Transcript Chip - Computer Science and Engineering

CS 2204
Lab 7
Experiment 3 - 4
Spring 2008
Digital
Logic and
State Machine Design

Experiment 3 Lab 7 Outline
 Presentation

Semiconductor technology overview
 Future aspects of the technology



A machine playing strategy
Analysis of Block 2 of the term project
Digital product development overview
 Block partitioning
 Implementing blocks

An overview of today’s individual work
 Individual work

Experiment 3
 Develop the Rightmost zero display circuit of the Ppm term
project

Experiment 4
 Develop the Rightmost largest display circuit of the Ppm term
project
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4
Page 2


Digital Hardware Evolution
Switches since 1920s
1) Electromechanical : Relays
2) Electronic switches
i. Vacuum tubes
AMD Opteron die
ii. Discrete transistors
iii. Integrated circuit transistors : Die contains
transistors
a) Small Scale Integration (SSI) (< 64 transistors on chip),
1960s
 AND gates (7408), OR gates (7432), NOT gates (7404) chips
b) Medium SI (MSI) (< 2K transistors), 1960s
 Decoder, encoder, multiplexer, counter chips
c) Large SI (LSI) (< 64K transistors), 1970s
 Micro-controller, special-function chips (calculator chips)
d) Very Large SI (VLSI) (< 2M transistors), 1980s
 Memory, special-function chips
Intel Dual-Core
Itanium 2 die
e) Ultra Large SI (ULSI) (> 2M transistors), 1990s
 Memory, microprocessor chips
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4
Page 3

Semiconductor Technology Today
 Moore’s Law holds since the 1960s :

Every two years the number of transistors on a chip doubles
 Transistors become smaller ≡ Process length becomes smaller
• We have been able to reduce the process length
• The process is 65nm now
• The process will be 45nm soon

Smaller transistors are susceptible to alpha particles
 Programs crash if we do not detect faults !

More transistors can be defective
 Programs would not run if we do not test chips well
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4
Page 4
Digital Hardware Evolution


Switches since 1920s
2)
Electronic switches
A transistor
nano size
iii. Integrated circuit transistors : Die contains transistors
f)
Today : Multi-chip module, MCM (>1 die on chip), Giga Scale, etc.
(200M–2B transistors)
Transistor size determined by process length
3)
4)
Optical switches ?
Molecular switches ?

5)
Biological ?
???
MIPS R10000 die
Sun Niagara die
IBM Power 6 die
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4
Page 5

Silicon Technology Today
Intel Dual-Core
Itanium 2 die
Intel Dual-core Itanium ® 2 processor (>26MB cache)
2006
1,720,000,000
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4
Page 6

Every two years the speed of microprocessors doubles
 The processor speed increases 50% a year !


But, memory speed increases 10 % a year !
Microprocessor speed for an application depends on
 Number of operations in the application (lower better)

The quality of the code

Do more operations in parallel

Perform each operation faster
 Number of parallel operations performed (higher better)
 Clock frequency (higher better)

 Because of Moore’s Law : transistors are smaller and wires are shorter
Until 2005 increasing the clock frequency was the main way to
increase the speed
 Power consumption (heat generation) increases with the frequency

The chip has to be cooled
 A heat sink or a fan or a liquid
 Since 2005 power consumption changed way to increase speed
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4
Page 7
Power Density Increasing Exponentially!
 Power Density Increasing Exponentially !
1000
Power doubles every 4 years
Rocket
Nozzle
Watts/cm 2
Nuclear Reactor
100
Pentium® 4
Pentium® III
Pentium® II
Hot plate
10
Pentium® Pro
Pentium®
i386
i486
1
1.5m
1m
0.7m
0.5m
0.35m
0.25m
0.18m
0.13m
0.1m
0.07m
Courtesy : “New Microarchitecture Challenges in the Coming Generations of CMOS Process Technologies” – Fred
Pollack, Intel Corp. Micro32 conference key note - 1999. Courtesy Avi Mendelson, Intel.
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4
Page 8

Multi-Core Microprocessors
 Microprocessor speed depends on


Number of operations in the code (the quality of the code)
Number of parallel operations performed
 Dual-core microprocessors with reduced frequency consume
less power (generate less heat)
• Two/Four/Eight cores perform more operations in parallel
The speed increase continues into the future with more cores on chip


Clock frequency
Number of cores per chip doubles every two
years
 The memory can become a bottleneck


The memory speed increases 10% a year
The memory wall problem
 Parallel Programming

Major concern now
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4
Page 9

Next 10 Years
 Next 10 Years : Double number of cores every two years
Make sure to handle
errors due to
Alpha particles
Defective transistors
Make sure to handle
Memory Wall
Make sure to improve
Parallel Programming
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 10

Next 10 Years
 Traditional computing will continue

A C/C++/Java program for an application becomes
Software
 Applications
• Intel : Recognition, Mining, Synthesis as platform
2015 Workload Model (on massively parallel core
chips)
• IBM : Presence information, knowing where and
things are and how to best match them, people are
sensorized
• Microsoft : Intention machine, computer predicts
user intentions and delivers useful information
• CMU : Computational thinking, computer science
based approach to solving problems, designing
systems, understanding human behavior
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 11

FPGAs for only Prototyping ?
 Are they used in commercial products ?


FPGAs are becoming cost competitive with microprocessors
FPGAs are becoming speed competitive with custom chips
 FPGAs are increasingly used for applications where


Speed and programmability matter
In the future engineers/programmers will write
code that will be converted to two parts :
 A machine code that will be run by processors and
 A bit file to program the reconfigurable area (CLBs)
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 12

New computing types
 A C/C++/Java program becomes

Hardware (no need to write VHDL/Verilog programs to
design hardware)
 A custom chip corresponding to program runs application OR
 A reconfigurable chip (FPGA) with a reconfigurable area runs
application

Part software and part hardware
 Reconfigurable chip (FPGA) with cores and reconfigurable areas runs
application
• Software is run by processor cores and
• Hardware is in the reconfigurable area
Reconfigurable area
to do operations in
hardware
Hybrid computing
Processor core
to run software
These FPGAs are available
now but we need much
better tools
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 13

Long Term Forecast
 Microelecromechanical systems (MEMS) with
computing elements

Microcameras, microsensors, micromirrors, micromotors,..
with computing elements
 Smart Dust at UC Berkeley

Bio MEMS
 SEMATECH : Consortium of semiconductor
manufacturers from America, Asia & Europe

Predictions for year 2020 (from its updated 2006 ITRS,
study)
 On-chip clock speed : 12 GHz (From 2007 Final draft)
 Number of transistors on a high-speed microprocessor chip :
17 Billion
 32 Gbit DRAM chips
 Process length : 14 nm

http://www.sematech.org
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 14


Longer Term Prospects
Nanotechnology
 Programmable materials
 NEMS
 Bio NEMS

Nano medicine
 Drug delivery
 Smart diagnosis
 Nanocomputing

1 Watt supercomputer
What is software
for them ?

 Quantum computing
IBM Blue Gene/L molecular dynamics demo
 Molecular computing
• Molecular self assembly
• Testing of molecular structures
• Adaptive molecular structures
Merger of bio and non-bio structures
 Synthetic biology
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 15

Longer Term Prediction by Individuals
 By 2019 a $1000 computer will match the processing
power of the human brain

Raymond Kurzweil, KurzweilAI.net, 9/1/1999
 His keynote speech at the Supercomputing Conference (SC06)
in November 2006
• The title of his talk is “The Coming Merger of Biological and
Non-Biological Intelligence”
 Singularity point ?
 Brain downloads possible by 2050

Ian Pearson, Head of British Telecom’s futurology unit,
CNN.com, 5/23/2005
 Computers will be used as virtual brain extensions ?

