#
(In Mano & Kime unless otherwise noted)
---
---
After peer grading we will review for the exam.
4/30
at
5:00 pm
5/01
Do 11-1, 11-2*
You may turn this assignment in for regular grading by
placing it in a box which is located near Prof. De Boer's
office door.
This assignment will be graded overnight and placed in
the plastic bin labeled "De Boer's Courses" by noon on
Saturday. You may pick up your homework at your
convenience on Saturday afternoon
4/30
5/01
zoned recording and logical block addressing.
Do 12-1*, 12-3, And answer this question:
12-A What are the advantages of packet-based
I/O such as in USB?
Note errata on page 601.
Note errata on problem 12-1*.
This assignment will be graded overnight and placed in
the plastic bin labeled "De Boer's Courses" by noon on
Saturday. You may pick up your homework at your
convenience on Saturday afternoon
4/28
4/30
Do 10-27*, 10-29
Note errata on problem 10-27*.
4/26
4/28
Do 10-23*, 10-24, 10-25*
Hint: For Problem 10-23* assume that the
"16" bit register is named "R".
Also, this supplement written for an older
edition of our text can be helpful to
understanding table 10-7, page 527 in our text.
(The example in the supplement assumes an 8-bit
register, not a 16-bit register. Be sure to
make the appropriate change.)
Hint: For Problem 10-25* a computer does
subtraction A – B as A +
+ 1
(details: text page 345) and assumes the numbers are twos
complement signed integers. Then "borrow" is
the NOT of the carry out of that addtion. This
borrow is saved in the carry flag bit in the
place of the normal carry bit.
4/23
4/26
Do 10-18, 10-20*, 10-21
Note: On 10-18 work the problem in base ten
using 13 places right of the radix point.
Note: On 10-20* the answer posted on the web
is wrong. The meaning of E is incorrect on the
first and last lines. You should list all the
binary bit patterns possible (e in binary) and
the value of the exponent (E in decimal) in
a style similar to the outer two columns of
Table 10-6 on page 526. Also note carefully
the text underneath Table 10-6.
Note: On 10-21 for part (b) list
E = (the smallest), -1, 0, 1, and (the largest)
values possible. On part (c) the calcuation
should be an estimate in base ten and in
scientific notation, to at least three significant
figures of the largest and smallest positive
normalized numbers.
Optional:
Prof. De Boer offers the links below not that
he expects you to read them all, but rather to
give you an appreciation for the complexity of
representing floating point numbers. Just open
the links and page through them a bit to get a
sense of how much work has been done by
others.
Wikipedia: IEEE 754-1985, a summary
What every CS should know about floating point
And the above is just the tip of an iceberg!
4/21
4/23
Do 10-3*, 10-4, 10-17*
4/19
4/21
Do 10-5, 10-6*, 10-7
Note errata on page 510
4/16
4/19
Do 10-1, 10-2*
Hint: You might find it easier to work on
10-2* first.
For both 10-1 and 10-2*,
SUB and DIV are not described in the text.
Some examples are shown below:
Three Address Instructions:
SUB R1, R2, R3 R1
R2 – R3 DIV R1, R2, R3 R1
R2/R3 Two Address Instructions:
SUB T1, T2 M[T1]
M[T1] –
M[T2] DIV T1, T2 M[T1]
M[T1]/M[T2] One Address Instructions:
SUB X ACC
ACC – M[X] DIV X ACC
ACC/M[X] Hint for 10-2: Part (b) requires only two
temporary memory variables.
Note errata on page 499
4/14
4/16
Do 9-12, 9-15
Hints for problem 9-15:
It might be easier to do part (b) first.
For part (a) fill unused fields with "X" for
"don't care." Note that address fields are the
only kind that can be unused since Figure 9-6
shows explicit logic for the other fields.
For part (a), line 3, in "R[5] + 2" the "2" arrives
from the "Constant in" lines to MUX B. (See
Figure 9-15 on page 471.)
In part (a) line 5 the "PC + se PC" should be
"PC + se AD". Also let AD = 25 (base ten--
convert it to binary). The value of AD is supplied
in a pair of fields, split to "Left" and "Right." The
left is most significant. This forms effectively a
six-bit field. (See Figure 9-14 on page 467.)
