Am I missing something about Ivy Bridge?

[Vesku]

*Image of an astronaut with a Haswell laptop playing Crysis while being sucked towards the event horizon emanating from the blackhole caused by the notebooks transition to a 4004.*

Pure awesome popped into my head with your humorous post, thank you kind sir.

 

[ZOXXO]

It folds the very fabric of space-time itself, so much so that it catches up with itself and the gate-last xtors become gate-first while the whole thing implodes and turns into a 4004. But yeah, it can do crysis, so, its bitchen

Still yet Bulldozer will be %50 faster.

 

[Edrick]

Still yet Bulldozer will be %50 faster.

ROFL! That was great.

 

[epidemis]

Do we know anything about Haswell yet? Is it too early to start speculating?

Some tentative parts of information. Fx it is said it's designed with mobile computing in mind, and will probably have a lower TDP than hitherto archs.

 

[gramboh]

OP: Buy SB now, I assume your laptop is not giving you the performance you want (especially if you play games). Who knows when IB and BD will actually come out and forget about Haswell.

I was going to build an SB 2600k system but decided I'm happy enough with my PC after upgrading my 280 to a 560 Ti, so I am going to wait for IB (assuming BF3 doesn't destroyed my PC too badly haha).

 

[Idontcare]

*Image of an astronaut with a Haswell laptop playing Crysis while being sucked towards the event horizon emanating from the blackhole caused by the notebooks transition to a 4004.*

Pure awesome popped into my head with your humorous post, thank you kind sir.

LOL that's a good one :thumbsup: Sucks for the astronaut, guess he shouldn't have been wasting taxpayer dollars though, serves'em right :awe:

Still yet Bulldozer will be %50 faster.

Damn straight, but the SOI layer turns out to be a metaphysical equivalent of the Casimir effect, insulating the CPU and its user from the black-hole thingy that Vesku reminded us ignorant noobs about :biggrin:

Intel's engineers only get to dream about building Casimir-enabled bulldozers, they'll have to get their jollies from fiddling with their fins for now

 

[Zap]

IB's turbo sounds like BD's turbo in that it monitors the temp.

i just wonder what the motherboard power circuitry will need to be built to? people have burnt out their AMD 95W boards when overclocking their processors.

That's their own fault for putting 125W CPUs into 95W motherboards. :twisted: Actually, I've done that for about five minutes without overclocking. Just at idle the motherboard VRMs got really hot!

 

[Nemesis 1]

Do we know anything about Haswell yet? Is it too early to start speculating?

From wik

Haswell is expected to have:[2][4]

A 14 stage pipeline.
Up to 8 cores available.
New advanced power saving mechanisms.
Possible on-package vector coprocessor.
FMA3 (Fused multiply-add) instructions.
An entirely new cache design.
64 kB data + 64 kB instruction L1 cache per core, 4-way associativity.
1 MB L2 cache per core, 4-way associativity.
Up to 16 MB L3 cache shared by all cores, 8-way associativity.




Link It will take you to an 800 page PDF . That should keep ya busy.

http://software.intel.com/file/35247/

 

[ydnas7]

Haswell instructions
http://software.intel.com/en-us/blogs/2011/06/13/haswell-new-instruction-descriptions-now-available/


What is important about about IvyBridge is that it is the tick where Intel stays on Moore's law for xtor speed, whereas for GloFo, 32nm is the end of the line for increases in xtor speed. the FD SOI will allow GloFo to keep improving energy useage and density, but for xtor speed, 3D will be a must.

 

[ShadowVVL]

so anyone know if IB is expected to cost more or less then sb?
Also what of names, maybe i7 2800k+ or i8?

Im looking forward to the IB i3 if they make one.

 

[bridito]

so anyone know if IB is expected to cost more or less then sb?
Also what of names, maybe i7 2800k+ or i8?

Im looking forward to the IB i3 if they make one.

