I spent a bit of time recently putting together technical documentation for Intel’s 4-series chipsets, partly motivated by research into Intel’s support of 4 GB DDR2 memory modules, partly driven by idle curiosity about one of Intel’s many hyped up and failed projects, Turbo Memory.
There are plenty of mentions of the Intel GM47 chipset. And that Turbo Memory article explicitly says that Intel’s Turbo Memory 2.0 is available on the Cantiga GM47 chipset as announced by Intel on July 15, 2008. Although the phrasing in the Wikipedia article may be misleading in that Turbo Memory 2.0 was probably available on the Cantiga (Mobile 4 Series) chipset and not just GM47. of course there’s no reference given for the statement… so who knows.
At any rate, there were certainly rumors that GM47 was to be released in 2008, and then in Q1 2009. But then 2009 came and there was no GM47 chipset. It appears to have gotten Intel’s typical second-class burial—a decree comes from up on high, the victim product is no longer mentioned, must not be spoken of, and references to it are scrubbed. Thus GM47 is notable only by its absence, say, here on Intel’s ark, or in the Mobile 4 Series Chipset Specification Update.
Perhaps the somewhat confusing status of the GM47 chipset’s existence is what led to curious statements like the one in this Italian Wikipedia article. “Cantiga (the 4-series mobile chipset) is marketed in three variants: GM47, GM45, GS45, PM45.” Now which one of the four listed was not actually marketed?
Was the GM47 chipset some kind of an elaborate hoax? No, of course not. Although a search engine indicates that “GM47” is not mentioned anywhere on Intel’s web site, except in community discussion forums. And even there, most mentions are of “GM45/GM47”, which is something that probably came from Intel’s own chipset INF files for Windows (as is common, several chipsets share PCI IDs for various built-in devices).
But in the past, Intel most certainly did mention the GM47 chipset; for example in the release notes for production Vista drivers version 188.8.131.524 (Sep 9, 2008) or 184.108.40.20609 graphics drivers (Nov 13, 2008). That indicates the GM47 chipset was more or less completed.
What’s the real story of the GM47 chipset then? I can only guess. GM47 was meant to be a performance mobile chipset for the Core 2 platform. It was probably delayed enough that Intel decided not to release it, since the Ibex Peak chipset was right around the corner and the Penryn CPUs could not compete with Clarksfield Core i7 processors on performance.
It is quite possible that laptop OEMs told Intel that they were not interested in releasing new “performance” GM47 based models in early 2009 only to have them obsoleted by Ibex Peak platforms circa half a year later.
The most interesting evidence of the GM47 chipset remains in Intel’s Mobile 4 Series chipset datasheet. For reasons that are not clear, that datasheet is currently (Feb 2021) nowhere to be found on Intel’s web site, although the previously mentioned corresponding specification update is.
There are of course enough copies elsewhere. In the datasheet’s revision history, on page 10 in the PDF, we see that in December 2008, document order no. 320122-003 “Added GM47 Feature Support”. But oops, the next revision 320122-004 from January 2009 “Removed […] GM47 Feature Support”! There must have been some confusion and last-minute decisions at Intel if the GM47 chipset was actually documented in the public datasheet and then deleted again the next month.
Sadly, the datasheet with Intel order number 320122-003 does not appear to have survived. That would be the best documentation of the phantom Intel GM47 chipset. Or does someone still have it?
The funny thing is that at the time DDR2 price was at its lowest, while DDR3 prices was higher (though not as high as in 2007).
I recall reading various reviews and user comments from circa 2007-2008 talking about how DDR3 was much more expensive and with the CPUs and DDR3 speeds then available, not really faster than DDR2 either. But Nehalem put an end to that, if you wanted a new PC, you had to have DDR3 🙂
Same with servers, the first-gen Nehalem server memory was insanely expensive. And at least in the batch we had in our server lab, a lot of those early registered DDR3 modules failed within a few years.
Very early DDR3 in 2007 had a large die size premium and was of course much more expensive, but even by 2008 this was ending.
1Gbit DDR3 based server memory was more expensive in 2009, but not insanely. 2Gbit DDR3 was much more expensive than 2Gbit DDR2 though, and this did not change until 2010.
I’m pretty sure the modules we had would have been 2Gbit DDR3, it was brand new stuff. And for the first time in ages, multi-socket Intel servers did not suck.
Surely you mean “sucked less” 🙂
DDR3 still had a significant price premium in April of 2008 when I built my Core2Duo system. Benchmarks at the time were showing no real speed advantage of going DDR3 vs. DDR2 either. In some cases it was actually slower due to slower timings on period DDR3. Over time that obviously changed.