It is really uncompleted description.
Even today no one produce single DSP but multi-core DSP which typically integrated with ARM devices.
There is also the issue of the prices and specifically price/performance ratio which hardly achieved using Intel Embedded devices.
note for example that TI ARM+DSP multicore based device is only about 10-25 Watt comparing to almost 90 Watt for Ivy-Bridge (8 cores).
Additional advantage is that DSP device are completely SOC (system on chip) which reduced significantly the total price for processing unit.
See HP moonshot project .
An application using a TI ARM+DSP would never be moved to a 8-core Ivy Bridge targeted for servers. But what about a 17 Watt 4 core Ivy Bridge like that in the Macbook Air coupled with a GPGPU? And a similar Haswell processor will soon be less than 10 Watts.
Could you kindly explain definitions of embedded processors and of embedded systems
Intel (computing), Freescale (communications), AMD (gaming) - they all offer embedded processors
I am quite lost in definitions used and would appreciate your guidance. Many thanks in advance.
Sure, the traditional definition of embedded is any sort of computer system that is not a PC/notebook/server in itself but is a computer/comms element built into (embedded) in something else.
Thus the field is an all-others, catch all that includes all sorts of military systems (radars, satellites, signal intelligence, missile guidance, etc.) as well as industrial (machine vision, process control, process automation), medical (various implants or external monitors and pumps including hospital test systems and monitors) automotive (drivetrain and infotianment) and much more.
They are generally similar in that they have more hard requirements in terms of operation at broader temperature and weather conditions, need for longer use life time, rugged conditions, better real time operation (lower latency) and higher safety requirements etc. that traditional PCs.
It's an old term that may need a marketing refresh, but its still a market reality.
Embedded System == A computer that does not look like a computer.
This looks like an article written by an Intel shill. I don't see much in the way of Intel processors in most embedded space. They are just too power hungry and expensive for most embedded markets.
Perhaps Intel still holds ground in some areas that are not power and price sensitive, eg. military and aerospace.
Rick, could you please define the word "cut" in terms of goverment budgets. I ask because it's been redefined to mean something non-intuitive. In other words instead of spending the proposed 10% increase, they spend 7% and call that a 3% 'cut'.
A classic example is my illustrious governor. When he took over this state's budget it was $28B. Now it is $37B and yet in his preparation for running for president he is claiming to have "cut" $8B in his tenure.
In government-speak that's budget "cutting" for you.
At the Federal level this PR chicanery is even worse because they can blur the lines by printing money, or borrow from future generations, and current Chinese ones.
Overall, even with the so-called sequester, we will still spend more in absolute dollars than last year. But does that apply to the military portion of the budget? Are they actually going to spend less of our cash this year than last year? Or, like most departments, will they just see a reduction in the rate of spending growth?
Yeah, I'm much more in line with C VanDorne on this "cuts" business. What will really set things straight is when we see more long-lasting signs of the economy recuperating as a result of sequestration, rather than vv.
Every tax-payer dollar the government doesn't spend is a dollar someone or something else can spend or invest. We're fortunately not YET a country of pathetic incompetents who depend on the government for "leadership." Although the more time passes, the more that appears to be happeneing.
In the US, the government is supposed to implement what people want, not the other way around. Let's see just what this sequestration really does to the economy, before buying into the dire predictions of those who want to take control over the people. Shall we?
The market pie charts showed xTCA, cPCI, VME, VPX, and Other. Does the Other category include IBM Blade Center? My company works with that ecosystem, even though it has a relatvely small number of vendors.
The market rank chart for VITA also mentions Extreme Engineering at the bottom. It would be good to include their trends, since they are a strong competitor to GE, Mercury, Emerson and others.
I think the embedded market can only increase in the immediate future, simply because everything is going that way. Not just automotive, military hardware, and industry, but everything. Even home appliances such as refrigerators, washing machines, stoves, HVAC systems, only recently still the province of mechanical timers and relays, are switching to embedded processing en masse.
So IMO, there's lots and lots of opportunities for all of the architectures, given that many applications have special needs. One size does not fit all, by a long shot.
The market Rick mentioned is really oriented towards high end embedded computing: defense, avionics, telco, etc. Look at the suppliers (Kontron, C-W, GE) and at the physical formfactors: armored boxes with MIL spec Amphenol connectors. This is a high margin market---I get their trade magazines and they just speak high profit (glossy paper, ads with flags and ships and fighter jets).
The flip side is the low volumes. If Intel is basing its embedded strategy on this market, I think ARM is going to win in the long term. In order to compete, Intel would have to significantly increase integration by offering SoC chips rather than CPU+chipsets, and reduce prices dramatically. A $50 CPU that requires $30 chipset is just not enough anywhere but on the highest end.
I can't imagine that Intel will fair well in "traditional" embedded design wins. It uses too much power compared to ARM and other options. While I agree that FPGAs can provide significant performance that comes at a cost in $. I can see the combination of DSP/floating point processor with an ARM core as a viable performer both in terms of cost and power.
What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Are the design challenges the same as with embedded systems, but with a little developer- and IT-skills added in? What do engineers need to know? Rick Merritt talks with two experts about the tools and best options for designing IoT devices in 2016. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.