Podcast: Play in new window | Download
Subscribe: Apple Podcasts | RSS
Welcome, Vincent Himpe of Silicon Valley Garage!
- Belgian kits – Velleman, German kits – Operaman
- Won Rotary Club contest for 2 month, got an internship in Finland with Salcomp/Nokia
- D2MAC chip – Analog HD chip – 1989
- Vincent says he wasn’t for school, ended up helping people with their senior projects.
- Had to do mandatory military service, started a job as a maintenance tech at a fab in Oudenaarde around 1993. OnSemi owns the facility today.
- Vincent’s first online name was Ping00751. He had a well known document about the fledgling protocol for i2c.
- The company he was working for was absorbed by Alcatel (later acquired by Lucent).
- Team of 7, invented DSL/ADSL
- The early boards used a very heavy duty I960 CPU and a bunch of ASICs to squeeze out performance from a 16 MHz clock.
- The systems also used Indeo video, which had a whopping 320×240 resolution.
- The ADC had an ~17.4MHz clock with 16 bit ADC output with a 4 bit opcode
- Xilinx FPGA 95 – learned verilog, the program wasn’t called ISE (Exemplar logic)
- Started using Altera (Max Plus) -> later Quartus2
- AstroPhysicist starts at the lab
- At Alcatel in 2000, there were 216,000 people working there
- Started using Protel/Altium in 93 (Dave did in 89)
- Bought a pick and place for the lab
- After all the layoffs in early 2000s/dot com crash, employment at Alcatel was down to 40,000
- Sold the fab/design team to ST Micro
- Exar chip 2206, designed to be used with 2211. Similar to MAX038. Both are obsolete not, yet people still clamor for them.
- Rochester Electronics could keep making them, but likely won’t unless you’re a military customer.
- Vincent bought 6 different PM2811’s off eBay and were sold ot someone for their firmware. This allowed the buyer to not need to recertify his test rig.
- Chris referenced an article on artificial intelligence, how the basis is usually grounded in patern recogintion based on past experience.
- “Signature DMMs” will give you a coded output for possible failure modes.
- Vincent doesn’t use Magic/IRSIM, nor is he a fan of KiCad.
- The 6502/8051 is used because…it’s royalty free.
Though we couldn’t possibly capture ALL of the awesomeness that Vincent talked about on the show, the audio file can! Be sure to take the time to listen to all 3+ hours!
Robert McNamara says
Three hours!! Cannot wait to listen to this. Listened to them all back to back from day one over the last 4 weeks and now eagerly await them every week..
Phil Macphail says
Can’t agree with the Bluetooth comments – I have a BT keyboard and mouse that last months between battery changes, and are several years old. BTLE offers even longer lifetimes, and streaming audio to satellite speakers has been working for some time (but the Apple BT implementation is incomplete, as noted in the show).
The comments that individual networks reduce the throughput of neighbouring networks is also misleading, as BT frequency-hops specifically to avoid this.
Hi Dave and Chris
This is you’r best episode ever!
Keep up the fantastic work you both do.
Chris Gammell says
Heh, we hardly spoke! Guess we’ll have to have Vincent back more often!
Ronald Lijs says
Good program, enjoyed it and keep them coming!
Eric Wasatonic says
I’m the one who asked about the XR2206 on the forum: http://www.eevblog.com/forum/projects/what-happened-to-xr-2206-funct-gen-ic-and-xr-2212-precision-pll/ . In my defense, before asking the forum, I knew nothing about the 2206’s history. No one here knew they had gone obsolete until I tried to buy more. It was simply a convenient chip for the students to use for the lab experiments. We have enough to last us through this semester, but starting next next year, we will certainly redesign the lab procedures using more modern parts.
Thank you, Vincent and Forum
– Eric Wasatonic
Electrical Lab Supervisor
Penn State Harrisburg
Mark J. Blair says
I just listened to this episode while I’m pushing polygons around on a PCB design at work, and it was fascinating! It certainly is hard to use some of the free (or even less expensive) EDA tools after getting spoiled by the pro-grade tools at work.
I’m curious about the specific free 8051 core in Verilog that Vincent mentioned near the end of the episode. I’d appreciate it if you could share a link to it.
David Bley says
Really enjoyed this podcast. They keep getting better and better. I have not had anymore problem with KiCAD than I have had with Protel/Altium/Tango/PCAD, FutureNET Dash PCB, EE Designer or any of the other PCB CAD software that I have used. They all have problems. I can get the job done with KiCAD and the price is right.
Vincent himpe says
A couple of replies: The 8051 core is from http://www.oreganosystems.at.
Bluetooth. All those profiles were there fromt he start. Some are only now being used, 15 years after design. Some are deliberately crippled. Keyboard and mouse run now for a few months with the newer ower profiles. The original profiles did barely a few weeks. But a 27Mhz or 435MHz band keyboard and mouse can run years ! The problem is that you cant join and ujoin a bluetooth node fast.
Vincent himpe says
An ism band keyboard sends the keystroke and sleeps inbetween. The bluetooth cant do that. Even typing fast the few microseconds the transceiver is used is a very low duty cycle compared to deadtime between keystrokes. This is in the benefit of ism band keyboards. The bluetooth keyboard can’t shutdown for such short periods. Unjoin and rejoin take seconds ! You would be missing lots of keystrokes.
Vincent himpe says
@eric. No need to defend yourself. If you don’t know it went obsolote then that is that. You can’t know everything. I brought this up for a different reason : why is the course still done with this old relic in the first place ? Doesn’t the course evolve ? I would assume that, a teacher with a genuine interest in the quality of his course, would keep up with the evolution and review the course content from time to time. In that case he would have seen10 years ago it was time to adapt…. One of my frustrations is that the content of many courses is stagnant… Which is bad for innovation and the quality of education
Eric Wasatonic says
Some teachers are more familiar with the unchanging theory than they are with the ever-changing practice. In this case, the teacher started here in 2004 after the retirement of the previous electromagnetics teacher who taught the course using the Laboratory Manual for Electronic Communications Systems: Fundamentals through Advanced, by Wayne Tomasi, 1994. The new electromagnetics teacher continued using the manual and the experiments within.
I believe he initially wanted to have the students use some more expensive equipment, but we did not have the budget for it. So, over the years, both he and I have become comfortable with the Tomasi lab manual. We never thought anything was wrong with the lab experiments (all done on breadboard with basic equipment) because the focus of the course is to learn the concepts – filters, VCO, PLL, AM/FM modulation, etc., regardless of the specific components the students use.
Mark J. Blair says
Thank you very much! That 8051 core is giving me some new ideas for a hobby project I’m working on.
I did fell asleep! But came back today to finish listening 😛
Ioannis Andrianakis says
Greatest episode ever! Kudos Vincent! Thank you David and Chris!
Best episode ever! Its taken 3 days to listen to all of it, but Vincent’s stories were enchanting.
This was awesome! Thanks Chris, Dave and Vincent, great content.
Guillermo A. Amaral (@gamaral) says
Yep, best episode ever. I hope you guys bring back Vince (even if his open-source software stance is a bit of a bummer). 😉
Great stories Vincent! Very interesting career. Thanks for sharing it on the Amp Hour.
Mike Perigo (@Retrophile) says
Absolutely Fantastic Episode!
I did have the advantage of living through the eras covered and being involved in the home and military electronic fields so understood all the acronyms, products and other references. As a result, every word was a delight.
I am however disappointed that it was so short. 🙂 I started listening without realising how long the episode was going to be (it should have had a warning at the start) and I only realised how much time had passed when Chris called time on it. It was obvious that Vincent still had lots more to recount and questions to answer.
Many thanks to Vincent and you guys for spending the time and sharing it with us. I look forward with great anticipation in the hope of a part 2 … part 3 etc. At the very least Vincent should be considered as a candidate next time either of you need a stand-in.
Yes! Fantastic show! Do more! This show and the one with the guy doing Power meters had great stories. Vincent has made me wonder which box in my garage has the 8051 and other boards bought on ebay and never used, i gotta dig them out and get to work.
One of you should wrestle Vincent to the ground and implant a chip in his head so to stream his thought processes to the web so we can hear more….LOL…make it a bluetooth chip just to see what happens….Thanks!
Loved it. It’s long… but never got dull. Also, listeners have the power to skip and leave, so don’t worry about “making” people listen. It sounded like you guys had to run, but don’t be afraid of lengthy content when it’s flowing so well.
Look forward to having Vincent again.
The One True Stickman says
Here’s the Solder printer Vincent predicted:
Jelle Haandrikman (@jhaand) says
Was a great show. Finally got around to finish it. Now on with the next 2 shows.
YES! I had kinda given up on your podcast, I got tired of fast forwarding and digging for any talk about electronics, but holy crap this one was awesome, and so was the Forrest Mims one, I’m back on-board!