E6520 Owner's Thread

That is right, but the Intel CPU lineup says 1600MHz for the 2720qm!
And the RAM is directly connected to the CPU without FSB.
Intel Core i7-2820QM - Sandy Bridge For Notebook PCs - Intel Sandy Bridge Comes to Laptops! - Legit Reviews

Intel Core i7-2820QM - Sandy Bridge For Notebook PCs - Intel Sandy Bridge Comes to Laptops! - Legit Reviews
So I am a little bit disapointed, but I think it will be changed to 1600MHz
by a future Bios update.

 

The 1333MHz came from minimal Dell.com documentation. It may not have included the quad CPU. Plus, it is not uncommon for Dell docs to fall short of actual system capability, particularly downstream as new specs eclipse initial specs.

So you could be right... a BIOS update might unleash the memory in quad applications.

GK

 

However, the higher bandwidth will be largely offset by higher latency (in terms of clock cycles, to give the same physical time).

John

 

I'm also very happy with the battery life. From my past experience, battery life estimate is way off unless you have used laptop for an hour or two. Try to browse for 3 hours and see what % of battery you have used. Or do one of the battery run down tests.

Next Monday I'm going to get another E6520 (with Windows installed), and I have a kill-a-watt power meter, so I should be able to post some power counsumption numbers soon.

 

Yes, you are right- I will do a longer test but first I will complete other tests and my installations.
By the way - directly at the power plug my E6520 draws about 20 Watt
(a mearure using my Wattmeter - a mystery to be solved in the future).
Offcause ther is the loss from the switching power supply, but 66% loss is not ok.
So either another power mode when using the plug or Windows numbers are wrong indeed.

Yes, if the software is very bandwidth relevant (e.g. Winrar) I do calculate the performace with
1333 CL9 = 100%
1333 CL7 = 104%
1600 CL7 = 106%
Source (German):
Test: Welcher Speicher für ?Sandy Bridge?? (Seite 13) - 28.02.2011 - ComputerBase

As most of real world applications are by far less bandwidth dependent, the effects come close to only 1% or even less.
So the CPU model is a lot more relevant compared to the 1333/1600 memory or the CL9/CL7 timing.
If I would have known before that 1600MHz is not working, I wouldn't have bought the more expensive memory.
Maybe someone here can use my results for his decision.

 

The best way to measure power consumption is to run on battery and then using a program (eg HWiNFO32) to report the power drain from the battery. You can then try different operating conditions and display brightness to see what affects the power drain.

John

 

Thank you, I am now using HWiNFO32 and it is doing a good job.
With the right mouse key Real-time graphs of the CPU-temperature can be selected.

For the last years I loved to use I8kfanGUI for CPU-graphs and the fan control at my DELL E6500, D830, D800. I used RmClock to optimize the voltage of the CPU. But both programs aren't supported and updated any more.

 

Can anyone please post the result of the Windows 7 Experience Index

(2) Graphics
Desktop performance for Windows Aero ?
(3) Gaming graphics
3D business and gaming graphics ?

I only get 4,7m (maybe because the internal Intel-HD-3000 via Optimus is selected)
It should be about 5,6 with the NVS 4200.
What is going wrong? I tried to select the NVS GPU at the Nvidia utility (no change).

 

I'm getting 6.4 for graphics, and 6.4 for gaming graphics. Thats with Optimus enabled - when it starts running the benchmark, the nvidia gpu kicks in

 

WEI for my e6520