5% accuracy for amplitude and rise-fall time is often good enough. Unless the circuit involved absolutely requires extreme precision, most analog designs should work using 10% tolerance parts without difficulty. If they do not, producing them can result in a lot of problems.
Amplitude, frequency, rise-fall time and all that is only the beginning of the amount of information available in the waveform display of a good analog CRT O'scope.
The vertical channel out can be routed to a counter, Spectrum Analyzer, cascaded into the other vertical channel to increase gain trading off noise and BW and... (user beware as the amplifier output on most every instrument can have distortions & noise that can give false reading in instruments like a spectrum analyzer or ...)
If greater accuracy is needed for these types of measurements. There are far better instruments to use like a hp 3403C thermal converter based AC voltmeter, RF power meter, frequency counter, network analyzer, impedance analyzer and...
The traditional basic four items of electronics instrumentation would be Power Supply, Volt-Ohm Milliamp meter (digital or analog), Signal generator and O'scope. What appears to have happened in recent times is the heaping on of features into the modern DSO where it is trying to become time domain instrument, frequency domain instrument, signal source, precision volt meter, precision frequency counter and every other instrument all crammed into one box becoming Jack-Of-All-Trades, Master of None. This is not a bad thing in itself, but users MUST be aware of the inherent limitations when all this stuff is crammed into a single box of instrumentation.
Mostly made possible due low cost reasonable performance to A/D and D/A conversion and lots of computing power with low cost memory, But it still has every limitation baked into a sampled data conversion system.
One more item, most analog CRT O'scopes do not broadcast much RFI conducted or transmitted. I'm not convinced modern PC based instruments produce near zero RFI pollution into it's operating environment which might or might not be a problem.
Bernice
In most applications, the accuracy of 5% of voltage's and rise time's measurements is sufficient.
That is what is obtained with a calibrated analog oscilloscope. ![ThumbsUp :-+](https://www.eevblog.com/forum/Smileys/default/icon_smile_thumbsup.gif)
But for frequencies, this is often insufficient.
Some oscilloscopes have an Y amplifier output for the A channel on the rear panel.
This is the case with the Hameg HM605.
I have a HP5315A counter plugged into the amplifier output Y so I have a precise measurement of the frequency.