next topic »

[on:]

From atmel:

http://www.atmel.no/webdoc/sam4s16xplained/sam4s16xplained.introductionsection_lvq_mjg_xf.html

[Reply #1 on:]

Not quite interested yet

There are so many great boards coming out I'm suffering from GABOSS (Great Arm Board Overload Stress Syndrome).

I'm going back to my Z80, I understood that.

______
Rob

[Reply #2 on:]

LPC800-Xpresso board is now available, but it doesn't look like the LPC812 can be had separately yet.

[Reply #3 on:]

I just wish I had an immediate use for an 800 it looks like such a great chip. But I'm up to my arse writing low-level drivers for the LPC1227 at present.

______
Rob

[Reply #4 on:]

I'm going back to my Z80, I understood that.

I've still got "Programming the Z80" by Rodnay Zaks on the bookshelf. I keep thinking that somehow it might still come in useful in the future, but I'm not sure exactly how. 

[Reply #5 on:]

My copy is long gone, that was in another life. Every time this comes up I say I'd like to make a retro Z80 board. One day.

_____
Rob

[Reply #6 on:]

I just wish I had an immediate use for an 800 it looks like such a great chip. But I'm up to my arse writing low-level drivers for the LPC1227 at present.

______
Rob

I know, I'm trying to figure out something too. I did go so far as to download the datasheet and user's guide. I might have to try one out anyway. If a person wanted to get started with LPCs, do you think the LPC812 is as good as any, or is there a model you'd suggest? I'm really interested to see the pricing on these chips, I wonder how far off they are. The datasheet still has some TBDs in the electrical characteristics! The switch matrix is a very cool feature, I've wondered whether anyone did such a thing; now I know. Makes the 8-pin DIP a little more understandable, I think I'd still go for one of the 20-pin packages just for general goofing around though.

[Reply #7 on:]

do you think the LPC812 is as good as any

Probably, although it's note available yet. Although as was pointed out above the LPCXpresso board is and to start with that's all that matters. The LPCXpress system is very good.

or is there a model you'd suggest?

Well I'm pretty much into the LPC1227, at 48 or 64 pins and with 2 UARTs I find it's about right for what I'm doing now. I liken it to the Mega1284 which I think is the best AVR around. I like those chips so much my current PCB has one of each

I've written a lot of code for the 1227, originally just to learn the chip but it's slowly turning into a proper Arduino-style HAL. See here

http://lard.robgray.com/index.php

I've spent the last couple of days getting the serial Tx working, it's all FIFOs and interrupts and an unfamiliar chip but it's going now. See here (my 3-day old blog) for today's trials

http://ardweenet.blogspot.com.au/

So, at the risk of being biased I'd recommend the 1227, but one thing I like about the LPC range is that they go from 8 to 217 pins so there's something for every one.

______
Rob


 

[Reply #8 on:]

Thanks for the links and info. Will let you know if I make the plunge/progress. I see the LPC812 has no ADC, but depending on, that's not a show-stopper.

[Reply #9 on:]

I see the LPC812 has no ADC

If it wasn't for that it would be the perfect small processor IMO.

______
Rob

[Reply #10 on:]

There are a few factors to consider:

1) resources: NXP is quite stingy when it comes to onboard resources. Some of the ST chips for example have 20+ timers, and the nxp chips tend to be limited.

2) retargetting: you don't typically know the eventual target for your code. For a development board, you want a chip that is as big as possible. 1768/69 are fairly popular for that reason.

3) debugger: you want to use a fairly standard debugger (jlink/ulink). I would be worried about some no-name debugger.

4) ide/support: if you are developing commercially, you want to have a human on the other end of a phone call.

If it is a hobby, well, the weirder the better, I guess.

[Reply #11 on:]

I was reading a bit on the ARM architecture the other evening, found this in a Wikipedia article. I thought, only 35,000? Wow! Contrast to 1.4B transistors in Intel's Ivy Bridge chips. Not exactly an apples-to-apples comparison, but the gap in performance is much less than the transistor counts might lead a person to believe.

Using the RISC approach, the core ARM processor requires only 35,000 transistors, compared to the millions in many conventional processor chips, resulting in lower power usage and making it very attractive in smaller devices.

I've often wondered about the ATtiny and ATmega chips, but have never seen a transistor count for them.

[Reply #12 on:]

Looks like the LPC81xM chips are getting closer, Mouser now lists them as on order, although no delivery dates. Prices (single quantities) range from $0.78 for the LPC810 to $1.16 USD for the most expensive LPC812. Unfortunately the development board went up from $15.00 to $18.75 since I last checked. I may just order one up anyway. I also discovered that the free LPCXpresso IDE only does C, not C++. Major bummer.

[Reply #13 on:]

I also discovered that the free LPCXpresso IDE only does C, not C++. Major bummer.

Yep, and restricted to 256k (or is it 128k?) code although that's seldom a problem.

To get C++ you have to buy the > 256k code size version for $256, major PITA and not a good way to entice people onto the platform I feel.

______
Rob

[Reply #14 on:]

Yeah the free version is 128K, which as you say doesn't seem like much of a restriction. Then there are 256K and 512K versions, for $256 and $512 respectively. Cute. Obviously not a serious pricing policy, I wonder what they were thinking.