How fast is the Intel 10th generation processor

Ice Lake: Intel's first 10th generation CPUs with strong graphics performance

Sebastian Schenzinger, Mark Hachman, Gordon Mah Ung

Intel has presented its first Ice Lake CPUs. With a very strong integrated graphics chip, which in ultrabooks should come close to the performance of discrete graphics chips. Plus: New manufacturing process, new architecture and a new graphics engine combined with “new” technologies in the areas of communication and I / O.

EnlargeEleven new mobile CPUs with Intel Ice Lake.

Intel has introduced its first 10th generation processors (code name: Ice Lake). We summarize the most important information: Eleven new processors, most of them are quad-cores with eight threads and a TDP of 9 watts, 15 watts or 28 watts. The standard clock rates range from 700 MHz to 2.3 GHz, with a maximum turbo clock of 4.1 GHz. The clock rate of the integrated graphics chip is always around 1 GHz market with 32, 48 or 64 EUs and increasing performance. We already have the first comparison values ​​for the graphics performance.

EnlargeIntel ice Lake basic specifications

nomenclature

Since the old 14nm architecture is still available on the market, it is not that easy to find your way around the product names. With the 10th generation, Intel is introducing a new nomenclature. There are still Core i3, i5 and i7 processors to define the performance levels. The following sequence of characters begins with a 10, to illustrate the 10th generation. The previous designation of the U and Y series, which indicated the TDP level, is no longer applicable. The fourth number now stands for this: “5” stands for the previous U series and “0” for the previous Y series. The graphic level is indicated with the designations G1, G4 and G7 from Entry to High-End. G4 and G7 stand for the new graphics chips of the Iris Plus generation.

EnlargeIce Lake nomenclature in detail

The lineup in detail

All new Ice Lake processors rely on Wi-Fi 6 and Thunderbolt 3 support. Up to four Thunderbolt ports can be installed, two on each side of the device. The 11th generation Intel graphics chips are divided into three performance groups according to the number of EUs installed (32, 48 or 64). EUs are Intel's programmable shader units for 3D rendering, computing or media functions. The performance level of the graphics chips differs primarily in the number of built-in EUs and less in the clock rate of the GPU.

EnlargeFull specifications for Ice Lake

The processors of the Y-series (with 0 in the name) can nominally allow themselves up to 9 watts, with a margin of up to 12 watts. The U series (5 in the name) may normally use 15 watts, the maximum limit here is 25 watts. The only exception is the i7-1068G7 with a TDP of 28 watts. The level 3_Cache is still determined by the processor level: i3 = 4MB, i5 = 6MB and i7 = 8MB. According to a representative from Intel, these eleven processors comprise the entire Ice Lake lineup, which is due to hit the market this year.

Even if we don't know how much the device manufacturers have to pay for the individual CPUs, we still got the first figures about the prices from Intel:

  • Core i7-1065G7: $ 426

  • Core i5-1035G7: $ 320

  • Core i5-1035G4: $ 309

  • Core i5-1035G1: $ 297

  • Core i3-1005G1: $ 281

As a comparison: The 8th generation Whiskey Lake processors were introduced for $ 409 for the i7-8565U and $ 297 for the i5-8265U.

Differences from Whiskey Lake and Amber Lake

The most noticeable change is the increased TDP value of the low-power Y series from 5 watts for Amber Lake to up to 9 watts for Ice Lake. It may be because there were very few devices that relied on Amber Lake Y-series CPUs and that the increased performance level wanted to appeal to more end customers. In contrast to the predecessor Whiskey Lake, the standard clock rates of the 10th generation are lower, the boost clock is significantly higher. That means the increase in performance depends primarily on how long the CPU can maintain the boost clock, and thus directly on the cooling performance of the device. With the new processors, the number of supported Thunderbolt connections increases from two to four and the transfer rate of the WLAN increases thanks to the Wi-FI 6 module. The control of external monitors has also improved, there are options for connecting either three monitors with 4K and 120 Hz, two monitors with 5K and 60 Hz or one 8K monitor with 30 Hz.

performance

Intel has provided our colleagues at PC-WORLD with a Software Developer System (SDS) for this test. A 256GB SSD Intel 7600P, 8 Gb LPDDR4X / 3733 RAM in dual Chanel mode and a Core i7-1065G7, which can be operated in both 15-watt and 25-watt mode, are installed. The processor has four cores with Hyper-Threading, a base clock of 1.3 GHz and a maximum boost clock of 3.9 GHz. The graphics unit clocks at 1.1 GHz and has 64 EUs. The L3 cache is 8MB. A simple single heat pipe solution with two fans is installed in the test model for cooling. The test model is compared with a Dell XPS 13 9380 and an HP Specter x360 13 - both with an i7-8565U.

EnlargeCooling solution of the Software Developer System

Rendering

In the Cinebench R15 multi-core benchmark, the Core i7-1065G7 calculates about six percent faster than the i7-8565U in the Dell XPS 13 9380 and a full 24 percent faster than the i7-8565U in the HP Specter x360 13. Here you can see very well how the performance strongly depends on the cooling capacity. In the Cinebench R20 Multi Core, the computing power is five percent slower than the Dell device, but it beats the HP Specter by 41 percent. The i7-1065G7 with 15W TDP also manages to overtake the i7-8565U in the HP Specter with eight percent more power. In terms of single-threaded performance, the test device with 15 W TDP (184 points), with 25 W TDP (181 points) and the Dell XPS 13 8380 (183 points) are virtually on par.

Geekbench

One of the greatest advantages of the Ice Lake generation is the significantly increased bandwidth of the RAM. The clock rates increase from 2133 MHz for LPDDR3 to 3733 MHz for LPDDR4X. In the Geekbench we achieve a performance increase of 34 percent. The integrated graphics unit primarily benefits from a faster working memory because it does not have its own graphics memory, but uses the working memory.