lol...
Runners up for that name was the Z28, because Intel believes 99% of its CPU user base is a fan of the Camero Z28 and... it sounds fast!
i think i got it!
i7 = i-seven --> ise-even --> ice heaven!
they want to hint its running very cool!
No?...well...i tried...
i7 = i-seven --> ise-even --> ice heaven!
they want to hint its running very cool!
No?...well...i tried...
Well there are 8 threads for Bllomfield...which would be numbered 0 - 7 in geeky computer numerology.
So perhaps a 2core/4thread Havendale will be referred to as i3?
And an 8core/16thread Beckton will be dubbed i15?
(I'm not serious)
Come to think of it, perhaps they chose an odd number specifically so it couldn't be confused with the # of cores in the product. AMD's 3-core chips slightly muddle this, but those are a largely failed product that rather few people know about.
/shrug
/shrug
videogames101
Diamond Member
- Aug 24, 2005
- 6,783
- 27
- 91
I like the name. Very simple and different. It actually really stands out and will be easier to market.
alkalinetaupehat
Senior member
- Mar 3, 2008
- 839
- 0
- 0
IIRC, many a snow cone was smashed against the car of the man at Intel who struck down the Core Z71 Edition. Offroad truck = All-terrain CPU?
I immediately thought of the 7 as denoting the number of the last virtual core in the OS. Although I couldn't for the life of me figure out why they would do that.
Anyways, Core i7 sounds dumb. And they might just get sued by Apple (so might the Islamic Republic of iRan, one of these days). Why not just call it Core Pro or something? They already made a mistake by putting the word "core" and a numeral in there, and i7 simply continues that tradition. No one ever thought the "Pentium 4" had 4 cores in it (although people confused by the marketing hype around HT DID think it had 2), so they really messed up when they named the uArch Core to begin with.
Anyways, Core i7 sounds dumb. And they might just get sued by Apple (so might the Islamic Republic of iRan, one of these days). Why not just call it Core Pro or something? They already made a mistake by putting the word "core" and a numeral in there, and i7 simply continues that tradition. No one ever thought the "Pentium 4" had 4 cores in it (although people confused by the marketing hype around HT DID think it had 2), so they really messed up when they named the uArch Core to begin with.
Has there ever been one? For any CPU, from any CPU manufacturer?
I love thread necros, always a blast to read the past.
TurtleCrusher
Lifer
- Apr 20, 2008
- 10,067
- 990
- 126
Comparing the architectural deltas between successive i#'s in that list and I don't see there being a sufficient delta between Core and Nehalem to warrant Nehalem being awarded its own distinct i# rating.
Going by what (little) we know of Intel's future architectures and ISA's, I'd say Sandy Bridge is probably the first (next) chip to contain a sufficient amount of changes to deserve a new iteration in the i# scheme.
Think about how many process nodes and ISA changes went into the evolution of the various flavors of the P5, from 60MHz to 1.4GHz, and they all get bundled together as the "i3" technology generation.
Nehalem is sweet, but its not ALL that different from Core when looking at this list.
Not trying to take away anything from what Nehalem is and does. I am just arguing over the lack of self-consistency in the list of i#'s I quoted. The list isn't the real rational of course for i7 existing.
Look at the i3 in that list. The P5 spanned from 1993 0.8um 60MHz Pentium to 2001 0.13um 1.4GHz Tualatin.
And for all the architecture and ISA changes that when into the P5 evolution over that timeframe, the author of the list of i#'s is saying "yeah, it's all i3".
Given this, I look to the core versus platform changes of Core and I see similar "minor" ISA improvements (SSE4.1/4.2) and architecture improvements from Kentsfield to Yorkfield to Nehalem so I'm not sure why Nehalem would suddenly merit the creation of its own distinct i#.
In the list i6 contains platform enhancements, faster FSB, larger L2$, DDR2->DDR3 transition...all i6. i5 contains products that did not have HT and products that introduced HT. So HT is not merit worthy of a new i# (per the logic within that fictitious list).
So looking at Nehalem, architecture enhancements to the "i6" are in-line with those that evolved over the P5 line, but P5 retained the i3 designation. Hyper-threading was reintroduced to the i6, as it was to the i5 but Netburst retained the i5 designation. Platform enhancements were introduced to the i6 (faster FSB, DDR3, larger L2$) and with nehalem the IMC moved to the CPU...still inline with an evolving i6.
I am not making the argument that Nehalem is not a vastly superior product to Yorkfield, out of boredom I am being pithy over the application of self-consistency on a fictitious list that has never existed in reality. I am merely saying if i3, i5, and i6 span the architecture, ISA, and platform enhancements as indicated then Nehalem is an i6. Sandy Bridge with AVX might contain enough ISA changes to merit the creation of a new i# classification.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter