( ESNUG 592 Item 09 ) ------------------------------------------------------ [06/27/23]
Subject: MENT Ravi on Siemens' 14 new Aprisa PnR customers in only 12 months
The live DAC'22 Troublemakers Panel
Cooley: Ravi. Was buying Atoptech (now "Avatar") a mistake?
Ravi: So, with Atoptech? It's a brilliant technology.
Atop was well known already for its outstanding PnR technology.
Yes, you could say, well ...
"Why would you (Mentor) be crazy enough to enter this market?"
The PnR market is one of those few markets in EDA where inside
someone's (CDNS' & SNPS') moat, you can build an island.
And you can build a pretty healthy island because PnR is a very
large business.
[ Editor's Note: total digital PnR sales from CDNS/SNPS/MENT
worldwide ballparks to around $950 million annually. - John ]
And digital PnR is a socket that is going to allow you to create
that growth.
So, you (MENT) actually can be number one in the PnR island in
someone else's moat. History has shown that to be the case.
So, with Atoptech (Aprisa) after the acquisition, there has been
significant investment into the product. Aprisa has TSMC
certifications at 7nm and 5nm -- and we are coming up with more.
And Samsung certifications at 5nm and 4nm to be announced shortly,
probably at the DAC'22 show.
[ Editor's Note: since that DAC'22, TSMC has now certified
Aprisa at both 5nm and 4nm and Samsung has at 4nm. - John ]
And Aprisa has over 14 customers in the last 12 months.
Cooley: Is that new customers, or existing customers?
Ravi: Brand new customers. Not previous customers of Atoptech.
Cooley: Which customers are going to you (Aprisa PnR)?
Because,I mean Cadence Innovus PnR is really well known, and
Synopsys Fusion Compiler PnR is really known. You guys are ...
Ravi: Right. And when Cadence Innovus started, they were a new PnR tool
and they demonstrated as the standard -- there's a methodology of
working with early customers -- who are teaching you what are the
new challenges as you go to from 7nm to 5nm to 3nm -- and what
needs to happen with the EDA tooling.
So absolutely, Innovus is there -- and there is room for more.
Customers ultimately will say, we have no business looking at a
third solution, in which case there will be no 3rd solution.
But today, customers are asking for a 3rd solution.
Cooley: Do you have a specific niche you go after?
Ravi: Do we have a specific niche we go after?
No.
We have 14 new customers at 12 months, and PnR evaluations are one of
the most intensive evaluations in EDA.
It's a significant investment of product engineering, R&D, and
application engineering. So, what you need to do is to be able
to have a robust enough framework on the PnR tool to be able to
quickly do the learning on each node.
It is node-by-node learning [12nm-10nm-7nm-5nm-4nm-3nm-2nm-1nm],
and that learning needs to essentially become scalable.
So, the new node introduction offers a new opportunity.
If the level of physics problems are such that it warrants a completely
new PnR solution.
If the existing [CDNS/SNPS] PnR solutions have difficulty scaling ...
Cooley: Right. If you're going to 2nm, it's a total re-evaluation.
Ravi: Exactly. So, what the world has presented right now is a new opportunity
in terms of the physics changing so much.
And that opens up the door for Siemens Aprisa, right?
Cooley: Wow.
---- ---- ---- ---- ---- ---- ----
Related Articles
Sawicki on Siemens Avatar PnR vs. Innovus, ICC2, and Fusion Compiler
Avatar Aprisa benchmarks vs. Innovus/ICC2 at 7nm is Best of 2019 #7b
Avatar/AtopTech's big comeback in digital PnR is #4a "Best of 2018"
Chi-Ping Hsu answers the 6 big technical doubts about Avatar PnR
Synopsys layoffs means ICC2 rewrite is unknown for 3 to 4 years out
92% of EDA users say Justice was NOT served in SNPS-Atoptech case
Join
Index
Next->Item
|
