After a ton of speculation, a bit of denial, and a ton of testing we finally have our answer. The thermal interface material change made by Intel when it went from Sandy Bridge to Ivy Bridge is indeed the cause of the excessive temperatures we’ve observed while overclocking. The first accusation was made in late-April by Overclockers.com, however proving it wasn’t easy. The Japanese division of PC Watch somehow managed to remove the integrated heat spreader from a Core i7 3770k, along with the stock binding and grease. They then proceed to replace it with aftermarket alternatives, and the results speak for themselves. 

The findings are undisputable. By swapping out the TIM with OCZ Freeze and Coolaboratory Liquid Pro, they were able to lower the temperatures by up to 18% off a stock clocked chip, and a whopping 23% when overclocked from 3.5GHz to 4GHz. Chips with the aftermarket TIM we also able to sustain higher core voltages, which of course will lead to better overclocking potential.

The report concludes that Intel’s decision to move away from the fluxless solder used in Sandy Bridge has unfortunately crippled the overclocking potential of this generation of CPU’s. Its possible Intel may eventually change the design back, however with AMD lagging pretty far behind these days they really don’t have much incentive to do so. Enthusiasts really only have one place to shop for CPU’s these days.     

Ivy Bridge is still an amazing push forward performance wise, it’s just unfortunate that the potential for even more is there and we can’t get at it.

(Image Credit = PC Watch)

Sony suffered through its worst year ever in 2011, and not just because of the high-profile hacker attacks that compromised millions of user accounts and resulted in extended downtime to the PlayStation Network. The real reason Sony can lament 2011 is because of the fact that it posted a net loss of $5.7 billion, the company’s largest loss ever in its 66-year history, and the fourth straight year of sitting in the red.

If there’s a silver lining to the dark and gloomy cloud, it’s that Sony had previously forecast a loss of $6.5 billion. Overall, however, it was an obviously crummy year for the electronics juggernaut, and not for any single reason. Sony blamed the poor performance on foreign exchange rates, earthquakes and tsunamis in Japan, floods in Thailand, and general “deterioration in market conditions in developed countries.”

According to VentureBeat, Sony CFO Masaru Kato indicated that “This year remains crucial for a recovery in our electronics business. A fifth straight year of losses should never be tolerated.”

Investors reacted negatively to the news with shares of Sony on the Tokyo Stock Exchange sliding 6.43 percent to 1,135 yen, a 38 percent drop from this year’s high of 1,832 yen back in March, and the lowest it’s been in three decades, The Register reports.

The full financial report is available here (PDF).

Image Credit: Flickr (Jami3.org)

Today we are looking at two of the latest graphics cards from HIS, the confusingly named HD7870 IceQ Turbo and the IceQ X Turbo X. Both of these cards feature custom cooling solutions and are supplied in a pre-overclocked state. We have been impressed with the IceQ coolers in the past, especially in regards to low noise levels, so we have hopes that these cards will prove a tempting solution for the discerning enthusiast gamer.

HIS have some of the most dramatically designed coolers on the market. While companies such as XFX and MSI opt for distinguished metallic coolers, HIS have designed their ICE-Q coolers in the past with bright plastic aqua blue and transparent colour schemes. The appearance will certainly split opinion, however technically they are actually in the top 10% of cooling solutions on the market.

Above, the HIS HD7870 ICEQ Turbo and ICEQ X Turbo X. No longer have they bright blue semi see-through coolers. Good move, we say.

The HIS HD7870 ICEQ Turbo and ICEQ X Turbo X are both shipped with a 100mhz core clock increase from 1,000mhz to 1,100mhz. The ICEQ X GDDR5 memory also receives a boost, from 1,200mhz (4.8Gbps) to 1,250mhz (5.0Gbps).

Nvidia today rolled out the welcome mat for the newest addition to its Kepler family, the GeForce GTX 670. The new 670 is “engineered from the same DNA as the recently announced GTX 680,” but is a more affordable part with prices starting at $399 for cards built around Nvidia’s reference design. And according to Nvidia, the 670 is a full 45 percent faster in gaming performance than the closest competitive product (i.e., AMD’s Radeon HD 7950).

“Plus, the GeForce GTX 670 ties the competition’s much higher-priced flagship product on 25 of the world’s most popular games and benchmarks, a testament to the overall performance efficiency of the Kepler architecture,” Nvidia claims.

In other words, the GTX 670 is all that a bag of chips, in Nvidia’s eyes. Performance claims aside, the GTX 670 sports 1,344 CUDA cores, 112 texture units, and 32 ROP units. It has 2GB of GDDR5 memory clocked at 6,008MHz on a 256-bit bus resulting in 192.2GB/s of memory bandwidth. The GPU has a base clockspeed of 915MHz and a boost clockspeed of 980MHz.

For comparison, the GTX 680 features a few more CUDA cores (1,536), more texture units (138), and a faster GPU (1,006MHz base, 1,058MHz boost). The GTX 680 is also a longer graphics card; the GTX 670 measures 9.5 inches long.

Technically, the GTX 670 is available to purchase now, but like all Kepler cards, that’s contingent on being to find the darn thing in stock.

Image Credit: Nvidia

The battle for the high end cooling throne has reached fever pitch this year. The Noctua NH D14 ruled supreme for a long time, although it recently had to make way for a new king, the Phanteks PH-TC14PE. Today we are looking at the latest flagship cooler from Thermaltake, The Frio Extreme. Is this a serious challenger for the top spot?

Thermaltake have released a plethora of high end coolers in the last couple of years, including the original FRIO, the Frio OCK, Frio Advanced and the latest Frio Extreme which we are reviewing today.

The Thermaltake Frio Extreme uses the high end construction of combining two tower heatsinks with a dual fan cooling system.

The Frio Extreme highlights that Thermaltake are pulling out all the stops. They are using two large 140mm fans which can be adjusted with the supplied fan controller.

Additionally they have finally ditched the plastic shroud, opting for a naked dual tower heatsink methodology.

We will compare today against the class leading Noctua NH D14 and the Phanteks PH-TC14PE in a heavily overclocked state.

Features:

Ultimate Over-clocking Design Structure, supports up to 250W

• Dual tower heat-sink with 0.4mm aluminum fins provide large surface for heat dissipation.
• 6 x Ø6 mm-U-shape copper heat pipes accelerate heat conductivity.
• Mirror-finished copper base, provide perfect contact with CPU.
• Premium thermal grease maximizes heat transfer from the CPU onto the copper base for faster dissipation.

The Combination of VR and PWM Fan Control

• Dual 14cm high performance blue blade designed fans, spins from 1,200 to 1,800RPM.
• Combination of VR and PWM functions, switchable upon user’s preferences.
• Tool-less and Easy installation design for quicker disassemble and assemble the fan module.

Universal Socket Compatibility & Accessory Package

• All-in-one back-plate design, support all Intel and AMD platform
• Universal socket support :
  Intel: LGA 2011, 1366, 1155, 1156, 775
  AMD: FM1, AM3+, AM3, AM2+, AM2