Heh, maybe. I know about this a few days. It is not the first time but Crossroads starts topic about end of support of full swing oscillator on some ATmegas and he really scared me. 
I do not want PIC and AVR owned by the same company >.>
Stock up now if really concerned. I think the notice said end of Feb for ordering?
So Atmel is going the way of Nokia? Somewhat sad.
To me both AVR and PIC chip architectures are going to struggle surviving the longer term.
The reason being is that for many markets, including things like the IOT market, cost and density is a big factor.
The way to get the cost down and reduce footprint size is to go to custom silicon.
I don't think either AVR or PIC32 (mips) is available as a core to integrate into your own full custom ASIC, and that is why ARM has gotten so popular.
When doing custom silicon you can easily include a processor core inside your chip to lower the overall cost of your design.
Why do you think ARM is so popular in the cell phone market?
They all went custom ASIC to get the cost and size down and used an ARM core inside their custom ASIC for the processor.
If you have to use discrete components like discrete micro-controllers whether AVR or PIC, it costs more since more components are involved. And it will consume more room on the PCB.
Mass volume is all about the cost.
I went through this exercise nearly 20 years ago, when my company did a custom ASIC with an ARM core.
This allowed us to reduce the cost of our DSL modem by more than 70% which allowed us to take over 90% of the world ADSL market at the time.
There was no way to get the production costs that low using off the shelf components and that is why we won all the business. Almost nobody was doing custom ASICs back then.
Our sale price to the Telcos was lower than many competitor build costs.
If I were an investor, I'd really want to know how companies like Atmel, MicroChip, Dialog, etc... are going to play in the new world of custom silicon which is where the big volumes are going to be.
Sure, there will probably always be a market for discrete processors, but if I were an investor I'd much rather be investing an company that collects royalties like ARM ltd, that doesn’t have to actually make the chips vs one that has to not only design them but make and fab them as well.
I also think that the days for 8 bit processors the like AVR is pretty numbered.
Not that they aren't useful, but 32 bit processors are starting to have the same price points.
So if a 32 bit part can provide the same costs and meet the needed power requirements, why mess with an 8 bit processor and have to deal with all the 8 bit limitations?
--- bill
I also think that the days for 8 bit processors the like AVR is pretty numbered.
People have been saying that for 20 years.
I'm more worried about Microchip's decidedly unfriendly approach toward beginners and experimenters: expensive development tools, custom programmers, etc.
I am confident that 8-bit processors will never die, they can always be optimized for low power usage better than a 32-bit. The days of ASIC are more likely numbered. Consider if Intel were to get Atmel, they could put an x86, AVR and FPGA on the same chip. The AVR could do things with very little power (e.g. customized FPGA loading) while the hulking x86 was sleeping quietly. Intel apparently resurrected an ancient x86 and called it a Quark to do this, but it is not nearly low enough power (nor is it small).
Microchip has not been able to get there older PIC (I think of it more as a POS nowadays, but I was once a fanboy) devices to work with GCC. I don't think they are unfriendly to beginners, it is just the tools that work with PIC are. Note PIC32 is an MIPS and does work with GCC, but it has clearly been a struggle for them. Perhaps the explosion of ESP8622 has clarified the importance of GCC for even the hard headed (I'm in that crowd).
jremington:
People have been saying that for 20 years.
The point I was making with respect to standalone micro-controllers is that 8 bit days are limited once 32 bit is the same cost
It is only a recent phenomenon that 32 bit micro-controllers can compete with 8 bit parts in terms of cost and power.
If 32 bit is the same cost and offers the needed power capabilities, why would anyone continue stick with 8 bit?
But for sure we are already seeing that for mass production products, stand alone processors are no longer being used and the definition of what unit quantity "mass production" determines going full custom is falling all the time.
I'm more worried about Microchip's decidedly unfriendly approach toward beginners and experimenters: expensive development tools, custom programmers, etc.
In terms of "expensive" development tools and being unfriendly to experimenters goes, there are low cost options for the pic products available to hobbyists.
While Microchip does have their own tools which they want ridiculous amounts of money for, I've never used them since the gnu tools are available for no cost.
If you really want to, you can also modify some of their tools from the open source that they have to provide, (since some of them are using the gnu tools) to rebuild them yourself so you don't have to purchase pre-built ones from them.
The latest PicKit which can be used for source level debugging is similar but a bit lower cost than the Atmel ICE. (around $50 USD)
No its not as low cost as a $2 USBasp device, but the PicKit and the ICE include source level debugging capabilities.
Microchip does have low cost 32 bit DIP based parts available like the
PIC32MX250F128B which can be purchased at low cost in single quantities (like 1 if you want) at less than $4 USD from their microchip direct site:
http://www.microchipdirect.com/ProductSearch.aspx?Keywords=TCHIP-USB-MX250F128B
At least for the pic32, it seems to be fairly friendly to hobbyists and I've always wondered why the Arduino guys never added support for pic32 to Arduino. (other than they did a deal with Atmel and came up with DUE)
There is already Arduino support available for the pic32 parts, available from the chipkit guys.
