5 ms·

The number of large semiconductor acquisitions in the past year has been pretty insane... just further consolidation with very few new semi startups coming out.

rpcope1 11y ago
It's far easier to build a start-up doing software than pretty much any hardware, so current giants have a huge advantage over anyone trying to break in. Consider for a moment that a new fab with a recent process is probably a many billion dollar investment, and that even fabless semiconductor companies (a la LSI, recently) there are huge amounts of IP, capital, and brain power tied up; everything also really needs to be really correct up front, and tooling is expensive (you wouldn't want to waste millions of dollars building a chip that's really unproven and undesired from a market perspective). Even technologies with less expensive (though not at all cheap) tooling, like hard disk drives, the IP is so tied up with the current giants (WD/HGST/Seagate), and the firmware needed to make the drives work at current capacities is so complex, that it is essentially impossible to break into the market.
- zokier 11y ago
  On the other hand the company in question here (Broadcom) was started by essentially two guys in a garage in early to mid-nineties.
  - geomark 11y ago
    That's the popular legend. But in reality one guy was a professor at UCLA and the other a well paid engineer at TRW. To get started with early development of much of their IP they used considerable resources available to them at their employers, which included funding from government IR&D projects and chip design engineers who were on TRW's payroll. So hardly a garage startup.
    - zokier 11y ago
      Why would TRW have given away such resources and IP to what would become Broadcom?
- trsohmers 11y ago
  As someone starting a fabless semiconductor startup on less than $2 million, I completely agree with you when it comes to starting a fab (Intel's, GF's, and TSMC's new 14-16nm fabs are in the $10 to $15 billion range), but a fabless semiconductor company can be done for the same range as a software startup. While it is not fair to use my startup as an example (as we don't have silicon in hand just yet), you can look here: http://www.adapteva.com/andreas-blog/semiconductor-economics-101/ http://www.adapteva.com/andreas-blog/semiconductor-economics... http://www.adapteva.com/white-papers/a-lean-fabless-semiconductor-startup-model/ http://www.adapteva.com/white-papers/a-lean-fabless-semicond...
  - rpcope1 11y ago
    Wow, this looks like a really interesting read. Thanks for sharing.
  - minthd 11y ago
    It could be that the adapteva model(which talks about a highly focused design) fits a small number of start-ups, that are as a generalization, less appealing to VC's due to smaller markets and returns, and most other companies require lots of funding - say $30+ million ?
qwerty-- 11y ago
Capitalism at work. Collusion, consolidation and no competition.
- adventured 11y ago
  So what existed prior to the consolidation then?
L_Rahman 11y ago
The only notable recent entrant I can think of is Spreadtrum from China. But I don't know that it counts as a startup when it is backed and partially owned by the Chinese government.
silverpikezero 11y ago
You should expect this trend to continue. Semiconductors have become a hideously expensive business, aggravated by fewer fabs and more draconian fab/IP acquisition costs. It's a well known that nobody on Sand Hill road will touch semiconductor startups at all, leading to a stagnant collection of semiconductor companies founded 20+ years ago. There was recently announced a semi-only incubator here in SV[1] because nobody can raise any traditional VC money, although time will tell how many successful exits it will have. [1] http://www.eetimes.com/document.asp?doc_id=1326501 http://www.eetimes.com/document.asp?doc_id=1326501
- trsohmers 11y ago
  There are quite a few people interested in semiconductors on Sand Hill Road (have talked with many of them), but you are right that they are very cautious. Thankfully, you should be hearing about a well known VC funding a semiconductor startup in the near future. Silicon Catalyst is great, and I look forward to silicon coming back to Silicon Valley.
- bsder 11y ago
  > although time will tell how many successful exits it will have. And that's exactly the problem. Semiconductors has no exit. Atheros was probably the last company to go from startup to big hit, and look how long it took them to exit. Why will a VC throw money at a company that has a 5 year exit, at best, when they can throw a lot less at stupid 20-somethings, flog them like slaves, and sell to Yahooglezonsoft.
  - loosescrews 11y ago
    What about SandForce? They were founded in 2006 (vs 1998) and acquired in 2012 (vs 2011).
- Keyframe 11y ago
  This is such a shame. I think,well I want to believe there os still a lot of things that can be done in that space, both in ip only dev and processing tech. It would be so nice to see a startup doing startupy thing in that space, either ip core, chemical processing or similar.

