01:00.37 | *** join/#htc-linux ndavey (n=neil@pc075650.workstation.griffith.edu.au) |
01:18.14 | *** join/#htc-linux _ecze (n=ecze@eczema.ecze.com) |
03:39.52 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfdb21.pool.einsundeins.de) |
03:45.50 | *** join/#htc-linux Foxdie` (n=foxdiefu@87-194-16-145.bethere.co.uk) |
04:02.51 | *** join/#htc-linux joschkop179 (n=joschi@AC9E34EE.ipt.aol.com) |
04:32.35 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
05:03.41 | *** join/#htc-linux joschkopf79 (n=joschi@AC9E34EE.ipt.aol.com) |
05:14.14 | *** join/#htc-linux goxboxlive (n=goxboxli@195.159.97.196) |
05:35.45 | *** join/#htc-linux kiozen (n=kiozen@rgnb-4db1dd51.pool.einsundeins.de) |
06:20.54 | *** join/#htc-linux kiozen (n=kiozen@p5492A283.dip0.t-ipconnect.de) |
06:23.34 | *** join/#htc-linux kiozen (n=kiozen@p5492A283.dip0.t-ipconnect.de) [NETSPLIT VICTIM] |
06:23.34 | *** join/#htc-linux hlbot (n=adm@iclem.net) [NETSPLIT VICTIM] |
06:23.34 | *** join/#htc-linux thinko (i=jdoe6alp@smaug.rackdragon.com) |
06:23.34 | *** join/#htc-linux SuN (n=SuN@82-170-225-106.ip.telfort.nl) |
06:25.50 | *** join/#htc-linux ltxda0 (n=ltxda@unaffiliated/ltxda) |
06:31.37 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfdb21.pool.einsundeins.de) |
06:31.37 | *** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz) |
07:14.26 | kiozen | cr2: patch applied |
07:29.44 | *** join/#htc-linux the_sys0p (i=the_sys0@gateway/tor/x-320ec47b60ad4a8a) |
07:50.50 | *** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz) |
08:37.10 | *** join/#htc-linux rob_w (n=bob@X04d9.x.pppool.de) |
08:38.30 | *** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
08:38.57 | *** join/#htc-linux Marex (n=marex@vasut.kolej.mff.cuni.cz) |
08:52.06 | BabelO | hi |
09:07.07 | *** join/#htc-linux patoche (n=patoche@77.197.28.145) |
09:08.43 | *** join/#htc-linux swc666 (n=no@24.143.67.18) |
09:25.17 | *** join/#htc-linux SmallR2002 (n=SmallR20@79-77-82-70.dynamic.dsl.as9105.com) |
09:31.16 | Marex | BabelO, hi |
09:31.21 | Marex | cr2, hi |
09:45.56 | *** join/#htc-linux swc-666 (n=no@24.143.67.18) |
09:47.14 | *** join/#htc-linux ndavey (n=neil@60-240-227-199-vic-pppoe.tpgi.com.au) |
10:34.56 | cr2 | kiozen: excellent program :) i've referenced a picture with 75 GCPs yesterday. just for fun. |
10:35.39 | kiozen | cr2: well say thank you to gdal. GT is just a humble frontend |
10:35.53 | kiozen | you ever tried a really distorted map |
10:35.58 | kiozen | ? |
10:36.19 | cr2 | it was not distorted |
10:37.02 | cr2 | i've "auto"created the GCPs with names and coordinates. with awk |
10:37.07 | *** part/#htc-linux swc-666 (n=no@24.143.67.18) |
10:37.15 | kiozen | maybe I try this weekend with a paper map scanned without closed and weighted lid |
10:37.23 | cr2 | and then just moved then into the right places. |
10:38.32 | kiozen | but good to know that gdal digests such an amount of gcps :) |
10:38.59 | cr2 | the biggest problem was to push the GCPs into the picture bounding box |
10:39.15 | cr2 | 75 is not much. |
10:39.47 | cr2 | it's 6x6 or 9x9 linear system after all |
10:40.15 | cr2 | TPS will need much more resources. |
10:40.48 | cr2 | another nice to have feature is "delete all gcps" |
10:41.15 | cr2 | does the incremental gcp loading work ? |
10:43.29 | kiozen | should work |
10:43.33 | cr2 | i've also thought about the direct ozi .map loader |
10:43.35 | cr2 | ok |
10:43.57 | kiozen | yes I have to implement strg+a and del |
10:44.53 | kiozen | with ozi I would prefer the same as with mph/mpr -> an external converte |
10:44.55 | cr2 | the addRef should not be void, but return refpoint * |
