Snapdragon Processor
•Transferir como PPTX, PDF•
10 gostaram•8,537 visualizações
Snapdragon Processors History Snapdragon 820 Intro Snapdragon Processor Architecture
Recomendados
Mais conteúdo relacionado
Mais procurados
Mais procurados (20)
Destaque
Destaque (17)
Semelhante a Snapdragon Processor
Semelhante a Snapdragon Processor (20)
Snapdragon Processor
- 1. 1 14-10-2015 By : Krishna Gehlot
- 2. 2 14-10-2015
- 3. 3 14-10-2015 Snapdragon SoC is a powerful smart processor. The Snapdragon CPU, dubbed from Scorpion First on 28nm process technology[Latest is 14nm] First Fully Integrated 3G/4G Uses ARM instruction set Superior CPU performance [in modem, graphics, power efficiency performance]
- 4. 14-10-2015 4 1. Snapdragon S1 2. Snapdragon S2 3. Snapdragon S3 4. Snapdragon S4 5. Snapdragon 200 series 6. Snapdragon 400 series 7. Snapdragon 600 series 8. Snapdragon 800 series
- 5. 14-10-2015 5 Semiconductor Technology 65 nm/45 nm CPU instruction set ARMv6/ARMv7 CPU Speed From 528 MHz, Up to 1 GHz CPU ARM11,Cortex-A5 and Scorpion CPU Cache Type L2 CPU Cache Max 256 KB GPU Adreno 200 GPU Support 2D support RAM Technology 200 MHz LPDDR1 RAM Speed 200 MHz Bluetooth -
- 6. 14-10-2015 6 Semiconductor Technology 65 nm/45 nm CPU instruction set ARMv6/ARMv7 CPU Speed From 528 MHz, Up to 1 GHz CPU ARM11,Cortex-A5 and Scorpion CPU Cache Type L2 CPU Cache Max 256 KB GPU Adreno 200 GPU Support 2D support RAM Technology 200 MHz LPDDR1 RAM Speed 200 MHz Bluetooth -
- 7. 14-10-2015 7 Semiconductor Technology 45 nm CPU instruction set ARMv7 CPU Speed Up to 1.7 GHz CPU Dual-core Scorpion CPU Cache Type L2 CPU Cache L2 512 KB GPU Adreno 220 GPU Support 2D support RAM Technology Single-channel 500 MHz/333 MHz LPDDR2 RAM Speed 200 MHz Bluetooth -
- 8. 14-10-2015 8 Semiconductor Technology 45 nm/28 nm LP CPU instruction set ARMv7 CPU Speed Up to 1.7 GHz CPU Dual-core Krait/quad-core Krait CPU Cache Type L0, L1, L2 CPU Cache Upto 2MB GPU Quad-core Adreno 320 GPU Support FHD/1080p and FWVGA/720p RAM Technology Dual-channel 500 MHz LPDDR2 RAM Speed 200 MHz Bluetooth Bluetooth 4.0
- 9. 14-10-2015 9 Semiconductor Technology 45 nm LP/28 nm LP CPU instruction set ARMv7 CPU Speed Quad-core Up to 1.4 GHz CPU Quad-core Cortex-A7 CPU Cache Type L0,L1,L2 CPU Cache Up to 2 MB GPU Adreno 304 GPU Support WXGA/720p RAM Technology LPDDR2/LPDDR3 533 MHz Bluetooth Bluetooth 4.1 WiFi - Other
- 10. 14-10-2015 10 Semiconductor Technology 28 nm LP CPU instruction set ARMv8 CPU Speed Up to 1.7 GHz Octa-core CPU Octa-core Cortex-A53 CPU Cache Type L1,L2 CPU Cache Up to 2 MB GPU Adreno 405 GPU Support Up to FHD 1080p RAM Technology LPDDR3 933 MHz Bluetooth Bluetooth 4.1 WiFi Multi-User MIMO (MU-MIMO) WiFi Other GPS
- 11. 14-10-2015 11 Semiconductor Technology 28 nm LP CPU instruction set ARMv8 CPU Speed Up to 1.7 GHz Quad-core CPU Dual-core+Quad-core Cortex- A53 CPU Cache Type L0,L1,L2 CPU Cache Up to 2 MB GPU Adreno 510 GPU Support Quad HD2560x1600 RAM Technology Dual-channel LPDDR3 933 MHz Bluetooth Bluetooth Smart v4.1 WiFi VIVE 1-stream 802.11ac WiFi Other GPS
- 12. 14-10-2015 12 Semiconductor Technology 14 nm FinFET CPU instruction set ARMv7/ARMv8-A CPU Speed 2.0 + 1.55 GHz Quad-core CPU 4+4 cores Hydra and Cortex CPU Cache Type L2 CPU Cache Up to 2 MB GPU Adreno 530 GPU Support 4K UHD video upscale RAM Technology LPDDR4 1866 MHz Bluetooth Native Bluetooth 4.1 support WiFi 802.11ac (2.4 and 5 GHz) WiFi Other Native 4K Support, HDMI 2.0,USB 2.0 and 3.0,64-bit Architecture
