Analog Research/AR-T

Analog Research => General Discussions => Topic started by: Analog on May 12, 2019, 03:16:58 PM

Title: "Why don't you have my frequency, and when are you going to get it?"
Post by: Analog on May 12, 2019, 03:16:58 PM
We get this question all the time, so having a forum is one way to answer this question once, and be done with it.

In a nutshell, if we don't have your frequency, the odds are we are never going to have it.

Basically, we sell the frequencies that we have, in stock, for our use. We make them available for sale, as there is no way we are going to use all the ones we have on hand.

And why is that?

Because you have to buy these parts in reels of 1000. If you are lucky. If not, then it is a reel of 2000 or 3000. Which, I can promise you, is never going to happen. We are never going to use 1000, so how the hell are we going to use 2-3k? We aren't.

On top of that, we only buy parts that we are confident are going to meet our needs. If they are not, then why buy them? Of course, finding parts that do, when you can only buy reels of 1k is tricky. No one has any of these parts in stock. We have to special order them. Which means we pay for them, up front, and wait for delivery. Which is typically 12 weeks.

So, do you really expect us to plunk down a large pile of coins, wait 3 months, and hope like hell the parts are good, when we are not going to use them? Seriously? Just so we can sell them to you? 

And what if they suck? Are you going to donate funds, to offset our loses? What are we going to do with a reel of 1000 parts that are just as crappy as what any of you can buy, for $1, from someone that has stock? Seriously?

You know the answers.

Yet, I get lots of folks contacting us, asking when we are going to get their frequency. And how good they are going to be, and how much they will cost.

And there is the other problem................................how much they cost.

When we buy a reel, we are out at least $1k. (Which we spend at least 3 months before we even have them! Which also means before we can even dream of selling any.) But, not all of those are usable. Let's say 10% of them are usable. (It is actually much less.)

So, to break even, we have to sell them for $10. And then we have to add in the cost of  testing those parts. For starters, it takes a machine that costs >$10k. And then you have to add in the cost of paying someone (usually me!) who knows what the hell they are doing, can run the machine, and know what to do with the results.

When you take all of that into account, you should be able to figure out why the parts we sell are so expensive, and why there isn't any realistic way it will be lower.

Unless you are willing to accept something that is just a bit less than the best.

Which, to be honest, is so much better than anything that any of you are likely to have used, that there should be a higher demand for them. But, no, everyone wants the very best, and a lot are not willing to wait for us to find them.

Let me explain how this works, in greater detail.

When we buy a reel, we have to measure a sizeable number of the parts, to determine what the yield will be. Having done that, not only can we come up with a price break-down, but we also have more than we will use in the short-term. But, once those are all gone, we need to sort a large number of more parts. Of course, at this point, we usually have a rather substantial number of very good units, that are readily available. But, no, folks only seem to want the very best. While large numbers sit around, collecting dust, that are still very good.

I know what the sonic differences are between the very best, and the ones just a notch below. I'm not sure any of you have equipment good enough to be able to reveal those differences. If not, you have wasted your money. And probably some time.

But, that is your choice. You guys need to be aware of this, and either be willing to wait, or save a few bucks and still be pretty damn happy.
Title: Re: "Why don't you have my frequency, and when are you going to get it?"
Post by: Analog on May 12, 2019, 03:36:29 PM
And that is just for the frequencies that we can use...........................

Putting aside that at some point, we may find some 45/49 MHz parts, that we can sell, that will be better than what any of you can find, even though we will never use them. But only if they are good enough to use, and we can make money selling them.

So, back to "when are you going to get my frequency?"

Like the above post says: "NEVER!"

And there is a pretty good reason for this.

And not just the "We don't need any clocks at that steeeeenking frequency!" excuse.

Folks want 12 MHz...................24 MHz...................25 MHz.....................and others, that are not a multiple of any audio sampling frequency. What the hell are we going to do with them? And more importantly, what the hell are you going to do with them?

Well, about the same. Since all of these frequencies are not multiples of audio sampling frequencies*, what the hell are they doing in your system?

If you have one of these oddball frequencies, they you either have some kind of PLL (and you should be able to figure out why that sucks, for clocks) or you are running in an asynchronous mode. Which also (probably) sucks.

Hopefully, this will help to quell the questions about when we are going to get your oddball frequency in stock.

If not..................

Then post it, here, where I will only have to answer it once and only once. OK?

* = I am surprised at the large number of guys who do not understand the basic clock frequencies used in digital audio. If you guys do not understand this, then maybe you should either put some effort into learning this stuff, or just leave the damn clocks in your gear alone.

Capisce?

I'm good either way.

Well, ok, maybe not.

"Maybe not? How 'bout 'probably never'? Pretty obvious someone crapped in your Cheerios, this morning. Or are you always this cantankerous?"

Uh.....................I would say the latter is the overwhelming favorite.

On any day.

Seriously!
Title: Re: "Why don't you have my frequency, and when are you going to get it?"
Post by: Analog on June 05, 2020, 09:18:07 AM
If we got hold of the so-called "low phase noise" version of the NDK clocks, would you guys be interested?

No idea how many (if any) will be good enough for us to use, but just throwing that idea out there.