splitted tool buttons have something |
| 20:33.40 | kiozen | at least very efficient with GUI real estate |
| 20:34.57 | kiozen | does it load geotif, too? |
| 20:35.49 | kiozen | wow it has a quit button! |
| 20:37.41 | cr2_ | i think it should. |
| 20:38.14 | cr2_ | but not quite sure about tiled+deflate. |
| 20:38.39 | kiozen | hm not in the test version, only bmp |
| 20:38.41 | cr2_ | they want to force you to use ozf2. |
| 20:38.46 | cr2_ | ok. |
| 20:40.15 | kiozen | despite the fact that ozf2 is proprietary |
| 20:40.31 | kiozen | does it have an advantage compared to geotiff? |
| 20:40.47 | kiozen | the 64x64tiles? |
| 20:52.59 | cr2_ | you can do 64x64 in geotiff |
| 20:53.29 | cr2_ | the advantage is the money you get from the desktop version sold ;) |
| 20:53.46 | kiozen | but that's my money - not good |
| 20:53.52 | cr2_ | :) |
| 20:54.04 | cr2_ | there is absolutely no advantage. |
| 20:54.13 | cr2_ | you can do all the same with geotiff |
| 20:54.32 | kiozen | that's what I wanted to hear |
| 20:54.45 | cr2_ | well, you don't need libgdal to read ozf2. |
| 20:54.48 | cr2_ | only zlib |
| 20:55.07 | kiozen | now we just need a conversion tool, |
| 20:55.17 | kiozen | a GUI like QLandkarte |
| 20:55.21 | cr2_ | ozf2tiff ? |
| 20:55.26 | cr2_ | it exists. |
| 20:55.33 | kiozen | no any map to gtiff |
| 20:55.53 | cr2_ | hmm. it's the question of GCPs. |
| 20:56.07 | kiozen | GCPs ? |
| 20:56.22 | cr2_ | there is a georeferencer plugin in qgis. |
| 20:56.28 | cr2_ | GroundCalibrationPoints |
| 20:56.35 | kiozen | ah |
| 20:56.38 | cr2_ | but i don#t like it. |
| 20:57.40 | kiozen | shouldn't that be easy to substitute for maps like the topos |
| 20:57.55 | cr2_ | i've written a small patch, so it works like in GRASS. |
| 20:58.11 | cr2_ | you open a canvas with a map, click a point |
| 20:58.31 | cr2_ | and then click the same point on a referenced vector or raster map. |
| 20:58.44 | cr2_ | but the GUI may be better done. |
| 20:59.17 | kiozen | for the topo maps we got this info in the tfw, right? |
| 20:59.20 | cr2_ | something like that: |
| 20:59.22 | cr2_ | http://www.pcigeomatics.com/services/support_center/tech_papers/rpc_pci_cert.pdf |
| 21:00.58 | cr2_ | no, that's not the right link. |
| 21:00.59 | cr2_ | here: |
| 21:00.59 | cr2_ | http://www.pcigeomatics.com/services/support_center/tutorials/pdf/cartosat_tutorial.pdf |
| 21:01.43 | kiozen | you need that if you scan paper maps, right? |
| 21:02.07 | cr2_ | yes. |
| 21:02.45 | kiozen | ok, for the first run I would start with that Topo10 stuff. |
| 21:03.02 | kiozen | no need to slove all problems at once |
| 21:03.06 | cr2_ | ok. |
| 21:03.08 | cr2_ | :) |
| 21:03.47 | kiozen | and finally we would need a render unit fitting into 2MB together with Linux |
| 21:03.48 | cr2_ | all the dirty work is done by gdal anyway. |
| 21:04.16 | cr2_ | yes, a clever tiled vector lib |
| 21:04.23 | cr2_ | like RasterIO in gdal. |
| 21:04.38 | cr2_ | just VectorIO :) |
| 21:05.13 | kiozen | you really like to go for that Garmin maps, don't you :) |
| 21:06.05 | kiozen | but you are right for routing you have to |
| 21:06.56 | cr2_ | i'd prefer to adapt the roadmap *.rdm format ot match *img in its capabilities. |
| 21:07.13 | cr2_ | it already has routing and turn restrictions. |
| 21:07.31 | *** join/#htc-linux Foxdie` (n=foxdiefu@87-194-16-145.bethere.co.uk) |
| 21:07.42 | cr2_ | but misses POIs and cleverly designed polygon renderer. |
| 21:07.45 | kiozen | which one you mean |
| 21:07.53 | cr2_ | rdm. |
| 21:08.32 | cr2_ | http://roadmap.digitalomaha.net/manual.html |
| 21:09.16 | cr2_ | the docs are a bit outdated. |
| 21:09.44 | cr2_ | 12 MAP FORMAT |
| 21:10.08 | kiozen | little bit sparse of information |