- 13. 14-10-2015 13 Technology Snapdragon S1 Snapdragon S2 Snapdragon S3 Snapdragon S4 Semiconductor Technology 65 nm/45 nm 45 nm 45 nm 45 nm/28 nm LP CPU instruction set ARMv6/ARMv7 ARMv7 ARMv7 ARMv7 CPU Speed From 528 MHz, Up to 1 GHz From 800 MHz, Up to 1.5 GHz Up to 1.7 GHz Up to 1.7 GHz CPU ARM11,Cortex- A5 and Scorpion Scorpion Dual-core Scorpion Dual-core Krait/quad-core Krait CPU Cache Type L2 L2 L2 L0,L1,L2 CPU Cache Max 256 KB Min 256 KB Max. 384 KB 512 KB Up to 2 MB GPU Adreno 200 Adreno 205 Adreno 220 Quad-core Adreno 320 GPU Support 2D support 2D support 2D support FHD/1080p and FWVGA/720p RAM Technology 200 MHz LPDDR1 333 MHz Dual- channel LPDDR2 Single-channel 500 MHz/333 M Hz LPDDR2 Dual-channel 500 MHz LPDDR2 Bluetooth - - - Bluetooth 4.0 WiFi - - - -
- 14. 14-10-2015 14 Technology Snapdragon 200 Series Snapdragon 400 Series Snapdragon 600 Series Snapdragon 800 Series Semiconductor Technology 45 nm LP/ 28 nm LP 28 nm LP 28 nm LP 14 nm FinFET CPU instruction set ARMv7 ARMv8 ARMv8 ARMv7/ARMv8-A CPU Speed Quad-core Up to 1.4 GHz Up to 1.7 GHz Octa-core Up to 1.7 GHz Quad-core 2.0 + 1.55 GHz Quad-core CPU Quad-core Cortex-A7 Octa-core Cortex- A53 Dual-core+Quad- core Cortex-A53 4+4 cores Hydra and Cortex CPU Cache Type L0,L1,L2 L1,L2 L0,L1,L2 L2 CPU Cache Up to 2 MB Up to 2 MB Up to 2 MB Up to 2 MB GPU Adreno 304 Adreno 405 Adreno 510 Adreno 530 GPU Support WXGA/720p Up to FHD 1080p Quad HD2560x1600 4K UHD video upscale RAM Technology LPDDR2/LPDD R3 533 MHz LPDDR3 933 MHz Dual-channel LPDDR3 933 MHz LPDDR4 1866 MHz Bluetooth Bluetooth 4.1 Bluetooth 4.1 Bluetooth Smart v4.1 Native Bluetooth 4.1 support WiFi - Multi-User MIMO (MU-MIMO) WiFi VIVE 1-stream 802.11ac WiFi 802.11ac (2.4 and 5 GHz) WiFi Other GPS GPS Native 4K Support, HDMI 2.0,USB 2.0 and 3.0,64-bit Architecture
- 15. 14-10-2015 15 4G LTE With X12 LTE Camera Up to 55 MP Dual ISP & 4K support Display with 4K Ultra HD GPS Qualcomm IZat™ Gen8C CPU Quad-core ARM Cortex A57 at up to 2 GHz and quad-core A53 with 64-bit support GPU Adreno 530 GPU DSP Hexagon™ V56 DSP (up to 800MHz)
- 16. 14-10-2015 16
- 17. 14-10-2015 17 Latest Technology : LTE and Wi-Fi with X12 LTE modem Fastest: Built in 4G+ LTE connectivity Speedy : with Downlink Cat 12 it speeds up to 603.0 Mbit/s and with Uplink Cat 13 it speeds up to 391.6 Mbit/s Advance : 4G+ connectivity, enabled by LTE Advanced Carrier Aggregation MIMO : Qualcomm VIVE 2-stream 802.11n/ac with MU- MIMO Efficient : 4th gen power efficient LTE multimode transceiver Multiband : Tri-Band Wi-Fi(2.4GHz,5GHz & 60GHz)
- 18. 14-10-2015 18 LTE-Unlicensed (LTE-U): increasing mobile network capacity and user throughput by aggregating LTE in licensed and unlicensed bands LTE and Wi-Fi Link Aggregation (LWA) : LWA is another method of aggregating licensed and unlicensed spectrum. It is designed to allow mobile operators to utilize their Wi-Fi infrastructure to augment licensed LTE network capacity with unlicensed spectrum capacity. Next-generation Wi-Fi calling : The X12 LTE modem supports next-generation HD Voice over LTE (VoLTE) and Video over LTE (ViLTE) calling services using the IP Multimedia Subsystem, with support for call continuity between LTE and Wi-Fi. Making use of the.
- 19. 14-10-2015 19 RF front end innovations : a. Advanced Closed Loop Antenna Tuner b. Qualcomm RF360™ front end solution with CA c. Wi-Fi/LTE antenna sharing Antenna sharing : The X12 supports several antenna sharing schemes between LTE and Wi-Fi, designed to make it easier for manufacturers to design devices with advanced technologies like LTE-U, 4x4 LTE MIMO, and 2-stream Wi-Fi, with attractive form factors and minimal performance impact on either technology. Call Continuity across Wi-Fi, LTE, 3G, and 2G
- 20. 14-10-2015 20
