WHOA! Microsoft strikes again...

On both boards, it's on a 2x13 pin header on the motherboard. That goes to a ribbon cable, which ends in a DB25S receptacle mounted on a plate.

Also available on eBay. Both boards take 2nd generation i3/i5/i7 CPUs in socket LGA1155. DD3 RAM is quite common on eBay.

I don't see a problem with the parallel port. However, if you want a real problem, try finding a modern motherboard that will work with a

Oh! That works too!

Is DDR3 still the current state of the art? Or is it obsolete already?

Ack!

I've still got a 5.25" floppy somewhere in my garage.

Thanks,

Michael

Only idiots would wait until a machine breaks to have a way to repair it. The potential problem is obvious, so the means of fixing it should be explored before it breaks.

Again, waiting until it is too late is a poor way to run a company. I have a board in production which makes me a lot of money. One chip is EOL and I have to buy from remaining inventory at distributors. Should I wait until the supply is gone before I redesign the board? I think now.

DDR4 is the current state of the art. I haven't seen any yet.

Keep it. They make good book ends and door stops. I suggest you not tell anyone that you have a 5.25" drive, or you will surely be asked to transfer someones ancient 5.25" floppy data to 3.5" or flash memory.

So you have a crystal ball to know that electrolytic capacitor C127 will burst and burn the board on December 27 at 0-eighthundred hours?

Test equipment has nothing to do with EOL'd parts. It has lifetimes of decades. Might be hard to imagine for you but it's fact.

If I have a half million dollar machine that is 20 years old, I would make sure I could buy or had on hand every part needed to fix it.

I have a custom test fixture to production test the boards I produce. I'm not waiting until it breaks to build another. I have several spares ready to go in spite of the fact that they cost me about a kilobuck each.

This makes no sense. The point is I'm not waiting until I am stuck to anticipate and deal with a problem that has major impact. If you can't understand that you are missing a very significant point.

Ever tried that with ... a DC-3, for example?

With sensor fabrication and such a spare can easily cost half a megabuck. Boards of directors will have a cow if you present such inventory levels.

See above.

And as a result the plane can sit on a remote runway for weeks while waiting for the parts to repair it. Is that what you recommend for the half million dollar equipment? How many functional PCs do they have as a backup to the one in use?

Do you need a full replacement for the entire machine? You are actually contradicting yourself. Either the machine needs to keep running or you can tolerate the downtime while you wait for parts and repair. Which is it?

Exactly. You are saying you need to continue to use decades old designs because they can't afford down time when the machine breaks and I am saying they can't afford the down time when you finally have trouble getting replacements for these machines. So you are better off designing a replacement before the machine breaks and you find you can't get a replacement part(s).

No, they do the same we did: They know where to get refurbished vintage parts and they also know the places where they can obtain newly fabricated "replica" parts.

Much better than carrying half a million Dollars worth of inventory.

Four. But they would not be needed because I can easily have one shipped to me overnight. Except that re-installing everything takes a week with Windows 7. With Windows XP it took 1/2 day.

Neither. It is ok for a machine to go down for a day. It is not ok to go down for three weeks.

What we were talking about is vintage-compatible computer gear which you need for vintage production machines. Why would you want to stock something that you can easily buy?

What trouble?

That's not the most efficient way to run a business with production. You may not understand this because you seem to contract out most of your production. I ran a division with a large clean room and a systems production. That's where you quickly learn this stuff because you must report regularly at board meetings. High inventory levels and large capital expense gets you into the dog house, fast.

All that sounds wonderful, but you'll find it's not that easy. For one thing, it does not matter how objectively wonderful the replacement is, it'll never be as good as the original in people's perception.

Leadership has limited mental bandwidth, and this sort of thing never gets the proper due.

Many have, in fact. DC-3s have been re-engined with turbo-props for this very reason. They didn't wait until engines were unobtanium to do the work (though there are still engines available).

If it's required for production, you do what's needed. Halfa-megabuck is nothing compared to betting your company against it failing.

No, Joerg, you're making no sense here.

