Intel DX79SI "Primed and Ready!" for November 15

[bamtan2]

Includes:
• UV reactive SATA cables
• Extreme mouse pad
• Bluetooth*/Wi-Fi module
• Board and back panel I/O layout stickers
• NVIDIA* SLI* Bridge connectors
• Post code information card
• ESET* Smart Security 5 Antivirus (WTF? THEY OWN MCAFEE!)

Source: http://vr-zone.com/articles/intel-le...ite/13854.html

Source: http://m.intel.com/content/intel-us/...9SI.touch.html

Board info: http://m.intel.com/content/dam/intel...eSeries_PB.pdf

PS: How about them dual Intel LAN onboard!

[bamtan2]

IMO the reader should note two things:
1) There is a bios on the download page dated TWO MONTHS AGO. We know we've been looking at X79 boards for a long time now. And Sandy Bridge has been out almost a year. So what is taking Sandy Bridge-E so long?
2) There were TWO Intel X79 boards talked about a few months ago (DX79SI and DX79TO), so where did the other one go?

This, combined with the recent removal of features from X79 boards, AND the rumors of SNB-E Xeon chips being delayed for a 2012 RESPIN, AND how close we are to 22nm Ivy Bridge tells me one thing: AMD isn't alone with their problems. This Intel platform (chip and chipset) has been delayed and delayed because it is having problems, and so now it is starting to look a bit behind the times. It is still the only new Intel platform with enterprise features, but that won't stop it looking a bit silly next to Ivy Bridge Z77/Z75 with native USB 3 and whatever else, running at 5ghz with a paper clip for a heatsink.

I might buy one anyway LOL! We'll see in two weeks.

[Greg83]

That's why all blue tooth / wireless. Intel added it to the design.
I am not sure about two wireless functions in standby mode on a enthusiast chip set though.
Could be apple wanted it. But then why not Bluetooth 4.0 for the new phone.

[IDANPE]

All the boards I've seen so far contain only 2 SATA3 and 4 SATA2 connections from the X79.
Everything else is coming from the Marvel chip.

Why?
the X79 supposed to provide up to 10 SATA3 Connections, so why we get only 2 ?
and the rest come from the poor marvel chip or sata2..

[Systemlord]

I read somewhere (don't remember) that there's some special feature about the PCI Express 3.0 spec to where there's more encoding bandwidth with two way communication between CPU and GPU, can someone please confirm this? If PCI Express 3.0 doesn't bring anything new to the table, then the spec is not really necessary. No complaint though, we'll take what we can get.

[kl0012]

I read somewhere (don't remember) that there's some special feature about the PCI Express 3.0 spec to where there's more encoding bandwidth with two way communication between CPU and GPU, can someone please confirm this? If PCI Express 3.0 doesn't bring anything new to the table, then the spec is not really necessary. No complaint though, we'll take what we can get.

http://www.pcisig.com/news_room/faqs/pcie3.0_faq/#EQ1
Q: How does the PCIe 3.0 8GT/s “double” the PCIe 2.0 5GT/s bit rate?
A: The PCIe 2.0 bit rate is specified at 5GT/s, but with the 20 percent performance overhead of the 8b/10b encoding scheme, the delivered bandwidth is actually 4Gbps. PCIe 3.0 removes the requirement for 8b/10b encoding and uses a more efficient 128b/130b encoding scheme instead. By removing this overhead, the interconnect bandwidth can be doubled to 8Gbps with the implementation of the PCIe 3.0 specification. This bandwidth is the same as an interconnect running at 10GT/s with the 8b/10b encoding overhead. In this way, the PCIe 3.0 specifications deliver the same effective bandwidth, but without the prohibitive penalties associated with 10GT/s signaling, such as PHY design complexity and increased silicon die size and power.

[bamtan2]

the X79 supposed to provide up to 10 SATA3 Connections, so why we get only 2 ?

they changed it.

[sin0822]

how is this platform delayed? i thought it was on time for launch?

[bamtan2]

• ESET* Smart Security 5 Antivirus (WTF? THEY OWN MCAFEE!)

Also, x58 smackover has 2 esata and 2 more usb. What the hell intel! all that empty space on the back and there are 8 USB ports hidden as pins on the motherboard. cripes!

everyone has to buy this and run more stupid wires just to get access to what they already paid for: http://www.newegg.com/Product/Produc...82E16812200523

[IDANPE]

they changed it.

ty.
any link for that info ?

[Tenknics]

ty.
any link for that info ?

Use google. Do a random search on sandy-e. There will be a plethora of articles from sites you've never heard of. They all contain said cuts and delays and changes.

Its pretty easy, the internet.

[TheLostSwede]

Yes it's late and yes, Intel has royally screwed up.

As for the chipset, the current revision will be X79 and X79 only as the re-spin is fixing the SCU issues. The SCU is where the extra SATA/SAS 6Gbps ports are located and this is a must on the server/workstation versions of the chipset. It seems like Intel ran into similar issues that they had with the early 6-series chipsets, just with the SCU this time around.

As for the CPUs, steer clear until at least the C3 stepping which will be out in Q1 next year, as the C2 stepping is only going to be produced for the Core i7's as Intel broke the virtualization on the C2 stepping. There will be a very limited quantity of the C2 stepping, although I can't say exactly how limited. Then there's the thermal issues with the CPU's hitting their TDP with ease and then some. This will apparently be fixed in the D stepping. Overclocked CPU's are hitting 200W+ without even trying, so have fun with that on air cooling... Also, the stock cooler is a piece of junk, not a heatpipe in sight and makes more noisy than a small airplane...