Note errata on problems 9-12 and 9-15.
4/12
4/14
Do 9-8*, 9-9
Hints on problem 9-9:
Zero can be formed as the exclusive OR of two
identical words. For example, R0 XOR R0 = 0.
Assume numbers are signed integers in the twos
complement number system. Then
_
X - Y = X + Y + 1 It is possible that some bits in the control
word do not matter. Mark those as "X" for
"Don't Care"
Although 9-10* is not assigned, understanding
that problem and its solution (posted on the
web) might help you understand and do 9-9.
4/09
4/12
Do 9-1, 9-2*
Note: In 9-1, "selection lines" refers to
"Destination Select," "A Select," and "B Select"
Note: For problem 9-2* assume the ALU is
designed to work with numbers in twos
complement format. Overflow detection is
described on page 165 of your text. The result
of your logic should be that N = 1 iff the
output of the ALU represents a negative number.
Bit Z = 1 iff the output represents zero.
Bit V = 1 iff there was an overflow.
Bit C = 1 iff there was a carry out of
the most significant place.
4/07
4/09
Do 8-9, 8-10, 8-11
4/05
4/07
Do 8-7, 8.8*
Note errata on problems 8-7 and 8-8 .
4/02
4/05
Do 8-1*, 8-2, 8-3* EXCEPT change
"64K X 16" to "1 M x 4" and change
"32000" to "196865." Note: The solution
given online goes with the problem as printed
in the textbook. The assignment is to re-work
the problem with the organization and address
given here.
3/31
4/02
Do 7-11* Note errata on p404, Problem 7-11*
Note Figure 7-15 too.
7-16 and use hierarcy including a diagram of a
single cell of the register.
Note additional errata on page 259 posted
Wednesday 3/24. (Relates to PS #20 and lab.)
3/29
3/31
page 353 through the end of the Section.
Do 7-4*, 7-5*, 7-8
Note errata on problem 7-4* (Errata Posted 4/16)
3/26
3/29
on page 353.
Do 7-3, 7-B
---
---
Study for the test on Wednesday, 3/10
3/24
3/26
Do 7-A (click on link).
Here is a datasheet for the '175 (problem 7-A).
See especilly the function table on page 3 and
the block diagram on the right side of page 3.
3/08
3/24
Do 6-9. Answers should be in the style of. . .
"Signal RX at X ns has a <name problem>."
Hint: The times mentioned in the answers always
correspond to a clock edge.
6-15 and also specify how to connect the inputs
X, Y, and Z to the ROM and how to connect the
outputs A, B, C, and D to the ROM.
(Note errata on 6-15),
6-20 except mark fuses to remain intact on the
fuse map for a TIBPAL16L8-25C part. The fuse
map can be found on page 5 of the datasheet.
Note: Discussion of Problem 6-20
3/05
3/08
Do 5-35 using this state assignment:
A = 00, B = 01, C = 10, D = 11
Note that the problem statement continues on
page 291 underneath Figure 5-45.
Note errata on page 251. Copies of this
errata were handed out in class on 2/26. Extra
copies are available from De Boer's plastic bin
near the engineering pod.
Note errata on page 254.
Note errata on page 259 posted Wed. 3/24.
3/03
3/05
Read Ch 5 Sec 5-5 from page 240 to the end.
Do 5-12, 5-13*
Note errata on page 243.
Note: Discussion of Problem 5-12
3/01
3/03
"Designing With D Flip-Flops" on page 240.
Do 5-9, 5-11*
3/01
Do these problems:
5-A. Draw a gate-level logic diagram of the SR
master-slave flip-flop shown in Figure 5-9 on
page 216. (Note Figure 5-7 on page 213.)
5-B. (Click on the link.)
5-6 (On page 281 in your text.)
2/24
2/26
Do 5-4
2/22
2/24
Do 5-1, 5-2.
Note: For these problems use the input
sequences given in the tables below. Fill in
the outputs shown on these tables. The
complete table or the equivalent information
in a vector waveform file print-out is the
answer.
