( SNUG 05 Item 2 ) ----------------------------------------------- [12/20/05]
Subject: Customers Rate The Best & The Worst Synopsys/Avanti Tools
READING THE TEA LEAVES: The best way to find out what's hot and what's not
is to directly ask the users. Because 3 years ago, back in SNUG'01, 321
answered the 2001 survey, I proportionally scale every year's responses to
321 responses here. Out of the 1,045 users attending this year's SNUG'05,
504 answered this best & worst Synopsys tools question, so I scaled their
answers by 321/504 to enable apples-to-apples year-to-year comparisions.
"Which Synopsys product(s) are you *most* satisfied with? (write in)"
Design Compiler '04 #########################S S################### 96
Design Compiler '05 ################################### 69
PrimeTime '04 ###################################### 76
PrimeTime '05 ##################################### 73
VCS '04 ############## 28
VCS '05 ###################### 44
Avanti Astro '04 ################# 34
Avanti Astro '05 ##################### 41
PhysOpt '04 ######################## 48
PhysOpt '05 #################### 40
Vera '04 ### 5
Vera '05 ####### 13
TetraMAX '04 ####### 14
TetraMAX '05 #### 8
Avanti Star-RCXT '04 ######## 15
Avanti Star-RCXT '05 #### 8
Avanti Jupiter-XT '04 0
Avanti Jupiter-XT '05 ### 6
Power Compiler '04 ## 4
Power Compiler '05 ### 6
Native Testbench '04 0
Native Testbench '05 ### 5
PrimeTime-SI '04 ## 4
PrimeTime-SI '05 ## 4
Avanti HSPICE '04 ## 4
Avanti HSPICE '05 ## 4
DFT Compiler '04 # 2
DFT Compiler '05 ## 4
Magellan '04 # 2
Magellan '04 ## 4
IC Compiler '04 0
IC Compiler '05 ## 4
Formality '04 # 1
Formality '05 ## 4
Avanti Hercules '04 ## 4
Avanti Hercules '05 ## 3
RVM '04 0
RVM '05 ## 3
PrimePower '03 0
PrimePower '04 # 2
DC-FPGA '04 ## 4
DC-FPGA '05 # 2
NanoSim '04 # 2
NanoSim '05 # 2
Avanti AstroRail '04 ## 4
Avanti AstroRail '05 # 1
Avanti Apollo '04 ## 4
Avanti Apollo '05 # 1
Funny observation here -- all the 2004 "best" votes total to 351 and all
the 2005 "best" votes total to 351! Ha! No, there's no error here; it's
just a coincidence. But it's interesting because you can see how the
histogram of "best" tools is flattening out -- Synopsys is no longer just
the Design Compiler company -- instead it's showing as a "best" vendor in
verification tools, too.
"Which Synopsys product(s) are you *most* satisfied with? (write in)"
VCS '04 ############## 28
VCS '05 ###################### 44
Vera '04 ### 5
Vera '05 ####### 13
Native Testbench '04 0
Native Testbench '05 ### 5
Magellan '04 # 2
Magellan '04 ## 4
Formality '04 # 1
Formality '05 ## 4
RVM '04 0
RVM '05 ## 3
This data is also confirmed in my 2005 Verification Census where, out of
the blue, Synopsys jumped up to match Cadence in the tool use in this niche.
Now on to the Synopsys "worst" tools vote as seen by the customers.