Direct brain - Internet link ?
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 16

Longer Term Prediction by an Individual
Biochip Group at Mesa+,
University of Twente, Holland
Hans Moravec, 1998
Many ethical issues will be facing you ! Being prepared will help !
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 17

Analysis of the Term Project
 The term project black-box view
 The term project operation diagram
 The term project black box partitioning
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 18

The Term Project, Ppm
 The black-box view
From the input devices
13
From page 2 of the Term Project Handout
19
Ppm
To the output devices
Figure 1. The Ppm black box view.

Ppm is sequential (not combinational)
 A large number of FFs are used !
 We need to partition the Ppm based on major operations
• We have to obtain the operation diagram
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 19
Ppm
Input/output
relationship
Ppm
operation
diagram
From page 8 of the Term Project Handout
LD6-LD8 on the
FPGA board
show the
current state
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 20

The Ppm Digital System Partitioning
M1
M2
From page 9 of the Term Project Handout
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 21

The term project black box partitioning
• Six schematics for six blocks
•
•
•
•
Block 1 : Control Unit : ppm1.sch schematic file
Block 2 : Input/Output : ppm2.sch schematic file
Block 3 : Human Play : ppm3.sch schematic file
Block 4 : Play Check : ppm4.sch schematic file
• Experiment 1 is on a circuit in this block
•
•
Block 5 : Points Calculation : ppm5.sch schematic
file
Block 6 : Machine Play : ppm6.sch schematic file
• The Machine Play Block uses all other blocks except the
Human Play Block
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 22

A Machine Player Strategy
Largest reward = 0 ?
N
Y
Player 2 does not have (64)10 or more
points & there is a position with a zero
& RD is not zero
N
Play on the
(rightmost)
largest
reward
position
(directly
if equal)
Skip
Y
Play on the
(rightmost)
zero position
directly
(if equal)
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 23

A Machine Player Strategy
 Its Implementation
Largest reward = 0 ?
N
Y
Player 2 does not have (64)10 or more
points & there is a position with a zero
& RD is not zero
N
Play on the
(rightmost)
largest
reward
position
(directly
if equal)
Skip
Y
Play on the
(rightmost)
zero position
directly
(if equal)
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 24

Input/Output Block, Block 2
 Has 75 inputs and 38 outputs
 Controls input/output devices on the FPGA board and
generates timing signals
 Has sequential circuits
75
Block 2
38
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 25

The Ppm Data Unit
75
Block 2
38
From page 17 of the Term Project Handout
 Block 2, Input/Output Block
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 26

The Ppm Data Unit
 Block 2, Input/Output Block
75
Block 2
38
 Controls input/output devices on the FPGA
board and generates timing signals

Three major operations
 Controls Input/Output Devices
• I/O Buffer Subblock
• Display Subblock
 Generates timing signals
• Timing Subblock
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 27

The Ppm Data Unit
32-bit frequency divider
 Block 2, Input/Output Block
Timing
Subblock
I/O Buffer
Subblock
Display
Subblock
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 28

Block 2, Input/Output Block Development
 I/O Buffer Subblock implementation
SW1-SW4
P1SEL
Input buffers
BTN1-BTN4
Output buffers
Clock : 25 MHz
RD
Add
PD3 – PD0
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 29

Block 2, Input/Output Block Development
 Timing Subblock implementation
32-bit frequency divider
Clock from the board : 25 MHz
Sysclk
6 Hz
P2clk
192 Hz
48 Hz
Rdclk
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 30

Block 2, Input/Output Block Development
 Display Subblock implementation
Select
Displays,
Points,
next RDs
Output to
displays
one digit
at a time
a 4-bit
code
Convert the 4-bit code
of the selected display
to a 7-bit code
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 31

Developing a new PCB
1) Development Cycle on Computers
Major error : Redesign
DESIGN
TEST
MODIFY
Major error : Redesign or terminate the project due to TTM
2) Dev. Cycle with off-the-shelf chips
Mount
Test
Modify
Major error : Redesign or terminate the project due to TTM
3) Dev. Cycle on prototype PCB
Fabricate
Test
Modify
PCB
Which chips and how many ?
TEST : Simulating by applying input
combinations, test vectors, may not be
possible. It may be coarse grain simulation
During testing if you see MODIFYING
hardware to optimize it is possible, do that
after you correct logic and timing errors.
Then, test again to see if your minimization
has logic/timing errors
Mount : Chips are mounted on
bread/boards and wired
Test : apply test vectors to the chips
Modify : chip mounting/wiring is changed
and tested or a simple design change is
made on computers, simulated, then chip
mounting/wiring is redone and tested
Fabricate PCB at a fabrication facility,
mount chips and other components
Apply test vectors to the PCB
Modify means chip mounting/wiring is
changed and tested
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 32

Developing a digital product
 A new PCB

Which chips and how many is determined by
 The application (major operations)
 Available chips of the technology chosen
 Besides speed, cost, power, etc. : product goals
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 33

Development Cycle on Computers

PCB
DESIGN
1) Input/Output Relationship
a) A simple block
Obtain the truth table of the combinational circuit with
less than 5 inputs then move on to Implementation (2)
Obtain the state diagram of the sequential circuit with
less than 5 FFs then move on to Implementation (2)
b) A complex block

Obtain the operation table or the operation diagram
►Try to implement it in (2)

If it cannot be implemented immediately in (2)
► Partition it


2) Implementation Try to use registers, counters,
TEST
MODIFY
shift registers even if it is a
simple sequential circuit
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 34

Development Cycle on Computers
PCB
For the current (sub)block we can get a truth table or a state diagram ?
Step I (a)
N
Y
Obtain an operation table or an operation diagram
Step I (b)
Current (sub)block is implementable ?
Step II
Y
Implement the current (sub)block
Step II
N
Partition it into (sub)blocks
Step I (b)
Works and satisfies design goals ?
TEST
Y
Any other (sub)block
to implement ?
Y
N
A simple design change
MODIFY
N
Design complete
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 35

Designing a Complex Block
 Partition it into pieces based on major operations

Besides the design goals and the technology
 One block for each major operation

These major operations are often
 Additions, MUXings, comparisons, decodings, encodings,
DeMuxing, registering, counting, etc.

These operations are already implemented by available
components/chips :
 ADDers, Multiplexers, Comparators, Decoders, Encoders,
DeMuxes, Registers, Counters, shift registers, etc.

