Chip technology and Pakistan

[Falcon26]

@Bilal Khan (Quwa) i have heard you speak about semiconductor technology and the need for Pakistan to invest in such technologies before. I was recently reading about this company Open Silicon which was founded by a Pakistani technologist. He recently founded SiFive a chip design company focusing on RISC-5 technology. In fact, his company includes the pioneers and inventors of this technology. His company has presence in China, the US and South Korea. What role do you see for companies like these towards of making Pakistan a player in semiconductors? Would love to read your thoughts @Riz

https://www.sifive.com/

 

[khansaheeb]

All organisations evolve in similar ways and encounter similar obstacles whether they be in New York, Karachi, London or Silicon Valley. The difference is risks are encouraged and rewarded in the infant successful companies and supported by external consultants and financiers who gain financially by their investment and contribution. Its a whole chain and network that is the fertile ground for success; nurtured by team work, watered by common goals and grown by education to a flower of success.

 

[Falcon26]

Here’s another player in the semiconductor business led by Pakistani-American entrepreneurs

https://www.whizzsystems.com/

Cavium was started by a Pakistani-American and was acquired by Marvell for $6.5 billion

 

[ps3linux]

Micro-controllers are the workhorse of modern electronics from microwave to automobiles to defense applications from 8 bit to 32 bit. After that microprocessors/SoC/CPUs. Prior ones are easier to source cheaper to develop into an end application but they lack the speed and many a times memory. The later ones are much better but restrictive due to the fact that most of the cases one has to purchase proprietary development environment. An alternate is FPGA but for the powerful ones there is export ban by USA.

That leaves another option custom processors, like the ones used in Playstations and XBOX (there are others but I have taken these names just for reference) but that's a very expensive alternate.

RISC-V being an open architecture is much better option. There have been a few endeavors in developing RISC-V based human capital and solutions locally in Pakistan, both in the civilian and defense circles.

As the business environment and economy of Pakistan improves we will see companies working in this area.

COVID-19 has changed the world significantly in ways no one ever considered, there are lessons first and foremost is self reliance not just for Pakistan but rest of the world, how much of these lessons result in some concrete action plan is yet to be seen, lets see.

 

[Falcon26]

Micro-controllers are the workhorse of modern electronics from microwave to automobiles to defense applications from 8 bit to 32 bit. After that microprocessors/SoC/CPUs. Prior ones are easier to source cheaper to develop into an end application but they lack the speed and many a times memory. The later ones are much better but restrictive due to the fact that most of the cases one has to purchase proprietary development environment. An alternate is FPGA but for the powerful ones there is export ban by USA.

That leaves another option custom processors, like the ones used in Playstations and XBOX (there are others but I have taken these names just for reference) but that's a very expensive alternate.

RISC-V being an open architecture is much better option. There have been a few endeavors in developing RISC-V based human capital and solutions locally in Pakistan, both in the civilian and defense circles.

As the business environment and economy of Pakistan improves we will see companies working in this area.

COVID-19 has changed the world significantly in ways no one ever considered, there are lessons first and foremost is self reliance not just for Pakistan but rest of the world, how much of these lessons result in some concrete action plan is yet to be seen, lets see.

Thanks for the post. I find it incredible that a company led by a Pakistani is the market leader in this technology. The current Pakistani government deserves credit for attracting SiFive to Pakistan.

 

[Paul2]

I did research the topic in the past. To my big surprise, I found out that Pakistan had a working backend fab in late eighties, ran by a company in Lahore.

Unfortunately, the train has since departed. Catching up to the leaders now is not realistic, not only for Pakistan. Even China struggles, as I've wrote many times in our section.

But, even if it's impossible to meet the competition head-on, it doesn't mean that microelectronics industry as a whole is off reach to you.