"Which Synopsys product(s) are you *least* satisfied with? (write in)"
Avanti Astro '04 ######## 16
Avanti Astro '05 ########## 20
Formality '04 ###### 12
Formality '05 ######### 18
VCS '04 ##### 9
VCS '05 ####### 13
Design Compiler '04 ####### 13
Design Compiler '05 ###### 12
PhysOpt '04 ######## 15
PhysOpt '05 ###### 12
DesignWare '04 # 2
DesignWare '05 ##### 9
Avanti Jupiter-XT '04 ##### 11
Avanti Jupiter-XT '05 #### 8
Avanti Hercules '04 ##### 11
Avanti Hercules '05 ### 6
PrimeTime-SI '04 ### 6
PrimeTime-SI '05 ### 6
TetraMAX '04 ### 5
TetraMAX '05 ### 6
VirSim '04 ## 3
VirSim '05 ### 6
Magellan '04 # 1
Magellan '05 ### 6
Avanti Cosmos '04 ## 3
Avanti Cosmos '05 ## 4
DFT Compiler '04 ## 4
DFT Compiler '05 ## 3
Avanti Star-RCXT '04 # 2
Avanti Star-RCXT '05 ## 3
LEDA '04 # 2
LEDA '05 ## 3
Power Compiler '04 # 1
Power Compiler '05 ## 3
PrimeTime '04 #### 7
PrimeTime '05 # 2
Floorplan Compiler '04 ### 6
Floorplan Compiler '05 # 2
Avanti AstroRail '04 ## 4
Avanti AstroRail '05 # 2
NanoSim '04 ## 3
NanoSim '05 # 2
BSD Compiler '04 # 2
BSD Compiler '05 # 2
Vera '04 # 1
Vera '05 # 2
Avanti Astro CTS '04 # 1
Avanti Astro CTS '05 # 2
FPGA Compiler II '04 # 1
FPGA Compiler II '05 # 2
Pathmill '04 0
Pathmill '05 # 2
Native Testbench '04 0
Native Testbench '05 # 2
Innologic '04 0
Innologic '05 # 2
VCS-MX '04 0
VCS-MX '05 # 2
Saturn
PrimePower
Module Compiler
Test Compiler
DC-FPGA '05 # 1
For the 5th year in a row, overall there were much less "worst" tool user
votes than "best" tool user votes.
Synopsys
"Best" "Worst"
tool votes tool votes ratio
------------ ------------ -------
2001 306 144 2.13X
2002 344 131 2.63X
2003 371 172 2.16X
2004 351 155 2.26X
2005 351 167 2.10X
Cadence
2004 142 115 1.23X
That is, for every "worst" Synopsys tool that a user cited in 2005, there
were also 2.10 "best" Synopsys tools a user cited. As I said before on
this question in my 2004 Cadence-vs.its-rivals census, I thought Cadence's
ratio would be roughly in line with the Synopsys ratio -- but CDNS 1.2 is
noticeably off from the SNPS 2.3 average. Good news for Synopsys; not
good news for Cadence.
"Schadenfreude \SHOD-n-froy-duh\ [Ger schaden 'damage' + freude 'joy'],
n., a malicious satisfaction in the misfortunes of others."
- from Dictionary.com
Now the tea leave reading comes on specific tools. This is where I ratio
the "best" votes with the "worst" votes to see if there's any news. For
example:
VCS '05 ###################### 44 "best" tool votes
VCS '05 ####### 13 "worst" tool votes
gives VCS a 44 to 13 ratio -- which is roughly 3-to-1 -- VCS is doing well.
The "cash cow" tools (Design Compiler, PrimeTime, PhysOpt, VCS) all have
ratios ranging between 37-to-1 and 3-to-1. Avanti Astro has a 2-to-1.
Aart should be happy, the current money tools are doing well.
The "looking good" tools have more fans than detractors. Vera ratios
at 7-to-1, Power Compiler ratios at 2-to-1, PrimePower 2-to-1, and
DC-FPGA at 2-t0-1. Avanti Star-RCXT ratios at an impressive 8-to-0,
Native Testbench at 5-to-0, Avanti HSPICE at 4-to-0, and RVM 3-to-0.
Not one negative vote! They must be doing something right. And I
laughed when IC Compiler came in at 4-to-0, too. There were only 4
IC Compiler users at SNUG'05. Guess they liked it.
The "OK" tools, those that ratio roughly 1-to-1, are: TetraMAX, Jupiter-XT,
PrimeTime-SI, Magellan, DFT Compiler, and NanoSim. This is a big step up
for Jupiter-XT; last year it was a 0-to-11 tool!
The "needs improvement" tools have ratios of 1-to-something-greater-than-1.
That is, there's at least some small group of users liking it, but most
hate it. Hercules and Astro Rail both ratio at 1-to-2. In 2003, Formality
ratioed 1-to-20. In 2004, it was 1-to-12. This year, it's 1-to-5.
The "disasters", where the tool got 0 "best" votes plus some number of
"worst" votes were: DesignWare at 0-to-9, VirSim at 0-to-6, and LEDA
at 0-to-3. Avanti Cosmos still sucks at 0-to-4. Floorplan Compiler,
VCS-MX, and BSD Compiler all went 0-to-2. Module Compiler did 0-1. I'm
not sure what's going on with that DesignWare 0-to-9; it was only 0-to-2
last year.
"Not everything that is faced can be changed. But nothing
can be changed until it is faced."
- James Baldwin, American novelist, (1924-1987)
Index
Next->Item
|
|