- 21. 14-10-2015 21 Kryo CPU : Kryo is tightly integrated with the Adreno 530 GPU and Hexagon 680 DSP and is designed for high- performance mobile computing along with the latest in multimedia and connectivity Custom cores : Kryo is being manufactured on the latest 14nm FinFET technology and is designed to reach speeds up to 2.2 GHz Symphony System Manager : Symphony is designed to manage the entire system-on-chip in different configurations so that the most efficient and effective combination of processors and specialized cores are chosen to get the job done as quickly as possible, with the least amount of power
- 22. 14-10-2015 22 Heterogeneous Computing : With heterogeneous computing,CPU support the efficient use of these diverse engines, individually and together, for dynamic workloads and demanding requirements for new user experiences, all while maximizing battery life and thermal efficiency 64 bits of better processing : integrate 64-bit CPUs which enabled processors with features like 4K display and deliver up to 30 percent faster performance
- 23. 14-10-2015 23 L0 Cache L1 instruction Cache L1 Date Cache CPU Core L0 Cache L1 instruction Cache L1 Date Cache CPU Core Bi-Directional Bus Uncore L3 Cache Shared L2 Cache Memory Controller
- 24. 14-10-2015 24 Biologically Inspired Learning Enable Devices To See and Perceive the World as Humans Do Creation and definition of an Neural Processing Unit
- 25. 14-10-2015 25 Qualcomm Spectra camera ISP : Dual ISP is designed to provide superior camera image quality and experiences. Qualcomm Spectra supports the latest 14-bit sensors for a wider range of colors and more natural skin tones. computational photography : the hybrid autofocus framework and multi-sensor fusion algorithms that support next generation computational photography 4K Support :4K capture and playback with H.264 (AVC),4K Ultra HD on-device display with 4K Ultra HD output to HDTV. Integrated Dolby audio : It supports 7.1 advance surround also makes dialogue crisper, even with heavy explosion or loud effects in the background. Snapdragon Voice Activation : The least amount of power possible to listen and be “woken up” only by a specific word spoken by the voice of the device owner
- 26. 14-10-2015 26 Hexagon 680 DSP : Snapdragon processors use the DSP for offloading tasks that can be done faster and/or consume less power than the CPU . Real-time Sound : Audio content can be delivered with real-time sound processing and streaming for, audio without delay or distortion, and background noise removal Improved recognition : It allows text recognition, object recognition, image enhancement, and facial recognition within images, all while saving battery life. Multimedia Functions : It supports HVX (LightWave 3D Hypervoxel) Extension and utilizes heterogeneous computing to support always-on tasks, like audio and display
- 27. 14-10-2015 27 DSP Extensions for Advance Imaging and Computer Vision mW substantially better CPU
- 28. 14-10-2015 28 Qualcomm IZat Gen8C : Shows location services also show who and what are around you, whether you’re indoors or out. Augmented Positioning : Ubiquitous Wi-Fi and cellular networks ensure accurate positioning. GNSS Chipsets : Using GPS, GLONASS, and BeiDou satellites, IZat location technologies pinpoint your position anywhere in the world, even in challenging urban environments. Cloud-Based Assistance : IZat location technologies use a network of cloud-based assistance servers that provide industry-leading location performance for any mobile device, on any network, in any environment. s
