Qwerty.55 wrote:
Yes, I do have some. However, the analysis is slightly tricky because Casio wrote some security checks into the files that make them difficult to reproduce. We're still identifying those.
Fascinating, I'd be very interested to hear more about that for the sake of educational curiosity. Do you think they encrypted the full contents with a key of some kind, or simply have an obfuscated format with a bunch of checksums and key sequences?
We're not sure how they encrypted it(or even if they used normal formatting characters for the Prizm) because they've released no information about it other than than a few image and animation files. There are a few signs in the image header of the still pics that suggest a few methods, though. One thing we do know is that a few different parts of the file interact with one another and that's one way by which they check the validity of the data.
Qwerty.55 wrote:
We're not sure how they encrypted it(or even if they used normal formatting characters for the Prizm) because they've released no information about it other than than a few image and animation files. There are a few signs in the image header of the still pics that suggest a few methods, though. One thing we do know is that a few different parts of the file interact with one another and that's one way by which they check the validity of the data.
Very cool, keep us posted on your progress. It sounds like a fun demo exploration into the magic of reverse-engineering file formats.
I wouldn't exactly describe it as fun, but Casio made it rather easy once I figured out how they made it. Anyway, the .g3p format header has pretty much been cracked. I just need a Prizm to test a few hypotheses and I can start working on the picture data.
Qwerty.55 wrote:
I wouldn't exactly describe it as fun, but Casio made it rather easy once I figured out how they made it. Anyway, the .g3p format header has pretty much been cracked. I just need a Prizm to test a few hypotheses and I can start working on the picture data.
That's great to hear; congrats! Keep up the good work.
For some reasons, it seems like the Casio Prizm is out already. The release timeframe of January 2011 is gone from their website and the calculator appears to be available on their online store at
http://www.shopcasio.com/product/4063
At first I thought many Ebay sellers started selling it before the release date, but now I know why that many started selling them. Snake_X caught my attention because he mentions that a local store where he lives already sells them.
I guess that maybe they wanted to release it early to get a few additional sales for Christmas, but I hope it's not rushed either, because like TI, they're not known for bug-free OSes.
There's a great store very close to me, J&R, that has it available now for in-store pickup. I'm seriously considering grabbing it; I'm hoping to hear some positive proof of a C compiler for the device to make it worthwhile, though.
I personally would wait until the SDK has been released before buying it, but I'm probably not going to get the Prizm anyway
.
I personally like to play around with an SDK before I buy a product. Helps me see what it can do and sometimes even includes an emulator so I can test it out the functionality.
ParkerR wrote:
I personally like to play around with an SDK before I buy a product. Helps me see what it can do and sometimes even includes an emulator so I can test it out the functionality.
That's a good idea; you're more intelligent than most impulse-buying consumers, then. Souvik, indeed, I'm thinking the same; are you probably not going to get one for monetary reasons or lack of interest reasons?
KermMartian wrote:
ParkerR wrote:
I personally like to play around with an SDK before I buy a product. Helps me see what it can do and sometimes even includes an emulator so I can test it out the functionality.
That's a good idea; you're more intelligent than most impulse-buying consumers, then. Souvik, indeed, I'm thinking the same; are you probably not going to get one for monetary reasons or lack of interest reasons?
Monetary reasons.
I received this email in response to a question I sent Casio about the Prizm:
Casio wrote:
Mr. Mitchell,
Thank you for contacting Casio America. We do not currently have an SDK
available for PRIZM. At this time, I do not know when or if an SDK will
be made available in the future. SDKs for some of our other calculator
models can be obtained through the Casio worldwide educational site at
http://edu.casio.com, so I recommend that you continue to check that
site periodically to see if an SDK for this calculator becomes
available.
We hope that we were able to assist you with your question, and again
thank you for contacting Casio America.
Sal Manfredonia
CASIO AMERICA, INC.
570 Mt. Pleasant Avenue Dover, NJ 07801
973-361-5400 ext: 4339 Fax: 973-537-8972
e-mail:
smanfredonia@casio.com
www.casio.com
I do not know how good the prizm could really be. I bought the best casio on the market and I was not happy with the programming capability. It came with 24,000 bytes of free memory to play with, but, I used around 4,500 for one picture, and it was very slow. Not only that, I did not like the way the programming menus where set up. It was definitley cheaper for a reason ("Casio Basic Sucked"). The PRIZM, without a doubt would have to be a lot more advanced, but can it be possible that they took that big of a leap forward, or is it more cheap bull.
Basicman, and that's why you're a TI calculator coder now?
The jury is still out on the Prizm as far as I'm concerned; until they can verify the existence of and a release date for the SDK, I don't think I'll be getting one.
ParkerR wrote:
I personally like to play around with an SDK before I buy a product. Helps me see what it can do and sometimes even includes an emulator so I can test it out the functionality.
I've checked some of Casio's other SDK's and they make you use the serial code to download them.
Qwerty.55 wrote:
ParkerR wrote:
I personally like to play around with an SDK before I buy a product. Helps me see what it can do and sometimes even includes an emulator so I can test it out the functionality.
I've checked some of Casio's other SDK's and they make you use the serial code to download them.
Edit: What do you think about what that CSR told me?
It was a unique serial code, if I remember correctly.
Anyway, even if they don't release an SDK, remember that we can access the internal binaries of the device because of how they've set it up to be read as a Flash drive. Further, I suspect that they'll be releasing applications of their own. If worst comes to worst, we can always take a look at those.
Qwerty.55 wrote:
It was a unique serial code, if I remember correctly.
Anyway, even if they don't release an SDK, remember that we can access the internal binaries of the device because of how they've set it up to be read as a Flash drive. Further, I suspect that they'll be releasing applications of their own. If worst comes to worst, we can always take a look at those.
Wait, the entire contents of the device? I assumed that the device software was abstracted away from the whole flash drive concept; am I incorrect?
Goplat has written a program to extract the device binaries. I haven't seen them as of yet, but it works from what I am aware.
Also, I managed to find the machine code for each of the SH3 assembly instructions. It should be possible to write a stand alone assembler for use to make the process much easier for others.
imho, if this thing gets a SDK and such (basically, not in the direction of TI) it should do pretty good for itself. I also am waiting for this SDK to come out, along with a nice library of programs
