You are here

Sandy Bridge, Part 4: Core i5-2400, i5-2500K and i7-2600K CPUs

Sandy Bridge CPUs: Intel Core i5-2400, i5-2500K and i7-2600K

January 24, 2011 by Lawrence Lee

Product
Core i5-2400

LGA1155 Processor
Core i5-2500K

LGA1155 Processor
Core i7-2600K

LGA1155 Processor
Manufacturer
Street Price
US$195
US$225
US$330

In previous articles we discussed Intel's Sandy Bridge architecture, the new integrated graphics chip, and the budget H67 and enthusiast P67 chipset/motherboards. Today we take a look at the real stars of the show, the CPUs themselves. We have three quad cores under scrutiny, the Core i5-2400, i5-2500K, and the flagship i7-2600K, with street prices of US$195, US$225, and US$330 respectively. The i7-2600K is a bit of a departure for Intel in that its price isn't ridiculously inflated. In the past when new Intel chips were released, the fastest model would usually be priced two to three times higher than the next chip down the line.

The main differences between the three chips are that the "K" series feature a fully unlocked multiplier for easy overclocking, and the i7 has Hyper-threading (eight threads total rather than four). All of them support Turbo Boost, which by default overclocks by 100 MHz when all four cores are active and 400 MHz when only one core is at work. So none of these CPUs ever actually run at their rated clock speed. AMD has a similar featured called "Turbo Core" on their Phenom II X6 line.

Desktop CPU Comparison (US$160~$350)
Model
Speed
L3
Cache
TDP
Features
Street Price
Phenom II X4

965 BE
3.4 GHz
6MB
125W
UL
$165
Phenom II X6

1055T
2.8 GHz
6MB
125W
TC
$180
Phenom II X4

970 BE
3.5 GHz
6MB
125W
TC,UL
$185
Phenom II X4

975 BE
3.6 GHz
6MB
125W
UL
$195
Phenom II X6

1075T
3.0 GHz
6MB
125W
TC
$200
Phenom II X6

1090T BE
3.2 GHz
6MB
125W
TC,UL
$230
Phenom II X6

1100T BE
3.3 GHz
6MB
125W
TC,UL
$270
Core i5-661
3.33 GHz
4MB
87W
HT,TB
$200
Core i5-660
3.33 GHz
4MB
73W
HT,TB
$220
Core i5-680
3.6 GHz
4MB
73W
HT,TB
$300
Core i5-750
2.66 GHz
8MB
95W
TB
$200
Core i5-760
2.8 GHz
8MB
95W
TB
$205
Core i7-870
2.93 GHz
8MB
95W
HT,TB
$290
Core i5-2300
2.8 GHz
6MB
95W
TB
$185
Core i5-2400
3.1 GHz
6MB
95W
TB
$195
Core i5-2500
3.3 GHz
8MB
95W
TB
$210
Core i5-2500K
3.3 GHz
8MB
95W
TB,UL
$225
Core i5-2600
3.4 GHz
8MB
95W
TB
$300
Core i7-2600K
3.4 GHz
8MB
95W
HT,TB,UL
$330
Core i7-950
3.06 GHz
8MB
130W
HT,TB
$295
Core i7-965 XE
3.2 GHz
8MB
130W
HT,TB
N/A
Tested processors are in bold.

UL = unlocked multiplier, HT = Hyper-threading,
TB = Turbo Boost, TC = Turbo Core

So where do the new processors fall in the big scheme of things? If you take into account CPU pricing alone, it appears that the i5-2400 is destined to do battle with AMD's slower hex cores and fastest quads, and Intel's own Core i5-750/760. The i5-2500K isn't much more expensive, hanging with the same crowd plus the Phenom II X6 1090T Black Edition, which also has an unlocked multiplier. The i7-2600K is in the same league as LGA1156's i7-870 and the low end offerings for LGA1366.

To give you a better idea of what the different CPUs will actually cost you, we added the prices of a typical motherboard to the processors we're comparing today. The prices were taken from all compatible Intel/Asus/Gigabyte/MSI DDR3 microATX and ATX motherboards at Newegg and averaged out to US$103 for AM3, US$117 for LGA1156, and US$147 for LGA1155. Adding this cost changes things up a bit. For example, as a platform, the X4 975 is actually on average US$44 cheaper than the i5-2400. Keep this in mind later on when we judge the value of each Sandy Bridge CPU.

