NEW Keysight HD3

[egonotto]

14-bits is a market ploy that many engineers accept.  THe noise levels aren't anywhere near the 14-bit level...so the scope is digitizing several bits of noise.  the claim still sells as 99% of customers never do the simple math.

I have not tested it yet, but from screen shots I've seen, it can pull out signals it claims other scopes can't.

Hello,

then you have to get the noise significantly smaller, and in my opinion this is only possible through some kind of averaging.

Best regards
egonotto

[Someone]

14-bits is a market ploy that many engineers accept.  THe noise levels aren't anywhere near the 14-bit level...so the scope is digitizing several bits of noise.  the claim still sells as 99% of customers never do the simple math.

I have not tested it yet, but from screen shots I've seen, it can pull out signals it claims other scopes can't.

Hello,

then you have to get the noise significantly smaller, and in my opinion this is only possible through some kind of averaging.

Best regards
egonotto

With all the tricks floating around like forcing on bandwidth limits at highest input sensitivity, claims of low noise are always suspect for the * at the end

[EvgenyG]

I have not tested it yet, but from screen shots I've seen, it can pull out signals it claims other scopes can't.

Other scopes can't, perhaps, because those signals are not there?
P.S. I'd be really keen to see a demo like that.

[nfmax]

Noooo! The advantage of more ADC bits comes when you need to use a higher V/div setting because of the higher peak signal level, but you still want to look at low level details. High Dynamic (HD) range

[tggzzz]

Noooo! The advantage of more ADC bits comes when you need to use a higher V/div setting because of the higher peak signal level, but you still want to look at low level details. High Dynamic (HD) range

Classic example test case: observing the last 0.1% of opamp/DAC/etc settling behaviour.

At these levels, the linearity also needs to be considered.

[2N3055]

Noooo! The advantage of more ADC bits comes when you need to use a higher V/div setting because of the higher peak signal level, but you still want to look at low level details. High Dynamic (HD) range

That is if you presume there is no noise.. Which will be there... It just depends how much..

[CRTbrain]

at higher V/div, the scope will have more noise.  Often there's a probe and probes will add noise.  14-bits is a mkt ploy.

[egonotto]

Hello,

with all the scopes I know, the signal-to-noise ratio is better in the insensitive ranges.

Best regards
egonotto

[nctnico]

Hello,

with all the scopes I know, the signal-to-noise ratio is better in the insensitive ranges.

Very true. It would actually be better to specify signal to noise ratio for each attenuator setting (or range of attenuator settings) rather than a number which is obtained at the most sensitive setting.

[nfmax]

Hello,

with all the scopes I know, the signal-to-noise ratio is better in the insensitive ranges.

Very true. It would actually be better to specify signal to noise ratio for each attenuator setting (or range of attenuator settings) rather than a number which is obtained at the most sensitive setting.

In the old days, when an oscilloscope front end was a switched attenuator ahead of a fixed gain amplifier, this wasn’t the case, so the SNR viewed on screen was the same on each range. I think some people have “internalised “ this without realising scopes don’t work like that anymore

[EEVblog]

then you have to get the noise significantly smaller, and in my opinion this is only possible through some kind of averaging.

Their new ADC and front end is reportedly "half the noise" of competing scopes.

[EEVblog]

egonotto

With all the tricks floating around like forcing on bandwidth limits at highest input sensitivity, claims of low noise are always suspect for the * at the end
[/quote]

The noise is specced over the full bandwidth.

[EEVblog]

14-bits is a mkt ploy.

Not from the screen shots and claims I've seen.
Why is it so surprising? Keysight have always been leaders in scope ADC development.

[egonotto]

Hello,

assuming you have 20 mVpp at 1 V/div, this is 40 pp noise for 14 bits and still 10 pp noise for only 12 bits.  I therefore assume that the subtle differences that can be seen with 14 bit, but not with only 12 bit, are lost in the noise. And 20 mVpp noise would be enormously good.

