( ESNUG 563 Item 7 ) -------------------------------------------- [03/09/17]
Subject: Amit and Sawicki on Cadence Spectre, Synopsys HSPICE, Mentor BDA
DAC'16 Troublemakers Panel in Austin, TX
Cooley: Amit. What's going on in the SPICE market?
Amit: So as you know we have variation aware design tools, so
we drive a lot of the SPICE simulation...
Cooley: No, no, no, I'm not asking about your tools.
Amit: Yeah, yeah... So we did a SPICE survey -- we just
released the results on your website. (ESNUG 561 #4)
I've got the data here, we broke the market into three
areas: analog/mixed signal, standard cell and memory design.
For analog/mixed signal, Cadence is number one, but flat.
Synopsys is down, and Mentor grew more than 2x for their
(prior) leadership perception - Mentor (BDA) grew on the
analog side.
Cooley: So can you say the specific tool; Spectre versus... ?
Amit: So Cadence Spectre, Synopsys HSPICE/Custom Sim, and then
Mentor Eldo and BDA/AFS.
Cooley: Ok.
Amit: On the memory side, Synopsys' lead is shrinking to Cadence's
Spectre and they have a Spectre XPS tool. So they bought
Altos, I think it's been, what, five years now? So I think
they continue to see gains in memory as a result of Altos
acquisition.
And then, for Standard Cell, the same thing. Synopsys is
number #1, but they're dropping on the SPICE simulation
side and Cadence is growing. Overall, Cadence is growing
in memory and standard cell and Mentor is doubling in analog.
Cooley: So one of the things I noticed in your survey was that BDA
dropped significantly. What happened?
Amit: Yeah, in terms of SPICE usage, BDA dropped about six points.
Cooley: 6 out of a possible 17.
Amit: Right. So I can't really explain that, because they did do
better on the analog side. I would suspect that on the
memory they released a new tool called AFS Mega and it
hasn't done as well as they had hoped.
Cooley: So, Joe.
Sawicki: Yes, sir.
Cooley: Why is BDA dropping?
Sawicki: So, a couple of things. One, we saw BDA, our overall analog
stuff, where revenue grew by 2x, and AFS Mega actually doubled
off a small base. So in terms of financials, it's going well.
My sense in the memory market, is that AFS Mega is a relatively
new tool and is going through the maturity process. It's a
relatively new tool, and I think the market perception is
relatively unstable. I'll worry about AFS Mega next year if
it does the same thing and doesn't flip the other way. My
expectation is that the tool has gotten pretty solid right now.
We're now taking it to more than a couple of customers because
we'd been fundamentally targeting just a couple of customers
to get AFS Mega solidified. So I think next year will be a
much better result.
Cooley: Jim, you know a little bit about SPICE. What's going on there?
Hogan: What's going on in SPICE?
Cooley: You know, with the drop. BDA and Ravi was supposed to be doing
well.
Hogan: You know, there are pretty well-defined markets. There's analog
mixed signal, there's standard cell and there's memory. In our
data - whoever we are - is you find that the mixed signal and
analog markets are actually growing a little bit. Surprisingly,
my intuition would be that wouldn't be the case; however they are.
Standard cell and memory is kind of a replacement market. And
there was a comment; we were talking a little earlier about who's
winning and what not. It's a market share issue nowadays. It's
who can win the market share, right? Some of the major guys are
doing a great job of being very aggressive and going after more
market share. And as a result this skews the product numbers.
So what's happened to Ravi? Well, I think it's been a year and
a half since he was acquired by Mentor and he probably went
through integration process, got the team squared away. You
don't enter the memory market in one cycle -- it takes multiple
cycles. And typically you go to the big memory guy -- we all
know who that is, in Korea -- and you look at their budget
cycle and I think that has a lot to do with it. So if you miss
one cycle, you will lose some market share in memory, and that
skews your entire cycle.
Cooley: What type of cycle is that, is that a 3-year cycle?
Hogan: At least a couple of years.
Cooley: Okay so you have to catch the window.
Hogan: Yeah, so maybe Ravi missed this window.
---- ---- ---- ---- ---- ---- ----
Related Articles
Anirudh and Sawicki on IC Compiler II, Innovus, Nitro-SoC, Antun
Amit and Sawicki on Cadence Spectre, Synopsys HSPICE, Mentor BDA
Dean and Sawicki on Big Data tapeout predictors, John Lee's Gear
Raik and Hogan on formal apps, formal engines, and SNPS VC Formal
Anirudh and Sawicki on iffy Apache IR-drop #'s vs. Voltus/Innovus
Join
Index
Next->Item
|
|