This happens frequently for real-life applications
PCB
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 36

An Unusual Major Operation (an unusual block)
 Trying to implement a block

If it has < 11 inputs implement it by using programmable
chips that are memory chips
 ROMs, RAMs

If it has 11 to 20 inputs, implement it by using programmable
chips
 PLAs, PALs, GALs, FPGAs

Otherwise (complex or too many inputs)
 Break it up or
 Repartition one level up, or
• Two levels up, or,…
• All the way up (redesign !)
 Eventually, the resulting operations will be additions,
comparisons, multiplexing, decoding, etc.
PCB
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 37
Designing a New PCB


PCB
DESIGN
1) Input/Output relationship
a)
A simple block
Combinational circuit
A circuit with less than 5 inputs
Obtain a truth table
Obtain circuit expressions
Sequential circuit
A
circuit with less than 5 FFs
 Obtain a state diagram
 Obtain circuit expressions
Move on to the Implementation step, (2)
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 38

Designing a New PCB

PCB
DESIGN
1) Input/Output relationship
b) A complex block
Obtain the operation table/diagram
Try to implement it (Step 2)
If impossible, partition the block based on
 Application (major operations) : a subblock
for each major operation
 Design goals : speed, cost, power, size,…
► Speed, cost, power,… depend on the technology
 Available components : components of the
technology
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 39
Designing a New PCB


PCB
DESIGN
2) Implement each circuit
i.
One or more high density (ULSI/VLSI/LSI/MSI) chips
implement the circuit ? A few SSI chips with gates and FFs
here and there ?
• If yes, draw the schematic and move to the TEST step
ii. One or more Programmable (PLA/PAL/GAL/ROM/FPGA) chips
implement the circuit ? A few SSI chips with gates and FFs
here and there ?
 If yes, draw the schematic, program the chips and move
to the TEST step
iii. Simple enough to be designed quickly using Switching Theory
(less than 5 inputs or less than 5 FFs) so a few SSI chips
with gates and/orFFs needed ?
• If yes, draw the schematic and move to the TEST step
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 40
Designing a New PCB


DESIGN
PCB
2) Implement each circuit
iv. The circuit can be designed as a new chip ?
• A risky process since we are designing a PCB
• Time can be saved by licensing portions of the chip
• If yes, borrow it, place it, design the chip and move
to the TEST step
v. If no to all the above questions, go back to step 1(b) to
partition it further or repartition one level up, two
levels up,,, or, all the way up
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 41
Designing a New PCB


TEST

Test (sub)blocks separately


PCB
Functional and timing simulations by applying test
vectors
• Pick the right test vectors and the right order of
them
• Note down these combinations and output values to
use them during later testing steps
Combine (sub)blocks one at a time
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 42

Designing a New PCB
 MODIFY
PCB
A simple change
 Optimize the circuit after you think your circuit
does not have logic and timing errors

 After the optimization, test the circuit to make sure the
optimization does not introduce logic and timing errors
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 43

Developing a new chip
1) Development Cycle on Computers
Major error : Redesign
DESIGN
TEST
MODIFY
Major error : Redesign or terminate the project due to TTM
2) Development Cycle with FPGA chips
Mount
Test
Modify
Major error : Redesign or terminate the project due to TTM
3) Development Cycle on prototype chip
Fabricate
Test
Which components and how many ?
TEST : applying input combinations,
test vectors, and simulating
During testing If you see MODIFYING
hardware to optimize it is possible, do
that after you correct logic and timing
errors. Then, test again to see if your
minimization has logic/timing errors
Mount : FPGAs are mounted on
bread/boards, wired and programmed
Test : apply test vectors to FPGAs
Modify : either FPGA mounting/wiring
is changed or a simple design change is
made on computers, simulated, then
FPGAs are programmed and tested
Fabricate chip by sending a GDSII file
to a fabrication facility : tape out
Apply test vectors to the chip
Chip
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 44

Developing a digital product
 A new chip

Which gates/FFs and how many is determined by
 The application (major operations)
 Available components of the technology chosen
 Besides speed, cost, power, etc. : product goals
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 45

Development Cycle on Computers

Chip
DESIGN
1) Input/Output Relationship
a) A simple circuit
Obtain the truth table of the combinational circuit with
less than 5 inputs then move on to Implementation (2)
Obtain the state diagram of the sequential circuit with
less than 5 FFs then move on to Implementation (2)
b) A complex circuit

Obtain the operation table or the operation diagram
►Try to implement it in (2)

If it cannot be implemented immediately in (2)
► Partition it


2) Implementation Try to use registers, counters,
TEST
MODIFY
shift registers even if it is a
simple sequential circuit
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 46

Development Cycle on Computers
Chip
For the current (sub)block we can get a truth table or a state diagram ?
Step I (a)
N
Y
Obtain an operation table or an operation diagram
Step I (b)
Current (sub)block is implementable ?
Step II
Y
Implement the current (sub)block
Step II
N
Partition it into (sub)blocks
Step I (b)
Works and satisfies design goals ?
TEST
Y
Any other (sub)block
to implement ?
Y
N
A simple design change
MODIFY
N
Design complete
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 47

Designing a Complex Block
 Partition it into pieces based on major operations

Besides the design goals and the technology
 One block for each major operation

These major operations are often
 Additions, MUXings, comparisons, decodings, encodings,
DeMuxing, registering, counting, etc.

These operations are already implemented by available
components/chips :
 ADDers, Multiplexers, Comparators, Decoders, Encoders,
DeMuxes, Registers, Counters, shift registers, etc.

This happens frequently for real-life applications
Chip
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 48

An Unusual Major Operation (an unusual block)
 Trying to implement a block

If it has < 11 inputs implement it by using programmable
components
 Memory components
• ROMs, RAMs

Chip
Otherwise (complex or too many inputs)
 Break it up or
 Repartition one level up, or
• Two levels up, or,…
• All the way up (redesign !?)
 Eventually, the resulting operations will be additions,
comparisons, multiplexing, decoding, etc.
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 49
Designing a New Chip


Chip
DESIGN
1) Input/Output relationship
a)
A simple block
Combinational circuit
A circuit with less than 5 inputs
Obtain a truth table
Obtain circuit expressions
Sequential circuit
A
circuit with less than 5 FFs
 Obtain a state diagram
 Obtain circuit expressions
Move on to the Implementation step, (2)
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 50

Designing a New Chip

Chip
DESIGN
1) Input/Output relationship
b) A complex block
Obtain the operation table/diagram
Try to implement it (Step 2)
If impossible, partition the block based on
 Application (major operations) : a subblock
for each major operation
 Design goals : speed, cost, power, size,…
► Speed, cost, power,… depend on the technology
 Available components : components of the
technology
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 51
Designing a New Chip


Chip
DESIGN
2) Implement each circuit
i.
CS2204
One or more Xilinx Design Blocks, XDBs or Xilinx nonprogrammable macros (not gates and FFs) implement
the circuit ? A few gates and FFs here and there ?
• If yes, draw the schematic and move to the TEST
step
ii. One or more Programmable Xilinx macros implement
the circuit ? A few gates and FFs here and there ?
 If yes, draw the schematic, program the macros
and move to the TEST step
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 52
Designing a New Chip


Chip
DESIGN
2) Implement each circuit
CS2204
iii. Simple enough to be designed quickly using Switching
Theory (less than 5 inputs or less than 5 FFs) so a few
gates and/or FFs needed ?
• If yes, draw the schematic and move to the TEST
step
iv. The circuit can be licensed ?
• If yes, borrow it, place it and move to the TEST
step
v. If no to all the above questions, go back to step 1(b) to
partition it further or repartition one level up, two
levels up,,, or, all the way up
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 53
Designing a New Chip


Chip
TEST

Test (sub)blocks separately


Functional and timing simulations by applying test
vectors
• Pick the right test vectors and the right order of
them
• Note down these combinations and output values to
use them during later testing steps
Combine (sub)blocks one at a time
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 54

Designing a New Chip
Chip
 MODIFY
A simple change
 Optimize the circuit after you think your circuit
does not have logic and timing errors

 After the optimization, test the circuit to make sure the
optimization does not introduce logic and timing errors
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 55

Example block partitioning
Macro 2, M2
in Block 6
How can we
design
Macro 2, M2
in Block 6 ?
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 56

Designing Macro 2 in Block 6
 Macro 2 has 16 inputs, 2 outputs and is also
combinational
DISP
16
Macro 2
LRGDISPPOS
2
 Input/output relationship : It outputs the position
number of the rightmost largest display
C 2 7 1  11
3
2 1 0
9 9 3 5  10
3
2 1 0
A F 4 F  00
3
2 1 0
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 57

Designing Macro 2 in Block 6
 How can we design Macro 2 ?

