Hi George:
IMHO, reading through a thread before posting is a 'simple' pre-requisite to keeping posts that are "off-topic" from adding to thread size.
The OP has repeatedly stated that his existing Water and Blend Masks created in Photoshop CS3 do not align with his purchased aerial imagery.
He is now using GIMP 2.8 trying to achieve proper alignment and scaling to match image size and resolution of his custom, very detailed Mask images with his purchased aerial imagery.
I have previously suggested that Caleb use GIMP 2.8 to re-configure his Mask graphic file format into 8-Bit Grayscale TIFFs during final output for SDK Resample to reduce working data set size and BGL output size.
I have noted that both you and Roby have previously stated on more than one occasion, that such 8-Bit Grayscale TIFF Masks MUST use LZW Compression during final output by GIMP for compatibility with SDK Resample.
AFAIK that LZW Compression 'requirement' for 8-Bit Grayscale TIFF Masks was subsequently reported to have been due to a "bug" in the way that GIMP 2.6 output such a TIFF graphic file format (...that is now 'fixed' in GIMP 2.8).
IMHO, it is not longer considered a preferred practice to use the LZW Compression with 8-Bit Grayscale TIFF Masks during final output by ex: GIMP 2.8 or later for submission to SDK Resample, as it results in approximately a 2 to 3 times longer BGL compilation time.
However, IIUC, it may prove useful for Caleb to still use (non-lossy) LZW Compression with his 24-Bit RGB color purchased aerial imagery and his 24-Bit RGB color Masks when loaded concurrently in GIMP 2.8, as this may reduce his working data set size, while still allowing for anti-aliasing to be used with the GIMP 2.8 "Transform" tools, thus enabling him to maintain more of the details in his very detailed custom Mask images.
George, considering your past experience as a FS developer, and considering the likelihood that the OP is currently working with multiple 24-Bit RGB color raster images of approximately 1 GB each loaded concurrently in GIMP 2.8, would you actually not have something pertinent and helpful to offer to this thread which may assist the OP as a relative 'newcomer' to the processes under discussion here ?
GaryGB
IMHO, reading through a thread before posting is a 'simple' pre-requisite to keeping posts that are "off-topic" from adding to thread size.

The OP has repeatedly stated that his existing Water and Blend Masks created in Photoshop CS3 do not align with his purchased aerial imagery.
He is now using GIMP 2.8 trying to achieve proper alignment and scaling to match image size and resolution of his custom, very detailed Mask images with his purchased aerial imagery.
I have previously suggested that Caleb use GIMP 2.8 to re-configure his Mask graphic file format into 8-Bit Grayscale TIFFs during final output for SDK Resample to reduce working data set size and BGL output size.
I have noted that both you and Roby have previously stated on more than one occasion, that such 8-Bit Grayscale TIFF Masks MUST use LZW Compression during final output by GIMP for compatibility with SDK Resample.
AFAIK that LZW Compression 'requirement' for 8-Bit Grayscale TIFF Masks was subsequently reported to have been due to a "bug" in the way that GIMP 2.6 output such a TIFF graphic file format (...that is now 'fixed' in GIMP 2.8).

IMHO, it is not longer considered a preferred practice to use the LZW Compression with 8-Bit Grayscale TIFF Masks during final output by ex: GIMP 2.8 or later for submission to SDK Resample, as it results in approximately a 2 to 3 times longer BGL compilation time.
However, IIUC, it may prove useful for Caleb to still use (non-lossy) LZW Compression with his 24-Bit RGB color purchased aerial imagery and his 24-Bit RGB color Masks when loaded concurrently in GIMP 2.8, as this may reduce his working data set size, while still allowing for anti-aliasing to be used with the GIMP 2.8 "Transform" tools, thus enabling him to maintain more of the details in his very detailed custom Mask images.

George, considering your past experience as a FS developer, and considering the likelihood that the OP is currently working with multiple 24-Bit RGB color raster images of approximately 1 GB each loaded concurrently in GIMP 2.8, would you actually not have something pertinent and helpful to offer to this thread which may assist the OP as a relative 'newcomer' to the processes under discussion here ?

GaryGB
Last edited:


