burn.sir@gmail.com wrote:

> I am in no way surprised that this thread turned into a FPGA vs ASIC
> and then a Xilinx vs. Altera (or Xilinx vs. the world) war. 

Hmm - I don't see much of a war here  ? - have I missed something ?

Austin put some numbers out, and I queried some details of that,
but I had no real issue with the numbers themselves.

It all seemed quite civil to me ?
None of Austin's replies offended me, and I don't think I offended him ?

> What I  don't understand is how come Jim hasn't recommenced any Lattice devices
> yet ;)

and how would that be relevent, to this thread, or why should I ?

> 
> Jokes aside, please stop the madness

madness ?  - I'm sure I missed something ...

-jg

I am in no way surprised that this thread turned into a FPGA vs ASIC
and then a Xilinx vs. Altera (or Xilinx vs. the world) war. What I
don't understand is how come Jim hasn't recommenced any Lattice devices
yet ;)


Jokes aside, please stop the madness

Jim,

Could be (where is Hardcopy included/not?).  I used company reports for
the second posting, and an independent market firm's results for the
first posting.

I am pretty sure the marketing firm has all of Altera's revenues under
PLD (including Hardcopy).  The yearly report clearly states Hardcopy as
4% in 2005.

It is pretty picky of us to not include it (exclude from all
comparisons), but we do so to call attention to the volume of business
that they do, that detracts from their mindshare in FPGAs.  It is as if
they also make buggy whips.  Who cares how much money they make on buggy
whips.  They state in their yearly report, that Hardcopy could/might
become 10% or more of their business.  Given that LSI bailed, and had
$145 million, that might be too low an estimate.  If Altera gained all
LSI business, that would be almost $200 million for structured ASICs in
their basket.

No small number, but from our experience with Hardwire 1,2, 3 etc. that
business is a huge distraction, with poor to non-existent margins.

It does offer customers a path for cost reduction, and its primary value
is in the fact that most customers can't stop changing things, so they
must stick with FPGAs far longer than they wanted to.  Or, they can't
make the Hardcopy work, so they have to use the FPGA for much longer
than they wanted to.  Timing closure is still the number one problem
with any "hardened" approach.

However, when the hardened version doesn't work, the customer usually
demands a price concession on the FPGA until it does work.  Such
'difficult conversations' are conducted on an adversarial basis with
customers and generate negative "goodwill" and are never good for a
company.  I'd rather not be in a business where I have to get tough with
a customer to stay in business!  Once burned, they tend to go away, forever.

I'd rather not deal with a manufacturer that has tough guys used to very
angry yelling and screaming customers that I must negotiate with...

I am surprised that no one pointed out how good 2001 was and how bad
2002 was.  The dot-com bust really burst many bubbles.  Again, for
really good comparisons, you might want to go all the way back to 1985.

Austin

Austin Lesea wrote:
>>I was wondering why the Altera results differed in two areas, but
>>Xilinx's did not. Did the report not clarify that, or was it a typo ?
> 
> 
> Not sure what you are referring to.  One thing is that Hardcopy is
> including in those Altera numbers, and we don't see Hardcopy as anything
> but an ASIC, so it makes it hard to compare how the programmable logic
> sector is doing (is Altera falling more rapidly behind in FPGAs?).

#Looking at PLDs alone (no surprises here):
#Xilinx up 3.7% with 50.3%
#Altera up 7.6% with 33.1%
#Which makes these 5 with 96.8% of a $3.17 billion market
#So, what happens when you combine the Logic ASIC with the PLD markets,
#and look at them together?
#Xilinx up 3.7% with 10.5%
#Altera up 10.6% with 7.2%
#A combined 45.7% of the total market, which accounts for $7.13 billion.

In both blocks, Xilinx is +3.7%, but in one Altera is +7.6%, and the 
other +10.6%, so that was what caught my eye.

The 'second Altera' seems to be slightly larger, which may be tbe $45M
hardcopy you mentioned, included in that block, but not in the first ?

-jg

Jim,

Answers, below,

Austin

-snip-

> I was wondering why the Altera results differed in two areas, but
> Xilinx's did not. Did the report not clarify that, or was it a typo ?

Not sure what you are referring to.  One thing is that Hardcopy is
including in those Altera numbers, and we don't see Hardcopy as anything
but an ASIC, so it makes it hard to compare how the programmable logic
sector is doing (is Altera falling more rapidly behind in FPGAs?).

> Yes, and an even better year for Foundries, who some could consider the
> ultimate 'ASIC suppliers'  :)

Foundries are doing well, right now.  Tough business.  Low margins.
Cycles up and down.  Huge capital costs.  Glad I am not in it.

> Did those numbers include OnSemi/Philips/TI etc standard logic devices ?

No, they did not.  All standard logic products were not included in any
of the numbers.  Neither were memories, microprocessors, nor DSP.

Standard logic in 2005 was $1,579 million total.
DSP was $7,569 million total.
uP+Memories was $104 billion!

As I said, you can make the numbers tell any story you want.

For example, comparing 2005 with 2002: (using company reports)
2002:
Altera, $712 million revenue for the year (No Hardcopy back then)
Xilinx, $1,016 million revenue
2005:
Altera, $1,124 million (4% or $45 million is Hardcopy, really $1,079
million, as noted in their public filing)
Xilinx, $1,573 million

So we grew 1573/1016= 55% and Altera grew 1079/712= 52%.