Try to implement it : Any high-density Xilinx
non/programmable circuit that implements it ?
 NO ! It is an unusual operation !
• Then, we have to partition it based on its major operations
 Compare displays
 Generate the number of the right largest display
DISP
16
Compare Generate
Displays Number
LRGDISPPOS
2
 Compare Displays has 16 inputs and determines the largest display
 Any Xilinx high density non/programmable circuit that implements it ?
 NO ! It is an unusual operation !
 We need to partition it
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 58

Designing Macro 2 in Block 6
 How can we partition Compare Displays ?


Compare two sets of displays and select 2 larger displays
Compare the 2 larger displays
A
PD0
4
PD1
4
PD2
4
PD3
4

B
A
B
Compare
Displays 0 & 1 &
Select
Compare
Displays 2 & 3 &
Select
A<B
A
4
B
Compare
Larger
Displays
A<B
4
A<B
Compare Displays 0 & 1 &Select has 8 inputs and compares and
selects the larger of two displays
 Any Xilinx high density non/programmable circuit that implements it ?
• NO ! It is an unusual operation !
 We need to partition it
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 59

Designing Macro 2 in Block 6
 How can we partition Compare Displays 0 & 1 & Select?
Compare the two displays
Select the larger one


A
PD0
4
PD1
4
B
Compare
Displays
0&1
A<B
PD0
PD1

0
4
1
4
Select
Select
Larger
Display
4
Compare Displays 0 & 1 has 8 inputs and compares two displays
 Any Xilinx high density non/programmable circuit that implements it ?
•
YES ! It is a 4-bit unsigned binary compare operation !
 We use a Xilinx 4-bit Unsigned Binary Comparator : COMPM4

Select Larger Display has 9 inputs and selects one of two 4-bit inputs
 Any Xilinx high density non/programmable circuit that implements it ?
•
YES ! It is a 4-bit 2-to-1 multiplexing operation !
 We use a Xilinx 2-to-1 MUX : X74_157
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 60

Designing Macro 2 in Block 6
 How can we design Generate Number ?

It has 3 inputs and is combinational
Displays 0 & 1
A<B
Displays 2 & 3
A<B
Compare 2 Larger Displays


A<B
Generate
Number
LRGDISPPOS
2
Input/output relationship : It outputs the number of the
largest display in Unsigned Binary based on the three inputs
Try to implement it : There are three inputs
 One can obtain a truth table and obtain the two minimal SOP
expressions !
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 61
Do not leave the lab before your partners finish
► Help your partners
QUESTIONS ?
Read slides starting at the end on term project, Project
Manager, schematic design and other related topics
Continue reading the Term Project handout
Think about the machine player strategy
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 62

Today’s Individual Xilinx Work
 We will continue to study (analyze) the term project


We will use discussions in class to implement Macro 1 (M1) of
Block 6 : Experiment 3
We will develop the rightmost largest display circuit of the
Ppm term project in Block 6, based on our classroom
discussion on it : Experiment 4
 We will replace Macro 2 (M2) in Block 6 with our own circuits
 Help your partners complete today’s project
 We will continue reading the Term Project handout

Relate each term project (sub)block in the Term Project
handout to the Ppm schematic
 Study Ppm (sub)blocks by performing simulations
 Read slides at the end to learn more about the term
project, Project Manager, schematic design and
other related topics
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 63

Today’s Individual Xilinx Lab Work
1. Copy the exp2 folder and paste it in the
cs2204 folder as the exp3 folder

We will experiment with the Ppm schematics
2. Open the Ppm project in exp3
3. Look at the six Ppm schematics



If you copy a project completely as we did and then open
its schematics, the schematics will be all Non-Project
Therefore, close all these schematics and close the
schematics window
Then, open the schematics one by one on the Project
Manager window, by double clicking on the schematic name
on the upper left side
4. Place your team info on the schematics on
schematic 1 : ppm1.sch
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 64

Today’s Individual Xilinx Lab Work
5.
6.
7.
8.
Save schematic 1
Switch to schematic 6
Zoom into the lower mid area, containing M1
There is a custom macro designed by the
professor

It has three outputs



Two outputs that indicate the number of the rightmost
zero position : ZERODISP
An output to indicate a display is zero : Aposzero
• If there are no zero displays, it is 0
See ppm6.sch on the next slide
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 65
Today’s Individual Xilinx Lab Work


Ppm Schematic 6
Macro 1
M1
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 66

Today’s Individual Xilinx Lab Work
9. Analyze the macro to determine how it is
used

See the correspondence between the classroom
discussion and M1 inputs and outputs



Determine which input is “a”, which input is “b,” etc.
Determine which output is “Y1” which input is “Y0” and
which output is “z”
Do a Hierarchy Push and notice that its
implementation cannot be shown by Xilinx

A comment on the bottom of the schematic sheet reads
“Symbol is a primitive cell”
10. Perform functional simulations on this macro

Use your truth table that studied in the
classroom
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 67

Today’s Individual Xilinx Lab Work
11. Search for the inputs and outputs of the
macro by clicking on the Query window
button on top of the schematic sheet
In the Signal/Bus mode of the SC Query/Find
window that will pop up
Determine which components generate the inputs

Pos0zero, Pos1zero, Pos2zero, Pos3zero
Determine which components use outputs

ZERODISP1, ZERODISP0 and Aposzero
12. Delete the macro : M1 in schematic 6


Do not delete the wires
Save schematic 6, ppm6.sch

See modified ppm6.sch on the next slide
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 68
Today’s Individual Xilinx Lab Work


Ppm Schematic 6
Macro 1, M1
deleted
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 69

Today’s Individual Xilinx Lab Work
13. Switch to the Human Play Block, Block 3 or ppm3.sch
14. Draw the schematic of the macro on the lower mid side in
schematic 3 by using classroom discussions and your design






You will implement the ZERODISP1, ZERODISP0 and Aposzero
outputs by using two 2-level AND-OR gate networks
You will use the Symbols toolbox button on the leftmost side (or
F3) to get the component list
You will use the Draw wires button on the leftmost side (or F4)
to draw wires
To rotate components right press ctrl-r
To rotate components left, press ctrl-l
Note, wires cannot be rotated