- 29. 14-10-2015 29 Qualcomm Adreno GPU : Adreno 530 does lane detection as well as 2D and 3D map navigation work quickly and smoothly Next-generation visual processing technologie : Qualcomm Adreno GPU and Qualcomm Spectra camera ISP (image signal processor) to bring significant performance, power efficiency, and user experience Fastest : 40% improvement to graphics performance and compute capabilities. Video support Up to 4K HEVC at 60fps over HDMI 2.0 . Support for 64-bit virtual addressing, allowing for shared virtual memory. flexible hybrid autofocus framework and multi-sensor fusion algorithms Efficient : Up to 40 percent lower power consumption
- 30. 14-10-2015 30
- 31. 14-10-2015 31
- 32. 14-10-2015 32 Ease of authentication: helps address the inherent limitations of passwords and pin codes Enhanced security: Qualcomm Technologies’ FIDO implementation is based on Qualcomm SecureMSM technology Ease of implementation: contributes to lower development cost for OEMs Versatility: new types of biometric sensors, like iris scanning, can be used Adaptability: able to grow as mobile biometrics needs grow Reliability: carriers like DOCOMO can confidently base their services on Qualcomm Technologies’ platform
- 33. 14-10-2015 33 USB 3.0/2.0 supported Inbuilt Bluetooth 4.1 Near field communication (NFC) supported Qualcomm Quick Charge 2.0 technology supports Parallel Charging Dual-channel 64-bit LPDDR4 RAM with speeds up to 1688MHz Process Technology 14 nm FinFET Sensors: 3D accelerometers, 3-axis gyro, compass, ambient light, proximity, temperature & pressure, humidity, medical, chemical, ultrasound pen and gestures, finger print reader, etc
- 34. Smaller in size Thermal and power efficient Fastest CPU performance Latest GPU performance Best-in-class energy efficiency for long lasting battery life Rich multimedia and connectivity PROS High cost processors Assembling not possible Heavy weight CONS34 14-10-2015
- 35. 14-10-2015 35 https://en.wikipedia.org/wiki/List_of_Qualcom m_Snapdragon_devices https://www.qualcomm.com/ http://www.notebookcheck.net/ http://www.pcper.com/reviews/Processors/Qu alcomm-Hexagon-680-DSP-Another-Part- Snapdragon-820-Story
- 36. 14-10-2015 36
Notas do Editor
- 10 nano meters =1 × 10-8 meters A system on a chip or system on chip (SoC or SOC) is an integrated circuit (IC) that integrates all components of a computer or other electronic system into a single chip. 28nm technology delivers twice the gate density of the 40nm process and also features an SRAM cell size shrink of 50 percent. ARM, originally Acorn Risc Machine, is a family of reduced instruction set computing (RISC) instruction set architectures for computer processors, configured for various environments, developed by British company ARM Holdings. This approach reduces costs, heat and power use. Such reductions are desirable traits for light, portable, battery-powered devices—?including smartphones, laptops, tablet and notepad computers, and other embedded systems
- 10 nano meters =1 × 10-8 meters A system on a chip or system on chip (SoC or SOC) is an integrated circuit (IC) that integrates all components of a computer or other electronic system into a single chip. 28nm technology delivers twice the gate density of the 40nm process and also features an SRAM cell size shrink of 50 percent. ARM, originally Acorn Risc Machine, is a family of reduced instruction set computing (RISC) instruction set architectures for computer processors, configured for various environments, developed by British company ARM Holdings. This approach reduces costs, heat and power use. Such reductions are desirable traits for light, portable, battery-powered devices—?including smartphones, laptops, tablet and notepad computers, and other embedded systems
