You both are so humble. You say you just "bum around" when in fact, you are inspiring teachers and communicators that are giving back so much of your high quality, high signal to noise knowledge to the community Thank you.
Re-watching this on the occasion of the 500th (and sadly, last) issue of The Embedded Muse. I am truly humbled by his depth of knowledge and dedication to the craft of engineering. Yes, you too Dave.
Today is the first day my searching fingers introduced me to first to Dave and within three clicks and two hours at retired and 70 years old I digitally met the two guys I wish I had known over coffee my whole life... That said, you're both so much farther down, ( or up? ), the "Rabbit Hole" than my miles of experience in mechanoelectron life. But the tears in my eyes come from joining two minds that see as I see. And that may actually say volumes as to why we delve in to all the disciplines where we are led to the answers our imaginations require us to go! If it were a "choice" , we would choose not to go. But cause and effect coupled to dreaming leave it to inevitable.... Thank you Gentlemen. Christopher Workmon
Very interesting .... ....I've noticed in rf DSP that companies have resorted to using old analog crystal lattice filters in front of the DSP section to relieve pressure on the analog to digital processor .... on very dirty signals....so the beauty of hybridization ....beware of the analog domain ! hahaha The next big wave.......human imbedded computing !!! Just watch !
this is way more than just electronics... a spontanous conversation with a friend, a lesson of passion for work and respect for each other. Thanks for sharing, I really enjoyed it.
@@EEVblog I', Jack's age; so I can relate to almost everything he says!! Most enjoyable! Now a retired EE; I went into optical transmission and networking in the late 70s. Good work for 20+ years. But, some observations: like the declining cost of silicon, transmission costs are 8 - 9 orders of magnitude lower, but the result has not been an improvement in the quality of human interaction. Being honest, ethical, and a hard worker still gives almost no control of how your work will be used. Management usually demanded engineers have no social conscience. Really got a laugh out of Jack's comments, especially about our kids!
As a EE student, I have to say this is some pretty great stuff. At my school they put a decent amount of emphasis on ensuring students can handle coding. One class is dedicated to C programming basics, one class is dedicated to learning digital logic through FPGAs using verilog, one class is dedicated to programming microcontrollers using assembly and C, and a lot of the classes in between use MATLAB or some other language as is convenient. Even after all of that, it's nothing like what a dedicated computer science guy can do. A lot of schools nowadays have computer engineering degrees that really focus in on the embedded design aspect of EE and a lot more coding than EEs, so a bunch of schools are stepping up to the challenge.
Amazing content, I wish there's more. Some of the discussions: 6:58 What are the major advances in embedded over the years? 14:10 What are the fundamental differences between sequential embedded programming and HDL? 16:44 Did the FPGA concept fail? 23:27 Is there a niche out there for a talented embedded programmer? 24:45 Can one be the world's best embedded programmer without knowing the hardware. 26:39 Should beginners be relying on using libraries? 36:55 Is 8-bits dead? 48:15 Quantum computing mainstream in the future? 54:35 Hardcore designing experience.
I like how he always tangents off describing features of everything he talks about, it's useful because he's always in a teaching mindset, not just reflecting on his experience.
Loved the nostalgic banter. Brings back loads of memories. Programming in tight assembler then. Great video at showing our different attitudes from the younger generation.
As a young engineer (Mechanical) I think the biggest problem with how engineers are educated this days are we are not thought the "Keep it simple"-rule.
Could not agree more. The prime example are German over engineering of cars like BMW, Porsche, the downfall of German Engineering Keeping it simple is also the key to reliability, future proofing and Maintainability .
Really great vid! QA starts at the design phase, is inherent through peer design review, peer code review (which should enforce coding standards), and testing (which should be performed both by the design engineer and a separate QA entity) is your final insurance policy. (BTW: criticism in review should always be viewed as a learning opportunity and not a personal attack). For low level code (whether developing drivers for a large operating system or a small embedded project), where often part of the goal is that the user has an expectation of proper function and isn't necessarily going to notice that proper behavior has been achieved, the lack of complaints of improper behavior (and your own knowledge of the inherent beauty of the implementation that few others may be aware of) may be your only source of gratification.
