6

Given that microcontrollers can be reprogrammed multiple times, does this apply to CPLDs as well? In particular, I am interested in Xilinx's CoolRunner-II. Until now, I was positive about the re-programability, but after looking at the specs at one of the stores, I got confused:

enter image description here

So, does it mean that I can program it only once?

Nazar
  • 3,172
  • 4
  • 36
  • 64

1 Answers1

9

No, and it probably illustrates the risk of looking at something that's been transcribed from a list or whatever rather than the original source.

The original datasheet you linked (primary source of information) lists the endurance at 1,000 erase/write cycles minimum (page 14).

Spehro Pefhany
  • 397,265
  • 22
  • 337
  • 893
  • That's a relief... Thank you. In general, re-probrammability is the known feature for all CPLDs, right? – Nazar May 30 '14 at 16:41
  • Nope. Actel (for one) has an antifuse technology that is one-time nonvolatile. You're perhaps unlikely to run into it unless you're doing military or space stuff. http://www.pldworld.com/_actel/html/digital.library/q1_2003/PDFs/af_PIB.pdf – Spehro Pefhany May 30 '14 at 16:47
  • One more thing, please. Just to make sure I understand the specsheet right: the VCC must be 1.8V only, while the IO voltages can be up to 3.3V? – Nazar May 30 '14 at 16:53
  • 4
    Go to the data sheet and look at "IO Banking". The answer is yes. – WhatRoughBeast May 30 '14 at 17:08
  • I wonder if "reprogrammability support" is talking about the ability of the chip to have one part of it reprogram another? A few programmable logic devices have such abilities; they can be very handy in some cases, though I've never had occasion to use one personally. – supercat May 30 '14 at 18:31
  • @supercat I would call that self-programming. Altera has some special devices ("HardCopy Stratix") that have "Reprogrammability Support" removed, presumably for security reasons. – Spehro Pefhany May 30 '14 at 18:40
  • Hmm... could the term perhaps be referring to the ability of the type to be unconditionally reprogrammed regardless of what's programmed into it, as distinct from the ability to program the device in such a way as to be unalterable? Being able to secure an alterable logic chip against "bricking" even by malicious code, and being able to secure a device against accidental alteration are both useful, but generally mutually exclusive unless a pin is reserved to control such functionality. – supercat May 30 '14 at 18:56