Samsung RF711/Intel HM65 SATA problem

Posted: 2012-05-12

TL;DR

The Samsung NP-RF711-S07UK laptop is probably susceptible to the Intel SATA performance degradation problem looks like it’s susceptible to the Intel SATA performance degradation problem, but its BIOS is probably lying. If your model is affected then it should only be the optical drive bay that suffers; the second 2.5” bay (sold empty) should be unaffected. Either way, the warranty will likely have expired by the time you know.

In full

I just received my new Samsung NP-RF711-S07UK laptop through Amazon, which seems like a bargain (I will have paid £625 after the VAT rebate offer). Unfortuately, having poked around on it I’m reasonably sure it’s susceptible When I first poked around on it I was reasonably sure it was susceptible to the Intel HM65 chipset design issue (original press release) where the SATA revision 2.0 (3 Gbit/s) ports may degrade over time.

I haven’t yet confirmed this — Intel advise you to either check the chip marking (but I’m not going to take my laptop apart for this, nor ever remove the heatsink if it has one) or contact the manufacturer. Having researched the topic more thoroughly, it looks unlikely that any affected chipsets would have made it into laptops this late on; it looks more likely that the BIOS is fiddling the PCI device IDs for compatibility reasons. I e-mailed Samsung UK support on 2012-05-13 and I’m awaiting a response. However, all other signs currently point to it being susceptible.

Why I think the RF711 issuspected the RF711 was susceptible

Intel lists the affected PCI IDs — 8086:1c49 rev 04 — in its 6 Series Chipset Specification Update.

On Linux:

On Windows 7:

At the time Intel stated:

Only computer makers who have committed to shipping the Intel® 6 Series Chipset in PC system configurations that are not impacted by the design issue will be receiving these shipments.

 The plot thickens

Gnawing at my mind was the recollection that I’d read a later, more personal-sounding assurance that no buggy SATA ports ever got used, but it took me ages to find it in my brower history. Someone called Christian Wood, purportedly an Intel employee, said the same thing in September:

All the B2 board should have been recalled and sent back to us. The only way for a B2 to still be out there is if the company with unit permanently disabled the SATA II ports on the board. […] Just so you know I havent heard of anyone getting an old B2 board after the recall.

Although there are certainly revision 05 chipsets out there, some laptops apparently pretend to have revision 04 chipsets as a BIOS-based compatibility feature:

Lenovo has not shipped B2 Sandy Bridge silicon on production ThinkPad systems. The ID being referenced returns the Intel CRID – Compatibility Revision ID – of the chip, not the actual chipset revision.

I’ve already overwritten the Windows installation with Linux so I can’t run the suggested Intel utility (if you can and do, do tell me what it says).

Possible effects

As sold, only the Blu-Ray drive is attached to a SATA revision 2.0 port — perhaps by design — so that’s the only drive that shouldcould be affected. The 2.5” bay that is empty (as sold) is attached to a SATA revision 3.0 port and should be unaffected. Obviously this wouldcould also affect any other drive in that bay, so if you were planning to replace the Blu-Ray with an SSD then you might want to reconsider.

A few places mention the statistic that between 5% and 15% of susceptible chipsets are expected to fail in the first two to three years. With a very quick search I couldn’t find the original source of this statistic nor how it’s calculated. If true, very few people will be able to claim under the 1-year warranty this laptop is sold under in the UK.

My opinion

I expect that the laptop is affectedunaffected and lying to me. At that price, there had to be a catch, but I’m disappointed that Samsung weren’t more upfront about what it was. I was aware of the Intel recall before I bought the laptop, and had read about HP laptops shipping with the susceptible chipset even after the recall, but over a year later I find it incredible that any manufacturer would still quietly be shipping an affected board. I hope it’s my mistake and that I’ll do my homework properly (and be less cynical) next time. In my defence, I will say:

So I guess one of the following must be true:

Correspondence

I asked Samsung about this on 2012-05-13:

Subject: Intel HM65 product recall / SATA port reliability

Hi Samsung UK Support,

I recently purchased a (very competitively priced) NP-RF711-S07UK laptop from you via Amazon. Upon inspection, I think it may contain one of the defective Intel HM65 chipsets that Intel recalled March/April 2011 (see http://www.intel.com/support/chipsets/6/sb/CS-032521.htm ). I’d be grateful if you’d confirm whether or not this laptop’s HM65 chipset is susceptible. If it’s not susceptible I’d be grateful if you could shed any light on why it appears to contain the susceptible revision, and a practical way to demonstrate that it isn’t susceptible.

Intel’s “6 Series Chipset Specification Update” (http://www.intel.com/content/www/us/en/chipsets/6-and-c200-chipset-specification-update.html ) states that PCI ID 8086:1c49 revision 04 is the susceptible model. Windows 7’s Device Manager lists “PCI\VEN_8086&DEV_1C49&SUBSYS_COA5114D&REV_04” as the Hardware ID; similarly, on Linux, running ‘lspci -nn | grep HM65’ yields:

00:1f.0 ISA bridge [0601]: Intel Corporation HM65 Express Chipset Family LPC Controller [8086:1c49] (rev 04)

Back on Windows, the Gigabyte 6 Series SATA Check Utility states “Chipset: HM65”, “Revision: B2” and warns that there is a device on the problematic port.

Thanks in advance,

Isaac Wilcox

If I get a response I’ll post it here.