[PatRaceTin]

I feel that people in my country can buy Socket 2011 before AMD FX !!!! (FX Still no retail)

[Tenknics]

Ive been waiting for sandy-e for awhile and to base my entire new system on. Last time I did this it was for prescott2m and we all know how that turned out. Not making the same mistake twice...The delay will give me more time to save and hopefully a new GPU or two will trickle out before the respin next year..(not holding my breath tho)

I think Im cursed. All the chips I ever decide I want end up having issues...I love my little arrandale but I never lusted it after it like I did Prescott2m and now Sandy-E..


:/

[M.Beier]

IMO the reader should note two things:
1) There is a bios on the download page dated TWO MONTHS AGO. We know we've been looking at X79 boards for a long time now. And Sandy Bridge has been out almost a year. So what is taking Sandy Bridge-E so long?
2) There were TWO Intel X79 boards talked about a few months ago (DX79SI and DX79TO), so where did the other one go?

This, combined with the recent removal of features from X79 boards, AND the rumors of SNB-E Xeon chips being delayed for a 2012 RESPIN, AND how close we are to 22nm Ivy Bridge tells me one thing: AMD isn't alone with their problems. This Intel platform (chip and chipset) has been delayed and delayed because it is having problems, and so now it is starting to look a bit behind the times. It is still the only new Intel platform with enterprise features, but that won't stop it looking a bit silly next to Ivy Bridge Z77/Z75 with native USB 3 and whatever else, running at 5ghz with a paper clip for a heatsink.

I might buy one anyway LOL! We'll see in two weeks.

First of all; it is not delayed...
Second what took it so long? I hear that a certain vendor was struggling with getting their lineup ready, one of great importance for Intel.

[Leeghoofd]

IMO the reader should note two things:
1) There is a bios on the download page dated TWO MONTHS AGO. We know we've been looking at X79 boards for a long time now. And Sandy Bridge has been out almost a year. So what is taking Sandy Bridge-E so long?

Mmmm I seem to have a far more recent version seen at the a certain site... dating from not so long ago when another competitor launched it's mainstream product...

[bhavv]

Where does it say november 15th? I cant find it in the links.

[Leeghoofd]

bios dates from one day earlier mate... still many issues to iron out...

[Systemlord]

bios dates from one day earlier mate... still many issues to iron out...

I'm not going to be an early adopter, let someone else be the Guinea pig!

[cmanser]

+1 on this - wait for further revision of chips and boards.

I'm not going to be an early adopter, let someone else be the Guinea pig!

[bamtan2]

Also, x58 smackover has 2 esata and 2 more usb. What the hell intel! all that empty space on the back and there are 8 USB ports hidden as pins on the motherboard. cripes!

everyone has to buy this and run more stupid wires just to get access to what they already paid for: http://www.newegg.com/Product/Produc...82E16812200523

rofl: http://www.atomicmpc.com.au/Gallery/...erboard.aspx/8

Strangely all the USB 2.0 ports that should've been at the rear I/O panel have wound up here as headers. Uh...I guess we can use expansion brackets for these. Very quaint choice.

They aren't even in order! 4, 1, 2, 3. It is almost as though something went wrong... like they lost their minds!

[sin0822]

how is the chipset late? Has it been delayed?

[AFQ]

It isnt coming on 14th?

[HelixPC]

Nice looking board, hope the performance is good.

[jaafaman]

One personality "quirk" with this MOBO (rig in sig):

The BIOS provides three modes for setting DRAM speed and timings - AUTO, Manual and XMP.

XMP will only give you the one single XMP profile that's stored in the SPD. Since there are only "set" speeds for the BCLK, changing the BCLK multi to 1.25 causes the IMC to drop one internal multi to keep the frequency as close to XMP as possible. But either because the speeds are in the same ballpark or no one foresaw a need, it does not alter any of the presented timings.

Using the AUTO setting yields the same type results but uses only the fastest SPD profile, which for my 1600 MHz XMP RAM is 1333 MHz.

The Manual setting puts you into a truly manual mode. You are free to change any of the presented parameters - voltage, frequency multiplier, timings - but you must do it starting from whatever point the last set mode gives you, whether that be the XMP or the SPD settings (unless you're making adjustments to your own manual profile).

Hmmm...

I've been able to coax the RAM to 1866 MHz using the 1333 MHz timings at 1T CR, raising the voltage from 1.35 V to 1.50 V, by changing the 9-9-9-24 timings to 10-10-10-27. Since I don't know what's being set for the timings that aren't being presented, I'm trying to coax 2000 MHz speeds out of the looser XMP timings, but may not be successful without raising the voltage up to 1.65. For reasons I'll get into when I do a full review of the platform, I don't want to go beyond 1.575 V, so this may preclude anything higher than the current 1866. But that's OK for now, as although I'm only getting 20 GB/s copy speeds per dual-channel set as compared to the 24 GB/s I was able to coax from the Lynnfield, the effects seem limited to just those apps that use only a single channel - things like the RAMDisk. And the throughput at 1866 MHz is still about 18% greater than the 1600 MHz XMP speeds from any angle.

What I'm wondering now is:

Is this firm "stepping" in supported RAM speeds common to the X79 or just this MOBO? If it's an across-the-board situation, then it strogly implies a need to outright purchase faster RAM rather than rely on an OC if you want the extra speed.

What are the experiences of some of you that are also sporting X79s but from different vendors?...