intel ivy bridge-e price

[csbin]

http://translate.google.com.hk/tran...e-and-didnt-over-1000-usd-for-4960x-07312013/

i7-4960X's default frequency of 3.6GHz, but only the highest dynamic speed 3.7GHz (previous exposure engineering model dynamic acceleration up to 4GHz)

 

[Idontcare]

Intel can count on continued YoY declining PC sales with these kinds of YoY product improvements

Digging their own grave, they are.

 

[RaistlinZ]

I don't understand, why are they're going backwards in speed?

The 4820k, 4930k, and 4960X all have slower peak speeks than the 3820, 3930k and 3960X. Why release CPU's slower than the ones you're replacing?


Ugh...I guess this is what we have to deal with when there's no competition.

/RIP PC

 

[JDG1980]

That screenshot came from a sketchy foreign site, so I'm not sure it should be taken at face value. It contains errors regarding current products: the i7-3960X has a default non-turbo speed of 3.3 GHz (according to Intel's official page), not the 3.5 GHz listed in the chart. Also, the information provided on Ivy Bridge-E by Tom's Hardware is different: that site indicates that the i7-4960X will have a base clock of 3.6 GHz, with turbo to 4.0 GHz.
I suppose we'll see soon enough, but I'd take the word of Tom's Hardware over a random screenshot on some Chinese site.

 

[blackened23]

The HECK? Someone please tell me those turbo speeds are a mistake? I was pretty excited about IVB-E as well. :thumbsdown:

edit: thanks for the info JDG. I hope you are correct. Having lower turbo speeds would be absolute nonsense, doesn't seem believable.

 

[CHADBOGA]

I was pretty excited about IVB-E as well.

5% over SB-E excites you?

 

[crashtech]

Well, the expectation was that IB-E would be a decent overclocker, but those turbo speeds are worrisome in that regard, almost as if they were not soldered.

 

[PPB]

Would be nice to see 4820K's OC temps compared to 3770K's ones, considering 4820K should be soldered like the rest of the 6 core variants.

 

[SPBHM]

the 4930K is by far the most interesting, just like the 3930K was.
so it's a little bit cheaper, and a little bit faster, uses a little less power than it's predecessor, it could be worse I guess.

 

[frozentundra123456]

Well, the expectation was that IB-E would be a decent overclocker, but those turbo speeds are worrisome in that regard, almost as if they were not soldered.

Yea, if those numbers are correct, it is pretty strange. Seems like the trend is toward higher base clock and lower turbo. I would think if the chip had good overclocking potential they would have increased the turbo, not decreased it.

Seems to be a nice decrease in power usage, but who cares unless that means cooler running and better overclocking potential, which again does not seem likely given the turbo speeds.

 

[nehalem256]

Intel can count on continued YoY declining PC sales with these kinds of YoY product improvements

Digging their own grave, they are.

If the specs for the 4960X are correct(lke the 3.7 turbo) I think improvement is the wrong word.

 

[crashtech]

Why would a "K" part like the 4820K have no turbo at all?

What's to say this image is not a fake?

 

[SPBHM]

Why would a "K" part like the 4820K have no turbo at all?

What's to say this image is not a fake?

it can be fake but,

one reason would be, making sure it's not faster than the 60x part... not even for ST... so the 4960X would be absolutely the fastest 2011 i7...


anyway, it's a shame that the $1K part still have 5MB and 2 cores (or more?) disabled.

 

[SOFTengCOMPelec]

Intel can count on continued YoY declining PC sales with these kinds of YoY product improvements

Digging their own grave, they are.

There is the odd rumour here and there that says they (Intel) have been having increasingly bad technical problems, especially with the fabs, as the scales have become smaller and smaller.
Not enough rumors for anything definitive, but enough to stick in the back of my mind.

Another rumor (but probably based on facts, if Intels full product line are examined) is that these Socket 2011++ ('-E' = Extreme) parts are really rejected Xeon chips.

E.g. We have 6 cores in the existing 3960X, but would have 8(+) cores, on die, or if we went with top end Xeons.

