NEW Keysight HD3

[tv84]

There's something unfinished about the front panel design.  Like it's been rushed out, or the tooling modified at the last minute or a major rejig of the PCB.

This detail is beyond belief! WTF authorized such design for a multi-k$ KS scope? Not even my chinese DMM have something like this in their back panel!! I can't take my eyes of it... I'm sorry.

(I wonder, if Rigol/Siglent would do such a thing, how long would be the size of the thread at this point. Not to mention the amount of the not-included essential SW features described in that first video a few posts earlier...)

[Wolfgang]

The SDS3000X HD is specced at 125uV RMS @1GHz, and it's not the best on the market.

the SDS3104X HD has less than 6 mV RMS noise at 1 GHz and 50 ohms and 1 V/div and 1 ms/div and 40 Mpts and 4 Gpts/s. Only LeCroy can beat that. With other time and memory settings, the SDS3104X HD can also be brought below 5 mV noise.

When I get it, I'll measure it. But according the specs, the noise is indeed half of the best competitor on the 2mV range. 1V is not specified in the info I have.
And in case anyone has missed mention of it, the HD3 has more resolution than all the competitors.

Unless it is some kind of high-res mode. Realistically, having more than 12 bits with hundreds of MHz of bandwidth requires an insanely low jitter / low phase noise sampling clock. Otherwise the jitter / phase noise by itself messes up the ENOB big time. So for actual measurements it makes sense to trade bandwidth for more bits / resolution.

HD3 ENOB of 10.4 at 20MHz/-1dB FS is not really spectacular for a 14Bit converter. Maybe they still have too much noise ?
Their claim is they have 2 bits more resolution (factor 4), but they claim to have only "halfed" the noise ... OK, tells something.
When you look at RS MXO4/5 they have similar ENOBs with 12Bit converters.

[EEVblog]

Shooting the first impression video now, and yes I will edit it.
Anyway, found my first problem, the waveform veritcal knob doesn't work properly when you have math turned on. As in I can't even get it to the bottom of the screen if I move it too quickly!
Also, it only has a 40MHz channel bandwidth selection, not 20MHz. Why?

That's twice the bandwidth, bigger is better.
What do you mean you want to compare your measurement?

The industry has standardised on 20MHz for decades, and this is commonly used for noise measurements for example.

[shabaz]

There's something unfinished about the front panel design.

I was wondering what it was, too. That long recess containing the BNC sockets is off-center; it's protruding, for no apparent reason, into the control panel on the right. (EDIT: just realized @tv84 mentioned this above too). I think that's (maybe subconsciously) unsettling in the eyes. Keysight must have had an excellent reason since they are not novices to industrial design. It does feel that perhaps their plans changed at some point, and it was not possible to modify. Plus, their customers won't care about that too much.

If this is a 3-series 'scope, then the 10" display is possibly reasonable. I think it must be a 3-series at heart since there are some things that make no sense were it to be a 4-series 'scope. Another feature that's also quite unsettling is the AWG capability at 8k points. This is far less than my decade-old cheap Rigol AWG (which I sold recently, making it great value for money overall). On the MXO 4, it's 40M points. I needed at least a few hundred k points when I wanted to simulate a LoRa signal (at an intermediate frequency), and that was easily achieved on the MXO 4.

[nctnico]

There's something unfinished about the front panel design.  Like it's been rushed out, or the tooling modified at the last minute or a major rejig of the PCB.

This detail is beyond belief! WTF authorized such design for a multi-k$ KS scope? Not even my chinese DMM have something like this in their back panel!! I can't take my eyes of it... I'm sorry.

(I wonder, if Rigol/Siglent would do such a thing, how long would be the size of the thread at this point. Not to mention the amount of the not-included essential SW features described in that first video a few posts earlier...)

That is a good point for sure. If you'd say the release is half assed, I'd agree.    Having the same bandwidth limit applied to ALL channels 

[tszaboo]

Shooting the first impression video now, and yes I will edit it.
Anyway, found my first problem, the waveform veritcal knob doesn't work properly when you have math turned on. As in I can't even get it to the bottom of the screen if I move it too quickly!
Also, it only has a 40MHz channel bandwidth selection, not 20MHz. Why?

That's twice the bandwidth, bigger is better.
What do you mean you want to compare your measurement?

