Blog Review: April 21
Napkins or beer? For most of us, that would be an easy choice. Napkins, of course. You can’t draw a business model in beer. But in England, where tradition is very important, the answer to making Java run faster apparently was a couple pints of good ale. Check out the blog from ARM’s Rod Crawford.
Cadence’s Richard Goering takes a look at 3D stacking and offers a reality check. There seem to be a lot of reality checks going on with this approach. But reality may depend on an individual company’s interpretation of 3D—and its budget. This is a hot topic these days, and this is a great look at why 3D is important and where the hurdles are.
Mentor’s Tom Fitzpatrick raises an interesting question about why Accellera’s technical subcommittee decided to move so quickly on OVM 2 being the basis of UVM. It’s a good question and one that needs to be asked, particularly given the amount of opposition from the VMM camp. So here’s our prediction: Either this whole issue will fade away or it will get worse, but it doesn’t look like there’s a middle ground anymore.
Case in point: There were three blogs posted on VMM central within a week, two from Verilab’s Jason Sprott on VMM Factory (Part I and Part II) and one from John Aynsley about using TLM with VMM 1.2. This doesn’t sound like a group that is handing the keys to the kingdom over to Accellera.
ARM’s Kerry McGuire wonders if tablet computers, aka the iPad, will slice out the market for netbooks. It’s a good question, but the answer may vary by geography. Netbooks are still cheap, even if they are mostly underpowered PCs. But battery life seems to be less of a differentiator these days. A Macbook will last 10 hours, and Dell has introduced a PC with an Intel and an ARM chip (something McGuire doesn’t mention).
Si2’s Steve Schulz digs into why co-opetition doesn’t always work. Hint: It has something to do with the number of complex tools that are intertwined and interdependent.
Daniel Nenni looks at the future of Moore’s Law. While the number of transistors continues to double every couple years, the number of problems associated with Moore’s Law is rising even faster. Food for thought.
Mentor’s Colin Walls continues digging into the multicore, multiprocessing history with a look at what’s really changed. Not much, except that it really has gotten smaller, faster and cheaper. The interconnects are better, too.
Synopsys’ Eric Huang has put to rest a mystery we didn’t even know existed. Was the inventor of the USB actually in an Intel commercial? Short answer: No.
Tags: ARM, Cadence, Mentor Graphics, Si2, Synopsys
April 21st, 2010 at 12:32 pm
Unfortunately, you missed the point of my post. OVM was clearly chosen over VMM for a number of reasons. The question I was asking was why the TSC was stuck on OVM2.0.3 until last week. More importantly, why were my suggestions to use OVM2.1 shouted down right up until the TSC chose to move to 2.1?
There are no discussions, nor are there likely to be, about the TSC changing to VMM. OVM has clearly won. Synopsys can continue paying consultants to blog about VMM1.2 in an attempt to make it appear relevant, but that doesn’t change the fact that OVM was chosen as the clear favorite by a group of industry experts, including Synopsys and Verilab.