Both DDR2 and DDR3 are readily available. DDR3 is recommended, though, for new designs (if possible). It's cheaper, too.

SO do I - single-sided. It's in my 5150 PC. ;-)

Of course. But they don't care if the plane is down for a few weeks.

Sure you can get that today. Will they still be around in another 10 years? You talked about running DOS on these machines. At some point it just will no longer be possible because of BIOS differences.

Yeah, you can buy it today. What about in 10 years? But you said they have 4 spare PCs.

When there are no longer PCs capable of running DOS.

I know that in big corporations when you go to that meeting to explain why the line is down when you can no longer get the long obsolete replacements. The question they ask is why wasn't this day planned for? High inventory gets adjusted. Shutting down lines gets people fired.

Yup. Basler ..

When we tested them they had a long backlog, and they're going for a fairly hefty price tag.

The new engines alter the CofG a fair bit so they make some compensating changes.

--sp

They do. Their livelihood depends on them:

Same for the "freight dogs":

Yes.

It hasn't happened so far and I cannot see a reason why it should in 10 years.

I do, not them. Production facilities carry as little inventory as possible. Regarding PCs there is no need to because of the many industrial PC vendors. When a company in Europe was in such a need it took maybe 30sec to find a supplier within driving distance. Because they needed a replacement "now".

I doubt that you or I will be among the living at that point.

As I've explained and as has been evidenced umerous times in my career people are planning for that. You simply go to a legacy dealer and buy a replacement.

Corporate life under P&L responsibility is not that simple. BTDT, many years.

Exactly. This is one of the points Rick does not seem to understand. You can use new hardware with legacy OS, for example. Provided the hardware is designed correctly. You do not have to buy and mothball half a dozen IBM-XT machines.

We plan for that. Just like the DC-3 freight companies do. They have no spare engines on hand but they know where to get one of who can install a turbo-prop retrofit of that's their preference.

The guys (and execs) I worked with always thought I did. And the numbers came in as planned which was their main concern. Those decisions were usually not made by me alone but in a team.

I bought a Durabook "almost mil-spec" laptop a few years ago, running XP. Needless to say it has a native and real RS232 port which came in very handy one day.

A client had an urgent need for a special analyzer and the only way to get one fast was to rent whatever was available. That turned out to be one with a weird optical interface and an RS232 converter. This converter would not work with any of their laptops using any of their USB-RS232 adapters. They asked me to travel to the site anyhow,hoping we could find the problem even if we could never make this analyzer work. I took the Durabook with me, connected it and ... bingo ... we were in business.

Of course, you can still buy laptops with RS232 ports:

I think you're talking past each other now. Of course you don't have to mothball a dozen XTs but you do have to be aware and have a plan for when they do become unobtanium. You might have to, at some point. DC-3 owners *have* been keeping spare engines for decades, and trying to do their best to corner the market on existing engines. The re-engining program is a hail-Mary for when there aren't any viable spares left. It's quite an expensive proposition, as would rebuilding a tool to use a modern PC. Anyone using legacy hardware has to understand the issues and have a plan, which might just be having hot spares of all hardware available, just as DC-3 operators do.

The DC-3 operators still using them commercially *DO* have spare _original_ engines (XTs as the analogy goes) on hand. They're doing their best to corner the market on them, in fact. Re-engining a DC-3 with turbo-props is an extremely costly endeavor.

The numbers come in as planned until the plan doesn't. ...and then they go out of business? We're talking contingency, here. You'd better have a plan.

This is getting a bit silly. You say they need the plane but they can't afford to keep spares of parts they can't buy quickly. In that case they have to live with the down time. There's no other choice.

I have no idea how you can say that with certainty. I don't have a semiconductor on my current production board that I know for a certainty I will be able to buy in 10 years. When none of the parts are still being made for the machine you need, you have no idea if anyone will be able to build a compatible machine.

That's different, you not being able to see. From stock prospectuses, "Past performance is not an indicator of future success".

Seems like a poor tradeoff, $1,000 inventory vs. being down for a day or more while waiting for the replacement.

Until you can't.

Actually it *is* that simple. Just as "No one ever got fired for buying IBM".