10:45.42 | cr2 | kiozen: ok. but then it's probably easier to add the gcps to .tiff directly. |
10:45.43 | kiozen | what for? |
10:46.13 | cr2 | btw, you may check if the gcps are already present in the .tiff, and draw them. |
10:46.31 | cr2 | in the .ozf2 -> .tiff converter. |
10:46.48 | kiozen | yes that's somewhere on the TODO stack |
10:47.00 | cr2 | ok. |
10:47.44 | kiozen | but for what would you need refpoint * addRef() ? |
10:47.54 | cr2 | if i'll understand ozi datum -> epsg conversion, then the gcps will be easy to do. |
10:48.10 | cr2 | it will simplify the code |
10:48.24 | cr2 | <PROTECTED> |
10:50.20 | cr2 | the gcp loader ignores negative pixel coordinates, but will happily accept those outside the bounding box. |
10:51.28 | cr2 | it's ok in general, but not very useful in other cases. |
10:52.16 | kiozen | bad regex at line 219 :) |
10:53.22 | cr2 | when you "autogenerate" the gcps, both cases may be interesting. |
10:53.44 | kiozen | svn up (untested) |
10:53.55 | cr2 | the transform itself does not care where are the gcps. |
10:54.02 | cr2 | ok. |
10:54.22 | cr2 | but with the names gcps you may want to position then inside the bbox :) |
10:54.29 | cr2 | s/names/named/ |
10:54.55 | cr2 | to match what you see in the bbox. |
10:56.05 | cr2 | 'Del' the highlighted gcp is also good. |
10:57.00 | cr2 | filename.*Latin1() should be probably filename.*Local8Bit() |
10:57.25 | cr2 | just a braindump from yesterday' night. |
10:58.12 | cr2 | yeah, the program is certainly the on its way to world domination (tm) :) |
11:26.23 | *** join/#htc-linux patoche (n=patoche@77.197.28.145) |
12:44.48 | *** join/#htc-linux swc666 (n=no@24.143.67.18) |
12:59.40 | *** join/#htc-linux dcordes (n=snoopdog@unaffiliated/dcordes) |
13:18.10 | dcordes | no BA today ;( |
13:58.49 | *** join/#htc-linux CVirus (n=GoD@196.205.192.125) |
13:59.08 | *** join/#htc-linux GPFerror (n=gpferror@76.187.41.132) |
14:55.25 | *** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz) |
14:58.02 | *** join/#htc-linux TimRiker (n=timr@68-27-123-91.area1.spcsdns.net) |
15:02.00 | *** join/#htc-linux bernt (n=bernt@dslb-084-061-233-204.pools.arcor-ip.net) |
15:11.13 | *** join/#htc-linux Marex_Karlin (n=marex@auryn.karlin.mff.cuni.cz) |
15:36.02 | kiozen | cr2: now you can even load gcps from gcp-ed tif |
15:36.55 | kiozen | BabelO: what chipset does you Core 2 Quad use? |
15:55.15 | cr2 | kiozen: ok, i'll check in the evening. btw, is it possible to get the source image dimensions in CCreateMapGeoTiff ? |
15:57.05 | kiozen | cr2: not yet, but I can fix that in 5 min :) |
16:00.50 | kiozen | cr2: CCreateMapGeoTiff ::sizeOfInputFile |
16:03.30 | *** join/#htc-linux CVirus (n=GoD@196.205.192.125) |
16:03.41 | *** part/#htc-linux swc666 (n=no@24.143.67.18) |
16:04.49 | *** join/#htc-linux tsdogs (n=tsdogs@net-84-18-144-39.c.bkom.it) |
17:04.23 | *** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be) |
17:10.18 | *** join/#htc-linux BabelO (n=Fabrice@unaffiliated/babelo) |
17:26.28 | *** join/#htc-linux the_sys0p (i=the_sys0@gateway/tor/x-de42ec8419b85a25) |
17:46.25 | *** join/#htc-linux CVirus (n=GoD@196.205.192.125) |
17:50.08 | *** join/#htc-linux LunohoD_ (n=alex@e180071120.adsl.alicedsl.de) |
18:00.56 | *** join/#htc-linux Marex (n=marex@vasut.kolej.mff.cuni.cz) |
18:07.08 | *** join/#htc-linux goxboxlive (n=goxboxli@208.84-48-176.nextgentel.com) |
18:21.45 | *** join/#htc-linux bpedman (n=chatzill@nat/novell/x-2b8697e7729fa802) |
18:22.11 | *** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfdb21.pool.einsundeins.de) |
18:51.30 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
18:54.42 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
19:01.39 | *** join/#htc-linux ltxda0 (n=ltxda@unaffiliated/ltxda) |
19:12.48 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
19:25.02 | *** join/#htc-linux ltxda00 (n=ltxda@c-98-196-125-205.hsd1.tx.comcast.net) |
19:25.57 | *** join/#htc-linux aLeSD (n=alex@162.Red-83-33-46.dynamicIP.rima-tde.net) |
19:26.42 | *** join/#htc-linux dcordes_ (n=snoopdog@unaffiliated/dcordes) |
19:30.04 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
19:31.37 | *** join/#htc-linux gers (n=jtrs@p54A2C490.dip.t-dialin.net) |
19:34.40 | gers | who is creating the kernels for the handhelds on handhelds.org? |
19:48.32 | *** join/#htc-linux diogene31 (n=rj@mur31-2-82-243-122-54.fbx.proxad.net) |
19:50.02 | dcordes_ | gers, you mean who in this channel has commit access? |
19:50.34 | dcordes_ | I think there are plenty of people here contributing to hh.org. at least BabelO and cr2 are two of them |
19:53.40 | tsdogs | gres: ph5, sokovolsky (is the maintainer/coordinator) afair |
19:53.56 | tsdogs | ph5 is a developer :) |
19:53.58 | *** join/#htc-linux dcordes (n=snoopdog@unaffiliated/dcordes) |
20:01.10 | gers | are they earning money, or is this all private work? |
20:01.46 | dcordes | I doubt the first but I guess only they can tell you |
20:01.47 | tsdogs | gers: all for fun |
20:02.02 | dcordes | gers, what are your htc machines? |
20:03.57 | cr2 | lol |
20:06.15 | gers | dcordes: I bought an universal 2 month ago |
20:07.36 | gers | I looked for a linux handheld with phone |
20:07.42 | cr2 | gers: we are losing money :) |
20:08.03 | gers | lol |
20:08.40 | cr2 | universal is a good choice, if you want to run linux. |
20:12.44 | gers | Well, i like it. I compared to FIC Neo. And i think they are quite comparable in Hardware |
20:13.28 | cr2 | comparable ? |
20:14.24 | gers | well processor is faster on universal i think |
20:14.41 | gers | and has usb 2.0 support |
20:14.51 | cr2 | moko1 has 128MB RAM |
20:15.02 | cr2 | no 2.0 on the universal |
20:15.07 | gers | you can add on uni too, i heard |
20:15.17 | cr2 | moko1 has gps. closed source though. |
20:15.34 | cr2 | everything else is better on the universal. |
20:15.35 | gers | 2.0 on uni, i agree |
20:16.20 | cr2 | wifi, umts, 2 cams, flashlight, SD(io) |
20:16.38 | cr2 | keyboard. |
20:17.20 | *** join/#htc-linux ltxda (n=ltxda@unaffiliated/ltxda) |
20:17.24 | gers | wifi is equal, but bluetooth 1.1 |
20:22.20 | dcordes | did somebody in here ever attempt to use any sort of video conversation on uni? |
20:25.49 | *** join/#htc-linux kiozen (n=kiozen@rgnb-4db1dd51.pool.einsundeins.de) |
20:29.16 | dcordes | good evening kiozen. how is gt going? |
20:29.55 | kiozen | dcordes: scanning a map with an A4 scanner sucks :) |
20:30.13 | kiozen | now I have to georef loads of A4 tiles |
20:30.14 | dcordes | yea I hear you need an a2 scanner |
20:30.32 | kiozen | A1 or A0 would even be better :) |
20:30.45 | kiozen | but A2 would be nice too |
20:30.51 | dcordes | if I was you I would make a donation request. maybe there is are fans of your project which have some spare ones |
20:31.00 | dcordes | -is |
20:31.02 | kiozen | lol |
20:31.29 | kiozen | don't think there is a "spare" a2 scanner on this world |
20:31.44 | dcordes | why not? |
20:31.50 | kiozen | those things are rare and expensive |
20:31.54 | dcordes | I see |
20:32.14 | kiozen | the only one I could locate in Regensburg is at the university |
20:33.47 | dcordes | would you be granted access to it? |
20:34.05 | kiozen | not officially i think |
