Well, I've had a little bit of time to have a proper play with that HP3325A. I'm currently watching a 0.01Hz sine wave crawl across my scope. Rock solid all the way from 0.001 Hz to 20 MHz to better than 1 ppm frequency accuracy with a nice clean sine wave. The 20-60MHz output on the back is a bit ropey and un-sinewave like, but is still spot on frequency wise. Good waveforms on the triangle, square and ramps too. I haven't tried out the external amplitude modulation or phase modulation yet - I can't currently think what use I'd make of those except possibly for impromptu pirate radio on the AM band.
The built in sweep works well in both linear and log modes. It's got a 10V sweep control voltage output that you can use in XY mode with a scope (and a Z blank output too). That's presuming you have a scope with a decent XY mode, which basically implies an analogue scope, which I currently don't. Oh dear, looks like I need to buy one...
Two hundred sobs well spent I think, certainly better than anything new I could have got for twice the money. Oh, and a huge rarity here - all four feet complete with tilting bails.
For those of us that bought one of that batch of TTi TF930 frequency counters a year or two back I can now confirm that TTi's claim that the TF390 can measure down to below 0.001Hz is likely to be true - technically I've only tested it down to 0.001Hz but that worked just fine.
I am NOT going to be testing the HP 3325A's claim of producing a 1µHz sine wave as I don't have any sane way of measuring that. That's a zero crossing every 5 days, 47 minutes, 13 seconds and 333,333.33 µS. Couldn't even have a rough result until midweek, even if I started now.