NEW Keysight HD3

[maxwell3e10]

Perhaps a bit more detail on the options and noise
https://mezamall.com/goods/goods_view.php?goodsNo=1000007150

[jayk]

Had to set my VPN to Australia to access, but there is a lot of info here...

[jayk]

Even at this price point, looks like basic serial decode is still an option. 

[mikeselectricstuff]

Even at this price point, looks like basic serial decode is still an option. 

That's an utter fail - no excuse for that level of penny-pinching these days  - UART, I2C and SPI decode are NOT optional for any serious work, especially at this price level.
Though it says MSO license included presumably probes extra?

[jc101]

Actually surprised they don’t make you rent the licences. Seems all the rage these days, because that is what the market wants apparently. Not sure how many marketing people visit the real world, but it can’t be many.

[EEVblog]

Even at this price point, looks like basic serial decode is still an option. 

That's an utter fail - no excuse for that level of penny-pinching these days  - UART, I2C and SPI decode are NOT optional for any serious work, especially at this price level.
Though it says MSO license included presumably probes extra?

Perhaps at this price point they know that most customers aren't going to care?
But yeah, stupid when they are boasting about having a bunch of other big stuff included.

[EvgenyG]

Oh well, the more competition, the better. I'm looking at Siglent SDS3000X HD specs (1Ghz model) vs maxed out HD3:
- No wavegen vs 100meg wavegen
 -12bit vs 14bit
- 890k vs 1.3M waveforms/s
- 400M vs 100M points
- Full BW Noise:125uV at 1mV/div vs 50uV at 2mV/div
- Price: 6.6k vs 22k?

Would be great if someone could compare the 2 and perhaps the MXO4 that has similar specs/price? Ideally, show that MXO4 or SDS3000X fail to do something that HD3 can.

[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.

So it is "half the noise" now?

That's literally what my datasheet says (and shows in the noise table).

What happened to "up to 10x less noise"?

Someone in marketing was having a wank.

[EEVblog]

Would be great if someone could compare the 2 and perhaps the MXO4 that has similar specs/price?

I think you'll find the R&S MXO4 and the HD3 will be pretty close in price point vs spec. The biggies like Keysight, R&S and Tek have never tried to compete in price to the likes of Siglent and Rigol. They aren't going to start now.

But if you must:

Oh well, the more competition, the better. I'm looking at Siglent SDS3000X HD specs (1Ghz model) vs maxed out HD3:
- No wavegen vs 100meg wavegen
 -12bit vs 14bit
- 890k vs 1.3M waveforms/s
- 400M vs 100M points
- Full BW Noise:125uV at 1mV/div vs 50uV at 2mV/div
- Price: 6.6k vs 22k?

- 14 bit is obviously and innovation and better
- The 820k wfms of the Siglent is only in segmented mode, it's 200k normally in a small sweet spot, and even then it's going to be nowhere near the Keysight's Megazoom, even the older Megazoon IV in the 3000X will beat it.
- The 400M is single channel only, it can drop to 100M. Keysight is 100M all the time
- The noise is less than half of the Siglent's
- Keysight has the fully ASIC hardware accelerated serial decoding, FFT, zone trigger, and masking

[tautech]

Oh well, the more competition, the better. I'm looking at Siglent SDS3000X HD specs (1Ghz model) vs maxed out HD3:
- No wavegen vs 100meg wavegen
 -12bit vs 14bit
- 890k vs 1.3M waveforms/s
- 400M vs 100M points
- Full BW Noise:125uV at 1mV/div vs 50uV at 2mV/div
- Price: 6.6k vs 22k?

Would be great if someone could compare the 2 and perhaps the MXO4 that has similar specs/price? Ideally, show that MXO4 or SDS3000X fail to do something that HD3 can.

Add 500 MHz 2ch SDG6052X and still be ~1/2 the cost.

[maxwell3e10]

Old MSO3000X was never particularly low noise, especially at low frequency (around 20 nV/sqrt(Hz)), so I can believe a 10 times improvement is possible for some bandwidth settings. But noise on 1-2 mV/div scale only tests the pre-amp, not ADC. One would need a comparison at 1V/div scale to compare the ADCs.

[pdenisowski]

- Keysight has the fully ASIC hardware accelerated serial decoding, FFT, zone trigger, and masking

Is the FFT hardware accelerated?  Maybe I missed it, but none of the materials posted here so far indicate that.  I'd also be curious if the math functions are done in hardware.

[pdenisowski]

I think you'll find the R&S MXO4 and the HD3 will be pretty close in price point vs spec.

Which kind of begs the question why it's the HD3 instead of the HD4. 

Unless, of course, they're calling it a 3 series to make it "best in class"   1.3M waveforms per second is best in class for a 3 series, but not for a 4 series.

[EEVblog]

- Keysight has the fully ASIC hardware accelerated serial decoding, FFT, zone trigger, and masking

Is the FFT hardware accelerated?  Maybe I missed it, but none of the materials posted here so far indicate that.  I'd also be curious if the math functions are done in hardware.

Ah, I could indeed be wrong about that.

[EEVblog]

Unless, of course, they're calling it a 3 series to make it "best in class"   1.3M waveforms per second is best in class for a 3 series, but not for a 4 series.

Note the marketing term "uncompromised".