Few points:

1. I find Pakistani engineering education to be quite ok, old fashioned, but ok. In comparison to workers I saw when I worked in Canada. Pakistani engineering students are way more prepared to do fieldwork. I only had less than 2 months in the country during the assignment, and worked with few B.Eng. level students whom the client was hiring.
   In Canada, if I were to ask a student intern to solve even a very simple circuit, they will either have to pull their laptops to do a SPICE simulation, or they wouldn't be able to do it. When I had my Pakistani client's engineer to do something similar for a valve actuator circuit. They guy sat quietly for a few minutes, and did it with a pen and paper.
   I think, this is the same for most of my generation of engineers in China. People who studied a decade ago, had no luxury to work a job sitting behind a computer screen.
   
2. Will your government put emphasis on the "brainwork," or the "handwork?" This is an important point that is very often misunderstood.
   
   In Shenzhen, we had a lot of discussions with the municipality's industry section on whether the "brainpower," or "manpower" is more worth efforts to preserve in the city.
   
   What it came to was an understanding that high net worth professionals don't come to the city just for giggles, the come to the city because there is a lot of hands to man the assembly lines. If there will be nobody to build the products they develop, they will have no material incentive to stick around.
   
3. It's better to start with just anything, than do nothing. Even if the fabrication business is now outside of your reach, there should still be at least some niches available if you search for them thoroughly. Take Malaysia as an example, now they are not a contender in the fabrication space, but they managed to setup themselves well in the packaging business, and are a biggest packager around after Taiwan.
   
   There should be some openings life that for you too.

 

[Bilal Khan (Quwa)]

I did research the topic in the past. To my big surprise, I found out that Pakistan had a working backend fab in late eighties, ran by a company in Lahore.

Unfortunately, the train has since departed. Catching up to the leaders now is not realistic, not only for Pakistan. Even China struggles, as I've wrote many times in our section.

But, even if it's impossible to meet the competition head-on, it doesn't mean that microelectronics industry as a whole is off reach to you.

Few points:

1. I find Pakistani engineering education to be quite ok, old fashioned, but ok. In comparison to workers I saw when I worked in Canada. Pakistani engineering students are way more prepared to do fieldwork. I only had less than 2 months in the country during the assignment, and worked with few B.Eng. level students whom the client was hiring.
   In Canada, if I were to ask a student intern to solve even a very simple circuit, they will either have to pull their laptops to do a SPICE simulation, or they wouldn't be able to do it. When I had my Pakistani client's engineer to do something similar for a valve actuator circuit. They guy sat quietly for a few minutes, and did it with a pen and paper.
   I think, this is the same for most of my generation of engineers in China. People who studied a decade ago, had no luxury to work a job sitting behind a computer screen.
   
2. Will your government put emphasis on the "brainwork," or the "handwork?" This is an important point that is very often misunderstood.
   
   In Shenzhen, we had a lot of discussions with the municipality's industry section on whether the "brainpower," or "manpower" is more worth efforts to preserve in the city.
   
   What it came to was an understanding that high net worth professionals don't come to the city just for giggles, the come to the city because there is a lot of hands to man the assembly lines. If there will be nobody to build the products they develop, they will have no material incentive to stick around.
   
3. It's better to start with just anything, than do nothing. Even if the fabrication business is now outside of your reach, there should still be at least some niches available if you search for them thoroughly. Take Malaysia as an example, now they are not a contender in the fabrication space, but they managed to setup themselves well in the packaging business, and are a biggest packager around after Taiwan.
   
   There should be some openings life that for you too.

I agree. Electronics is a constantly expanding field, so you don't need to master the "root" areas (e.g., semiconductors) to actually succeed in the field.

There are many areas that build upon semiconductors right up to applications that Pakistan can try building expertise on and exporting (like Malaysia).

Absolute turnkey independence isn't feasible for any country but the US and China, the key for Pakistan is to figure how it can become indispensable in shared supply chains with as many countries as possible.

 

[Samlee]

I did research the topic in the past. To my big surprise, I found out that Pakistan had a working backend fab in late eighties, ran by a company in Lahore.

Unfortunately, the train has since departed. Catching up to the leaders now is not realistic, not only for Pakistan. Even China struggles, as I've wrote many times in our section.

But, even if it's impossible to meet the competition head-on, it doesn't mean that microelectronics industry as a whole is off reach to you.

