| Login | | Don't have an account yet? You can create one. As a registered user you have some advantages like theme manager, comments configuration and post comments with your name. | |
| Who's Online | There are currently, 52 guest(s) and 1 member(s) that are online.
You are Anonymous user. You can register for free by clicking here | |
 | |
|
Verification Guild: Forums |
|
| View previous topic :: View next topic |
| Author |
Message |
Fritz Newbie


Joined: Feb 11, 2004 Posts: 1
|
Posted: Wed Feb 11, 2004 8:43 am Post subject: Real numbers on cost of verification? |
|
|
Ok, we all know that verification takes time and costs money, but aside ITRS, are there any other credible sources that would give data worth referring into?
What portion of the time and NRE costs is taken by verification? What is the true impact of failure of verification in time and money (from overrun schedule to re-spin or re-call; direct costs and loss of revenue due to late entry)? What is the real volume threshold to move from FPGA to ASIC (vendors' data is hardly credible)? In the end is the threshold a question of manufacturing NRE cost, or do other issues outweigh it? What are the trends? |
|
| Back to top |
|
 |
richardbradley Senior


Joined: Feb 10, 2004 Posts: 73 Location: St Louis, Mo
|
Posted: Wed Feb 11, 2004 9:27 am Post subject: |
|
|
Well, that is a very complicated question. The answer is: it depends.
Effort in verification is EXPONENTIALLY related to the complexity of the project. That is what is so difficult to get most ASIC engineers to understand. Most experienced engineers want to take what they used on previous projects and apply them to MUCH bigger designs. It is hard to explain to them that what worked yesterday is doomed to failure today.
So, time in verification ranges from .5X to 3X design times in my experience. I don’t know many good papers on this subject, I wish I did. Most of what I’ve read that can be used on this subject I got from papers on software testing. Although even there it is hard to get good empirical numbers rather then hand-waving rules-of-thumb.
Depending on the nature of the design also. Some products just need more verification then others. Simple data-path designs (like some networking equipment) are much simpler to verify then say an SMP computer, or microprocessor.
Here is a good place for software papers, It may give you some ideas on how to present the problem of verification, but I doubt you will get the numbers you are looking for:
http://www.testing.com/writings.html
Also, The Mythical Man Month is a good primer to some of the problems of complex designs.
To me, it is not a matter of NRE costs, but in what needs to be done to get the design out and working. Again, that depends. Clearly not doing verification on large designs is setting your project up for failure.
~Rich |
|
| Back to top |
|
 |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
| |
|
|