Speed advantage

Parallel processing and the RISC set give transputers a considerable speed advantage over conventional serial processors for handling information or flows of data. 

Max. rpm Ratio Maximum Half speed speed Max. speed Half speed Max. Half speed speed Advantages Limitations... 

From any real input, this provides an output in the range 0,1. The Tanh function (Figure 2.23) has a similar shape, but an output that covers the range -1, +1. Both functions are differentiable in closed form, which confers a speed advantage during training, a topic we turn to now. 

It took the short time of one year or so to solve the structure of rhinovirus which causes the common cold. This relied on two major advances in methods. The first was the use of synchrotron radiation in data collection. Nearly a million reflections were collected on the protein crystallography facility at the Cornell Synchrotron source in a matter of days. This conveyed a speed advantage over data collection on a conventional source and also ameliorated an otherwise impossible problem of radiation damage when long exposure times were used. The far greater rate of radiation damage in the X-ray beam in relation to plant viruses is symptomatic of an inherently less stable protein capsid and the absence of quasi-symmetry. The capsid consists of 60 copies each of four proteins and the virus with about 30 % RNA has a total molecular weight of about 8.5 million. 

Apart from its speed advantage, a maglev system of either type is far superior to conventional trains for many reasons. Larry Johnson ticked them off ... 

Given the huge investments in time and resources in the development of serial processors and their software programs, it is unlikely that serial processors will be dropped in favor of transputers in the near future. However, in situations where large amounts of information must be handled in a very short time, transputers are used because they hold a distinct speed advantage over conventional processors. 

The most common type of emissions spectrometer used today is of the multielement type. This type of instrument has the speed advantage that the plant chemist wants as opposed to a sequential type of spectrophotometer. A multielement type can measure up to 60 elements at a time. When an arc or spark type of instrument is used, it is necessary to integrate and average the signal produced to obtain reproducible line intensities. 

A summary of the computer times and the number of trials required to solve a variety of examples by the methods enumerated above is presented in Table 4-17. These results show that for absorber-type columns (Examples 4-2, 4-3, and 4-4), procedure 1 requires fewer trials than do procedures 2 and 3. For columns which have a relatively small number of plates, all three procedures require about the same amount of computer time. As the number of plates and components are increased (see Example 4-12), the speed advantage of procedures 2 and 3 over procedure 1 becomes more pronounced. 

For distillation columns, the 6 method presented in Sec. 8-1 required about the same amount of computer time as did procedures 1 or 2 for Example 8-1. As was found for distillation columns without reaction, the computer times required by the 6 method and procedures 1 and 2 do not differ significantly. Hefoever, as the number of plates and components are increased, the speed advantages of the 0 method over procedures 1 and 2 can be expected. 

When a direct test was made of our first derivative program against Boyd s program for the n-hexane molecule, it was found that our scheme was approximately twice as fast. Since the calculation time required per iteration for the matrix method will increase at a rate which is proportional somewhere in between the square of the number of atoms (if calculation of the matrix elements is the slow step), or to the cube of the number of atoms (if diagonalization is the slow step), whereas for our method the rate is in between the first and second powers of the number of atoms, it would seem that the speed advantage of our method will increase with increasing molecular size. 

Large-diameter (>10-ft) columns. Here the cost and speed advantage of dry packing are the controlling considerations. 

Because the number of time-domain slices (and hence the number of recorded projections) is relatively small, the density of sampling points is far lower than the density used in the conventional experiment, which must examine every point on the complete Cartesian grid while satisfying the Nyquist condition and the requirement for adequate resolution. This is where the critical time saving occurs. With this limited radial sampling [13], the speed advantage increases by an order of magnitude for each new evolution dimension beyond the first. This opens up the... 

---