EmbeddedRelated.com
Forums

Microcontroller comparission

Started by bhavesh gharat March 2, 2006
i want to Compare two microcontoller .
perivously used NEC controller ( otp)
now i want to go for other micro contoller ( microchip)
so i need basis for the comparission like code porting ,cost, tools
etc
i would also like to code Porting( asm -asm) effort estimation basis
thanks 
bhavesh

bhavesh gharat wrote:

> i want to Compare two microcontoller . > perivously used NEC controller ( otp) > now i want to go for other micro contoller ( microchip) > so i need basis for the comparission like code porting ,cost, tools > etc > i would also like to code Porting( asm -asm) effort estimation basis > thanks > bhavesh
The Microchip parts are a law unto themselves. They bear no relation to anything else on the market. Porting asm from an NEC part to one of these will be a nightmare. You would be better off choosing a more conventional part like an 8052 derivative. ian
Ian Bell wrote:

> > i want to Compare two microcontoller . > > perivously used NEC controller ( otp) > > now i want to go for other micro contoller ( microchip) > > The Microchip parts are a law unto themselves. They bear no relation to > anything else on the market. Porting asm from an NEC part to one of these
While I agree about Microchip, any job involving NEC's casually documented and occasionally functional microcontrollers is such an exercise in sado-masochism that it is worth almost any pain to migrate into a different vendor's part. I wouldn't pick uChip as the target, but when compared with NEC it's actually a step up in ease of use.
Hi,
We are looking for 8-bit microcontrollers with following features:


Features	8-bit Microcontroller
Operating Frequency	10- to 40-Mhz
Program Memory	64KB
Data Memory	4kbytes
Data EEPROM Memory	1024 bytes ( optional)
I/O ports	54- to -60 ( with 3.3/5.0 v  compatiblity)
ADC / channel	10bit X 8Channels (100ksps)
Vcc (Volts)	2.0 -5.5
16 bit Timer	3
8 bit Timer	2
RTC ( on chip)	1
SPI	1
UART	1
I2C	1
ICSP/ICD	Yes
Brown-out-reset	Yes
Watch Dog Timer	Yes
On chip Clock	Yes
Hardware Multiplier	Yes
Interrupts	28
External interrupts	4
Cost Per Chip	~4$ to 5$
c
can any one suggest any beter part option are available with other
manufacturers

In comp.arch.embedded,
bhavesh gharat <bhavesh.gharat@math.net> wrote:
>Hi, >We are looking for 8-bit microcontrollers with following features: >
Why do you restrict yourself to 8-bit controllers? The rest of the requirements (which I snipped), look like there might be a suitable ARM out there for you. But a matching 8051 is a possibility as well. And price depends on quantity, so you should mention that. Your requirements also seem copied from those of an existing chip, but what are your actual requirements? Do you, for instance, require 28 interrupts? -- Stef (remove caps, dashes and .invalid from e-mail address to reply by mail) Things past redress and now with me past care. -- William Shakespeare, "Richard II"
larwe wrote:

> > Ian Bell wrote: > >> > i want to Compare two microcontoller . >> > perivously used NEC controller ( otp) >> > now i want to go for other micro contoller ( microchip) >> >> The Microchip parts are a law unto themselves. They bear no relation to >> anything else on the market. Porting asm from an NEC part to one of these > > While I agree about Microchip, any job involving NEC's casually > documented and occasionally functional microcontrollers is such an > exercise in sado-masochism that it is worth almost any pain to migrate > into a different vendor's part. I wouldn't pick uChip as the target, > but when compared with NEC it's actually a step up in ease of use.
Can you give more details of your NEC woes? First I have heard of them being problematic. Ian
bhavesh gharat wrote:
> Hi, > We are looking for 8-bit microcontrollers with following features: > >
ATmega6450
> Features 8-bit Microcontroller 8
bit
> Operating Frequency 10- to 40-Mhz 16 MIPS
max
> Program Memory 64KB 64 kB
Flash
> Data Memory 4kbytes 4
kB SRAM
> Data EEPROM Memory 1024 bytes ( optional) 2 kB EEPROM > I/O ports 54- to -60 ( with 3.3/5.0 v compatiblity) 68 I/O > ADC / channel 10bit X 8Channels (100ksps) 8 x 10 bit but
not 100 ksamples
> Vcc (Volts) 2.0 -5.5
1.8V - 5.5V
> 16 bit Timer 3
1 x 16 bit timer
> 8 bit Timer 2
2 x 8 bit timer
> RTC ( on chip) 1
32 kHz oscillator
> SPI 1
SPI
> UART 1
USART + USI
> I2C 1
TWI
> ICSP/ICD Yes
JTAG
> Brown-out-reset Yes
BrownOut Reset
> Watch Dog Timer Yes
Watchdog
> On chip Clock Yes
Calibrated Oscillator
> Hardware Multiplier Yes H/W
Multtiplier
> Interrupts 28
24 MultiInput Wakeup on all I/O
> External interrupts 4 > Cost Per Chip ~4$ to 5$
Negotiate with your disti...
> c > can any one suggest any beter part option are available with other > manufacturers
So it meets everything except the sample rate of the ADC. -- Best Regards, Ulf Samuelsson ulf@a-t-m-e-l.com This message is intended to be my own personal view and it may or may not be shared by my employer Atmel Nordic AB
bhavesh gharat wrote:
> Hi, > We are looking for 8-bit microcontrollers with following features: > > > Features 8-bit Microcontroller > Operating Frequency 10- to 40-Mhz > Program Memory 64KB > Data Memory 4kbytes > Data EEPROM Memory 1024 bytes ( optional) > I/O ports 54- to -60 ( with 3.3/5.0 v compatiblity) > ADC / channel 10bit X 8Channels (100ksps) > Vcc (Volts) 2.0 -5.5 > 16 bit Timer 3 > 8 bit Timer 2 > RTC ( on chip) 1 > SPI 1 > UART 1 > I2C 1 > ICSP/ICD Yes > Brown-out-reset Yes > Watch Dog Timer Yes > On chip Clock Yes > Hardware Multiplier Yes > Interrupts 28 > External interrupts 4 > Cost Per Chip ~4$ to 5$ > c > can any one suggest any beter part option are available with other > manufacturers
Some of your requirements show that you pretty much just copied a bunch of info from the data sheet. Why does your replacement have to be 8 bits? Why do you care how many interrupts there are? You ask for 28 interrupts and 4 external. Is 10 MHz a minimum speed or do you need a part that can work over the range of speeds? There are ARM MCUs that can do this job and give you a lot more speed. Check out www.gnuarm.com. Go to the Resources page and scroll down to ARM Chips, then select ARM device comparison chart. It works better with Firefox than IE.
Ian Bell wrote:

> Can you give more details of your NEC woes? First I have heard of them being > problematic.
Oh boy, where do I start? The LCE development kit includes a little board with a ZIF socket on it, for programming flash versions of the micro. Nobody at NEC knows how to use this board - the comment I got was "nobody ever got it working". Transpires that a clock signal provided by the LCE is somehow inadequate to drive the part being programmed; adding an external crystal and startup caps onto the board makes it work. Of course this was something I had to discover from first principles. The LCE gets blazingly, blazingly hot. Almost literally. I can feel the thing radiating onto my arm from a foot away. Perhaps related to this, it's terribly unstable; it rarely runs longer than an hour or two. The control software for the emulator is of that uniquely pleasurable type that needs a week of tweaking and calling tech support before it will work - each time you install it. The parts [at least the ones we use] have no JTAG or other in-system test interface. You have to solder down a fake chip that hooks into a full ICE. These fake chips are composed of a stack of interconnects that are incredibly unreliable. The parts have horrible EMI problems. Software toolchain support is poor because nobody is using these parts. All nontrivial questions go through Japan and are apparently filtered by the "ignore stupid gaijin" filter. They were originally speced in because they're cheap. Turns out that in the [literally] years it has taken to get our project working with their part, other vendors have come in with better parts at lower prices (and with other vendors we get flash for less than NEC charges for ROM).
larwe wrote:

> > Ian Bell wrote: > >> Can you give more details of your NEC woes? First I have heard of them >> being problematic. > > Oh boy, where do I start? The LCE development kit includes a little > board with a ZIF socket on it, for programming flash versions of the > micro. Nobody at NEC knows how to use this board - the comment I got > was "nobody ever got it working". Transpires that a clock signal > provided by the LCE is somehow inadequate to drive the part being > programmed; adding an external crystal and startup caps onto the board > makes it work. Of course this was something I had to discover from > first principles. > > The LCE gets blazingly, blazingly hot. Almost literally. I can feel the > thing radiating onto my arm from a foot away. Perhaps related to this, > it's terribly unstable; it rarely runs longer than an hour or two. > > The control software for the emulator is of that uniquely pleasurable > type that needs a week of tweaking and calling tech support before it > will work - each time you install it. > > The parts [at least the ones we use] have no JTAG or other in-system > test interface. You have to solder down a fake chip that hooks into a > full ICE. These fake chips are composed of a stack of interconnects > that are incredibly unreliable. > > The parts have horrible EMI problems. > > Software toolchain support is poor because nobody is using these parts. > > All nontrivial questions go through Japan and are apparently filtered > by the "ignore stupid gaijin" filter. > > They were originally speced in because they're cheap. Turns out that in > the [literally] years it has taken to get our project working with > their part, other vendors have come in with better parts at lower > prices (and with other vendors we get flash for less than NEC charges > for ROM).
We used to deal with Sunrise Electronics in the UK seemed pretty good when I was using or considering using an NEC part. What is your local distribution support like? Ian