The industry has standardised on 20MHz for decades, and this is commonly used for noise measurements for example.

Once again, it was a joke.

[Martin72]

HD3 ENOB of 10.4 at 20MHz/-1dB FS is not really spectacular for a 14Bit converter. Maybe they still have too much noise ?
Their claim is they have 2 bits more resolution (factor 4), but they claim to have only "halfed" the noise ... OK, tells something.
When you look at RS MXO4/5 they have similar ENOBs with 12Bit converters.

If I have understood correctly, the Hi-Res mode does not go beyond 16 bit.
As a rule, you get at least 3 bits more in hi-res mode, in some cases even more.
(Siglent SDS3000X HD +4bits).
In contrast, the 14+2 bits of the Keysight Scope seem a little “strange”.
But even that is just speculation, we basically know nothing about it.
This will probably come to an end in 2 weeks.

[nctnico]

The SDS3000X HD is specced at 125uV RMS @1GHz, and it's not the best on the market.

the SDS3104X HD has less than 6 mV RMS noise at 1 GHz and 50 ohms and 1 V/div and 1 ms/div and 40 Mpts and 4 Gpts/s. Only LeCroy can beat that. With other time and memory settings, the SDS3104X HD can also be brought below 5 mV noise.

When I get it, I'll measure it. But according the specs, the noise is indeed half of the best competitor on the 2mV range. 1V is not specified in the info I have.
And in case anyone has missed mention of it, the HD3 has more resolution than all the competitors.

Unless it is some kind of high-res mode. Realistically, having more than 12 bits with hundreds of MHz of bandwidth requires an insanely low jitter / low phase noise sampling clock. Otherwise the jitter / phase noise by itself messes up the ENOB big time. So for actual measurements it makes sense to trade bandwidth for more bits / resolution.

HD3 ENOB of 10.4 at 20MHz/-1dB FS is not really spectacular for a 14Bit converter. Maybe they still have too much noise ?
Their claim is they have 2 bits more resolution (factor 4), but they claim to have only "halfed" the noise ... OK, tells something.
When you look at RS MXO4/5 they have similar ENOBs with 12Bit converters.

Yep. And the Batronix Magnova is not far behind where it comes to noise and is ahead where it comes to ENOB according to the specs.

[2N3055]

There's something unfinished about the front panel design.  Like it's been rushed out, or the tooling modified at the last minute or a major rejig of the PCB.

This detail is beyond belief! WTF authorized such design for a multi-k$ KS scope? Not even my chinese DMM have something like this in their back panel!! I can't take my eyes of it... I'm sorry.

(I wonder, if Rigol/Siglent would do such a thing, how long would be the size of the thread at this point. Not to mention the amount of the not-included essential SW features described in that first video a few posts earlier...)

It was rushed out. Many things don't work yet. Monitor out does not. Lot's of stuff from datasheet is "yet to come"..
And that case does look funny...

[egonotto]

....

If I have understood correctly, the Hi-Res mode does not go beyond 16 bit.
As a rule, you get at least 3 bits more in hi-res mode, in some cases even more.
(Siglent SDS3000X HD +4bits).
In contrast, the 14+2 bits of the Keysight Scope seem a little “strange”.
...

Hello,

I assume they only have a 16 bit wide sample size.
Otherwise the binary data format of older devices would no longer fit.

Best regards
egonotto

P.S. Maybe this is wrong if the data is float.

[EEVblog]

cool! but the boot sequence was edited out! i wanted to see if it had the megazoom splash or something else, since they got rid of the logo and all!

The boot screen is, well, kinda embarrassing 

[EEVblog]

If I have understood correctly, the Hi-Res mode does not go beyond 16 bit.
As a rule, you get at least 3 bits more in hi-res mode, in some cases even more.
(Siglent SDS3000X HD +4bits).
In contrast, the 14+2 bits of the Keysight Scope seem a little “strange”.

I assume they only have a 16 bit wide sample size.
Otherwise the binary data format of older devices would no longer fit.

Correct 16 bit max only HD mode.

[EEVblog]

I looked briefly at the file, there are a couple of problems. First it has only 62500 points in it, not 1M.  Perhaps some decimation happens in saving. Also 100 MS/s rate is not sufficient to get accurate noise spectrum up to 1 GHz, so it becomes aliased. You can reduce the time/div until the sample rate is maximum or see what happens if one turns on the bandwidth limiting filter.

