>
> "Specific chip pricing has not yet been disclosed, but Xilinx says it
> will be in the range of $2-$35 for Spartan-6 and $57-$2100 for
> Virtex-6. This pricing is for large quantities in the second half of
> 2011, which is when the chips are expected to go into volume
> production."
>
> Second half of 2011????
> It's a long wait!
>
> Luiz Carlos
well it says: expected :)
if something doesnt go so well, the expected will be shifting even
further away into the future
i wonder if S3A ALL devices will be shipping in H1 2011?
or maybe they will be already be obsoleted by that.
S3A VQ100 seems to be still unavailable...
Antti
Reply by oen_br●February 19, 20092009-02-19
From InsideDSP:
http://www.insidedsp.com/Articles/tabid/64/articleType/ArticleView/articleId/294/Default.aspx
"Specific chip pricing has not yet been disclosed, but Xilinx says it
will be in the range of $2-$35 for Spartan-6 and $57-$2100 for
Virtex-6. This pricing is for large quantities in the second half of
2011, which is when the chips are expected to go into volume
production."
Second half of 2011????
It's a long wait!
Luiz Carlos
Reply by Antti●February 4, 20092009-02-04
On Feb 4, 6:15=A0pm, Simon <goo...@gornall.net> wrote:
> On Feb 3, 8:33=A0pm, Eric Smith <e...@brouhaha.com> wrote:
>
> > Simon <goo...@gornall.net> writes:
> > > well, this is all very nice, but more importantly, when are they goin=
g
> > > to actually start delivering these things, and how much will they
> > > cost ?
>
> > > I can't really believe they haven't *got* that information, so why th=
e
> > > staged release of info ?
>
> > When was the last time that a semiconductor vendor gave you such
> > information about a major new product release, and the information
> > actually proved to be reasonably accurate?
>
> Agreed, but then you take the price-estimate the vendor gave for the
> last series, get the 1-off, 100-off, 1000-off price (at launch) from
> your parts-source of choice, calculate the ratio and apply to the new
> vendor price as an approximation :)
>
> Simon
Simon,
simple math approx
cheapest S3A qty one digikey price is 6.8$
from that i would derive 10k price to fall below 4$ (after price
battle negotations...)
if we assume some % price drop for S6
then we get what? 3$
So the price given by Xilinx (smallest S6 qty 10k =3D 3$) and publisched
by Peter Clarke may actually be accurate.
Antti
Reply by Simon●February 4, 20092009-02-04
On Feb 3, 8:33=A0pm, Eric Smith <e...@brouhaha.com> wrote:
> Simon <goo...@gornall.net> writes:
> > well, this is all very nice, but more importantly, when are they going
> > to actually start delivering these things, and how much will they
> > cost ?
>
> > I can't really believe they haven't *got* that information, so why the
> > staged release of info ?
>
> When was the last time that a semiconductor vendor gave you such
> information about a major new product release, and the information
> actually proved to be reasonably accurate?
Agreed, but then you take the price-estimate the vendor gave for the
last series, get the 1-off, 100-off, 1000-off price (at launch) from
your parts-source of choice, calculate the ratio and apply to the new
vendor price as an approximation :)
Simon
Reply by -jg●February 4, 20092009-02-04
On Feb 4, 6:59=A0pm, rickman <gnu...@gmail.com> wrote:
> I will very surprised if they can really sell any of the Spartan 6
> parts for $3. =A0The smallest part they sell has 100 I/Os and who knows
> how much logic. =A0I think it is about 2K LUTs and 4K FFs. =A0From what I
> have been told most of the cost of a small part is from the testing
> and that is dominated by the I/O count. =A0The cheapest part is likely
> going to be around $10 at qty 1000 IMHO.
Sounds about right.
Smallest part says 3366 Logic cells, 9101 for larger sibling, and
largest in TQFG144.
I would paste in the table but Xilinx has disabled COPY in the PDF,
and I would
give a link to a google find that would allow you to View as HTML, but
google
cannot find the named pdf, even when pointed to Xilinx.com.
Seems Xilinx really is doing everything to prevent anyone sharing this
information!!
[I wonder if they even realize this....]
-jg
Reply by Kim Enkovaara●February 4, 20092009-02-04
Eric Smith wrote:
> When was the last time that a semiconductor vendor gave you such
> information about a major new product release, and the information
> actually proved to be reasonably accurate?
Under NDA for bigger customers the vendors give very accurate
information way before the launch and quite accurate estimates
on sample deliveries etc. Of course there are possible slips but
they estimate then new schedules. Many people do designs for
those chips way before the public release, and have own product
launch schedules also.
--Kim
Reply by rickman●February 4, 20092009-02-04
On Feb 3, 9:42=A0pm, -jg <Jim.Granvi...@gmail.com> wrote:
> > "One Spartan-6 FPGA, an LX16 device, is already sampling..."
> > "The Spartan-6 family is priced at between about $3 and $54 in high
> > volume of 10,000 units..."
>
> 10,000 is not really 'high volumes' ?
>
> Xilinx used to quote =A0"Over the Horizon" prices,
> eg for parts 250,000 pcs, delivered at the end of 2010.
>
> Maybe sense finally prevails....?
I agree that their previous pricing was pretty absurd. I wonder if
*any* customers ever got those numbers. Still, I think 10,000 is not
very realistic for many of us, but then anyone buying less than 10,000
is likely not even on the Xilinx radar screen. Personally, I like the
way TI does it listing the 1000 piece price and giving direct links to
the disti's web pages showing real prices.
I will very surprised if they can really sell any of the Spartan 6
parts for $3. The smallest part they sell has 100 I/Os and who knows
how much logic. I think it is about 2K LUTs and 4K FFs. From what I
have been told most of the cost of a small part is from the testing
and that is dominated by the I/O count. The cheapest part is likely
going to be around $10 at qty 1000 IMHO.
Rick
Reply by Eric Smith●February 4, 20092009-02-04
Simon <google@gornall.net> writes:
> well, this is all very nice, but more importantly, when are they going
> to actually start delivering these things, and how much will they
> cost ?
>
> I can't really believe they haven't *got* that information, so why the
> staged release of info ?
When was the last time that a semiconductor vendor gave you such
information about a major new product release, and the information
actually proved to be reasonably accurate?
Reply by -jg●February 3, 20092009-02-03
> "One Spartan-6 FPGA, an LX16 device, is already sampling..."
> "The Spartan-6 family is priced at between about $3 and $54 in high
> volume of 10,000 units..."
10,000 is not really 'high volumes' ?
Xilinx used to quote "Over the Horizon" prices,
eg for parts 250,000 pcs, delivered at the end of 2010.
Maybe sense finally prevails....?
-jg
Reply by Georg Acher●February 3, 20092009-02-03
In article <410e08bf-0eb4-4096-9755-53cfbedb13f1@k9g2000vbl.googlegroups.com>,
oen_br <oen_no_spam@yahoo.com.br> writes:
|> http://www.eetimes.com/showArticle.jhtml;jsessionid=N5KULYMEMNGWUQSNDLOSKH0CJUNN2JVN?articleID=213000271
|>
|> "One Spartan-6 FPGA, an LX16 device, is already sampling..."
|> "The Spartan-6 family is priced at between about $3 and $54 in high
|> volume of 10,000 units..."
According to my experience with Digikey et al., it will then be between $20 and
$300 for a single chip. The Digikey/Avnet/NuHorizon prices are ridiculously
high...
--
Georg Acher, acher@in.tum.de
http://www.lrr.in.tum.de/~acher
"Oh no, not again !" The bowl of petunias