• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD demos Bulldozer at Investors conference

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
For once, I do agree with your sentiments completely. I would personally prefer if such detailed information were not kept in secret.

Unfortunately, "keeping things under their hats" has been a standard practice of companies in practically every industry the world over, so it is not likely to change unless their respective management groups get a change of heart - or a change of paradigm.


I don't need all the circut details or a mask or die shot. What is helpful is fully documenting all the cpu functions etc and how to program them.

I can't imagine we would ever get reliable automobiles with such a paradigm of nondisclosure.

Its rediculous and it needs to stop.
 
Talking about reliable automobiles...

The Problem with Threads
Edward A. Lee

http://www.eecs.berkeley.edu/Pubs/TechRpts/2006/EECS-2006-1.pdf

"This work was supported in part by the Center for Hybrid and Embedded
Software Systems (CHESS) at UC Berkeley, which receives support from
the National Science Foundation (NSF award No. CCR-0225610), the
State of California Micro Program, and the following companies: Agilent,
DGIST, General Motors, Hewlett Packard, Infineon, Microsoft, and Toyota."
 
Yeah its a very good document about why the linux and Ms way of threading isn't a good idea.

we should all just face the facts that computer designs are hitting walls and hitting them quickly.

Talking about reliable automobiles...

The Problem with Threads
Edward A. Lee

http://www.eecs.berkeley.edu/Pubs/TechRpts/2006/EECS-2006-1.pdf

"This work was supported in part by the Center for Hybrid and Embedded
Software Systems (CHESS) at UC Berkeley, which receives support from
the National Science Foundation (NSF award No. CCR-0225610), the
State of California Micro Program, and the following companies: Agilent,
DGIST, General Motors, Hewlett Packard, Infineon, Microsoft, and Toyota."
 
And for what BTW ? AMD and INTEL can both reverse engineer each other products well enough to figure out how to pick their enemy apart.

While I don't know if a modern CPU can be reverse engineered, I can tell you it's not done. There's really no point to it, it doesn't help your architecture to know the very fine details of how your competitor is doing something, because you wouldn't be able to use it anyway.

CPU's haven't been reverse engineered since the 8 it days.
 
While I don't know if a modern CPU can be reverse engineered, I can tell you it's not done. There's really no point to it, it doesn't help your architecture to know the very fine details of how your competitor is doing something, because you wouldn't be able to use it anyway.

CPU's haven't been reverse engineered since the 8 it days.


I don't think they reverse engineer them in full detail but I do think they take a look under the hood.
 
I don't think they reverse engineer them in full detail but I do think they take a look under the hood.

A full deprocessing workup is done, naturally, as a means to both benchmark internal progress/goals with that of the competition.

(Companies like Chipworks and Semiconductor Insights make an entire business model around it)

In addition to providing very valuable "sanity checks" on one's own internal development roadmap the deprocessing work-up can help determine whether or not a competitor is violating your IP as well as ensuring your internal IP roadmap is not on a trajectory that will land yourself in violation of their IP.

As far as the architecture goes, individual circuits are selectively targeted and characterized for the same IP considerations.

But yes, this is usually not done for sinister reverse-engineering reasons (to violate IP)...but I have seen more than my fairshare of this actually being done (reverse-engineering with intent to violate IP) on a number of levels at a number of companies (globally, it is not region or culture specific).

The sheer scale of complexity in a modern IC though, anything with a billion transistors, pretty much prevents brute-force reverse-engineering of the entire product though. Kinda like having a 2048bit key, that can give even the NSA a bit of trouble.
 
Hey Nem, looks like your wish comes true.

You show me FMA on desktop CPU and I can show you intels new Itanic.

http://forums.anandtech.com/showpost.php?p=27753830&postcount=22
:biggrin:

16c3p82.jpg
 
Back
Top