Few points:

1. I find Pakistani engineering education to be quite ok, old fashioned, but ok. In comparison to workers I saw when I worked in Canada. Pakistani engineering students are way more prepared to do fieldwork. I only had less than 2 months in the country during the assignment, and worked with few B.Eng. level students whom the client was hiring.
   In Canada, if I were to ask a student intern to solve even a very simple circuit, they will either have to pull their laptops to do a SPICE simulation, or they wouldn't be able to do it. When I had my Pakistani client's engineer to do something similar for a valve actuator circuit. They guy sat quietly for a few minutes, and did it with a pen and paper.
   I think, this is the same for most of my generation of engineers in China. People who studied a decade ago, had no luxury to work a job sitting behind a computer screen.
   
2. Will your government put emphasis on the "brainwork," or the "handwork?" This is an important point that is very often misunderstood.
   
   In Shenzhen, we had a lot of discussions with the municipality's industry section on whether the "brainpower," or "manpower" is more worth efforts to preserve in the city.
   
   What it came to was an understanding that high net worth professionals don't come to the city just for giggles, the come to the city because there is a lot of hands to man the assembly lines. If there will be nobody to build the products they develop, they will have no material incentive to stick around.
   
3. It's better to start with just anything, than do nothing. Even if the fabrication business is now outside of your reach, there should still be at least some niches available if you search for them thoroughly. Take Malaysia as an example, now they are not a contender in the fabrication space, but they managed to setup themselves well in the packaging business, and are a biggest packager around after Taiwan.
   
   There should be some openings life that for you too.

Yes There Was A Company Called MEI I Think It Meant Microelectronics International

 

[S.Y.A]

There is a company, Lampro Mellon, has recently set up their office in lahore, they are funded by sifive i think. they intend to train pakistanis in asic design and provide fabless design services to china etc. they hired a lot of people recently for this purpose. hope they open offices all over Pakistan. also, another company just started a setup in khi, its by a guy from usa, they intend to design asics for networking, smartnics (forgot the company's name).

EDIT: its Sahil Semiconductors

hope these initiatives succeed, and other people follow in their footsteps. much like the IT industry. it wont be something major, but it still would be something in the right direction.

 

[Paul2]

Yes There Was A Company Called MEI I Think It Meant Microelectronics International

Yes, Philips was their partner for ELMAC venture.

But Philips went to Shenzhen by early nineties, when they gave up on Pakistan.

Philips was one of the biggest employer in early Shenzhen, and they were the reason SEG got the semi tech.

As it's said, the rest is history.

Question about Semiconductor Industry in Pakistan.​

936 views
Subscribe







Zaka U. Bhatti​

03/10/1991
to

Fellow Netters:

Does anybody know about the Semiconductor Industry in Pakistan?
For example names of companies, their products and their research
goals. This includes companies that do IC level circuit design, CAD
development, component level circuit design or IC fabrication.
Please respond to me by electronic mail. I will compile the
responses and post it on this group at a later date.
Zaka-=-

Tahir Ghani​

03/10/1991
to
In article <1991Oct3.1...@ee.eng.ohio-state.edu> za...@ee.eng.ohio-state.edu (Zaka U. Bhatti) writes:
>
> Fellow Netters:
>
> Does anybody know about the Semiconductor Industry in Pakistan?
>For example names of companies, their products and their research
>goals. This includes companies that do IC level circuit design, CAD
>development, component level circuit design or IC fabrication.
>
>Zaka-=-
I know of a company ( back in 1985 ) which use to do back end
processing ( mainly packaging ) of IC's. I have forgotten the name
but it was located in Lahore. I am not sure whether it is presently
in operation or not. As for the companies doing IC level circuit
design, CAD develpoment or IC fabrication, there are none, zilch,
zero!!
Most US Semiconductor companies prefer to do back end IC processing
( which inculdes testing and packaging of IC's ) in Singapore,
Malasyia, Thialand, Phillipenes and other east Asian countries because
the process is labour intensive. Pakistan should also be attractive
in this regard, but it needs to be politically stable and also needs
foresight on the part of our leadership.
The institutes such as NIE ( National Institute of Electronics ) and
Silicon Research Institute, which have been set up by the federal govt.
for the promotion of electronics/computers, provide nothing more than
lip service. The chairman of one of these institutes happen to visit
Stanford a few years back and indicated that he was more interested
in the GaAs technology rather than Silicon because of his perception
about GaAs being "the state of art", without having any understanding
of the real issues. I would love to hear the views of Pakistani EEs
who have worked at places such as NIE, SUPARCO etc.
Regards,
Tahir

