Max, I noticed when I try a link (and lose this page) and then try to come back to it, it starts SLOWLY rebuilding the conversation. But if I right-click and "reload" or "refresh" at that point, it almost instantly jets straight to the top of the list of posts. Maybe it will work for others... (i'm using Chrome)
I was just being funny.... I don't even use an oscilloscope, but I stumbled into that video and thought it might be a good resource to share. I do sympathize with people trying to learn new subjects, Sometimes it seems like there is too much info on the internet, a lot of it very specialized, and it can be difficult just trying to find a GENERAL INTRO on something that covers the bases without going too deep. That is one reason I appreciate DigiKey and Max for doing a series like this.
pcbJack - I'm the opposite; much rather read than watch (usually, anyway) - why take 5 minutes to watch something I can skim and read in 60 seconds? (but then, I'm old, as my kids keep reminding me...)
Errant - that doc pcbJack posted shows a bunch of equations to figure it out (page 4) but says rule-of-thumb is 5X the fastest clock in the system. That jives with my memory (but I didn't post an answer 'cause I wasn't sure my memory was right...)
I'm sorry everyone -- it will take me ages to recover (like 30 minutes to reload all the messages to come up to date) -- I'll have to sign off now -- I hope to see you tomorrow for the final session...
Re "Cray on an FPGA" .. it's easy to forget how far we've come so fast -- the article I mentioned says "1983, the Cray X-MP was the world's fastest computer -- Dual CPUs, 16MB of RAM, and a peak performance of 400 Megaflops" ... not so impressive by today's standards...
OSC. BANDWIDTH: It is simple. Decide how manu periods of the fastest clock on a digital circuit/system and/or the fastest harmonic component of an analog signal you want to reproduce and "measure" sith clarity on the osc. screen. That is the answer. Am I close, MAX?
yeah, I'm kiddin' ya... but seriously, I just fouind a LOT of info about that just with a general search on "oscilloscope bandwidth". The answer i saw more than once, specifically, is that your o-scope should be at least two times the highest frequency in the device. I'm sure there is more to it, just telling you what I saw
I asked Mike Dunn who is Editor in Chief of www.ScopeJunction.com -- he replied: In some ways bandwidth is the least important parameter. For a 100 MHz clock rate, you need a bandwidth of at least 350 MHz. More important are things like triggering, sampling rate (for digital scopes), pulse capture, etc. It would take at least an hour to explain all of the factors involved. Look here http://www.tek.com/learning/oscilloscope-tutorial for tutorials.
Re the mention of implementing a Cray on an FPGA -- you are right -- a Cray Supercomputer from I think the 1980s was recently replicated on a single (not so big) FPGA by a young designer -- check out thsi article: http://www.programmableplanet.com/author.asp?section_id=2340&doc_id=247408&
@Luizcosta: Re your question about natural language -- I'm not sure -- if we get to the point where th ecomputers knwo wha twe are saying, then we are at the point where they can design themselves -- check out my review of Robopocalypse by Daniel H. Wilson http://bit.ly/SW6A6j
@Samdisp06: Re looking for an inexpensive FPGA dev kit. Duane on APP (AllProgrammablePlanet) is uasing an $89 one (search for "Discovering FPGAs") -- but Gadget Factory gave us a special offer on one known as a Papilio Board for about $35 -- check out this article: http://www.programmableplanet.com/author.asp?section_id=1925&doc_id=247412
@Vinaya: Again. I think you will find the discussions on AllProgrammablePlanet.com to be agood place to start ... do a search on "Ask Max" and "Ask Adam VHDL" and "Discovering FPGAs" (that will keep you busy for a while :-)
@Vinaya: I think FPGAs are a REALLY good area to get into. Before you learn FPGAs you need to understand basic digital concepts -- then you are going to have to learn a hardware programming language like Verilog of VHDL (which is the best is aon on-going debate)
2 Questions:1. Is there a hobbyist/experimenter/workbench kit for "playing" with these devices and the associated logic much like the Arduino or is there a better way to learn this? 2. What common items have Zynq-7000's or similar FPGA's in them? Thanks for these classes.
@shorek - without pull-up, line will float unless it's being pulled up or down by a logic gate. floating means you won't be able to predict what it is. pull-up makes it predictable value. (usually a '1' - then logic gate pulls the signal down to '0')
Yes, I think the point is that Max is combining early history so we can see how the technology grew, and now the devices are so complex that you can't assimilate it all in your mind. The goal is to have a general understanding of where we came from and how fast the technology grew, but to really design ASICs and FPGAs you will ned a LOT of VERY specialized training. This is just an intro...
@bmatts - changing a logic line from "0" to "1" requires a small bit of charge (electrons), changing from "1" to "0" requires that to be drained to ground. Mulitply this by a gazillion logic gates and gigaherz clock speeds results in significant current flow. High current flow results in heat generation.
@RudySchneider: I appreciate your comment. however, if you look closely to the boolean equations in all the examples, you'll realize that those vertical lines with the cross-connect on the vertical line, they imply a three wire connection to the output OR gates. By the way, SOP stands for Sum of Products like in the boolean equation w = a & B & C | A & B | A & C, which requires 3 input OR gates.
Rich - some of the webinars I have attended do have a "the webinar will begin in X minutes" message in the period 5-10 minutes preceeding the event. This is super handy to determine if one's system is up to th task. You may suggest it if you get the opportunity.
Otherwise - these continue to be great sessions. Thanks to Digi-Key, yourself, and Max
@Emily, I heard about the courses through a co-worker back in January. He possibly was on a design news mailing list. I thought, why not give it a try and see what the courses are all about. That was 14 courses ago.
The streaming audio player will appear on this web page when the show starts at 2pm eastern today. Note however that some companies block live audio streams. If when the show starts you don't hear any audio, try refreshing your browser.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.