China
Africa
Explore chapters and articles related to this topic
Spin manipulations in magnetic nanostructures
Published in Guo-ping Zhang, Georg Lefkidis, Mitsuko Murakami, Wolfgang Hübner, Tomas F. George, Introduction to Ultrafast Phenomena from Femtosecond Magnetism to high-harmonic Generation, 2020
Guo-ping Zhang, Georg Lefkidis, Mitsuko Murakami, Wolfgang Hübner, Tomas F. George
A computer memory is any physical device capable of storing the binary information. There are two types of computer memory: volatile and non-volatile. A volatile memory requires electric power to maintain stored information. For example, an operating system of your computer uses random-access memory (RAM) to process your commands. Most RAM available in the market uses semiconductor technology, i.e., it is electric-current dependent and therefore inevitably volatile. Each bit of information in semiconductor RAM is stored in a memory cell which consists of a tiny capacitor and a transistor called MOSFET (metal-oxide-semiconductor field-effect transistor). Billions of such memory cells are mounted on an integrated circuit (IC) or so-called microchip. A typical speed of RAM in a personal computer is a few GHz, which means that it can read or write a few billion bytes per second. Roughly speaking, it is said that the number of transistors in an IC doubles about every two years, which is called Moore’s Law.a (There is of course a limit as to what extent in time this law is valid.)
The CPU and a microprocessor system
Published in Stuart Anderson, Microprocessor Technology, 2012
Computer memory devices come in two basic types, RAM and ROM. These words are not as helpful as they could be. RAM stands for ‘random access memory’, a term which can in fact be applied to both types. ROM is ‘read only memory’ which means that the computer system can obtain information stored in the memory (‘read’ it) but can't store anything new in it (‘write’ to it). RAM stores can both be read from and written to, but both ROM and RAM have the random access facility. Put simply, this means that any data byte stored in the memory chip can be obtained as easily as the next; it doesn't matter whereabouts in the chip it is.
Programming Techniques for Finite Element Analyses
Published in Jie Shen, Radhey Lal Kushwaha, Soil-Machine Interactions, 2017
Computer memory or storage is the place where program and data are located. In the prevalent von Neuman computer model as shown in Fig. 7.2, both program and data are stored in a single memory. Since the FEM applications in reality become more and more complex, how to allocate a limited memory space on a computer is a crucial factor to efficiently solve sophisticated problems.
The link transmission model with variable fundamental diagrams and initial conditions
Published in Transportmetrica B: Transport Dynamics, 2019
Jeroen P. T. van der Gun, Adam J. Pel, Bart van Arem
The computer memory requirements are also limited and independent of the duration of the simulation. For the calculation of within-link densities, only the most recent parts of the downstream boundary condition (last time) and the upstream boundary condition (last time) need to be known. Older traffic states from the boundary conditions cannot affect the current within-link densities anymore and can thus be forgotten, so that longer simulations do not require more computer memory. Note that the initial condition must be remembered, since initial densities exceeding the jam density may remain present on the link indefinitely. The computations of the receiving and sending flows additionally need to store cumulative inflow and outflow constraints for a finite number of time points in the near future. These data can be handled efficiently using a (variable-size) circular buffer data structure instead of a traditional array.
On orientational order in nematic and twist-bend nematic phases: a 2H-NMR study of binary mixtures of the odd dimer,1″,9″-bis(4-cyanobiphenyl-4′-yl) nonane, (CB9CB), and the monomer, 4-pentyl-4′-cyanobiphenyl, (5CB-d2)
Published in Liquid Crystals, 2018
Geoffrey R. Luckhurst, Bakir A. Timimi, Neil J. Wells, Herbert Zimmermann
Four mixtures of the dimer containing 5.70wt%, 14.70wt%, 25.65wt% and 36.65wt% of the deuteriated monomer were prepared by weight directly into standard 5 mm NMR tubes and were homogenised by heating into the isotropic phase and shaking repeatedly. These mixtures are denoted by DM(5.70), DM(14.70), DM(25.65) and DM(36.65), respectively. Our original measurements [18] were made on a Bruker AVII 400 MHz FT-NMR spectrometer using a single pulse sequence and a magnetic field of 9.40T; this high field made it possible to align the heliconical axis from a previously aligned nematic phase. Nonetheless, we found it more expedient to use a Varian Chemagnetics CMX Infinity 400 MHz spectrometer to record the 2H-NMR spectra because it gave greater temperature stability and enabled us to obtain a higher density of data. We used a single pulse sequence of 5μs width and a delay of 0.05s between successive pulses. The number of FID transients used was 4000–6000 depending on the strength of the deuterium NMR signal. The spectra were saved in 8192 words of computer memory. In these experiments, the original sample tubes were shortened and sealed so that they could be placed horizontally in the probe of the Varian spectrometer. It is the results obtained with the Varian spectrometer that are presented here. Before doing so we shall provide some general comments concerning NMR spectroscopy and the spectral analysis.
Mesh modeling and simulation for three-dimensional warp-knitted tubular fabrics
Published in The Journal of The Textile Institute, 2022
Haisang Liu, Gaoming Jiang, Zhijia Dong
An effective approach for warp-knitted tubular fabric simulation is proposed in this article. The structure and knitting technology of double-needle bed warp-knitted tubular fabric are introduced. The three-dimensional structure is regarded as a flat fabric according to the knitting technology by a vertical cut along one wale. Different from the common simulation method with one by one calculation for points, a set of matrices for coordinate space transformation including a translation matrix and a rotation matrix is created on the basis of the principle of coordinate translation and rotation in three-dimensional space. the translation matrix is used for the radius changing and the rotation matrix is applied to the stitch position and orientation on a course circle. For the convenience of expression, stitches of fabric are treated as particles in a particle-mesh model containing a flat model and a tubular model which is rotated from the former one. In the process of stitch transformation, control points for stitches are rotated independently on a topology-structure basis. According to the different rotation direction set on the flat fabric, both of the technical face and the technical back on the outside are simulated, which satisfy the requirements of different patterns and applications. In the meanwhile, the stitch deformation is taken into account using the mass-spring model theory so that the mesh effect can be better rendered. The tubular radius is set to be variable through the coordinate translation transformation of the initial stitch circle. Consequently, a tubular structure with a changeable tube diameter is able to be simulated. The results can be represented on the warp-knitting CAD software visually, which shows that using this method the regular warp-knitted tubular products can be well simulated. Nevertheless, due to the limitations of the rendering model, some more complex structures can’t be simulated, which is imperative to be studied in future research. Some tubular warp-knitted fabrics have large pattern width and pattern height, which results in a huge computation amount and data amount. The smoothness of curves connecting the control points and the yarn surface is determined by the rendering parameters. The more the rendering parameters are, the smoother the surface is. Meanwhile, it will also increase the rendering load. However, computer memory has a fixed capacity. Too much computation will cause the program to crash. Thus, it is a problem to be solved to reduce the amount of computation while ensuring the rendering effect.