Disclaimer: This is an ES CPU, retail may or may not have changes from this sample.
Goals and Objectives: The primary goal of this overclocking experiment is to complete a benchark of FutureMark®'s 3DMark06 with stable processor overclock, without lowering the multiplier from nominal 8x, no solder modding on the board and costly cooling requirements.
Success Indicator: There will be four(4) success indicators:
- Completion of 3DMark06 benchmark
- Reaching stable overclock on two cores
- Stable multi-threaded Prime95 with aircool heatsink for at least 5min
- Zero Physical Changes on the hardware (no solder mod, no lapping, etc)
The Intel® Core 2 Duo™ E6320 has the following specifications:
- sSpec: ES^
- Frequency: 1862 MHz (1.86GHz)
- Core: Conroe
- Process: 65nm
- Package: LGA775
- Core Voltage: 1.19v
- Bus Speed: 266MHz
- L2 Cache: 4096KB
- Multiplier: 7
- Stepping: B2-6
- Thermal Specification: 61.4°C
- Thermal Guideline: 65.0W
- Platform Compatibility: 06
- EM64T: Yes
- Execute Disable: Yes
- Hyperthreading Technology: No
- Virtualization: Yes
- Core per Package: 2
- L2 Cache per Core: Shared 4096KB
Test Set Up
Processor Intel® Core™ 2 Duo E6320(1.86GHz) Dual Core
Motherboard: Asus® P5B Deluxe BIOS 1011
Memory: Apacer Value DDR2-800MHz 2x1024MB(CL5-5-5-18 SPD)
Hard Drive: Seagate* Barracuda 120GB SATA RAID 0
Video Card: Asus® 7900GTX 512MB v93.74
Power Supply: Delta 670W EPS12V (server grade)
Ambient Temperature: 34c
For benchmarking, at least three (3) runs will be made for each benchmark. From the scores, the lowest and highest score will be discarded, reducing the random spike in results and ensuring better consistency of results. Note that more than three (3) benchmark runs can be executed to validate any inconsistency or abnormality that may arise.
For temperature gathering, Core Temp Beta 0.95 will be used along with SpeedFan v1.32 and the motherboard's own software. From the readings of all three software, assuming they all worked, an analysis will be done to derive the most accurate temperature reading. An external temperature reader, Coolermaster Aerogate II, will be used as well for additional reference. For multicore system, the temperature reading for both idle and load will be based on the highest reported temperature of all the cores.
Heatsink/Fan: Due to lack of funding, there will be no water cooling, or phase change from this Blog article and as such, part of the success indicator is just aircooled-based heatsink. The best I could muster is an aging Thermalright® XP-90 with a Delta® brushless fan model FFC0912DE (1.5a) and an LGA775 retention mechanism. I have no other heatsink available except the stock reference cooler. Thermaltake® Big Typhoon, Sunbeam® Tuniq Tower, Sycthe® Ninja Plus, Zalman® CNPS9700 or even Coolermaster® GeminII are all great heatsink, and arguably, the best right now in the market but availability and most importantly $$$ is scarce right now. But hey, I have used it in my Prescott days, so it should perform equally well (fingers crossed).
Motherboard: The motherboard selection is not much of a debate here, Asus® P5B Deluxe is selected as the motherboard of choice. While performance is far from equal compared to their i975 brethren and manual selection of bootstrap is not possible, the capability of this board to hit 400MHz FSB and beyond is unquestionable. However, like many other Asus® boards I have tried before it, this one also suffers with severe voltage drop of the vCore. This will be remedied with a vDroop pencil mod.
Memory: Apacer® memory will be used for now, as the ample size of 1GB on each module will hopefully be enough to boost 3DMark06 score. While this may not seem like a good choice, 400MHz FSB is really not that low, considering the size of RAM and timing that go with this RAM.
Hard Drive: Two Seagate® 160GB SATA/3Gbps/8Mb will be used and configured as RAID 0 using Intel® Matrix RAID configuration with their ICH8R chipset. Not the fastest, but it's the most reliable and compatible that I have used so far. The hard drive will be partitioned in two, one will be for the OS installation, and the other partition will just be for miscellaneous data.
Video Card: Since 3DMark06 is targetted, I decided that an Asus® 7900GTX/512MB is good enough, besides, that's the best I have right now since my 8800GTX is already returned from loan. The driver is the latest (dated November 2006) downloadable for this video card at the time of testing. No tweak is done on this video card, and operated on stock. However, I will be tweaking this if score is ruining the magnificence of the processor.
Power Supply: A Delta® 670w model DPS-670DSB REV: S2. It's a full featured power supply and actually stable enough to run even server processors. It's an industrial grade power supply, and would be more than enough for my specifications.
Chassis: The testing will be done on an open test bed so there will be no chassis.
::Overclocking and Benchmarking Notes::
There's not much sense in running this processor on stock, as I have already put this processor to several tests, however, the stock benchmark will be used to compare the results. The initial overclock will be pushed to 50% above nominal, so that will be at 2.793GHz. Not that high, but at this clock speed, the memory will already be reaching its stock speed of 400MHz or DDR2-800MHz (well, actually 50% is 399MHz so it's 798MHz). This will be good to start benching away and produce some good numbers.
Memory timing is set to CL4-4-4-15, with a measly 2.00vDimm. A fan is already put in place on top of the memory just to make sure I don't forget cooling them.
I will be focusing on 3DMark06 for now, and later on, I will update this Blog post for other benchmark results. Note however, that since I am not a Pro* user of FutureMark benchmarks, each time I submit a new score, the link will not be updated. In that case, check the latest Blog post on this subject matter to find out the latest score. This will happen since I will be finding the best settings to achieve a good 3DMark06 score.
Doom3: Low Quality-640x480
Cinebench: Render Score
Cinebench: Render Time
SANDRA: RAM Bandwidth
Loaded Temperature Screenshot
::Closing Thoughts with 50% Overclock::This is an early benchmark, and the score will continue to evolve as I try to find the sweet spot with the processor, motherboard, memory, and of course, video card. For now, enjoy the score, no matter how low it may seem to be. I will post the next installment of the journey in the next upcoming blog posts...