Intel Sandy Brdige Q4 2010?

ambam

Distinguished
Jun 5, 2010
1,166
0
19,290
I've read that the release date for Intel's "Sandy Bridge" is Q4 of 2010, or within the next Four months. Has Intel announced the actual name of this CPU? Have they dropped the "Core" prefix yet? The release model will only be quad-core, but in 2011 and 2012 there will be many 6 and 8 core models.

Intel as well as other hardware makers usually tease PC enthusiasts with advertisements and technical information about unreleased products shortly before they are released. But I have not seen or heard anything "official" from Intel about the Sandy Bridge, yet.
 
sandybridge_chart.png

from http://www.semiaccurate.com/2010/08/11/intels-sandy-bridge-cpu-roadmap-leaks/
 

eyefinity

Distinguished
Mar 6, 2010
1,106
0
19,310
Don't mistake intel shipping the chips for revenue with them being ready for general sale. Intel should start shipping them in late Q4, and systems with them should be available sometime in Q1 2011.
 


Its the same. Remember that when Conroe came out they were listed as Core 2 Duo/Quad E/Q 6XX0. Then the die shrink of Conroe was still Core 2 Duo/Quad E/Q but instead they went to the 9XX0 series and later went to 8X00 series as a low end. It only make sense to have Snady Bridge, the die shrink of Nehalem, to keep the same name scheem and numbers (Core i3/i5/i7) but change the numbers.

Its almost the same as AMD for Phenom, even though theirs was a tad confusing. The first Phenoms started at the 9XXX. Well Phenom II came out and started at 9XX instead. A lower number but second gen and better performance/clocks.

In all it makes sense. Now if they changed the name but kept the numbering scheme it would be kinda confusing.
 

jonpaul37

Distinguished
May 29, 2008
2,481
0
19,960


Only difference here is that it seems that intel is not going to allow these new Sandy Bridge CPU's to run on current 1156 and 1366 motherboards as the die shrink for core2/quad brands was still LGA 775. This may lead to some people getting disappointed

(not me, i knew i was holding back on nehalem for a reason)
 


This we do not know yet. Even if they drop a pin or two it can still be put into a older socket. AM3 has less pins than AM2+ but they work.
 


your guess is as good as ours. Although im guessing not very much for now as games are still more gpu than cpu dependent.

You'll probably see more performance gain out of the next generation of gpu's (like ATI Radeon HD 6xxx ) vs HD 5xxx than a Nehalem vs Sandy Bridge.


Although i am interested FPU performance increases in these cpu's.... :heink:
 

ambam

Distinguished
Jun 5, 2010
1,166
0
19,290


"FPU?" Floating point unit?

There aren't any games which make use of a quad core processor, let alone 6 or 8. It is very unlikely a PC game will exist which fully utilizes 6-8 cores for many, many years. Even Crysis only uses 2 cores. The ONLY PC game that uses a quad core is GTA4, due to a poorly coded port from consoles to PC.

Maybe Crysis 2 will use 3, hopefully 4 cores.
 


?? You got any Bulldozer previews you can share? IIRC it's still a year away, and AFAIK there are no preliminary performance numbers, just WAGs & speculation on a bunch of websites.
 

ares1214

Splendid
Yeah, while i do agree there isnt much info out on Bulldozer yet, i still think it deserves a bit more hype. Although it doesnt seem very fair, this lineup is just a overpowered current i3, i5, and i7, and seems to be more of a stepping stone to socket "2011". So comparing bulldozer to this seems to be like comparing intels flagship top dog to amds mid range. Bulldozer is a real (more or less) 8 core, so im very much so expecting it to beat these, however im not so confident how it will fair against "2011". BTW, if these chips cant oc like the test (or the motherboard manufacturers dont fix it) or intel actually does call it socket "2011, Intel deserves one of these:



