# 75th ECTC Friday Morning Panel

https://www.youtube.com/watch?v=qLC3oFilv_s

[00:00] Theme for this morning is ECTC at 75 celebrating the past innovating for the future.
[00:06] I have four luminaries of the packaging world with me.
[00:09] Together we have more than 200 years of packaging experience sitting on the podium.
[00:19] Our panelists are Ralph Tamala from Georgia Institute of Technology, John Laauo from Uni Micron, Bill Chen from ASSE Group, and Kitty Pearel from Capstand Technologies.
[00:30] I'm doing the introduction right now.
[00:32] We'll show a quick ECTC history video that will give you the facts and then our panelists will present and give you their personal um impressions and then we'll have a Q&A session.
[00:54] The electronic components and technology conference or ectc sponsored by ITLE E
[01:00] Electronics Packaging Society or EPS has a legacy spanning over 70 years shaping the future of micro electronics packaging components and system integration.
[01:12] In the post World War II era, electronics were generally unreliable.
[01:16] To address this, the symposium on the improvement of the quality of electronic components was launched in 1950.
[01:23] It later evolved into the electronic components conference in 1958.
[01:28] It was renamed ECTC as the event expanded in scope following the formation of ITLE E in the 1960s.
[01:37] The unified organization became a co-sponsor.
[01:39] More than half a century later, the conference has grown immensely in popularity and has transformed into a world-renowned, highly anticipated annual event spanning three full days and featuring 41 technical sessions with a total number of approximately 400 technical papers, including five interactive presentation sessions.
[01:58] On Tuesday, the conference featured 12
[02:00] Special invited sessions, a young professionals event, and 16 professional development courses.
[02:07] For its 75th anniversary, ECC introduced a new student engagement program to support workforce development along with a student and startup innovation challenge.
[02:17] The conference's current platinum sponsors Amcor, IBM, and ASSE are pioneers in the field of electronic packaging technology.
[02:26] Let's review the major packaging innovations of the past 75 years where ECTC played a significant role in disseminating the latest technologies.
[02:34] Throughhole technology which was introduced in the 1940s helped standardize the assembly of early electronic products.
[02:40] Dual inline package or DIP introduced in the 1960s featured two parallel pin rows simplifying integrated circuits or IC mounting and dominating electronics for decades.
[02:54] Surface mount technology or SMT from the 1970s enabled direct printed circuit board or PCB placement, increasing
[03:03] density, reducing size, and allowing automation.
[03:05] Flip chip technology refined from the 1960s mounted bare dies upside down for shorter paths and improved heat and speed.
[03:14] Ball area array from the 1990s used underside solder balls for better thermal or electrical performance and high-speed high pin count devices.
[03:25] Multi-chip modules or MCM first introduced on multi-layer ceramic substrates in the 1980s was later advanced to organic laminate substrates in the 1990s enabling more compact and higherformance systems.
[03:38] Chip on board or COV popular in the 1980s bonded IC's directly to PCBs saving space and cost in consumer and LED devices.
[03:47] Package on package or pop from the 2000s stacked IC's to save board space and allow separate memory or processor testing.
[03:59] System and package or SIP also from the
[04:03] 2000s integrated diverse functions into one package for compact smart devices.
[04:09] Wafer level packaging or WLP from the late 1990s packaged at wafer stage to reduce size and cost vital for mobile tech.
[04:20] This technology was further enhanced to fan out WLP or FOWLP in the 2010s, extending connections beyond chip edges for thinner higherformance packaging ideal for mobile and 5G devices.
[04:31] 3D IC's in TSVs gaining ground in the 2010s.
[04:36] Stack dies with vertical links for high bandwidth in memory and AI.
[04:41] Embedded die packaging from the 2010s placed IC's within substrates to boost thermal or electrical performance in power or RF apps.
[04:52] Heterogeneous integration rising in the 2020s combined varied dyes in one package for enhanced AI, edge, and IoT performance.
[05:01] Chiplet architectures popular in the 2020s
[05:04] modularized chips for better yield, flexibility, and scale as seen in the products of AMD, Intel, and AI accelerators.
[05:12] ECTC 2025 will continue to highlight new packaging innovations such as hybrid bonding and wafer scale packaging.
[05:20] The conference remains a key venue for knowledge sharing and collaboration across industry and academia.
[05:26] Along with our generous sponsors, ECTC extends its gratitude to the conference speakers, attendees, technical committee members, exhibiting companies, and PDC instructors for helping make it the world's premier packaging conference.
[05:42] Visit the website www.ectc.net or connect on LinkedIn, YouTube or Instagram.
[05:52] All right.
[06:01] Now, we'll have our speakers.
[06:03] I want to remind you feel free to use the
[06:05] app to put your questions in at any time while the speakers are talking.
[06:09] Um, we'll take questions both from the floor and from the app at the end.
[06:13] Our first speaker is Dr. Ra Tamala who's in invented or pioneered many technologies while at IBM in Georgia Tech including the industry's first LTCC substrate and the industry's first plasma display the industry's first 2.5D package with up to 144 chips that preage today's chip set chiplets the concept of SOP or system on package.
[06:38] Ralph set up the Georgia Tech packaging research center is the largest and most comprehive ensive research education and industry center in the US and created a consortium model for universities to develop manufacturable technologies.
[06:52] He's published over 800 articles, seven textbooks and has 100 patents.
[06:58] He's received more than 50 awards and I could summarize this all by saying for all of his contributions he
[07:06] was named as the father of modern packaging by the ILE E which created a technical field award in his name the E Ralph Tamala electronic packaging award.
[07:16] Ralph,
[07:24] good morning.
[07:27] Do you still remember me?
[07:29] I'm still around.
[07:32] Thank Thank you very much, Pat, for thinking of me, inviting me to be part of this wonderful special panel that comes once every 75 years.
[07:43] I was actually the president of this society when it was 50 years age, but I didn't run anything like that.
[07:52] I wish I did. Okay, every 25 years we should do something like that.
[07:54] So I've been part of this uh conference since 1982 having attended every one of them.
[07:59] So I find working when I worked at IBM developing these technologies
[08:07] this a great place to publish and inform the people about the technologies he developed.
[08:10] I'll talk a little bit more about that little later on.
[08:14] But this conference is a lot more valuable actually for academics.
[08:18] I became academics as as Pat said I left IBM 93 and I started this so-called PRC and so as a faculty and students it's all about publication publication publications
[08:34] so of course it e is the gold standard for publications peer-reviewed journals proceedings
[08:41] okay that everybody reads around the globe so so so clearly that's a one of the most valuable thing okay that uh that I can think of.
[08:49] I also think of this society as a gathering of some of the topnotch packaging technologies ever since 1950 I guess from around the globe.
[09:01] So all the people came here from around the globe and talked to each other and
[09:09] talked a little bit of what they're doing.
[09:10] So but we kept on increasing increasing that number and today as you heard okay yesterday it seems this conference you now being attended by 2500 people unbelievable from around the globe actually most of the people I like to think of most outside from outside US.
[09:23] So what I remember among many many other things is actually intense debates.
[09:36] Okay at IBM was developing the first industry's first smart module.
[09:38] In fact, that's that's a figure there.
[09:40] Okay.
[09:42] And uh so back if you go back and I started this actually with an idea in 1975 and I began to develop that seriously because that's the only way to get the performance.
[09:55] Okay, single chip packaging putting on a board never keeps the performance.
[09:59] I have no choice in a way.
[10:01] Systems companies had no choice even going back then other than integrate integrate integrate onto package.
[10:07] That's where we are today to have D3D is all
[10:11] about integration.
[10:13] But systems company knew even back then semiconductor companies their their job is to sell the chips not necessarily make a system.
[10:20] Therefore they did not grasp the value of integrated packaging.
[10:22] Okay.
[10:25] But now of course we're all in it in a way.
[10:29] So back then there's intense competition between IBM and many companies in US and Japan.
[10:34] Okay.
[10:37] I have a lot of Japanese friends in here.
[10:39] Back then the three three most important companies in Japan are Itachi, Fujitsu and so they all used to come here we used to have evening sessions debating and talking about all the technologies I particularly remember the one of course the one I'm doing is had is based on glass ceramic started the glass glass ceramic copper 60 layers eventually 100 layers with copper pastel TCC but Japanese pursued a different path of mixing glass and aluminina by Pujitsu, a friend of mine from Geneva.
[11:10] Okay.
[11:10] I still remember his name.
[11:13] Then NEC folks took a different route.
[11:15] They put RDL, believe it or not, in 1980s.
[11:17] They put RDL on a 300 square ceramic substrate.
[11:22] Okay.
[11:22] In 1980s and that became product for a long long time.
[11:26] Okay.
[11:30] And um Hachi folks took a different route.
[11:30] So I got to know them very well.
[11:33] Okay.
[11:35] Watari's son from NEC and Kobayashi's son from uh and from Hitachi and I got to know them so well.
[11:41] I traveled to Japan probably 100 times since then and that they became some of my very best friends even today.
[11:48] So moving on okay.
[11:52] Those are fond memories.
[11:52] Moving on if you ask me okay about the recent what made the industry some of the key innovations in industry.
[12:00] So to me com coming from the computing world okay packaging means interconnection.
[12:05] IOS, IOS, IIO, IOS.
[12:07] And as you know, I wrote the first modern textbook called micro
[12:14] electronic packaging handbook.
[12:17] And I define packaging to be interconnecting.
[12:19] Most important is interconnections, not protection.
[12:21] Obviously, you got protected interconnecting, powering, cooling, protecting.
[12:25] Okay.
[12:25] Chips.
[12:29] But now I I redefined it.
[12:31] Chips not enough.
[12:33] You really have to do that for system.
[12:35] interconnecting power and cooling protecting the entire system is where I am today with a concept I call SOP system on a package as opposed to system on a chip okay that I believe in that's where we are today and we've been in my view all along if you really want to do okay high performance computing as as I'm doing so the okay so given the iOS okay flip chip which now goes on to area array is is the single most important starting point of the whole industrial in my view.
[13:02] Okay, we're now for the first time able to put all the IO's on the chip and entire okay area of the chip.
[13:10] Now once you have the area array on the chip, now you got to have corresponding area array on the package.
[13:15] Therefore that okay that that forced us
[13:18] to develop substrate technologies
[13:20] initially ceramic that's what I was doing
[13:22] that quickly moved on to laminate buildup
[13:25] actually by IBM Japan my colleagues okay
[13:27] then uh then we moved on obviously to silicon interposer
[13:34] okay we're now in my view into glass-based packaging
[13:36] so I'm not sure where we going because most people never thought that someday we could actually reach one micron technology.
[13:46] In my view, we are less than one micron.
[13:49] We had demonstrated my my students demonstrated we can do on this panel less than one micron lithography
[13:56] very much like a back end of the line.
[13:58] Okay.
[14:00] So clearly therefore we can see okay soal panel packaging one micron with better better dilectrics okay with the direct cons and back end of the line.
[14:11] So I can clearly see the okay the value of uh okay some of the latest advances particularly glass-based
[14:15] packaging that that IP needed with my team at Georgia Tech.
[14:18] So the multip substrate then of course 3D we moved on to 3D 3D so clearly 2 and a halfD okay.
[14:27] just like we did not we did not get the performance in a board we had to move onto a package okay.
[14:30] putting in 2 and 1/2D is like a small board so we really have to do 3D 3D 3D so we've been at it now for 20 years 3D okay.
[14:39] it started with the intensity around around 2005 so in my opinion we still don't have a 3D we have 3D 3D logic we have 3D memory okay.
[14:51] 3D logic and memory for ultra high performance computing okay.
[14:54] we still don't have that the only 3D we have okay true 3D of logic and one stack is actually by Sony for uh image sensors okay.
[15:06] where they they're able to put a logical memory okay and and others into 3D stack but there ultra low power so we know I'll solve that problem but ultra
[15:16] high power we don't so we still we still have to.
[15:19] Okay. Uh beyond that.
[15:22] So the other one is glass glass panel packaging to one micron direct liquid cooling.
[15:25] Okay. We talk about direct liquid cooling.
[15:29] In fact that module there that you see on the screen is directly cooled.
[15:31] It's water cooled.
[15:33] Back then only three watts per chip.
[15:36] Okay. But by the time you had 144 chips you have so much power you had to go I had to go liquid cooling direct liquid cooling.
[15:44] Okay. copper sitting directly on the chip with close to zero thermal interface resistance.
[15:50] To me the critical technology is really thermal interface resistance.
[15:53] Okay, you really have to solve that problem.
[15:57] Okay, thermal grease and on those okay don't hack it.
[16:03] And uh so that's the problem I think we solved in a way.
[16:05] And the next one I think signal power distribution on the design tools methodology is clearly major major innovations that allowed all these technologies to come together.
[16:16] designed fabricated on down.
[16:20] This is my view of the key innovations okay to okay leading to where we are today.
[16:24] So if you ask me where do we go from here?
[16:26] Well we are coming to an end in my view in the electronic space.
[16:30] I'm looking from the research point of view as opposed to manufacturing.
[16:33] Manufacturing will go on forever of all the things we do today.
[16:35] But if you look at where we're going okay so people talk about musla going okay slowing down clearly it's been doing for a long long time.
[16:37] So in my view we really have to look for solutions in computing.
[16:39] Computing drives all the technologies incidentally.
[16:41] Initially it's all computing as the technology matures and goes to high volume reduce the cost to become applied to all the other segments.
[16:45] So computing drives so okay so if you look at what the computing needs are in my view we really have to look for a million times more computing power than where we're at today.
[16:50] Can we do with what we have
[17:17] today?
[17:17] Not really.
[17:17] Okay.
[17:19] Therefore what will be the next wave?
[17:22] So next week must be okay photonic based you can call it integrated opto electronics beyond CPO.
[17:28] CPO to me is a is a short gap we really have to go beyond beyond that almost photonic driven.
[17:35] Okay after electronics obviously okay.
[17:37] Could be that but highly integrated lead into quantum computing.
[17:42] Okay clearly some companies may skip all of this go directly into quantum computing to get that million times more okay but clearly a lot of challenges.
[17:51] But ultimate you know we can't keep on putting keep on putting power to get the performance we have known for 50 years okay.
[17:58] One way to get the more more performance putting more power and therefore we're now at gigawatt in a way so we really have to think of entirely different way.
[18:08] So I want to leave with this thought that the ultimate ultimate system to me is this one here.
[18:13] This thing has a 90 billion interconnections.
[18:18] as opposed to what we have today not even a million.
[18:20] So a long way to go and this thing instantly does not take more than 20 watts of power okay as opposed to megawatts of power that we are planning on.
[18:28] So somehow we need to find a way okay neuromorphic mimicking neuromorphic computing systems mimicking human brains in some way to me the ultimate solution that we all need to strive for.
[18:40] Thank you very much.
[18:42] Thank you, Ralph.
[18:51] Our next speaker is John Laauo.
[18:53] John has more than 40 years of R&D and manufacturing experience in semiconductor packaging.
[18:58] He's published more than 500 peer-reviewed papers and 24 textbooks.
[19:04] Has 52 issued and pending US patents.
[19:07] He's an elected ILE E fellow, an IMAPs fellow, and an ASME fellow.
[19:14] He's been actively participating in industry academy and society meetings
[19:19] and conferences to contribute, learn and share.
[19:33] John uh Pat asked me to talk about three topics.
[19:39] May I have my slide?
[19:42] The first topic is The first topic is early memories of ECTC and key innovations in the past.
[20:00] There are many good papers in the past ECTC.
[20:06] So I set up a criteria that has to be in high volume production.
[20:17] I will talk about four papers.
[20:21] The first one is in 1992
[20:28] IBM at Yasu
[20:31] Sukaden.
[20:33] He present the SLC
[20:38] technology
[20:40] which becomes the buildup package
[20:45] substrate. Today
[20:48] just about every electronics product
[20:52] use buildup package substrate.
[20:58] Actually if there were a Noel price for
[21:02] packaging this is it. Okay. The impact
[21:07] is this.
[21:10] The second paper is
[21:14] 2005.
[21:17] Leti
[21:19] they present the T system on wafer
[21:24] package
[21:26] which is by using a TS3 interposer
[21:32] to support multiple chips.
[21:35] Now a day we call it 2.5D
[21:41] IC integration.
[21:45] TSMC
[21:47] in 2013
[21:51] they have siling
[21:54] to ship the products.
[21:57] They have siling to ship the FPVGA
[22:02] in 2013.
[22:06] Today
[22:08] the package of choice for product driven
[22:14] by AI is 2.5D
[22:18] IC integration
[22:32] in
[22:35] two to 2010,
[22:39] 15 years ago,
[22:42] stack chip pack
[22:45] with Arison
[22:48] and Infinion.
[22:50] They propose to embed the application
[22:56] processor
[22:58] in fan epoxy molding compound.
[23:05] 2016
[23:08] TSMC
[23:09] make it into production.
[23:12] They help Apple to ship their iPhones.
[23:19] I always say
[23:22] TSMC
[23:23] and Apple they save the fan. Now
[23:29] before 2016
[23:32] the application of fan now is for very
[23:36] small chips very large live with and
[23:40] space RDL.
[23:44] But since 2016
[23:48] after
[23:49] TSMC
[23:51] helped Apple to ship their iPhone, it is
[23:55] for very large chip
[23:59] more than 120 mm very high performance.
[24:07] And the number four paper I would like
[24:12] to mention is in 2013
[24:17] SE chip pack again
[24:20] they
[24:22] propose to use the final redistribution
[24:28] layer to replace the TS3 interposer.
[24:37] This is what we call 2.3D
[24:40] [Music]
[24:41] IC integration.
[24:45] Last year TSMC present a paper to
[24:52] put it into production. So 2.3D is
[24:58] already in high volume production.
[25:08] So next topic
[25:10] Pat wants me to talk about is
[25:14] what is happening
[25:17] now. Excite me.
[25:22] I guess this is not difficult to answer.
[25:26] Nowadays copper to copper hybrid bonding
[25:31] is heartless.
[25:34] Last year there were 80 papers.
[25:39] The year before last there were 72
[25:43] papers.
[25:45] 2022
[25:47] there were more than 65 papers.
[25:51] So cuppper to cuppper is a hot list
[25:54] today.
[25:58] Next one
[26:01] is 3.5D
[26:04] IC integration.
[26:07] That is before you put the component on
[26:13] a single substrate. That means before
[26:16] you do historicis
[26:19] integration
[26:21] you should do some
[26:24] front end integration that is chiplet to
[26:29] chiplet copper hybrid bonding.
[26:35] This is called 3.5D
[26:40] that will reduce the package size and
[26:44] increase the electrical performance.
[26:54] The number three is the bridges
[26:59] for chiplet.
[27:02] Of course 2016
[27:05] IBM they present their EMIB
[27:11] that is the bridges embedded in buildup
[27:17] package substrate.
[27:20] 2023
[27:23] TSMC present two papers.
[27:28] They are going to replace their TS3
[27:33] interposer
[27:35] by silicon bridges embedded in fan epoxy
[27:43] molding compound they call chip on wafer
[27:47] on substrate L.
[27:50] By the way, it is going to be in
[27:53] production for Nadia AI product before
[27:59] end of this year.
[28:03] The number four topic exciting me is
[28:08] glass.
[28:11] There are at least three features of
[28:15] glass packaging.
[28:19] One is to replace the glass core. I'm
[28:23] sorry.
[28:25] One is to replace the organic core into
[28:30] glass core.
[28:33] We call glass substrate. That's 2D.
[28:38] Another
[28:40] one is to replace the TS3 interposer
[28:46] into glass interoser
[28:50] that is for 2.5D
[28:55] for 3D
[28:57] we can embed the chip into the glass
[29:03] substrate and become a 3D
[29:08] AC integration.
[29:12] Actually, glass
[29:16] works very well with optics.
[29:24] Okay. The last thing he wants me to talk
[29:29] about is your thoughts on what is coming
[29:36] on the future
[29:39] of ECTC.
[29:42] Again, I have four phase to propose.
[29:48] Number one, for AI and communications,
[29:55] we should use co- package optics
[30:02] like the one on your well the one on my
[30:07] left
[30:08] is for AI and the one for my right is
[30:14] for co- package optic.
[30:19] combine them together, you can do AI and
[30:24] good communications.
[30:29] The second one
[30:33] is called 3.3
[30:38] ID
[30:40] 3 I'm sorry 3.3D
[30:44] IC integration.
[30:49] The reason is the size of the TSV
[30:53] interposal
[30:54] getting bigger and bigger.
[30:57] The manufacturing you is lower and
[31:02] lower. The cost is higher and higher.
[31:07] So
[31:09] let the high bandwidth memory
[31:13] stack up on the SOC
[31:18] and then attach it to a glass core
[31:23] substrate without TS3 interposer.
[31:36] The third one is
[31:39] we should use glass substrate more and
[31:44] combine with the
[31:51] high bandwidth memory and the switches.
[31:58] For example, this one is to embed the
[32:02] PIC, the photonic IC
[32:06] in the glass substrate
[32:09] and then couple to couple hybrid bonding
[32:14] with the electronic IC.
[32:18] The reason we use the glass substrate is
[32:22] for the
[32:24] glass way.
[32:27] And the last one is
[32:30] quantum computing.
[32:34] In 1980
[32:37] Richard Fineman Freeman he said not many
[32:41] people understand quantum computing that
[32:45] excise many physics working in this
[32:48] area. I don't I know nothing about it. I
[32:52] only know quantum computing works like a
[32:58] refrigerator.
[33:00] That means it's very very low
[33:02] temperature.
[33:04] I wonder
[33:06] how is the packaging
[33:09] survive in that kind of environment.
[33:13] Thank you very much.
[33:16] [Applause]
[33:22] >> Thank you, John.
[33:24] Our next speaker is Bill Chen,
[33:27] also known as William Chen. holds the
[33:30] position of ASSE fellow and senior
[33:32] technical adviser at ASSE group. Prior
[33:35] to joining ASSE, he was director at the
[33:38] Institute of Materials Research and
[33:40] Engineering in Singapore following a
[33:42] distinguished career at IBM Corporation.
[33:44] He's a past president of ILE E
[33:47] Electronic Packaging Society as well as
[33:49] being a life fellow of ILE E and a
[33:52] fellow of ASME. He received the ASME
[33:55] interpac achievement award in 2007 and
[33:59] in 2018 he received the ITE electronic
[34:02] packaging field award. Bill shares the
[34:04] heterogeneous integration roadmap
[34:06] initiative co-sponsored by three ITLE E
[34:10] societies along with semi andme bill.
[34:17] [Applause]
[34:31] Um Pat asked me to say a few words first
[34:36] about uh ECTC.
[34:38] So ECTC has been a long
[34:44] history of uh um innovation in terms of
[34:50] bringing all of the team together.
[34:53] Now there's something happened 15 years
[34:56] ago for ECTC
[34:59] and uh
[35:01] so I would talk about 2010
[35:05] what happened at 2010 and then also talk
[35:08] about the key innovations that set the
[35:11] landscape today and
[35:15] observe that future is indeed unlimited.
[35:20] Now, ECTC
[35:24] for a long time was a joint effort
[35:28] between
[35:30] electronic pattern society
[35:32] at that time was CPMT
[35:35] and ECA.
[35:37] Now we each of us being a joint each of
[35:42] us has 50% of the vote.
[35:47] 50% of the vote sometime is difficult
[35:51] as you all know that uh if we want to do
[35:57] something and uh like for example uh
[36:02] nominate a treasurer
[36:04] and then if the other side doesn't agree
[36:07] we cast the vote 50 50 and then nothing
[36:11] happens.
[36:12] So the buyout happened 15 years ago.
[36:18] 15 years ago we
[36:23] I think it took three past president of
[36:26] uh uh CPNT plus a current president of
[36:30] CPMT in order to make that happen.
[36:34] Um I think uh as you show over here at
[36:37] that time the president was Ralph
[36:39] Ashenber and he signed the agreement
[36:42] with uh ECA and uh for E for CPMT to uh
[36:50] um to be a sole um owner of the ECTC.
[36:58] That is a buyout
[37:01] um by myself, past president, Phil Guru,
[37:05] past president and uh Jim Turella. Is
[37:08] Jin here?
[37:11] Yes, Jeene Terrella. So, it took all of
[37:13] us to make that happen. And once that
[37:17] happened of course you can see that
[37:19] there's a no decision need to be made
[37:24] 50/50 decision can be made by the
[37:28] committee over here like all of you here
[37:32] we have this wonderful wonderful ECTC
[37:34] program year after year and I think that
[37:38] was a time you that was a time that made
[37:43] a important decision invested
[37:47] during the buyout and we and we have
[37:51] outstanding ECDC 2025 and we look
[37:55] forward to going to
[37:58] another uh 25 years and I think we can
[38:03] be proud that all of us work together
[38:06] and made that happen. So that's the uh
[38:11] 2010 ECTC milestone. I still remember
[38:15] uh with some of the pictures taken of uh
[38:19] of the uh signing of of the contract.
[38:26] Now I'm going to talk about fan out.
[38:29] Now you can see originally fan out is a
[38:33] very low end
[38:36] uh application. It was the idea that you
[38:39] have wafer level CSP
[38:42] and then the wafer level CSP from one
[38:46] generation to another generation die
[38:48] shrink and doesn't work anymore. So in
[38:52] the same format because you have this
[38:54] product you want to have the same same
[38:56] footprint. So the fan art originally, if
[39:01] you look at it, was to to make the same
[39:05] footprint.
[39:06] And uh
[39:10] and then
[39:13] people found out that oh when we put one
[39:16] die in, we can put two D.
[39:20] The same factory can do at the same
[39:22] time.
[39:25] You know it took some years to figure
[39:28] out that this could happen. Now once we
[39:31] put two dye in you can say oh I can put
[39:35] a circuitry on top of it.
[39:44] So you go this step from the number A to
[39:50] B to C to D and then in 2016 we figured
[39:57] out that you can make a product
[40:00] at that time we made the product but
[40:04] people's design rules and things like
[40:06] that took years to follow.
[40:11] But it was something that looking at it,
[40:15] looking back and said fan out starts
[40:19] from this very very humble beginning
[40:23] and step by step with ideas and
[40:28] innovations from people in this room and
[40:32] then it happened.
[40:37] So that's how
[40:39] I look at it, looking all those papers
[40:42] on fan out and thinking about the people
[40:47] working on those and innovations
[40:50] pile one on top of each other
[40:55] and I saw so many good p important
[40:59] boutile paper today and I could think
[41:01] about that it takes all of us together
[41:06] as I just say innovation power on
[41:09] innovations driven by the need of the
[41:12] market.
