• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD's 3GHz K10 to break 30,000 3DMark06

To be fair, the 'Concept' of pipes is a fair analogy of a cpu's pipelines, with the length representing the pipelines depth, and the diameter representing the issue width of the cores.

However Hyperthreading is about making use of the nature of many programs being linear and often unable to make use of a wide core. And the wider the core, the more chance of underutilization.
 
Justintime said:
Short fat pipes? Long narrow pipes? huh? You know cpus don't actually have pipes in them right? :p AMD invented HyperTransport, not HyperThreading and its royalty free. Seriously would like to see your evidence on this. And as Corasik said,it was already embedded in the Williamette cpu from start, just not fully workable yet until later Northwood cpus. I do believe you got this as inside information though (inside your head ;) ).
Click to expand...

AMD Invented and then later PATENTED Hyperthreading in 2002 FACT, but as with some other CPU techs they have agreements to share, niether INTEL or AMD own the rights to the X86, evedence is Google (be it long ago) or the guy who used to actaully WORK for AMD over in ym IRC hang out (#winbeta), believe that guy over anyone as what he has told in past and not just about AMD, also MS stuff, Sata 150/300/600, EFI (replaces Eprom) Longhorn all before we had heard of elsewhere, ;)

And AMD trying to gety Reverse Hyperthreading to work aint a myth, never did say it did work but they did work on it, and it would be amazing if it did work.
 
Last edited:
Ok this is 1 source, Google came up with immed, remember this news is years old.

" AMD patent could enable hyperthreading "

"Apparently, way back in 1999 AMD filed a patent which gives them rights to release a processor with hyper-threading technology. Entitled "Microprocessor configured to execute multiple threads including interrupt service routines", the patent outlines how a CPU can execute multiple threads concurrently.

So, its not outwith the bounds of possibility that we could see an hyper-threading enabled Athlon64 processor in the works for some time in 2003. Certainly, for AMD to remain as competitive with Intel as possible, it would be a smart move. "



Filed in 1999, 2002 did come into it probably when it was all done and dusted and the posts of WWW appeared.


http://www.techspot.com/news/3556-amd-patent-could-enable-hyperthreading.html
 
Only people i ever heard suing Intel over HT was MicroUnity Systems Engineering and BIAX. Though Intel brought HT to the consumer cpu market similar systems were in use a long time ago before the AMD patent as well in IBM mainframe architectures, so because AMd patented something 'like' HT dosen't mean they invented it as you claim. Also Intels HT roots can be traced way back to Digital Equipment Corp (DEC).
 
Last edited:
And your info was still wrong, it took me 2secs in Google to find that info, so get over it and learn to look yourself next time before condemning a post. ;)

I guessed they invented it if they wanted to patent it, not the normal to patent something thats publically known to be invented by another.

I knew even after I posted it, there would be some "IF" or "BUT" to the reply.
 
helmutcheese said:
I guessed they invented it if they wanted to patent it, not the normal to patent something thats publically known to be invented by another.

.
Click to expand...

They did not invent it. Can you not get that HT systems were not first used or invented by AMD? They might have filed a patent to some similar technology but not the HT that Intel, whom btw has patented theirs, use. The premise of SMT was there a long time before either company experimented with it for retail chips and AMD might've filed a patent for some technology that was similar, but not the one Intel used in their chips which is thier own version of something already in use.
 
I never did say for a fact they invented, I assumed as they wanted to patent it, as you would see if you read my last post. :rolleyes:

I DID say for a fact that they PATENTED it, you did not want to believe this and were too lazy or narrow minded to even bother to use Google.

You are hitting back with small no real valid points simply as you were wrong, topic is actually off topic and is over for me, good day. :)
 
helmutcheese said:
I never did say for a fact they invented, I assumed as they wanted to patent it, as you would see if you read my last post. :rolleyes:
I DID say for a fact that they PATENTED it, you did not want to believe this and were too lazy or narrow minded to even bother to use Google.
Click to expand...

helmutcheese said:
AMD Invented and then later PATENTED Hyperthreading in 2002 FACT
Click to expand...
:p
 
Hey PinkFloyd, I learn every day, its nice to pass on some to others. :D

I really hope the K10 is that good, remember just before AMD 1st 64 release, the press was all doom and gloom and end of AMD was near.
 
CPU speeds don't count for an awful lot in 3DMark06 in my experience.. I get a "massive" ~600 points boost going from a 3.6Ghz C2D to 4.3Ghz.

Assuming that an Xfire HD2900 XT system with a regular C2D at any clockspeed would probably be in the region of 20k (a bit of a guesstimate), are they seriously suggesting that the extra 10k is due to the CPU?
 
September 10th or 11th. I think thier for servers not sure, but whatever you do dont get the b1 version. They suck and need a lot of voltage and dont clock very high. The b2 version is very good and the one to go for.
 
You must log in or register to reply here.
Back
Top Bottom