20:34.19 | kiozen | but I will give it a try |
20:34.40 | kiozen | befor it catches dust ;) |
20:35.00 | dcordes | you could offer free geo coding lectures in change :) |
20:36.00 | kiozen | oh gosh, better not. I don't think I am really an expert on that area |
20:37.04 | kiozen | damit, you never find anything in Gimp |
20:39.02 | BabelO | kiozen: you buy the core2 quad too ? |
20:39.38 | kiozen | well, my pc at work gets slow |
20:39.44 | *** join/#htc-linux patoche (n=patoche@lns-bzn-44-82-249-192-208.adsl.proxad.net) |
20:39.53 | kiozen | thus I consider to buy new equipment |
20:40.22 | kiozen | but I would like to have a small caseing |
20:40.38 | BabelO | kiozen: ok, |
20:40.43 | kiozen | BabelO: do you know your chipset? |
20:41.19 | BabelO | kiozen: yes it is a Q6600 with fsb 1066Mhz, the only one, next generation will have FSB at 1333 Mhz |
20:41.54 | kiozen | BabelO: not the cpu, the north and south bridge |
20:41.59 | dcordes | hehe I have 1333 |
20:42.08 | dcordes | and dd3 support |
20:44.02 | BabelO | ah, it is a p5K mother board, with fsb 1333, wait i look version |
20:44.36 | dcordes | BabelO, I have p5kc |
20:46.30 | BabelO | kiozen: all is ICH9 Family -> 82801I and 82801IB and 82G33/G31/P35/P31 |
20:46.46 | BabelO | dcordes: don't know what p5kc have more ? wifi ? |
20:46.59 | dcordes | hm no idea |
20:47.05 | dcordes | no wifi |
20:47.05 | BabelO | dcordes: mine had rsata |
20:47.11 | dcordes | what's that? |
20:47.18 | BabelO | dcordes: raid on sata |
20:47.30 | dcordes | oh ok. I thought new rastafarian bus |
20:47.51 | BabelO | dcordes: so i think we have same |
20:48.02 | dcordes | you have dd3? |
20:48.19 | kiozen | thought about buying a shuttle barebone, but it's getting too hot -> loude |
20:48.50 | kiozen | now thinking about uATX board, but there are no barebones for uATX |
20:49.02 | dcordes | make a casing yourself |
20:49.03 | kiozen | and buying everything on your own sucks |
20:49.20 | kiozen | dcordes: don't have the time |
20:49.23 | BabelO | dcordes: no just ddr2 |
20:49.35 | dcordes | that's the difference: p5kc can take dd3 and 2 |
20:49.45 | BabelO | ok |
20:50.01 | dcordes | and 1333 mhz bus speed |
20:50.18 | dcordes | but I only have some mid-range ddr2 ram. but at least it is geil ram |
20:50.25 | BabelO | dcordes: ok bus memory speed is only1066, but i have only ddr2 800 :) |
20:50.43 | dcordes | I think also my geil is 1066 |
20:50.59 | BabelO | geil ? |
20:52.36 | dcordes | yea :) it's some ram label |
20:52.49 | BabelO | ok |
20:54.42 | dcordes | it's a bit odd because in german geil signifies horny http://geilusa.com/ |
21:08.01 | *** join/#htc-linux cr2 (n=konversa@crpl6.physik.uni-wuppertal.de) |
21:15.56 | cr2 | kiozen: running the latest svn GT version. |
21:16.20 | kiozen | me too |
21:16.39 | cr2 | kiozen: 2 notes. it shows the latest referenced image at startup. |
21:16.54 | cr2 | it's impossible to load gif and 8bit png now ? |
21:17.11 | kiozen | never tried :) |
21:17.20 | kiozen | but as lon as it is 8bit |
21:18.01 | kiozen | referencing A4 tiles sucks... |
21:18.13 | cr2 | the selection box says 'Raw *.tif' now. |
21:18.41 | cr2 | if they have the margins, it's even worse. |
21:19.20 | kiozen | the problem is that they need to overlap |
21:19.27 | kiozen | thus you have to cut 2 maps |
21:19.30 | *** part/#htc-linux patoche (n=patoche@lns-bzn-44-82-249-192-208.adsl.proxad.net) |
21:19.34 | kiozen | and do the job twice |
21:20.14 | cr2 | i've one more idea. mark some gcps as 'corner markers', so they are just defining the rectangular bbox, and not used as gcps. |