I worked for a short time project managing for the UK home Office, and the project I worked on required a software review by the security services, so it does happen when it really counts, which can be a very good thing.
Really interesting discussion. I'm a retired mechanical engineer but life long electronic fiddler. I can remember in college when Maxwell's Equations suddenly clicked and it all seemed so obvious. Around the same time I read a series of articles in Popular Electronics? about how transistors can make simple gates and counters ect. and can be combined into registers, memory, stacks and into a computer. I lost the feeling of familiarization with Maxwell's Equations within weeks but have never gotten over my love for microprocessors. Guess I'm not alone.
Jack is finally retiring, and after taking a look at the latest 'The Embedded Muse', I went back to watch this video, and honestly, I'm blown away by how they both anticipate the arrival of AI, talking about these disruptive advances down the road. Also, I've always admired how clear he is when it comes to the embedded world, especially when it comes to quality reviews, code recycling and learning from bugs and errors.
I'm not a very successful financially, but I don't think we are going to get quantum computing anytime some. It is always going to be 10 years off - just like fusion. I also was wrong about on demand video, so maybe I'm wrong about fusion, and quantum computing. Also, I totally agree with the mathematics curriculum was ridiculous for the EE degree I recieved; I remember getting through Calculus III, and then finally taking Analytical Analysis; On the first day of this Analytical Analysis class the instructor, said "all those Calculus classes you took isn't really how we solve integrals and derivatives; we just use computers to solve these equations with analytical approximations- down to about 99.999999 accuracy. I was so furious when I learned these Calculus classes I took were arcane in the sciences today.
I used to work on a 2900-series bit slice processor, IIRC it had a 5-deep hardware stack. It was forbidden to EVER add code with a call because it was almost impossible to diagnose how deep that stack was at any point, so any patches or new code had to jump and then jump back.
High quality video about the embedded sys industry! It's very interesting to listen to two experts' musings about the field, a field that I wish to move up to as a career someday! Signed, an Electronics technologist
As a retired EE enjoyed your video very much. Identified with much of your conversation. I was a "Vital Systems Engineer" (Application Engineer) with a Railroad. I started out using safety relays with 3KV isolation between contacts and when I retired we were using vital processors for the same function.
Thanks Dave, I really enjoyed listening to Jack and you - everything was there - electronics, nostalgia, common sense and real gold nuggets of information and experience.
Jack Ganssle is the most persuasive and hieratic guy in the electronic world, he's a legend, a sacred institution. And a bit less garrulous than Dave Jones. (Yeah, just joking, I love both these badass guys)
4:55, I too was living in a VW Microbus (in Boulder) with hair to my shoulders designing systems around the 8748 micro-controller. Just set it in sun light or UV to erase the eeprom. The in-circuit emulator was massive and floppy disks really were floppy.
I loved the comment "oh my god, there's registers". Granted, it was more profound when I actually figured out what the hell a register was. When you get that and understand what's really physically going on inside the chip, it's like you've been stumbling through a giant room with a flashlight that only reveals a tiny area at a time and then you find the light switch.
They mentioned software in automotive, I recently had the software updated in the cvt transmission for my Honda Civic. And now it drives better for some reason.
A good conversation. Got me to thinking about the ethical problem at Volkswagen that began it and the social media engineering that ended it. Engineers make things work, yes, but today the unintended consequences of designs emerge much faster than they used to. I am talking about large side-effects far beyond just technical possibilities, but social and political ones. I think that any day now some engineers who works for one of the social media companies may find him or herself standing in front to a judge and facing criminal liability not just for an ethical lapse of the kind we just saw with Volkswagen, but for a tort that resulted from a design. It would not surprise me if a legal case emerged against something Facebook or Google did for which some engineer applying "Signals and Codes" problem solving to natural language and social engineering where it could be shown that there was a moral issue. Just because something is possible to do does not mean it should be done. I have rubbed shoulders with software and electrical engineers for years around Silicon Valley and I find them to be too specialized both intellectually and in training. They suffer from immense hubris and too often spout off on topics they have too little knowledge while thinking that they know the science to hold opinions. Engineers are not scientists mostly because the imperative to fix things, while nobel, is exactly the opposite of the suspension of judgement that is central to science. I am trained as a geologist and I know that to do that discipline requires an broad background in many physical and biologic disciplines and to spend most of your time defining critical tests of ideas and wait for the data to be obtained, often at great cost. One does not go off half cocked.
Fantastic video that really made me consider imbedded systems as a possible specialty. Really loved the point about the maker movement, it's great to get individuals involved in working with their hands and encourages a lot of people to possibly go for engineering how these things actually work is lost on a lot of people who just pull up a library ask it for the data. It makes debugging a lot more difficult as there is a vast amount of information about how this thing works that without it finding the answer to your problem may be impossible.
As a firmware/software guy of some 35 years, I have a slightly different take on the divide between hardware and software. In my experience, hardware guys are not very good at software (although they often think they are) and vice versa. Often, hardware guys end up writing the firmware and software because there was no-one else to do it, and all credit to them for knuckling down and getting on with it! But writing good quality software is a craft that takes a lifetime to hone if you do it full time. For embedded designs, what works best in my experience is when a hardware and software guy work in close partnership, where each understands something about the other's field but it isn't their job to do it. There was also a comment in the discussion along the lines of "software isn't something that just gets thrown in, it is something we have to really get on top of" and I completely agree with that. If the electronics is the engine of a system, the software is the fit and finish and too many systems are let down badly by poor software. It is something that companies should value much more as an asset, rather than just treating it as a cost centre!
9 лет назад
Fascinating and funny at the same time. Super cool, gonna share it. Thanks Dave.
In my university, the Electrical and Computer Engineering Department they want to do something to get teenagers to EE, even with 220 new students per year in Electrical and Computer Engineering, but they want better students to get into the department, because in here all good students go to med school. Funny to here at 33:00 about silicon doping, that is what we learn in the first semester in chemistry (and we learn about photovoltaic panels and lcd technology), we also learn vhdl programing, and it is one of the most interesting parts of the freshman year.
I must say. I have been really digging back into your stuff. Loving it as a layman who wants to learn some fundamentals as you go through all types of stuff. With that said, this interview. BEAUTY!
I'd be curious to hear your position on Computer Engineering as a major. It is really designed to be a bridge between electrical engineering and computer science, but both are really stripped down to be geared towards electronics. I personally have liked it a lot, but it does seem like it skips useful parts of both to trim it down to 4 years.
I'm 15 years old and love electronics. This video, especially around 15-20:00 is exactly what I needed to put my desire for an EE degree in concrete. Thanks so much Dave and Jack. :)
i'm having this problem right now . i can do arduino and all that jazz but I wanna go in deeper . for example i don't have problems writing directly to ports . But for all else i'm having a really hard time understanding the documentation . And since everyone is using the libraries i'm having a hard time finding well explained low level code aswell .
I've read a lot of Jack's articles and been an Embedded Muse subscriber for years. He's a great guy that has given a lot to the industry. Glad you got a video in with him.
Dave and Jack, thank you for an exciting and riveting chat. This vid is a good example of one of the better uses for social media. Also, before this vid I had never heard of Jack.
Super interesting talk! I really enjoyed it. I couldn't agree more with the advice that Jack gave. In fact, I am a software engineer with almost no EE background. Two years ago I changed job to work on embedded software (ARM SoCs). Colleagues of mine are hardware designers and I got curious about HDL and that sort of stuff. So I bought myself an FPGA board and started to hack around. I can confirm that writing Verilog may not be so easy when you have a C background but developing e.g. an FM stereo encoder on FPGA is certainly fun and enlightening :) Now I'm a bit surprised that Open Source was not mentioned as a potential way of mitigating some issues with embedded software (reliability/robustness/compliance...). You know, Linus's law: "given enough eyeballs, all bugs are shallow"...
Excellent video !!! Thank you very much for the info you share with us. While I enjoy all the teardowns it was also nice to see a kind of different type of video, I've always enjoyed hearing experienced people views on their subject.
So good to listen to some old hands talk shop! And I agree with Jack about 'software quality' being a constant core issue that shakes out in many negative ways - reliability and security being the main ones. (And VW now adds 'trust'.) The essential tension is between two sides that say "we must be able to see the source code, because our lives depend on it" and the paranoia camp who say "we can't release the source code, because people's lives depend on it!" In both cases, manufacturers are reluctant to show their work, rather than proudly display it for all to see and verify. Perhaps code that needs to be 'protected' isn't good code. Or perhaps embedded systems are inherently more vulnerable, because they simply don't have the CPU to fend off concerted external attacks. It's a discussion that needs having. "Software is not just something you throw in there." Never a truer word said.
Fantastic, I'm a big fan of Jack, have been for decades. Always look forward to his articles in embedded-dot-com and prior to that the actual print magazine. - R
Nice format. It would be cool, if there would be something like "EEVBlog - Double Teardown / Vintage Time". Jack brings some old stuff with him and both tell us something about it. By the way: a big thumbs up!!!!
You both are so humble. You say you just "bum around" when in fact, you are inspiring teachers and communicators that are giving back so much of your high quality, high signal to noise knowledge to the community Thank you.
I go back and watch this every few months. So real, so down to earth, these two guys together. Brilliant.
Re-watching this on the occasion of the 500th (and sadly, last) issue of The Embedded Muse. I am truly humbled by his depth of knowledge and dedication to the craft of engineering. Yes, you too Dave.
Today is the first day my searching fingers introduced me to first to Dave and within three clicks and two hours at retired and 70 years old I digitally met the two guys I wish I had known over coffee my whole life...
That said, you're both so much farther down, ( or up? ), the "Rabbit Hole" than my miles of experience in mechanoelectron life. But the tears in my eyes come from joining two minds that see as I see. And that may actually say volumes as to why we delve in to all the disciplines where we are led to the answers our imaginations require us to go! If it were a "choice" , we would choose not to go. But cause and effect coupled to dreaming leave it to inevitable....
Thank you Gentlemen.
Christopher Workmon
One of your best blogs, I really enjoyed that one
Very interesting .... ....I've noticed in rf DSP that companies have resorted to using old analog crystal lattice filters in front of the DSP section to relieve pressure on the analog to digital processor .... on very dirty signals....so the beauty of hybridization ....beware of the analog domain ! hahaha
The next big wave.......human imbedded computing !!! Just watch !
Fantastic interview Dave. Jack is a very inspiring person. I went to one of his seminars once, and must say, the guy knows what he is talking about.
this is way more than just electronics... a spontanous conversation with a friend, a lesson of passion for work and respect for each other. Thanks for sharing, I really enjoyed it.
Absolutely the most enjoyable vid I've seen in AGES - thanks Dave. Great hosting, great great great guest, great topics. :-)
+TrollingAround Glad to hear, thanks.
@@EEVblog I', Jack's age; so I can relate to almost everything he says!! Most enjoyable!
Now a retired EE; I went into optical transmission and networking in the late 70s. Good work for 20+ years.
But, some observations:
like the declining cost of silicon, transmission costs are 8 - 9 orders of magnitude lower, but the result has not been an improvement in the quality of human interaction.
Being honest, ethical, and a hard worker still gives almost no control of how your work will be used. Management usually demanded engineers have no social conscience.
Really got a laugh out of Jack's comments, especially about our kids!
This was a lot of fun to watch, hope you can have more of these sit downs again.
As a EE student, I have to say this is some pretty great stuff. At my school they put a decent amount of emphasis on ensuring students can handle coding. One class is dedicated to C programming basics, one class is dedicated to learning digital logic through FPGAs using verilog, one class is dedicated to programming microcontrollers using assembly and C, and a lot of the classes in between use MATLAB or some other language as is convenient. Even after all of that, it's nothing like what a dedicated computer science guy can do.
A lot of schools nowadays have computer engineering degrees that really focus in on the embedded design aspect of EE and a lot more coding than EEs, so a bunch of schools are stepping up to the challenge.
Amazing content, I wish there's more. Some of the discussions:
6:58 What are the major advances in embedded over the years?
14:10 What are the fundamental differences between sequential embedded programming and HDL?
16:44 Did the FPGA concept fail?
23:27 Is there a niche out there for a talented embedded programmer?
24:45 Can one be the world's best embedded programmer without knowing the hardware.
26:39 Should beginners be relying on using libraries?
36:55 Is 8-bits dead?
48:15 Quantum computing mainstream in the future?
54:35 Hardcore designing experience.
Thanks Dave & Jack for such a good conversation :D
a briliant contribution. Conversation to such people can tell you so much more than your own perception of the electronics development process
I like how he always tangents off describing features of everything he talks about, it's useful because he's always in a teaching mindset, not just reflecting on his experience.
Loved the nostalgic banter. Brings back loads of memories. Programming in tight assembler then. Great video at showing our different attitudes from the younger generation.
As a young engineer (Mechanical) I think the biggest problem with how engineers are educated this days are we are not thought the "Keep it simple"-rule.
Could not agree more. The prime example are German over engineering of cars like BMW, Porsche, the downfall of German Engineering
Keeping it simple is also the key to reliability, future proofing and Maintainability
.
Awesome and entertaining show! Thanks Jack and Dave!
Excellent questions and excellent answers. Thank you both.
I love this video, hooked on Jack's words.
Thanks Dave!
Really great vid! QA starts at the design phase, is inherent through peer design review, peer code review (which should enforce coding standards), and testing (which should be performed both by the design engineer and a separate QA entity) is your final insurance policy. (BTW: criticism in review should always be viewed as a learning opportunity and not a personal attack). For low level code (whether developing drivers for a large operating system or a small embedded project), where often part of the goal is that the user has an expectation of proper function and isn't necessarily going to notice that proper behavior has been achieved, the lack of complaints of improper behavior (and your own knowledge of the inherent beauty of the implementation that few others may be aware of) may be your only source of gratification.
Found Jack Ganssle and this video today, what a great blog.
I worked for a short time project managing for the UK home Office, and the project I worked on required a software review by the security services, so it does happen when it really counts, which can be a very good thing.
"I was afraid of quantum computing, I mean I knew nothing about quantum physics! But then I realized: There's gonna be an API!"
Jack is a god :D
I'm so motivated by these guys. I just hope I get accepted into a Computer Engineering MSc. I will focus completely in embedded systems!
Really interesting discussion. I'm a retired mechanical engineer but life long electronic fiddler. I can remember in college when Maxwell's Equations suddenly clicked and it all seemed so obvious. Around the same time I read a series of articles in Popular Electronics? about how transistors can make simple gates and counters ect. and can be combined into registers, memory, stacks and into a computer. I lost the feeling of familiarization with Maxwell's Equations within weeks but have never gotten over my love for microprocessors. Guess I'm not alone.
Loved it! One of my favourite episodes of EEVBlog. It's been quite time when I was googling so many stuff while listening to this talk.
Great interview and great sense of humor! I like when people are passionate about that are engaged. Great job Dave thank you very much!
Jack is finally retiring, and after taking a look at the latest 'The Embedded Muse', I went back to watch this video, and honestly, I'm blown away by how they both anticipate the arrival of AI, talking about these disruptive advances down the road. Also, I've always admired how clear he is when it comes to the embedded world, especially when it comes to quality reviews, code recycling and learning from bugs and errors.
Man, listening to you guys shoot the shit is awesome.
I'm not a very successful financially, but I don't think we are going to get quantum computing anytime some. It is always going to be 10 years off - just like fusion. I also was wrong about on demand video, so maybe I'm wrong about fusion, and quantum computing. Also, I totally agree with the mathematics curriculum was ridiculous for the EE degree I recieved; I remember getting through Calculus III, and then finally taking Analytical Analysis; On the first day of this Analytical Analysis class the instructor, said "all those Calculus classes you took isn't really how we solve integrals and derivatives; we just use computers to solve these equations with analytical approximations- down to about 99.999999 accuracy. I was so furious when I learned these Calculus classes I took were arcane in the sciences today.
This is gold! Thanks Dave.
Cheers from Italy!
Thanks Dave. Awesome interview !
How I wanted either of you as a prof when I was student :)
Cheers from Belgium.
A ham radio? LOL That is nonsense. One might be able to make a lemon battery, but a radio made from ham? Seriously...
And they expect us to find this mythical crystal pig .... where?
I'm a big fan of Jack. Glad to see him visit the EEVBlog.
I used to work on a 2900-series bit slice processor, IIRC it had a 5-deep hardware stack. It was forbidden to EVER add code with a call because it was almost impossible to diagnose how deep that stack was at any point, so any patches or new code had to jump and then jump back.
Subbed to Jack's channel. Great interview and interesting perspective
High quality video about the embedded sys industry! It's very interesting to listen to two experts' musings about the field, a field that I wish to move up to as a career someday! Signed, an Electronics technologist
I felt this one like a conversation with an good old friend...
As a retired EE enjoyed your video very much. Identified with much of your conversation. I was a "Vital Systems Engineer" (Application Engineer) with a Railroad. I started out using safety relays with 3KV isolation between contacts and when I retired we were using vital processors for the same function.
Inspiring, I'd say. Thank you Dave & Jack!
Regarding 37:00 There is also the j-core processor where you don't pay tax but it's not finished yet.
As an Electronic & Software Engineering student this was quite reassuring...
Fantastic video! Please give us more like these.
Thank you for this interview, I really enjoyed it, Thanks again.You are great.
Yeah this was fantastic. I am blown away by the quality of this.
I really liked the side by side interview format, it worked well. Good job on this one Dave!
Thanks Dave, I really enjoyed listening to Jack and you - everything was there - electronics, nostalgia, common sense and real gold nuggets of information and experience.
I was a bit worried you might start disassembling Mr. Ganssle.
Fantastic Interview Dave. Really enjoyed it, felt like an AmpHour Show with faces associated with it
Omg this is legendary 🙌
Jack Ganssle is the most persuasive and hieratic guy in the electronic world, he's a legend, a sacred institution. And a bit less garrulous than Dave Jones. (Yeah, just joking, I love both these badass guys)
4:55, I too was living in a VW Microbus (in Boulder) with hair to my shoulders designing systems around the 8748 micro-controller. Just set it in sun light or UV to erase the eeprom. The in-circuit emulator was massive and floppy disks really were floppy.
Fun thing you brought it up Jack; actually most of us Software Engineers at Audi (VW) do speak English ;)
And some watch your and Daves blog :)
I loved the comment "oh my god, there's registers". Granted, it was more profound when I actually figured out what the hell a register was. When you get that and understand what's really physically going on inside the chip, it's like you've been stumbling through a giant room with a flashlight that only reveals a tiny area at a time and then you find the light switch.
Enjoyed this episode very much. Is this going to be a semi-regular thing? Interviews and conversations with interesting people?
can you generate a spice file for an hdl code in order to have some access to the circuit's "internals"?
Fantastic ! One of the most interesting blogs to date.
couple of red flags @ 52:15
"I never solved an integral" - Jones
"I never solved a differential" - Jack
Excellent vid. Why didn't i find this one earlier - two legends behind one camera. - thanks guys
Jack's awesome, I discovered him when I got in to embedded development! Thanks for posting this.
They mentioned software in automotive, I recently had the software updated in the cvt transmission for my Honda Civic. And now it drives better for some reason.
Hi,
And many thanks for a wonderful upload. An absolutely wonderful discussion!
A good conversation. Got me to thinking about the ethical problem at Volkswagen that began it and the social media engineering that ended it. Engineers make things work, yes, but today the unintended consequences of designs emerge much faster than they used to. I am talking about large side-effects far beyond just technical possibilities, but social and political ones. I think that any day now some engineers who works for one of the social media companies may find him or herself standing in front to a judge and facing criminal liability not just for an ethical lapse of the kind we just saw with Volkswagen, but for a tort that resulted from a design. It would not surprise me if a legal case emerged against something Facebook or Google did for which some engineer applying "Signals and Codes" problem solving to natural language and social engineering where it could be shown that there was a moral issue. Just because something is possible to do does not mean it should be done.
I have rubbed shoulders with software and electrical engineers for years around Silicon Valley and I find them to be too specialized both intellectually and in training. They suffer from immense hubris and too often spout off on topics they have too little knowledge while thinking that they know the science to hold opinions. Engineers are not scientists mostly because the imperative to fix things, while nobel, is exactly the opposite of the suspension of judgement that is central to science. I am trained as a geologist and I know that to do that discipline requires an broad background in many physical and biologic disciplines and to spend most of your time defining critical tests of ideas and wait for the data to be obtained, often at great cost. One does not go off half cocked.
Kind of Digikey warehouse was available for me when I was kids, 40 years ago, called Akihabara. Same day delivery, no shipping cost.
Is there something wrong with the video / RUclips the video always seems to get stuck at 21:21
Fantastic video that really made me consider imbedded systems as a possible specialty. Really loved the point about the maker movement, it's great to get individuals involved in working with their hands and encourages a lot of people to possibly go for engineering how these things actually work is lost on a lot of people who just pull up a library ask it for the data. It makes debugging a lot more difficult as there is a vast amount of information about how this thing works that without it finding the answer to your problem may be impossible.
As a firmware/software guy of some 35 years, I have a slightly different take on the divide between hardware and software. In my experience, hardware guys are not very good at software (although they often think they are) and vice versa. Often, hardware guys end up writing the firmware and software because there was no-one else to do it, and all credit to them for knuckling down and getting on with it! But writing good quality software is a craft that takes a lifetime to hone if you do it full time. For embedded designs, what works best in my experience is when a hardware and software guy work in close partnership, where each understands something about the other's field but it isn't their job to do it. There was also a comment in the discussion along the lines of "software isn't something that just gets thrown in, it is something we have to really get on top of" and I completely agree with that. If the electronics is the engine of a system, the software is the fit and finish and too many systems are let down badly by poor software. It is something that companies should value much more as an asset, rather than just treating it as a cost centre!
Fascinating and funny at the same time. Super cool, gonna share it. Thanks Dave.
good job, I like the interviews
When I search for Jack Ganssle on RUclips I get this video :-)
In my university, the Electrical and Computer Engineering Department they want to do something to get teenagers to EE, even with 220 new students per year in Electrical and Computer Engineering, but they want better students to get into the department, because in here all good students go to med school.
Funny to here at 33:00 about silicon doping, that is what we learn in the first semester in chemistry (and we learn about photovoltaic panels and lcd technology), we also learn vhdl programing, and it is one of the most interesting parts of the freshman year.
I must say. I have been really digging back into your stuff. Loving it as a layman who wants to learn some fundamentals as you go through all types of stuff.
With that said, this interview. BEAUTY!
I'd be curious to hear your position on Computer Engineering as a major. It is really designed to be a bridge between electrical engineering and computer science, but both are really stripped down to be geared towards electronics. I personally have liked it a lot, but it does seem like it skips useful parts of both to trim it down to 4 years.
I'm 15 years old and love electronics. This video, especially around 15-20:00 is exactly what I needed to put my desire for an EE degree in concrete. Thanks so much Dave and Jack. :)
As an electronics engineering and applied computer science student I really like this video.=)
I loved to listen to you guys. Thank you.
i'm having this problem right now .
i can do arduino and all that jazz but I wanna go in deeper .
for example i don't have problems writing directly to ports .
But for all else i'm having a really hard time understanding the documentation .
And since everyone is using the libraries i'm having a hard time finding well explained low level code aswell .
I've read a lot of Jack's articles and been an Embedded Muse subscriber
for years. He's a great guy that has given a lot to the industry. Glad
you got a video in with him.
DREAMS DO COME TRUE. JACK AND DAVE IN THE SAME VIDEO!
Nice cozy talk.
Dave and Jack, thank you for an exciting and riveting chat. This vid is a good example of one of the better uses for social media. Also, before this vid I had never heard of Jack.
"Software folks are resistant of typing..." Best comment of the video.. True but still lolzzz
Great combination Sir
Which PIC's have a 'puddle of gates'? Would really like to try that out.
+Tzadvantage Bkk I think they were referring to CLC, which is available on the more recent PIC parts. Also check out Cypress PSoC and SiLabs EFM32.
+Beer Tower YEs that was the right name for it. Makes searching a lot easier. :) Thanks.
Super interesting talk! I really enjoyed it.
I couldn't agree more with the advice that Jack gave. In fact, I am a software engineer with almost no EE background. Two years ago I changed job to work on embedded software (ARM SoCs). Colleagues of mine are hardware designers and I got curious about HDL and that sort of stuff. So I bought myself an FPGA board and started to hack around. I can confirm that writing Verilog may not be so easy when you have a C background but developing e.g. an FM stereo encoder on FPGA is certainly fun and enlightening :)
Now I'm a bit surprised that Open Source was not mentioned as a potential way of mitigating some issues with embedded software (reliability/robustness/compliance...). You know, Linus's law: "given enough eyeballs, all bugs are shallow"...
Excellent video !!! Thank you very much for the info you share with us. While I enjoy all the teardowns it was also nice to see a kind of different type of video, I've always enjoyed hearing experienced people views on their subject.
Great interview Guys.
So good to listen to some old hands talk shop! And I agree with Jack about 'software quality' being a constant core issue that shakes out in many negative ways - reliability and security being the main ones. (And VW now adds 'trust'.) The essential tension is between two sides that say "we must be able to see the source code, because our lives depend on it" and the paranoia camp who say "we can't release the source code, because people's lives depend on it!" In both cases, manufacturers are reluctant to show their work, rather than proudly display it for all to see and verify. Perhaps code that needs to be 'protected' isn't good code. Or perhaps embedded systems are inherently more vulnerable, because they simply don't have the CPU to fend off concerted external attacks. It's a discussion that needs having. "Software is not just something you throw in there." Never a truer word said.
awesome video man. thanks so much!
What a great interview
Brilliant vid - two greats line up for a wonderful time. Like a two body syzygy.
the video was a pile of golden knowledge for me as a "young player" as DAVE calls us new to the game.
Agreed.
That was fantastic! Probably the only youtube video I've ever watched for over an hour
I was surprised to see Jack in your neck of the woods - this was an excellent video.
Fantastic, I'm a big fan of Jack, have been for decades. Always look forward to his articles in embedded-dot-com and prior to that the actual print magazine.
- R
This was just fantastic ! Best one.
Thank you Dave, that was Excellent.
Wonderful interview. Thanks, Dave and Jack Ganssle.
What a incredable great video, finaly a relevant discussion between guys that understand stuff, realy spot on.
Nice format. It would be cool, if there would be something like "EEVBlog - Double Teardown / Vintage Time". Jack brings some old stuff with him and both tell us something about it. By the way: a big thumbs up!!!!