Nvidia Tegra X1 vs Apple A10 Fusion

Nvidia Tegra X1

► remove NVIDIA X1

The NVIDIA Tegra X1 (Tegra 6, Codename Erista) is a 64-Bit high performance ARM based SoC (System on a Chip) for (mainly Android based) tablets and embedded systems (like cars). It contains four ARM Cortex A57 and four ARM Cortex A53 cores in the processor part. As a graphics card, Nvidia integrated a Maxwell based GPU with 256 CUDA cores (see for more details and benchmarks). The integrated memory controller supports LPDDR3 and LPDDR4 with a maximum bandwidth of 25.6 GB/s (2x 32 Bit LPDDR4-3200). The Tegra X1 is the successor to the exactly one year earlier announced Tegra K1 SoC.

Similar to the Symsung Exynos 5433, the Tegra X1 uses a big.LITTLE combination of four power saving and small Cortex-A53 cores and four powerful Cortex-A57 cores. Nvidia however uses an own cluster migration solution instead of the ARM Global Task Scheduling (GTS) and therefore only one of the two quad-core clusters is active at a time.

The performance of the processor cores should be in the high-end segment of ARM based SoCs. Nvidia states that the X1 offers a twice as high energy efficiency compared to the Exynos 5433 (same ARM cores at 1.9 GHz manufactured in 20nm).

Due to the big Maxwell graphics card the SoC is mainly aimed at tablets with good cooling solutions and wont be commonly seen in smartphones. The Tegra X1 is produced at TSMC in 20nm.

Apple A10 Fusion

► remove Apple A10 Fusion

The Apple A10 Fusion is a system on a chip (SoC) from Apple that is built into the iPhone 7 and iPhone 7 Plus. It integrates four 64 Bit cores that are divided in two clusters. Two high performance cores are clocked at up to 2.34 GHz and should be around 40% faster than the Apple A9 (according to Apple) and two low power cores (up to 1.1 GHz?). Up to now its unclear if all four cores can be used at once (that need only 1/5 th of the energy in some use cases). At the release of the iPhone 7 it looks like that IOS is only using two cores at a time and automatically switches between the two clusters. Therefore, apps are seeing only a dual core. The principle is similar to the first generation of ARMs big.LITTLE concept.

The integrated graphics card of the SoC will most likely stem from PowerVR (again) and perform 50% faster at 2/3 of the power consumption (according to Apple). 

All in all the chip includes 3.3 billion transistors, which is more than a current AMD Bristol Ridge (3.1) and Skylake Quad-Core (1.75)  X86 SoCs.

Sources: Apple Keynote, Ars Technica

Nvidia Tegra X1Apple A10 Fusion
NVIDIA TegraApple Apple A-Series
Erista (Cortex-A57 + A53)Cyclone 4?
2000 MHz2340 MHz
576 KB
2.5 MB
8 / 84 / 2
20
4x Cortex-A57 + 4x Cortex-A53, 2x 32 Bit LPDDR4-1600 (25.6 GB/s, max. 4 GB), Maxwell-GPU (256 Shader, OpenGL ES 3.1, OpenGL 4.5, DirectX 11.2, CUDA 6.0), Dual ISP (1.3 GP/s), H.264/H.265 4K60 Encoding/Decoding, HDMI 2.0, eMMC 5.1ARMv8 Instruction Set
iGPUNVIDIA Tegra X1 Maxwell GPU ( - 1000 MHz)Apple A10 Fusion GPU / PowerVR
ArchitectureARMARM
: Apple A-Series Cyclone 4?
Apple A10X Fusion compare2.39 GHz6 / 3 Cyclone 4?
» Apple A10 Fusion2.34 GHz4 / 2 Cyclone 4?
3300