21:20.56 | cr2 | it's trivial to cut the outside margins with just an gdal_translate parameter. |
21:21.33 | kiozen | well we could do extra cut marks |
21:21.58 | kiozen | but I still think it's better do start with a propper processed image right from the beginning |
21:22.35 | cr2 | yes, if there is only junk in the margins. |
21:23.06 | cr2 | but if you have the coordinate grid, you lose all the numbers, and need to keep a gimp window opened at a side |
21:25.10 | kiozen | yes but that is happening for the center tiles anyway |
21:25.20 | kiozen | gues what I am doing right now |
21:26.04 | kiozen | this work reminds me at some stupid homework from school |
21:28.02 | cr2 | yeah. |
21:28.20 | cr2 | i've done rotate+rectangular selection |
21:29.01 | cr2 | before i've thought about select by path. |
21:29.25 | cr2 | cut off some parts of the map in the process, because of the scan area being nonrectangular. |
21:29.44 | cr2 | need to add this to a FAQ. |
21:29.54 | *** join/#htc-linux CVirus (n=GoD@196.205.192.125) |
21:31.08 | kiozen | cr2: http://www.qlandkarte.org/shot.png |
21:31.15 | kiozen | that's not working real nice |
21:33.02 | cr2 | same pixel size ? |
21:33.43 | kiozen | yes but you always cut a few pix away while scanning |
21:34.49 | kiozen | the best would be to scan the map as a whole |
21:34.52 | cr2 | i mean the same pixel size after referencing them. |
21:36.51 | kiozen | cr2: |
21:37.02 | kiozen | Pixel Size = (3.085777933651771,-3.085777933651771) |
21:37.02 | kiozen | Pixel Size = (3.084406479966723,-3.084406479966723) |
21:37.02 | kiozen | Pixel Size = (3.085552695796509,-3.085552695796509) |
21:37.02 | kiozen | Pixel Size = (3.083412881096579,-3.083412881096579) |
21:37.05 | kiozen | not really |
21:37.46 | cr2 | well, these are minor differences. 1mm :) |
21:38.42 | *** join/#htc-linux BabelO (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
21:38.47 | kiozen | it's impossible to scan two sheets exactly on the pixel |
21:39.09 | kiozen | thus you have to cut a few pixel. leading to whit estripes |
21:39.24 | cr2 | don't they overlap ? |
21:40.05 | cr2 | ah. you cut the big map. |
21:40.05 | kiozen | they can't if you cut 1 map |
21:40.12 | cr2 | ok :) |
21:40.57 | kiozen | I think I will spare the 2nd map for a better scanner |
21:41.00 | cr2 | get an A3 300dpi scanner, like i've done. |
21:41.14 | kiozen | a2 or a1 would be better |
21:41.24 | cr2 | too much €€€ |
21:41.52 | kiozen | not to buy, better to do this with a scan service |
21:42.01 | kiozen | the problem is just to find one |
21:42.14 | cr2 | kiozen: btw, filename.*Utf8() will probably break on mac/win |
21:42.39 | cr2 | ok. i've never heard of such service, actually. |
21:42.53 | cr2 | A4 and A3 is never a problem. |
21:43.26 | cr2 | kiozen: qgis uses filename.*Local8Bit() |
21:44.12 | kiozen | ok |
21:56.23 | BabelO | lol, i just understand why last day pH5 says : installed scratchbox (urgh) |
21:59.27 | cr2 | BabelO: too simple for you ? |
22:00.18 | BabelO | cr2: yes a bit crazy even with the automatics install script |
22:05.22 | *** join/#htc-linux BabelO (n=Fabrice@unaffiliated/babelo) |
22:31.06 | *** join/#htc-linux CVirus (n=GoD@196.205.192.125) |
22:32.15 | *** part/#htc-linux CVirus (n=GoD@196.205.192.125) |
22:36.52 | *** join/#htc-linux BabelO (n=Fabrice@unaffiliated/babelo) |
23:17.37 | BabelO | cr2: around ? |
23:18.10 | BabelO | cr2: any quick idea about cp parameters to keep symlink and others strange file ? |
23:24.17 | cr2 | BabelO: cp -a ? |
23:26.18 | BabelO | cr2: i just try with cp -R -p |
23:26.34 | BabelO | ok your is better ;) |