Infineon Microcontrollers

[iso9001]

Does anyone have any experience with the XC16 or XC88 series ?

They are really attractive processors with exactly the peripherals I want (low pin count with multiCAN) but I can't seem to find anyone else using them,

So I fear the community suppport and 'Gotcha!'s would be a problem,

Thanks

 

[iso9001]

I think some of them run a new/modified 8051 core.

The datasheets all claim the perfect combination of RISC, CISC, and DSP. I'm guessing that translates to the main core running CISC, peripheral and coprocessor core running RISC, and since it has single cycle hardware multiply and a Mutiply And Accumulate register (mac) they can say its like a DSP.

Comments ?

 

[blueroomelectronics]

Since it's like a DSP why not just use a DSP like the dsPIC3x, tools are cheap too.

 

[iso9001]

Well,

I have experience with Microchip's CAN and ECAN modules and they have been OK but not stellar. The ECAN modules appear to be lacking features the Ininfeon series has in all chips.

The 4ma dsP vs 5ma sink/source is about the same. Except that there is 2.5-5V output on IO lines for the Infineon.

I don't technically 'want' a DSP but if my processor had DSP features like hardware multiply that would be fine.

Another plus is that the Infineon line seems very constant. For instance, Port 5 is always Analog Input, port 9 always has the CAN and addition IO lines. Where Microchip seems to change port names, numbers, and locations with models even in the same series. It appears extremely easy to switch code written for the XC8 to XC16 and even a couple XC22 (32bit).

Another factor is that the Infineon line goes to 5xCAN and 5xSerial/SPI. The microchip line goes all the way to 2. I don't think I need 5 can lines, but 3 could be handy here or there.

Microchip support is phenomenal, no doubt. However, I think I've graduated to a big boy's uC I'd rather not get into Freescale since their pin count is 80 at the min where I only need 20 or 30, and I despise paged memory locations.

I don't know, I do need to consider the dsPic series, but something just doesn't sit right with me. Maybe its the Chinese translations of the datasheets? maybe it's that I know I wouldn't use even 1/4 of its DSP features so why use a DSP ? They seem a little more fragile then a full on uP.

Arg, looking into new processors sucks!

 

[blueroomelectronics]

Well I've never read the Chinese versions of the datasheets, but I've read plenty Chinese to English datasheets that are pretty bad.

 

[dknguyen]

How are you affording the development tools for these ICs? I seem to be stuck with Microchip even though I've really wanted to step into Coldfire, ARM, or Blackfin but I can't afford the cost of the tools. Sometimes the programmer is expensive (but I can still take it), but nowhere near the cost of the compiler itself and the free versions are ridiculously limiting (if there even is a free version). I'd be lucky to get them even grant me a 30 day trial but have never gotten any responses.

 

[iso9001]

Good question about the cost. Fortunately for me, my company recognizes that these are expenses.

However, when its just for me, or when i need a better test than the demo can offer, I look for more creative means of evaluation if you get my drift. (PM me if you need some more info.) That, and we all read and write assembly here, not much is off limits when you get to that point


Yea, I meant the to English translations.

In the Infineon sheets I've read so far it seems informative and, almost a little candid. Stuff like "Load the register with a single word into CTRL0BG Register ...." then later "This means you'll need to load two bytes even though the first 4 bits will be overwritten later"

I think the dsPic series is about the top of what I need right now. The Infineon series looks like I would have room to grow.

 

[iso9001]

Hmmm.... I just found out that the XC16 series IS the ST10 from STM. How about that.

I'm not sure if STM got the core from Infineon or the other way around. But, it appears that opens a little bit of support up for that chip.

It also appears the prefered IDE/Tool chain "Keil" is very extensive with the XC16, ST10, and 805X and ARM/Coretex chips so going up with the same tools would be cake.

I apparently have a lot of reading to do tomorrow.

 

[Papabravo]

Infineon is a rotton company to do business with. They renegged on a committment to deliver 1000 pieces of a C505 with a 28 week lead time and told us to go *bleep* in our hats when after 28 weeks we asked where our parts were. They even had the cheek to tell us to stop calling them! Turns out NA customers were on allocation while their German customers could get all they wanted.

I vowed that if I ever had a chance to pimp them in the future I would not pass it up.

Search Google news to find out about the sacking of the CEO, and the rumor they might be takeover bait as they miss their numbers. Great company to depend on.

 

[iso9001]

Well papabravo, I'll certainly have to take that into consideration. You were using an Infineon 8bit mirco in Michigan huh? That almost has to be automotive work.

Since the part I am looking at is almost identical (in code and features) to the ST10 I'm a little less worried about allocation. My volumes are also incredibly small compared to auto suppliers. I'm looking at maybe 1000 parts a year. If its a problem, I just go to Germany and export 1000 out

How are the parts / architecture to work with? The C505 is an 8051 right? Are they pretty close to other 8051 based parts like Atmel, Dallas/Maxim, Analog Devices? I take it you were too far along to switch devices (even to another 8051 device?)?


(As an aside, that is 'trashing' them, not 'pimping' them. Pimping them would be recommending, but its ok, I know what you're saying)

 

[Papabravo]

It was automation related and we switched immediately to the Atmel AT89C51CC01 for all projects after April 2002. We haven't bought so much as a transitor or a FET from them since that. Serves 'em right the rat bastards.

In my vernacular 'to pimp' also means to bad mouth or speak ill of.