Could have fun all day with numbers, but I need to get some work done now.

Austin Lesea wrote:

> Jim,
> 
> Any snapshot of just the last year, at the calendar year mark, can be 
> safely ignored if you try to get too much from it.
> 
> If you take the last five years, then maybe you get to see the average 
> trend (which is evident when you look at the overall market share -- 
> hard to get to 50+% without actually showing some pretty impressive 
> growth...).  That Altera is (still) ~33% is also not lost on anyone. The 
> whole market has more $$$, so we all show growth in $$$, even if some of 
> us are not really gaining market share.

I was wondering why the Altera results differed in two areas, but 
Xilinx's did not. Did the report not clarify that, or was it a typo ?

> A quick review of "How to Lie with Numbers" would prove instructive to 
> anyone who wanted to spin the data to their advantage.  That said, I 
> have my own bias, and the way I chose to post the numbers could be 
> considered highly biased (not intentionally, really).
> 
> So, the whole reason for the post was to show that 2004 vs 2005 was a 
> generally good year for PLD suppliers compared with digital ASIC suppliers.

Yes, and an even better year for Foundries, who some could consider the
ultimate 'ASIC suppliers'  :)

> 
> It was also to show that PLDs are about 1/5th the total logic market, so 
> VHDL and verilog are pretty useful to know.

Did those numbers include OnSemi/Philips/TI etc standard logic devices ?

-jg

Jim,

Any snapshot of just the last year, at the calendar year mark, can be 
safely ignored if you try to get too much from it.

If you take the last five years, then maybe you get to see the average 
trend (which is evident when you look at the overall market share -- 
hard to get to 50+% without actually showing some pretty impressive 
growth...).  That Altera is (still) ~33% is also not lost on anyone. 
The whole market has more $$$, so we all show growth in $$$, even if 
some of us are not really gaining market share.

A quick review of "How to Lie with Numbers" would prove instructive to 
anyone who wanted to spin the data to their advantage.  That said, I 
have my own bias, and the way I chose to post the numbers could be 
considered highly biased (not intentionally, really).

So, the whole reason for the post was to show that 2004 vs 2005 was a 
generally good year for PLD suppliers compared with digital ASIC suppliers.

It was also to show that PLDs are about 1/5th the total logic market, so 
VHDL and verilog are pretty useful to know.

Austin

Austin Lesea wrote:

> All,
> 
> Lately I got a database of ICs, analog, digital, etc. market research.
> 
> It allows you to ask questions like:
> 
> "Who are the top 5 digital logic IC companies, in order of 1 to 5, and
> how did they fare compared to last year (2005 vs 2004)"
> 
> Intel +27.2% up from last year to 10.3% of market
> Qualcomm up 7.5% with 5.4% of market
> Broadcomm up 10.4% with 5.4%
> nVidia up 23.8% with 4.4%
> and
> ATI up 6% with 4.4%
> 
> Which makes those top 5 with only 29.8% of the total market, but with
> $14.21 billion in revenues from all Logic ICs.  This is everyone.  Us
> included.
> 
> But, since that is not so narrow as just ASICs, I then narrowed it to
> Logic ASICs:
> 
> IBM down 8% with 15.7% of market
> Fujitsu down 3.8% with 10.7%
> NEC down 26.8% with 9% of market
> Toshiba even with 6.4% of market
> and
> Agere down 3.1% with 6.3% of market
> 
> Which makes for 48.1% of the market, which comprises $5.9 billion (total
> logic ASIC here is $12.34 down 7.2% from last year.
> 
> Looking at PLDs alone (no surprises here):
> 
> Xilinx up 3.7% with 50.3%
> Altera up 7.6% with 33.1%
> Lattice down 7.1% with 6.4%
> Actel up 7.8% with 5.5%
> and
> QuickLogic up 6.7% with 1.5%
> 
> Which makes these 5 with 96.8% of a $3.17 billion market, which overall
> grew 3.6% over last year.
> 
> 
> Hmmm.  Logic ASIC ~4X PLD's ...that's interesting.
> 
> 
> So, what happens when you combine the Logic ASIC with the PLD markets,
> and look at them together?
> 
> IBM down 8% with 12.4%
> Xilinx up 3.7% with 10.5%
> Fujitsu down 3.8% with 8.4%
> Altera up 10.6% with 7.2%
> and
> NEC down 26.8% with 7.1%
> 
> A combined 45.7% of the total market, which accounts for $7.13 billion.
> 
> 
> Conclusion:
> 
> So, if you are wondering if you should learn verilog or VHDL, the
> figures really say you should know both.  Well.  Really well.

.. and learn Quartus, as Altera's growth is well above Xilinx's ? :)

  Indeed, Xilinx's growth seems to be well below the average of the top 
5 in this sector ?

  In one table above, you have Altera as appx 2x Xilinx growth, but
in another shows ~3x - is that because the first table
excludes their hardcopy ASICs ?

  The second table also shows a larger Altera (lower ratio 
Xilinx:Altera) so something extra seems to be included here, that is 
missing in
what you call PLDs alone ?

-jg