Parts of texture are being built at a lower resolution?

This is new and rather peculiar.

Parts of the texture are coming out mosaic-ed

*side note: forgive the massive amount of threads, I am working on several maps for my game and I was doing a big push at the time of the update so I have many maps that have issues. I will be posting source files tomorrow after my workout.

http://www.smoth.net/files/worldmachine/GRTS_Northern mountains_004_source.7z

Sorry but my net is a bit slow tonight so I cant download the file…but!

Is that all done with noise generator like Advanced perlin? or did you use a DEM? Sometimes I have found with DEM’s that you have to blur them a little so that the square artifacts don’t get carried through the rest of the device chain.
That’s what it looks like to me:/

Not sure what a DEM is.

I am not using a noise generator, it is colorizer combined with a texture import. Take a look in basic coverage.

the left and right side of the image are using the same technique. something I spent months getting down and correct last year. It has been broken by something in this version of wm2. :\

In your northernmountains_pre.tmd file, erosion is being applied based on a mask; I believe what you are seeing is simply the pixelation of the base.png file being expanded to higher resolution.

To demonstrate this, I opened the above file and applied a blur filter between the file input and the erosion device. I believe you will see the mosaic disappear in the end product.

A DEM is a “Digital Elevation Model”…which is what your png file is…and even if its a very high res image mostly they will have small square artifacts like you have…what Remnant said is basically what I thought and should be the answer. If the image is a png you can try to resize the image in photoshop by using bicubic interpolation to get rid of the squares…but a little bit of Gaussian blur in worldmachine should also work. Basically you have to get rid of the square artifact’s before you start to add nice effects like erosion or it will get carried down the device chain…sometimes though those squares add for some interesting interaction between the erosion.

That is really interesting. Thanks for the info!