Latest Core i7 release date info

[WaitingForNehalem]

Originally posted by: Beavermatic

Ah, Ah, Ah....

If i deem the i7 a worthy upgrade to my q9550 when it is released, you better beleive i'll be the first one in line for one. Its pocket change for the price of a processor nowadays, as long as they are $1000 or less. I want to see benchmarks first, though. I don't run of hypothetical scores, or virtual scores for that matter.

Though I keep hearing these words of a six core i7 due to be released within the first few months of 2009, and that they wont be making an apperance with the first releases on November 17.

Going from a Quad core to another Quad core when 6 cores are right around the corner would be a mistake easily made, especially if the initial quad core i7 release does not oust the qx9770.

You say I need to keep up on my game. May i remind those who commented that the article that claimed the 2009 release is only a month and a half old, and the November 17th release are skeptical leaks from supposed non-disclosure agreement violators at best, or at least it was, until I checked with a buddy who deals with intel reps on a daily basis for a hardware distrubtor company.

So I stand corrected, indeed, you were right, the highend quadcores of I7 series are indefinetely being released in November.

Though I do know they are being released on November 17th myself now, its where I stand as far as going to a slightly faste quad core, or just waiting around for the 6 core.

I know the high end quad cores are definitely releasing on the 17th, but no true dates on the six core yet ***sigh***. Unfortunately, intel hasn't been so kind to slip any info on that set.

But for you, I'm happy to say you'll have a machine thats probably going to rock the pants off anyone around you come November

There won't be more processor releases until Q4 2009. These quadcores that are coming out have HyperThreading which makes them have 8 threads. Normal quads have 4 threads. Westmere should have an 8 core processor, but like the Core 2 generation, you only need the original amount of cores released. It will be like Pentium D and Core 2 Quads, they are just two dies on one MCM, they aren't native. BTW, I consider anything over one day old news.

 

[IntelUser2000]

You know how people always argue how x86 is the inferior instruction set and how its bloated?? Nowadays that talk is mainly gone but it still surfaces when talking about x86 for graphics or for making it extremely small/low power like the Atom.

Well, there was an article about Nehalem few months ago from PCWatch and they were talking about what the Intel execs and the designers of the architecture had to say. We know by now Nehalem won't bring big single thread performance increase. Funny thing is how people are talking like AMD will easily catch up to Nehalem's single thread performance.

The article basically said that now they are really reaching the limits of how much performance they can extract out of CPUs, and one of the limiters is the x86 instruction set. It may look like 5% performance improvement is a trivial thing to do, however it is not. The things that brought biggest performance increases for the last couple of generations of CPUs(both AMD and Intel) had been related to memory and memory parallelism. Pentium III coppermine, Athlon 64, Core 2.

Easy 5-7% performance increase had been brought on by ever increasing cache sizes. Now we are facing a point of diminishing returns. If you ever read very detailed studies about IPC and CPUs, the CPU designers go quite a lot to extract 1% IPC. Things like Micro/Macro Op Fusion, Branch prediction improvements, greater issue capability, better reordering capability all add up to mere single digits combined together.

Nehalem is the last big single threaded integer IPC focus(the most important performance metric in CPUs) that we will see. Hence their focus on turbo mode. The next tock, Sandy Bridge, will bring hefty FP performance improvements but in single thread, it won't advance. Sandy Bridge will have much better turbo than Nehalem that will clock far higher.

Say every process technology brings greater clock speed and 2x the cores. Sure, it'll continue for couple of generations with ever more aggressive power management. But its time where it has to be decided whether to increase more cores or increase single thread performance.

 

[Quiksilver]

Memory bandwith benchie.

 

[Tempered81]

single vs. multi thread in cinebench:

http://publish.it168.com/2008/1007/images/1184844.JPG

 

[AuDioFreaK39]

Originally posted by: Quiksilver
Memory bandwith benchie.

looks like they pulled the Hexus review to keep it wraps under NDA. :laugh:

(this was discussed over at XS as well)

 

[Tempered81]

