That could be possible. There are a lot of similarities in specs between all Mobilon models, with key differences in certain specs like color vs grayscale, RAM/ROM specs. Even the original manual is a shared manual across several Mobilon models.
Though just to clarify, WinCE System Info is reporting MIPS R4000, but all written documentation that I'm finding is reporting MIPS R3000. So a part of me initially felt suspicious of online specs, where the source specs could have been wrong and then re-transmitted many times over across the web. I even saw one site saying that it used an ARM R3000, which is just really wrong.
However, my initial suspicion of online documentation has been fading because there also seems to be a lot of correlation in specs in old
(late 90s
) and new docs alike, where the various sources are all different enough that it's clear that their specs were NOT just copy/pasted from a single, potentially incorrect source.
When I open up the device, the IC is listed as a Sharp SC3912. While I can't find anything online about this chip model number itself, I am finding that the chip's form factor appears to match the form factor and pin count detailed within the PR31700 documentation
(final page
):
http://datasheets.chipdb.org/Philips/PR31700.pdf
Also, other than the text on my IC, it appears to be an identical, visual match to the PR31700 in this photo, even down to the three large die circles in the three square corners:
http://www.bugs.nl/velo/velomainboard.jpg
So at this point, I'm inclined to think that this chip is a white label flavor of the PR31700
(as Rich mentioned earlier
) and for some reason or another the processor type in the System Properties is just wrong. I have a hard time believing that Philips would make a hybrid PR31700 for Sharp that runs a MIPS R4000 instead of MIPS R3000.
The last thing I'm trying to determine is what Windows CE 2.0 uses as the source of truth for the System Properties data. Since Windows was originally created to be functional on a lot of different machines, I always assumed that this data somehow propagated up through the abstraction layer from the chip. Though considering that each CPU seemed to need a different version of Windows CE 2.0, it's possible that this is just a value set somewhere within the registry.
If anyone knows where this data comes from, I would appreciate your expertise on that.
Sorry for the insanely long response, but thanks for sticking with me!