有關腹瀉

治療腹瀉的藥物有抗菌藥物、止瀉劑、吸附藥及腸黏膜保護劑、微生態制劑等

止瀉劑速治靈 (Imodium) 主要成份為洛哌丁胺 (Loperamide)。作用于腸壁，是通過抑制胃腸平滑肌收縮、抑制腸蠕動，減少排便次數來達到止瀉效果。 主要用于非感染性的一般腹瀉，對于感染性腹瀉的急性期、炎症及中毒症狀較明顯時，應視止瀉劑為禁忌。不能單獨使用于伴有發熱和便血的細菌性痢疾。

吸附藥及腸黏膜保護劑常用于治療消化不良、急性腸炎、潰瘍性結腸炎等引起的腹瀉，主要有藥用活性炭、鞣酸蛋白(Albuminate, find in Apple)、次碳酸鉍等。作用原理是藥物口服到大腸後，通過吸附或收斂作用減少腸內容物對腸壁的刺激，或形成保護膜，減少腸黏膜滲出而發揮止瀉作用，這些藥物宜空腹服用。

微生態制劑有雙歧桿菌(Bifidobacterium)、嗜酸性乳桿菌、酪酸菌、地衣芽孢菌無毒菌株等，可用于急慢性腸炎、消化不良、腸道菌群失調等。其能大量補充腸道內正常菌群的數量，並能促進其增殖，從而糾正菌群失調，抑制腸道內致病菌的生長，並以此達到止瀉目的。

Ref: Imodium

Common viral infections that cause diarrhea include rotavirus (especially in children), norovirus (a.k.a. Norwalk virus), and others.

Bacterial infections that cause diarrhea are typically from contaminated food or water and include salmonella and E. coli. The most common cause of Traveler's Diarrhea is a bacterial infection.

Loperamide HCl is a white to faintly yellow, amorphous or microcrystalline powder. It is classified as an antiperistaltic antidiarrheal agent.

Loperamide is an orally administered, noncentrally acting antidiarrheal agent that has been shown to be effective for relief of acute and chronic diarrhea of diverse etiology. It acts locally in the small and large intestines to decrease motility and, consequently, increase gastrointestinal (GI) transit time by inhibiting peristalsis. ... This occurs primarily via an opioid ef- fect, enhancing circular segmental intestinal muscle contractions,15,42 retarding forward peristaltic motion and increasing intestinal transit time.

Loperamide also reduces daily fecal volume output, inhibits intestinal secretion of fluid and electrolytes, and increases anal sphincter tone.

Ref: Bifidobacterium

Some Bifidobacterium strains are considered as important probiotics and used in the food industry. Different species and/or strains of Bifidobacteria may exert a range of beneficial health effects, including the regulation of intestinal microbial homeostasis, the inhibition of pathogens and harmful bacteria that colonize and/or infect the gut mucosa, the modulation of local and systemic immune responses, the repression of procarcinogenic enzymatic activities within the microbiota, the production of vitamins, and the bioconversion of a number of dietary compounds into bioactive molecules.

Mother's milk contains high concentrations of lactose and lower quantities of phosphate (pH Buffer). Therefore, when mother's milk is fermented by lactic acid bacteria (incl. bifidobacteria) in the infant's GI tract, the pH may be reduced, making it more difficult for Gram-negative bacteria to grow.

双歧杆菌Bifidobacterium是1899年由法国学者Tissier从母乳营养儿的粪便中分离出的一种厌氧的革兰氏阳性杆菌，末端常常分叉，故名双歧杆菌。双歧杆菌在母乳喂养儿肠道内大量存在，对婴幼儿有许多好处，如营养、免疫及抗感染作用。并且还具有抗过敏、抗肿瘤、调整肠道功能及改善营养的作用等。在临床上，双歧杆菌具有调整肠道功能紊乱作用。可以预防腹泻，减少便秘，即双向调节。这种调节能起到预防和治疗各种肠道疾病的效果。

藥食同源 - 具有止泻作用的食物

1. 马齿苋 - 以马齿苋的水煎液对大肠杆菌、痢疾杆菌、伤寒杆菌以及常 见 致病性皮肤真菌有明显的抑制作用，对预防痢疾有良好作用，其治疗效果不亚于黄连素和其 他肠用抗菌素。
2. 白扁豆 - 主治霍乱呕吐、肠鸣泄泻、炎 天暑气、酒毒伤胃。花可治血痢。但由于它含有对人体有 影响的植物血凝素和溶血性皂素，需高温方能破坏之，因此食服时必需烧熟煮透。对生扁豆 加热不够，食后可发生恶心、呕吐、腹痛、头晕等中毒症状。
3. 茶叶 - 药理作用主要由其所含的黄嘌呤衍生物(咖啡因及茶 碱)所产生，另外尚含大量鞣酸，有收敛、抑菌及维生素p样作用。
4. 山楂 - 山楂的体外抑菌试验表明，生山楂有很强的抑制福氏痢疾杆 菌、宋内氏痢疾杆菌、变形杆菌、大肠杆菌及绿脓杆菌的作用。
5. 乌梅 - 对肠道致病菌 以及呼吸道多处致病菌有较强的抑菌作用。
6. 苹果 - 經煮熟的苹果中含有鞣酸和有机酸，两者有收敛作用。
7. 石榴 - 石榴皮止泻效果比石榴果实更佳。
8. 胡椒 - 大辛热，纯阳之物，肠胃寒湿者宜之。
9. 肉桂 - 能补元阳、暖脾胃，对阳气不足、腹冷泄泻者尤宜。
10. 栗子 、榛子

Texture Mapping

Game Engine

A game engine is a framework that used to make up a video game. It consists of different systems:

Input

Rendering

• Resource management
• Scene management - statement management, LOD, game updates, optimizing assets, geometry culling, etc.
• Game update - AI characters update, physics and collision detection, etc.
• Optimizing assets - compressing textures, reducing texture resolution, index geometry, dynamic branching for early-out execution, etc
• Geometry culling - occlusion culling, view frustum culling, geometry partitioning, etc

Sound

• Sound effects, ambient sounds, musics tracks, speech
• AI - situation-based audio

Scripting

• AI - animation paths, situational animations, etc
• Level editor
• Material editor

Physics

• Geometry collisions - point, rigid, soft masses
• Friction / drag / lift / buoyancy support for various objects

Cinematic System

• Real-time cut-scenes - move objects and the camera along predefined paths to tell part of a story
• key-frame animation, skeleton animation, etc.

