M16C62 distributor in US or Japan

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View
Hi all.
I am currently researching the Renesas M16C62 part for a product, but
it seems like all their sales distributors do not stock the part. It
does not bode well for me since I have to guarantee availability as
well as functionality. Does anyone have some suggestions on a good
distributor in US or Japan?

Am I looking for the correct part or is this part being phased out and
a new one is coming in?

Thanks for all the past help this board has offered me. It has saved
me a lot of time and frustration.
Chris

Re: M16C62 distributor in US or Japan
I'm designing a project now for a huge customer, using the M16C/62P family,
specifically the  M30624F8PFP (64K flash, in single chip mode) My
distributor is Arrow.com. The 62P is the 3rd die shrink in the family, just
starting production this year. The 62A and 62N are still available, about 2
years old now. I think it's a well supported product. The technical and
sales support I've received from Renesas engineers, reps, and distributors
was a little slow to start with, but has ramped up and is excellent now
(especially as my inquiries have became more focused)  The performance is a
little better,  and price point is a LOT better than the part we are using
now, the Hitachi H8/2357. The 62P is full of peripherals, and it's
instruction set is optimized for C stack handling and bit operations. We are
using the IAR Embedded  Workbench V.2.11A, and the Renesas ICE PC7501.

Of course selecting any part depends entirely on your application and cost
restraints, but we are gearing up to use this part (unless something derails
us) and are confident that the supply chain will deliver thru the entire
product cycle, at least a few years long.

(Nope I'm not affiliated with Renesas or any of their agents, I'm a
customer)

chris wrote:
Quoted text here. Click to load it



Re: M16C62 distributor in US or Japan
Thanks for your valuable input Coat. If it is possible, I am curious about
the debug toolchain you are using. Is the Renesas ICE PC7501 a JTAG
debugger? Also, does it use C-SPY on IAR? Last question, do you know the
approximate cost of the ICE? Thanks very much. I am kind of leaning towards
it right now. I need to locate some parts though because all distributors I
talked to claim to be out of the part.
Chris




Re: M16C62 distributor in US or Japan
The PC7501 is a traditional ICE with a pod that replaces the entire CPU for
debugging. You can read about it and other Renesas development tools on the
Renesas website:
http://www.renesas.com/eng/products/mpumcu/toolhp/mcu/m16c_e.htm
The cost of the emulator and pod for the 62P is around $8000 USD (not
allowed to say what we pay)

But, there are cheaper (almost free!) alternatives to using the PC7501. The
M16C family doesnt have JTAG as such, but it does have an address-matching
breakpoint unit built in, with debugging thru a serial port. Using one of
the debug "monitors" (either the C-SPY from IAR, or the PD30 from Renesas)
that you link into your code, you run the debugger front-end on your PC. You
can set up to 6 hardware breakpoints using the capabilties built in to the
CPU, adding hardly any overhead. It's really quite good, if that's all you
need.

If you are like us, doing complex real-time control in a high volume
product, where safety is an issue (and where a product recall would kill
you) then you really must invest in, and carefully use the PC7501. One of
the time-saving optimization features of the M16C family is that it has an
instruction queue, where it reads in the next 4 instructions while it's
processing the first one, in anticipation that they might get executed
(unless a branch happens), so you have to be careful not to do any accurate
timing loops in software, always do them with one of the hardware timers
(there are 11 of them!)  So the cycle counter in the PC7501 is the only
accurate way to see exactly what the timing is, short of instrumenting your
code with writes to a port pin and measuring it with an oscilloscope. The
PC7501 also lets you set breakpoints on a variety of conditions, or to
capture a trace of what was exectued, or to see which lines of code were
covered (we have to stop execution to set variables to artificially force
exection of every possible line of code, just to see what happens) and lots
of other features.

There aren't a lot of alternatives to the PC7501, it's the most recent and
up to date tool of any I have found,  but you can check out the ADViCE from
Yokogawa, the M16C from Nohau, and the HyperStac from Sophia Systems (Sophia
is owned by Nohau now). I like the PC7501 because it's a recent addition to
the Renesas tool chain (replacing the older 4701U) and they are releasing
updates to the debugger (one just came out Sept 1/03) so the support is
present. And if you start with the cheaper PD30 debug monitor, then move to
the PC7501, you are still using the same debugger interface on the PC you
are familiar with, nothing to relearn! (except for the additional features).

If you are having trouble getting parts from the distributor, place a
request for some samples, and meanwhile you can order an Evaluation Kit,
it's a great way to get started, only around $100, and at least you can try
out some stuff while your boards are getting made and the samples are on the
way. (I see from browsing around on the Renesas website they've recently
added a newer "Starter Kit", dont know anything about it ) The evaluation
kit comes with a limited Renesas toolchain. The compiler is either code size
limited, or unlimited but with a 120 day lifespan, I forget which, but I
have seen both.  We chose the IAR toolchain after a lot of benchmark testing
with Sieve, Dhrystone, and our own application code comparing it to Renesas.
For now we are making a daughter card for development purposes with our
application circuts on it to attatch to the connector on the Evaluation Kit,
before designing our production boards.

I hope this helps.
Good luck.