[egonotto]

Hello,

the noise at 2 mV/div is really impressive (you can then work with a good preamplifier), but I find the noise at 1 V/div much more important. I haven't heard anything there yet.
I wonder if it's not quite as good?
At 50 ohms you may have several preamplifiers in parallel to keep the noise to a minimum.
Is the bandwidth at 1 MOhm really 1 GHz? Many scopes have 500 MHz or the new MXO[4,5] 700 MHz.

Best regards
egonotto

[CRTbrain]

I just did the math... Keysight 14-bits.... KS gives noise of 55 uV @ 2mV/div which is 16 mV full scale.   Noise is 56X more than smallest ADC resolution @ 2 mV per division.    Many customers will be fooled by the "14-bit" spec as it appears that the instrument is digitizing 6 bits of noise.  Stunning.  Even with 8, 10, and 12-bit scopes noise is typically a bigger issue than ADC resolution, but users can understand an ADC spec much easier than a noise characterization chart.

   

[CRTbrain]

Keysight had a "temporarily" 1 GHz passive probe that was fairly expensive.  They stated they would build it into the next generation scope.  I'm expecting it with HD3.   Historically, Keysight scopes have had more noise on 1 MOhm path than 50 Ohms.   Typically users have a probe when in 1 Mohm path mode...and the probe's noise severely outweighs the scope's inherent measurement noise.

[CRTbrain]

in class ... Keysight slides compare HD3 against Tek MSO 4B and R&S MXO4 who are the leaders in the 4 class.   The "fastest update rate" seems to be just a false claim as R&S MXO4 does 4.5 M wfms/s.   Maybe Keysight will correct before intro.   I like that R&S MXOs show update rate values in the UI.   I would like to see this in Tek and Keysight so it's easier to see what settings impact update rate.

[CRTbrain]

I think they are calling it a HD "3" because they don't have a replacement 3000T anywhere near the price point of InfiniiVision...they have a 50% price premium for HD3 so they have to compare against Tek MSO4 and R&S MXO4 as the HD3 pricing doesn't come close to the 3000T price points.

[CRTbrain]

comparing prices... R&S MXO4 is 15% to 20% less expensive than the new Keysight HD3 in Euros and seems to be superior in specs.  It's priced more like the Tek 5 B or R&S MXO5.   HD3 also seems to be much more expensive than Tek MSO 4 Series B.  All are way more expensive than InfiniiVision 3000T.

4-ch
                     Keysight HD3           R&S MXO4       Tek 4 MSO B
200 MHz        9876 EUR                  8280 EUR       $ 9180
350 MHz       15787 EUR                13080 EUR      $14400
500 MHz       19872 EUR                16900 EUR      $18800
1     GHz       23849 EUR                20930 EUR      $22900
1.5  GHz           NA                        24200 EUR      $28900

[mikeselectricstuff]

comparing prices... R&S MXO4 is 15% to 20% less expensive than the new Keysight HD3 in Euros and seems to be superior in specs.  It's priced more like the Tek 5 B or R&S MXO5.   HD3 also seems to be much more expensive than Tek MSO 4 Series B.  All are way more expensive than InfiniiVision 3000T.

4-ch
                     Keysight HD3           R&S MXO4       Tek 4 MSO B
200 MHz        9876 EUR                  8280 EUR       $ 9180
350 MHz       15787 EUR                13080 EUR      $14400
500 MHz       19872 EUR                16900 EUR      $18800
1     GHz       23849 EUR                20930 EUR      $22900
1.5  GHz           NA                        24200 EUR      $28900

Do R&S include any decodes ?

[Someone]

comparing prices... R&S MXO4 is 15% to 20% less expensive than the new Keysight HD3 in Euros and seems to be superior in specs.  It's priced more like the Tek 5 B or R&S MXO5.   HD3 also seems to be much more expensive than Tek MSO 4 Series B.  All are way more expensive than InfiniiVision 3000T.

4-ch
                     Keysight HD3           R&S MXO4       Tek 4 MSO B
200 MHz        9876 EUR                  8280 EUR       $ 9180
350 MHz       15787 EUR                13080 EUR      $14400
500 MHz       19872 EUR                16900 EUR      $18800
1     GHz       23849 EUR                20930 EUR      $22900
1.5  GHz           NA                        24200 EUR      $28900

Do R&S include any decodes ?

No, they offer "low speed serial" as a 2.5k option. Still better than the individual licenses of Tek where Uart and SPI+I2C are separate at 2.5k each.

[Someone]

Keysight had a "temporarily" 1 GHz passive probe that was fairly expensive.  They stated they would build it into the next generation scope.  I'm expecting it with HD3.   Historically, Keysight scopes have had more noise on 1 MOhm path than 50 Ohms.   Typically users have a probe when in 1 Mohm path mode...and the probe's noise severely outweighs the scope's inherent measurement noise.

The PP0001A "passive" probe that couldn't connect directly (passively) to ANY oscilloscope? If that amplifier doesn't make it into the front end of the HD3 then it's looking pretty dire. Also just catching up to Teks "passive" current mode probes:
https://www.eevblog.com/forum/testgear/can-tektronix-tpp0500-probe-be-used-with-non-tek-oscilloscopes/msg2986066/#msg2986066

[Martin72]

Prices are sound and smoke....
We order directly from Lecroy, for example, and the discounts you get there are something you can only dream of with distributors.
The same applies to R&S, as I recently discovered.
And it won't be any different with Keysight, there's just one big disadvantage.
As a private individual, you can't get access to these discounts.