RE: SCE-MI 2.0

All three of you had the same reaction:

What I was trying to convey is that proxy models are optional
for all 3 interfaces but I realize the diagram does not
depict this other than the yellow comment that says

"and/or proxy models".

I suppose I could show the proxy models in giant [square brackets] :-)

-- johnS


-----Original Message-----
From: Per Bojsen [mailto:per.bojsen@amd.com]
Sent: Thu 5/17/2007 3:47 PM
To: Shabtay Matalon
Cc: Stickley, John; brian_bailey@acm.org; itc@eda.org
Subject: Re: SCE-MI 2.0
 
> 6.	Nothing precludes Proxy Model companions to the transactors when
> using Pipes. Added optional proxy model box.

For DPI and pipes an arrow is shown bypassing the proxy model box.  Why
isn't there one for the macro-based use model?  Actually, should we just
get rid of the proxy model and show the use models connecting directly
to the testbench?  In effect this would hide the details of how you
architect your testbench which is not that relevant to this picture.

On the other hand, I do think it is important to note the distinction
between transactor and DUT on the HDL side since the transactor components
are part of the HDL side of the testbench.

In summary, I suggest getting rid of the proxy model boxes to make the
picture simpler.

Per




-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.