Comments for System Verification Blog http://blog.freemodelfoundry.com Where a system is more than one chip Sun, 06 Jul 2008 05:29:28 +0000 http://wordpress.org/?v=2.5.1 Comment on EDA Bloggers Meet at DAC by SKMurphy » Bloggers Covering Electronic Design Automation http://blog.freemodelfoundry.com/2008/05/26/eda-bloggers-meet-at-dac/#comment-180 SKMurphy » Bloggers Covering Electronic Design Automation Thu, 29 May 2008 07:44:16 +0000 http://blog.freemodelfoundry.com/?p=26#comment-180 [...] EDA Bloggers Meet at DAC [...] [...] EDA Bloggers Meet at DAC [...]

]]> Comment on The Need to Exceed by munden http://blog.freemodelfoundry.com/2007/05/15/the-need-to-exceed/#comment-175 munden Fri, 15 Feb 2008 00:06:08 +0000 http://blog.freemodelfoundry.com/?p=14#comment-175 The code you quote does indeed look suspicious. I have asked the original author of the model to review it. I suggest further discussion of this model be moved to the <a href="http://www.freemodelfoundry.com/phpBB3" rel="nofollow">forum</a> area where it will be easier for other interested parties to find. The code you quote does indeed look suspicious. I have asked the original author of the model to review it.

I suggest further discussion of this model be moved to the forum area where it will be easier for other interested parties to find.

]]> Comment on The Need to Exceed by gnuarm http://blog.freemodelfoundry.com/2007/05/15/the-need-to-exceed/#comment-174 gnuarm Thu, 14 Feb 2008 19:28:08 +0000 http://blog.freemodelfoundry.com/?p=14#comment-174 Wow, that was a quick reply. I see that I should have made my way to the white paper link first. Talk about drinking from the fire hose! I'll have to wade through some of that. Thanks. About the model, I guess the first thing I tried to verify is either not correct, or maybe I am not looking at the code correctly. I see that the assignments for the analog inputs are like this. LchData := INTEGER((AINL - AINLNeg) / 10.0 * 8388607.0);--16#7FFFFF# This says to me that the analog input range is +10 to -10 volts. But the data sheet says... well, I can't copy the line because the data sheet is copy protected... (grumble, grumble). But it says the input range is +2.45 to -2.45 volts with 0.15 volts of tolerance. So is the model supposed to be accurate on the analog input range, or am I missing something? Wow, that was a quick reply. I see that I should have made my way to the white paper link first. Talk about drinking from the fire hose! I’ll have to wade through some of that. Thanks.

About the model, I guess the first thing I tried to verify is either not correct, or maybe I am not looking at the code correctly. I see that the assignments for the analog inputs are like this.

LchData := INTEGER((AINL - AINLNeg) / 10.0 * 8388607.0);–16#7FFFFF#

This says to me that the analog input range is +10 to -10 volts. But the data sheet says…

well, I can’t copy the line because the data sheet is copy protected… (grumble, grumble).

But it says the input range is +2.45 to -2.45 volts with 0.15 volts of tolerance.

So is the model supposed to be accurate on the analog input range, or am I missing something?

]]> Comment on The Need to Exceed by munden http://blog.freemodelfoundry.com/2007/05/15/the-need-to-exceed/#comment-173 munden Thu, 14 Feb 2008 19:06:59 +0000 http://blog.freemodelfoundry.com/?p=14#comment-173 gnuarm, Thank you for your comment. We assume the engineer using a model will also have the datasheet for the component. That said, it would not be all that difficult to add a few more comments that spell out some key specifications. Would that be sufficient or do you have something more extensive in mind? Have you seen our white papers at http://www.FreeModelFoundry.com/papers.html ? Do you think it has enough to get someone started? If not, please describe what you would like to see. Since I have been using these models for so long, it is difficult to see things from the perspective of a newcomer. Suggestions are very welcomed. gnuarm,

Thank you for your comment.

We assume the engineer using a model will also have the datasheet for the component. That said, it would not be all that difficult to add a few more comments that spell out some key specifications. Would that be sufficient or do you have something more extensive in mind?

Have you seen our white papers at http://www.FreeModelFoundry.com/papers.html ? Do you think it has enough to get someone started? If not, please describe what you would like to see. Since I have been using these models for so long, it is difficult to see things from the perspective of a newcomer. Suggestions are very welcomed.

]]> Comment on The Need to Exceed by gnuarm http://blog.freemodelfoundry.com/2007/05/15/the-need-to-exceed/#comment-172 gnuarm Thu, 14 Feb 2008 18:39:38 +0000 http://blog.freemodelfoundry.com/?p=14#comment-172 You asked for feedback, so here goes... I don't know how others like to use models, but I prefer a certain amount of documentation. I am new here, so maybe I just have not found it yet. I am looking at using the ak5393 ADC VHDL model. I see that the analog inputs are provided as using real numbers. However, I have to reverse engineer the code to figure out what the input range is for a full scale. Turns out it is not the same as the chip, it is +-10 volts rather than +-2.45 volts. This may seem like a nit, but with no documentation that I could find, I am sure there are many other questions on these models that can only be answered by digging into the code. Also, as a first time user, it would be great to have an example program that I could look at to see everything required to make use of these models. Even better would be a short how-to just to help with the initial learning curve. Perhaps I should have started off by saying, thanks for even doing this. I am impressed with the general quality of the models and even their mere existence. Thanks! You asked for feedback, so here goes… I don’t know how others like to use models, but I prefer a certain amount of documentation. I am new here, so maybe I just have not found it yet. I am looking at using the ak5393 ADC VHDL model. I see that the analog inputs are provided as using real numbers. However, I have to reverse engineer the code to figure out what the input range is for a full scale. Turns out it is not the same as the chip, it is +-10 volts rather than +-2.45 volts.

This may seem like a nit, but with no documentation that I could find, I am sure there are many other questions on these models that can only be answered by digging into the code.

Also, as a first time user, it would be great to have an example program that I could look at to see everything required to make use of these models. Even better would be a short how-to just to help with the initial learning curve.

Perhaps I should have started off by saying, thanks for even doing this. I am impressed with the general quality of the models and even their mere existence. Thanks!

]]> Comment on MemCon and Different Folks by Don Larson http://blog.freemodelfoundry.com/2007/07/20/memcon-and-different-folks/#comment-164 Don Larson Sun, 22 Jul 2007 19:35:14 +0000 http://blog.freemodelfoundry.com/?p=17#comment-164 Hi Rick, In my opinion, most web sites that become popular are usually just read. It is not that easy to get people to write on another's site. Even then, it is hard to get people to work towards a common goal using only the web. I used to have a discussion group with about 30+ people on the list. I hardly received any posts from the majority of them. They preferred to read or lurk, as the case may be. I do not know a compelling approach to get people to participate together online. It seems that to get real work accomplished, use the online medium to inform people and then hold a meeting at a real world location. From my twelve years of using the Internet, that seems to work the best. Don Hi Rick,

In my opinion, most web sites that become popular are usually just read. It is not that easy to get people to write on another’s site. Even then, it is hard to get people to work towards a common goal using only the web.

I used to have a discussion group with about 30+ people on the list. I hardly received any posts from the majority of them. They preferred to read or lurk, as the case may be.

I do not know a compelling approach to get people to participate together online. It seems that to get real work accomplished, use the online medium to inform people and then hold a meeting at a real world location. From my twelve years of using the Internet, that seems to work the best.

Don

]]> Comment on Language Wars Revisited for Component Modeling - VHDL or Verilog? by snttdxwrpv http://blog.freemodelfoundry.com/2006/12/04/language-wars-revisited-for-component-modeling-vhdl-or-verilog/#comment-24 snttdxwrpv Tue, 19 Jun 2007 15:47:11 +0000 http://blog.freemodelfoundry.com/?p=8#comment-24 Hello! Good Site! Thanks you! Hello! Good Site! Thanks you!

]]> Comment on Why Should IC Companies Supply Models? by wangwil http://blog.freemodelfoundry.com/2007/03/21/why-should-ic-companies-supply-models/#comment-143 wangwil Fri, 01 Jun 2007 03:37:12 +0000 http://blog.freemodelfoundry.com/?p=12#comment-143 Your work is great, as a test engineer, I have been suffering from the absence of the simulation model for the devices under test. while the simulation model creation may count for 40% of test application develpment time. Your work is great, as a test engineer, I have been suffering from the absence of the simulation model for the devices under test. while the simulation model creation may count for 40% of test application develpment time.

]]> Comment on Language Wars Revisited for Component Modeling - VHDL or Verilog? by Ravi http://blog.freemodelfoundry.com/2006/12/04/language-wars-revisited-for-component-modeling-vhdl-or-verilog/#comment-19 Ravi Thu, 12 Apr 2007 12:38:45 +0000 http://blog.freemodelfoundry.com/?p=8#comment-19 Hi, This is nice practical blog for using verilog in general and iverilog in perticular. Check out other tool to give graphical view of your verilog code http://vlsi.cs.iitm.ernet.in/veriviz/index.html it is a totally new approch. Regards Hi,
This is nice practical blog for using verilog in general and iverilog in perticular.
Check out other tool to give graphical view of your verilog code http://vlsi.cs.iitm.ernet.in/veriviz/index.html it is a totally new approch.
Regards

]]> Comment on Why You Need Timing in Your Simulation by Magnus Danielson http://blog.freemodelfoundry.com/2006/09/02/why-you-need-timing-in-your-simulation/#comment-4 Magnus Danielson Mon, 12 Mar 2007 00:10:09 +0000 http://blog.freemodelfoundry.com/?p=5#comment-4 While initial work will not require timing for functional simulation (during this phase your simulation speed will certainly benefit from not having the overhead), board-level simulation and especially handling high-speed interfaces will be meaningless without proper timing.This is also important in initial special analysis simulations early out to ensure the strategy works. So, while not all simulations you do require timing, there are simulation needs throughout the design process that does need them. When you only do functional simulations, timing is rarely important and the gained simulation speed is the benefit. Like all good tools, wise usage and timing of use is the key. You want as many cool tools in the toolbox, but you also want to know when to apply which tool. While initial work will not require timing for functional simulation (during this phase your simulation speed will certainly benefit from not having the overhead), board-level simulation and especially handling high-speed interfaces will be meaningless without proper timing.This is also important in initial special analysis simulations early out to ensure the strategy works. So, while not all simulations you do require timing, there are simulation needs throughout the design process that does need them. When you only do functional simulations, timing is rarely important and the gained simulation speed is the benefit.

Like all good tools, wise usage and timing of use is the key. You want as many cool tools in the toolbox, but you also want to know when to apply which tool.

]]>