I've created a couple of kml files showing where I think the eastward and westward routes might go. Yellow line in the image is Tesheiner's route map.

Nice map!

Why not go another km or so South to the end of the salt-flats, and then zig zag East on the path of Ridges?

note: I don't know how to set up my computer to edit kml files, so I'm just describing it in words. Sorry.

Techniques to advance:

• On Bedrock : Maybe it would be the best.
• At the bottom betwen crest : almost good since there is very thin layer of sand. Maybe it is better since it causes less vibration to rover.
• Crosing the crest: The worst, make more effort to wheels motors.

I may well add that route... once I finish filling the gaps in this!

I'm trying to make an extension of the Hirise coverage that is in sync with the existing frame, has enough resolution to see individual ripples, covers the terrain ahead only and is under 20MB in size. This is what I've done so far, should finish a few more bits on Monday/Tuesday and then I'll upload the kmz. It's annoying that I can't get perfect registration with both the existing Hirise image and the HRSC and MOLA data.

Wow, I can see Oppy high-tailing as she bounds through the ripples, dust flying up into the air as she 6-wheel drifts into that turn east toward the prize. Yee-haw!!!!

My apologies to those of you who can't stand a little unrealistic anthropomorphizing.

That certainly is unrealistic anthropomorphizing!
More realistic would be a pigeon-toed oldster shuffling along at one meter per minute.

I finished the map extension on Monday but have only now found a place to post it- it's located here and is split into three parts.

This isn't something I had originally intended to do. If I had planned this from the outset I would have used a more sophisticated method of making the tiles than just saving images from the IAS viewer! Initially all I wanted was to see just a little bit more terrain off to the west but I found that increasing the coverage was a bit like pulling at a loose thread and before I knew it I was all the way to Endeavour.

Anyway, I've decided to stop at this point. I may decide to do things properly in the future and start over again. First I will need to find some software to break the JP2s into full resolution 8192X8192 chunks that slightly overlap each other.

I have some scripts to do this, let me know if you decide to do it and I can probably make the chunks for you and throw them on my webspace.

James

Just thinking out loud but it would be great to have this map coverage implemented using nested regions (regions + LOD). It's a LOT of work and I presume the only (sane) way to do it is with some automated software.

Cheers for that, that would be great! First image to do should probably be the map-projected ESP_011765_1780. Each tile should have something like a 5 % overlap with all the neighbouring tiles. I won't be able to start right now though as I have some issues to fix with my PC.


I'm also wondering whether it would be worth completely re-doing the pre-existing HiRISE frame as it just does not seem to match up well with the HRSC and MOLA data. That's why the topography of the hills at the edge of Endeavour looks a bit odd in my map extension- the overlay is too far to the NE.

I'm now thinking that it might not be that difficult. Sure, there will be 16 times as many pixels to map but the real issue is how many tiles there will be. As each tile (at 8192X8192) will have about 37 times as many pixels as my original tiles it might be 2.3 times easier!

I'll see what I can do - babies take up a lot of time so no promises!

What format do you want the output? jpg, png? presumably 8bit?

PM sent to SFJCody with the tiles. If anyone else has a use for full resolution 8192x8192 JPG chunks of ESP_011765_1780 let me know.

While doing that I also ran my ripple mapping tool on that image.



I await to see what the 'possible' route is - maybe folloing the bedrock out crops something like this...



(Both images link to same blog post with larger versions)

James

I'm looking into various ways of doing nested regions. Also I'm wondering if I should abandon trying to match the existing HiRISE frame and switch to matching introduced HiRISE images to either the CTX image or the HRSC image. Either of these two look like they will provide a better fit to the topography data around Endeavour than extending the positions of the current embeded image, which seems out of alignment. Of course, the downside is that the entire route map will have to be redrawn and the DTM of Victoria created from rover data will be out of sync!

Actually, you can keep both options.
The BIG amount of data are the JPEGs and they would be the same on both a) HiRISE to HiRISE match and HiRISE to CTX match. The "only" change would be on the control information in the KML files where the position of the pictures would have to be offset in options a) and .
I still prefer option a) for obvious reasons.

Good point, good point. I guess I could make the first mosaic fit the existing HiRISE and then simply batch modify all the kmls to make a version that provides a better alignment around Endeavour.

Version 2 preview, not at full resolution. I'm trying various pieces of overlay software to find out which is the best at automating the process of making nested regions. Will upload later this week.

Well, I've finally decided to upload version 2.0! What with Opportunity about to rove off the current HiRISE frame there aren't really any excuses left to be made. I kind of abandoned this in July because I wasn't happy with how it was running but hopefully there will be some kml experts out there who can suggest ways to reduce the time it spends loading images (which will then feed into 2.* and so on).
It's an unholy chimera of a thing created using a combination of methods because the super-overlay tool I was using had a bizarre problem when I tried to create region files within a particular latitude band on Mars. Anyway, the final product more-or-less sort-of works. Just don't try to fly horizontally at high speed over the landscape because it might crash Google Earth.

It's available as a torrent here:

Torrent link

If you don't have a BitTorrent client to download this with I recommend using µTorrent.

Google Earth recommendations: If you have not done so and your computer is capable of handling it it would be a good idea to increase Google Earth's memory cache size to 500Mb (go to tools -> options -> cache).
It's probably best to remove version 1.0 first, otherwise it might prove a bit taxing.


To run, unzip the whole folder and open the file "MER-B map extension version 2.0.kml" located within.

Thanks again to James Canvin for splitting the HiRISE frame.

Some screenshots (with Tesheiner's route map):

Thanks for the torrent - it's coming down now, and I'll be seeding it all night if anyone wants to try and grab it - for about the next 14 hrs.

Thanks too!
It takes a bit of time loading on GE but after that, it is smooth navigation and will be very useful in this part of the trek.

Glad it's working. The bit Oppy is currently near is OK though, it's the area between lat -2.14 and lat -2.20 that has the really long loading times. That's because I had to fudge it and use a different method. Why the superoverlay software fails to work properly in that latitude range I've no idea.

How did you get it BTW? I've seen four downloads so far and each one was in a different country: Denmark, India, the USA and the UK.

Well, we are using torrent aren't we?
My BitTorrent client found two seeds and was continuously connected and downloading from only one of them ... from Denmark. It might explain.