[41:15] Now there's another one that we see is
[41:19] uh um 2.5D
[41:23] and u
[41:25] on the right over here you could see
[41:28] that the world's first 2.5D with four
[41:32] HBMs
[41:34] it's released on June 2015 same 10 years
[41:39] ago
[41:41] uh it was AMD and ASC
[41:44] And uh it took a long time, eight years.
[41:48] We worked on this together with AMD.
[41:52] Many times
[41:54] um I go to the factory, go to the lab
[41:58] and see their progress, but it took
[42:01] eight years um from the beginning to the
[42:05] very end. But we learned a lot of
[42:08] things. Some of the things such as what
[42:11] mean how do you how do we measure
[42:13] warpage? Warpage become a big problem.
[42:16] How do we learn to wage? How do we learn
[42:19] how to model? So in developing these
[42:26] difficult
[42:28] but pro but innovative ideas one also
[42:32] learned a lot of lessons and th those
[42:35] lessons
[42:37] kept us going and I'm sure some of those
[42:40] lessons you saw are reported today.
[42:45] I've looking at the future the this is a
[42:49] slide from uh Kevin at TSMC and on April
[42:54] 25th he showed uh this was uh from the
[42:58] TSMC press briefing you could see that
[43:02] we look at the industry and industry
[43:04] will be $1 trillion.
[43:08] We all together
[43:12] will be helping to make this thing to
[43:14] happen.
[43:16] And I'm believe that if we lock out like
[43:21] pet asked to say locking out another 10
[43:25] years
[43:27] this number is going to be multiplied.
[43:31] It will not be on all the ad on all the
[43:36] uh um
[43:38] what we call advanced packaging areas.
[43:41] you will trickle down so that you will
[43:43] benefit
[43:45] all the
[43:47] less
[43:49] advanced packaging products in the world
[43:52] and making things for us that benefit
[43:56] society in whole everybody. So that's
[43:59] what I believe is contributing for the
[44:04] society, contributing
[44:06] us working together and thank you very
[44:10] much.
[44:12] [Applause]
[44:19] >> Thank you, Bill. Our fourth speaker is
[44:22] Kitty Pearl. Kitty earned her PhD degree
[44:25] in mechanical engineering and materials
[44:27] from UT Austin in 1983. She worked for
[44:30] IBM from 1972 to 2013 where she became
[44:34] both an IBM distinguished engineer and a
[44:37] member of the IBM Academy of Technology.
[44:40] She was a process consultant and subject
[44:42] matter expert working on strategic
[44:45] initiatives impacting component quality
[44:47] and end quality of procured commodities.
[44:50] She's currently an independent
[44:52] consultant for Capstand Technologies.
[44:55] While at IBM, Kitty received four
[44:57] outstanding technical achievement
[44:59] awards. She holds 12 US patents. She's a
[45:03] licensed professional engineer and the
[45:05] recipient of the UT Austin Cochril
[45:08] Engineering Distinguished Engineering
[45:10] Graduate Award in 2007
[45:13] and is a charter member of the UT
[45:15] Mechanical Engineering Department
[45:17] Academy of Distinguished Alumni. She's
[45:20] an active member in IT E and EPS and as
[45:23] well as a member of TMS and ASM and WIE.
[45:28] She is the longtime chair of the PDC's
[45:33] at ECTC and was the EPS president in
[45:37] 2223 and is currently the EPS junior
[45:40] past president. Kitty
[45:43] [Applause]
[45:53] I'm going to approach this a little bit
[45:55] different. Um, from my perspective, um,
[45:59] we've heard a lot of, uh, fascinating
[46:01] things this morning about the current
[46:04] and the future environment uh, for our
[46:07] advanced packaging world and processing.
[46:10] But what I want to do is share with you
[46:14] my perspectives of how it was when I
[46:17] started getting involved and how I
[46:19] believe it is today and what I think
[46:22] from my viewpoint it will be like by the
[46:25] time we reach 100 the ectc
[46:34] right and one of the things that Oh, I
[46:37] got to do it this way.
[46:39] Okay. One of the things that I I want to
[46:42] start with, it was a very different
[46:44] landscape then than it certainly is
[46:48] today. If you think back to uh I started
[46:51] actually in actual engagement in ECTC in
[46:56] 1992
[46:58] and I was a very m materialsdriven
[47:02] type of individual and we were very much
[47:06] impacted by what was going on around us.
[47:10] You know at that time the EU Montreal
[47:13] protocol was being pushed. there were 26
[47:17] companies that signed up to meet this
[47:21] target of reducing you know the the
[47:25] CFC's that were in our environment that
[47:28] were destroying the uh ozone layer. Now
[47:32] um
[47:34] in 19 we we were supposed to meet these
[47:36] challenges by 1992.
[47:38] Uh IBM we beat that by two years. So I
[47:42] thought that was pretty good.
[47:46] a lot of regulatory requirements. The
[47:48] CFC's uh which were the ozone depleting
[47:52] uh chemicals, these were the uh focus of
[47:55] the Montreal protocol. We then moved
[47:58] into the toxic release uh inventory,
[48:03] excuse me, in chemicals, inventory
[48:06] chemicals, which okay, there's more than
[48:08] CFC's that are being put out there. We
[48:10] need to see what else is being released
[48:12] into the environment.
[48:14] Moving on to the assembly processes.
[48:16] This is where you start looking at
[48:18] ROHOS. I don't know how many people here
[48:19] have had the experience with the
[48:21] restriction of hazardous substances. It
[48:24] meant several things. We had to change
[48:27] from our tried andrue and working for
[48:31] years tinled rosinbased solders and we
[48:36] transitioned to the water soluble tinled
[48:40] solders. And then we had to tradition
[48:43] again uh to transition again to the low
[48:46] residue tin lead and where we are today
[48:50] leadf free and there's a variety of
[48:51] those materials. Um also this wie we
[48:57] were running into as a manufacturing
[48:59] facility we had to do all the waste uh
[49:03] for from electrical electronic equipment
[49:06] and I'm sure any of you in those kind of
[49:08] companies you were dealing with that
[49:10] also. So at that time I believe both uh
[49:14] well specifically Ral and and John they
[49:16] were pointing out I guess bill too where
[49:19] we started we started with pin and hole
[49:22] technology we went to definitely um
[49:25] surface mount and the surface mount at
[49:28] the time we're covering our AS6 or MCMS
[49:31] chip scale packages and on and one of
[49:35] the things I like it's just a favorite
[49:37] of mine I just threw it up there for me
[49:38] is to look at how did we disseminate all
[49:41] the knowledge that we gathered here at
[49:44] ACTC over these 75 years. And I I just
[49:49] show it's on the left, bottom left, you
[49:52] will see that's how we used to come to
[49:54] ECTC. You'd sign in and they'd give you
[49:57] this big heavy book. That was your
[50:00] knowledge. And you would take this and
[50:02] you would go into the various sessions
[50:04] and you'd be looking at it and reading
[50:06] it and you already knew. So, it was in
[50:08] your hands before you went into the
[50:10] session. Um, then of course we went to
[50:13] our cute little memory sticks and now
[50:16] it's all online. So, I think it's just
[50:20] kind of a sign of the times.