Heres a little overclocking information from Andre at NordicHW:

http://www.xtremesystems.org/f...p=3351585&postcount=46

"The QPI bus is not going to be your main focus, but it's not hard if you want to.
You just should forget everything you've read at Fudzilla and TheInq about how this is done and what problems there might be. The stories about how Intel would lock overclocking through synchronized bus controllers in the CPU and northbridge, problems the synchronized voltages etc, they have just been a bunch of lies. Some of the stories haven't don't even contain a single ounce of truth or even the slightest fact from reality. Maybe in a slightly distorted dimension on another planet, but the reality is that Nehalem overclocks and considering how much trouble we had when overclocking Kentsfield and Yorkfield, Nehalem has been a lot more stable.
You are going to overclock Nehalem just like any other new system. Just a new bus and new voltages to get used. Overclocking is here to stay.
//Andreas"

And Gary Key on the Voltage & ram blog:

http://www.xtremesystems.org/f...=3349010&postcount=112

"As you know and several others here, we could write a few pages on the technical details and how to get the memory clocked up with higher voltages without affecting the IMC (to a certain point). I wanted to go further, but no need in upsetting the blue team this close to launch. I hate having a sock in my mouth for this long and seeing all the "High VDimm DDR3 will kill your Nehalem" articles.

For everyone else-

Can High VDimm damage the CPU? Yes it can, and it has, but nothing really different than what occurred on Phenom/A64 until the users and BIOS engineers were educated. Intel is playing it safe and rightfully so up front with the 1.60V~1.65V guidelines. I do not blame them as this platform will be a significant change for users weaned on FSB for the past decade.
I over simplified the HT to QPI comparisons in the blog, but really, anyone with clocking experience on the A64/Phenom platforms will feel right at home. There are a couple of additional/important settings to learn and the terminology is a little different, but I think a good A64/Phenom clocker will be about 80% there on day one.
That said, I feel really safe right now that up to 1.8 VDimm and correct BIOS settings will be fine for the general enthusiast (time will tell). For the more adventurous overclockers, the right settings will result in 2.0V~2.1V for benchmarking purposes (not so sure for long term usage as I had my hand on two CPUs today that did not survive 24/7 benching at higher than 2.0 VDimm settings). This is more than enough for current Samsung ICs, probably a bit much for the new Qimonda ICs. I am testing Micron DDR3 now and will report further on it next week.
For the rest of us, getting CAS5 at 1333 or CAS6 at 1600 in Tri-channel operation with 1.65V will result in some phenomenal bandwidth, write speeds, and latencies (see Tony's post above). So much so, it has been extremely difficult to even locate a program (commercial application) that can utilize the bandwidth Intel is about to give us. To be honest, we were at this point a few years ago with A64 in the beginning, but the jump this time is just incredible."

 

[solofly]

I'm not a big overclocker but sometimes I do overclock a little. I've been doing some testing with my Patriot DDR3 RAM and the best it will do is 1333MHz at 8-8-8-20 without going over the recommended voltage of 1.65v. Now if it proves all that extra memory bandwidth won't do much for gaming, I will drop the voltage all the way down to 1.5v and run it either at 1333@9-9-9-24 or 1067@7-7-7-20... (benchmarks will decide which one)

 

[Tempered81]

Triple channel DDR3 performance in Crysis:

http://www.techpowerup.com/img/08-10-13/35d.jpg

and in Cod4:
http://www.techpowerup.com/img/08-10-13/35c.jpg



link:
http://www.xtremesystems.org/f...howthread.php?t=204544

 

[WaitingForNehalem]

Well, now we know November 17 is the release date.

http://www.pcstats.com/NewsView.cfm?NewsID=71048

 

[Idontcare]

Originally posted by: WaitingForNehalem
Well, now we know November 17 is the release date.

http://www.pcstats.com/NewsView.cfm?NewsID=71048

I know this is the historic timeframe that Intel likes to release new CPU's but does anyone know why they chose this timeframe?

It is oddly late in Q4 from a christmas shopping perspective IMO. Surely Oct 17 would allow them to snag more holiday units.

 

[bryanW1995]

they're probably balancing the need to come to market for xmas with the need to have a reasonably stable system available.

 

[Idontcare]

Originally posted by: bryanW1995
they're probably balancing the need to come to market for xmas with the need to have a reasonably stable system available.

That's a reality that exists because everyone has spent the past 4 yrs working towards an expected release date of mid-Nov. Had the anticipated (internally) release date been mid-Oct then you can bet things would have been staged and planned so they met the mid-Oct release with just as stable a platform as they are planning to release come mid-Nov.

