Hmm... some big moves are in the making in the EDA business - Mentor needed to attach itself to a key piece of ESL technology while more or less disconnecting itself from its own faltering CatC, just to be able to be a player in these coming moves (that may or may not happen). Good move by Mentor, and if it gives Calypto some deeper pockets, good for them too. God knows Calypto has burned through a mountain of cash.
By the way, your colleague Max has a good opinion piece on the EE Times blogs with an interesting comment from Brian Bailey ...
In this deal, Mentor's equity from the Catapult C group and existing minority investment in Calypto yields two seats on the board of directors. So it appears that MGC has a controlling stake in an otherwise independent Calypto.
On a practical level, this frees the Catapult C team to interact with other ESL partners at will. At Mentor, they were part of the same organization that housed Vista and Precision.
I understand from talking to Calypto's CEO that the company will continue to support all of the tools it supports now. I understand your concern about the motivation for doing so, but the CEO went out of his way to emphasize this point (probably to head off any concerns such as yours).
"Aitelli emphasized that Calypto remains an independent company, supporting many different tools and industry-wide interoperability. "This doesn't change our outward stance on how we work with other parts of the industry," Aitelli said."
What's to become of support for non-Catapult high-level synthesis tools? I don't see any motivation on Calypso's part to continue supporting Forte Design System's Cynthesizer or Cadence's C-to-S tools.
Interesting. Normally it's the big fish swallowing up the small ones thru M&A. This time it's a small fish eating up a small piece of a big fish.
I have seldomly seen anything like this in American business landscape. Spinoff is common in Asia but American companies always want to be bigger & bigger.
Blog Doing Math in FPGAs Tom Burke 18 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 ...