Label the wires (inputs and outputs) based on your analysis in part
(9)


But, by pulling from one end of a wire, it can be rotated !
Label the gates starting at U314
See modified ppm3.sch on next slide
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 70
Today’s Individual Xilinx Lab Work


The modified ppm3.sch
ZERODISP0
ZERODISP1
Aposzero
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 71

Today’s Individual Xilinx Lab Work
14. Draw the schematic of the macro on the
upper right side in schematic 3 by using
classroom discussions and your design
In the Instance mode of the SC Query/Find
window that will pop up
Determine that there is no component labeled U314 and
above
Label the components starting at U314
The last component label is U323
Save schematic 3, ppm3.sch
See modified ppm3.sch on next two slides
First, the ZERODISP circuits
Then, Aposzero circuit
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 72

Today’s Individual Xilinx Lab Work
 The ZERODISP circuits in ppm3.sch
ZERODISP0
ZERODISP1
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 73

Today’s Individual Xilinx Lab Work
 The Aposzero circuit in ppm3.sch
Aposzero
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 74

Today’s Individual Xilinx Lab Work
15. Perform an Integrity Test to check for
errors

Integrity tests do not catch all the errors

That is why after the Integrity tests we have to
perform
• Functional simulations
• Xilinx IMPLEMENTATIONs
• Timing simulations
16. Perform functional simulations on this macro
in schematic 3 to verify that it is working
Use the truth table you have developed
Make sure the circuit is beautified and the
schematic is saved again
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 75

Today’s Individual Xilinx Lab Work
17. Do a Xilinx IMPLEMENTATION

Make sure there are no errors


Make sure the IMPLEMENTATION options are
changed so that a better IMPLEMENTATION is done
Read the Implementation Log File to confirm
that

The number of warnings 26
• These warning are OK, we can continue
• Note that there are 26 warnings not 25 as it was
the case in Experiment 1 since a wire in Block 5 is
not used
WARNING:NgdBuild:454 - logical net '$Net00202_' has no load
•
This wire is the wire that connected the unused
data inputs of the Xilinx 4-bit ADDer to GND in
Block 5
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 76

Today’s Individual Xilinx Lab Work
17. Do a Xilinx IMPLEMENTATION

Read the Implementation Log File to see that

The FPGA chip utilization is 98%
•
The Xilinx IMPLEMENTATION maps the design to 191 to
193 CLBs, hence 97% to 98% utilization, after an
IMPLEMENTATION, a feature peculiar to FPGA testing
 The conversion of the schematic to the bit file is
“randomized” to have a better mapping of the logic to
CLBs, but it leads to this situation
 That is why we fabricate the prototype chip before we mass
produce it to test the design one more time to make sure
the design is correct
 Nevertheless, the utilization is high since two gate networks
implement a full adder and this implementation is worse than
the Xilinx implementation
 That is why it is better that we use Xilinx components if
they are available
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 77

Today’s Individual Xilinx Lab Work
17. Do a Xilinx IMPLEMENTATION

The Project Manager window looks like this after the
IMPLEMENTATION is completed successfully :
Make sure the options
for IMPLEMENTATION
are “High Effort” “50”
and “5”
The checkmark for
IMPLEMENTATION
can be delayed a few
minutes sometimes
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 78

Today’s Individual Xilinx Lab Work
18. Do you think there is a possibility of a glitch
by the full adder circuit ?



If yes, which output(s) would have the glitch ?
Which input combination pairs would generate
the glitch ?
Observe the glitch and show it to the TA
19. Download the Ppm project to the FPGA chip
and play the game and to verify that the
schematic works correctly
If it does not work, inspect your circuit in Block
3 and correct your circuit
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 79

Today’s Individual Xilinx Lab Work
1. Copy the exp3 folder and paste it in the
cs2204 folder as the exp4 folder

We will experiment with the Ppm schematics
2. Open the Ppm project in exp4
3. Look at the six Ppm schematics



If you copy a project completely as we did and then open
its schematics, the schematics will be all Non-Project
Therefore, close all these schematics and close the
schematics window
Then, open the schematics one by one on the Project
Manager window, by double clicking on the schematic name
on the upper left side
4. Place your team info on the schematics on
schematic 1 : ppm1.sch
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 80

Today’s Individual Xilinx Lab Work
5. Save schematic 1
6. Switch to schematic 6
7. Zoom into the upper right area, containing
M2
8. There is a custom macro designed by the
professor


It has two outputs that indicate the number of
the rightmost largest display position
See ppm6.sch on the next slide
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 81
Today’s Individual Xilinx Lab Work


Ppm Schematic 6
Macro 2
M2
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 82

Today’s Individual Xilinx Lab Work
9. Analyze the macro to determine how it is
used

See the correspondence between the classroom
discussion and M2 inputs and outputs



Determine which inputs are for display 0, which inputs
are for display 1, etc.
Determine which output is “Y1” and which output is “Y0”
Do a Hierarchy Push and notice that its
implementation cannot be shown by Xilinx

A comment on the bottom of the schematic sheet reads
“Symbol is a primitive cell”
10. Perform functional simulations on this macro

Use also your notes that cover the discussion in
the classroom
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 83

Today’s Individual Xilinx Lab Work
11. Search for the inputs and outputs of the
macro by clicking on the Query window
button on top of the schematic sheet
In the Signal/Bus mode of the SC Query/Find
window that will pop up
Determine which components generate the inputs

DISP0, DISP1, DISP2, DISP3,…, DISP15
Determine which components use outputs

LRGDISPPOS1 and LRGDISPPOS0
12. Delete the macro : M2 in schematic 6


Do not delete the wires
Save schematic 6, ppm6.sch

See modified ppm6.sch on the next slide
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 84
Today’s Individual Xilinx Lab Work


Ppm Schematic 6
Macro 2,
M2
deleted
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 85

Today’s Individual Xilinx Lab Work
13. Switch to the Human Play Block, Block 3 or ppm3.sch
14. Draw the schematic of the macro on the lower right side in
schematic 3 by using classroom discussions and your design

You will implement the LRGDISPPOS1 and LRGDISPPOS0 outputs by
using as many Xilinx design blocks as possible and as few gates as
possible







Note that what is discussed in the presentation must be followd
where we try to use as many available components as possible
 M2 uses three Xilinx comparators, two Xilinx multiplexers and a
few gates
 Note that what is learned in designing M1 can be used for M2
You will use the Symbols toolbox button on the leftmost side (or
F3) to get the component list
You will use the Draw wires button on the leftmost side (or F4)
to draw wires
To rotate components right press ctrl-r
To rotate components left, press ctrl-l
Note, wires cannot be rotated

But, by pulling from one end of a wire, it can be rotated !