So if they are rejected Xeons, and they are having difficulties with the later chips, then it is harder for them to spare "good" chips for the 'Extreme' market.
Hence the slightly lower turbo clock, and relatively lacklustre clock speeds, compared with the previous chips.

So there could be reasons behind this information leak (or it is FAKE ?).

I can't remember who said it (it might of been you, IDC), but the finFet process, may not be fully debugged/sort out yet, especially with the major diversion of resources to the small non-desktop devices markets.

(I twisted the words round, the original information said that Sandy Bridge was the 'LAST' battle tank strength, bullet proof design to come from Intel, by inference it implies the finFet and/or something about 22nm is troubled)

 

[crashtech]

So we are supposed to believe that none of these presumably soldered Ivy Bridge chips can even match the stock turbo speed of a 3770K? If it's true, it's about the most pathetic thing I've seen from Intel in a long, long time. I sure hope this is wrong, hell it just about has to be a joke.

 

[beginner99]

Fake or intel is trying to commit suicide. Or maybe 22 nm tri-gate is just not made for chips like these how knows?

 

[ShintaiDK]

Wrong info, wrong MSRP.

Really, they couldnt even get the 3970X 3.5Ghz number right. or the 3960X.

http://ark.intel.com/products/70845...ssor-Extreme-Edition-15M-Cache-up-to-4_00-GHz

The 4960X is a 3.6Ghz base, 4.0Ghz turbo 130W chip.
To compare:
The 3970X is a 3.5Ghz base, 4.0Ghz turbo 150W chip.

 

[greenhawk]

So we are supposed to believe that none of these presumably soldered Ivy Bridge chips can even match the stock turbo speed of a 3770K?

IIRC there was a similar issue with the SB-E when they where originally released, so not that surprising.

 

[escrow4]

I used to have no problem upgrading every few months but now its all pfffft. I await Skylake E, assuming there will be one and Intel doesn't just switch everything to soldered, I don't want to go back a quad after a hexa. Lower power consumption is irrelevant for me, the tiny IPC gains are also yawn worthy and even a fresh new chipset with more ports is meh, yes X79 only has 2 SATA 3, but 2 is 1 one than I need anyhow.

 

[Nothingness]

Looks like a fake if we are to believe what Tom's Hardware wrote: http://www.tomshardware.com/reviews/core-i7-4960x-ivy-bridge-e-benchmark,3557.html

 

[MisterMac]

In this segment for those who have 2011... why would they even care?


Any of you 2011 owners run stock settings? at all?


The only thing to note is how high they clock to perhaps extrapolate the extra headroom for OC - but even so that's extremely speculative and could mean nothing in terms of headroom.


So why would we care as Extreme buyers if it was a little higher or lower?
Those i would imagine that buy this - are going to crank up to max speed @ desired voltage from day one no?


Just a thought i had.

 

[pelov]

Does anyone feel like it may be time for Intel to have 3 separate architectures? server/consumer/mobile? I genuinely feel like they're shooting themselves in the foot here.

 

[Idontcare]

Does anyone feel like it may be time for Intel to have 3 separate architectures? server/consumer/mobile? I genuinely feel like they're shooting themselves in the foot here.

Itanium, Haswell, Knights Ferry, Atom? How many more does Intel need? Getting rather segmented already.

 

[NTMBK]

Itanium, Haswell, Knights Ferry, Atom? How many more does Intel need? Getting rather segmented already.

Oh come on, Itanium is dying. The 22nm part was cancelled, as was the "common platform". Either HP couldn't afford to pay for further development or Intel just flat out wanted it gone.

 

[R0H1T]

Itanium, Haswell, Knights Ferry, Atom? How many more does Intel need? Getting rather segmented already.

I think what he meant was that Intel needs a new microarch for the consumer segment & separate it from their server biz. Itanium is dead(pretty much) & the current consumer/server chips are derived from the core microarch, with several ISA additions & minor improvements over the years, also the current Atom is based on Pentium(correct me if I'm wrong here) so maybe he wants something like a seachange akin to K10 -> Bulldozer, minus the failings of course, but Intel won't bet on their bread & butter nor risk a P4 like unmitigated disaster especially in a shrinking market !