Do Xilinx Fix Their Prices?

I prefer the majority of the definitions on here:

The quick definition on that page sums up my understanding:

"noun: (economics) a market in which control over the supply of a commodity is in the hands of a small number of producers and each one can influence prices and affect competitors"

You don't describe the relationship between X and A like the above definition? Wow!

-- Steve

So far in this thread I've been accused of not understanding economics, pricing or capitalism; seemingly just because I've had the audacity to question Xilinx's low quantity prices. Basically, if they're going to patronise me then I'm not going to just sit here quietly and take it.

-- Steve

Steve,

Question away. I think the thread has allowed folks to vent a little, and to learn why selling FPGAs is not as simple as they may have thought. Those that are paranoid, I am afraid we can not help. Regardless, others benefit from an open and frank discussion. I quite often take a mildly extreme position to help focus the discussion and to entertain (after all, why would anyone read this stuff if it wasn't somewhat entertaining?). Anything I have said is not to be taken as an accusation, but rather as a challenge to explain your views (which you did).

If I offended, I apologize, as that was never my intent.

Austin

You are not making any sense. By definition X and A have an oligopoly. So what is your point? Your questions have no point. Your statments are about the obvious. You are not telling anyone here anything they don't already know. You are just acting like a spoiled brat throwing a tantrum because he can't have dessert.

Are you just trolling or do you have a point?

I was responding to Peter Alfke's comments about X and A not being an oligopoly:

"The Columbia Encyclopedia describes oligopoly as: ...the control of supply by a few producers...or by agreements among members of an industry to restrain price competition...

Does that describe your impression of the relationship between X and A ? Wow !"

From:

"The 3S50, 3S200, and 3S400 Spartan-3 devices with 50,000, 200,000, and 400,000 system gates respectively, are available for less than $6.50*. The 3S1000 Spartan-3 device with 1 million system gates is also available for under $12.00*."

The cheapest XC3S400 (400k gates) in small quantities here:

is $28.90, that's 4.45 times as expensive. The difference will narrow when production is ramped up to maximum, but how much will it narrow to? Three times the price, twice the price?

Oh, so you don't mind paying highly inflated prices? Fair enough.

To be perfectly honest the only person ranting round here is yourself.

I've made my point and I can't be bothered to make it again.

-- Steve

I've still not read a good reason why your highest volume products are so expensive in small quantities.

The reasons that have been given up to now have been that small companies should pay the same for support costs as large customers whether or not they use the support. The simple solution to that would be to charge small companies 75 cents per minute when they ring up for support, and let the larger companies have their support unchanged.

The other reas "But rest assured that we are seriously looking at ways to improve the plight of the low-volume customer. Some of your complaints did not fall on deaf ears."

so only time will tell, and it'll be interesting to watch the price of Spartan 3's over the next year or so...

I admit accusation was too strong a word.

No offence taken.

-- Steve

Then later in the year when the best-selling Spartan 3 chips are being shipped in their millions we can all compare the high and low volume prices again....

-- Steve

I think everyone who is interested in using these parts, know the prices. That has been discussed here before. Your posts shed no new light on the matter. Also, as I and others have mentioned before, if you pay list price it is your own fault. Like when you buy a new car, anyone can get a price cut just by asking. But instead of asking, you start spewing here.

Actually, if I am only building 10 boards, no I don't mind paying more for the parts because my time and expense far over shadow the cost of the chips. As others have pointed out, this is not a factor of there being an oligopoly. This is a simple fact of volume production. But then that has been explained to you before.

Good, you have already said it more than enough times.

Why is this such an issue with you? You have never explained how this pricing affects you. Is there a board you want to sell that is priced too high because the qty 10 prices on FPGAs are too high?

He's always got the option of not using FPGAs if he feels there is a better or cheaper approach. If he cannot find a cheaper alternative, then there should be no reason to complain, right? If he can, then he should use that cheaper alternative, and again, there is no reason to complain. Obviously, the price point for FPGAs is not so far off the mark since people do use them.

--

--Ray Andraka, P.E. President, the Andraka Consulting Group, Inc.

"They that give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, 1759

That sounds good comming off the tongue, but could actually be counter-productive to the problem, which is high low volume prices. Much of that business hits MOQs from vendors, and also the 'can't be bothered' threshold in Distis. Adding MORE distis is less efficent, as now their chance of selling stocked devices is less, since the number of customers has not increased. Most would respond by saying - "Sure, we can get that in for you - MOQ is xxx pcs, leadtime is yy weeks"

That's why a web/transaction based model would work better. You focus one stock point for the low volume orders, and rationalize it further by stocking only best spec region devices, with maybe a sprinking of 'absolute cheapest' (slowest, worst temp range). Semiconductor Companies are _already_ doing this - just not yet the big FPGA ones.

-jg

people do use them.

Ray, I think Steve is screaming at the delta between low and high quantity pricing, not necessarily at any specific price. In other words, if the high volume guys paid 50% less than low volume buyers he might not object. He is seeing a 400% differential and that bothers him.

The rationalization might be that there are no real differences in the cost of manufacture for the chips that the small guy is buying (vs. the large buyer). Or the production of tools, documentation, etc. And, armed with that, I think he's saying that he can't see what in the process of getting chips to a small operator can justify a factor of 4x or more.

I can't say that I am in complete disagreement with the idea of questioning pricing practices. However, I have yet to decide not to do a product based on the price of small quantities of FPGA's. Today, thanks to FPGA's, we can design systems that were impossible to realize not too long ago. On a price-per-FF basis you can't beat them. And, to repeat what I said in an earlier post, communication with the chip manufacturer is very important. They are sensitive to your needs and I've found Xilinx (through Austin and others) to be genuinely interested in helping the little guys.

Well, their hearts aren't bleeding for yours either.

They spend 90% of their time going for 10% customers. Or the one customer having 50% of their business...

Homann

Given the calamitous drop in the dollar recently, find a distributor who quotes in dollars and accepts payment in Euros...