To me a great product would be a "uno" type board based on the PIC32MX250F128B.
I think for many users it would be better than the DUE given it has 5v tolerance so it can play in both 5v and 3v worlds - something that DUE can not do.
That chip as well as a few others come in a 28 pin DIP so it would be way more hobby friendly than the DUE.
I guess the point I'm making is I don't see Atmel being acquired or going away (even if absorbed by Microchip) as all doom and gloom for the beginner and hobby market even if Microchip completely eliminated the 8 bit AVR parts.
--- bill
ron_sutherland:
The days of ASIC are more likely numbered.
Perhaps your definition of "ASIC" is different than mine.
I'm referring to Full custom ASICs, with integrated processor cores and in some cases even mixed mode with analog sections well. This is what all the phone makers are doing, including Apple with its Ax "processors" as it is the only way to crunch everything down into the small space available.
Plus with full custom, you get to customize to your needs which reduces the overall cost and power consumption - it becomes a value proposition as to will you have enough volume to justify and recover your upfront costs for the ASIC development.
I do agree with you about the older PICs vs the pic32 - which uses the mips core.
I wouldn't touch that old stuff (at least from a hobbyist project perspective) since it can't run run gcc.
I'm more worried about Microchip's decidedly unfriendly approach toward beginners and experimenters: expensive development tools, custom programmers, etc.
Huh? Microchip STARTED the hobbyist-friendly microcontroller industry. Parallel port programmers, "ponyprog", and the PIC16C84 were wowing hobbyists while Atmel's flash microcontrollers (with the dubious distinction of being "first") were still next to unobtainable. The PIC-based "Basic Stamp" occupied the niche that Arduino has taken over for a LONG time, and the PIC-based PICAXE is still a viable competitor. Microchip samples were an online form away for nearly anyone, back while Atmel samples were nearly unheard of.
While there has never been a great "free" C compiler for PICs, there have been quite a few low-cost and/or limited size compilers in C, BASIC, and other useful languages. (the lack of a good free C compiler has been the PIC's big weakness...)
While the 8-bit pics are sort of gross, architecturally, the 16bit PICs are a lot nicer, and the 32bit PICs are based on the industry standard MIPS architecture. You can get a Microstick2 that is logically similar to an Arduino zero (on-board debug support, 32bit cpu with 128k flash and 32k RAM), but it's only $35 and includes 3 other microcontroller chips.
And there are the ChipKits and MPIDE, that give you an open source compiler/libraries and Arduino-compatible core and libraries for the PIC32s.
I don't know if a Microchip/Atmel merge is a good thing or not. But this particular complaint is off the mark...
The PIC-based "Basic Stamp" occupied the niche that Arduino has taken over for a LONG time, and the PIC-based PICAXE is still a viable competitor.
Oh come on. Those were developed by companies who used PICs but had otherwise nothing at all to do with Microchip. And, they had to write their own serial bootloaders, so that hobbyists could easily program them.
I've had 20+ years experience with PIC products and until recently, all the "student oriented" free C compiler offerings were so crippled as to be nearly useless. My first programmer was the PicStartPlus, which I recall cost over $200 (~ two decades ago). Programmers for the latest 12F and 16F chips, like the Pickit3, are still $50.
I was SO relieved to discover Atmel, free avrgcc and inexpensive programmers.
until recently, all the "student oriented" free C compiler offerings were so crippled as to be nearly useless.
Until recently (ie until they acquired HiTech), Microchip didn't have a C compiler. The C compilers from 3rd parties (including Hi-tech, but I was using CC5x from Knudsen Data), were less crippled than Microchip's initial offerings after the acquisition. 16-bit and 32-bit PICs have always used a variation of gcc, so you can get/use uncrippled free compilers (especially for PIC32, because MIPS gcc is well-established.)
My first programmer was the PicStartPlus, which I recall cost over $200 (~ two decades ago). Programmers for the latest 12F and 16F chips, like the Pickit3, are still $50.
And what was the cheapest programmer/debugger you could get from Atmel 20 years ago? An stk-500, at about the same price? (Interesting, some vendors still sell both of these, and they're still about the same price!)
Meanwhile, the latest Atmel (official) programmers are also "about $50" (Atmel ICE "basic"), and "homebrew"/eBay programmers are essentially equally cheap (ponyprog vs parallel port programmer (both useless, these days), usbasp vs pickit2 clones (the pickit2 is more expensive ($10 vs $5), but it also does more.)
There was even the "USB bitwhacker" at roughly the same time as the original USB Arduinos, which didn't catch on as well mostly because (IMO) the people behind it didn't have the vision to recognize their audience.
westfw:
The PIC-based "Basic Stamp" occupied the niche that Arduino has taken over for a LONG time, and the PIC-based PICAXE is still a viable competitor.
Hmmm.
| BS2 | Arduino Mini |
|---|---|
 |
 |
Heh. I never noticed the BS2/Mini pinout similarities before.
For better or worse, Arduino did NOT particularly copy the BS language primitives...
Not just similarities - it was clearly deliberate (as has been mentioned here before).