[50:24] All right. Today we have so many
[50:27] innovative technologies uh and and we're
[50:31] faced with a lot of the advanced
[50:33] packaging that you heard the earlier
[50:36] presenters talked to and it's to the
[50:39] point that uh companies are having to
[50:42] use all of this to enhance their
[50:44] competitiveness.
[50:46] They want to be the one that's just a
[50:48] little bit ahead of the other one. And
[50:50] so uh we're going to have a lot more
[50:53] focus. We've already seen the focus on
[50:54] smart manufacturing,
[50:56] you know, smart factories, etc. Uh that
[51:01] has to increase and it is increasing.
[51:04] We're seeing, you know, the uh
[51:06] artificial intelligence and machine
[51:08] learning coming in that's really going
[51:10] to be there. It we're we're creating and
[51:13] collecting massive amounts of data about
[51:17] our assembly process pro uh processes,
[51:20] how we can optimize.
[51:22] We're utilizing as I said AIT AI and IoT
[51:26] to gather this data. We're fast moving
[51:29] in. There's people already doing they're
[51:31] they're in-depth of analysis of how can
[51:34] we use digital twins to improve the
[51:39] product and it starts collecting data as
[51:42] you build the particular product.
[51:45] collecting that data as you go through
[51:47] the uh pro processing uh process with
[51:50] the manufacturing. You collect all that
[51:53] so that you know so that you have the
[51:56] ability to see around the corners what's
[51:58] going to be happening at the end of that
[52:00] line on the end product based on what
[52:03] you started with and you can make some
[52:05] real time adjustments. That's where the
[52:07] AI and ML really come in for analyzing
[52:13] and knowing what we should be expecting.
[52:16] I'm really excited to see all the
[52:18] modeling uh and simulation uh
[52:21] presentations that have gone on here
[52:23] this week to show us how can we model,
[52:26] how can we then predict.
[52:29] And of course the the other one is
[52:32] enhancing our cyber security. anyone in
[52:35] here that has been impacted as a result
[52:38] of uh poor security.
[52:42] This is where we are hoping that the AI
[52:45] can come in and really help protect us.
[52:48] So specifically for the semiconductor uh
[52:51] supply chain
[52:53] of course nowadays we're seeing a lot of
[52:55] emphasis on our supply chain for you
[52:58] know we could spend hours talking about
[53:00] that but the whole thing is we need to
[53:03] protect the supplier data the supplier
[53:06] processing uh data and the quality and
[53:09] reliability output of the finished
[53:11] product
[53:13] definitely goal ensure continuity of
[53:16] supply for the bill of materials
[53:18] And of course, what's been going on the
[53:21] last three or four years, introduction
[53:22] of all these new advanced packages,
[53:29] it didn't go.
[53:33] Okay. So, looking at, you know, what do
[53:36] what do I see coming up? Of course, I've
[53:39] got a materials engineering perspective,
[53:41] so I'm approaching it in that way.
[53:44] Definitely we're going to have as we
[53:46] already see the expansion of the
[53:47] electronic vehicles uh station excuse me
[53:51] electronic uh vehicles the EV global car
[53:55] market share is uh projected to grow by
[53:59] 40% by the end of 2030 and if you look
[54:03] at that the global sales projected to
[54:07] exceed the $20 million that we have in
[54:10] 2025 and so that's pretty exciting I
[54:13] mean you Oh, wow. Um, but what's going
[54:16] to make that even better? What will make
[54:19] that better? So, with the EVs, you're
[54:23] going to need expansion
[54:26] of the EV charging stations that we have
[54:29] out there today. We don't have enough to
[54:33] really realize that if we eventually get
[54:36] to where this is all there is, you'll be
[54:38] able to maintain your charges.
[54:41] uh you know but this has got to be
[54:45] controlled I think well government and
[54:47] businesses are out there to help us meet
[54:50] those projected EV growth. Another thing
[54:53] one of the problems that we're seeing
[54:55] with the EVs is the increase well the
[54:59] lack of driving distance that we would
[55:02] like to have. So we need to see that
[55:04] increasing and right now it's being
[55:07] limited by the current battery design
[55:11] and that is something you know we should
[55:13] talk to. We need to be looking forward
[55:15] at solidstate batteries. We all know
[55:18] about the regular automobile out uh
[55:21] batteries where we are going from
[55:23] electrolytic cell type batteries. We
[55:25] then move to u a sort of polymer gel
[55:30] separating the anode and and cathode and
[55:33] the batteries, the lithium ion
[55:34] batteries. But we need to get to the
[55:37] point where you actually have a solid
[55:39] state material separating those. And um
[55:44] this is something that
[55:46] even with the the Tesla product out
[55:48] there, I could not find uh any
[55:51] prediction on their part that says we
[55:54] will have our solid state battery by
[55:56] this state. But uh if you look up here,
[56:00] it's Toyota is promoting that they will
[56:03] have their first solid state battery in
[56:07] 2027.
[56:08] And then you've also got Volkswagen and
[56:11] Renault that are close behind. So this
[56:14] is something we need to be looking at as
[56:16] the the uh power generation the
[56:20] electronic vehicle charging as I talked
[56:23] about that it has to grow. We have to be
[56:26] able to get those news new batteries in
[56:28] there get the charging stations that are
[56:31] going to support them and we'll be able
[56:32] to move on from there. all of us. I
[56:34] mean, I'm certainly a horrible driver,
[56:36] so I really like working on this
[56:39] advanced driver assistant systems. You
[56:42] know, if you see Star Trek, be me
[56:44] aboard, Scotty. I'm all for that. Um,
[56:47] also with everything, we need standards.
[56:50] If we don't have standards, we can't
[56:53] control what's going on around us. So,
[56:56] we have to have some standards out there
[56:58] being developed for these batteries. So,
[57:00] it's across the board. It's it's they're
[57:03] going to be ind industry um independent.
[57:06] Everyone has to meet the standards. And
[57:08] this way we'll also then ensure that our
[57:11] global supply challenges are met for the
[57:15] new materials and advanced packages for
[57:17] EVs. Basically, you know, we looked at
[57:20] how it was when I came. Here we are
[57:23] today. Look at around this room and
[57:25] where I think we need to go in the
[57:28] future for materials. Thank you. All
[57:32] [Applause]
[57:46] right.