| 21:10.14 | kiozen | guess there is code, too |
| 21:10.25 | cr2_ | of course. |
| 21:10.36 | cr2_ | there is a qtopia4 port too. |
| 21:10.42 | cr2_ | thanks to tsdogs |
| 21:11.17 | cr2_ | works on qt3/qt4/qtopia/wince/win32/gtk2 |
| 21:14.27 | kiozen | are there many maps in rdm format? |
| 21:20.56 | cr2_ | in the us it's tiger. |
| 21:21.17 | cr2_ | and i've mostly done a mp2rdm converter. |
| 21:21.36 | *** join/#htc-linux hlbot749 (n=adm@iclem.net) |
| 21:21.51 | cr2_ | need to properly escape the sql. |
| 21:22.22 | kiozen | ok |
| 21:22.23 | cr2_ | it's bash+awk+psql+postgis+buildmap |
| 21:22.34 | kiozen | uhh |
| 21:22.46 | cr2_ | but mp has more features as you can guess. |
| 21:22.59 | cr2_ | POIs and sane polygon implementation. |
| 21:23.15 | cr2_ | yes, you need postgres+postgis installed. |
| 21:23.25 | cr2_ | it's possible to use shapefiles, |
| 21:23.30 | cr2_ | but i prefere postgis. |
| 21:23.38 | cr2_ | it#s much more flexible. |
| 21:23.51 | cr2_ | shapefiles+dbf(or CSV) tables |
| 21:24.23 | cr2_ | and sql is ascii, unlike dbf and shapefiles. |
| 21:24.42 | cr2_ | well, text, no necessarily ascii. |
| 21:25.06 | cr2_ | utf8 . |
| 21:26.53 | cr2_ | there is something common between mp2rdm and mkgmap :) |
| 21:27.09 | kiozen | what? |
| 21:27.19 | kiozen | java? |
| 21:27.25 | cr2_ | the purpose :) |
| 21:27.32 | cr2_ | no java here. |
| 21:27.58 | cr2_ | img is a better format(right now) |
| 21:28.09 | cr2_ | but rdm supports routing. |
| 21:28.22 | cr2_ | which really works. |
| 21:28.43 | tsdogs | ho all |
| 21:29.00 | cr2_ | hi tsdogs |
| 21:29.06 | tsdogs | BabelO: goxboxlive sayd that suspend is working on opieII |
| 21:29.10 | kiozen | better means more efficient in memory, does it |
| 21:29.12 | tsdogs | hi cr2_ |
| 21:29.16 | cr2_ | do you have a good raster map of your area ? |
| 21:29.40 | cr2_ | read the logs, we have went the whole process through with kiozen. |
| 21:30.06 | tsdogs | I read them, for what I can understand :) |
| 21:30.26 | cr2_ | do you have a map ? |
| 21:30.28 | tsdogs | so geotiff is the format |
| 21:30.34 | cr2_ | yes. |
| 21:30.37 | kiozen | who |
| 21:30.40 | tsdogs | who you asking? |
| 21:30.41 | tsdogs | :) |
| 21:30.53 | cr2_ | kiozen: you already have a map :) |
| 21:31.08 | kiozen | not for my area, just ordered |
| 21:31.32 | cr2_ | tsdogs: it's always easier to use a local map which you know well. |
| 21:31.43 | tsdogs | :) yep |
| 21:31.59 | cr2_ | kiozen: at least something . |
| 21:32.20 | cr2_ | tsdogs: you can even grab it from the net. |
| 21:32.21 | kiozen | hope it arrives befor my vacation |
| 21:32.34 | kiozen | landesvermessungsamt is not amazon |
| 21:32.35 | cr2_ | then we can discuss how to convert it to geotiff. |
| 21:32.49 | cr2_ | kiozen: you never know. |
| 21:33.36 | cr2_ | tsdogs: i've searched for a map of verona, but did not find anything useful. |
| 21:33.45 | tsdogs | :( |
| 21:34.30 | cr2_ | you can grab many google screenshots, and then glue them together. |
| 21:34.36 | cr2_ | but it's a lot of work. |
| 21:34.43 | tsdogs | what format should I look for? geotiff tfw ? |
| 21:34.55 | cr2_ | png/gif/jpg |
| 21:35.00 | tsdogs | ok |
| 21:35.16 | cr2_ | tfw is a text file with 6 numbers. |
| 21:35.30 | cr2_ | not a map :) |
| 21:35.37 | tsdogs | google does not have the detail we need (at least free one) |
| 21:35.47 | cr2_ | ok. |
| 21:37.39 | cr2_ | tsdogs: i think the qpegps can be ported to qt4. at least the drawing canvas. |
| 21:37.58 | kiozen | ok, guys, it's time for me to go to bed, good night... |
| 21:38.02 | cr2_ | the gpsd and all that datum/fetch stuff can be dropped, |
| 21:38.06 | cr2_ | good night. |
| 21:38.07 | tsdogs | ok |