TEST METHODOLOGY

Common Test Platform:

Intel LGA1156:

Intel LGA1366:

Intel LGA1155:

AMD AM3:

Measurement and Analysis Tools

Timed Benchmark Test Details

  • NOD32: In-depth virus scan of a folder containing 32 files of varying
    size with many RAR and ZIP archives.
  • WinRAR: Archive creation with a folder containing 68 files of varying
    size (less than 50MB).
  • iTunes: Conversion of an MP3 file to AAC.
  • TMPGEnc: Encoding a XVID AVI file with VC-1.
  • HandBrake: Encoding a XVID AVI file with H.264.
  • Photoshop: Image manipulation using a variety of filters, a derivation
    of Driver Heaven's Photoshop
    Benchmark V3
    (test image resized to 4500x3499).


1080p | 24fps | ~14mbps

x264 1080p: Spaceship is a 1080p x264 clip encoded from the
Blu-ray version of an animated short film. It features a hapless robot
trying to repair a lamp on a spaceship.

Testing Procedures

Our main test procedure is a series of benchmarks, timed tests of real-world applications. System power consumption (AC) is measured with a Seasonic Power Angel during these tests (an average of the first 10~15 seconds) as well as at idle, during playback of a 1080p H.264 encoded clip, and during full CPU load. To stress the CPU we use either Prime95 (large FFTs setting) or CPUBurn depending on which
produces higher system power consumption. The AC system power is then later converted to DC.

Certain services and features like Superfetch and System Restore are disabled
to prevent them from affecting our results. Aero glass is left enabled if supported.
We also make note if energy saving features like Cool'n'Quiet and SpeedStep
do not function properly.

Estimating DC Power

The following power efficiency figures were obtained for the
Seasonic SS-400ET used in our test system:

Seasonic SS-400ET Test Results
DC Output (W)
21.2
41.6
60.2
81.9
104.7
124.1
145.2
AC Input (W)
32.0
58.0
78.0
102.0
128.0
150.0
175.0
Efficiency
66.3%
71.7%
77.1%
80.3%
81.8%
82.8%
83.0%

This data is enough to give us a very good estimate of DC demand in our
test system. We extrapolate the DC power output from the measured AC power
input based on this data. We won't go through the math; it's easy enough
to figure out for yourself if you really want to.

Operating Voltages

Before we jump into our test results, please note the operating voltages
of the processors tested today as sort of a disclaimer. Different samples of
the same processor often run at slightly different operating voltages which
can affect energy efficiency. Different motherboard models do not apply the exact same core voltage either.

TEST RESULTS

System Power

Our first test is low load system power consumption, at idle and during video playback. While some users work their systems hard, many PCs are often left idling or used for mundane tasks like watching video or surfing the web for hours on end.

In his area, Sandy Bridge shows a slight improvement over Lynnfield/Clarkdale chips, using about 4W less than most of the LGA1156 lineup idling, and about the same power when playing video.

Power consumption on a theoretical full load was also impressive, about equivalent to that of the Core i5-750 and only cleanly beaten by the i5-661 and i5-680, a pair of dual core processors.

Thermals

For our thermal test, each CPU was paired with the Scythe Kabuto with its stock
fan spinning at ~800 rpm. AMD temperatures were taken using the motherboard's
sensor via SpeedFan and Intel temperatures were taking with RealTemp.



*Direct temperature comparisons between Intel and AMD processors should not be made.

Given their superb energy efficiency we are hardly surprised to find that the 2400/2500K/2600K all run much cooler than the LGA1156 and LGA1366 Core i7's.

PERFORMANCE

To test performance, we pit the processors against one another in a short series of timed tasks using real world applications. Power consumption was measured when applicable.

Not content to take the energy efficiency crown, the Sandy Bridge processors dominated Photoshop by a more than a healthy margin. The dual core Clarkdale i5-680 chip did very well in this test, as did Nehalem's Core i7-965, but both were felled by the new arrivals.

The NOD32 isn't a threaded test, favoring high clock speeds, but with Turbo Boost, the newcomers edge out the competition. The Phenom II X4 975 and X6 1100T put up a fairly good fight, but with unflattering power consumption. They pulled 50~60% more wattage than Sandy Bridge.

In WinRAR, Sandy Bridge beat out all comers once again, and did so with power consumption close to that of dual core i5-661/680.

Our iTunes encoding test favored the high speed Core i5 dual core chips and yet the Sandy Bridge offerings still won out in the end.

Video Encoding

Our video encoding benchmarks are representative of how well the CPUs respond to multithreaded applications. This is where quad and hex core processors typically shine.

