HOME  |  NEWS  |  BLOGS  |  MESSAGES  |  FEATURES  |  VIDEOS  |  WEBINARS  |  INDUSTRIES  |  FOCUS ON FUNDAMENTALS
  |  REGISTER  |  LOGIN  |  HELP
Comments
You must login to participate in this chat. Please login.

once again, thanks sir brian for the lecture...excellent topic being delivered...

thanks to the DIGI-KEY & DESIGN NEWS for giving this Continuing education to us...we owe it to you...

i will be going to the next lesson...thanks...

i am now on slide 10...

i am now about to start with the lecture...

i am now downloading the slides for sir brian lecture today..

good morning sir brian! once again to continue your lecture...

Very informative lecture

Iron

Great to have access to the recorded sessions!

Iron

Missed the live lecture

Iron

Taking the class offline

Iron

hello all from chilly Edmonton, Alberta.

Iron

Another great lecture

Looking forward to another good day of lectures

Abstraction is not software friendly, it usually contradicts with hardware

Hi from OR, happy archive!

Good overview - helps relate to implementation concepts. What about approach / process concepts and issues?  Coverage/Path/Thread, Regression ... some of which you hinted at or alluded to here.  Still wanting to tie Specification to Test Plan to how to transition test plan to functional verification on target tools - a few slides would help if any of that makes sense.

Better late than never....thaks for the presentation.

 

Iron

I that case, Analog sensing needs to be modelled separately and tested, once done, abstraction can be used to bypass it.. 
Thanks Alot Mr.Brian

Right. In case i can emulate the analog sensnig with digital flags that trigger events, I can write a test bench using another controller to test responses on different stimulus

Just going to send you an email shortly

Iron

Thanks brian. It was awesome

Iron

Ok, thanks Brian! Great talk by the way!

Iron

I would ask if the analog portion can be abstracted up to operate in a digital environment. Take a look at real-number modeling as an example. This gives you digital types of performance but the ability to include voltages etc. The bad news is that it does not handle feeedback well.

For large systems, you will use simulation/emulation for most of it.

Thank you. I will. Well, gotta go.  Thanks for a great presentation.

Iron

Thanks alot Brian for the wonderfull presentation. 
I have one question, Considering an application in which you have a device that does some analog sensing and complex decision making based on the raw and processed data, how should one proceed with firmware verification??

@Suraj - if you can ever say - if this does not work we are dead! It must be 100% certain that we have no problems with this. Then you should look at formal. Typically this is for bus protocols, life critical functionality etc.

There is a free version of OrCad tools I believe - take a look at this http://www.edn.com/electronics-blogs/practical-chip-design/4406145/Book-review--Analog-Design-and-Simulation-using-OrCAD-Capture-and-PSpice

 

@Brain: I understand the differnce beteween Dynamic and Formal/Static Verification  - but because both methdologies offer different capabilities - I believe we would need to use both when trying to verfiy the design spec and functionality of a system? If true, which of the two is more beneficial/critical when modeling large systems?

Iron

They are written for their own brands but both allow you to create your own SPICE models.

Iron

@Kentj - I am surprised. I will see what I can dig up by tomorrow.

 

@NSL22

Take a book on digital design using verilog. They cover pretty much every thing you are looking for

Iron

@ NLS22 What kind of system are you trying to design?

I'm not sure I know what to ask.  Still trying to wrap my head around the concept.

Iron

@Kentj - is that for SPICE simulators?

@NSL22 - I would need a little more context to answer that question.

 

I googled it and those were the only two I came up with that were not just free for a trial period.

Iron

I would be interested in anyone that can get better than a 50% utilization above 8 cores

 

This is true. I have run into this myself.

 

great point Brian

Iron

@ tariq786 - exactly. The good news today is that there is more concurrency in designs compared to when I was working on the problem. That makes the improvements better, but at some point when trying to do a whole chip, simulators are too slow for most verification objectives.

Brian, can you give an example of a project that would require verification at the digital level? And how does one come up with a Model of the system that the project is trying to control?

Iron

SPICE and SystemC are both open source, so should be plenty of free versions available.

for designs that have enough parallelism which means that parts of the design do not exchange data periodically, multicore simulation results in speedup over conventional simulation 

but you get the point of diminshing returns soon

Iron

For free analog SPICE, not on the cloud, is TINA from TI and LTSPICE from Linear Technology.

Iron

I would be interested in anyone that can get better than a 50% utilization above 8 cores.

Multicore is better than multi-processor because they have faster communications between them, but they will still be killed by the need to synchronize and do load balancing.

