Celestia Forums

Dont get your hopes up for a quick download Jack,unfortunately my FTP has upload to you capacity of about 8k/s so if your the only one on line is pretty slow,if there are a few of you it gets split evenly between you's.There is already 'Spec' and 'Normal' maps on the same site.I have them zipped up or in folders so you can maybe download some every day and make AltSurface for it.Jestr

Ah- right. I forgot that FTP site is pretty slow...

One of my friends actually owns a net cafe with a satellite DSL connection- so hopefully I can get him to download it all bit by bit. (He's been looking for a 64k earth for a long time- I told him there wasn't any so he'll be really impressed to see this!)

Sombrehombre,

I have a 32k VT spec map for earth too, see the www link below.

By the way: I agree with your remarks about colouring. Many corrected Blue Marble textures are too pale. The main problem with the Blue Marble texture is the "fluorescent" greens, not the yellows and reds of desert areas. Photographs by astronauts show that vegetation (forest) is very dark green. And the oceans should look blue, not purple or grey. Thanks for the link to arcscience!

Don. Edwards wrote:BTW,
I have yet another new 16k texture ready for release very shortly, along with 16k cloudmap if all goes well.

Don. Edwards

Ohhh! what's new this time Don

Darkmiss,
I have desided to plug-in the shadow-less continental data into the previous texture. This of course will make the use of a really good bumpmap or normalmap imperitive to bring out detial. So it will not be for everyone. I am also tinkering with color values again, but just a little.

Don. Edwards

Users with interest in accuracy -or in running Jupiter related scripts- may want to know that in order to match Celestia 1.3.1 central meridian alignment, this remarkable Io texture needs an horizontal shift of 183 pixels "to the right"

Bye

How do you do that ?

Cham wrote:How do you do that ?

If you have the GIMP installed, you can use the 'Offset' command under the 'Image->Transforms' menu.
You must select the option 'turn around' or something like this. Or you can use my GIMP script, that accepts input in pixels and degrees.

If you have Photoshop, there must be an offset command also. (but I don't know its name or location)

If you only have a basic image manipulation tool, perhaps you can do it manually. It gets more difficult as image size increases, though.

1) Select the right-most 183 pixels wide block (there should be a pixel counter on the lowest bar of the program window) and copy it to the clipboard.
2) Then select all the remaining image (invert selection command, if it exists) and move it 183 pixels to the right.
3) Finally, paste the clipboard's content and move it to the left until it reaches the image's begining.
4) Save the image with ANOTHER name. If you make a mistake (most probably, given the manual work) then you can revert to the original image, without having to redownload it.

Bye

Toti wrote:Users with interest in accuracy -or in running Jupiter related scripts- may want to know that in order to match Celestia 1.3.1 central meridian alignment, this remarkable Io texture needs an horizontal shift of 183 pixels "to the right"

If you have the GIMP installed, you can use the 'Offset' command under the 'Image->Transforms' menu. You must select the option 'turn around' or something like this. Or you can use my GIMP script, that accepts input in pixels and degrees.

If you have Photoshop, there must be an offset command also. (but I don't know its name or location)

Thanks for the info Toti. I noticed the shift but was too lazy to figure it out, thought it might have just been that twitch of mine.

Photoshop does indeed have an offset command. It's found under Filter->Other->Offset. Just enter the number of pixels to offset and set the "Undefined Areas" option to "Wrap Around".

Paint Shop Pro 8 (don't know about earlier versions) also has an offset command. Go to Effects->Image Effects->Offset and enter in the appropriate stuff, "wrap" should also be selected.

Dave

It looks like I did in fact leave a residual offset in my published texture. My base texture, to which everything else was aligned, was the uncompressed version of this:

Global mosaic of Voyager hi-res and med-res images of Io, resampled to 16 pixels per degree or 1.99 km/pixel, size 5760 x 2880 pixels. From Tony Rosanova at USGS Flagstaff. Longitude range 360 - 0 degrees, latitude range 90 - -90 degrees.
http://www.lowell.edu/users/ijw/maps/iomaphires.jpg

My understanding is that this is the most geometrically refined global mosaic of Io, and substantially geomerically better than the orange/violet Voyager color mosaics most textures are based on. I did my best to align the positions of everything else (the 3 Galileo low-res projovian color fragments and the high-res antijovian hemisphere) to this (this is NOT easy without more speciallized cartographic software), while aligning the colors of everything to that of the high-res true-color anti jovian hemisphere imaged by Galileo.

Checking 6 different features against Tony Rosanova's mosaic above, it appears that 0 longitude is between pixel (GIMP ordering) of 1871 and 1872 from the seam, whereas it should occur between pixels 2047 and 2048.

That indicates an offset to the right (East) by 176 pixels is actually correct. It would technically be better to do a pixel correct adjustment at my working resolution of 16 pixels/degree (5760 across), but unfortunately, that version has been lost to an overambitious disk-drive cleanup. I decided it wasn't quite good enough for an 8k Io.



I have made a correction to offset the Io textures to the East by 176 pixels, and am currently uploading that corrected versions as both .png and .dds to the website http://laika.012webpages.com/celestia. Thanks to Toti for identifying this error.



If you still think its 7 pixels (about half a degree) off, I disclaim copyright to this work, and you are free to do whatever gets you through the night with it.

sombrehombre wrote:I have made a correction to offset the Io textures to the East by 176 pixels, and am currently uploading that corrected versions as both .png and .dds to the website http://laika.012webpages.com/celestia. Thanks to Toti for identifying this error.

If you still think its 7 pixels (about half a degree) off, I disclaim copyright to this work, and you are free to do whatever gets you through the night with it.

Thanks for your speedy work in correcting the texture.

Sombrehombre:

There is no error at all.
The alignment problem is really easy to correct (and most people aren't interested in it, anyway). My above post was not intended to despise your (highly precise and methodical) work at all, but to give a hint on how to align the texture with respect to Celestia 1.3.1 standards.

The shift of 183 pixels -as I wrote in my post- is incorrect. There is too much difference when changing alt surfaces at low altitudes (I can't believe that I didn't check it this way).
I arrived at this value by comparing your 4K Io with an upscaled-to-4K version of the "Celestia Io" at the small "bridge/peninsula" in Loki Patera. Not the best method: with such a low resolution frame, a lot of measurement error is involved.

I tested your 176 pixel value and it is much more accurate.

Keep your excellent work
Bye

Toti wrote:Sombrehombre: There is no error at all. The alignment problem is really easy to correct (and most people aren't interested in it, anyway).

Hello: don't forget that even a LITTLE missalignement with the current version of Celestia may show curious informations when your turn the locations on, have a look on an AltSurface, use VT's, and so on
So, even with my poor computer i find this Io texture superb, SombreHombre, thanks. Jeam

Right you are Jeam...

As I found out recently while working with the SaturnTour script, if your locations are off as little as 5 degrees in longitude (Maybe 10 pixels on the image I was working on), you're location markers are going to identify the wrong feature on a surface. The Central Meridian alignment must be correct in order to properly identify the features on a planetary object.

Take care, Bob

sombrehombre wrote:That indicates an offset to the right (East) by 176 pixels is actually correct.......

That's what I also found.

Dave