Label the gates starting at U324
Label the wires (inputs and outputs) based on your analysis in part
(9)
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 86

Today’s Individual Xilinx Lab Work
14. Draw the schematic of the macro on the
lower right side in schematic 3 by using
classroom discussions and your design
In the Instance mode of the SC Query/Find
window that will pop up
Determine that there is no component labeled U324
and above
Label the components starting at U324
Save schematic 6, ppm6.sch
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 87

Today’s Individual Xilinx Lab Work
15. Perform an Integrity Test to check for
errors

Integrity tests do not catch all the errors

That is why after the Integrity tests we have to
perform
• Functional simulations
• Xilinx IMPLEMENTATIONs
• Timing simulations
16. Perform functional simulations on this macro
in schematic 3 to verify that it is working
Make sure the circuit is beautified and the
schematic is saved again
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 88

Today’s Individual Xilinx Lab Work
17. Do a Xilinx IMPLEMENTATION

Make sure there are no errors


Make sure the IMPLEMENTATION options are
changed so that a better IMPLEMENTATION is done
Read the Implementation Log File to confirm
that

The number of warnings 26
• These warning are OK, we can continue
• Note that there are 26 warnings not 25 as it was
the case in the previous experiment since a wire in
Block 5 is not used
WARNING:NgdBuild:454 - logical net '$Net00202_' has no load
•
This wire is the wire that connected the unused
data inputs of the Xilinx 4-bit ADDer to GND in
Block 5
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 89

Today’s Individual Xilinx Lab Work
17. Do a Xilinx IMPLEMENTATION

Read the Implementation Log File to see that

The FPGA chip utilization is 98%
•
The Xilinx IMPLEMENTATION maps the design to 191 to
193 CLBs, hence 97% to 98% utilization, after an
IMPLEMENTATION, a feature peculiar to FPGA testing
 The conversion of the schematic to the bit file is
“randomized” to have a better mapping of the logic to
CLBs, but it leads to this situation
 That is why we fabricate the prototype chip before we mass
produce it to test the design one more time to make sure
the design is correct
 Nevertheless, the utilization is high since two gate networks
implement a full adder and this implementation is worse than
the Xilinx implementation
 That is why it is better that we use Xilinx components if
they are available
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 90

Today’s Individual Xilinx Lab Work
17. Do a Xilinx IMPLEMENTATION

The Project Manager window looks like this after the
IMPLEMENTATION is completed successfully :
Make sure the options
for IMPLEMENTATION
are “High Effort” “50”
and “5”
The checkmark for
IMPLEMENTATION
can be delayed a few
minutes sometimes
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 91

Today’s Individual Xilinx Lab Work
18. Do you think there is a possibility of a glitch
by the full adder circuit ?



If yes, which output(s) would have the glitch ?
Which input combination pairs would generate
the glitch ?
Observe the glitch and show it to the TA
19. Download the Ppm project to the FPGA chip
and play the game and to verify that the
schematic works correctly
If it does not work, inspect your circuit in Block
3 and correct your circuit
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 92

Today’s Individual Xilinx Lab Work
20. Help your partners complete today’s project
21. Continue reading the Term Project handout

Relate each term project (sub)block in the Term Project
handout to the Ppm schematic


Study Ppm (sub)blocks by performing simulations
Play the other two versions of the term project to
refresh your memory
•
•
Ppm human vs. human : ppmhvsh
Ppm machine vs. machine : ppmmvsm
22. Read slides at the end to learn more about the
term project, Project Manager, schematic design
and other related topics
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 93

Understand Critical Wires
RD : 4 bits
 The random digit
R1D : 4 bits
Next random digit
R2D : 4 bits
The random digit after next random digit
P1add, TRD : 4 bits
Altogether they form a random digit manually input to the machine player to
test it
In order to input it, one of SW1 – SW4 must be 1
DISP : 16 bits
 They represent the four position displays
 In Hex
 DISP15-DISP12 : the leftmost position display, PD3
 DISP11-DISP8 : position display PD2, etc
TDISP : 16 bits
Next display bits after the current random digit is played
SELTPD : 4 bits
Selects between DISP and TDISP to add the current or next random digit
If it is 0, it selects DISP, otherwise TDISP
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 94

Understand Critical Wires
TADDDISP : 16 bits
The result of selection between DISP and TDISP
NPDISP : 16 bits
 TADDDISP digits plus RD
NDISP : 16 bits
 New DISP bits
 In Hex
BRWD : 4 bits
 Basic reward
 In Hex
 The digit played and also minimum points earned
Brwdeqz : 1 bit
 BRWD is zero when it is 1
PDPRD : 4 bits
 Display overflow bits after addition
Pdprd : 1 bit
The display overflow bit of the position played
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 95

Understand Critical Wires
Selplyr : 1 bit
 The current player
 If it is 0, it is the human player, otherwise, it is the machine
player
P1SEL : 4 bits
 The position played by the human player
P2SEL : 4 bits
 The position played by the machine player
PSEL : 4 bits
 Position Select bits of current player
ENCPSEL : 2 bits
 The number of the position played
EQ : 4 bits
 The equality of the four displays to the digit played
NSD : 2 bits
 The number of similar digits, i.e. the adjacency information
of the position played
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 96

Understand Critical Wires
RWD : 8 bits
 The reward points calculated based on adjacencies
 In Unsigned Binary
P1PT : 8 bits
 Player 1 points
 In Hex
P2PT : 8 bits
 Player 2 points
 In Hex
PT : 8 bits
 The points of the current player
 In Hex
NPT : 8 bits
 New player points for the current player
 In Hex
Ptovf : 1 bit
The points overflow
 if it is 1, the new player points is above (255)10
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 97

Understand Critical Wires
P1add : 1 bit
 Player 1 adds when it is 1
P2add : 1 bit
 Player 2 adds when it is 1
Add : 1 bit
 The current player adds when it is 1
P1skip : 1 bit
 Player 1 skips when it is 1
P2skip : 1 bit
 Player 2 skips when it is 1
P1played : 1 bit
 Player 1 played when it is 1
P2played : 1 bit
 Player 2 played when it is 1
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 98

Understand Critical Wires
Clear : 1 bit
 Clear FFs, registers, counters, etc. during reset in Block 2 and Block 4 so
that it can play again
Clearp2ffs : 1 bit
 Clears Player 2 FFs, counters and registers
Shp1rds : 1 bit
Shows next two digits to Player 1 in state 1
Add : 1 bit
Shows that the current player has selected to add
Stp1pt : 1 bit
 Store Player 1 points
Stp2pt : 1 bit
 Store Player 2 points
Grd : 1 bit
 Signals to generate a new random digit
 The random digit counter output is stored as P2RD while P2RD and P1RD are
shifted to generate the new P1RD and RD
Bpds : 1 bit
Blink one or all displays slowly
Bpdf : 1 bit
Blocks a display fast after a display overflow
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 99

Understand Critical Wires
Clff : 1 bit
 Clears FFs in Block 2 so that the next player can play if
there is no overflow
S1 : 1 bit
 State 1 where when it is 1, the Ppm is in state 1
P2sturn : 1 bit
 Signals that Player 2 has the turn
 It is 1 when the Ppm is in state 4
Sysclk : 1 bit
 System clock of the operation diagram at 6 Hz to the digit
played
 P2clk : 1 bit
 The clock signal of Player 2 at 48 Hz
 Rdclk : 1 bit

The random digit counter clock at 192 Hz
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 100

Project Manager Actions and Reminders
 Make sure there is a CS2204 folder
 Make sure there is an experiment folder for
the current experiment

You can check the folder the current project is in
by selecting File -> Project Info
 Make sure the FPGA chip and its model are
correct when a new Xilinx project is created

You can check the FPGA chip and its model by
selecting File -> Project Type…
 The selections must be as follows
• The chip : Spartan
• The model : S10PC84
• Speed : 3
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 101

