Hi Steve,
Yes, I was referring to "production scopes", and more specifically to
"conventional analog production scopes". Few people other than you would
know of the few, exotic, exceptions that exceeded the 1GHz limit. I am aware
of scopes other than you cited that exceeded the 1GHz bandwidth but as you
point out they can be quite strange due to the engineering constraints
reality imposes at that frequency.

For example B&H Engineering in Yonkers NY modified the 7104 to push it to
3GHz. They were able to do this because the incredible MCP CRT of the 7104
has a 3dB deflection point in excess of 3GHz (this is documented in the 7104
Service manual). The vertical system of the 7104 was completely removed by
B&H and they developed a single ended 3GHz vertical plugin based on a DC to
3GHz module they designed (which had over 20% ringing at some frequencies).

This vertical amplifier had to bypass the interface connector on the back of
the plugin via SMB connectors that were mounted in 2 of the 4 holes at the
top of the shroud on the rear of the plugin. Once past the interface, I
believe the vertical signal was mated directly to the 7104 delay line. Only
one CRT deflection plate was driven. I have two different B&H vertical
plugins (one has a variable delay line and the other does not) but I have
yet to find a B&H mainframe. As far as I know their scope relied on Tek 7B10
and 7B15 time bases which were not modified so this scope would have had
trouble triggering at anything beyond what those time bases are capable of.

It may seem like the 3GHz capability of the 7104 MCP CRT would make it
possible for Tek to have built an even faster 7104 but that was not the case
for many reasons: The 7000 series was not designed for it and the cost would
be prohibitive to adapt it to higher frequencies; Digital scopes were
catching up rapidly; The low cost dedicated scopes were taking their toll on
the more expensive general purpose laboratory scopes like the 7000.

I don't believe that the 519 has better linearity than the 7104. I haven't
done a test personally although I briefly had a 519 in my home. It wasn't
mine so I didn't dare turn it on for fear that I might damage it. I use the
7104 extensively and have never noticed a single anomaly when compared to 3
7603s, a 7844, 7854, or 7934. John Addis and Bruce Hofer at Tek were issued
US patent 4132958 for the Feedbeside Compensation circuit that corrects the
amplifiers of the 7104 for thermals which is what I think you are referring
to.

I have seen a 519 working at the VintageTEK museum and it would be virtually
impossible to measure linearity on that tiny 2cm x 6cm screen accurately
enough to compare it to the 7cm x 9cm of the 7104. If you can get a more
precise definition of what the EG&G folks were referring to by 'linearity' I
think we can persuade the folks at the museum to do a direct comparison of
the 519 to the 7104 and take some 'linearity' measurements.

Dennis