In TMPGEnc, Sandy Bridge pulled out yet another victory, but with slimmer margins in both time to completion and energy efficiency.

In our last test, the three new high-end Sandy Bridge chips finally met some stiff competition from AMD. HandBrake is one of the few applications that utilizes all of the 1100T's six threads, giving very good results. Unfortunately, it also used about 50W more power.

Overall Performance

The Core i5-2400, i5-2500K, and i7-2600K dominated all portions of our test suite, an impressive feat considering there were both threaded and unthreaded tasks involved. The three completed our test suite in 74~82% of the time required by AMD's flagship Phenom II X6 1100T.

The total power consumed running our test suite was even more impressive than the pure performance. They were effectively twice as efficient as the Phenom II's and the i7-965.

SPCR Score

In our opinion, CPU selection should be based on a combination of performance, cost, and energy efficiency. With this in mind we created a weighted scoring system, with each aspect having equal worth. The best performing processor in each category received 33.33 points and the rest scored proportionally. The perfect CPU would ideally have a total score of 100. For the energy efficiency portion of score, we assumed that half of its operating life would be at low load (the average of idle and video playback consumption) and the other half at high load (the average consumption during our timed tests).

When all three aspects are taken into account, the i5-2400 and i5-2500K seem to be the best all-around CPUs on the market. The i7-2600K doesn't pack as much of a punch for its higher price, but it still scores very well. The Core i5-750 and dual core Core i5-661 are solid choices, depending on your usage pattern. Multitasking and threaded-applications obviously favor the quad core i5-750. As for AMD, they were struggling to remain competitive before the launch of Sandy Bridge, and now they've fallen further behind.

Not all of our readers will agree with our weighting system, but it gives you a good starting point. If you care more about performance, pad the score of the Sandy Bridge processors and the i7-870. If you believe energy efficiency is more important, knock the Phenom II's down a peg or two. If cost is an issue, bump the AMD chips up instead, as well as the LGA1156 chips.

FINAL THOUGHTS

With Sandy Bridge, Intel has set new standards in both performance and power consumption. Not only is the new hardware faster and more efficient, it is also not that expensive. These LGA1155 chips obliterate their comparably priced competition, with the Core i5-2400 and i5-2500K offering the best overall value. The i7-2600K offers a slight clock speed bump and Hyper-threading, but with most applications, the performance boost is generally not worth the US$100 extra over the i5-2500K. However as premium Intel processors go, the markup is fairly low. Also note that the "K" series reportedly overclocks very well.

With all this goodness, end-users will rejoice, but there are also those who will be ticked off. Enthusiasts who bet on LGA1366 (and paid a healthy premium doing so) are now stuck with hardware that is worth significantly less than what they invested. LGA1366 motherboards and processors still start at about US$170 and US$300 respectively, but LGA1155 has made the platform almost dead in the water unless serious price cuts are enacted. The only LGA1366 processors that haven't yet been cannibalized by Sandy Bridge are the uber-expensive six core/twelve thread models. Ultimately, however, even deep price cuts are not likely to help, nor are the likely to be in Intel's plans — the replacement for socket 1366 is scheduled to appear in Q3 this year.

Meanwhile AMD fans are seeing their horse fall even further behind in the race. The top-of-the-line Phenom II's can't touch Sandy Bridge's speed, nor can they compete in energy efficiency. There is a possible light at the end of the tunnel as AMD is slated to launch the new processor, Bulldozer, this summer. But given the prowess of Sandy Bridge, AMD has to knock it out of the park. This is especially true as Intel has yet another CPU line coming out in Q3 for socket LGA2011, the official successor to LGA1366. Currently, AMD on the desktop seems advisable only if you're on a tight budget.

Our thanks to Intel and AMD
and for Core i5/i7 and Phenom II X4/X6 samples used in this review.

* * *

Articles of Related Interest

Sandy Bridge, Part 3: Gigabyte P67A-UD4 & Intel DP67BG P67 Motherboards

Sandy Bridge, Part 2: Intel DH67BL & Asus P8H67-M EVO H67 Motherboards

Sandy Bridge, Part 1: Intel GMA HD 3000/2000 Graphics

AMD Athlon II X3: Affordable Compromise

Athlon II X4 610e & Phenom II X4 910e: 45W & 65W Quad Cores

Intel Core i5-661: A 32nm CPU with Integrated Graphics

* * *

Discuss
this article in the SPCR forums.

Google

www SPCR