Where did the time go? ImageCraft Propeller C project status

ImageCraft

Whoa, it seems like only yesterday that I talked to the Parallax folks at ESC and now, Summer's end is in sight already!

I just notice that I have not given any update to you fine folks on the Propeller C project, so here's a short report: the project is under way and we have a good idea on the runtime architecture (*) I have reorg'ed the development process in our company so unfortunately some times were lost on bringing the resources up to speed. We are still planning for a beta release by the end of October and a production release soon afterward. The Parallax staff has been very supportive of our endeavor and we know our product will be a good complement to the Propeller chips.

Hang in there all ye C programmers. We will be there relatively soon.... As suaul, if you have any questions or comments, you can leave them here or send me an email richard @imagecraft.com.

(*) Recall that with Propeller C, we will be using a "large programming model" implementation such that the program memory size is limited by the HUB memory and not the COG memory, at a cost of about 25% of native code performance in the worst cases and 80%-90% of native code performance in some routines.

Gavin

Any tips/suggestions on hardware to get the max from the compiler?
Need to start working on PCB's if the compiler is only months away.

rjo_

I hope that you are planning to hit the Tech/University circuits with this... I can't imagine a more attractive way to learn about C... than learning about it in conjunction with the Propeller.

Rich

ImageCraft

First of all, we actually got a REALLY EARLY basic pre-alpha compiler prototype working to work out the runtime architecture issues. This means that we can handle really simple expressions and condition codes.... The Propeller is a different target than we are used to. With other targets, usually we stuck our heads in the datasheet and wham! a month later, we can get a basic integer only code generator working. With the Propeller, we could have done that too, IF we are targeting COG programs only. Obviously we want to do more than that so the design cycle is a bit longer, in addition to the other things I mentioned.

As for what the compiler needs, I presume you are asking about the Host requirement: currently the host is Windows (hiss, I know, I know) and really any half decent XP class machine will work fine. We even have people using ancient 486 machines... The IDE/compiler should work under VMs such as Parallel and Wine, but there is no guarantee yet, we will just have to wait and see how it plays out.

As for Universities etc., Parallax strong point is the educational market, so we definitely plan to leverage their expertise and market penetration in the area. Wait and see for now though...

// richard