Rigol DG4000 series counter

[jpb]

I've a feeling this has already been covered but I couldn't find it by searching.

I was just looking at the specs for the Rigol DG4162 and I see the counter is specced at 6 digits/sec (gate time of 1sec) but it shows 12 digits all the time!?

The default gate time is only 1.3ms which presumably means only the first three digits are meaningful?

To get 12 digits worth you'd need to set a gate of 11.5 days? (Or is my maths wrong?)

I remember on another thread someone saying that they got 9.999 999 ... for a Rubidian 10MHz source but this apparent accuracy is presumably
a fortuitous combination of it using binary and counting a whole number that is pretty close.

[AndyC_772]

You can set the gate time - but if you're looking for accurate frequency measurements I couldn't recommend the DG4xxx family. There are definitely some rounding / quantisation issues with the displayed frequency, it's not nearly as precise as it could or should be.

[EV]

I have also wondered this earlier and tested it again. The counter gives 7 correct digits with Rubidium standard as external clock and 10 s gate time. The accuracy is quite the same with internal clock. So why are there 4..5 irrelevant numbers? 

[jpb]

I'm also taken aback by the number of digits. I would guess it is similar to scope counters which are typically 6 digits but only show 6 digits. It seems to be a marketing rather than engineering addition. I can imagine one of the Rigol design engineers suggested that they could easily add a simple six digit counter, the marketing department then looked around and saw top quality counters had 12 digits and it all morphed from there.

In a similar vein, last time I was in Tokyo I noticed in the taxi the dimensions, such as height, on some official plaque were given to the nearest mm!

[KedasProbe]

The Standard Deviation shown in the statistics gives you a clear view of the accuracy.
So I don't mind more digits.