Somewhere here is a BS1 (in its original box, no doubt), but I must confess, I never really did anything serious with it, certainly not compared to the Arduino. Mind you though, many years in between ...
Tuesday it [Atmel] plans to end its merger agreement with the British company [Dialog] and accept the Microchip offer unless Dialog raises its bid for Atmel by 4 p.m. Tuesday. Atmel must pay Dialog $137.3 million if it ends the merger agreement.
pito:
Atmel gets sweeter offer from Microchip; no word on Colorado Springs plant's fate
When it comes to a merger/buy-out deal. Cash is king. The Microchip deal is $7 cash + $1.15 in stock whereas the Dialog deal was stock.
Executives don't make their money on salary. They make it on the stock appreciation since they award themselves stock instead of pay because the cap gains tax rate is so much lower (like half) the tax rate of income.
There are typically trigger clauses on unvested options for executives based on a change of control so all their unvested options typically mature once the deal is completed.
What a cash buyout does is allow an executive to instantly convert ALL his shares including his unvested options into immediate cash without having to wait the option period. These agreements can be re-done as part of the deal but seldom are.
So even the Microchip and and Dialog were the same dollar value, Microchip would win because it has such a large cash component which is a better deal for executives, particularly those that will be eliminated or want to leave after the acquisition.
To retain critical executives, acquiring entities typically issue new/additional options in the acquiring company.
What we see in this case is a bit strange in that it appears the CEO announced his resignation in May 2015, for Aug 2015, but continued to remain in control. He also immediately dumped a large slug of shares just after his resignation announcement back in May 2015.
Since he appears to still be with the company he has retained all the unvested options he received as CEO even though technically he is not really the CEO, and I'm guessing that since he is still acting CEO, he will ensure that all the unvested options (at least his) get vested and turn to cash as part of the deal.
Had he actually resigned and left the company, he would have lost all the unvested options- typically 30-90 days after departing from the company.
Ah.... the games that CEOs play to enrichen themselves.
--- bill
Comment: A skeptic's view of the Microchip/Atmel merger
Bernie Cole, 10/3/2008 EETimes
With this history as a guide, if Microchip's acquisition of Atmel is successful, Microchip will have to make choices about which architectures to support fully and which to either let die for lack of support or sell off to interested buyers.
Based on watching electronics companies try to merge and manage two different product lines over the years, I don't think Microchip's plans for supporting combined 8/16- and 32-bit MCU product lines, if its attempted acquisition of Atmel succeeds, is realistic.
History suggests that a company with two 8/16-bit microcontroller architectures and at least two different 32-bit CPU architectures will not survive very long. One of three outcomes is likely: the company will crash and burn in the effort, one or the other architecture will go into the dustbin of history, or the company will sell off one of them, the way Microchip plans to sell Atmel's nonvolatile memory and RF and automotive businesses to On Semiconductor.
A look at the industry's history reveals that trying to support more than one architecture just does not work. In the recent past, Intel Corp. was juggling three different architectures, targeting three different markets: the DEC StrongARM-derived PXA2xx targeting mobile and consumer designs, the IXP4xx family of network processors it also acquired from DEC, and the X86 architecture targeted at desktops, laptops and servers. What we have now is a single architecture from Intel, targeting all these segments.
Further back there is Advanced Micro Devices, which in the 1990s had to make a choice between supporting its X86 strategy in desktops and servers and its internally developed 29000 RISC architecture, which at the time had a strangle-hold on the high-end printer market.
More recently AMD tried to juggle two different efforts, its X86 in desktops and servers and the MIPS-based architecture acquired from Alchemy for mobile and consumer devices. It is now focused on a single X86 strategy.
In the 32-bit arena Atmel will have to make a choice between two different RISC architectures, the MIPS-based one it uses in its new PIC32 and the ARM-core based 32-bit family that Atmel has proliferated throughout the embedded market.
The MIPS architecture is the one EE and CS students read about in their text books. On the other hand, the ARM RISC is widely supported and licensed to virtually every serious player in the market. And while the MIPS architecture has an impressive repertoire of development tool support, it can't compare with the support available for the ARM. It will be a tough choice, but at some point Microchip will have to make a decision as to which to support totally.
Microchip faces the same problem in the 8/16 bit microcontroller business. Despite the fact that Microchip is No. 3 in the MCU market with its PIC family, Atmel's AVR family is a well-respected RISC MCU architecture, especially in Europe and in certain market segments such as automotive and industrial. There is no way the two architectures will coexist within the same company.
History again tells us that it just doesn't work. Intel, when it was a serious player in the microcontroller business, had two separate architectures, the 8/16-bit 8048/8051, now discontinued by the company but second-sourced by eight to ten other manufacturers, and the 16-bit 80196, which now survives only within the company as a specialized CAN communications controller.
The same story is repeated at Motorola and STMicroelectronics. Both at one point had multiple MCU families and each now supports only one.
With this history as a guide, if Microchip's acquisition of Atmel is successful, Microchip will have to make choices about which architectures to support fully and which to either let die for lack of support or sell off to interested buyers.
However, insofar as we believe that many of the chips we actually buy are "clones", it will not necessarily matter if Atmel were to cease production!