This may not look like much but it took a lot more evenings to compile than the lists of ten best and ten worst decisions.
74 Series TTL
741
709
16R6
1103
4004
1702
TMS32010
ARM1
28F256
NEXT WEEK: Ten best CEOs
« Win-Win Deal For Infineon and India | Main | You're having a laugh, aren't you Paul? »
This may not look like much but it took a lot more evenings to compile than the lists of ten best and ten worst decisions.
74 Series TTL
741
709
16R6
1103
4004
1702
TMS32010
ARM1
28F256
NEXT WEEK: Ten best CEOs
TrackBack URL for this entry:
http://www.electronicsweekly.com/cgi-bin/mt/mt-tb.cgi/5607
This page contains a single entry from the blog posted on April 4, 2007 8:03 AM.
The previous post in this blog was Win-Win Deal For Infineon and India.
The next post in this blog is You're having a laugh, aren't you Paul?.
Many more can be found on the main index page or by looking through the archives.
Subscribe to this blog's feed
Comments (20)
David - I know I'm professionally fixated on this stuff, but are you going to tell us who first made each of these chips?
Peter
Posted by Peter van der Sluijs | April 4, 2007 10:18 AM
Posted on April 4, 2007 10:18
Peter, I assumed you knew.
The manufacturers are: TI, Fairchild, Fairchild, Monolithic Memories, Intel, Intel, Intel, TI, ARM, Intel.
The chips are: discrete logic, op amp (Dave Fullagar), op amp (Bob Widlar), bipolar PAL (programmable logic), DRAM, Microprocessor (Ted Hoff), EPROM (Dov Frohmann), DSP, Microprocessor (Steve Furber and Sophie Wilson), Flash memory.
all the best
David
Posted by David Manners | April 4, 2007 11:10 AM
Posted on April 4, 2007 11:10
Where does the 555 figure, cornerstone of many a high school project and introduction to the industry for students?
Posted by Stephen Waddington | April 4, 2007 3:46 PM
Posted on April 4, 2007 15:46
Funnily enough the 555 was in, then out, of the list. It was basically a toss-up between the 555 and the 28F256 for 10th place. Bushy and I argued over this, and eventually came to the conclusion that the 28F256 was the more important because it was the start of the flash business which has become huge.
But it could have gone either way. Thanks for saying it.
Posted by David Manners | April 4, 2007 5:09 PM
Posted on April 4, 2007 17:09
Not a single ADC or DAC, modem or codec, UART or serial comms device?
How do you think all these digital chips communicate with the analogue world in real applications, as if by magic?
The world nowadays is driven by communications and interfaces to audio, video, serial data, but there isn't a single chip listed in any of these areas.
It's no use being able to process data or signals if you can't get the inputs in from and the results out to a human being, or send them to another person...
Ian
Posted by Ian Dedic | April 5, 2007 9:12 AM
Posted on April 5, 2007 09:12
How can you leave out the 555 timer chip and its variants and successors? Must be one of the biggest sellers of all time! Signetics first produced it about 35 years ago, and almost every other manufacturer has since. It's so versatile, there are hundreds of applications for it. I once used it in an SMPS when a suitable purpose-built device wasn't available.
John
Posted by John Roscoe | April 5, 2007 9:40 AM
Posted on April 5, 2007 09:40
David, it was VLSI Technology rather than ARM or Acorn that actually built the first ARM1.
Best wishes
Robin
Posted by robin saxby | April 5, 2007 9:45 AM
Posted on April 5, 2007 09:45
The ARM1 was made by Acorn and not ARM.
Posted by Anonymous | April 5, 2007 10:02 AM
Posted on April 5, 2007 10:02
David,
Hard to beat the INMOS transputer because of its influence on other microprocessors, parallel processing in other chips and spin-out companies.
Chris Ryan
Posted by Chris Ryan | April 5, 2007 10:39 AM
Posted on April 5, 2007 10:39
Robin,
Surely Steve Furber and Sophie Wilson were employed by Acorn at the time the ARM1 was built.
All the best
David
Posted by David Manners | April 5, 2007 10:47 AM
Posted on April 5, 2007 10:47
Chris,
You are absoutely right about the Transputer but it's only now, about 30 years after it was launched, that it's influence is beginning to be felt with other discrete multi-core microprocessors coming onto the market.
Best wishes
David
Posted by David Manners | April 5, 2007 10:56 AM
Posted on April 5, 2007 10:56
Anonymous
You're right, and my recollection is that it was called Acorn Risc Machine (ARM).
Best wishes
David
Posted by David Manners | April 5, 2007 10:59 AM
Posted on April 5, 2007 10:59
Interestingly, the 555 timer was designed by an outside consultant for Signetics (I went to the original design review, and later worked for the consultant company, where a seat-belt ASIC design was replaced by the 555!).
I see some omissions there also. How about CMOS logic gates (an early RCA device would do), and some early ADCs. My vote would go for the (old) Intersil parts ICL7106&7, which established the digital DVM market (one second-source part even included a mask bug needed for an early fix!), but I am biased, since I contributed the data sheet and applications for this part. Also, how about some programmable logic device?.
Maybe you need 15 again!
Posted by Peter B | April 10, 2007 9:45 PM
Posted on April 10, 2007 21:45
Ian, Was any ADC more important than the 709?
Best wishes
David
Posted by David Manners | April 11, 2007 11:18 AM
Posted on April 11, 2007 11:18
John,
As I pointed out to Stephen Waddington (above)it was a toss up whether to go for the 555 or the 28F256
Best wishes
David
Posted by David Manners | April 11, 2007 11:20 AM
Posted on April 11, 2007 11:20
Peter,
You're right about needing 15!
There is actually a programmable logic device in the top ten list - Monolithic Memories' 16R6 bipolar PAL.
Best wishes
David
Posted by David Manners | April 11, 2007 11:22 AM
Posted on April 11, 2007 11:22
Your list seems to be in rough chronological order.
If that was your intent, the 709 preceeded the 741 ... which was basically a compensated 709 op-amp. Perhaps one or the other could be replaced by the 555 and thus address that issue.
Otheriwse, I agree totally with your choices.
Thanks for the "memories".
Stan
Posted by Stan Schiller | April 18, 2007 8:40 PM
Posted on April 18, 2007 20:40
David asked "Was any ADC more important than the 709?"
I was pointing out that you had 8 digital chips/families and 2 analogue circuits (both of which were similar opamps) but nothing in between.
Without picking out any specific examples, I would have said that single-chip A-law/u-law codecs (electronic telephones) or UARTs (serial interfaces) should have replaced one of them -- these might not appear as headline IC developments, but have had a huge impact on the way that the world works.
It does seem that 10 chips isn't enough, there are more fundamental categories than this, all of which have changed the world.
Posted by Ian Dedic | May 1, 2007 1:25 PM
Posted on May 1, 2007 13:25
I thought that the 555 - or it may have been the CMOS variant? - was designed out of hours, and quietly fabbed on a spare bit of a die actually devoted to another design. There was certainly something unorthodox in its genesis - perhaps Peter B can put me right.
I'd also like to propose the chips that powered the 8-bit revolution - the GE AY38910, the AMD 7910 modem, 765 floppy controller, 6845... they made everything possible.
Posted by Rupert Goodwins | June 14, 2007 4:53 PM
Posted on June 14, 2007 16:53
I must say i hadn't heard that story about the 555 though it's normal to use second hand silicon in the analogue business.
And you're right about those chips but for a top 10 slot they have to compete with the first DRAM, first flash memory, first microprocessor etc which spawned big businesses.
Posted by david manners | June 17, 2007 10:47 AM
Posted on June 17, 2007 10:47