3DMark - 3DMark Ice Storm Physics
0 Points (0%)
3DMark - 3DMark Ice Storm Extreme Physics
0 Points (0%)
3DMark - 3DMark Ice Storm Unlimited Physics
100%
X1 +
min: 22527     avg: 23316     median: 23071 (24%)     max: 24349 Points
min: 9189     avg: 14249     median: 15479 (16%)     max: 15626 Points
3DMark - 3DMark Sling Shot Extreme (ES 3.1) Unlimited Physics
min: 1252     avg: 1437     median: 1434.5 (18%)     max: 1627 Points
3DMark - 3DMark Sling Shot (ES 3.0) Unlimited Physics
min: 1408     avg: 1525     median: 1518 (26%)     max: 1690 Points
Geekbench 3 - Geekbench 3 64 Bit Multi-Core
100%
1 X1 +
min: 4208     avg: 4318     median: 4216 (7%)     max: 4530
Geekbench 3 - Geekbench 3 64 Bit Single-Core
100%
1 X1 +
min: 1410     avg: 1460     median: 1411 (29%)     max: 1558
Geekbench 4.0 - Geekbench 4.0 64 Bit Single-Core
min: 3476     avg: 3483     median: 3483 (54%)     max: 3490
Geekbench 4.0 - Geekbench 4.0 64 Bit Multi-Core
min: 5585     avg: 5608     median: 5607.5 (14%)     max: 5630
Geekbench 4.1 - 4.4 - Geekbench 4.1 - 4.4 64 Bit Single-Core
100%
X1 +
1510 (21%)
min: 2672     avg: 3233     median: 3499 (48%)     max: 3527
Geekbench 4.1 - 4.4 - Geekbench 4.1 - 4.4 64 Bit Multi-Core
100%
X1 +
4264 (7%)
min: 4508     avg: 5461     median: 5922 (9%)     max: 5952
Geekbench 5.0 - Geekbench 5.0 64 Bit Single-Core
min: 767     avg: 771     median: 771 (46%)     max: 775
Geekbench 5.0 - Geekbench 5.0 64 Bit Multi-Core
min: 1426     avg: 1429     median: 1429 (9%)     max: 1432
Geekbench 5.1 - 5.3 - Geekbench 5.1 - 5.3 64 Bit Single-Core
777 (45%)
Geekbench 5.1 - 5.3 - Geekbench 5.1 - 5.3 64 Bit Multi-Core
min: 1421     avg: 1424     median: 1423.5 (8%)     max: 1426
Linpack Android / IOS - Linpack Single Thread
0 MFLOPS (0%)
Mozilla Kraken 1.1 - Kraken 1.1 Total Score *
100%
X1 +
min: 3844     avg: 4689     median: 4688.6 (8%)     max: 5533.1 ms
min: 1025.5     avg: 1178     median: 1113.4 (2%)     max: 1531 ms
Sunspider - Sunspider 1.0 Total Score *
100%
1 X1 +
min: 597     avg: 715     median: 715 (8%)     max: 833 ms
Octane V2 - Octane V2 Total Score
100%
X1 +
min: 7563     avg: 8032     median: 8031.5 (14%)     max: 8500 Points
min: 19783     avg: 25316     median: 26053 (44%)     max: 27967 Points
Vellamo 3.x - Vellamo 3.x Metal
100%
1 X1 +
2276 Points (61%)
Vellamo 3.x - Vellamo 3.x Multicore Beta
100%
1 X1 +
2535 Points (56%)
Vellamo 3.x - Vellamo 3.x Browser
100%
1 X1 +
6271 Points (83%)
Quadrant Standard Edition 2.0 - Quadrant Standard 2.0 Total Score
100%
1 X1 +
20655 points (48%)
AnTuTu v5 - AnTuTu v5 Total Score
100%
1 X1 +
min: 59811     avg: 63053     median: 63052.5 (64%)     max: 66294 Points
AnTuTu v6 - AnTuTu v6 Total Score
100%
X1 +
min: 90568     avg: 112682     median: 112682 (38%)     max: 134796 Points
min: 142532     avg: 153966     median: 153966 (52%)     max: 165399 Points
AnTuTu v7 - AnTuTu v7 MEM
min: 6141     avg: 9979     median: 10439 (32%)     max: 12895 Points
AnTuTu v7 - AnTuTu v7 UX
min: 32746     avg: 45782     median: 47936.5 (59%)     max: 54510 Points
AnTuTu v7 - AnTuTu v7 GPU
min: 45772     avg: 54850     median: 52519 (16%)     max: 68591 Points
AnTuTu v7 - AnTuTu v7 CPU
min: 50770     avg: 65083     median: 60084.5 (37%)     max: 89394 Points
AnTuTu v7 - AnTuTu v7 Total Score
min: 160578     avg: 175694     median: 165802 (29%)     max: 210596 Points
AndEBench - AndEBench Java
100%
1 X1 +
2699 Iter./s (83%)
AndEBench - AndEBench Native
100%
1 X1 +
15220 Iter./s (50%)
PassMark PerformanceTest Mobile V1 - PerformanceTest Mobile CPU Tests
100%
X1 +
129085 Points (17%)
min: 49294     avg: 49329     median: 49329 (6%)     max: 49364 Points
PCMark for Android - PCM f. Android Work Score
100%
1 X1 +
min: 6858     avg: 7492     median: 7491.5 (37%)     max: 8125 Points

Average Benchmarks Nvidia Tegra X1 → 100% n=7

Average Benchmarks Apple A10 Fusion → 149% n=7

-
-
* Smaller numbers mean a higher performance
1 This benchmark is not used for the average calculation

Add one or more devices and compare

In the following list you can select (and also search for) devices that should be added to the comparison. You can select more than one device.

restrict list:

show all (including archived), 2020, 2019
v1.16
Please share our article, every link counts!
> Notebook / Laptop Reviews and News > Benchmarks / Tech > Processor Comparison - Head 2 Head
Redaktion, 2017-09- 8 (Update: 2017-09-11)