Xbit: TSMC vows to ramp up 28nm production in 2012,start manufacturing 20nm in 2013

[Lonbjerg]

I'm more interested in GFLOPS for what I do though.

I don't care what you do...especially when you seem fond of moving the goal posts fallacies.
You asked me (to overconfident)...and when the answers didn't suit you...you retorted to fallacies...says it all.

Have a nice day.

 

[jhu]

I don't care what you do...especially when you seem fond of moving the goal posts fallacies.
You asked me (to overconfident)...and when the answers didn't suit you...you retorted to fallacies...says it all.

Have a nice day.

I'm telling you what I want out of my computer, which is GFLOPS. Can an Octeon provide cheap GFLOPS within standard programming models the way x86 can?

You're an advocate for non-x86. While we're at it, you could theoretically use an Octeon processor for general computing purposes, but are you?

 

[Lonbjerg]

I'm telling you what I want out of my computer, which is GFLOPS. Can an Octeon provide cheap GFLOPS within standard programming models the way x86 can?

You're an advocate for non-x86. While we're at it, you could theoretically use an Octeon processor for general computing purposes, but are you?

I think I let your post stand...just as an example of how much one views can be misrepresented.

Kinda funny...but not construtive.

 

[jhu]

That graph is really sad...it's the part of the reason we are still stuck with x86 and all it's legacy crap and havn't moved on to a more future oriented ISA...without all the "baggage".

So, you still haven't elaborated on what "baggage" you want out, what you actually want from your processor, and how removing said "baggage" will help things out. What future oriented ISA would you want that doesn't break existing software? I can tell you now that ARM, POWER, and MIPS all have "legacy baggage" within their ISAs as well.