• 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.

Intel confirms existence of X86-64 Yamhill chip

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Originally posted by: jhu

not quite. ie don't expect an itanium clone from anyone in the foreseeable future.
Click to expand...

In theory AMD could if they wanted to. If the x86-64 ultimately fails and Intel has nothing but IA64 chips, then AMD would also develop IA64 chips.

Originally posted by: NightCrawler

The Pentium 3.2 is 1200mhz faster then the Athlon 64 3000+ and it doesn't beat it in benchmarks...but that inflated clock does market well.
Click to expand...

It doesnt matter if its 100Ghz faster. The simple fact of the matter is that it can scale that fast. High IPC doesnt do you any good if you cant scale to high clock speeds.

OTOH, while I was browsing around in Frys/Best Buy, a lot of people avoid high clocked Celerons (2.4-2.8Ghz) because even though its high clocked, the public still sees that as a value chip. Ask most poeple buying stuff in those stores, and they'd probably accept a P4-2Ghz over a Celeron 2.8Ghz. Reminds me of a client that wouldnt expect any celerons because he thought his P2-233 was better than a Celeron 300a@450.
 
omfg Venomous, the FX-51 at 2800Mhz! How did you do that! 😱

As for the liscensing and how they can use whatever they want to....then why didn't AMD include SSE2 instructions to the AXP line with the Bartons or something...

and devex I've noticed the same thing, but only when comparing Celeron/P4. I know people who will buy 2.6Ghz P4 over a 2.8Ghz Celeron, but they would opt for a 2.6Ghz Celeron when compared with a 2500+

Even though people are getting the hang of it, you cannot deny the fact that Inte'ls marketing is soo good that a superior processor is often dismissed because it isn't "Intel"
 
Originally posted by: magomago

As for the liscensing and how they can use whatever they want to....then why didn't AMD include SSE2 instructions to the AXP line with the Bartons or something...

and devex I've noticed the same thing, but only when comparing Celeron/P4. I know people who will buy 2.6Ghz P4 over a 2.8Ghz Celeron, but they would opt for a 2.6Ghz Celeron when compared with a 2500+

Even though people are getting the hang of it, you cannot deny the fact that Inte'ls marketing is soo good that a superior processor is often dismissed because it isn't "Intel"
Click to expand...

It takes time to develop something. When the P4 came out with SSE2, I doubt AMD could just patch SSE2 support in Tbred's. It took a core revision (Athlon64) to get it in. As for the X86-64, the standard was laid out several years ago.

AMD's marketting isnt as good. People just dont know the AMD brand. And startlingly, it depends on the salesmen.

Its kind of like me when I buy home appliances. I have no idea wtf is good, so I just buy the brand that I know (Kenmoore, GE, etc).
 
The report quotes Intel representative Robert Manetta as saying it has "a working prototype of a 64-bit X86 design that it could bring to market when customers request it".
To the best of our knowledge, if this remark is reported accurately, it is the first time Intel has publicly acknowledged the existence of an X86-64 alternative.
Click to expand...

Now I don't know about you guys but I very wary about Intel and x86-64... what i am actually looking for intel to do is sell a x86+IA64 chip, somewhat a dual core implementation that reuses parts where they can be between 32bit x86 and 64bit IA64.

Next comment. I sure hope to hell that x86-64 never really makes it. Not because I do not like AMD, but because there is a reason why Intel completely dropped x86 in their IA64 architecture, and it wasn't because they couldn't do what AMD did w/ x86-64.

Next, while everyone speculates Yahmill is x86-64 but never puts out the possibility of it being x86+IA64. I just think Intel would be marking the death to IA64 if they do x86-64 and I really don't think they would do such a thing. (Also there is a 64bit version of windows for IA64 so there isn't a problem there...)

Everything I just wrote was just my rambling and my feelings on the topic, and obviously I cannot back anything up.

Now if AMD's x86-64 is Yahmill, more props to AMD b/c they made the giant change their strategies, but i really believe that is the wrong road and that we should be looking toward a totally new architecture that IA64 provides...

Josh
 
Originally posted by: dexvx
Remember that well into the Pentium era (where machines had up to 128MB memory), the base memory was still 640KB; everything on top of that was just extended.
Click to expand...

