From: <ki...@ba...> - 2005-01-21 04:41:32
|
Quite recently, I have encountered two MR97310 cameras which use a slightly different communication protocol and which use a new (for the Mars cameras) compression format. Unlike the older Mars cameras, there does not seem to be any choice to run without compression. I have had some preliminary success in getting past the download of the configuration block, which works pretty much the same way on both the old-style cameras and on the newer ones, and I am able now to download some data after that. Observed differences between the old cameras and the new ones include: -- the older cameras downloaded the photos one at a time, and there was data in the configuration block, which indicated the size of the download for each. Thus it was possible to download the photos separately. -- the newer cameras seem to dump out in a continuous stream whatever there is in the camera. The way that one photo is delineated from the next is not clear. One would think that there is some kind of marker or separator between the successive photos, but it is not clear what that marker must be. Also it is not so simple as to say that the memory of the camera simply gets dumped, because more bytes are downloaded if there are lots of photos than are downloaded if there are only one or two photos on the camera. Also the data in the config block which used to give the data sizes of the respective photos seems to have shifted in meaning and/or function. -- the old cameras had a rather strange 3:1 fixed ratio compression mode, but one was not forced to use it. The newer cameras have a (or more than one?) compression mode(s) and it is not possible to avoid using it (them). And the compressed output does not seem always to be of the same size, though it is too early to be certain about that. Another thing which is perhaps unfortunate from the user's or purchaser's point of view is that the newer cameras seem to be using a USB Vendor:Product ID already used by the older ones, namely 0x093a:0x010f Anyone who wants to work on this new frustration of mine or thinks that he/she has something to offer is welcome. I would suspect that the worst problem would be to unscramble the data. I expect that sooner or later I will probably figure out where a photo begins and where one ends, though someone could well be quicker than I am about that, too. Currently known "new" Mars MR97310 cameras include the Argus DC-1620 (and 1610?) and an unknown camera from Sakar. I have been informed that the camera is the Sakar iConcepts camera, but after seeing one in the package and the specs written on it ("up to 152 photos") I have the suspicion that the iConcepts is probably an sq905; that is the exact capacity of many sq905 cameras running at 176x144 resolution, with 2:1 compression. Therefore I am not sure about the ID of the second Mars camera. In any case, whoever does not wish to spend his treasure in the service of the Linux revolution should not buy one of these, just now. But if you do buy one, then let's work on it. Theodore Kilgore |