............................................______ __
....................................,.-‘”...................``~.,
.............................,.-”...................................“-.,
.........................,/...............................................”:,
.....................,?........................... ...........................,
.................../.................................................. .........,}
................./.................................................. ....,:`^`..}
.............../.................................................. .,:”........./
..............?.....__............................ .............:`.........../
............./__.(.....“~-,_..............................,:`........../
.........../(_....”~,_........“~,_....................,:`..... ..._/
..........{.._$;_......”=,_.......“-,_.......,.-~-,},.~”;/....}
...........((.....*~_.......”=-._......“;,,./`..../”............../
...,,,___.`~,......“~.,....................`..... }............../
............(....`=-,,.......`........................(......;_,,-”
............/.`~,......`-...................................../
.............`~.*-,.....................................|,./.....,__
,,_..........}.>-._...................................|........... ...`=~-,
.....`=~-,__......`,.................................
...................`=~-,,.,...............................
................................`:,,.............. .............`..............__
.....................................`=-,...................,%`>--==``
........................................_........ ..._,-%.......`
...................................,
 
^ Wasn't intending to be snarky or anything, just thought it was too soon for any BD ES benchies to appear.

Anyway AMD really should bring BD forward as fast as possible, at least the server part, since according to IDC they just dropped a whopping 36% in the highly-profitable server marketshare last quarter. Down from 10.1% to 6.3%, and still posting a loss (although just #143M due to their stake in GF). While Intel had a blowout Q2, particularly in the server segment. So it seems Magny Cours did not help AMD any.

IIRC AMD had a high of nearly 30% marketshare in servers as recently as 2007. So they have shrunk about 600% in 3 years.
 


Would this be what you're be referring to?

http://blogs.amd.com/work/2010/08/02/what-is-bulldozer/
 


Lets say that BD has a 12.5% per core performance improvement. Its hard to say since it has 2 more cores. But lets say its right. And lets speculate that Intels preliminary 18.8% per core performance gain is correct.

Lets say that current AMD CPUs are behind Intel by about 10%+/- (being nice) that means that BD would be about +2.5% better than Nehalem but -16.3% worse than Sandy Bridge per core.

That doesn't give me any confidence. Of course its all speculation based on, speculation. It could be right or it could be wrong.

According to a lot of info, BD is supposed to at least catch up to nehalem in pure performance. And Sandy Bridge is a evolution of Nehalem. Its said by Intell to give close to that Netburst -> Conroe leap.

But of course we have to take it with a grain of salt.
 

ares1214

Splendid
Well thats also the server market. AMD didnt even mention desktop BD, so its not terribly valid. Also, i think those difference shown for sandy bridge were overall. This is per core. Going from 4 core at 100% performance to 8 cores with 112.5% performance...well that doesnt even sound that bad! :lol: Although improvement in the desktop end were suppost to be a lot higher, AMD touted 50% increase per core. I doubt it, MAYBE 50% performance increase theoretical (4 more cores, 12.5%, 50% by their math i guess.) but no use speculating until more info is out. One interesting observation though. These are 32nm just like core i3. But if you notice, they are starting their bottom normal cpu at more than 200 MHz what they started i3 at, and even more as you go up the line. Im thinking 1 of 3 things.

1. Intel needs to clock these higher to get them to perform on par with what they want. Like AMD right now

2. Intel is scared of BD, and so is clocking them high just to make sure :lol:

3. Intel DID limit overclocking, and therefore try to compensate for it by clocking them higher at stock.

If its number 3, i dont care if bulldozer is 2.5% performance increase, im buying it! That will be both smart, and dumb if intel doesnt allow oc'ing. Smart since it keeps curious 10 year olds from trying to OC it and complaining to HP and bestbuy their computer has an ugly blue screen. Smart as it likely might lower costs for motherboard manufacturers. DUMB for us overclockers and enthusiasts. But we must remember we are but a sliver of a percent of the market.
 


Oh so true. :lol: