English | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Home page | Services | Past achievements | Contact | Site map |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Page d'accueil | Services | Réalisations précédentes | Contact | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Français | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
pi | Mathematical constant | 4*atan(1) | 3.14159e+00 | |
Dw | Diameter of the wafer | 0.150 | m | 1.50000e-01 |
Aw | Usable area of the wafer | pi*(Dw-0.02)^2/4 | m2 | 1.32732e-02 |
lw | Line width | 1.0e-6 | m | 1.00000e-06 |
D | Defect density | 1.35e6 to 1.65e6 | 1/m2 | 1.35000e+06 to 1.65000e+06 |
Ap | Area of a processor cell | 9900*lw^2 to 10100*lw^2 | m2 | 9.90000e-09 to 1.01000e-08 |
Yp | Processor yield | exp(-(D*Ap)^0.5) to exp(-D*Ap) | 8.78892e-01 to 9.86724e-01 | |
Npw | Number of processors on the wafer | Aw/Ap | 1.31418e+06 to 1.34073e+06 | |
Ngpw | Number of working processors per wafer | Yp*Npw | 1.15502e+06 to 1.32293e+06 | |
IPSp | Instructions per second for each processor | 1.5/lw | 1/s | 1.50000e+06 |
IPSw | Raw instructions per second for the wafer | Ngpw*IPSp | 1/s | 1.73253e+12 to 1.98440e+12 |
Pp | Power consumption per processor | 1.35e6*lw^2 to 1.65e6*lw^2 | W | 1.35000e-06 to 1.65000e-06 |
Pw | Power consumption for the wafer | Ngpw*Pp | W | 1.55928e+00 to 2.18283e+00 |
Over all, ABOTEC is probably no longer necessary, since its role can be largely performed by conventional spreadsheet programs. ABOTEC does offer two major advantages, though. One is the ability to refer to parameters by their user-defined symbol name, rather than the obscure "C15" type of naming that is used in conventional spreadsheet programs. The other is the use of the "to" operator to handle engineering tolerances, and error bounds (and could also be used to indicate the range been two different theoretical models, such as for the processor yield calculation, Yp, in the above example, using the Seeds and Poisson models). For this, each arithmetic operation takes a pair of values (the minimum and maximum) for each input parameter, and generates a pair of values for its result. The program correctly takes account of the appropriate boundary conditions (when evaluating in the arithmetic and trigonometric expressions, for example, such as between sin(0.9*pi/2) to sin(1.1*pi/2) for example, or even tan(0.9*pi/2) to tan(3.1*pi/2)).