Project Manager Actions and Reminders
 If you copy a project completely and paste it as a
new project, its schematic files cannot be worked on
right away




After you open the schematics, they are all Non-Project
schematics
Close all the schematics
Close the schematics window
Open the schematics one by one on the Project Manager
window
 Double click on the schematic name on the upper left side for
each schematic file
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 102

Project Manager Actions and Reminders
 When you do the first Xilinx
IMPLEMENTATION or after clearing the
implementation data, you need to change
implementation options before clicking on
“Run” in the Implement Design Window

You can change the options by selecting Options…
in the same window and then
 Increase the Place & Route Level to the Highest Effort
on the “Options” window
 Click on the Edit Options… button for Implementation: in
the Program Options area of the “Options” window
 Click on Place and Route on the “Spartan Implementation
Options: Default” window
 Increase Router Options to 50 and 5 for both Routing
Passes and Delay-Based Cleanup Passes
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 103

Project Manager Actions and Reminders
 After a successful IMPLEMENTATION




The schematic files have a check mark next to them
The Design Entry button will have a check mark
The IMPLEMENTATION button has a check mark (after a
delay of minutes sometimes)
The PROGRAMMING button is highlighted
 If not, just click in anywhere in the Flow tab area of the
Project Manager window, it will be highlighted
 If the IMPLEMENTATION is not successful due to
errors, the IMPLEMENTATION button will have an
“X” mark

The error can be because of wrong chip selection or
schematic design errors
 Correct them then !
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 104

Project Manager Actions and Reminders
 After a Xilinx IMPLEMENTATION, read the
Implementation Log File for errors, warnings and
FPGA chip utilization

You can read the Implementation Log File by selecting
Reports -> Implementation Log File
 All No driver warnings must be corrected
• No Driver means, the wire is not connected to any
component output
 All Multiple drivers warnings must be corrected
• Multiple Drivers means, a wire is connected to multiple
component outputs
 Most No Load warnings can be ignored
• Because, the software warns that a component output is
not used, because you do not need the output
• But, if a component output is needed, and not connected,
then it is an error, the output must be connected to the
input of a component
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 105

Project Manager Actions and Reminders
 After performing several Xilinx IMPLEMENTATIONs, clear the
implementation data, by selecting Project -> Clear
Implementation Data

Back to back Xilinx IMPLEMENTATIONs use previous
implementation data that is unchanged to save time
 Over time, this implementation data becomes corrupt and the bit file
has errors
• Correct designs do not perform correctly on the FPGA board

Clearing the implementation data changes the implementation
options to the default ones





The schematic files will keep their check marks
The Design Entry button will keep its check mark
But, the IMPLEMENTATION button will have a question mark
The PROGRAMMING button will not be highlighted
The implementation options must be changed to the required ones again
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 106

Schematic Design Actions, Shortcuts &
Reminders
 Place team info on schematics

You can enter the team info by selecting File -> Table
Setup…
 Place your name & a partner name on Line1:
 Place names of the other two partners on Line 2:
 On Line3: place CS2204 – Section A/B/C/D/E/F – Spring 2007
 Press F2 to enter the Select & Drag Mode

Only, in this mode components can be deleted, rotated,
copied and pasted
 You can press ESC to enter the Select & Drag Mode
 Press F3 to get component library on screen



VCC is logic 1
GND is logic 0
To quickly locate a component, enter the first few letters of
the component in the bottom area of the SC Symbols window
 To locate XOR gates, just enter letter “X” and “O”
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 107

Schematic Design Actions, Shortcuts &
Reminders
 Press F4 to draw wires
 Press F5 to draw buses
 Press F7 to search for wires and components

To search for wires, select the Signal/Bus mode
 If the wire does not have a name, the software assigns one
that starts with a “$” symbol and ends with a “_” symbol
• Use the whole name to search for a wire

To search for a component, select the Instance mode
 If a component does not have a name, the software assigns one
that starts with “$I” symbols followed by a number
• Use the whole name to search for the component
 Press F8 to start simulation quickly
 Press F10 to refresh the screen
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 108

Schematic Design Actions, Shortcuts &
Reminders
 Press ctrl-c to copy a wire or a component selected


When components are copied, their labels are not copied !
You can copy from a schematic that belongs to another
project
 To open the schematic of another project, click on
button
in the upper left corner, then select the schematic file which
will be in another folder
 Press ctrl-v to paste a wire or a component
 Press ctrl-r/ctrl-l to rotate components right/left

Wires cannot be rotated !
 You can see how a Xilinx macro is designed (the
internal structure), do a Hierarchy Push, by selecting
Hierarchy -> Hierarchy Push
 You can close the macro internal design screen, by
selecting Hierarchy -> Hierarchy Pop
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 109

Schematic Design Actions, Shortcuts & Reminders
 Unless otherwise stated, use Xilinx macros instead of designing
them to save time
 Use buffers to rename wires
 Do not use unnecessary input/output buffers
 Do not use unnecessary input/output pads
 If you copy and paste components, their labels are not copied
and pasted by the software

You will need to “source” the schematic file to copy and paste
component labels as explained in the Advanced Xilinx and Digilent
Features handout
 Xilinx does not have high density ROM memory components

16x1-bit and 32x1-bit
 They may not be used at all
• If needed, its usage is described on page 9 of the Advanced
Xilinx and Digilent Features handout
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 110

Schematic Design Actions, Shortcuts &
Reminders
 Drawing buses by using Draw Buses button on the left
side :


Ppm buses are type None
Individual wires of a bus must have names the same as the
bus name
 The indices of individual wires start at 0 and are up to the
number of bus wires minus 1
• Bus NPT has 8 wires : NPT7, NPT6, NPT5,…, NPT1, NPT0

If a component generates a bus, there is no need to draw
the individual wires of the bus, unless a components needs
those individual wires
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 111

Schematic Design Actions, Shortcuts &
Reminders
 Beautify the schematic for documentation purposes

Place components of different sub/blocks separate from
each other to recognize them
 Write Comments, draw lines and rectangles and label
sub/blocks to identify them on the schematic for
documentation purposes
• Use the Graphics Toolbox button on the left :

Label components appropriately
 Wire names follow application and block partitioning naming
requirements
• Except for wires that are connected IBUFs, OBUFs, IPADs and
OPADs
 Component names start with a U
• Except if it is a BUF, IBUF, OBUF, IPAD or OPAD
 To label a component, right click on the component and select
Symbol Properties…
• Give the name in the Reference: section of the Symbol Properties
window
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 112

Schematic Design Actions, Shortcuts &
Reminders
 Beautify the schematic for documentation purposes


Do not leave components unused
Draw short wires and label them with the same name
 To label wires double click on the wire and enter the name in
the Net Name: area of the pop up window






Draw wires without unnecessary turn
Draw wires without tangling
Draw wires around components/labels/names
Do not short circuit input lines
Do not short circuit output lines
Do not have labels/attributes/components overlap
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 113

Schematic Design Actions, Shortcuts &
Reminders
 Perform integrity tests to catch simple
errors

You can do an integrity test of the current
schematic sheet, by selecting Options -> Integrity
Test for Current Sheet
 After the completion, a window may tell you to look at
the Project Manager window to read about warnings
detected, even if it says the test passed successfully
• Look at the Project Manager window, you will see warnings
in blue
• If the last line has the Schematic Contents OK line, there
is no need to correct anything
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 114