Chris wrote:
Quoted text here. Click to load it



Re: M16C62 distributor in US or Japan
Hi Coat.
Thanks for your advice. I think it is swaying me towards the Renesas
(Mitsubishi) M16C26 micro. Maybe you should contract for them to receive
commission when you help with sales. Anyways, I am going to purchase a
starter kit first and download the demo of IAR for the chip. If things go
well, then I will probably go with the Renesas as opposed to the other chips
I am looking into (MOT 68HCS12, Atmel ATMega64, Zilog EZ80F93). Anyways,
hopefully I can end my painful trek of comparing micros soon. It is too damn
time consuming to do this. I wish someone could just have a database that
compares all microcontrollers known to man at once.
Chris





Re: M16C62 distributor in US or Japan
Chris:
for a M16C toolchain chack out: www.tasking.com.
I am sure we can compete.

If this does not suit you, send an inquiry to: snipped-for-privacy@altium.com.

I am sure you will like the TASKING toolchain better.


/jan

bjbi3j$nhe$ snipped-for-privacy@news01cd.so-net.ne.jp...
Quoted text here. Click to load it
chips
damn



Re: M16C62 distributor in US or Japan
omg yes, I wouldn't want to take anything away from Tasking by neglecting to
mention them. Sorry about that. Tasking is also very excellent!
Unfortunately for us, we had very limited time to do our evaluation and make
a decision, and now we are committed to IAR for this product, for better or
worse.  I wish I had a chance to evaluate Tasking against Renesas and IAR
when I had the chance.

Years ago I used the BSO compiler for the 8051, I believe it was, around the
time when BSO merged with Tasking, so I have not been keeping up with the
Tasking offerings. But, Tasking toolchain support for Renesas is there! For
example, when you open up your Renesas debugger, it asks you if you're using
a Renesas, IAR, or Tasking generated object file! That's what I call
compatibility! Also, Tasking is now bundling the CodeWright editor as the
front end of their IDE, which is awesome! Codewright is my editor of choice,
but I had to buy the standalone version from Borland, and then I had to fill
in all the batch calls to the IAR compiler myself.

Another reason why IAR and Tasking are great is that they both have very
responsive technical support. For example, when I was evaluating IAR, I
noticed they didn't handle bit operations the best way possible in some
cases: They did read-modify-write where they could have used bit ops... In 4
weeks they came out with a new release that fixed the problem! They rewrote
all the SFR headers to take advantage of the new scheme too! Crazy amazing!
Now IAR beat Renesas in our benchmark. hehe! And if Tasking is still
anything like I remember them, they are every bit as responsive as IAR is.
Unfortunately with Renesas, it seems if you have issues with their
toolchain, the request has to filter back to Japan for consideration before
you get action. Well, that's the way it seems.So by all means please
consider Tasking too! (Especially if you don't already have the CodeWright
editior, it rocks!)

That's it.


Jan Homuth wrote:
Quoted text here. Click to load it



Re: M16C62 distributor in US or Japan
Hi Jan and Coat.
Thanks for your input and all your help. Also, I must agree with you, Coat.
I use Codewright all the time and it is by far one of the best editors I
have used.

I still have a question regarding the M16 chain. I have been  searching high
and low for the development board so I can evaluate the chip. I already have
the IAR IDE demo installed on my computer and have played around with it.
The IAR rep said that I can request an extension once I get the actual
hardware. However, I have yet to locate the development board.
I checked both Arrow and Avnet for the part number listed on the Renesas
site for their Starter Kit 2 board. It is part of the M16/C62 family. The
part number I used was:

M3A-0654G01/G02

However both listed the part unknown. It is getting down to the wire and I
need to port the design soon. Is this the part number of the starter kit you
have gotten? The IAR website also does not list this part number in their
supported hardware platforms for their ROM debugger.

Anyways, sorry to bug you so much.
Chris





Re: M16C62 distributor in US or Japan
Chris,
this used to be the MSA 0654 starter kit board of Mitsubishi (now Resesas).

It is equipped with a M16C62 CPU.
Do not worry about the number.
A long as the M16C62 is on it it's all the same anyway.
Mitsubishi (now Renesas) also has a neat little board: the Three Diamonds.
You can get it in different flavors:
with thje M16C62A, the M16C6N (CAN derivative) and the M16C62P (newer 24Mhz
derivative with even more RAM and flash on chip).

Setup is different for these derivatives.
Mitsubishi supplies ROM monitors with all these boards.
TASKING has a ROM monitor implemented in C that comes along with the
toolchain - for all derivatives.

ragards
/jan

bjgo7c$at4$ snipped-for-privacy@news01db.so-net.ne.jp...
Quoted text here. Click to load it
Coat.
high
have
you



Re: M16C62 distributor in US or Japan
Hi Jan.
Thank you very much for your help. I appreciate it a lot and will look into
the boards and it may save me a lot of time and effort.
Chris



Re: M16C62 distributor in US or Japan


There is more than one kind of kit for the M16C/62P from Renesas.. You =
inquired about a Starter Kit-2 part number M3A-0654G01/G02. This part =
just recently showed up on the Renesas website (Aug 25) The original =
starter kit which is quite different, is part number MSV1632/62P-SKP... =
Renesas just today stripped the remaining references to the MSV off =
their website, but I'll bet the distributor will be happy to try and =
order one for you.

The older kit has a lot of features on it.. 2 line LCD display, EEPROM, =
4 buttons, a pot, 2 serial ports with driver, It doubles as a Kit for =
the 80 series cpu as well, it has a footprint for that chip. And a 150 =
pin DIN expansion connector (extra pins for the 80 series chip). RC =
reset circuit.

The new kit has two 7-segment displays instead of LCD. 96 pin expansion =
connector (not populated),  3 switches, a pot. It has a proper reset =
chip instead of an RC.




Quoted text here. Click to load it

Site Timeline