Blog Review: Dec. 23
By Ed Sperling
It may be the end of the year, but it certainly isn’t the end of the story. Blogging is alive and well as we try to put a really ugly year in the rearview mirror.
Mentor’s Tom Fitzpatrick takes a swipe at Synopsys’ VMM labeling. He says it should have been VMM 1.2, not VMM 2.0. We have the same problem with Web 2.0, Windows 7 and the rounding error for trans fat on food labels (less than 1 gram is not the same as zero.) It’s interesting that each time we report any dispute between the OVM and VMM camps we get chided for blowing this whole issue out of proportion. Seems we’re just gluttons for punishment. Read on, if you dare, but make sure you check out the cartoon.
And what’s your opinion of Brussels Sprouts? In one of the strangest encounters this year, a TSA agent asked Synopsys’ Karen Bartleson if the stalk was a weapon. That’s when you know you’ve been flying too much.
In the case of Aptina’s David Talbot, he hasn’t been flying much at all. So it came as a surprise when Denali thanked him for attending it’s DAC party. Talbot is based in the U.K. and hasn’t crossed the pond in several years. That’s just one of the many surprises from DAC (yes, it was last summer). Still, it’s worth checking out John Cooley’s rather raucous, totally subjective and somewhat controversial “best and worst” column.
Mentor’s Thomas Bollaert trumpets one of the findings in Cooley’s blog about Catapult C’s leadership in high-level synthesis, which should give you some idea of how this stuff gets misused. Synopsys wasn’t even in this market when DAC was held last summer, so it’s not surprising. The question is what the market will look like in two years as both of the big guys are battling with Forte in this space. Our guess is the market will be much bigger by then, but it also will be much more hotly contested.
Mentor’s Jon McDonald looks at how some engineers view ESL modeling, which is not always in the best light because the models aren’t always accurate.
And Cadence’s Steven Brown says don’t forget about all the stuff that Cadence has done in the ESL world. Normally you leave that to the balance sheet, but apparently he’s indignant that others are claiming what Cadence has already done.
BAE’s Daniel Nenni takes a look at the precipitous decline in semiconductor manufacturing jobs in the U.S. and what the trends look like going forward. While semi manufacturing isn’t exactly upbeat, the U.S. Labor Department is predicting big job growth. It’s just a question of what and where.
Mentor’s Colin Walls talks about portable electronic devices and how to make them more responsive to their owners, as in less boot time. Given the focus on powering down and multiple states, this is an important topic—for hardware and software engineers.
Virage Logic’s Mike Thompson talks about the new inflection point and the blurring of boundaries between microcontrollers and processors. Whenever someone mentions inflection points you can substitute the word “war.”
Can the power intent of a design be automated? You’ll have to ask Synopsys’ Godwin Maben because it doesn’t look like he fully answers that question in his latest blog.
Si2’s Sumit Dasgupta looks at the idea of an open process design kit (PDK) from the foundries. We’ve been hearing a lot more about PDKs, lately, but an open PDK is an interesting concept. The jury is still out whether it’s a realistic one, though.
Also on the subject of standards, Cadence’s Richard Goering does a great job of pulling together all the standards efforts under way in 2010. This is a great piece of information to file away.
Synopsys’ Avinash Agrawal takes digs into VMM log details in waveforms. This is deep in the verification weeds, and great stuff for VMM fans.
December 23rd, 2009 at 10:51 am
You have it backwards. Tom said the VMM revision should have been 2.0 (due to so many new features) rather than the 1.2 they called it – which he claimed implied a small delta in functionality.