Batboard

OK, folks, for those of you who care, I'd like to get some info about a feature I'm working on for the IFR 2975 service monitor. This is going to get a bit technical, so bear with me please.

Our SINAD and distortion calculations are done using an FFT (fast fourier transform) - the pratical upshot is that we COULD measure SINAD on any frequency, not just the 1kHz default (in fact, if you use the box's remote control language you can change this right now.)

I'm thinking of making some of the parameters of this available to the user on the user interface, namely:

The notch frequency start and stop (this is the range of frequencies the unit considers to be "signal" when computing the SINAD).
The noise frequency start and stop (this is the range of frequencies the unit considers "noise" when computing SINAD)
The window used to condition the FFT (currently NONE or KASIER, but we might add more in the future....)
The alpha for the Kaiser window.

Now, if you don't understand the last two things that's OK - unless you are a signal processing guy you might not care.

Now, here's the deal: if you screw with those settings, you can get bogus readings - so the defaults will be set "sensibly". However, would you guys rather see this:

a) in the zoomed meter windows
b) as a button in the zoomed meter windows that opens a config window
c) as a screen under the System menu
d) not at all.

If you've worked with a dynamic signal analyzer (audio spectrum analyzer) you may have encountered this sort of thing in the past.

I like to set up Fail/Pass setups. It depends on how good your Techs are.
Sinad is a good test for that macro. Also on the test parameters I use indurtry standards. I don't like to weight the results.
If you have good techs, then solution " B " would be my choice.
I think that systems options should be reserved for future options.. Bit error rate and so forth.

Two cent worth.

mm

"I like to set up Fail/Pass setups."

That is already in the system now. You define the limits on the meter, and we will indicate if you are out of range.

" I don't like to weight the results. "

I'm not sure I understand what you are saying here. If by "weighting" you mean applying a C-weight filter or CCITT filter, that is an option we already give you.

If you are confusing the windowing with weighting - Windowing is a different animal all together - it's a function of doing an FFT, rather than a time-continuous filtering of the signal.

"Also on the test parameters I use indurtry (sic) standards"
So do we. However, "the great thing about standards is that there are so many to choose from." That's why we have to make our equipement so flexible.

Thanks for your input - obviously real-world input is the most valuable thing I can get in designing this gear.

Rafael says you are going to make the spectrum analyzer hundreds of times faster. How close to reality is that statement?

Bill says you are thinking about making the box do DHCP. Is that as a DHCP server or client or user choice?

OK, it's official. I must kill Rafee.

We are always looking to improve the box. We have had several complaints about the spectrum analyzer update rates (not the least of all from ME - I did the COM-120B, so if you've worked with a 120B you understand where I place the bar!). However, the update rates on the 2975 are due to some hardware issues that we are looking into. However, I cannot make any promises at this time (and neither should Rafee, damnit!)

As for DHCP:

I've been wanting to put a DHCP client in the box, to allow it to get its IP address via DHCP, for years (it is bug #1 in the bug database, entered 2001-07-12 16:07:09 by yours truely). However, considering that marketing didn't even think anybody would WANT TCP/IP addressability in the box when we started (or a mouse, or scriping, or...) it was never given priority. I'm starting to get some support for that, but it still has not been made a priority (hint: if you all start pounding upon Rob, this may happen.)

Now, as for a DHCP server: That gets to be nasty, in that if we try to be a DHCP server on a network that already has a DHCP server, that is a sure way to a) confuse the world, b) piss off the Network administrator, and c) get us banned from a network.

The only real time us being a DHCP server would be useful would be the case where you are hooking a laptop up to us with a crossover cable and the laptop is configured for DHCP.

Personally, what I suggest is getting a cheap Linksys DSL router/hub and using that - that gives you several ports, DHCP server, and firewalling.