[57:47] Now, it's time to take uh questions from
[57:50] from the audience for a panel. Please
[57:52] come up to the mic or put one in the
[57:54] app. I'm going to start with the one on
[57:56] the app and this is for the panel. Could
[58:00] you share your thoughts on how you
[58:01] believe AI will influence the future of
[58:04] electronic packaging?
[58:18] Okay. Well,
[58:21] AI is going to influence every aspect of
[58:23] what we do. Not just packaging, of
[58:26] course, everything else too. So
[58:28] packaging packaging to me is all about
[58:30] advancing the next generation of
[58:32] technologies. So AI will enable that
[58:34] much faster because we have all the
[58:37] information we need much faster and all
[58:40] the data you ever need. So in a way you
[58:43] should be able to model and simulate
[58:46] taking the data existing and predict
[58:49] okay predict not only electrical which
[58:52] we know very well but uh mechanical
[58:54] thermal all of those
[58:57] and uh even optimize for uh across all
[59:01] these so as to end up with a tradeoff
[59:03] architecture
[59:05] that allows new technologies to be
[59:07] developed. So AI is the only way in my
[59:11] opinion because the technologies are so
[59:13] so complex.
[59:14] So I've been writing articles lately and
[59:17] and telling the world that to make a
[59:19] product it requires 12 different
[59:21] technologies.
[59:23] Most of us are experts in one of those
[59:25] okay AI is the only way to bring all of
[59:28] them together
[59:30] and okay and propose
[59:33] design and architecture option to solve
[59:35] a problem that need to be solved. Okay.
[59:38] In my view,
[59:46] um,
[59:48] AI is going to create new jobs. AI is
[59:53] also going to change the job market
[59:57] because some of the industry will be
[01:00:00] changed dramatically because of AI. We
[01:00:04] saw it last night. If you see last
[01:00:08] night, you see the drone. Isn't it a
[01:00:11] wonderful, wonderful display?
[01:00:14] And when I thought about it, what it
[01:00:17] meant is that
[01:00:20] two or three years ago, instead of those
[01:00:23] drone, we would have fireworks.
[01:00:27] And so overnight, I saw that the
[01:00:31] firework industry is going to be
[01:00:36] That's what's
[01:00:37] at the same time you could see that
[01:00:40] there's new job that's created because
[01:00:42] the person who designed the
[01:00:47] display
[01:00:49] is going to be a very much sought after
[01:00:54] person because he understand management
[01:00:58] of the drone. He understand how to
[01:01:01] multiply the database and make that
[01:01:04] happen.
[01:01:06] So AI as I just said will create new
[01:01:11] jobs, will create new industries. At the
[01:01:15] same time it will change
[01:01:19] what may be some industry that could not
[01:01:23] would not survive today.
[01:01:27] >> Uh one of the things that Bill mentioned
[01:01:30] about increased jobs in the workplace of
[01:01:35] Every one of us here as we approach AI
[01:01:39] and as we approach learning and smart
[01:01:41] learning manufacturing, we need to
[01:01:43] seriously make a commitment that we are
[01:01:46] going to build this workplace, you know,
[01:01:50] for uh excuse me, build our workplace
[01:01:53] and our technical development of the
[01:01:56] young sprites coming up
[01:01:59] into this world. We need to get more
[01:02:03] development challenges for them. We need
[01:02:06] to get them ready and able to accept all
[01:02:09] these challenges from, you know, AI and
[01:02:11] ML. It's a different it's a different
[01:02:14] learning process. You know, you can't
[01:02:16] just, yes, you can get it from, but they
[01:02:19] really need to know what's going on
[01:02:21] around them. They need to be heavily
[01:02:23] involved. What does AI and digital uh
[01:02:26] reality mean and how am I going to use
[01:02:28] that in in the near years? So I I'm
[01:02:32] pretty excited about AI and machine
[01:02:35] learning and I do believe that we've got
[01:02:38] to get more into the digital realities
[01:02:41] with these uh learning where you can
[01:02:44] actually test things out before you can
[01:02:47] generate all these use cases and they
[01:02:50] can you know you can take them and say
[01:02:52] ah if this happens
[01:02:55] or if something is no longer available
[01:02:58] in my supply chain I already knew that
[01:03:00] maybe 24 hours earlier and I put a plan
[01:03:03] in place to do it. But we need to have
[01:03:05] the the workforce development behind
[01:03:09] that in order for us to truly be, you
[01:03:11] know, be successful.
[01:03:17] Uh you heard all the answers. I agree
[01:03:23] with those
[01:03:25] from packaging points of view.
[01:03:30] uh SOC such as uh graphic processor
[01:03:37] will be require much finer feature size.
[01:03:43] If we will go down to two narome meters.
[01:03:48] Also the high bandwidth memory
[01:03:51] the bandwidth
[01:03:54] larger and the requirement for the high
[01:03:57] bandwidth memory is much much higher.
[01:04:08] Um, this questioner says, "In my
[01:04:11] opinion, the success of the A IML
[01:04:14] community is mostly given."
[01:04:17] >> Oh, all right. Go ahead. Sorry, I didn't
[01:04:18] see you in the light.
[01:04:20] >> Thank you. My name is Jaka Ding. I'm
[01:04:22] from the Swedish University of Northern
[01:04:24] Northern Sweden. uh I hear a lot talk
[01:04:28] about packaging and materials and
[01:04:33] basically upstream the business but I
[01:04:36] don't hear hardly anything about the
[01:04:39] downstream part and in particularly the
[01:04:41] interaction with the software domain
[01:04:44] which is utilizing all the power that we
[01:04:47] are putting together in this domain but
[01:04:51] wouldn't what about that dimension
[01:04:55] Is there anyone that has a comment on
[01:04:56] how that potentially could further
[01:04:59] enhance our developments?
[01:05:22] So let me try and rephrase the question
[01:05:25] you're asking about the interaction
[01:05:27] between the software use of the hardware
[01:05:30] that we're building,
[01:05:31] >> right? Or are you you wanting to hear
[01:05:33] more about the back end of the line
[01:05:36] manufacturing process versus
[01:05:38] >> No, I think I'm I'm I'm very much
[01:05:41] realized that the software guys can
[01:05:42] change things very very fast. On the
[01:05:46] hardware side, as we heard earlier on,
[01:05:47] it took eight years to actually go from
[01:05:50] the ID to product. Uh software guys can
[01:05:53] do that in days.
[01:05:56] And but the software guys, they don't
[01:05:58] understand the hardware. that don't have
[01:06:01] any good training in physics.
[01:06:09] >> Oh, sorry.
[01:06:15] >> Maybe I'm speaking too close to the mic.
[01:06:20] >> Okay.
