Interesting article about new micro without onboard FLASH and the reasons why...
Cluso99
Posts: 18,069
in Propeller 2
Here is an extract...
eejournal.com/article/nxps-new-i-mx-rt1050-gets-edgy/
Maybe the P2 will not be the only micro without internal flash after all.
BTW Note the price... ~$3
For the full article...What it doesn’t have is any flash memory. This was, obviously, a deliberate decision on the part of NXP’s design team, and I got the sense that the company is a bit defensive about it. Certainly, most other MCUs have on-chip flash. Indeed, the Semiconductor Industry Association (SIA) treats on-chip flash as the very definition of a microcontroller; without it you’re a CPU, not an MCU. Great quantities of on-chip flash have become de rigueur for microcontrollers for many years. Why leave it off?
Cost and speed, to name two reasons. Flash manufacturing processes are a little different from pure-logic processes, so you wind up compromising processor speed and/or flash capacity when you mix the two. Flash-capable fab processes also cost more. A pure-logic process (in this case, its 40nm LP process) is well amortized, reliable, and cheap. That’s how you get a Cortex-M7 to run at 600 MHz. The company is already hinting strongly that 1-GHz versions aren’t far behind, once it jumps to the 28-nm node.
eejournal.com/article/nxps-new-i-mx-rt1050-gets-edgy/
Maybe the P2 will not be the only micro without internal flash after all.
BTW Note the price... ~$3
Comments
Back in the day we had big DIP micro-processors that needed big DIP RAM and ROM and peripherals and glue logic to get them working. A whole huge board full of DIPs.
Then came the micro-controllers that put all the RAM and ROM etc into one big DIP packages as a stand alone compute device. Think 8051. Great. If you only need something very small.
Today that non-volatile store is a teeny weeny chip on the side. Many ARM SoCs do not have FLASH on chip. Think Raspberry Pi for example. Many FPGAs do not have their non-volatile configuration on chip.
The Propeller was obviously ahead of its time
It does not matter in the vast majority of P2 targets.
There are plenty of MCUs out there without FLASH, it does not take much effort to find them.
* Cypress FX2 parts are quite old now, and they deliver HS-USB with a Loaded-RAM Code scheme.
* NXP have done FLASH-less versions of MCUs for close to a decade
* ESP8266 & ESP32 are Flash-less.
* XMOS is Flash-less
* Analog devices chose to retire FLASH DSPs some years ago, (flash just not able to deliver the speed needed)
Then there are parts that merely Boot-Flash, not run-from-flash - can that still be called 'on board flash' ?
* NUC505 has an in-package SPI memory, boots into RAM
* FTDI's new FT51, FT9xx series, boot into RAM
Which is why parts that push the speed envelope, often avoid execute-from-flash, it's just too slow...
With Serial Flash parts now sub 20c, designing parts to talk-better with external flash is the new trend.
JEDEC have recently released a xSPI standard, that may be beyond P2's fully supporting.
(P2 has no native DDR support)
While I am disappointed, there seems to be some reasoning behind it in some areas, which is why I pointed out the article. The ARM M7 is on 40nm. Perhaps cheap FLASH is not there yet. Cheap FLASH is currently being produced on 180nm and 90nm IIRC. This is why I expect OnSemi will have it available for their ONC18 process.
Most of the small and cheap micros have flash on chip. I know they have the huge volumes to support it. Same applies to onchip oscillators.
But, FWIW, an extra $0.20 in BOM cost does not translate to $0.20 in board cost. It's more like $1.00. Same goes for resistors and capacitors. There is a handling fee for assembly, inspection and testing that far outweigh the physical cost of small components.
ESP do make a chip with onboard flash. IIRC its something like ESP8285 or ESP8286.
Ah, ok, so that ESP8285 slots into the Boot-Flash, same as NUC505, which also has in-package SPI memory.
That's something P2 might be able to do, depending on what OnSemi says...
Needs bonding pad allowances, but maybe they do not need to be full ESD designs...