No, i was not quoting a server situation regarding the memory tests., i don't need to
put on a defense against intel in server situations as Intel routinely digs themself
into a hole. I mentioned server arch only because that's where all the shiny desktop
hardware is usually derived from. AMD has more often than not come out ahead in
memory performance (by which i mean, real world memory performance) going back to the
barton core. Intel got a slight lead when they had thier 3.6ghz chips with quad
pumped 800mhz FSB going up against AMD's 1.8ghz - 2.2ghz with 333mhz -400mhz FSB. But
that was quickly fixed with AMD 64's launch.
Btw "Mr.Factboy" I actually pointed out intel's adventures into designing an IMC.
But, the way you say "who's coping who" seems to indicate you think you've achieved
some sort of victory in insinuating that AMD copied Intels IMC. IF that is in fact
the case...then your title of Factboy is just....well it's about as acurate and
truthful as Intel's marketing and bussiness practices.
See, what happens is one company comes up with an idea. They see the idea through,
put a product on the market that recieves favorable feedback and makes them a tidy
profit. This having happend a rival company takes notice, and thinks "well that would
have been a great move for us....if only we had thought of it first."
But of course they hadn't thought of it first. So instead the rival company gets a
new idea. "well, if it worked so well for those guys, why try and fix something that
isn't broken" They then proceed to copy the idea, almost exactly. They leave a few
minor differences in thier design, so they can slap their logo on it and get a
patent, in hopes of getting similar praise and financial gain.
But in order for that type of thing to happen, the first guys have to have an idea
that...ya know...works. So, you see as AMD's IMC not only worked but gave tramendous
performance gains over not only the competition, but their own previous product line;
Intel failed, not once but TWICE.
The more commonly know Intel chip with IMC was the Timna which was announced in 99,
and set to be launced in summer of 2000 i belive. But as the design was based around
using Rambus RDRAM which never actually became affordable they had to scrap the
design and rework it for use with SDRAM, pushing the launch back several times to
final confirmed spring of 2001. But this was around the time that intel was getting
ready to launch another industry milestone....netburst-ing P4. So it got pushed back
again....along with the P4. Then of course when a test batch was rolled out...there
was a pesky fatal design flaw. Which meant they would have to go and do more work,
when really it wasn't worth putting so much effort into a cpu for a $500-$600
desktop, because thats not where the market was at, the consumers wanted to have to
pay more money, not less! So they scrapped it. Hmm...i'm sure that's where AMD got
the idea...the twice failed thrice delayed chip that never actually made it to
launch.
Of course there is also the 386SL and 486SL. Those chips were launched with an IMC in
early '91. Back when those 20Mhz speeds were blazing. The chip was delayed several
times, discovered to have a manufacturing flaw preventing a large number of the chips
from running 32bit applications....ironically this was at the same time Intel lost
it's monopoly on 32bit arch, Despite AMD being little more than a nusince at the
time, it was evident to everyone that the SL chips were trash and no one bought
them...Intel cancled production in '92. But hey, maybe that's where AMD got the idea
from....
But then again...the EV7 was the first high performance chip to have an IMC. Hmmm.
Who designed that...
Oh yes, it was the Alpha design team from DEC. The same group of engineers that went
to AMD and designed the Opteron/939 chips...along with the socket A chips. But i
guess that makes it quite clear the only possible source of inspiration for the IMC
was Intels wildly successful chips.
Fun fact though, the EV8, another of those Alpha designs, was actually the first
processor with multi-threading. Was ready to go to production long before the P4, but
got cancled when DEC was being passed around. But i wonder where intel got that idea.
Of course there is also the fact that near every sneak-peek at Nehalem has noted that the IMC implimentation is startlingly similar to AMD's IMC design. So the IMC, the x86-64 arch, the low power high performance, the quad core on a single die, all AMD designs and ideas that have been around going back 5 years...but i'm sure intel just had a moment of clarity.
Also it's nice to see the consistency in the Nehalem benches...intels latest and greatest 64bit quad core...benched in 32bit applictions and coming soon! Now the litte comparison below, is why i get annoyed at a lack of 64bit benches. I was reading the Nehalem' review on Anandtech, and noticed the 2.66 quad was running in 32bit version of cinebench. So just to get an idea i clocked my 9850 BE down to 2.7ghz, and compared the 32bit and 64bit results.
64Bit 32bit
1cpu= 2874 2245 21.88% performance gain
xCpu= 10550 8230 21.99% performance gain
speedup= 3.67 3.67
openGL 4564 4568
Incidently, the 2.66ghz Nehalem got 3048 in single core, and 12,400 i think running all 4 cores. I realize it's based off of results running on a unsupported x58 board...but seeing as how cinebench actually takes advantage of IMC 5.7% lead on a single core and 15% running 4 cores really isn't too impressive for a 45nm chip with twice the on die Cache....while running 1366mhz DDR3.
Oh regarding my previous post, i have a somewhat unfair advantage when it comes to the details about who did what when and how the differences in how the chips handle code. That last post was what i found out from talking to my uncle, who aside from having a masters in micro-processor engineering, was a Vice president at Compaq from 94-2001. So aside from him actually having designed and physicall made computer processors, a gpu, and a couple raid cards he worked with the Alpha guys while he was at compaq, was in the room during those meetings at intel in 2005 as part of the team to get them to a point of making something to compete with AMD again(which depresses me to no end) he's spent the last 25-30 years or so working with people like Dave cutler, Steve Ballmer, Bill Gates, Michael dell. and even Ross Perot before he sold EDS.
Though the man usually tell me little about upcoming hardware beyond "sorry, had to sign an NDS" and then proceed to enjoy taunting me with silence, once it's public knowledge he can give better explinations about the stuff we come and bicker about on forums like this, than any of the hardware reviewers the hardware enthused tend to frequent. A fair amount of that has to do with the fact that i know he has nothing to gain unlike internet sources, or those actually affilated with either company; aside from the fact that he gets to play with the hardware while it's still being developed.
Benchmarks, and company preferences can be bickered about till the end of time.
So the last post, wasn't an opinion, or an interpertation or even something to be argued about, short of someone else being privy to the same group of coworkers. It's just fact.
Facebook
Twitter
Instagram