Page 1:Overclocking And Efficiency Go Hand-In-Hand
Page 2:Intel Core i7-2600K For Mainstream Overclockers
Page 3:Turbo Boost 2.0 And The PCU Manage Overclocking
Page 4:Overclocking Settings
Page 5:Test Setup And Benchmark Settings
Page 6:Benchmark Results: Audio/Video
Page 7:Benchmark Results: Office/Graphics/Rendering
Page 8:Benchmark Results: Archiving
Page 9:Idle/Peak Power Consumption
Page 10:Efficiency Using One Core
Page 11:Efficiency Using Multiple Threads
Page 12:Combined Single/Multi-Thread Efficiency
Page 13:Overall Overclocked Power Efficiency
Page 14:Conclusion: Overclocking Becomes Efficient
Conclusion: Overclocking Becomes Efficient
This article doesn’t set out to look at the highest clock rate you can achieve on a Sandy Bridge-based processor. For that, we'd want to use more aggressive cooling, higher voltages, and generally throw the entire efficiency story out the window completely. At the high end, existing BIOSes support 5700 MHz through a 57x multiplier and a bit more through modest BCLK increases. This is the top end for now, though Intel's engineers have told us that they intend to edge the multiplier limits up slightly.
Practically, though, you should be able to reach anywhere between ~4.5 and roughly 5 GHz on air cooling with all Core K-series processors based on the 32 nm Sandy Bridge architecture.
There are three conclusions we can draw from this article:
1. Sandy Bridge overclocks well
It certainly didn’t take this article to find that Sandy Bridge overclocks very well, at least as long as we’re talking about the K-series processors in Intel's Core i5/i7 portfolio. Going beyond 4 GHz is easily possible, even without a voltage increase, and our processor sample scaled reliably all the way up to 5 GHz on the standard Intel cooler.
2. Overclocking no longer trades efficiency for performance
While all previous processor generations caused an increase in power consumption that was always more substantial than the added performance (especially at higher, more difficultly-achieved frequencies), Sandy Bridge is the first processor architecture where clock speed and power consumption scale almost linearly.
Effectively, this means that your overclocking attempts are neutral to total power consumption. If you speed up the processor, you will consume more power, but the workload will terminate quicker, which in the end saves time. Stock speeds and the overclocking settings both require a very similar amount of total power. This is made possible through persistently low idle power consumption paired with great performance per clock.
3. Overclocking can be fool-proof now
There is a paradigm change going on, as performance is no longer strictly defined by clock rate, but by the defined processor power limit. Once you understand that this power limit is a neat way of keeping a K-series Core i5/i7 processor within a thermal envelope, you understand that overclocking governed by the power control unit is like adding a safety net to your overclocked setup. Provided that your CPU cooler is aligned to your selected power limit, you can crank up clock speeds and you will typically get a very reliable configuration that automatically decreases clock speeds once it fully utilizes the defined power limit.
Intel’s next architecture step will be a shrink of Sandy Bridge into 22 nm, code named Ivy Bridge. We don’t expect it to introduce fundamental architectural changes, but I’m definitely curious if Intel can continue to improve power efficiency. Will the 22 nm Haswell architecture, to follow Ivy Bridge, possibly ignite a new clock speed race because it might make sense again from an efficiency standpoint? What’s your opinion?
- Overclocking And Efficiency Go Hand-In-Hand
- Intel Core i7-2600K For Mainstream Overclockers
- Turbo Boost 2.0 And The PCU Manage Overclocking
- Overclocking Settings
- Test Setup And Benchmark Settings
- Benchmark Results: Audio/Video
- Benchmark Results: Office/Graphics/Rendering
- Benchmark Results: Archiving
- Idle/Peak Power Consumption
- Efficiency Using One Core
- Efficiency Using Multiple Threads
- Combined Single/Multi-Thread Efficiency
- Overall Overclocked Power Efficiency
- Conclusion: Overclocking Becomes Efficient