_ _ _ _
S R | Q Q C S R | Q Q
-----+----- --------+-----
1 1 | 0 0 0 |
1 0 | 0 0 1 |
0 1 | 0 1 0 |
1 1 | 0 1 1 |
0 1 | 1 0 1 |
0 0 | 1 0 0 |
0 1 | 1 1 0 |
1 0 0 |
Table for 5-1 1 1 0 |
1 1 1 |
1 1 0 |
1 1 1 |
1 0 0 |
Table for 5-2
Note: Discussion of Problems 5-1 and 5-2 Optional supplemental reading: RS Nand Latches.
2/19
2/22
Do 4-23
Hint: Although 4-20* is not assigned, look at
that problem statement and its solution which
is posted on the web.
Another Hint: The sentence, "Note that
complemented inputs are available." means that
although the signal a_n is a separate input
from a, you should assume that a_n is always
the logical not of a. Wherever you might like
to use
you
should instead use the signal a_n. The same goes for the other similarly
named pairs of signals.
2/17
2/19
Do 4-4 except do it in this order:
1.) The given numbers are unsigned binary.
Pad each with leading zeros until it is
eight bits wide. This converts the
numbers from unsigned binary to signed
two's complement (and all of them are
positive).
2.) Negate subtrahends by taking the
two's complement.
3.) Add. The result is the answer without
any further manipulation. The answer is
in the two's complement notation.
4.) For each case, note if there is
"overflow" or "no overflow."
4-45 Note: this problem is not in the
textbook, click the link to view it.
2/15
2/17
Do 4-2* (Use manual verification),
4-3* and note errata on 4-3*
Also for each note if there is overflow or not.
Definition: Overflow upon negation occurs iff
negating a non-zero number does not change
the sign of the number.
Note that by definition the number zero cannot
overflow upon negation. (–0 = +0 = 0)
Note: In 4-3* the given numbers are mostly
negative since most start with a 1 bit.
Note errata on Page 166
---
---
Study for the test on Wednesday, 2/10.
2/12
2/15
Do 4-1 except begin with a truth table, then
use K-maps to derive equations for outputs
C2, S1, and S0. Stop with equations in SOP form
since drawing a schematic will be too tedious.
Note that the inputs are A1, A0, B1, B0, and C0.
You will need five-variable K-maps for this
problem. Hint: Drawing the K-maps in a
horizontal style works best for this problem.
FYI "Karmah" is an online Java applet that can
do K-maps up to 8 inputs. It uses a slightly
different style (overlays) than Prof. De Boer
illustrated in class.
Optional Reading on five-variable K-maps.
Explains both the grey code style preferred by
Prof. De Boer (so that "folding" works) and
the overlay style that generalizes (poorly) to
even larger K-maps.
2/08
2/10
Do 3-35*, 3-42*, 3-46, 3-47*
2/05
2/09
2/10
handed
back
during
labs
(PS#10
too)
Do 3-24, 3-25, 3-21,
3-32 except use a type '138 decoder and NAND
(instead of OR) gates.
Hint 1) Output a logic-1 to a LED to turn it on.
Hint 2) Read part (b) before doing part (a).
Hint 3) Use a block symbol for the '138
decoder rather than drawing all the gates.
This decoder is comparable in function to the
the one shown in your text in Figure 3-33.
Hint 4) Use value fixing on unneeded gate inputs.
Note: Professor De Boer assigned this problem
because it poses an interesting technical
challenge, but he does not endorse gambling.
Note errata on Figure 3.33
2/03
2/05
Do 3-2*, 3-16, 3-20
2/01
2/03
Read Ch 3 Sec 3-1
Do 2-8, 2-30, 3-1
1/29
2/01
Do 2-24, 2-25*, 2-34
1/27
1/29
Do 2-15*, 2-16, 2-19*, 2-31
1/25
1/27
Do 2-7*, 2-11, 2-12*, 2-14
Note errata on problem 2-12*
Note errata on page 60, Figure 2-9(b)
Note on 2-14: "Optimize" means to "simplify"
as shown by several examples in section 2-4.
1/22
1/25
Do 2-1*, 2-2*, 2-10*
1/20
1/22
Also read about binary prefixes from NIST
Optional--read more about binary prefixes here.
Do 1-10*, 1-22, 1-24, 1-25*
1/18
1/20
Do 1-4, 1-5, 1-6, 1-7*, 1-8, 1-9*
Note: The answer to 1-5 must be exact.