Schematic Design Actions, Shortcuts &
Reminders
 Perform logic simulations to catch logic errors

Press F8 to start simulation quickly
 You will see the SC Probes window
 To select the input wires to be simulated, click on the
Stimulator tool button of the SC Probes windows
 Then click on the input wires by precisely clicking on their
names to select them
• There will be a square gray box shown on the left side of the
input wire name
• Wires that have no name cannot be simulated, therefore, they
must be given names for simulation
• When selecting input bus wires, click on the bus wires in the
increasing index order : ABUS0, ABUS1, ABUS2,…
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 115

Schematic Design Actions, Shortcuts &
Reminders
 Perform logic simulations to catch logic errors

Press F8 to start simulation quickly
 You will see the SC Probes window :
 To select the output wires to be simulated, click on the Probe
tool button of the SC Probes windows :
 Then click on the output wires by precisely clicking on
their names to select them
• There will be a square gray box shown on the left side of
the output wire name
• Wires that have no name cannot be simulated, therefore,
they must be given names for simulation
• When selecting output bus wires, click on the bus wires in
the increasing index order : OBUS0, OBUS1, OBUS2,…
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 116

Schematic Design Actions, Shortcuts &
Reminders
 Perform logic simulations to catch logic errors

Press F8 to start simulation quickly
 You will see the SC Probes window :


To start the simulation, click on the Simulator button of the
SC Probes window :
Once you have the simulation window on the screen
 You will see the input wires listed and then the output wires on
the left side of the Logic Simulator window
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 117

Schematic Design Actions, Shortcuts &
Reminders
 Perform logic simulations to catch logic errors

Separate the input rows from the output rows by placing a
blank row between the input and output wires sets
 Click on the top output wire
 Make selections Signal -> Empty Rows -> Insert

Combine bus bits to reduce the number of rows
 Click on the top bus wire which has the lowest index (ABUS0)
 Press shift and simultaneously click on the highest order bus
wire (ABUS7) to select all the wires of the bus
• A turquoise rectangle covers the bus wires
 Right click on the turquoise rectangle and make the following
selections Bus -> Combine
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 118

Schematic Design Actions, Shortcuts &
Reminders
 Perform logic simulations to catch logic errors

In order to simulate the circuit, the input wires must be
first given new names
 Click on the Select Stimulators button :
• A keypad window will be shown
 Select an input wire by clicking on it (it will be covered by a
turquoise rectangle) and then click on any letter key on the
keypad, such as “q”
• To the right of the input wire, the new name “q” is shown
• To the right of “q”, the current value of the wire is shown
►
If it is a single wire, the value is Hi-Z
◊ This has to be changed to have correct simulations
► If it is a bus, the value is shown as capital letter “Z”
◊ This has to be changed as well for correct simulations
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 119

Schematic Design Actions, Shortcuts & Reminders
 Perform logic simulations to catch logic errors

To change the values of wires on the simulator window
 If it is a single wire, the value is Hi-Z :
• Just click on the Hi-Z line to make the value 0
►The value is shown to the right of name “q” as 0
• Click on the 0 value line again to make the value 1
►The value is shown to the right of name “q” as 1
 If it is a bus, the value is shown as capital letter “Z”
• Click on Logical States to give a value to the bus :
►The Stimulator State Selection window will be shown
• Click on the bus name, such as ABUS
• Enter an appropriate Hex value in the Bus State area, such as “FA”
► Appropriate means the Hex value must fit the width of the
bus : “FA” implies, the bus has at least eight wires
• Click on the Bus button of the Stimulator State Selection window :
►The value assigned is shown to the right of name “q” as “FA”
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 120

Schematic Design Actions, Shortcuts & Reminders
 Perform logic simulations to catch logic errors

To change the values of wires on the simulator window
 To have a clock signal as an input follow the steps below :
• Make sure the input signal is not renamed as “q”, “w” etc.
• Click on the input signal to select it
• Click on the Select Stimulators button :
• Click on Formula…
• Double click on C1: under Clocks
• Enter the following in the Edit Formula area :
• 100ns=H 100ns=L
► This means a periodic signal which is 100 ns 1 and 100 ns 0 is generated
► The periodic signal has a period of 200ns or a frequency of 5MHz
• Click Accept
• Click Close
• You will see the C1 button on the Select Stimulators window
highlighted
• Click on C1 so that the input signal is renamed C1
• Click on the Simulation Step button several times :
• You will see the periodic signal automatically generated and the
output values in response to that
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4 Page 121

Schematic Design Actions, Shortcuts &
Reminders
 Perform logic simulations to catch logic errors

Start simulating the circuit for different input combinations
 If the circuit has 4 or less inputs, then simulate the circuit for
all input combinations (test vectors)
• 16 or less number of input combinations (test vectors)
 If the circuit has more than 4 inputs, select a number of input
combinations (test vectors) then simulate the circuit for these
test vectors
• Which test vectors to choose is a very important task !


To simulate the circuit, click on the Simulation Step button
several times :
Observe the outputs
 If they are correct, try another input combination
 If wrong, return to the schematic and try to figure out why it
is wrong !
 If an output value is Hi-Z or Unknown, there is an error,
correct it
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 122

Schematic Design Actions, Shortcuts &
Reminders

Printing schematics
1) Double click on the Printer227 icon on your desktop and
wait about a minute to allow it to affect the printing
option
2) Zoom into an area of the schematic to print the area
3) Select File -> Print on the schematic window
4) Change the option to Current View Only on the Print
window
5) Click on Setup on the Print Window
6) Change the printer to HP Printer 8150 in Room 227
7) Click on Options to select Landscape printing if
necessary
8) Click OK as many times as needed to print the page
9) Print one copy of each area and then make copies of the
printed schematics for your partners
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 123


What to do if the testing on the board gives wrong
results even thought the design is correct ?
If the design is absolutely correct, here are the steps to
follow in sequence :
1)
2)
3)
4)
5)
6)
7)
8)
The FPGA board is turned on ?
SW9 is in the PROG position ?
The Bitronics Data Switch selects your PC ?
The FPGA type and model are correct ?
The implementation options are changed ?
There are not too many levels of folders to reach the project on
the PC ?
Clear the implementation data, close the software, restart the
software and do a new Xilinx IMPLEMENTATION

Does it work now ?

Delete the project, recreate the project, copy the schematic design
from the saved schematic file
Save the schematic file worked on in a separate folder

•
Does it work ?
•
Does it work ?
Download the zipped project from the course web site, unzip it, copy
the schematic design from the saved schematic file
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 124

What to do if the testing on the board gives
wrong results even thought the design is
correct ?
9) Repeat step 7, by using your partner’s working
schematic
10) Login to another PC and try steps 5 - 8
11) Ask from the TA to help you
a) The TA will login to your original PC and try steps 5 – 8
by using your schematic design and his/her S drive
b) The TA will login to another PC and try steps 5 – 8 by
using your schematic design and his/her S drive on the
new PC
c) The TA will inform the professor
12)If the project works on the second PC, inform
the lab supervisor, Mr. Keni Yip that the original
PC has a problem
CS 2204 Spring 2008 Lab 7 Experiment 3 - 4Page 125