Teensy 2 users and the disconcerting sale of Atmel
- Thread starter stevech
- Start date
- Status
PaulStoffregen
Well-known member
Microchip has acquired other companies over the years, which is how they created such a diverse product line, especially the analog stuff. Discontinuing Atmel's successful products would definitely be out of character for them!
On thing I have noticed about Microchip, especially in talking with them and also (separately) Digilent at Maker Faire is their tendency for very tight control over their branding and trademarks. Rumor has it there's even an approval process required to publish almost anything with "ChipKit" in the name. It'll be interesting to see how they react to a hobbyist world filled with non-Atmel creations with names like AVRDUDE.
Regarding Teensy 2.0 and Teensy++ 2.0, just yesterday Robin put in purchase orders for enough parts to likely take us all the way through 2017. Assuming they're still willing to sell us the chips, you won't have to worry about those models running out anytime soon.
On thing I have noticed about Microchip, especially in talking with them and also (separately) Digilent at Maker Faire is their tendency for very tight control over their branding and trademarks. Rumor has it there's even an approval process required to publish almost anything with "ChipKit" in the name. It'll be interesting to see how they react to a hobbyist world filled with non-Atmel creations with names like AVRDUDE.
Regarding Teensy 2.0 and Teensy++ 2.0, just yesterday Robin put in purchase orders for enough parts to likely take us all the way through 2017. Assuming they're still willing to sell us the chips, you won't have to worry about those models running out anytime soon.
Constantin
Well-known member
Unfortunately, this sort of consolidation is a common trend in uC land. As fixed costs keep going up, so do the barriers to entry and the tendency to consolidate more and more. No different than the auto or aviation industry, for example.
That's not to say that you can't go it alone (for-hire, external fab houses make it possible) but your product has to have some pretty impressive margins, your customers have to be aware of and clamoring for the product, etc. Plus, buyouts tend to make management rich, ditching long term growth for short term profit.
That's not to say that you can't go it alone (for-hire, external fab houses make it possible) but your product has to have some pretty impressive margins, your customers have to be aware of and clamoring for the product, etc. Plus, buyouts tend to make management rich, ditching long term growth for short term profit.
Parallax and the Propeller is still around, but I guess there aren't too many other examples like that.
This greatly concerns me. My employer purchases thousands of PIC processors per month. And dealing with these people is sheer hell in both technical and logistical terms. Microchip reps have lied to me and other engineers several times about roadmaps, chip specs for new stuff, and availability. The coup de gras was when they shipped a batch of a new variation of PIC16s where some peripherals were inop. PICs now only go into the old (legacy) stuff - have not been used in new stuff for at least two years, but have been getting pressure from sr manglers to go back to mickeychips.
Do not see anything good coming out of this.
Time to standardize - between ST and FS(NXP) (and am not certain about continued trust in the latter). Or bite my tongue and go with TI...
Do not see anything good coming out of this.
Time to standardize - between ST and FS(NXP) (and am not certain about continued trust in the latter). Or bite my tongue and go with TI...
PaulStoffregen
Well-known member
Microchip blatantly lied to me too, in the late 1990s. I'm still bitter.
I will admit, I've used a couple of their analog chips and eeproms at times when I had little other choice. But I never want to use another PIC microcontroller!
I will admit, I've used a couple of their analog chips and eeproms at times when I had little other choice. But I never want to use another PIC microcontroller!
But with PIC8's, it's so much simpler with no stack pointer! And heavily paged memory. And no GCC
Last edited:
PaulStoffregen
Well-known member
Yup, only 33 instructions to learn!
Nothing difficult, like add or subtract with carry input!
Nothing difficult, like add or subtract with carry input!
Constantin
Well-known member
Long memories here!
My PIC8 tangent began before flash.. the window-topped DIPs had a UV port.
Crash-and-burn programming.
EEPROM UV eraser in constant use with rotating DIPs.
No C.
The absurd architecture of the PIC8 soured it with me.
It wouldn't even make a good controller for a kitchen blender. Oh, wait! Blenders didn't need more than buttons.
Crash-and-burn programming.
EEPROM UV eraser in constant use with rotating DIPs.
No C.
The absurd architecture of the PIC8 soured it with me.
It wouldn't even make a good controller for a kitchen blender. Oh, wait! Blenders didn't need more than buttons.
PaulStoffregen
Well-known member
In the mid 1990s, before and right around the time Atmel released their first AVR chip (the '1200 one without any peripherals), I designed a commercial product using the PIC chip. It was the only microcontroller on the market at the time with all the required peripherals and low enough power consumption and without the terrible logistics of masked rom.
By the time is was done, I'd written approximately 20000 lines of PIC assembly, running in a chip with 4 RAM and 4 EEPROM banks. The old architecture was really horrible!
By the time is was done, I'd written approximately 20000 lines of PIC assembly, running in a chip with 4 RAM and 4 EEPROM banks. The old architecture was really horrible!
- Status