1 month on a 4yr schedule is pretty easy to accomodate when you have Intel's resources.

I was more assuming there was some marketing facet to the rationale, or accounting rationale in terms of not undermining inventory builds from Q3 that need a holiday season of frenzied buying in order to flush out of the system. Don't want Nehalem causing existing Penryn sales to crash.

I really don't know the reasoning, was kinda hoping someone else who was "in the channel" could illuminate the subject. But I do feel we can safely rule out "systems aren't stable come every Oct these past years" as the systematic reasoning why Intel prefers to launch CPU's in Nov with consistency.

 

[WaitingForNehalem]

We were able to obtain the final, confirmed launch date! The Intel Core i7-965 Extreme Edition, i7-940 and i7-920 processors, as well as the Intel X58 Express chipset will be released at 9:00 pm Pacific Time, November 16, 2008. This will be followed by the official Intel press release and worldwide launch events on November 17.

Although OEMs are apparently restricted to launching their products only at 5 pm Pacific Time, November 17 (after the Intel launch), Intel will allow the new Core i7 processors and X58 Express motherboards to be sold as early as 12:01 am local time, November 16, 2008.

To summarize :

November 16 (12:01 am local time) - End user sales, shipment and ads

November 16 (9 pm Pacific Time) - Worldwide product release

November 17 - Intel press release and launch events

November 17 (5 pm Pacific Time onwards) - OEM press release and launch events

http://www.techarp.com/showarticle.aspx?artno=589

 

[WaitingForNehalem]

People keep on saying they're unimpressed with the performance advantage of Core i7. What they fail to realize is that Core 2 came 6 years after P4. The reason AMD beat Intel with K8 was because it was a new generation. AMD went through Athlon, Athlon XP, Athlon 64, Athlon 64 X2, and Athlon 64 FX. Intel went through Pentium 4, Pentium 4 HT, and Pentium D. Core i7 hasn't been released yet so everyone should wait and see. Plus software is so behind that it will be a while before all the advantages of Core i7 are shown.

 

[WaitingForNehalem]

Core i7 owned the QX9770 in Lost Planet: Colonies.

IDF Video

 

[AuDioFreaK39]

Originally posted by: WaitingForNehalem
Core i7 owned the QX9770 in Lost Planet: Colonies.

IDF Video

Awesome, finally some leverage after all the negativity Nehalem has been receiving in respect to gaming.


I've known this for quite some time now, and I just didn't know if it was still under NDA or not, but - EVGA will be releasing an X58 motherboard that will *hopefully* release at the launch of Nehalem. I've talked to a few people at EVGA about this, and they are very hush hush about timeframes but they did say there is an EVGA X58 platform in the works.


Absolute confirmation can be found in NVIDIA's latest press release:

http://www.nvidia.com/object/io_1224586792565.html

?EVGA is continuing its tradition of bringing true enthusiast platforms to the market following the success with NVIDIA nForce-based SLI motherboards supporting Intel CPUs,? said Joe Darwin, Director of Technical Marketing at EVGA. ?By licensing NVIDIA SLI technology, the EVGA X58 motherboard will deliver the ultimate 2-way and 3-way SLI platform to, once again, meet the enthusiast?s demand.?


Hardware Akutell - EVGA is Working on X58 Mainboard (Translated)


With that said, don't get your eyes set on the ASUS P6T or Gigabyte EX58 offerings because, as stated from my sources, "they will suck compared to this monster!"

In addition, here is a blog entry written by Theo Valich from Wordpress describing the situation:

WORLD EXCLUSIVE: EVGA to launch Intel X58 motherboards http://theovalich.wordpress.com/2008...-motherboards/ When it comes to add-in board vendors, EVGA is probably the most faithful company in the business. Ever since the company launched, Nvidia was the only name EVGA wanted to hear about. But, things are about to change.

