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
Overclocking Lynnfield at Stock Voltage: We're PCIe Limited
Remember the on-die PCIe controller? Yep. It's to blame.
Lynnfield is Intel's first attempt at an on-die PCIe controller and it actually works surprisingly well. There are no performance or compatibility issues.
The on-die PCIe controller needs more voltage as you overclock Lynnfield, limiting Lynnfield's stock vt overclocking potential.
Unfortunately the PCIe controller on Lynnfield is tied to the BCLK. Increase the BCLK to overclock your CPU and you're also increasing the PCIe controller frequency. This doesn't play well with most PCIe cards, so the first rule of thumb is to try and stay at 133MHz multiples when increasing your BCLK.
The second issue is the bigger one. As you increase the BCLK you increase the frequency of the transistors that communicate to the GPU(s) on the PCIe bus. Those transistors have to send data very far (relatively speaking) and very quickly. When you overclock, you're asking even more of them.
We know that Bloomfield can easily hit higher frequencies without increasing the core voltage, so there's no reason to assume that Lynnfield's core cannot (in fact, we know it can). The issue is the PCIe controller; at higher frequencies those "outside facing" transistors need more juice to operate. Unfortunately on Lynnfield rev 1 there doesn't appear to be a way to selectively give the PCIe transistors more voltage, instead you have to up the voltage to the entire processor.
Intel knows the solution to Lynnfield's voltage requirement for overclocking, unfortunately it's not something that can be applied retroactively. Intel could decouple the PCIe controller from BCLK by introducing more PLLs into the chip or, alternatively, tweak the transistors used for the PCIe interface. Either way we can expect this to change in some later rev of the processor. Whether that means we'll see it in the 45nm generation or we'll have to wait until 32nm remains to be seen.
The good news is that Lynnfield can still overclock well. The bad news is that unlike Bloomfield (and Phenom II) you can't just leave the Vcore untouched to get serious increases in frequency.
Facebook
Twitter
RSS
343 Comments
View All Comments
yacoub - Tuesday, September 8, 2009 - link
lol, what a stupid comment. yes it's "cheating" to benchmark the processor the way it comes out of the box, which also happens to be how it is used in the real world environment.Voo - Tuesday, September 8, 2009 - link
Well there are many users who don't bother with overclocking so the tests aren't "illegal" or anything.But I tend to agree that most users who would be interested in buying an i7 920 or i7 860 would overclock it, so turbo mode wouldn't help at all, as we see with the OC results.
I'm curious if PCI-e on die is the only problem and if we'll see new chips who benefit from turbo mode even when overclocked. After all the principle behind turbo mode doesn't change if you overclock, does it?
james jwb - Tuesday, September 8, 2009 - link
IF that's true, i'm not at all happy with this review. But i'll wait for someone else to confirm this for obvious reasons... anand, confirm!Voo - Tuesday, September 8, 2009 - link
You read the text, didn't you? It was mentioned several times..james jwb - Tuesday, September 8, 2009 - link
i don't have time to read through all of it right now, was just flicking through and immeditaly thought to ask the question. I will read it fully later on, though.Hence why i asked the question. You say "it", as in which way, benches had turbo, benches didn't?
snakeoil - Tuesday, September 8, 2009 - link
yes again, turbo was on for all the benchmarks which is illegal and biased.maxxcool - Tuesday, September 8, 2009 - link
yes, the federal government says making a feature that makes your product better is legal.JarredWalton - Tuesday, September 8, 2009 - link
Illegal and biased? Yes, Intel is illegally making their CPUs run better at all workloads for normal users that don't overclock. Someone should arrest them! What would be biased is to test these CPUs in a fashion that artificially limits performance. Sure, it would be nice to see performance compared with and without Turbo enabled, but generally there's not enough time to run every potentially interesting test scenario.snakeoil - Tuesday, September 8, 2009 - link
there you go, finally you said it.all the benchmarks have at least 600 mhz over the processor's stock speed.
that is outrageous, then if you want to compare the result with phenom 2 you have to overclock phenom 2 at least 600 mhz over stock speed.
just to be fair
Anand Lal Shimpi - Tuesday, September 8, 2009 - link
The processor's stock speed is variable according to the workload it's running, that's what turbo mode does. AMD will enable similar functionality in 2011. This is the out-of-box performance of Lynnfield. Turbo mode is a feature of the processor as it has been since the mobile Penryn days (and more recently Nehalem). There's no reason to disable it as no end user would, unless you want to make Intel look worse for some reason.We also ran Turbo on vs. off numbers in the review: http://anandtech.com/cpuchipsets/showdoc.aspx?i=36...">http://anandtech.com/cpuchipsets/showdoc.aspx?i=36...
Take care,
Anand