Gentoo Archives: gentoo-embedded

From: Peter Stuge <peter@×××××.se>
To: gentoo-embedded@l.g.o
Subject: Re: [gentoo-embedded] OT: HiTech-C question
Date: Mon, 15 Nov 2010 19:06:40
Message-Id: 20101115185307.4013.qmail@stuge.se
In Reply to: Re: [gentoo-embedded] OT: HiTech-C question by wireless
1 wireless wrote:
2 > > It's not a realistic spec for any microcontroller. Please try again,
3 > > with more care. You can get most of what you want in a single package
4 > > but not all of it. Unless of course you make your own.. Take an Actel
5 > > M1A3P250 with an ARM Cortex-M1 hardcore, then you could easily fit
6 > > all those peripherals in one package.
7 >
8 > Oh sure it is, but not in the 32 bit world.
9
10 You can get one, but will end up with a much larger chip, in order to
11 find one which has all the things you needed, and it'll also have a
12 ton of other stuff that you don't need.
13
14 To a degree I think this goes for all processor makers, but granted,
15 Microchip really have very many parts with only small peripheral
16 differences. :)
17
18
19 > > M1A3P250 starts at $11.99 at Future Electronics. (MOQ=180, was 90 before)
20 > > But maybe you'll be able to put something else on the board into the
21 > > FPGA to balance that extra cost.
22 >
23 > yes, 32 bit and dsp processors have come way down on price.
24
25 The M1A3P250 is not a processor, it's a processor and FPGA combined.
26
27
28 > But, when you look at building a complete embedded system,
29 > those high end processors eat you alive on external
30 > component count and manufacturing costs.
31
32 The point that this thread tries to make is that all 32-bit
33 processors are not "high end" as you might be used to.
34
35 In particular the Cortex-M products are quite fuss free. A handful of
36 caps is really all you need. That goes for the M1-enabled FPGA too.
37
38
39 > That board I just spec'd cost less that $30 to manufacture, with a
40 > PIC and every thing else that I did not require, like molex
41 > connectors and such.
42
43 I think the cost would not be significantly higher if using something
44 more powerful than a PIC, and the other point this thread tries to
45 make is that the development work would be significantly easier,
46 netting a total reduced cost.
47
48
49 > > As you see, part cost is no problem for ARM, but you'll need more
50 > > than one component for your project however you do it.
51 >
52 > PRECISELY!; a 32 bit part can never compete with a micro if
53 > specs are tight and cost/power requirements are astringent,
54 > which most are. Certainly anything that is manufacutured in
55 > lots of 10 or more, every penny counts and cost reduction
56 > rules the decision process, never what some employee or
57 > consult "likes". They (32+) only compete when you actually
58 > need all those mips and mops, which is rare for the vast
59 > majority of uP based products.
60
61 I think you would benefit from re-evaluating this position, quickly.
62
63 And of course it is simply folly to save on production cost in a
64 small (1k, 10k) run if there is a noticeable tradeoff to be made with
65 software/firmware development effort.
66
67 For lots of 10, 100, 1000 and even 10000, pennies in production are
68 irrelevant, they translate to just a few hours worth of development
69 time.
70
71 I haven't looked closely at the power numbers for M0, so for power,
72 physical size and mass production I agree that it remains very
73 important to choose parts very carefully.
74
75 But ARM cores have quite significant benefits in development, and
76 especially with Cortex-M0 they are eating up big parts of what used
77 to be an 8- or 16-bit only market.
78
79
80 > Don't believe me, just do a little research into the numbers,
81
82 This is my point too.
83
84
85 > Fairchild and such won't even talk to you about
86 > anything less than 1M in qty per quarter.
87
88 That's certainly not my experience from (in particular) Fairchild.
89
90
91 > For large companies, those (8/16)uP are sub $1, for qty 10k or
92 > more....... Some companies sell uP for pennies, just
93 > to get the supply contract for the passives and such
94 > on really large deals.
95
96 Of course it may be significant to save $1 (vs. the $1.46 ARM in
97 100qty, assuming you can get down to $0.46 for something else) for a
98 10k run, but certainly not for a 100qty run. It buys just one hour of
99 development time.
100
101
102 > 8/16 STILL rules the world and dominates the economics of embedded.
103
104 The state today is mostly uninteresting IMO, I find what happens
105 tomorrow all the more interesting. ARM is quickly taking a big part
106 of the market.
107
108
109 > Granted 32 bit cores that run linux are very cool and preferred by
110 > most embedded folks, but, that's a very small number of design wins
111 > with big quantity (cell phones for example), compared to their
112 > mature brethren (8/16).
113
114 That's comparing apples and oranges. I think you should really take
115 a look at the smallest ARM cores.
116
117
118 > and yes, I like ARM very much, particularly in areas of
119 > low power design, relative to intel or amd.
120
121 While more on-topic for gentoo-embedded that is only the Cortex-A
122 parts, which is on the opposite end of ARM's line card. Look into
123 the Cortex-Ms.
124
125
126 //Peter