SAFWAN SHAH​

04/10/1991
to
In article <1991Oct3.2...@EE.Stanford.EDU> gh...@EE.Stanford.EDU (Tahir Ghani) writes:
>In article <1991Oct3.1...@ee.eng.ohio-state.edu> za...@ee.eng.ohio-state.edu (Zaka U. Bhatti) writes:
>>
>> Does anybody know about the Semiconductor Industry in Pakistan?
***some stuff deleted*** >>Zaka-=-

>
>I know of a company ( back in 1985 ) which use to do back end
>processing ( mainly packaging ) of IC's. I have forgotten the name
>but it was located in Lahore. I am not sure whether it is presently
>in operation or not. As for the companies doing IC level circuit
>design, CAD develpoment or IC fabrication, there are none, zilch,
>zero!!
***some stuff deleted***
> Tahir
Why there are none, zilch, zero?
This article attempts to generate a discussion on Mr Tahir's statement. It
may lead to some insight into the circumstances which incapacitate a "HI-TECH"
industy in Pakistan. Though he is specifically talking of CAD and IC
development I have slightly augmented the base to include all products which
are state-of-the-art and could help propel Pakistan into the 'Developed'
nation category. Please there is no room for wars and flames!

The name of the company which "packaged" IC's was Micro Electronics
Technology (MET) Inc. It operated from the premises of Micro Electronics
International (Pvt.) Ltd. It was actually a joint venture between a
company based in Fort Lauderdale, Florida and MEI. In 79-80 the
regime of President Zia encouraged a Pakistani in the US to return
to Pakistan and set up a wafer fab plant, since the infrastructure for
such an industry did not exist it was planned that an existing plant set
up by Philips under the name ELMAC would be modernised for this purpose.
At that time ELMAC was not operational (aftermath of the Islamic Bomb
controversy), Philips had come under a lot of pressure for supporting
clandestine development of hi-tech against the COCOM and GATT treaties
in Pakistan and they had scuttled their operations after an aggravated
labour strike. Prior to shut down ELMAC was manufacturing diodes and
transistors and was also in a position to package IC's. It may be
mentioned here that it was because of ELMAC that the government had levied
custom duties of 80 to 100 % on electronic components.
In 1980 MET and MEI became operational with the intention to set up a
Wafer Fab plant, several US companies including the first few ASIC
companies were participants in this venture. Unfortunately at that
very stage the US embassy in Islamabad was attacked by angry crowds and
the US government ordered stoppage of equipment ($ 40 M worth) to
Islamabad. The participating US companies were discouraged and the project
was in jeopardy. Ironically, within a few months the same companies started
construction of a wafer fab plant in China. With the "so called" foreign
help out of the deal MET began to import wafers from US companies in
Phillipines, Singapore, Taiwan etc. 64K RAMS, 74XX series, and many other
simple chips were packaged. The capacity of the plant was 1 Million chips
per week per shift. Obviously this capacity could not be reached because
the investment to buy the required wafers was not there and also the
chip market was in a slump. Tragically the US companies which were selling
wafers were also not doing well and none of the Japanese companies were
ready to sell wafers when they could make more profits by selling chips
themselves. MET dumped over 15 million RAM chips in US through HK traders
under MSJ and MET markings and consequently the US banned export of
several important constituents of the packaging proccess to Pakistan,
particularly a gas which is used to seal the epoxy over the bonded wafers.
So much for this tale of sorrow. MET managed to live by procuring
wafers from the Chinese and packaging some chips for local (inhouse)
needs. MEI the sister company went into the manufacture of HF/VHF
radio equipment with great success. The initial plans and dreams to
make semi custom chips were basically dead for the time being.
In 1985-86 MET/MEI launched another project with KDD (Kokusai Denshi Denwa)
the Japanese Telephone Co. and Harris semiconductor to manufacture
custom IC's for the manufacture of cellular phones in Pakistan. The