Best regards
egonotto

[2N3055]

then you have to get the noise significantly smaller, and in my opinion this is only possible through some kind of averaging.

Their new ADC and front end is reportedly "half the noise" of competing scopes.

So it is "half the noise" now?
What happened to "up to 10x less noise"?

 

[pdenisowski]

It has new feature I'm interested in trying out.

Fault hunter?

[tszaboo]

then you have to get the noise significantly smaller, and in my opinion this is only possible through some kind of averaging.

Their new ADC and front end is reportedly "half the noise" of competing scopes.

So it is "half the noise" now?
What happened to "up to 10x less noise"?

 

10 times less than MSOX 3000 with high res mode disabled.
Half of competing modern ones.
I like that scope but it's not very quiet in modern standards.
And standard probe just as noisy, quiet one 3K EUR.
At least it's my 2c

[2N3055]

then you have to get the noise significantly smaller, and in my opinion this is only possible through some kind of averaging.

Their new ADC and front end is reportedly "half the noise" of competing scopes.

So it is "half the noise" now?
What happened to "up to 10x less noise"?

 

10 times less than MSOX 3000 with high res mode disabled.
Half of competing modern ones.
I like that scope but it's not very quiet in modern standards.
And standard probe just as noisy, quiet one 3K EUR.
At least it's my 2c

My 3104T has 300µV RMS noise at 1mV/div... so more like 5x

I mean don't get me wrong. If you have half the noise that is not something to sneer at.  But no need for outrageous 10x claims without defining the reference point.  I'm sure that I could engineer artificial circumstances where any of the scopes I have here in the lab would have 1/10th of the noise against something...

[tszaboo]

My 3104T has 300µV RMS noise at 1mV/div... so more like 5x

I mean don't get me wrong. If you have half the noise that is not something to sneer at.  But no need for outrageous 10x claims without defining the reference point.  I'm sure that I could engineer artificial circumstances where any of the scopes I have here in the lab would have 1/10th of the noise against something...

I have 1.27mV RMS noise with 20M bandwidth with the standard 1:10 probe connected and shorted out with the ground loop.
129uV RMS with my PRP1 prototype probe connected and shorted input.
Regular 50 Ohm input is 111uV. So it all depends on how you measure it.
So it could be x10 because, after all, it comes with the x10 probes...

[2N3055]

My 3104T has 300µV RMS noise at 1mV/div... so more like 5x

I mean don't get me wrong. If you have half the noise that is not something to sneer at.  But no need for outrageous 10x claims without defining the reference point.  I'm sure that I could engineer artificial circumstances where any of the scopes I have here in the lab would have 1/10th of the noise against something...

I have 1.27mV RMS noise with 20M bandwidth with the standard 1:10 probe connected and shorted out with the ground loop.
129uV RMS with my PRP1 prototype probe connected and shorted input.
Regular 50 Ohm input is 111uV. So it all depends on how you measure it.
So it could be x10 because, after all, it comes with the x10 probes...

My measurement is with 1:1 probe and full 1GHz BW because that is what was claimed for HD3.

[ITArchitect]

Ah, these times make me wish I was rich and famous so I could afford it.  I better start saving cause I'm sure it's way out of my league!   

[Martin72]

For me it's the other way around, I haven't read anything here that would knock my socks off.
In both cases, we'll have to wait until actual information is available, which won't be long now.

[electr_peter]

There is blank space for fifth BNC (?) or something on the front panel. What is it intended for?

[NE666]

There is blank space for fifth BNC (?) or something on the front panel. What is it intended for?

This is to allow the customer to bring out a temperature proportional drive signal for a replacement cooling fan, to be held on by a self-printed bracket, since the OEM fan noise will of course be too loud for the ears of serious scope collectors..

[Martin72]

This can also simply be an open space.
If the Scope has an external input at the back, it would have been great if keysight had moved it to the front in this free area.
It has become a bad habit to put something like this at the back.