Here are the facts:
1) EVGA does not want to miss the Core i7 train
2) Nvidia is not making a chipset for Intel Core i7
3) EVGA poached excellent engineering team from now-defunct EPoX and does not want that team to do nothing until MCP8-series show up

Well, those facts end with a really simple result. EVGA is preparing to launch its first non-Nvidia based motherboard, but it will still have Nvidia chips on it. You?ve guessed it right - X58+nForce 200 bridges for full Triple-SLI capability. The motherboard is being designed by ex-EPoX engineering team, bringing plentiful of overclocking capabilities, Digital PWM, and fully solid-state caps across the board. Special attention is being given to providing top juice to graphics cards, so if you decide to go for the gold and grab 3-SLI setup with three water-cooled GTX270 boards, be our guest. But, that is not all. This is the first motherboard, which will be followed by ultimate motherboard for this Christmas, no questions asked. The enthusiast motherboard is actually being designed by a world class overclocker. Yes, the one and only Peter Tan a.k.a. Shamino, who is making ?Shamino special?. For those unaware ? Brian Flood from Mushkin, Shamino, Kinc and I share a special connection. According to German police, we all had the honor of dying and resurrecting. Last year, we were all pronounced ?missing presumably dead? when some east-German thieves stole helluwa stuff from our brand new VW Passat Variant, en route to the airport. Well, VW is one car I will *never* buy. Piece of alarm-unsecured junk. Getting back to the subject, EVGA is bringing several things to high-end X58 market that nobody has. First of all, the company will offer limited lifetime warranty (just as with all of their high-end products) and yes, 90-day Step up program. If that is not making ASUS, Gigabyte, MSI and other motherboard makers sweat, we don?t know what will. Lifetime warranty on a motherboard? Free upgrade program? Bloody hell, I am buying that one.

P.S. I'd also like to note that this will be EVGA's first Intel chipset board.

 

[WaitingForNehalem]

The EVGA motherboard sounds very expensive. So far I know the Foxconn Renaissance and MSI Performance will be $200-$250. Am I missing anymore?

 

[aigomorla]

omg...

why did evga go with epox.

yeah, its gonna be one hell of an expensive board. :X

 

[Idontcare]

Originally posted by: AuDioFreaK39
P.S. I'd also like to note that this will be EVGA's first Intel chipset board.

Uhg, rarely (if ever) a good sign for the end consumer to buy the first of any hardware from a company. I pity the first round of beta testers who get to pay for the pleasure of ramping EVGA up their internal learning curve.

 

[Slappy00]

Originally posted by: aigomorla
omg...

why did evga go with epox.

yeah, its gonna be one hell of an expensive board. :X

My Epox 939 Nforce 4 is one of the best boards I have ever owned. They made an awesome product that had zero problems from installation to daily use... coming on 3 years now. I was going to buy Epox forever (low price for an awesome part) too bad they are not in the motherboard business anymore in the US.

 

[AuDioFreaK39]

Originally posted by: Idontcare

Originally posted by: AuDioFreaK39
P.S. I'd also like to note that this will be EVGA's first Intel chipset board.

Uhg, rarely (if ever) a good sign for the end consumer to buy the first of any hardware from a company. I pity the first round of beta testers who get to pay for the pleasure of ramping EVGA up their internal learning curve.

There is no such thing as a "learning curve" when you have the legendary Shamino as well as a dozen (confirmed) or so ex-Epox engineers working on this board. It's just going to fly. Additionally, don't forget there will be an "Ultimate" edition releasing shortly after in December, which early adopters such as myself will be able to upgrade to through the EVGA Step-Up program. What can I say? EVGA FTW.

 

[WaitingForNehalem]

Oooo, Intel deleted all the 11-3-2008 posts as the release date on XtremeSystems.

 

[VirtualLarry]

Originally posted by: WaitingForNehalem
Oooo, Intel deleted all the 11-3-2008 posts as the release date on XtremeSystems.

Intel corp. has moderator rights on XS???

 

[Idontcare]

Originally posted by: VirtualLarry

Originally posted by: WaitingForNehalem
Oooo, Intel deleted all the 11-3-2008 posts as the release date on XtremeSystems.

Intel corp. has moderator rights on XS???

Not deleted by Intel...deleted by the post originator at the request of Intel.

I.e. the posters elected to become (presumably) unpaid shills and scurried themselves into self-censorship at the behest of their masters.

Me loves me a monopoly.

Here's the original:

Originally posted by: OBR
Launch date is 03.11.2008 - Source (mail from Intel representative about NDA)

http://www.xtremesystems.org/f...howthread.php?t=205362

Here's what it was edited to:

Originally posted by: OBR
Deleted by request of Intel

http://www.xtremesystems.org/f...howthread.php?t=205362

 

[bryanW1995]

those are the same thing...