plan was to train Pakistani engineers to design gate arrays and with
the support of a Japanese bank the project was launched. Calma, Daisy
and Iris workstations were procured, automatic production lines and SMT
(surface mount tech.) equipment was set up. This time the problem
arose when the State Bank of Pakistan refused to counter guarantee the
Japanese loan (6%) and local banks declined all support on the premise
that they could not use electronic equipment as collateral. Their
reasoning was that if this had been a textile or cement factory the
banks would have no problem because that equipment can be consumed
locally. MET/MEI had already sunk a large sum in the project and the
foreign partners withdrew. The equipment is still in Pakistan and
wasting away. The only remaining hope was that the new digital exchange
would be made in Pakistan and local companies would have a significant
role in its development. This hope was first shattered by PM Junejo when
he proudly announced that this exchange would be built by SIEMENS (FRG)
(it was later cancelled), and then recently the highly qualified
communication minister of Pakistan (Mr Murtaza Jatoi) granted the
project to Ericsson and some other foreign company. The modules will
be transferred to Pakistan semi knocked down and the local role
will be the construction of the premises and large commissions.
Anyway, this was just a fragment of the whole picture based on one
companies experience. There are those who have been successful, but
only if they managed to beat the system. It is still extremely doubtful
if a medium/or large scale hi-tech establishment can make ends meet in
Pakistan. Unless one trades in foreign names (Pagers traded by Digital
Comm.: Cellular Phones by Instaphone and another co.), it is very
difficult to survive, because then you are compelled to
not inflate the project cost as eventual manufacture would reflect that
expense. While simple trading allows you to generate the funds required
to give the ministers their kicbacks.
The above stated tale of woe does not imply hopelesness. It is a
neccessary stage and probably every nation has to go through it before
someone comes and breaks the various barriers. I base my arguments on
two things.
1) Assembly and progressive manufacture is far more essential
for progress then simple trading.
2) Govt. controlled large scale projects, like the
Digital Exchange can provide the impetus to encourage
the Dawoods, Habibs and others to go into manufacture of
basic electronic components (Res, Cap, Diodes, Tr,
Switches etc.). Inexperienced expatriates and local
TV manufacturers and product assemblers cannot do both
make products and also finance the downstream industry.
Until banks and bureaucrats develop a concept 'electronics' this tale
of woe will repeat like the proverbial 'bad penny'
Regards,
Safwan Shah
--
====================================================================
= SAFWAN SHAH =
= Bioserve Space Technologies =

 

[peagle]

Everyone will wake up to this thread now,
or at least I hope lol

 

[ssethii]

There is a company, Lampro Mellon, has recently set up their office in lahore, they are funded by sifive i think. they intend to train pakistanis in asic design and provide fabless design services to china etc. they hired a lot of people recently for this purpose. hope they open offices all over Pakistan. also, another company just started a setup in khi, its by a guy from usa, they intend to design asics for networking, smartnics (forgot the company's name).

EDIT: its Sahil Semiconductors

hope these initiatives succeed, and other people follow in their footsteps. much like the IT industry. it wont be something major, but it still would be something in the right direction.

www.dreambigsemi.com

 

[S.Y.A]

www.dreambigsemi.com

yeah, they renamed it to dream big, it has now expanded to lahore as well.

 

[ssethii]

yeah, they renamed it to dream big, it has now expanded to lahore as well.

coming to isb soon

 

[ABBASIA]

coming to isb soon

We will like further progress on chip designing in Pakistan by establishing more and more companies and training large pool of engineers who design chips for US/Chinese markets and eventually for local Pakistan own designed products. Government should step in more efforts on this.