会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 3. 发明专利
    • BINARY CODED DECIMAL NUMBER DIVIDING SYSTEM
    • JPS5987543A
    • 1984-05-21
    • JP19661882
    • 1982-11-09
    • HITACHI LTD
    • YABE HIDEAKIOOSHIMA YOSHIOISHIKAWA SUKETAKAOOTSUKI TOORUFUKUDA MASAHARU
    • G06F7/491G06F7/493G06F7/496G06F7/52
    • PURPOSE:To select directly a desired multiple value register, to reduce the number of gate stages, and to execute the processing at a high speed by modifying in advance a data to be stored in a quotient forecasting table. CONSTITUTION:A value modified to (0110)2-(1111)2 containing a redundant part of a binary coded decimal representation is stored in a quotient forecasting table 4 of a binary coded decimal number dividing system. Also, each upper bit of a dividend register 1 and a divisor register 3 is set as an address of the table, and a value read out of the table 4 is set as a value which modifies +6 to the forecasting quotient. The lower three bits of four bits of the forecasting quotient read out of this table 4 are used directly as a selecting signal of a multiple value register 7, and a multiple is selected. Also, at the same time, 6 is subtracted by a -6 circuit 10 of a quotient determining circuit 11, and thereafter, it is stored in a forecasting quotient register 5. Subsequently, a multiple value of a divisor from the register 7 is added to a decimal adder and subtracter 8, is subtracted from a dividend value of the register 1, a value derived by the circuit 11 is stored in a register 2, and the processing can be executed at a high speed.
    • 6. 发明专利
    • DECIMAL MULTIPLYING DEVICE
    • JPS6011927A
    • 1985-01-22
    • JP11955683
    • 1983-07-01
    • HITACHI LTD
    • OOTSUKI TOORUOOSHIMA YOSHIOISHIKAWA SUKETAKAYABE HIDEAKIFUKUDA MASAHARU
    • G06F7/496G06F7/491G06F7/506G06F7/508G06F7/52G06F7/527
    • PURPOSE:To obtain a multiplying device which can perform a decimal loop addition at a high speed by using the double input of a 3-input carry save adder as the input of an intermediate product which is shown in the form of a carry and another input as the input of the product of a decimal part. CONSTITUTION:The double input of a 3-input carry save adder is used as the input of an intermediate product shown in the form of a carry, and another input is used as the input of the product of a decimal part. For instance, 6 is added every decimal digit by a 6-addition correcting circuit 28 as the initial setting to start the loop addition. Then the intermediate product held in the form of a sum and a carry is added with the product to a decimal part obtained by another arithmetic means by a 3-input carry save adder CSA29. Then the correction is given to the sum output as well as to an upper decimal digit from a sum output correction circuit 31 and a decimal upper digit corretion circuit 32 for each decimal digit of the CSA29. A decimal addition is carried out for the first intermediate product and the product of a decimal part after the bit positioning is carried out to the sum obtained after a carry correction.
    • 7. 发明专利
    • Dividing device
    • 分体装置
    • JPS59173843A
    • 1984-10-02
    • JP4714783
    • 1983-03-23
    • Hitachi Ltd
    • FUKUDA MASAHARUOOSHIMA YOSHIOISHIKAWA SUKETAKAOOTSUKI TOORUYABE HIDEAKI
    • G06F7/493G06F7/496G06F7/508G06F7/52G06F7/535
    • G06F7/535
    • PURPOSE:To read out an estimated quotient at a high speed by using an RAM having small storage capacity by dividing a quotient estimating table into plural parts and selecting one output out of plural ones from the quotient estimating table by bit strings of the divisor and divident. CONSTITUTION:A bit string consisting of respective upper bits of a divident register 1 and a divisor register 3 is inputted to a logical circuit 40 as a bit string 41. A bit string excluding the lower 2-bit of the bit string 41 is inputted to four quotient estimating tables 4200, 4201, 4210, 4211 as a common address through a line 41a. The 2nd bit from the lowmost bit of the bit string 41 is inputted to comparators 4300, 4301, 4310, 4311 through a line 41b and the lowmost bit is inputted to AND gates 4400, 4401, 4410, 4411. All outputs of the AND gates 4400, 4401, 4410, 4411 are inputted to an OR gate 45 and the output of the OR gate 45 is inputted to an estimated quotient register 5.
    • 目的:通过将商估计表分成多个部分,通过使用具有小存储容量的RAM来高速读出估计商,并通过除数和除数的位串从商估计表中选择一个输出 。 构成:将由除数寄存器1的各高位组成的位串和除数寄存器3作为比特串41输入到逻辑电路40.将比特串41的下位2除外的比特串输入到 四个商估计表4200,4201,41210,41211作为通过线41a的公共地址。 来自比特串41的最低位的第二比特通过线路41b输入到比较器4300,4123,4310,41311,并且最低位被输入到与门4400,4401,4107,411。AND门的所有输出 4400,4401,4410,4411被输入到或门45,或门44的输出被输入到估计商寄存器5。
    • 10. 发明专利
    • Rounding arithmetic system
    • 圆形算术系统
    • JPS59184944A
    • 1984-10-20
    • JP5924083
    • 1983-04-06
    • Hitachi Ltd
    • ISHIKAWA SUKETAKAOOSHIMA YOSHIOYABE HIDEAKIFUKUDA MASAHARUOOTSUKI TOORU
    • G06F7/38G06F7/48G06F7/483
    • G06F7/48G06F7/49963
    • PURPOSE:To facilitate an easy rounding arithmetic system by adding a data pattern where only the position corresponding to the operand data and its prescribed bit is set at (1) with other positions set at (0). CONSTITUTION:The operand data consists of (m) bits and requires up to (n) bits (m>n) as a result of a rounding operation. In such a case, only (n+1) bits are set at (1) with other bits set at (0) in a data pattern. Such a pattern is added with the operand data. Then the carry-up is produced when the (n+1) bits of the operand data are set at (1). Then the generation of the carry-up is inhibited when said (n+1) bits are set at (0). Thus the rounding operation is facilitated for the operand data of (n+1) bits.
    • 目的:为了方便一个简单的舍入算术系统,通过添加一个数据模式,其中只有与操作数数据相对应的位置及其规定的位被设置为(1),其他位置设置为(0)。 构成:操作数数据由(m)位组成,由于舍入操作需要最多(n)位(m> n)。 在这种情况下,在数据模式中,只有(n + 1)位被设置为(1),其他位设置为(0)。 这样一个模式加上操作数数据。 然后,当操作数数据的(n + 1)位被设置为(1)时产生进位。 那么当所述(n + 1)位被设置为(0)时,禁止产生加载。 因此,对于(n + 1)位的操作数数据,舍入操作变得容易。