Do you like Virtex-5 ?

On Sat, 17 Feb 2007 08:34:30 -0800, Austin <austin@xilinx.com> wrote:

>ricky,
>
>Peter and I take this issue of pricing and availability quite seriously.
>
>We have begun discussions on the issue.
>
>If you think you are upset by this, try to place yourselves in the shoes 
>of the hundreds of hardware ICDES folks, and hundreds of software folks, 
>and hundreds of production test and characterization engineers...you get 
>the picture.
>
>I went to some of the other nominee's websites, and on one, I could: 
>order a sample, order parts, have a representative contact me (the 
>buttons I could click on).  10K pricing was listed.  I think there is 
>significant room for improvement here.

I've said it before - not about Xilinx but about distributors; however
it applies equally well to the online store:

you'll know the store wants to help the small guy when you see a PayPal
button...

- Brian

Jim Granville <no.spam@designtools.maps.co.nz> wrote:

>Peter Alfke wrote:
>
>> Do you like Virtex-5 ? Then please vote for it...
>> 
>> The editors of Electronic Design News think Virtex-5 is an innovative
>> product;
>> they have nominated for their 17th Annual EDN Innovation Awards:
>> 
>>  the Xilinx Virtex-5 Design Team for "Innovator of the Year," and
>> 
>>  the Virtex-5 LXT platform for "Innovation of the Year" (in the
>> Digital ICs category)
>> 
>> EDN is asking their readers to select the winner.
>> 
>> &#4294967295;         Learn more ( http://www.xilinx.com/prs_rls/2007/silicon_vir/
>> 0727_ednawards.htm )
>> 
>> &#4294967295;         Vote now! ( www.edn.com/innovationballot.asp )
>> And confirm your vote as soon as you receive an automated email from
>> EDN.
>> 
>> Peter Alfke, Xilinx Applications, and proud of it !
>
>I see a pretty small category :
>( where do these guys get their candidates ? - could it be advertiser 
>revenue ? )
>  They also miss a [none of the above] vote, so readers cannot indicate 
>if they agree with the shortlists.
>
>Digital ICs, programmable logic, and memory:
>  PEX 8548 PCI Express switch (PLX Technology)
>  MR2A16A MRAM (Freescale)
>  Virtex-5 LXT FPGAs (Xilinx)
>
>and since the award is for innovation :
>
>Adj.	
>1. innovative - ahead of the times;
>2. innovative - being or producing something like nothing done or 
>experienced or created before;
>
>then the clear winner (by a large margin) is the MRAM.
>
>That's far closer to innovative than "another iteration in FPGAs" ?

MRAM is not new either. Ramtron has been making similar parts for
years. 

I think the PCI Express switch is the most innovative part because it
links hardware like a real network. Since PCI express is a serial bus
it can be extended by several feet, it might replace USB and firewire
in the very near future. A PCI Express switch which has configurable
ports is a key element for such a replacement.

-- 
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl

Nico Coesel wrote:
> Jim Granville <no.spam@designtools.maps.co.nz> wrote:
>>
>>then the clear winner (by a large margin) is the MRAM.
>>
>>That's far closer to innovative than "another iteration in FPGAs" ?
> 
> 
> MRAM is not new either. Ramtron has been making similar parts for
> years. 

Ramtron's devices are FRAM, which I believe use a different method.
They have a volatile read, and smaller devices / slower speeds.
Freescales devices are the first of a technology line.

I also saw recent news from Hitachi/renesas that their Phase Change 
memory was in a real device
http://home.businesswire.com/portal/site/google/index.jsp?ndmViewId=news_view&newsId=20070216005599&newsLang=en

20ns Tacc, and  416KB/s write speeds - no mention yet of joules/byte 
read, or yields, but a 4MBit/1.5V device is good to see.


> I think the PCI Express switch is the most innovative part because it
> links hardware like a real network. Since PCI express is a serial bus
> it can be extended by several feet, it might replace USB and firewire
> in the very near future. A PCI Express switch which has configurable
> ports is a key element for such a replacement.

So that moves Virtex-5 into last place ?

-jg

Jon Elson wrote:
> The distributors have warehouses full of non-RoHS compliant parts they
> need to get
> rid of before they can fill them back up with more stuff they won't be
> able to move.
The way I see it, they don't even have warehouses anymore in most cases.
They forward your order to the chip manufacturer and put a new sticker
on the package before they forward it to you.

That's why you have lead times of 20 weeks and why it's so hard to get
chips in decent quantities (i.e., not hundreds of thousands of parts),
because that would require them to a) really stock some parts (because
you can't bother the manufacturer with thousands of small orders for a
handful of chips) and b) repackage them in the quantities ordered. And
that's too much effort nowadays.

They used to do it, and still do it for $1000 FPGAs, but I guess with
the myriad of different packages, speed grades, RoHS/non-RoHS and so on,
there's no way to stock nearly enough different parts...

If you need things like cheap SDRAM-memory chips, you're screwed in most
cases. Getting a few samples is OK, but after that it's buy 1000 or nothing.

-- 
My email address is only valid until the end of the month.
Try figuring out what the address is going to be after that...

Sean Durkin wrote:

>Jon Elson wrote:
>  
>
>>The distributors have warehouses full of non-RoHS compliant parts they
>>need to get
>>rid of before they can fill them back up with more stuff they won't be
>>able to move.
>>    
>>
>The way I see it, they don't even have warehouses anymore in most cases.
>They forward your order to the chip manufacturer and put a new sticker
>on the package before they forward it to you.
>  
>
Umm, maybe.  Check the return address on sample parts you get from TI and I
think Analog Devices, at least.  Hmm, 701 Brooks Ave. South, Thief River 
Falls,
MN.  Isn't that the address of Digi-Key?  What a coincidence!

