Seminar at UoCambridge Dept of Materials Science and Metallurgy 3mar15. Today, we are making the most complex machines that the world has ever know; tomorrow they will be even more so. We sell them in large volumes to customers who value their functionality but have no recognition of the technology that enables them; even less, our roles in their creation. Technology products don't grow on trees, they are painstakingly designed and constructed one molecule at a time by man(kind). And the more complex the machines the ever more complex they will have been to design and construct. Alas, out of sight is out of mind; so as our technology becomes less visible to the user, then their support for its evolution will diminish! To correct this we must educate the public; but before we can do that we must understand what we do ourselves ... This talk could just be the start of that process.
16. 16
! Need:&A$Mechanism$for$enhancing$human$memory$(Camera)&
! Technology:&...$
! Digital(Logic((CPU+I.O.)(
! Sokware(
! Memory((NV(and(RAM)(
! Excellent(Lenses(and(Displays(
! Analogue(Electronics((Network(&(GPS)(
! Sensors(and(Transducers((CCD(&(MEM)(
! Precision(Mechanics(
! MicroMMotors(
! BaNeries(and(Energy(Storage(
! LEDs(and(Discharge(Tubes(
! Precision(forming(of(Plas+cs(and(Metal(
! Electronic(Packaging(
! Robo+c(Assembly(...(Technology(External(to(the(Product!(
...((All(of(these(Technologies(are(available(to((
21c(Businesses(today(...(
2014$6$Canon$EOS$5D$$$(Today)$
Incorporating DIGIC5+ (ARM)
System-Level
Computation
Q: So why couldn’t ARM make a camera
to compete with Canon?
A: Because Canon has Capabilities in
place that we don’t have!
ARM-based
Computer
17. 17
So$CapabiliOes$actually$limit$a$Business's$Product$PossibiliNes...$
...They are also an Entry Barrier to your Competitors
CAPABILITIES
Product
Development
PRODUCT
! Business$needs$to$Know$it$can$make$a$Product$before$they$Start$to$design$it.$
! CapabiliIes(help(to(Manage(the(Risk((There’s(frequently(no(second(chance!)(
! Making$a$Product$is$sNll$Work;$but$based$on$CapabiliNes$it$is$KnownRWork!$
! Unknowns(lead(to(Protracted(Timescales,(Blown(Budgets,(Lost(Opportuni+es(and(Panic!(
! Business$need$and$Appropriate&set&of&CapabiliOes,$before$they$
commences$a$Product&Development&
(...(But(is(the(Process(Linear((as(depicted(here)?(
((((((…(Does(a(given(Set(of(Capabili+es(lead(to(a(Specific(Product?(…!NO!!
19. 19
! A$Capability$is$an$Installed$$‘Unit’$of$$Technology,$such$that$it$may$$
be$relied$upon$for$Mission6CriNcal$Business$purposes.$
! They(are(Installed:(Reliable,(Trusted,(Predictable,(Usable.(
! They(may(Cross(Technology(Boundaries((HW,(SW,(Op+cs,(Mul+MChip,(etc)(
! They(are(NOT(Product(Specific(...(
! Eg:$$Need$Fortran$capability$by$2016.$$Need$to$handle$Logic$Reliability$by10nm$node.$$Need$a$Bus$
architecture$for$use$in$and$between$3D/Chiplet$by$2020.$$Need$10x$system6power$efficiency$
improvement$$by$2020.$$Need$a$JIT$Java$Compiler$by$2016.$
! They(can(be(Technology(Answers(...(
! Eg:$$How$will$our$Partners$handle$the$end$of$Moors$Law?$How$can$we$increase$our$Partners$ProducNvity$
in$our$Product$Offering?$Will$Transistor$Reliability$become$a$serious$issue$at$14nm?$
(...(Capabili+es(are(the(Founda+ons(for(Product(Development(
( (...(And(they(are(established(by(Research((“Finding(out”)(
Capability&6$A$unit$of$Deployable$Science$
A Capability is the Bridge between a Science and its Exploitation
20. 20
! Short$Term:$From$the$KnownRSet&...(
! Refine!what!we!know!(Including(Individual(knowledge,(Team(knowledge,(and(Tools).(
! Acquire!what!others!know!(Recruit,(Develop,(Contract,(Licence,(Acquire,(Espionage,(etc)(
(...(Specific,(internal,(secrecy,(lowMnumber,(highMcost,(confidence(
(
! Longer$Term:$From$the$UnknownRSet$...(
! Targeted!Research!M(Specific,(pointMfocus,(commercial(advantage((CrownMjewels)(
! Partnered!Research!M(Specific,(focused,(indirect(commercial(advantage((ARM(Partners)(
! University/InsItute!Research!M(Generic,(wide(spectrum,(nonMdifferen+a+ng(knowledge(
! ARM$Funded$Research$Projects$
! NaNonal$Support$Programs1$(TSB,$H2020,$Research$Councils,$DARPA,$etc)(
( (...(Generic,(broadMspectrum,(openness,(highMnumber,(lowMcost,(partnering,(sharing,(uncertainty(
$
(...(Guided(by(Road(Map(from(longer(term(Business(Objec+ves$
Research$6$Establishes$CapabiliNes$
... for Development Engineers to use for Product Development
21. 21
Development$–$Exploits$CapabiliNes$to$Deliver$a$Product$
... (Tangible or Intangible) to fulfil a Customer Need and a Business Objective
! Design$is$a$creaNve$role$within$Development$
! Engineers$Deliver&an$effecNve$Technical6Product$soluNon$...$
! MeeIng!FuncIonal(and(NonOFuncIonal(criteria(...(
! Eg:$Quality,$TTM,$Budget,$Reproducibility,$Testability,$Cost,$Power.$$
! Maximising!Product!DifferenIaIon;(Minimising(Product(Risk,(Cost(and(Uncertainty(
! Using(Available!Capability!(Technology,(Methodology,(Tools(and(KnowMHow)(
! Engineers$Innovate$by$pushing$the$bounds$of$Available$CapabiliNes$...$
! Using(the(opportuni+es(presented(by(the(Par+cular(Product(Brief(
! Trading(between(Available(Capabili+es((Eg:(HW,(SW,(Analogue)(
! U+lising(his/her(wider(sphere(of(knowledge,(knowMhow(and(experience(
(...((Not(all(implementa+ons(are(‘equally(good’(
(...(Innova+on(is((&(will(remain)(a(Fundamental(Engineering(role(
36. 36
Software
development
tools
Physical IP – Design of
the building blocks of
the chip
Processor and
Graphics IP – Design
of the brain of the
chip
Power Mgmt
Bluetooth
Cellular Modem
WiFi
SIM
GPS Flash Controller
Touchscreen
& Sensor Hub
Sensor Hub
Camera
Apps Processor
ARM$Technology$
! Advanced$consumer$products$are$incorporaNng$more$and$more$ARM$
technology$–$from$processor$and$mulNmedia$IP$to$souware$
Early software
development on
Virtual Platforms
39. 39
ARM’s$Vision$…$
… Is a world where all electronic products and services are based upon energy
efficient technology from ARM, making life better for everyone