Intel's Core i7 870 & i5 750, Lynnfield: Harder, Better, Faster Stronger
by Anand Lal Shimpi on September 8, 2009 12:00 AM EST- Posted in
- CPUs
Homework: How Turbo Mode Works
AMD and Intel both figured out the practical maximum power consumption of a desktop CPU. Intel actually discovered it first, through trial and error, in the Prescott days. At the high end that's around 130W, for the upper mainstream market that's 95W. That's why all high end CPUs ship with 120 - 140W TDPs.
Regardless of whether you have one, two, four, six or eight cores - the entire chip has to fit within that power envelope. A single core 95W chip gets to have a one core eating up all of that power budget. This is where we get very high clock speed single core CPUs from. A 95W dual core processor means that individually the cores have to use less than the single 95W processor, so tradeoffs are made: each core runs at a lower clock speed. A 95W quad core processor requires that each core uses less power than both a single or dual core 95W processor, resulting in more tradeoffs. Each core runs at a lower clock speed than the 95W dual core processor.
The diagram below helps illustrate this:
| Single Core | Dual Core | Quad Core | Hex Core | |
| TDP |  |  |  |  |
| Tradeoff |  |  |  |  |
The TDP is constant, you can't ramp power indefinitely - you eventually run into cooling and thermal density issues. The variables are core count and clock speed (at least today), if you increase one, you have to decrease the other.
Here's the problem: what happens if you're not using all four cores of the 95W quad core processor? You're only consuming a fraction of the 95W TDP because parts of the chip are idle, but your chip ends up being slower than a 95W dual core processor since its clocked lower. The consumer has to thus choose if they should buy a faster dual core or a slower quad core processor.
A smart processor would realize that its cores aren't frequency limited, just TDP limited. Furthermore, if half the chip is idle then the active cores could theoretically run faster.
That smart processor is Lynnfield.
Intel made a very important announcement when Nehalem launched last year. Everyone focused on cache sizes, performance or memory latency, but the most important part of Nehalem was far more subtle: the Power Gate Transistor.
Transistors are supposed to act as light switches - allowing current to flow when they're on, and stopping the flow when they're off. One side effect of constantly reducing transistor feature size and increasing performance is that current continues to flow even when the transistor is switched off. It's called leakage current, and when you've got a few hundred million transistors that are supposed to be off but are still using current, power efficiency suffers. You can reduce leakage current, but you also impact performance when doing so; the processes with the lowest leakage, can't scale as high in clock speed.
Using some clever materials engineering Intel developed a very low resistance, low leakage, transistor that can effectively drop any circuits behind it to near-zero power consumption; a true off switch. This is the Power Gate Transistor.
On a quad-core Phenom II, if two cores are idle, blocks of transistors are placed in the off-state but they still consume power thanks to leakage current. On any Nehalem processor, if two cores are idle, the Power Gate transistors that feed the cores their supply current are turned off and thus the two cores are almost completely turned off - with extremely low leakage current. This is why nothing can touch Nehalem's idle power:
Since Nehalem can effectively turn off idle cores, it can free up some of that precious TDP we were talking about above. The next step then makes perfect sense. After turning off idle cores, let's boost the speed of active cores until we hit our TDP limit.
On every single Nehalem (Lynnfield included) lies around 1 million transistors (about the complexity of a 486) whose sole task is managing power. It turns cores off, underclocks them and is generally charged with the task of making sure that power usage is kept to a minimum. Lynnfield's PCU (Power Control Unit) is largely the same as what was in Bloomfield. The architecture remains the same, although it has a higher sampling rate for monitoring the state of all of the cores and demands on them.
The PCU is responsible for turbo mode.