Brain fart.
Here it is again at 3.2GS/s 1M memory depth setting.
But I'm still only getting the same file size .BIN file (125KB). Not sure what's going on here.

[egonotto]

Hello,

thank you, that's 32 kpts. Exactly the data that corresponds to the screen.
The noise is really very low.

Best regards
egootto

[SWB]

The subtle awkwardness in the industrial design is striking when compared with the rest of the family.

Clearly something's missing that was either originally intended or still yet to come.

Also, even if they need(ed) to keep that space, they could have shifted the entire row to the left a bit to center it under the screen. That would have looked better. (However, they'd either have to move the power button somewhere else or increase the blank space to the left of the screen to make room for it, either of which would be unfortunate.)

[EEVblog]

thank you, that's 32 kpts. Exactly the data that corresponds to the screen.
The noise is really very low.

Unfortunately I cannot get it to single shot trigger on the noise, where resumably it would take the full 1M capture.
Keysight have mentioned improved trigger sensitivity is coming, but at the moment I can't do it 

[EEVblog]

thank you, that's 32 kpts. Exactly the data that corresponds to the screen.
The noise is really very low.

Unfortunately I cannot get it to single shot trigger on the noise, where resumably it would take the full 1M capture.
Keysight have mentioned improved trigger sensitivity is coming, but at the moment I can't do it 

I single shot captured another waveform with fixed 1M point memory and even specified the Save size to the 1M point and it STILL gave me a 125KB binary file! 
Presumably another bug?

[maxwell3e10]

That's annoying, if one can't get raw data out of a digital scope then it doesn't exist. Strangely, the first binary file at 100MS/sec was 250kB.

[Anthocyanina]

There's something unfinished about the front panel design.  Like it's been rushed out, or the tooling modified at the last minute or a major rejig of the PCB.

This detail is beyond belief! WTF authorized such design for a multi-k$ KS scope? Not even my chinese DMM have something like this in their back panel!! I can't take my eyes of it... I'm sorry.

(I wonder, if Rigol/Siglent would do such a thing, how long would be the size of the thread at this point. Not to mention the amount of the not-included essential SW features described in that first video a few posts earlier...)

and did you see this bit yet at the trigger controls? the spacing between slope and its light indicator, and mode and its light indicator isn't aligned at all with the buttons above. and the force trigger button also not centered with the knob, and not centered with the knob's text below. it looks unfortunately sloppy

[EEVblog]

and did you see this bit yet at the trigger controls? the spacing between slope and its light indicator, and mode and its light indicator isn't aligned at all with the buttons above. and the force trigger button also not centered with the knob, and not centered with the knob's text below. it looks unfortunately sloppy

THey didn't bother me. Obviously they had to move the button to fit the indicators next to those buttons.

[KE5FX]

Seems like a bit of a faux pas, all right.  You never put two or more objects or details almost in line with each other.  Either separate them or align them.  Even the masters seem to screw this up every so often:



It'd bug me, but not enough to cost them a sale if the rest of the product is well-executed.

[artur0089]

I would like to see the trigger frequency on different time bases? Preferably with an oscillogram. A list (at least a few points faster than 100 µs/div) is possible
https://www.eevblog.com/forum/testgear/first-impressions-and-review-of-the-rigol-ds2072-ds2000-series-dso/msg160064/#msg160064
or a graph https://www.eevblog.com/forum/testgear/rigol-mso4000-and-ds4000-tests-bugs-firmware-questions-etc/msg973064/#msg973064 .
1.3M waveforms/sec is not an indicator for the entire frequency path...

[Someone]

Unfortunately I cannot get it to single shot trigger on the noise

the force trigger button

A dedicated button right there on the front panel!

[georgd]

Let's steer the discussion from subtle design flaws and unfinished software to finding what's so revolutionary about this device, or are the key capabilities not yet unveiled?
Excessive expectations tend to lead to disillusionment.

Georg

[EEVblog]

Let's steer the discussion from subtle design flaws and unfinished software to finding what's so revolutionary about this device, or are the key capabilities not yet unveiled?

14bit ADC, half the noise, and way faster updating. I just tested it in comparison to the MXO doing noise measurments (yes, it less than half the noise) and the measurement update was over 7 times faster on the HD3.