This is a great idea. Storing for example into a ZIP file as uncompressed allows the extraction of the data by the MCU during reading with relatively simple access code. The ZIP file includes a header in front of each file image in the archive that is only a small number of bytes in length. This header includes the file image size, date, attributes, name and the CRC-32 for the image. Having this in the header means that there is no modification of the file image data at all. The information in the first header includes enough information to easily determine the offset within the archive to find the next image so the MCU could support multiple file images within one ZIP with ease.

The headers of such uncompressed ZIP would be also very easy for the MCU to create if it were to be used in a mode where it created file images for sending back to a PC environment. Even if one did not have documentation of the header field contents it is easy enough to decode by making a series of test ZIPs with various sized files and then looking at the archive with a hex viewer/editor utility.

Michael Karas