Facebook
Twitter
RSS
343 Comments
View All Comments
tajmahal - Tuesday, September 8, 2009 - link
You fail to mention that Microcenter prices are for IN STORE PURCHASE ONLY. If you live about 6 driving hours away from a Microcenter like i do, then you're screwed.Chlorus - Tuesday, September 8, 2009 - link
Shhh! Don't spoil his self-righteous post with your troublesome facts!strikeback03 - Tuesday, September 8, 2009 - link
Does Microcenter have a limit on how many processors people can buy? If not, why isn't anyone buying these things and reselling them for less than the ~$280 that Newegg (and every other online retailer) do?Solema - Tuesday, September 8, 2009 - link
Anand,I know those prices you quoted are per-unit prices from Intel, but they are much more expensive than actual CPU costs. Given that I can get the following from Micro Center, and that I plan to overclock and run two 8800GTS 512's in SLI, what would you recommend?
i5 750 - $179
i7 920 - $199
i7 860 - $229
It still seems to me that the additional overclocking flexibility of the 920 (especially on stock voltage), coupled with the better multi-GPU performance would make that the best CPU to purchase, no? Given that P55 motherboards currently only retail for about $50 cheaper than many x58 boards, wouldn't the extra $70 cost for x58+i7 920 over a P55+i5 750 be worth it? You get better multi-GPU performance, better overclocking, better RAM performance, and future upgradeability to 6-core CPU's. What am I missing that would tip the scales in favor of the i5?
Pneumothorax - Tuesday, September 8, 2009 - link
the 920 for sure as you get a HT CPU for even cheaper than the 860. Both should overclock the same.dman - Tuesday, September 8, 2009 - link
So, do they support Hardware Virtualization? And don't give slack about not targeted at that market, specifically, does this support Windows 7 virtualization mode?I searched and didn't see it covered. I've read that the i5 and lower do not support vt-d, but, I'm not sure how that translates to Windows 7 "XP mode" support... I do need to review a bit more, would be nice if this was covered in the review.
I do know that this IS something that the Phenom family does support.
http://www.virtualization.info/2009/07/intel-core-...">http://www.virtualization.info/2009/07/...core-i3-...
ash9 - Tuesday, September 8, 2009 - link
No Virtalization!! That maybe huge for corporate setups,how did I miss that.
It should have been reported
has407 - Wednesday, September 9, 2009 - link
VT-x != VT-d. You want, and may need, VT-x. Most people don't need VT-d, much less know what to do with it or have a system that can make use of it--if you do, you're very unlikely to be using one of these CPUs.VT-x is Intel's name for processor virtualization features; it is part of the processor. All Core iX CPUs support it. VT-x is required for some hypervisors, (e.g., MSFT HYperV), but not all, although most (all?) require it for running 64-bit guests.
VT-d is Intels name for for IO virtualization (specifically "directed IO"); it is, or has been, part of the northbridge. For VT-d to be useful, you need a chipset that supports it; a MB/BIOS that supports it; and a hypervisor that knows how to use it. VT-d is primarily of interest to VM's that want to dedicate direct access to hardware by guests, and avoid the overhead of the hypervisor for that IO.
When you see "CPU X supports vT-d", it means the chipset for CPU X supports VT-d (the P55 supports VT-d). Whether MB/BIOS vendors choose to support it is another matter. Moreover, support for VT-d isn't simply yes or no; support varies by chipset (e.g., the P55, like the rest, support virtualizating a subset of interfaces).
In short:
1. Based on Core iX chipset capabilities (e.g., P55, X58), VT-d support is an MB vendor decision--not a function of the CPU model.
2. Which vendors support VT-d, and to what extent, is more often than not clear as mud, and the topic of much discussion in some threads.
3. If VT-d is important to you, you're probably running a heavy virtualized workload on an MP system with 10Gbe or very fast DAS--certainly not a Core iX. (Only exception of interest to others might be for access to GPU's by VMs)
What the new processors throw into the mix is an integrated PCIe controller, which also means an integrated DMA controller (at least I hope it does). Whether that supports VT-d is unknown (I haven't been able to find a definitive answer).
Gary Key - Tuesday, September 8, 2009 - link
VT-d is enabled on the i7/870-860. It is not enabled on the i5/750, just VT-x is available on it. I am working on Windows 7 XP mode as we speak.Jakall78 - Tuesday, September 8, 2009 - link
Reading this site for years, but there is something wrong going on here. Besides some slideshow pictures from Intel and 2-3 tests... there is nothing. That is not the way reviews are done. Look at the SSD reviews, THAT is a review(both of them actually). Now please look at this reviewhttp://translate.google.com/translate?prev=hp&...">http://translate.google.com/translate?p...mp;sl=ro...
and say it`s not better...
* I`m not making any false advertising here, I just found a better review.