| 21:38.18 | tsdogs | good night. hmm too late |
| 21:38.28 | cr2_ | good night. |
| 21:39.28 | tsdogs | cr2_: was for kiozen :) |
| 21:39.58 | tsdogs | cr2_: can I use garmin maps? |
| 21:40.29 | cr2_ | hmm. |
| 21:40.50 | cr2_ | not in the binary form. |
| 21:41.12 | tsdogs | ok |
| 21:42.21 | cr2_ | we need to match rdm with img features. |
| 21:42.43 | cr2_ | kiozen's klowledge can be very useful here :) |
| 21:44.02 | tsdogs | I saw it. He defenatly nows much more than I do |
| 21:46.11 | cr2_ | ok. |
| 21:46.32 | cr2_ | so look for a map :) you will need it for testing. |
| 21:46.56 | cr2_ | it'd be nice to have it really big. like 10000x10000 pix |
| 21:47.29 | tsdogs | cr2_: I was wrong google, is detailed. need to find a way to easily get all the tiles. |
| 21:48.17 | cr2_ | the sat layer ? |
| 21:48.26 | tsdogs | nope |
| 21:48.29 | tsdogs | the street one |
| 21:48.30 | cr2_ | the map ? |
| 21:48.32 | cr2_ | ok. |
| 21:49.24 | tsdogs | should I need the sat map? |
| 21:49.29 | cr2_ | no. |
| 21:50.04 | cr2_ | but it depends on what you want and how good is the map layer. |
| 21:51.09 | tsdogs | streets are pretty detailed. |
| 21:51.43 | cr2_ | ok, then take a big monitor and use 'xwd' :) |
| 21:52.09 | tsdogs | :) |
| 21:52.59 | cr2_ | i guess you#ll have problems grabbing the tiles with wget. |
| 21:53.13 | tsdogs | at maximum detail I guess |
| 21:53.31 | tsdogs | can't qpegps do this for me? |
| 21:53.51 | cr2_ | it#s not much better than wget. |
| 21:54.00 | tsdogs | ok |
| 21:54.44 | cr2_ | http://sourceforge.net/forum/forum.php?forum_id=206322 |
| 21:55.47 | cr2_ | http://sourceforge.net/forum/forum.php?thread_id=1215966&forum_id=206322 |
| 21:57.18 | cr2_ | lol |
| 21:57.21 | cr2_ | http://sourceforge.net/forum/forum.php?thread_id=1372643&forum_id=206322 |
| 21:59.02 | tsdogs | it's commercial :( |
| 21:59.26 | cr2_ | ok. |
| 21:59.47 | cr2_ | what about the upper one ? |
| 22:00.16 | tsdogs | not responding :( |
| 22:02.00 | cr2_ | now i see where z7.png comes from |
| 22:02.04 | cr2_ | http://sourceforge.net/forum/forum.php?thread_id=1324463&forum_id=206322 |
| 22:03.27 | cr2_ | http://gtm.tel.uva.es/ztep/maps/dmap.htm |
| 22:04.54 | cr2_ | hehe. overlap 5% |
| 22:05.05 | cr2_ | stone age software :) |
| 22:05.40 | tsdogs | ok. so it ain't good |
| 22:06.24 | cr2_ | qpegps is not good. |
| 22:06.45 | cr2_ | but the grabber is ok. just too much overhead. |
| 22:07.22 | tsdogs | ok |
| 22:07.40 | cr2_ | http://homepages.phonecoop.coop/4oznuff/maps/qpegpsmaps.html |
| 22:11.46 | tsdogs | I downloaded some maps. (very small one) |
| 22:12.05 | tsdogs | but I now have 25 gifs. |
| 22:14.54 | cr2_ | have you written down what was written at the top ? |
| 22:15.12 | tsdogs | ok generating a new one with 324x324. |
| 22:15.31 | tsdogs | I'm using the wget one. |
| 22:15.40 | cr2_ | ok. |
| 22:15.52 | cr2_ | but the number are important. |
| 22:16.26 | tsdogs | but I now have 25 gifs. |
| 22:16.41 | tsdogs | :(sorry |
| 22:16.54 | tsdogs | the 324 is using javascript :( |
| 22:17.14 | tsdogs | ok I need the numbers above, and the map itself. what zoom should I do? |
| 22:17.33 | cr2_ | well, can you get them into gimp and stitch together manually ? |
| 22:18.02 | tsdogs | sure. |
| 22:18.05 | cr2_ | the numbers are per image. |
| 22:18.45 | cr2_ | ok. then you'll only need to defile 2 points with known coordinates (GCP) |
| 22:18.46 | tsdogs | ok, if I use the whole one? |
| 22:18.52 | cr2_ | yes. |
| 22:19.26 | cr2_ | so put 25 files together in 1 file. |
| 22:59.37 | cr2_ | ok, it's late. good night |
| 23:00.36 | tsdogs | good night, just finished the map |
| 23:00.38 | tsdogs | :) |
| 23:25.37 | BabelO | tsdogs: ok, so can send package to ljp |