That's how it is to this day, AFAIK.

 
Now if AMD's x86-64 is Yahmill, more props to AMD b/c they made the giant change their strategies, but i really believe that is the wrong road and that we should be looking toward a totally new architecture that IA64 provides...

that's pretty difficult given the amount of x86 code there is out there. if you think x86 is bad, take a look at ibm's current mainframes that have to be able to run binaries compiled in the '70s
 
Remember that well into the Pentium era (where machines had up to 128MB memory), the base memory was still 640KB; everything on top of that was just extended.

that's because during the boot process, the processor starts out in 16-bit mode. sun did make a few proprietary x86 machines in the late '80s/early '90s that likely booted straight into 32-bit mode (i could be wrong though).

but this is a moot point since most people don't use a 16-bit os anymore.
 
In theory AMD could if they wanted to. If the x86-64 ultimately fails and Intel has nothing but IA64 chips, then AMD would also develop IA64 chips.


there is no licensing agreement between amd and intel with regard to ia64.
 
I can only tell you that Intel was working on this chip over two years ago - my brother was one of the Engineers who worked on the production line for it. This work was in the theoretical phase - like how they would convert an existing line to produce this chip. I am not an engineer so I didn't bother asking anything more about it.
 
Originally posted by: SocrPlyr
Next comment. I sure hope to hell that x86-64 never really makes it. Not because I do not like AMD, but because there is a reason why Intel completely dropped x86 in their IA64 architecture, and it wasn't because they couldn't do what AMD did w/ x86-64.
Click to expand...
While no one is going to argue that x86 brings a lot of unwanted cruft with it, that's not to say that x86 doesn't have a lot of strengths too. IA64 is an EPIC arcitecture, meaning that to maximize performance, you need to be able to run things in a massively parallel configuration, a style not native to current programmers nor current applications. Even for future applications, a lot of stuff most consumers(and geeks alike) do is not really parallel; games, reading email, etc are either really simple in what needs to be done, or CPU intensive and serial(draw frame, update world, draw frame, etc); all of the situations result in sizable ineffeciencies in the EPIC design, and IA64 in turn. This isn't to say that IA64 wouldn't still "work" for most people, but it's silly to say that such a design is effecient for the consumer market compared to x86(-64).
 
Originally posted by: ViRGE
Originally posted by: SocrPlyr
Next comment. I sure hope to hell that x86-64 never really makes it. Not because I do not like AMD, but because there is a reason why Intel completely dropped x86 in their IA64 architecture, and it wasn't because they couldn't do what AMD did w/ x86-64.
Click to expand...
While no one is going to argue that x86 brings a lot of unwanted cruft with it, that's not to say that x86 doesn't have a lot of strengths too. IA64 is an EPIC arcitecture, meaning that to maximize performance, you need to be able to run things in a massively parallel configuration, a style not native to current programmers nor current applications. Even for future applications, a lot of stuff most consumers(and geeks alike) do is not really parallel; games, reading email, etc are either really simple in what needs to be done, or CPU intensive and serial(draw frame, update world, draw frame, etc); all of the situations result in sizable ineffeciencies in the EPIC design, and IA64 in turn. This isn't to say that IA64 wouldn't still "work" for most people, but it's silly to say that such a design is effecient for the consumer market compared to x86(-64).
Click to expand...
Hey Virge, it has been a while...
Good point. I guess it is really hard to say b/c what is better not running something w/ massive parallelism that was designed to or running something with a lot of excess baggage. I guess only time will tell.

Josh
 
IA64 is an EPIC arcitecture, meaning that to maximize performance, you need to be able to run things in a massively parallel configuration, a style not native to current programmers nor current applications.

that's true. however, intel's position is that the extraction of parallelism from compiled code be shifted from hardware to software, specifically the compiler. so from a programmer's perspective, he shouldn't really have to change the way he programs because the compiler is supposed to extract parallelism (unless he's the compiler programmer; what a fun job that should be). even today's compilers for x86 do that to some extent to take advantage of modern x86 cpus' superscalar and out-of-order design.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top