[01:06:32] So, so the the question is very much uh
[01:06:35] we are speaking here about uh the
[01:06:38] packaging and the materials meaning
[01:06:40] upstream but what about downstream part
[01:06:44] of this business since what we are doing
[01:06:47] is actually providing the downstream
[01:06:50] people downstream engineers which is to
[01:06:52] a very large extent software engineers
[01:06:55] doesn't understand very much about the
[01:06:57] physics, the mechanics that we are
[01:06:59] dealing with. But what about that
[01:07:02] interaction? How do we help them and how
[01:07:05] can they help us?
[01:07:14] >> Okay.
[01:07:16] One of the things that, you know, I'm
[01:07:18] hoping to see when you're starting to
[01:07:20] talk about those software and
[01:07:22] development engineers, you know, because
[01:07:24] they need to know what the materials are
[01:07:26] that they're going to be working with so
[01:07:28] that they are better able to understand
[01:07:31] what kind of use cases they need to put
[01:07:33] together. What is it we need to be
[01:07:35] simulating? Um, you know, there is a lot
[01:07:38] of materials involvement. Uh that's why
[01:07:41] I thought you know today mine was a
[01:07:43] little bit different because I'm
[01:07:44] basically a metallergical engineer and
[01:07:47] um how I got with all these doubles I'm
[01:07:48] not sure but anyway anyway so yes uh we
[01:07:53] do need to you know
[01:07:57] shore up that physical or the physics
[01:08:00] that are behind what the software
[01:08:03] developing team is doing. Uh we we need
[01:08:07] to hear from them. we need to get back
[01:08:09] to them and we need to have better
[01:08:11] better collaboration.
[01:08:13] So yes, it is something that maybe is
[01:08:15] not as tight as we would like it today
[01:08:17] but it needs to be worked on. I don't
[01:08:21] know answered your question but
[01:08:24] >> and I have go on and follow up and give
[01:08:27] my sales pitch.
[01:08:28] >> You okay so sales pitch is heterogeneous
[01:08:32] integration road map. You have to have a
[01:08:35] road map so that you could give a vision
[01:08:39] of what the future is whether it is five
[01:08:43] years, 10 years, 15 years. If you think
[01:08:47] about originally the national technology
[01:08:50] road map, the international technology
[01:08:53] road map, they were there for that
[01:08:55] purpose.
[01:08:57] Um,
[01:08:59] originally Baltimore Moore was the
[01:09:02] person who started the NTS.
[01:09:06] And what he said was that he looked out
[01:09:09] into the three years cadence
[01:09:13] of uh um semiconductors
[01:09:16] and he said he we need to have the
[01:09:20] materials,
[01:09:22] the equipment
[01:09:24] and as equally important the science the
[01:09:28] science need to be developed. So the
[01:09:31] road map
[01:09:33] is the direction is the way in which we
[01:09:39] can get this information
[01:09:43] put together.
[01:09:45] People talk about materials and but I
[01:09:48] think science is equally important and
[01:09:51] all of you in this room over here who
[01:09:54] have
[01:09:56] who have uh research
[01:09:59] you have great graduate students they
[01:10:03] are the future.
[01:10:05] Yeah, my my two cents I think what
[01:10:07] you're asking is we talk about a lot of
[01:10:09] packaging here but we don't talk much
[01:10:11] about physics chemistry materials all
[01:10:14] the science scientific based to make
[01:10:17] packaging happen in this society we are
[01:10:20] here mostly talk talk about engineering
[01:10:23] engineering aspects of of electronics so
[01:10:27] there are other societies like MRS many
[01:10:30] many other societies that deal with the
[01:10:32] materials and the science and technology
[01:10:34] behind behind those. So this society
[01:10:37] doesn't concentrate as much in
[01:10:38] fundamentals
[01:10:40] but other societies do and uh but
[01:10:43] clearly whatever we do here is is all
[01:10:46] based on science okay even though we
[01:10:49] don't report okay we assume we do lot of
[01:10:53] scientific work but you only here we
[01:10:56] only talk about what we achieved in
[01:10:58] terms of engineering achievements as
[01:11:00] opposed to scientific mechanisms in this
[01:11:03] society.
[01:11:05] Thank you.
[01:11:07] >> Can you hear me? Okay.
[01:11:10] >> Okay. So, uh you know, you're sitting
[01:11:12] behind the
[01:11:33] So in 1950 our industry was in its
[01:11:36] infancy. Um and a group of engineers
[01:11:41] from leading companies recognized that
[01:11:45] and I'm going to use the word standard
[01:11:47] here. There were no standards. There
[01:11:49] were no rules.
[01:11:51] We were building material but it wasn't
[01:11:54] lasting. And there was a recognition
[01:11:56] among some of the visionaries of our
[01:12:00] industry that we needed to get together
[01:12:03] and come up with ways to make this work
[01:12:06] or it was going to be a failure. Um and
[01:12:09] so in 1950 this seminar was held in
[01:12:12] Washington DC by members of companies
[01:12:15] like IBM and they sat down and talked
[01:12:19] about how are we going to get
[01:12:21] consistency in our materials,
[01:12:23] consistency in our processes? How do we
[01:12:26] assess? How do we what metrics are we
[01:12:30] going to use so that when we build this
[01:12:32] equipment these things for people to use
[01:12:36] that they will be fit for use. So that
[01:12:39] if you want to go back 75 years that was
[01:12:41] the inception of um what is today ECTC.
[01:12:46] It was a two-day seminar uh with about
[01:12:50] 10 sessions and it was successful in
[01:12:55] that people met and they started what
[01:12:57] turned out to be standards bodies and
[01:13:00] the like in our industry. Um there was a
[01:13:03] second seminar in 1952
[01:13:07] uh that covered some of the same
[01:13:10] activities and then that gradually
[01:13:13] worked into what became the ECC in 1958.
[01:13:18] Even in the 1950s we had topics like
[01:13:22] materials, advanced packaging, emerging
[01:13:25] technologies.
[01:13:27] Um and so that if you want to go back 75
[01:13:31] years, that's how we came from our
[01:13:33] roots.
[01:13:35] >> And if you remember the name CPMT,
[01:13:40] component packaging manufacturing
[01:13:43] society that's that
[01:13:47] >> Yeah. And Bill, I'm going to do one
[01:13:49] better. CHMT before CPMT and that was
[01:13:53] components hybrids in manufacturing
[01:13:56] technology because 75 years ago leading
[01:13:59] edge technology were all discretetss and
[01:14:02] passives and it as we've gone through 75
[01:14:07] years of evolution hybrids were
[01:14:09] important and then components and now we
[01:14:12] have electronic components technology
[01:14:15] conference that covers packaging at all
[01:14:17] levels. These are great questions. I'm
[01:14:19] sorry I'm I'm a minute over and they're
[01:14:22] going to take us off with hooks. Thank
[01:14:24] you so much for attending. Please thank
[01:14:26] our panelists for their vision.
[01:14:34] Feel free to address any of these uh
[01:14:36] one-on-one with your questions about
[01:14:38] history in the future. Thanks so much.