Networking

• Multiplayer gaming
• Downloading game fixes, additional contents, or entire game, etc.
• Statistics tracking

Core

• Various data structures - geometry partitioning (octrees, quad trees, BSP trees, portals, scene graphs, etc), arrays, link lists, stacks, queues, hash tables, trees (binary, k-dimensionals, etc), graphs, etc.
• Common game math such as vectors, matrices, quaternions, bounding volumes, rays, and planes.
• Timers
• Memory management
• Resource management - keep track of shared resource, e.g. sound, textures, etc
• Journaling services
• File logging
• Application profilers
• Depreciation facilities
• Compression / decompression algorithms
• Encryption / decryption algorithms

CD & Laser Optics

A CD is made from polycarbonate plastic. The program area is 86.05 cm².

• 1.2 millimetres (0.047 in) thick and weighs 15–20 grams.
• Pit size is ~100 nm deep by 500 nm wide, 850 nm to 3.5 µm in length.
• Distance between the tracks, the pitch, is 1.6 µm.
• The length of the recordable spiral is (86.05 cm2 / 1.6 µm) = 5.38 km.
• Scanning speed of 1.2 m/s. Playing time is 74 minutes, or 650 MB of data on a CD-ROM.
• CD uses 780 nm (near infrared) semiconductor laser
• Binary data representation uses non-return-to-zero, inverted method - a change from pit to land or land to pit indicates a one, while no change indicates a series of zeros.
• Data encoding / decoding uses eight-to-fourteen modulation - data to be stored is first broken into 8-bit blocks (bytes). Each 8-bit block is translated into a corresponding 14-bit codeword using a lookup table.
• Error detection and error correction uses cross-interleaved Reed-Solomon code (or CIRC) - adds to every three data bytes one redundant parity byte.

DVD & Blu-ray

• DVD pit size is 400nm. Blu-ray pit size is 150nm.
• DVD track pitch is 740 nm. Blu-ray track pitch is 320 nm.
• DVD uses 650 nm red laser. Blu-ray Disc uses 405 nm "blue" (actually in violet range) laser directly (without frequency doubling or other nonlinear optical mechanisms) from GaN (gallium nitride) laser diodes.

Laser Optics:

The optics may include a collimating lens, diffraction grating (to produce the three beams in a three beam pickup), beamsplitter prism or mirror, turning mirror (for horizontally mounted optics), and focusing (objective) lens.

Typical CD laser optics put out about 0.3 to 1 mW at the objective lens though the diodes themselves may be capable of up to 4 or 5 mW (max). For the Blu-ray player, the output power is 40mW - 100mW. Blu-ray burner uses higher power, its output is over 500mW.

The minimum "spot size" on which a laser can be focused is limited by diffraction, and depends on the wavelength of the light and the numerical aperture of the lens used to focus it. By decreasing the wavelength, increasing the numerical aperture from 0.60 to 0.85, and making the cover layer thinner to avoid unwanted optical effects, the laser beam can be focused to a smaller spot, which effectively allows more information to be stored in the same area.

Blue laser pointers have the same basic construction as DPSS (Diode-pumped solid-state) green lasers. They emit light at 473 nm (sometimes reported as 474 nm), which is produced by frequency doubling of 946 nm laser radiation from a diode-pumped Nd:YAG or Nd:YVO4 crystal. The principal wavelength of Neodymium-doped crystals is 1064 nm. With proper reflective coating mirrors, it can be made to lase at other non-principal neodymium wavelengths, such as the 946 nm. As with green DPSS lasers, use of a 1000 mW IR diode usually results in approximately 300 mW of visible blue light.

For high output power BBO (Beta barium borate (β-BaB₂O₄)) crystals are used as frequency doublers; for lower powers, KTP (Potassium titanyl phosphate (KTiOPO₄)) is used. Output powers available are up to 1000 mW, but this usually is the total output including the infrared.

A few higher-powered (120 mW) 404–405 nm "violet" laser pointers have become available which are not based on GaN, but uses DPSS frequency-doubler technology starting from 1 watt 808 nm gallium arsenide infrared diode lasers being directly doubled, without a longer-wave neodymium laser interposed between diode laser and doubler-crystal. As with all high powered lasers, such devices are able to pop balloons and light matches.

Freezing Temperature of Icewine & Freezing Distillation

The extent of freezing point depression depends only on the alcohol concentration, for small concentration, the value can calculated by the equation: ΔT_F = K_F · m · i , where

ΔT_F = T_{F (pure solvent)} - T_{F (solution)} = T_{F (water)} - T_{F (wine)}

K_F = cryoscopic constant, for water = 1.853 K·kg/mol

m = molality (mol solute per kg of solvent)

i = van 't Hoff factor (number of solute particles per mol), for alcohol i = 1.

Molar mass (g/mol) of alcohol (ethanol), CH₃–CH₂–OH

= 2 x 12.0107 + 5 x 1.00794 + 15.9994 = 46.06844 g/mol

For ice wine of 6% of alcohol, per 100g of wine

the mole of ethanol = 6/46.06844 = 0.13024

the mass of water = 0.094 kg

molality, m = 0.13024/0.094 = 1.38553

T_{F (wine)} = 0 (=-273K) - 1.853 x 1.386 x 1 = -2.568°C

Freeze distillation

A process of enriching a solution by partially freezing it and removing frozen material that is poorer in the dissolved material than is the liquid portion left behind. Such enrichment parallels enrichment by true distillation, where the evaporated and recondensed portion is richer than the liquid portion left behind. Freeze distillation is a kind of partial crystallization of liquid.

The best-known freeze-distilled beverages are applejack and ice beer. Ice wine is the result of a similar process, but in this case, the freezing happens before the fermentation, and thus it is sugar, not alcohol, that gets concentrated.

However, thermodynamics (eutectic system) shown that,

• unless the removal of solid material carries away liquid, the degree of concentration will depend on the final temperature rather than on the number of cycles of removing solid material and chilling
• even if temperatures somewhat below the freezing point of ethyl alcohol are achieved, there will still be alcohol and water mixed as a liquid, and
• at some still lower temperature, the remaining alcohol-and-water solution will freeze without an alcohol-poor solid being separable

Notes:

1. Ethanol also called ethyl alcohol, drinking alcohol, or grain alcohol, is a straight-chain alcohol. Its molecular formula is C₂H₅OH. Ethanol is often abbreviated as EtOH, using the common organic chemistry notation of representing the ethyl group (C₂H₅) with Et. Freezing point is -114°C and boiling point is 78°C.

2. Colligative property - consider solvent freezes to a very nearly pure crystal. This typically occurs simply because the solute molecules do not fit well in the crystal, i.e. substituting a solute for a solvent molecule in the crystal has high enthalpy. For dilute solution, the freezing point depression depends solely on the concentration of solute particles, not on their individual properties. The explanation for the freezing point depression is then simply that as solvent molecules leave the liquid and join the solid they leave behind a smaller volume of liquid in which the solute particles can roam. The resulting reduced entropy of the solute particles thus is independent of their properties.

3. This approximation ceases to hold when the concentration becomes large enough for solute-solute interactions to become important. In that regime, the freezing point depression depends on particular properties of the solute other than its concentration.

4. Through the procedure called cryoscopy, a known constant can be used to calculate an unknown molar mass.

5. van 't Hoff factor is a measure of the effect of a solute upon colligative properties, such as vapor pressure, osmotic pressure and freezing point depression. i = the actual number of particles in solution after dissociation ÷ the number of formula units initially dissolved in solution. Means the number of particles per formula unit of the solute when a solution is dilute. For ideal solutions, i is essentially 1 for most non-electrolytes dissolved in water, e.g., Glucose in water. For most ionic compounds dissolved in water, the van 't Hoff factor is equal to the number of discrete ions in a formula unit of the substance, e.g., i=2 for NaCl.

Wine Serving Temperature

Ref: http://wineintro.com/basics/temperatures.html

Human Body Specifications - Cardiac

Weight - BMI (Body Mass Index)

BMI is a measurement of body fat based on height and weight that applies to both men and women between the ages of 18 and 65 years.

BMI = ( Weight in Pounds / ( Height in inches x Height in inches ) ) x 703

BMI = ( Weight in Kilograms / ( Height in Meters x Height in Meters ) )

BMI related diseases (http://www.bmi-calculator.net/bmi-related-disease.php)

Heart Rate

Maximum Heart Rate (HR) is about 220 minus your age.

The American Heart Association recommends that you do exercise that increases your heart rate to between 50 and 75% of your maximum heart rate (your maximum heart rate is 220 beats per minute, minus your age). They recommend getting at least 30 minutes of exercise on most days of the week (AHA, 2006).

Blood Pressure

Blood Pressure is the force of the blood against the walls of arteries. It includes two measurements:

Systolic pressure (top number): The pressure as the heart beats and forces blood into the arteries

Diastolic pressure (bottom number): The pressure as the heart relaxes between beats

Average value for young people: 120/80 mmHg

Average value for old people: 140/90 mmHg

Healthy blood vessels dilate to allow more blood to flow through more easily. When you exercise, your heart speeds up so the blood can reach your muscles. It may be possible for your heart rate to double safely, while your blood pressure may respond by only increasing a modest amount. When you stop exercising, your pulse does not immediately return to normal; it gradually returns to its resting level. The greater your fitness level, the sooner your pulse rate will return to normal.

(Ref: http://www.medindia.net/patients/calculators/bp_chart.asp)

Health Tools:

http://www.medindia.net/patients/calculators/index.htm

Direct3D - DirectX

Direct3D uses hardware to perform advanced graphics computing, including z-buffering, anti-aliasing, alpha blending, mipmapping, atmospheric effects, and perspective-correct texture mapping, if it is available on the graphics card. It offers full vertex software emulation but no pixel software emulation for features not available in hardware.

The aim of Direct3D is to abstract the communication between a graphics application and the graphics hardware drivers. It provides a low-level interface to every video card 3D function (e.g. transformations, clipping, lighting, materials, textures, depth buffering). It is presented like a thin abstract layer at a level comparable to GDI.

History

Direct3D 8.0 introduced programmability in the form of vertex and pixel shaders. Its programmable shading capabilities were the first major departure from an OpenGL-style fixed-function architecture, where drawing is controlled by a complicated state machine. Direct3D 8 contained many powerful 3D graphics features, such as vertex shaders, pixel shaders, fog, bump mapping and texture mapping.

Direct3D 9.0 added a new version of the High Level Shader Language, support for floating-point texture formats, Multiple Render Targets (MRT), and texture lookups in the vertex shader.

HLSL is analogous to the GLSL shading language used with the OpenGL standard. It is the same as the Nvidia Cg shading language.

Direct3D 10 defines a minimum standard of hardware capabilities which must be supported for a display system to be "Direct3D 10 compatible". The API features an updated shader model 4.0 and optional interruptibility for shader programs. In addition to the previously available shader stages, the API includes a geometry shader stage that breaks the old model of one vertex in/one vertex out.

HLSL programs come in three forms, vertex shaders, geometry shaders, and pixel (or fragment) shaders. A vertex shader is executed for each vertex that is submitted by the application, and is primarily responsible for transforming the vertex from object space to view space, generating texture coordinates, and calculating lighting coefficients such as the vertex's tangent, binormal and normal vectors. Optionally , geometry shader takes as its input the three vertices of a triangle and uses this data to generate (or tessellate) additional triangles, which are each then sent to the rasterizer. Group of vertices (normally 3, to form a triangle) come through the vertex shader, their output position is interpolated to form pixels within its area; this process is known as rasterisation. Each of these pixels comes through the pixel shader, whereby the resultant screen colour is calculated.

Direct3D 10 features:

• Programmable pipelines (often referred to as unified pipeline architecture)
• New state object to enable (mostly) the CPU to change states efficiently
• Shader model 4.0 adds instructions for integer and bitwise calculations
• Geometry shaders, which work on adjacent triangles which form a mesh
• Texture arrays enable swapping of textures in GPU without CPU intervention
• Rapid occlusion culling - prevents objects from being rendered if it is not visible or too far to be visible
• Instancing 2.0 support, allowing multiple instances of similar meshes, such as armies, or grass or trees, to be rendered in a single draw call

Direct3D 10.1 features:

• Finer control over antialiasing (both multisampling and supersampling with per sample shading and application control over sample position)
• More flexibilities to some of the existing features (cubemap arrays and independent blending modes)
• hardware must support the following:
• - Mandatory 32-bit floating point filtering
• - Mandatory support for 4x anti-aliasing
• - Shader model 4.1
• - In 2011, Intel chipsets started supporting Direct3D 10.1 with the introduction of HD graphics 2000 (GMA HD)

Direct3D 11 was released as part of Windows 7, features:

• Tessellation — to increase at runtime the number of visible polygons from a low detail polygonal model
• Multithreaded rendering — to render to the same Direct3D device object from different threads for multi core CPUs
• Compute shaders — which exposes the shader pipeline for non-graphical tasks such as stream processing and physics acceleration, similar in spirit to what OpenCL, Nvidia CUDA, ATI Stream achieves, and HLSL Shader Model 5 among others
• Two new texture compression algorithms for more efficient packing of high quality and HDR/alpha textures
• Increased texture cache

Direct3D 11.1 is an update to the API that will initially ship with Windows 8, features:

• Shader tracing
• Support for video playback
• Shader processing of video resources
• On-the-fly swapping between Direct3D 10 and 11 contexts and feature levels
• Minor updates to the shader language, such as
• - larger constant buffers and optional double-precision instructions,
• - improved blending modes and mandatory support for 16-bit color formats to improve the performance of entry-level GPUs such as Intel HD Graphics

Cost Effective Notebook PC for Gaming

3DMark06 - DirectX 9.x

3DMark06 includes advanced Shader Model 2.0 and 3.0 graphics tests as well as single CPU, multiple core and multiple processor tests as part of the 3DMark score.

1. SM2.0 Graphics Test 1 : Lighting (total 26 lights)

• Two directional with one casting CSM (Cascading Shadow Maps),
• 12 point lights provide fill, most cast shadows using 1024x1924x6 cube depth maps and hardware shadows, some are masked and animated
• 12 small spot lights with non-overlapping shadow-maps

2. SM2.0 Graphics Test 2: CSM & shadow-mapped lighting

• One directional light casting CSM to create a moonlight effect
• Two illuminated fireflies are shadow-mapped point lights with cube-map masks
• A procedural light scattering shader
• Materials have Blinn-Phong reflections using lookups for materials
• Ground material has diffuse, diffuse detail, normal and normal details maps added
• Rock surfaces have an added specular map
• Branches are similar to the rock materials but with an added diffuse cube map and no specular or bump map

3. SM3.0 Graphics Test 3: HDR (High Dynamic Range) rendering and dynamic shadows

• Sky and atmospheric effects rely on a complex atmospheric light scattering algorithm with cloud blending
• Heterogeneous fog effect creates deep humidity
• Water effects are recreated using HDR reflections and refractions
• Water surface is distorted with two scrolling maps and Gerstner wave functions
• Subsurface scattering and a Blinn-Phong shading model with two normal maps and one color map are used for the sea-monster
• Strauss shading is used for the captain, crew and airship
• Canyon wall material uses three color maps, three normal maps and Lambertian diffuse shading
• One directional light, casting CSM is used to create the bright sunlight, producing dramatic dynamic shadows

4. SM3.0 Graphics Test 4: CSM with dynamic long, soft shadows

• As the sun sets, shadows dynamically increase in length for a very realistic effect
• Snow storm - heterogeneous fog and particles with an added diffuse cube map
• Blinn-Phong shader with two normal maps, one color map and subsurface scattering is used for the snow material
• Metallic and other surfaces use the Strauss shading model
• One directional light casting CSM recreates the sunlight.
• Blending a pair of cube-maps, (one diffuse, the other specular) an ambient effect is created

5. CPU Test 1: tight AI synchronization intervals

• AI, physics and game logic to generate a multi-threaded workload. Uses a high level of path finding complexity, tight AI synchronization intervals over 40 frames, locked to fixed frame rate of 2FPS with a Shader Profile of 2_0 and a resolution of 640x480

6. CPU Test 2: lax AI synchronization intervals

• Uses a lower level of path finding complexity, lax AI synchronization intervals over 60 frames, locked to a fixed frame rate of 2FPS with a Shader Profile of 2_0 and a resolution of 640x480

7. Feature Tests

• Fill Rate (single-texturing)
• Fill Rate (multi-texturing)
• Pixel Shader
• Vertex Shader (simple)
• Vertex Shader (complex)
• Shader Particles
• Perlin Noise

Pixel Fill Rate

Pixel fill rate is the total number of pixels the card can output and is calculated as the number of raster operations (ROPs) multiplied by the clock frequency.

Texture Fill Rate

Texture fill rate is calculated differently by ATI and Nvidia. Both are correct methods as Nvidia has one texture unit per pixel shader unit or one per pixel pipe.

• Nvidia multiplies the number of pixel pipelines by the clock frequency.
• ATI multiplies the number of texture units by the clock frequency.

The most cost effective Notebook PC for gaming is : Intel i5 with NVIDIA GT540M

Baseline : MacBook Pro 13" mid 2009; w1 = 0.35, w2 = 0.65

MacBook Pro, 13-inch Mid 2009

MacBook Pro, 13-inch Mid 2009 (MacBookPro 5,5) OSX 10.7.2

Intel Core 2 Duo P7550 @2.266GHz

- 2 cores, 2 threads

- L1 cache, 32KB + 32KB

- L2 cache, 3MB

- SIMD, 1

- Fab, 45nm

- FSB, 266MHz

- Bus/Core rate, 8.5

- HT/QPI/DMI, 1066MHz

- Instruction set, 64-bit

- Max TDP, 25W

- Floating Point:

Dot Product single-threaded scalar -- 3088, , 1.49 Gflops

Dot Product multi-threaded scalar -- 6435, , 2.93 Gflops

Dot Product single-threaded vector -- 2470, , 2.96 Gflops

Dot Product multi-threaded vector -- 5628, , 5.85 Gflops

NVIDIA GeForce 9400M, 256MB

OpenGL 2.1 NVIDIA-7.12.9,

OpenCL 1.1 (July 25 2011) Full_Profile

Shading Language v1.2

Max texture size: 8192 x 8192


Max texture coordinates: 8


Max vertex texture image units: 16


Max texture image units: 16


Max geometry texture units: 16


Max anisotropic filtering value: 16


Max number of light sources: 8


Max viewport size: 8192 x 8192


Max uniform vertex components: 4096


Max uniform fragment components: 4096

Max geometry uniform components: 2048


Max varying floats: 60


Max samples: 8


Max draw buffers: 8

Mobile Processors - Benchmarks

(Ref: NoteBookCheck)

GPU - Specifications

1. xx:xx:xx:xx - Vertex shader : Pixel shader : Texture mapping unit : Render Output unit

2. xx:xx:xx - Unified Shaders (Vertex shader/Geometry shader/Pixel shader) : Texture mapping unit : Render Output unit

NVIDIA

AMD

Xbox

Xbox

Bill Gates unveiled the Xbox at the Game Developers Conference in 2000. At that time, Sega's Dreamcast sales were diminishing and Sony's PlayStation 2 was just going on sale in Japan.

The Xbox was the first gaming product to feature Dolby Interactive Content-Encoding Technology, which allows real-time Dolby Digital encoding in game consoles. Previous game consoles could only use Dolby Digital 5.1 during non-interactive "cut scene" playback.

The system software may have been based on the Windows NT architecture that powered Windows 2000. It exposes APIs similar to APIs found in Microsoft Windows, such as DirectX 8.1.

A derivative of the NVIDIA GeForce 3, known as the NV2A (233MHz), co-developed by Microsoft and Nvidia was used in the Microsoft Xbox game console. GeForce 3 was the first Microsoft Direct3D 8.0 compliant 3D-card. Its programmable shader architecture enabled applications to execute custom visual effects programs in Microsoft Shader language 1.1. With respect to pure pixel and texel throughput, the GeForce 3 has four pixel pipelines which each can sample two textures per clock. Features:

- Lightspeed Memory Architect (LMT) - reduce overdraw, conserve memory bandwidth by compressing the z-buffer (depth buffer) and better manage the memory bus.

- support multi-sampling and Quincunx anti-aliasing

- texture sampling units were upgraded to support 8-tap anisotropic filtering, distant textures can be noticeably sharper

NV2A: (Nvidia ceased production of the Xbox's GPU in August 2005)

- 4 pixel pipelines with 2 texture units each

- 932 megapixels/s (233 MHz × 4 pipelines)

- 1,864 megatexels/s (932 MP × 2 texture units)

- Peak triangle performance (32pixel divided from filrate)

= 932M/32 = 29,125,000 32-pixel triangles/s raw or w. 2 textures and lit.

= 970,833 triangles per frame at 30 frame/s

- Bilinear, trilinear, and anisotropic texture filtering

- 8 textures per pass, texture compression, full scene anti-aliasing (NV Quincunx, supersampling, multisampling)

- 64 MB DDR SDRAM at 200 MHz; in dual-channel 128-bit configuration giving 6400 MB/s

CPU: 32-bit 733 MHz, custom Intel Pentium III Coppermine-based processor.

- SSE floating point SIMD. Four single-precision floating point numbers per clock cycle.

- MMX integer SIMD

- 32 KB L1 cache. 128 KB on-die L2 cache

- uses uses out-of-order execution

- 133 MHz 64-bit GTL+ front-side bus to GPU

Xbox 360

The Xbox's successor, the Xbox 360, was officially unveiled announced on May 12, 2005. Xbox 360 uses the triple-core IBM designed Xenon (the cores of the processor were developed using a slightly modified version of the PlayStation 3's Cell Processor PPE architecture) as its CPU, with each core capable of simultaneously processing two threads, and can therefore operate on up to six threads at once. Graphics processing is handled by the ATI Xenos, which has 10 MB ofeDRAM. Its main memory pool is 512 MB in size.

CPU: 3.2GHz Custom IBM Central Processor, named Xenon at Microsoft and "Waternoose" at IBM.

- triple-core, two threads per core

- VMX unit per core

- 128 VMX Registers per thread

- 1MB L2 Cache (Lockable by Graphics Processor)

- uses in-order execution

- 9.6 billion dot products per second

- Unified Memory Architecture (UMA), 700MHz GDDR3 512MB RAM, 128 bit interface

- 22.40 GB/s (=700MHz*2*128/8) of memory interface bus (FSB)

Xenos: Although R500 has long since been mentioned in relation to the graphics behind Xenon, the actual development name ATI uses for Xenon's graphics is "C1". The PC graphics processor R520 and Xenos are very different.

The Xenos graphics processor consists of two distinct elements: the graphics core (shader core) and the eDRAM module. Both units running at 500MHz within the single die.

- Unified Shader Core

- 48 ALU’s for Vertex or Pixel Shader processing

- 16 Filtered & 16 Unfiltered Texture samples per clock

- 10MB eDRAM frame buffer, 32 GB/s between the eDRAM die and the GPU, 500 MB/s south-bridge

- eDRAM internal logic to its internal memory bandwidth is 256 GB/s ==> 4× FSAA, z-buffering, and alpha blending with no appreciable performance penalty on the GPU

- 48 ALU operations (one vector4 and scalar op per clock), 16 texture fetches, 32 control flow operations, and 16 programmable vertex fetch operations with tessellation per clock = 48*2 + 16 + 32 + 16 = 160 operations per cycle or 160 * 500 = 80 GOps per second

The Xbox 360 launched with 14 games in North America and 13 in Europe. The console's best-selling game for 2005, Call of Duty 2, sold over a million copies. Five other games sold over a million copies in the console's first year on the market: Ghost Recon Advanced Warfighter, The Elder Scrolls IV: Oblivion, Dead or Alive 4, Saints Row, and Gears of War.

Xbox 360 S was officially announced on June 14 2010, marketed simply as the Xbox 360, feature redesigned internal architecture with the Valhalla motherboard, which allows for around 30% more space than previous motherboards,and the XCGPU, an integrated CPU/GPU/eDRAM chip using a 45 nm fabrication process. Microsoft have said that they believe that the console is only mid-way through its life-cycle and will continue through 2015.

Xbox 360 system software (or Dashboard), is the updatable software and operating system for the Xbox 360. It resides on a 16 MB file system and has access to a maximum of 32 MB of the system's memory. Version 2.0 (11/11) supports:

- Stereoscopic 3D

- Dual play

- Updated Standard and Kinect Dashboard with Full Voice Dashboard Commandment Support

DirectX

Windows 98 and Windows NT 4.0 both shipped with DirectX. DirectX is a collection of APIs for handling video and gaming programming. Its functionality is provided in the form of COM-style objects and interfaces. The components of DirectX are:

- DirectDraw, Direct3D, Direct2D, DXGI, DirectWrite, DirectCompute,

- DirectInput, DirectPlay, DirectX Diagnostics (DxDiag), DirectSetup

- DirectSound, DirectSound3D, DirectMusic (replaced with XAudio2 and XACT3)

- Direct Media Objects, DirectMedia, DirectShow.

DirectX versions

5.2 (8/97) - Available as a beta for Windows 2000 that would install on NT 4.0

7.0 (2/00) - Windows 2000

8.1 (11/01) - Windows XP, Windows Server 2003 and Xbox

8.2 (02) - same as 8.1b (fix to DirectShow on Win2K) but includes DirectPlay 8.2

9.0 (12/02) - 9.0c Windows XP SP2 (8/04), SP3 (4/08)

10 (11/06) - Windows Vista exclusive

10.1 (4/09) - SP2 for Windows Vista, Windows Server 2008, includes Direct3D 10.1

11 (2/11) - SP1 for Windows 7, Windows Server 2008 R2

11.1 (9/11) - Windows 8 Developer Preview

DirectX 8 & 9

DirectX was also used as a basis for Microsoft's Xbox and Xbox 360 console API. The API was developed jointly between Microsoft and Nvidia. The Xbox API is similar to DirectX version 8.1. In 2002, Microsoft released DirectX 9 with support for the use of much longer shader programs than before with pixel and vertex shader version 2.0. In 2004, DirectX 9.0c was released with shader model 3.0.

DirectX 10

Windows Vista includes a new version of Direct3D, called Direct3D 10. It adds scheduling and memory virtualization capabilities to the graphics subsystem. The Direct3D 10 incorporates Microsoft's High Level Shader Language 4.0. The API introduces unified vertex and pixel shaders. In addition, it also supports Geometry Shaders, which operate on entire geometric primitives (points, lines, and triangles), and can allow calculations based on adjacent primitives as well. The output of the geometry shader can be passed directly onwards to the rasterizer for interpolation and pixel shading, or written to a vertex buffer (known as 'stream out') to be fed back into the beginning of the pipeline.

Direct3D 10 is not backward compatible like prior versions of DirectX. Its functionality requires WDDM (Windows Display Driver Model) and new graphics hardware. The graphics hardware will be pre-emptively multithreaded, to allow multiple threads to use the GPU in turns. It will also provide paging of the graphics memory.

Many former parts of DirectX API were deprecated in the latest DirectX SDK and will be preserved for compatibility only:

- DirectInput was deprecated in favor of XInput,

- DirectSound was deprecated in favor of the Cross-platform Audio Creation Tool system (XACT) and lost support for hardware accelerated audio, since Vista audio stack renders sound in software on the CPU.

- DirectPlay DPLAY.DLL was also removed and was replaced with dplayx.dll; games that rely on this DLL must duplicate it and rename it to dplay.dll.

Direct3D 10.1 is an incremental update of Direct3D 10.0, requires video card supports Shader Model 4.1 or higher and 32-bit floating-point operations - adds support for cube map arrays, separate blend modes per-MRT, coverage mask export from a pixel shader, ability to run pixel shader per sample, access to multi-sampled depth buffers.

DirectX 11

Direct3D 11 is a strict superset of Direct3D 10.1 — all hardware and API features of version 10.1 are retained, and new features are added only when necessary for exposing new functionality. New features: GPGPU support (DirectCompute), and Direct3D11 with tessellation support and improved multi-threading support to assist video game developers in developing games that better utilize multi-core processors. Hardware tessellation and Shader Model 5.0 require Direct3D 11 supporting hardware.

Direct3D 11 runtime introduces Direct3D 9, 10, and 10.1 "feature levels", compatibility modes which allow use of only the hardware features defined in the specified version of Direct3D. For Direct3D 9 hardware, there are three different feature levels, grouped by common capabilities of "low", "med" and "high-end" video cards; the runtime directly uses Direct3D 9 DDI provided in all WDDM drivers.

DirectX 11.1 is included in Windows 8.

It supports WDDM 1.2 for increased performance, features:

- tighter integration of Direct2D, Direct3D, and DirectCompute

- DirectXMath, XAudio2, and XInput libraries from the XNA framework

- features stereoscopic 3D support for gaming and video

Steve Jobs 1955-2011

Steve Jobs' 2005 Stanford Commencement Speech

"No one wants to die. Even people who want to go to heaven don’t want to die to get there. And yet death is the destination we all share. No one has ever escaped it. And that is as it should be, because Death is very likely the single best invention of Life. It is Life’s change agent. It clears out the old to make way for the new. Right now the new is you, but someday not too long from now, you will gradually become the old and be cleared away. Sorry to be so dramatic, but it is quite true.

Your time is limited, so don’t waste it living someone else’s life. Don’t be trapped by dogma — which is living with the results of other people’s thinking. Don’t let the noise of others’ opinions drown out your own inner voice. And most important, have the courage to follow your heart and intuition. They somehow already know what you truly want to become. Everything else is secondary."

股票必勝法

股神葉天臨死前授予方展博的股票必勝法：

「貪心輸錢贏，輸血不輸錢；

人人去輸血，殺人不打風；

落雨去長洲，長洲飲煉奶；

見人分遺產，出獄嫌錢腥；

快樂冇人格，出家冇人工；

攻無不克，戰無不勝矣！」

取勝唯一法則 ──「及早離去」

中秋大坑火龍 2011

火龍全長二百二十英尺 , 除龍頭龍尾外 , 龍身分三十二節 , 全用珍珠草包紮而成 , 全身滿插長壽香 , 故又名 「火龍」。

火龍晚上6時左右在大坑蓮花宮點睛開光，再於安庶庇街插香後起龍，先到浣紗街對嘉賓作致敬禮，多數會打龍餅（喜結龍團），之後會按傳統遊街（火龍會在大坑的街道上舞動），途徑浣紗街、京街、新村街、銅鑼灣道等。

遊街完畢會到浣紗街拔香和重新插香，後會在浣紗街表演至10點左右。整個儀式可分為「火龍過橋」、「火龍纏雙柱」、「綵燈火龍結團圓」三部份。

節日完畢後，習慣把火龍拋下銅鑼灣避風塘的海底，以示「龍歸滄海」，但近年為免污染海水，於深夜用貨車送到焚化爐去，變作「飛龍在天」。

引路開場

見龍在田

飛龍在天

亢龍有悔

曲終人散

EMI/RFI Shieding

Metal Casing / Coating / Plating

M.G. Chemicals

A general purpose EMI / RFI shielding spray for use on plastic electronics enclosures. Consists of a tough, durable acrylic base pigmented with a high purity nickel, silver coated copper or silver flakes.

Common-Mode Choke

Narrowband interference arises from intentional transmissions - radio and TV stations, pager transmitters, cell phones, etc.

Broadband interference comes from incidental radio frequency emitters - electric power transmission lines, electric motors, thermostats, etc.

High-speed switching devices - TV, computer, and other digital equipment, etc.

Putting common-mode chokes on your feedline, power, and other cables will substantially reduce your received noise level. Simple RF choke works as a balun by preventing signals passing along the outside of the braid.

Balun can be considered as simple forms of transmission line transformers. A transmission line transformers use small ferrite cores in toroidal or "binocular" shapes. Something as simple as 10 turns of coaxial cable coiled up on a diameter about the size of a dinner plate makes an extremely effective choke balun for frequencies from about 10 MHz to beyond 30 MHz.

Toroidal-core

Binocular-core

Ferrite bead is used as a passive low-pass filter. It filters out the high frequency noise in the circuit by dissipating it as heat. The material used to construct the bead becomes highly resistive at the design frequency range and the induced current inside the bead is dissipated as heat instead of inducing an opposing current back in the signal cable (magnetic field within the bead is unable to establish properly at that specific frequency range).

Clamp-on cores can be attached without wrapping the wire at all. Although the wire is not coiled around the core, the introduction of the ferrite core around the wire increases the self-inductance of the wire thus still has the effect of absorbing energy from the noise traveling in the wire.

PCB Design

From "Key Design Techniques Reduce EMI/RFI Effects in Military Applications"

Extra (for fuses)

See the Discussion on Home Power Line Noise at below.

Discussion on Home Power Line Noise

Discussion & Conclusion

There was a thread in Hi-Fi forum about residential power line noise. The writer concluded from his study that low noise preamp benefit most from the audio grade fuses.

The following items were tested to see the effect on AC power by using Tektronix TDS2012 digital oscilloscope:
1. PS Audio - Soloist In-Wall Power Conditioner
2. PS Audio - xStream Premier SC power cord
3. PS Audio - Power Plant Premier
4. MDA ceramic fuse
5. Isoclean fuse
6. Littlefuse 313 fuse
7. HiFi Tuning fuse

The writer found that, both FFT measurement of the output AC power and audio listening to the music playing, fuses sounded better if they were mounted in the direction of energy flow (current).

In this post, I used some of his posted FFT diagrams to render two animated sequences then from which I come up with the followings:
1.) The audio improvement, if it was matched with the greatest changes in FFT, is mainly from the reduction of noise in the frequency range 10Hz - 200Hz. The most significant decrease in noise level is in the range 10Hz - 60Hz.
2.) Audio grade fuse do a good job in the frequency range 10Hz - 60Hz.
3.) Audio grade fuse mounted in the energy direction reduce the noise the most, followed by ceramic fuse and audio grade fuse in reverse direction.
4.) High quality power cord reduce some stray frequencies in FFT
5.) Ceramic fuse while introduce some low frequency harmonics to the power output also reduce some of the harmonics of 40Hz.

6.) I am not sure what physics explain the difference in noise by changing mounting direction.

Filtering

1. PS Audio - Soloist In-Wall Power Conditioner
2. PS Audio - xStream Premier SC power cord
3. MDA ceramic fuse
4. Isoclean fuse in reverse direction
5. Isoclean fuse in forward direction

Re-generation

1. PS Audio - Power Plant Premier
2. Littlefuse 313 fuse
3. HiFi Tuning fuse in reverse direction
4. HiFi Tuning fuse in forward direction

Item description
PS Audio - Soloist In-Wall Power Conditioner
The Soloist utilizes a balanced power coupler known as a balun for filtering. A balun is a donut of powdered iron with a wire wrapped around each half of the donut. When an AC signal comes into the balun’s two wrappings of wire, the balanced power coupler design cancels out anything in common to the wires. Unlike series filters that only work on noise over a specified frequency range, a well-designed balun cancels a great deal of noise without a lot of wire in the way, thus doing so without any type of dynamic restrictions.

The Soloist’s power coupler reduces common and differential mode noise by up to 40 dB, meaning that whatever noise is on the line will be reduced by over 100 times.


xStream Premier SC power cord
Concocted from a blend of the world’s finest conductors: - PCOCC single crystal copper and solid 99.9% pure silver solid core - it’s surrounded by multiple shields and encased in a luxurious cloth outer jacket. The wire gauge of Premier is 7. 25% of the it is solid silver and the balance is PCOCC single crystal copper.



Power Plant Premier
The Premier Power Plant is the smaller of the three Power Plants in the AC regenerating line. The Premier produces up to 800 continuous watts of regulated power at an efficiency of 85% with little to distortion or noise. The Premier has ten outlets, five IsoZones™, a built in power sequencer, THD analyzer, voltage meter, MultiWave, CleanWave and a remote control to access all the features.


Hi-Fi Tuning Fuse

From audible frequency to electrical noises

The humans audible range of frequencies is in between 20Hz - 20kHz. There is considerable variation between individuals, especially at the high frequency end. Sensitivity also varies a lot with frequency, as shown by equal-loudness contours. By definition two sine waves, of differing frequencies, are said to have equal-loudness level measured in phons if they appear equally loud to the average young person without significant hearing impairment.

EMI

Large electric motors create powerful electromagnetic fields and cause Electromagnetic interference (EMI) problems. Electric motors, which may cause EMI, are found in various kinds of equipment-refrigerators, air conditioners, washing machines, furnaces, copier, elevators, and machine tools, etc.

RFI

Radio-frequency interference (RFI) is in general caused by radio or television sources. A weaker source, such as in-house wireless telephone, fluorescent lamp and lamp dimmer can also interfere with the electronic device if they are placed too close.

Harmonic Distortion

It is a form of pollution in the electric plant that can cause problems if the sum of the harmonic currents increases above certain limits. All power electronic converters used in different types of electronic systems can increase harmonic disturbances by injecting harmonic currents directly into the grid. Common non-linear loads include motor starters, variable speed drives, computers and other electronic devices, electronic lighting, welding supplies and uninterrupted power supplies.

The effects of harmonics can be overheating of transformers, cables, motors, generators and capacitors connected to the same power supply with the devices generating the harmonics. Electronic displays and lighting may flicker, circuit breakers can trip, computers may fail and metering can give false readings.

Figure shows how the current harmonics (ih) in the input current (is) of a power electronic converter affect the supply voltage (ut).

In a theoretical case where output current can be estimated as clean DC current, the harmonic current frequencies of a 6-pulse three phase rectifier are n times the fundamental frequency (50 or 60 Hz), where n = 6k ± 1 and k=1,2,... .

(Ref: Guide to Harmonics with AC Variable Frequency Drives by ABB)

Total harmonic distortion (THD) indicates the strength of electromagnetic noise generated. THD in the current is the root mean square (rms) of all the harmonic currents as a percentage of the fundamental current.

Radio Frequencies

Radio frequency

3 - 30Hz Extremely low frequency ELF

30 - 300Hz Super low frequency SLF

300 - 3000Hz Ultra low frequency ULF

3 - 30kHz Very low frequency VLF

30 - 300kHz Low frequency LF

300kHz - 3MHz Medium frequency MF

3 - 30MHz High frequency HF

30 - 300MHz Very high frequency VHF

300MHz - 3GHz Ultra high frequency UHF

3 - 30GHz Super high frequency SHF

30 - 300GHz Extremely high frequency EHF

ELF

The United States Navy utilized extremely low frequencies (ELFs) as radio band and radio communications. Because of the electrical conductivity of seawater, submarines are shielded from most electromagnetic communications. The Submarine Integrated Antenna System (SIAS) was a research and development effort to communicate with submerged submarines.

SLF

This frequency range includes the frequencies of AC power grids (50 hertz and 60 hertz). The radio services Seafarer (USA) on 76 hertz and ZEVS (Russia) on 82 hertz operate in this range. They both provide communication services for submarines at a certain depth.

ULF

Many types of waves in the ULF frequency band can be observed in the magnetosphere and on the ground. These waves represent important physical processes in the near-Earth plasma environment. This band is used for communications in mines, as it can penetrate the earth. The speed of the ULF waves is often associated with the Alfven velocity that depends on the ambient magnetic field and plasma mass density.

VLF

VLF waves can penetrate water to a depth of roughly 10 to 40 metres (30 to 130 feet), depending on the frequency employed and the salinity of the water. VLF is used to communicate with submarines near the surface (for example using the transmitter DHO38), while ELF is used for deeply-submerged vessels. VLF is also used for radio navigation beacons (alpha) and time signals (beta), and electromagnetic geophysical surveys.

LF

In the western hemisphere, its main use is for aircraft beacon, navigation (LORAN), information, and weather systems. In Europe, and parts of Northern Africa and of Asia, part of the LF spectrum is used for AM broadcasting as the longwave band. Some radio frequency identification (RFID) tags utilize LF. These tags are commonly known as LFID's or LowFID's (Low Frequency Identification).

Electronic ballasts for fluorescent lamp employ transistors to alter mains voltage frequency into high-frequency AC, typically from 25–50kHz, while also regulating the current flow in the lamp. EMI at the electronic ballast fundamental frequency and its harmonics propagate from the ballast’s electronic circuits to the line conductors. This EMI may interfere with other electrical equipment on the same distribution network. Most electromagnetic ballasts have THD between 18% and 35%. Most of them are below 20%.

Switching-mode power supply (SMPS) is an electronic power supply that incorporates a switching regulator. Unlike a linear power supply, the pass transistor of a switching mode supply switches very quickly (typically between 50 kHz and 1 MHz) between full-on and full-off states, which minimizes wasted energy. Voltage regulation is provided by varying the ratio of on to off time. In contrast, a linear power supply must dissipate the excess voltage to regulate the output. Output current flow depends on the input power signal, the storage elements and circuit topologies used, and also on the pattern used to drive the switching elements. Disadvantages of SMPS, include greater complexity, the generation of high-amplitude, high-frequency energy that the low-pass filter must block to avoid electromagnetic interference (EMI) - (properly earthed) EMI/RFI filter is connected between the input terminals and the bridge rectifier, and a ripple voltage at the switching frequency and the harmonic frequencies thereof.

MF

Radio stations are allocated an AM broadcast band from 526.5 kHz to 1606.5 kHz in Europe; in North America this extends from 535 kHz to 1705 kHz.

There is an amateur radio band known as 160 meters or 'top-band' between 1800 and 2000 kHz (allocation depends on country and starts at 1810 kHz outside the Americas). Amateur operators transmit CW morse code, digital signals and SSB voice signals on this band. 2182kHz is the international calling and distress frequency for SSB maritime voice communication (radiotelephony). It is analogous to Channel 16 on the marine VHF band. There are a number of coast guard and other ship-to-shore frequencies in use between 1600 and 2850 kHz.

HF

Also known as the decameter band or decameter wave. Since the ionosphere often refracts HF radio waves quite well (a phenomenon known as skywave propagation), this range is extensively used for medium and long range radio communication. Some radio frequency identification (RFID) tags utilize HF. These tags are commonly known as HFID's or HighFID's (High Frequency Identification).

Noise, especially man-made interference from electronic devices, tends to have a great effect on the HF bands. For example, frequencies on which broadband over power lines (BPL) operates (typically corresponding with the HF band) and the tendency for the BPL "signal" to leak from power lines. Some BPL providers have installed "notch filters" to block out certain portions of the spectrum (namely the amateur radio bands), but a great amount of controversy over the deployment of this access method remains.

VHF

Common uses for VHF are FM radio broadcast, television broadcast, land mobile stations (emergency, business, and military), long range data communication with radio modems, Amateur Radio, marine communications, air traffic control communications and air navigation systems (e.g. VOR, DME & ILS). In most of the world, FM radio stations broadcast in the band goes from 87.5 to 108.0 MHz (band II).

British television originally used VHF band I and band III. Television on VHF was in black and white with 405-line format. British colour television was broadcast on UHF (channels 21–69), beginning in the late 1960s. From then on, TV was broadcast on both VHF and UHF. The last British VHF TV transmitters closed down on January 3, 1985. VHF band III is now used in the UK for digital audio broadcasting, and VHF band II is used for FM radio, as it is in most of the world.

(Band I ranges from 47 to 88 MHz, and it is primarily used for radio and television broadcasting. Channel spacings vary from country to country, with spacings of 6, 7 and 8 MHz being common. Band III ranges from 174 to 230 MHz.)

UHF

It is widely used in two-way radio systems and cordless telephones, whose transmission and reception antennas are closely spaced. UHF signals travel over line-of-sight distances. A repeater propagates UHF signals when a distance greater than the line of sight is required. Civilian applications, such as GMRS, PMR446, UHF CB, 802.11b ("WiFi") and the widely adapted GSM and UMTS cellular networks, also use UHF frequencies.

Universal Mobile Telecommunications System (UMTS) is a third generation mobile cellular technology for networks based on the GSM standard. UMTS employs wideband code division multiple access (W-CDMA) radio access technology to offer greater spectral efficiency and bandwidth to mobile network operators. Developed by the 3GPP (3rd Generation Partnership Project), UMTS is a component of the International Telecommunications Union IMT-2000 standard set and compares with the cdma2000 standard set for networks based on the competing cdmaOne technology. The specific frequency bands originally defined by the UMTS standard are 1885–2025 MHz for the mobile-to-base (uplink) and 2110–2200 MHz for the base-to-mobile (downlink).

A microwave oven works by passing non-ionizing microwave radiation through the food, usually at a frequency of 2.45 gigahertz (GHz)—a wavelength of 122 millimetres (4.80 in).

Since 2006, UMTS networks in many countries have been or are in the process of being upgraded with High Speed Downlink Packet Access (HSDPA), sometimes known as 3.5G. Currently, HSDPA enables downlink transfer speeds of up to 21 Mbit/s. Work is also progressing on improving the uplink transfer speed with the High-Speed Uplink Packet Access (HSUPA). Longer term, the 3GPP Long Term Evolution project plans to move UMTS to 4G speeds of 100 Mbit/s down and 50 Mbit/s up, using a next generation air interface technology based upon Orthogonal frequency-division multiplexing.

SHF

Also known as the centimeter band or centimeter wave as the wavelengths range from ten to one centimeters. This frequency is used for microwave devices, WLAN, most modern radars. The commencing Wireless USB technology will be using approximately 1/3 of this spectrum.

EHF

This band is commonly used in radio astronomy and remote sensing. This band has a wavelength of ten to one millimetre, giving it the name millimeter band or millimetre wave, sometimes abbreviated MMW or mmW.

Signals in the 57–64 GHz region are subject to a resonance of the oxygen molecule and are severely attenuated. Satellite-based remote sensing near 60 GHz can determine temperature in the upper atmosphere by measuring radiation emitted from oxygen molecules that is a function of temperature and pressure.