Oscilloscopes reach 12-bits, finally

Yawn ... :-)

I have a $600 brick for that, a receiver that runs from close to DC to

3.3GHz and has some real demodulators and real analog filters in there. It has found noise sources that fancy analyzers couldn't.

So who'd be buying the bulk of those? A few R&D folks at the telco infrastructure vendors won't be enough to recoup the NRE.

R&S isn't exactly a newcomer when it comes to RF signal analysis. That company was founded in 1933, it is over a decade older than Tektronix.

--
Regards, Joerg

http://www.analogconsultants.com/

A sigma-delta ADC does not work by just oversampling. It has a feedback which allows to increase the resolution of the one bit converter.

If you could use an 8 bit ADC as a 16 bit ADC by oversampling, they would not sell 16 bit ADCs. Heck, no-one would bother to make anything beyond 1 bit and we'll all be using LM339s to digitize signals.

If the noise is real noise, then your signal will still be signal+noise. By averaging or filtering you just remove the unwanted spectrum from the signal but the inaccuracy remains. It looks cleaner on a scope, but you should ask yourself whether you are seeing the signal or are you filtering away most of the effects caused by choosing a bad grounding point for the probe.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

I know, I was kidding somewhat. Nevertheless it is possible to improve the resolution of even a 1-bit ADC by adding dither and digitally filtering the result.

It is not unheard of to see 10mV of noise when a digital scope is switched to is 10mV/division range (even with its input shorted). Yet one LSB might be 0.3mV. When working on low frequency signals it is a useful feature to be able to filter out this excess noise.

--

John Devereux

Hi JW,

Don't remember that. I used it with high-voltage differential probes and normal 1 MOhm inputs. Ma signals where single shot up to 1 us/div and the pixel line was quite clear (no interpolation)

Never used such a pretty scope later.

Marte