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
Jamahl - Tuesday, September 8, 2009 - link
Digusting! How much money did intel bung you for this disgrace?strikeback03 - Tuesday, September 8, 2009 - link
How is that disgusting? It is the stock configuration of the processor. They are not doing all this testing as an e-pissing contest of who has better performance per clock, it is a comparison of retail products in real-world applications. If (and according to the review, when) AMD has something similar, I'd imagine they will test with that turned on too.Jamahl - Tuesday, September 8, 2009 - link
Why not benchmark the Phenom 2 with fusion for gaming anand???Gary Key - Tuesday, September 8, 2009 - link
"Why not benchmark the Phenom 2 with fusion for gaming anand??? "Have you actually tried using Fusion with Windows 7 x64? It is a total mess. I will be happy to show some results with it, most will say DNF, but that might not make you happy. ;) That said, AMD is working on it, especially trying to get it to play nice with AOD.
In the meantime, here is the current list of items to watch out for - http://game.amd.com/us-en/drivers_fusion.aspx?p=3&...">http://game.amd.com/us-en/drivers_fusion.aspx?p=3&... .
Jamahl - Tuesday, September 8, 2009 - link
What? It's AMD's fault that an unreleased 64-bit os is causing issues with their software?How many people are using Window 7 64 bit who visit this website?
Gary Key - Tuesday, September 8, 2009 - link
It is just as bad in 64-bit Vista. I imagine a fair amount of people that visit the site are using the RTM version of Win7.Jamahl - Tuesday, September 8, 2009 - link
were you benchmarking a processor at 2.66 gigahertz or a processor at 3.2 gigahertz?Gary Key - Tuesday, September 8, 2009 - link
Do you and SnakeOil live in the same house or is the IP address just that similar? LOL.. The processors were benched as they come out of the box. For Lynnfield that means turbo was on as we stated in the article and here in the comments. For Bloomfield, that also means turbo was on, just as it comes out of the retail box.If you check all the other reviews on the web at the main sites, everyone tested with turbo on in the primary benchmarks. So I guess you can say there is a huge conspiracy between us to actually utilize the processors as Intel intended for the users.
Apparently, we all failed at covering it up, so congratulations on discovering the Freeturbomasons. A now not so secret fraternal organization bent on world domination through the use of turbo frequencies inside processors carrying the blue "i" logo.
For the AMD Phenom II x4 series, they were benched with all cores enabled just as they come out of the box, even though you can disable each core in the BIOS just like you can disable turbo on the i7/i5. I guess to make things fair, we should disable the cores on the 965 BE as having that "feature" turned on is cheating.
Anyway, thanks for making my day, I needed some much deserved laughter. :)
Chlorus - Tuesday, September 8, 2009 - link
Seeing as how Fusion isn't even out yet, that would be hard to do...go troll somewhere else. What is with the idiots coming out of the woodwork on this post? You've got the standard fanboys, as well as insecure LGA-1366 owners who feel the need to defend their purchase, and insecure purchasers of 1156 products who are afraid their choice might be bested in a benchmark somewhere.Roland00 - Tuesday, September 8, 2009 - link
Would you please rerun page 9 with an overclocked 975 and the 870. I am wondering how much the difference will grow when the gpus are fed more information due to the faster cpus. Something like 3.8 to 4.0 ghz on both cpus with turbo off (a good overclock yet not in the unreasonable area)If you are investing 400+ dollars in gpus, and you are building it yourself you are probably going to overclock.