So, at least in some cases, the manufacturer doesn't keep their own 
stock, or just
doesn't want to deal with the hundreds of little envelopes going out.

>  
>
Jon

On Feb 18, 12:34 am, Austin <aus...@xilinx.com> wrote:
> ricky,
>
> Peter and I take this issue of pricing and availability quite seriously.
>
> We have begun discussions on the issue.


Well, I guess finally someone notice it..

Obviously, Avnet and NuHorizon has bad attitude towards SME and
consultants (and to those who dont want to buy 1000 pieces for
prototyping).

If I run Xilinx, I would choose another distributor that's willing to
help
prototyping industries or there are no more mass quantity order for
Xilinx's. (no prototype, no mass product).

-kunil

kunil,

Just one small problem:  what other distributor(s)?  Seems that the
distributor business has "consolidated" down to two major distis.

One carries Altera, and one carries Xilinx.

I know that in some parts of the world, there are still more than these
two choices, but increasingly we will see consolidation even in those
marketplaces.

A distributor is increasingly challenged to provide value, and not add
cost.  How is this possible?  If customers are unwilling to pay any
premium whatsoever, and demand the lowest possible prices, they leave no
room for anyone to profit, and as a result there is no shelf stock at
all (anywhere).  And, no services.

"Just in time" becomes "just late."  How does a distributor pay their
200 Field Applications Engineers? (rough estimate of how many FAEs
support Xilinx, we also have a similar number of factory FAEs, and how
do we pay for their work effort if not through the sales price?)

Stockholders freak out if there is more than 120 days of inventory at
the factory (and rightly so).  If we overbuild, at the rate prices erode
on a new process/product, we end up taking a loss if we don't ship
everything as it is built.

Imagine that this week is costs 1/2 as much to sell V5 as it did when it
was introduced.  Well, then all the inventory is now worth 1/2 as much.
 The result is a loss must appear on the financials.  But, if we ship
all the parts before the prices erode, we do not have to write off anything.

Peter and I also understand that small companies sometimes become big
companies.  There are many examples of remarkably successful companies,
and many of them (most of them) use FPGA technology.

How do we continue to support the "onsy-twosy" crowd while also
supporting the Nokia, Cisco, Huawei, Alcatel (etc) crowd?

How do we keep customers happy?

How do we provide the necessary engineering support?

How do we make shareholders happy?

Stay tuned....

Austin

"kunil" <kunilkuda@gmail.com> wrote in message 
news:1171987031.968482.203320@p10g2000cwp.googlegroups.com...

> Obviously, Avnet and NuHorizon has bad attitude towards SME and
> consultants (and to those who dont want to buy 1000 pieces for
> prototyping).

I am a consultant, and have had good luck with Avnet and with
the local rep organization. There's never been too much of a problem
getting prototypes and pilot-production Xilinx parts for any of my
clients.

Have you tried getting in touch with the local rep organization?
A good relationship with them could work wonders. 

Purchase a few hundred chips internally, sell them on your online store 
as not for production use (preferably with the aforementioned paypal 
button) and write off the declining book value as a marketing expense?

Austin Lesea wrote:
> kunil,
> 
> Just one small problem:  what other distributor(s)?  Seems that the
> distributor business has "consolidated" down to two major distis.
> 
> One carries Altera, and one carries Xilinx.
> 
> I know that in some parts of the world, there are still more than these
> two choices, but increasingly we will see consolidation even in those
> marketplaces.
> 
> A distributor is increasingly challenged to provide value, and not add
> cost.  How is this possible?  If customers are unwilling to pay any
> premium whatsoever, and demand the lowest possible prices, they leave no
> room for anyone to profit, and as a result there is no shelf stock at
> all (anywhere).  And, no services.
> 
> "Just in time" becomes "just late."  How does a distributor pay their
> 200 Field Applications Engineers? (rough estimate of how many FAEs
> support Xilinx, we also have a similar number of factory FAEs, and how
> do we pay for their work effort if not through the sales price?)
> 
> Stockholders freak out if there is more than 120 days of inventory at
> the factory (and rightly so).  If we overbuild, at the rate prices erode
> on a new process/product, we end up taking a loss if we don't ship
> everything as it is built.
> 
> Imagine that this week is costs 1/2 as much to sell V5 as it did when it
> was introduced.  Well, then all the inventory is now worth 1/2 as much.
>  The result is a loss must appear on the financials.  But, if we ship
> all the parts before the prices erode, we do not have to write off anything.
> 
> Peter and I also understand that small companies sometimes become big
> companies.  There are many examples of remarkably successful companies,
> and many of them (most of them) use FPGA technology.
> 
> How do we continue to support the "onsy-twosy" crowd while also
> supporting the Nokia, Cisco, Huawei, Alcatel (etc) crowd?
> 
> How do we keep customers happy?
> 
> How do we provide the necessary engineering support?
> 
> How do we make shareholders happy?
> 
> Stay tuned....
> 
> Austin

S,

This is not the "rocket science" part of the business.  Even with all of
the constraints, there should be a way to get our samples to our customers.

Breaking packages into smaller packages means that customers must bake
their parts before they use them, as moisture may have seeped into the
parts when the packages were being re-packaged.

Moisture is death to the part, as the device will "popcorn" when solder
reflowed to the pcb without a pre-bake.

No one wants to open the package (which may contain hundreds of parts),
take one out, put it into a new package, and place it on a shelf for a
customer.

That said, it is done, and done right by people, so it isn't like this
is not possible, it is done everyday.

I wish it was as easy as you suggest, but to insure that we don't
deliver popping corn, we do need to pay attention to the proper handling
of the parts in the distribution channel.  This is not a job for amateurs.

Austin