rpcope1 11y ago

It's far easier to build a start-up doing software than pretty much any hardware, so current giants have a huge advantage over anyone trying to break in. Consider for a moment that a new fab with a recent process is probably a many billion dollar investment, and that even fabless semiconductor companies (a la LSI, recently) there are huge amounts of IP, capital, and brain power tied up; everything also really needs to be really correct up front, and tooling is expensive (you wouldn't want to waste millions of dollars building a chip that's really unproven and undesired from a market perspective). Even technologies with less expensive (though not at all cheap) tooling, like hard disk drives, the IP is so tied up with the current giants (WD/HGST/Seagate), and the firmware needed to make the drives work at current capacities is so complex, that it is essentially impossible to break into the market.

zokier 11y ago
On the other hand the company in question here (Broadcom) was started by essentially two guys in a garage in early to mid-nineties.
- geomark 11y ago
  That's the popular legend. But in reality one guy was a professor at UCLA and the other a well paid engineer at TRW. To get started with early development of much of their IP they used considerable resources available to them at their employers, which included funding from government IR&D projects and chip design engineers who were on TRW's payroll. So hardly a garage startup.
  - zokier 11y ago
    Why would TRW have given away such resources and IP to what would become Broadcom?
trsohmers 11y ago
As someone starting a fabless semiconductor startup on less than $2 million, I completely agree with you when it comes to starting a fab (Intel's, GF's, and TSMC's new 14-16nm fabs are in the $10 to $15 billion range), but a fabless semiconductor company can be done for the same range as a software startup. While it is not fair to use my startup as an example (as we don't have silicon in hand just yet), you can look here: http://www.adapteva.com/andreas-blog/semiconductor-economics-101/ http://www.adapteva.com/andreas-blog/semiconductor-economics... http://www.adapteva.com/white-papers/a-lean-fabless-semiconductor-startup-model/ http://www.adapteva.com/white-papers/a-lean-fabless-semicond...
- rpcope1 11y ago
  Wow, this looks like a really interesting read. Thanks for sharing.
- minthd 11y ago
  It could be that the adapteva model(which talks about a highly focused design) fits a small number of start-ups, that are as a generalization, less appealing to VC's due to smaller markets and returns, and most other companies require lots of funding - say $30+ million ?

qwerty-- 11y ago

Capitalism at work. Collusion, consolidation and no competition.

adventured 11y ago
So what existed prior to the consolidation then?

L_Rahman 11y ago

The only notable recent entrant I can think of is Spreadtrum from China. But I don't know that it counts as a startup when it is backed and partially owned by the Chinese government.

silverpikezero 11y ago

You should expect this trend to continue. Semiconductors have become a hideously expensive business, aggravated by fewer fabs and more draconian fab/IP acquisition costs. It's a well known that nobody on Sand Hill road will touch semiconductor startups at all, leading to a stagnant collection of semiconductor companies founded 20+ years ago. There was recently announced a semi-only incubator here in SV[1] because nobody can raise any traditional VC money, although time will tell how many successful exits it will have. [1] http://www.eetimes.com/document.asp?doc_id=1326501 http://www.eetimes.com/document.asp?doc_id=1326501

trsohmers 11y ago
There are quite a few people interested in semiconductors on Sand Hill Road (have talked with many of them), but you are right that they are very cautious. Thankfully, you should be hearing about a well known VC funding a semiconductor startup in the near future. Silicon Catalyst is great, and I look forward to silicon coming back to Silicon Valley.
bsder 11y ago
> although time will tell how many successful exits it will have. And that's exactly the problem. Semiconductors has no exit. Atheros was probably the last company to go from startup to big hit, and look how long it took them to exit. Why will a VC throw money at a company that has a 5 year exit, at best, when they can throw a lot less at stupid 20-somethings, flog them like slaves, and sell to Yahooglezonsoft.
- loosescrews 11y ago
  What about SandForce? They were founded in 2006 (vs 1998) and acquired in 2012 (vs 2011).
Keyframe 11y ago
This is such a shame. I think,well I want to believe there os still a lot of things that can be done in that space, both in ip only dev and processing tech. It would be so nice to see a startup doing startupy thing in that space, either ip core, chemical processing or similar.