- ARMv6 is aimed at the very low end of the 32-bit microcontroller space, enabling very low gate-count designs with very simple and highly efficient micro architecture. Several features of ARMv7 are not available to enable this simplicity. The following is a summary of the major changes: The ARMv6-M instruction set is the smallest supported by any ARM processors, numbering just 57 distinct instructions. With the exception of 6 OS-type instructions, all are 16-bit. Privileged execution is an implementation-option (meaning that it may be included or excluded from the device during the chip design process). This is referred to as the “Unprivileged/Privileged Extension” The SysTick timer is an implementation-option. The maximum number of external interrupts is limited to 32 and only four priority levels are available. Halting debug support is optional. Memory accesses must always be naturally aligned. Exclusive accesses are not supported. The Memory Protection Unit is available as an implementation-option (the PMSA Extension). It is currently supported only by the Cortex-M0+. ARMv7-M supports only the Thumb-2 instruction set. Coupled with a simple two-mode programmer’s model, privilege (which does not have to be used at run-time) and a very simple register banking scheme, this makes for very compact and efficient designs ARMv7-M adding a set of saturating and SIMD instructions ARMv7E-M is currently supported by the Cortex-M4 processor. It builds on ARMv7-M adding a set of saturating and SIMD instructions. This (called the “DSP extension”) significantly increases the capability of the core in DSP applications. These instructions cover operations like signed and unsigned saturated arithmetic, byte and halfword packing/unpacking, dual 16-bit and quad 8-bit operations, and extended halfword multiply-accumulate. There is a further variant of the Cortex-M4 which includes the “Floating-point extension”. Incorporating a further group of instructions for single-precision floating point, these are supported by the addition of an FPU to the standard Cortex-M4. These instructions operate on an extended register bank of 32 single-precision registers and provide single-precision floating point arithmetic, comparison, data transfer between the extension registers, core registers and memory. Of course, the standard Cortex-M4, without the FPU extension, can still handle floating point arithmetic in software but this will take longer and require extra code. The low power (LP) process is the first available 28nm technology. It is ideal for low standby power applications such as cellular baseband. The 28LP process boasts a 20 percent speed improved over the 40LP process at the same leakage/gate cache memory is a high speed memory kept in between processor and RAM to increase the data execution speed. It is kept near to the processor. There are different levels of cache. L1-cache is the fastest cache and it usually comes within the processor chip itself. The L1 cache typically ranges in size from 8KB to 64KB and uses the high-speed SRAM (static RAM) instead of the slower and cheaper DRAM (dynamic RAM) used for main memory. The Intel Celeron processor uses two separate 16KB L1 caches, one for the instructions and one for the data. L2 cache comes between L1 and RAM(processor-L1-L2-RAM) and is bigger than the primary cache (typically 64KB to 4MB). L3 cache is not found nowadays as its function is replaced by L2 cache. L3 caches are found on the motherboard rather than the processor. It is kept between RAM and L2 cache. So if your system has L1,L2 and L3 cache data fetching will be L1->L2->L3->RAM ie. If data is not there in L1 it will check L2 then L3 then RAM..
- AVC-Action Video Camera H.264 or MPEG-4 Part 10, Advanced Video Coding (MPEG-4 AVC) is a video coding format that is currently one of the most commonly used formats for the recording, compression, and distribution of video content Ultra HD - 3840×2160