Lpc2000

lpc2100_fan wrote:

>
> Al,
>
> when you wrote this
> "Whoops, sorry, thought I weas replying to that group, should read
> more closely."

I simply didn'yt see the point in swapping since Rick is in both groups. 
No ulterior motive. I don't expect Philips to listen to what I have to say.


>
> You actually had the option to do exactly that, responding to the
> group for Analog Devices ARM chips, but you needed to tell this group
> once more about your previously expressed frustration that Philips
> does not build the part you need with high precision analog (expensive
> to implement). Instead they implement of many comms (cheap to
> implement) that you thinks the majority does not need.

More like the bulk of the microcontroller market suggests aren't 
necessary. Yes Philips are doing well with this family, for good 
reasons. It has great bang for buck ratio, lots of memory (far too much 
for me) and plenty of comms. It's a common platform that will be come 
fairly ubiquitous over time, everybody and their uncle will know it, a 
bit like PICs,  but all those comms have to communicate something don't 
they? Most often that is data gathered from sensors, with the Philips 
parts any price advantage, to me, is killed by the need for external 
circuitry. Comms and sensors are the two fastest growing fields in 
electronics right now. Surely it makes sense to combine them.


>
> Guess the popularity of the LPC2000 which is reflected in this group
> speaks for itself.

Yes it does, but it isn't even close in  popularity to, say the PIC or 
AVR, and the reason for that isn't price, it is cheaper than many of the 
higher end parts of these families. Don't get me wrong, qwhen I first 
heard about the LPC2100 a year or more ago I was straight on the phone 
to my disti chasing samples parts info, anything. It looked like the 
answer to performance issues I was chasing. When the first parts were 
announced I was a bit disappointed by the configuration of peripherals, 
but thought, no, they'll test the water then come back with an 
assortment of peripheral mixes/sizes. But the mix they came back with 
was, to say the least, disappointing to me. Maybe not to anybody else, I 
accept that. In most popular micro families though, (and this is one 
reason I believe they are popular, along with cheap tools, good support 
etc) you find the very lowest end parts have simplest peripherals, small 
memory, low to medium pin count, the very highest end parts have comms, 
timing, analog and often other peripheral classes combined with high pin 
counts, large memory etc. In between there are low (20) to medium (64) 
pin count devices with various combinations of analog, timing and comms, 
all on one part, with various memory sizes. For example PIC, AVR, Ti, 
Mot, etc all have micros with 40 or less pins that have UART, IIC, SPI, 
A/D, Timer cap/com, PWM plus others like comparators etc. In a range of 
memory sizes. The smallest memory available with Philips is 128K flash, 
16K RAM. Other than for data logging purposes I've never needed that 
much memory. perhaps I will with the ARM, I don't, know, but I doubt it. 
that is just wasted silicon from my perspective. I am going to run an 
LPC2138 trial, I had to go that far to get ADC and DAC, because I may 
yet need the full 60Mips CPU speed. But, whereas I'd envisaged this part 
taking over the next build of 4 or 5 existing designs, the reality is 
that the ADuC looks a better mixed peripheral option, and the MAXQ2000 
looks like a better low end option.

heck, I've got the Keil MCB2130 dev tools AND the IAR 2106 dev kit, so 
you can't say I wasn't keen to give it a fair go.

>
>
> May be you want to look at the LPC2131 or LPC2132 both have 10-bit
> ADCs but no DAC. What about DAC through PWM?

I need 4 DACs, preferably of 16 bit resolution, although 12 bits and 
oversampling will work for the first stage. I'll probably end up using 
the 16 bit  ADuC parts when they become available, but for now will 
settle for the 12 bit parts. Size limitations precludes external codecs.

>
>
> No the best analog chips around but probably close to best price
> performance that I know off. 

The prices I've been quoted so far don't reflect that. In fact the 
budgetary rpice I saw for 1k ADuC70xx was slightly over half the price I 
was quoted for a simple LPC2106.


> There is a reason why the other company
> is called Analog Devices and there are also reasons why those devices
> are popular with less people than the LPC2000 series. 

They aren't available yet, and AD haven't run a world wide sales seminar 
as Philips has done. (I hope to hell they do!) I'd say the amount of 
traffic I'm seeing on the parts now reflects the LPC21xx at a similar 
stage in its development.

> Guess Philips'
> decision with the "poorly executed" family wasn't so bad after all.

Poorly executed in my opinion,(which counts to noone except me) and, 
when they get to be in the top 3 for number of sales of micros, I'll 
believe they made the right choice, because I believe that they had the 
potential to achieve this with the improvements they've made to the core 
speed, and their price point. After all Microchip went from obscurity to 
number 1 in sales on the back of a severely brain damaged piece of 
silicon, and Ti showed, with the MSP430, that a well thought out family 
of parts could succeed beyond expectations in the market, despite the 
best efforts of the company not to support it at times.

>
>
> Please no offense ;-)  Bob

I never TAKE offense ;?} What use would such a group as this be if we 
couldn't express and exchange our thoughts without acrimony? Nobody has 
to agree with me, and they probably don't.

Al