I have extensive inside Intel knowledge (NOT!) but from what I've seen rumored:

- They'll stay with the odd number i series and no i1 or i9 in sight. Their naming convention makes less sense than Sarah Palin For President, but let's not go there. :'(
- They'll likely push SB-E to the 3000 series so maybe IB will be a 4000?
- The initial IBs will be LGA1155 so they will continue the Sandy Bridge introduction manner where they bring in their lower end CPUs first and then later launch the heavy duty equipment, unlike how they've done it in the past (the high end LGA1366 first then the low end LGA1155).

This information comes with absolutely no guarantees.

 

[garagisti]

Check this post...

http://www.xtremesystems.org/forums...-info-fans-!&p=4883691&viewfull=1#post4883691

This guy owns a SR2 rig and uses it for a desktop. He says BD would beat a 990X. Given he owns a SR2, and he has access to BD, i'd believe him.

 

[tweakboy]

2012 Ivy Bridge . 22nm


Everyone hold their horses and the people that already have quads ,, relax your fine no need to upgrade a quad with another quad LOL ,,, wait for Ivy 6 core 6 HT 12logicals.. 2012 Ivy Bridge will pownz the universe soo ya,,

 

[Roy2001]

IB won't be out for another year. If you need a PC, get SB know.

 

[Nemesis 1]

IB won't be out for another year. If you need a PC, get SB know.

Were talking Intel here not AMD. IB will be 1st QT 2012

 

[Ken g6]

My understanding is that IB's integrated GPU will also support OpenCL. If Llano and Trinity go over well using OpenCL on the on-die GPU, then this will matter for IB as well. If not... :|

 

[Kingkazma]

im stuck right now because of the chipsets... i actually want pcie 3.0...

 

[Phynaz]

Were talking Intel here not AMD. IB will be 1st QT 2012

Which means you will not be able to buy one until Q2. And it just so happens we are in Q2 now, so that makes it a year.

Please turn down the anti-AMD rhetoric, it's getting very old. Some would say boorish.

 

[ydnas7]

http://spectrum.ieee.org/semiconductors/design/the-origins-of-intels-new-transistor-and-its-future


Chenming Hu: Staying with the hose analogy, when you have a big thick hose, you can carry more current, so it’s good for high speed. So that’s why I think the large companies that can make the investment in FinFETs will probably do it, because FinFETs are versatile. For the companies that need to have a quick way to get beyond 22 nm, I think UTB is a viable technology, especially for those companies that already have experience with the SOI. The shortcoming, of course, is the flat hose—with such thin silicon, less current goes through, which translates to lower speed.

 

[ydnas7]

this is why IB will be important, because SOI (AMD) may help lower power consumption but it will hit a speed wall, but Finfets (Intel) IB will keep going

 

[Idontcare]

this is why IB will be important, because SOI (AMD) may help lower power consumption but it will hit a speed wall, but Finfets (Intel) IB will keep going

Yup, we call them "red brick wall" in the industry. The only way to circumvent the barrier is to develop new technology.

If you don't need or require high performance CMOS then you don't need to worry about finfets and HKMG for a while, you just keep focused on reducing feature sizes (lowering production costs by shrinking your chips) and power-consumption by reducing Vcc but you won't get much more out in terms of micro-amps/micron driver currents.

Intel is still trying to solve the problems with process tech innovations. IBM threw in the towel and will just turn up the Vcc to juice up drive-currents (and power-consumption) and resort to water cooling as the stock cooling solution to their 200W+ TDP beasties.

IEEE Spectrum: How did the industry react to the FinFET paper?
Chenming Hu: It was an instant hit. I remember Cathy and I were invited to Intel Santa Clara just a couple of months after the publication, and in that same year, 2000, I was invited to Intel Oregon twice. At the time, people were asking me how long it would take for the idea to get into production. I said about 10 years, so I guess I was off by one.

This guy and his two buddies are going to get the nobel prize in physics, no question.