( ESNUG 475 Item 5 ) -------------------------------------------- [09/18/08]
Subject: Magma Talus/Hydra, Atoptech, Solido, MunEDA, Synopsys ZRoute
WHERE IS MENTOR SIERRA?: Again, Magma did well in the digital backend space
with users noticing Talus & their newly announced Hydra floorplanner. New
routers from AtopTech and Synopsys (Zroute) also got user attention along
with the DFM/Yield oldies Solido and MunEDA.
The only blindingly obvious hole was Mentor's; only one user commented on
Calibre and not one person noticed Mentor's Olympus-SoC nor Pinnacle tools
which they had bought from Sierra for $90 million last year. Not one!
What if you held a party and nobody came? Ouch...
"What were the 1 or 2 or 3 INTERESTING specific tools that you
saw at DAC this year? WHY where they interesting to you?
(If any were under NDA say it and I'll keep you anon on them.)"
---- ---- ---- ---- ---- ---- ----
One demo that comes to mind is Magma's Talus-Vortex. The AE explained
that Talus-Vortex is all brand new and not to be thought as Blast 6.0 (!)
Talus reads the PrimeTime generated SDF directly for optimization.
Reduces the iterations to sign-off. Has concurrent setup and hold
optimization. Our previous design had difficulty closing for set up
and hold when SI was thrown in.
I look forward to seeing if it delivers what we saw in the demo.
- [ An Anon Engineer ]
---- ---- ---- ---- ---- ---- ----
We attended the Magma Hydra and their low-power presentation. We are
using Magma already for CTS, placement and routing. We wanted to check
out what Hydra is about.
Report wise, we felt that Hydra was mainly for people doing hierarchical
flow throughout. We do that partly. So, didn't apply for us. But it
was good to know, that the tool had some wonderful features.
Low power presentation was good, but we don't do power-analysis much with
Magma tools. But, was checking out what they had in their tool compared
with what we use.
- Shankar Nagarajan of Qualcomm
---- ---- ---- ---- ---- ---- ----
I know it isn't earth shattering or anything but I was kind of impressed
with the schematic editor that Synopsys was demoing. I had used Cadence
in the past and had minor things that annoyed me. It seems like many of
those annoyances were taken care of with the "new" Synopsys editor.
Compared to what i'm used to doing, watching how easy it was to make
edits and move around the hierarchy made me drool a bit.
I attended Magma's automated hierarchical design demo (Hydra). I think
this it'll be useful especially in multicore systems. I'm just not sure
if it's going to be as easy to use as they make it sound. I hope so.
- Joel Woolf of Marvell Semi
---- ---- ---- ---- ---- ---- ----
My #2 was Atoptech, which seems to be the first tool that really can
smoothly deal with the problem of hierarchical physical design. Not
sure how real it is, though - will need to do an in-house evaluation
to determine that.
- [ An Anon Engineer ]
---- ---- ---- ---- ---- ---- ----
AtopTech Aprisa: claims/showed better timing, area, runtimes than some
current mainstream PnR vendors. We want to look at this tool soon.
- [ An Anon Engineer ]
---- ---- ---- ---- ---- ---- ----
Solido Design Automation: most of the other folks on the show floor were
dealing with the ability to RUN MCMM in their design. Solido actually
provides an engineering tool which is focused interpretation of the data
that comes out. They have a somewhat complicated, but straight forward
interface to making useful decisions as a result of an MCMM analysis.
The great thing is they are moderately, simulation engine agnostic, so
the designer can use the tool they most believe in, rather than having
to adopt just one of the camps in order to use their tool.
- [ An Anon Engineer ]
---- ---- ---- ---- ---- ---- ----
2.) Second place was a tie between MunEDA and Solido for the DFM/DFY,
design cockpit, space. MunEDA "Wicked" allows the circuit designer to
analyze for the best circuit topology while at the same time optimizing
for the best performance against given constraints and the robustness
and yield to various process corners and statistical variations.
First prepare the circuit by defining the parameters, then perform
feasibility optimization and lastly design centering. MunEDA has an
exhaustive list of user interface windows.
For a lighter option, try SolidoSTAT from Solido Design Automation.
Proximity effects such as WPE and STI can be specifically targeted for
yield improvement. Since these effects lead to variation in the design
the hope is that if the design can be improved to vary less then it will
yield more good die. SolidoSTAT can be used to identify problematic
sources of variation using standard process corners, statistical
variation and proximity variation. Changes are then recommended if
possible while remaining within the original user's constraints.
3.) Third has to be Synopsys ZROUTE even though it is not available
until 2009. 10X speed improvement, better DFM and overall better QoR.
If Synopsys pulls this off, we will all be a lot happier.
4.) I still need to extract the substrate reliably, however, when trying
to tackle such a big problem maybe the best approach is to limit the
sensitive areas of concern. OptEM is doing just that by given the
user control over what area(s) get extracted (hopefully reducing the
size to something manageable) and then reporting sensitivity to a few
carefully defined points. For example, we are all sensitive about PLL
substrates. One analysis may be focused on just that one sensitivity
thus eliminating all of the massive far away portions of the very
complex substrate parasitics that may have no affect.
- [ An Anon Engineer ]
---- ---- ---- ---- ---- ---- ----
Synopsys ZRoute. The fact that it is a router designed from ground up to
address the new technology complications.
- [ An Anon Engineer ]
---- ---- ---- ---- ---- ---- ----
Interesting 1: Mentor Calibre nm DRC, for its hints on computation
reduction for litho-checking
Interesting 3: Mentor Calibre nm LVS, for its waking-up on debugability.
- Yongcheng Miao of Fujitsu
Index
Next->Item
|
|
