This may be the first processor-based certification but I see calls for certification come and go periodically. I have yet to see one which was worth the effort for someone doing embedded software and I expect this one to be no different. Contrary to Kris's assertion, I don't expect it become a requirement; I've never worked with an ARM core and don't expect to in the foreseeable future.
Are there any prerequisites before taking the exam, for example certain years of experience in working with ARM based processor hardware/ software? I feel, mandating a year or a couple of years of experience in working with ARM based processor for taking the exam makes practical sense. But, that might create "chicken and egg" situation for the engineers fresh from college.
As far as I can tell there are no requirements for taking the tests....apart from a working credit card.
Never-employed students can take the tests and are encouraged to do so. But you have to know the stuff to pass the test.
I doubt if it is real locking. I have passed some Cisco exams, but over the longer period this was a terrible mistake. They are overpriced and focused only to Cisco features with have very small connection to real problems in the network. I suspect those courses and exams will no make the enginner understand how to design and programm products. They will only teach some templates and thats all. Of course they can teach some specific naming convension or any other things that typical engineer is aware of.
I agree elektryk321, I doubt it is real locking too...but I can easily see why ARM would be trying, it is very tempting to corner the market...Cisco used to be dominant networking box maker, but even at its peak they had 10 or so competitors so being certified in Cisco gear had limited value...in microprocessor space there is only ARM, Intel and Mips so the market is more concentrated, it might be easier to pull certification trick...it will be interesting to see what Intel does, I am sure they are watching...Intel certification to follow? Kris
To say that only those three are in the microprocessor 'space' shows a lack of knowledge of what is shipping out in the real world. All the older 8 and 16 bit cores are still shipping in high volume, and the fpga/asic IP cores are also shipping in high volume. And most are NOT based around Arm and Mips. I designed high volume products for over 20 years before using an ARM part and that was based on the cost benefit, not any architectual superiority. Certification only has meaning IF the person can apply the knowledge used to gain his piece of 'paper' in the real world.
Far more neutral but good real test is the WCET conducted by the IEEE to certify wireless engineers. Cisco certification, which is a joke (does not need even a basic degree and is limited to Cisco products), has brought a bad rap to certification. The IEEE certification is conducted worldwide and expects a good background both in RF and EE. It covers both cellular and WiFi but quite well in depth.
Good point @KRagh...I think certification should be left to left to IEEE or similar bodies that has no incentives to push certain technologies or products...so we need microprocessor certification program IEEE! Kris
For details on the Wireless certification by IEEE, please visit: www.ieee-wcet.org/
In response to comments by elektryk321, I would say many people have fallen pray to Cisco's advertising tactics. Cisco is not a wireless company and does not have the background to conduct wireless certification. One look at the syllabus by IEEE exam will con tests. One look at the IEEE site will convince any engineer what is needed (prep needed is quite extensive) since the exam includes latest topics. Unless one is a praciticing wireless engineer, it is hard to attempt this exam.
I think this is a very bad thing for engineers. Now company HRs who have no idea what embedded systems programming is all about will start denying jobs to engineers with 20 years' embedded systems development in favor of a recent grad who studied for the ARM test because they have the piece of paper. Brilliant move by ARM, but a scheme that will ultimately, I suspect, will make us see lower quality code being produced "in the wild".
Well the good news is engineers are not compelled to take the test.
But if employers and HR people start asking for AAE certification as a pre-condition for certain jobs and work then those that have it will get the work.
But engineers with 20 years experience should be able to schedule a test and pass easily, right?
So then it will be back to a level playing field with the experienced engineers beating out the recent grads that only have the AAE.
I believe that this is a good step by ARM, not because it will generate lot of good programmers but it will certainly proliferate ARM's architecture among the student community and be helpful as a guide for employers in ascertaining a candidates profile. The certification has multiple levels so it's not just about processors but includes software, hardware and systems knowledge which makes for an all round skill for intending to go all the way.
It would probably also help in increasing the arm talent pool in the job market and also aid free lancer embedded programmers who are new to ARM.
Blog Doing Math in FPGAs Tom Burke 15 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...