 

There are some people who have been trying with the cloud. I am talking with one company at the moment that is attempting to do that for analog, but I can't mention the name yet.

Let me clarify. I do not  mean multiprocessor.

 

Iron

Are ther simulation services available in the cloud yet?

Iron

Just google my name and you will find me!

Multi-core is not new. I was working on it 25 years ago.

brian how to contact you?

Iron

In some respects an emulator is a highly parallel processor specifically designed for the purpose of simulation. It is built with a very high-speed and dedicated interconnect fabric to get the necessary speed.

multi-core simulator technology is new. Infact synopsys has rolled out VCS multi-core. It does offer speedup over single-core simulation.

 

What are your thoughts?

Iron

(My thanks seems to have gone into the nether world.) Thanks for the info.

Iron

@tariq786. Maybe we should talk about that. I do have some very specific opinions but the industry does not always agree with me for a while.

multi-procesors simulators do not provide enough speed. This has been tried for decades with digital simulation and can only produce modest amounts of speedup. It is because of the way they operate. Analog simulators can be sped up more.

Mentor Graphics offers ModelSim free tailored for FPGA vendor tools. You need to register

Iron

brian another side question. What are the good topics of research in verification + design for graduate students?

Iron

free verilog simulator is icarus verilog. google for it

Iron

Can you name some of the free VHDL/Verilog simulators?

Iron

instead of using emulators, can we not use parallel simulators running on multi-core machines

This thing seems to be hot these days?

Or are there any issues with it?

 

Iron

I do not know of any free UVM/OVM implementations. There are free SystemC simulators and free VHDL/Verilog simulators.

 

There are many compromoises between the systems. Most companies find they have to use all of them for large, complex chips.

Maybe I am anticipating a little but I would like to know if there are free tools with which one can test  OVM and/or UVM.

Iron

Thanks Brian and Chuck!

Iron

Thank you. See you tomorrow

Iron

Hi everyone - yes turn around time is how long it takes from making a change in the design to how long you can be verifying again.

Simulators are fast, emulators are a little slower, prototypes are slow.

Thank you also Chuck

Iron

Thanks! Very good talk Brain!

Iron

Brian, thank you for Today's presentation 

Iron

Thanks very much Brian & Chuck

Iron

Thanks alot, Great presentation

Iron

Thank you again Brian! Great presentation

Iron

Thank you Brian, very good presentation!

Iron

Compared vs silicon you need weeks to months to have a new chip to test. Simulators are very fast. You may only need to tweak a couple of lines and there you go. Also as typically it is for snall blocks you may be able to contact a few designers to do a change. Opposite to a large DUT in an emulator, you more likely need to contact more layers of designers (digial and analog ones).

Iron

@Alka

It means you do your fix, run simulation again and see the result fairly quickly (after minutes or hours at most).

Iron

@Alka:  How long it takes from the time you make a change to the time it takes to test it agani if I understand correctly.

Iron

What is the meaning of fast turn around time in simulator?

 

Iron

******* Slide 10 ******

Iron

System Verilog, SPICE

Iron

Verilog, Embedded C, ASM

Iron

As in slide9, execution engine can be a simulator, an emulator or a FPGA. It is where you run your verification.

Iron

VHDL, Verilog, System Verilog

Iron

Coming from Europe: Matlab/Python, C+ /SystemC, VHDL, SPICE

Iron

Verilog/VHDL/C+

Hspice/MATLAB

Iron

Verilog, C, C+ , Matlab, and others as the project requires

Iron

Verilog, Verilog-A, SPICE

Vanilla C, believe it or not (early days).

Iron

I'm using Systemverilog.

Iron

******* Slide 9 ******

Iron

******* Slide 8 ******

 

Iron

******* Slide 7 ******

Iron

Interesting blog post

Iron

******* Slide 6 ******

Iron

Is execution engine equivalent to testbench?

 

Iron

******* Slide 5 ******

Iron

When mine doesn't work I use a different browser or a different computer.

Iron

install a new version of Flash.

Iron

******* Slide 4 ******

Iron

I guerss that I will have to listen to the archived version.  Nothiing is working.

Silver

What do you mean by dynamic verification?

 

Iron

Good morning from Guadalajara.

Iron

@John E: did you update your browsers flash plugin after yesterday? Newer updates tend to break the audio.

Iron
Be sure to click 'Today's Slide Deck' under Special Educational Materials above right to download the PowerPoint for today's session.
 
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.
Blogger

Please use either Firefox or Chrome. Many people have trouble with the audio player when using IE.

Audio is UP and loud and Clear! 

Iron

Refreshing your browser can help with audio problems

Iron

Fro me today audio gets on and off.

Iron

Below solutions work unless your company blocks steaming audio.

Iron

I Have and nothing happens. It worked fine yesterday.

Silver

The audio player is the bar under the "Apr 02 - Day 2: Models & Model Execution" banner.  Just click the "play" button.

 

Iron

Restart your browser.  Then click on the audio bar

Iron

I cant hear the audio

 

Iron

I don't even see the player :(

Iron

Has the audio started yet?  I see the player but don't hear anything.

Silver

Hi all.


Thanks for slide 16

Iron

A GPS accurate down to a few mm would sure be handy for my "smart" lawn mower.   :)

Good afternoon, everyone

Iron

Hello from snowy Toronto!!!

Iron

Hi all - Audio is live!

greetings from Newport Beach, California!

Iron

Brian, you are correct; GPS was the brainchild of a DOD scientist and MIT scholar, Dr. Ivan Gething. The Raytheon Corporation where he worked was instrumental in development of the first prototype circa 1951. Not sure there was a lot of micro circuitry back then, but throughout the years contractors like Raytheon, Booze Allen, Lockheed Martin and Boeing, have done considerable work refining GPS and making it what it is today. Hope this helps!

 

Nelso7926 from Texas.

Iron

Good evening all.. Kartik from India :)

Iron

Re GPS: True the military did the initial development, but when they decided to allow commercial use, the industry took off. Someone had to develop the chipsets that are in every Garmin,smart phone, surveying instrument, marine navigation, etc.

Good evening and greetings from Munich.

Iron

Getting everything ready for the broadcast now. See you in a couple of minutes.

Hey..11.30p.m in India!!

 

Iron

Good afternoon all!

Iron

@brian_baily GPS came from the military.  Chipset came along in the early 90's for the military version (very propriatary!), late 90's for the commercial versions.

Iron

Good morning everyone... 10:53 a.m. Here in Chihuahua Mexico =)...

Iron

Hello from Michigain

Iron

Well GPS came out of a military program back in the late 70's I believe.

Brian: Re chip design... Do you know  any history of how the chipsets for Global Positioning (User end) were developed and which consortium did the design work?

Good afternoon/morning from PA.  I'm not sure what season it is - the calendar says spring, the window says it's winter.

Iron

LPM = a low power fab process

@brian_bailey We're just ready for sunshine and nice weather...  Winter is just hanging on..had 12" snow a week ago.

Iron

LPM from your blog.  Since it has to do with chip design I know it's not Linear Panel Meter.

Iron

@brian_bailey:  LPM?? Context??

From your blog posting (I would also like to know):

The approach they used enabled them to integrate the TSVs with minimal disruption to the 20nm-LPM platform technology, demonstrating SRAM functionality with critical device characteristics in line with those of standard 20nm-LPM silicon.

Iron

@Kentj

That's funny. At least your are standing out!

Iron

Yuck - So glad I live in Portland, OR

Naw...summer is coming...100 deg and 95% humidity...LOL!

Iron

Hello from Albuquerque.

Iron

@paul_b - guess you should call that summer!

Afternoon from sunny St. Louis (two days in a row!)

Iron

Good morning all from CA

Iron

I was a little odd.  I wanted to be a chemist all my life and ended up designing specialized test equipment.

Iron

I wanted to be an astronaut all my life but ended up a chip designer...

Iron

As kids we all wanted to "go to the moon" but nobody wanted to improve the FinFET

Yes - I think that is it. Rocket science has a "wonder" about it, even though technologically it is very backwards.

Hi everyone, good morning.

Iron

Rocket science and chip science are both interesting to me.

Iron

Good evening from Iasi, Romania

Iron

Hello from sunny Atlanta.

 

Iron

@brian_bailey:

I read your blog and it's very good. Quite interesting indeed are the advances in chip design. But I'm afraid that to most poeple, including myself, rocket science is more inspiring ;)

Iron

Hello from Ottawa ON

Iron

Good Morning from Sunny Valdez. Ready for Models and Model Execution..

Just posted a blog on EDN - would love to hear your thoughts- http://www.edn.com/electronics-blogs/practical-chip-design/4411128/Chip-design---it-s-not-rocket-science

Yes - had two cups already. Will get another one before we start! Good job I have my own espresso machine!

 

@gordonmx It seems to be the first 25 that login to the chat.  The first three or four classes I attended I didn't login until a few minutes before the lecture starts.  During them I never got a Starbucks card.  The last class or two I started logging in around noon Eastern and I got a card today.

Iron

Yes, they should give all lecturers free coffee.  They need to stay attentive than anybody.  (Staying awake isn't a problem.)

Iron

They should give me free coffee as well!

I will wait until the start of the session before going back to slide 16 otherwise some poeple may miss it.

 

At first it was all attendees but I bet that got too expensive.

Iron

It used to be the first 25 now they just say 25 attendees so I'm assuming it's random.  I assume a lot.

Iron

Hello from windy cold upstate NY  temp 29F wind chill 12F

Iron

I don't think it reads first or last, just 25.

Iron

There seems to be a number who log in just for the Starbucks card.  Some have logged in almost a week in advance.  So I'm curious, is it the 1st 25 who log into the site or the 1st 25 who click on the audio at the start of the session?

Iron

Good morning

Are we having an early jump start with slide 16 from yesterday?

 

Iron

Good afternoon all.

Iron

Hello from sunny, cool Huntsville, Al.

Iron

Hi from Austin TX...

Iron

Good evening everyone!

Iron

Good afternoon everyone!

Iron

Good (period of time of day where you're at) everybody from Richmond, TX

Iron

Greetings from Chicago!

Iron

Greetings from Raleigh, NC.

Iron

Good Morning from Panama City, FL

Iron

Cold and sunny in Aurora - Ontario

Iron

Glad you didn't delete the coffee you will need to keep up!

I received one of the Starbucks eGift cards from Design News in my email today.  Because I usually just look at the from and delete anything that looks like it's from marketing I almost  deleted the email without reading it.  Fortunately I glanced at the subject and it mentioned Design News but with all the spam we get it would be easy to miss one of the eGift card emails.

Iron

Good morning from sunny (for a change) Brick, NJ.

Iron

Good morning everyone. I hope you are ready for another fast ride today! I will also try and go over slide 16 from yesterday in a little more detail.

Good Morning from partly cloudy Boston, and the Red Sox 1 game Up on the Yankees,LOL JO'L

Iron

Good morning from Chicago

Iron

Good morning from Scottsdale, AZ

Iron

Thanks to DesignNews, I just received a new Starbucks Card today. It really makes my day:)

Iron

Good morning from Edmonton, AB

Iron

Good morning from Milwaukee!

Iron

Good morning from Mobile, AL

Good Morning from GA

Iron

Indeed.  It is important to be jacked on caffeine when writing complex verification components. ;)

Looking for the coffee ..



Partner Zone
Latest Analysis
The promise of the Internet of Things (IoT) is that devices, gadgets, and appliances we use every day will be able to communicate with one another. This potential is not limited to household items or smartphones, but also things we find in our yard and garden, as evidenced by a recent challenge from the element14 design community.
Researchers have developed a new flexible fabric that integrates both movement and sensors, introducing new potential for technology-embedded clothing and soft robots.
Made by Monkeys highlights products that somehow slipped by the QC cops.
If you didn't realize that PowerPoint presentations are inherently hilarious, you have to see Don McMillan take one apart. McMillan -- aka the Technically Funny Comic -- worked for 10 years as an engineer before he switched to stand-up comedy.
The first Tacoma Narrows Bridge was a Washington State suspension bridge that opened in 1940 and spanned the Tacoma Narrows strait of Puget Sound between Tacoma and the Kitsap Peninsula. It opened to traffic on July 1, 1940, and dramatically collapsed into Puget Sound on November 7, just four months after it opened.
More:Blogs|News
Design News Webinar Series
11/19/2014 11:00 a.m. California / 2:00 p.m. New York
11/6/2014 11:00 a.m. California / 2:00 p.m. New York
10/7/2014 8:00 a.m. California / 11:00 a.m. New York
12/11/2014 8:00 a.m. California / 11:00 a.m. New York
Quick Poll
The Continuing Education Center offers engineers an entirely new way to get the education they need to formulate next-generation solutions.
Dec 1 - 5, An Introduction to Embedded Software Architecture and Design
SEMESTERS: 1  |  2  |  3  |  4  |  5  |  6


Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.
Last Archived Class
Sponsored by Littelfuse
Learn More   |   Login   |   Archived Classes
Twitter Feed
Design News Twitter Feed
Like Us on Facebook

Sponsored Content

Technology Marketplace

Copyright © 2014 UBM Canon, A UBM company, All rights reserved. Privacy Policy | Terms of Service