NEW 100MHz Keysight Scope Release on 1st March 2017

[Neganur]

I don't think he actually confirmed that it does anything though, did he? Some "...separate code..." talk

[Keysight DanielBogdanoff]

Not mentioned in the manual, but uou can plug in a USB keyboard for entering label names, annotations

Apparently it's Windows CE based. What about a mouse?

Tried but doesn't do anything....

During the Q&A section of Keysights live draw today, someone asked about a mouse and Daniel confirmed it should work.

My understanding is that a mouse should work. The "zone trigger" question was the "separate code" discussion.

Going back a page or two to the Tek video (because Tek never plays dirty  ), our ASIC is designed to use the on-screen data for measurements, and it wouldn't be trivial to make measurements happen on acquisition data. The 6000X has a "precision mode" which does give more resolution at a cost update rate. But realistically, any scope with any memory depth will have a loss of measurement precision when zooming out far enough. For ours, it's screen-dependent, but for some others it's full-memory dependent. If you're trying to make precise measurements, no matter what scope you're using the signals will have to be scaled appropriately.

For the comparison stuff, that's more of a legacy item that we have to do for non-technical purchasers. We don't really expect engineers to make purchase decisions based off of them, but if you were trying to build a case to your purchasing dept it could come in handy.

@Mike, I have no idea why your scope didn't have the detents, I'd be happy to swap it out for you.

[TheAmmoniacal]

 @Mike @Keysight_DanielBogdanoff So if Mike swops it.. I can get it used and at a discount, right?  Do it, do it!

[RGB255_0_0]

@Mike @Keysight_DanielBogdanoff So if Mike swops it.. I can get it used and at a discount, right?  Do it, do it!

I think Mike's is chipped 

[JPortici]

Not mentioned in the manual, but uou can plug in a USB keyboard for entering label names, annotations

Apparently it's Windows CE based. What about a mouse?

Tried but doesn't do anything....

During the Q&A section of Keysights live draw today, someone asked about a mouse and Daniel confirmed it should work.

My understanding is that a mouse should work. The "zone trigger" question was the "separate code" discussion.

Going back a page or two to the Tek video (because Tek never plays dirty  ), our ASIC is designed to use the on-screen data for measurements, and it wouldn't be trivial to make measurements happen on acquisition data. The 6000X has a "precision mode" which does give more resolution at a cost update rate. But realistically, any scope with any memory depth will have a loss of measurement precision when zooming out far enough. For ours, it's screen-dependent, but for some others it's full-memory dependent. If you're trying to make precise measurements, no matter what scope you're using the signals will have to be scaled appropriately.

For the comparison stuff, that's more of a legacy item that we have to do for non-technical purchasers. We don't really expect engineers to make purchase decisions based off of them, but if you were trying to build a case to your purchasing dept it could come in handy.

@Mike, I have no idea why your scope didn't have the detents, I'd be happy to swap it out for you.

FWIW i think that decoding with screen data is more a concern rather than measurements! (see the endless discussions about it)

[mikeselectricstuff]

@Mike, I have no idea why your scope didn't have the detents, I'd be happy to swap it out for you.

I can tell you exactly why - the two encoder types are very hard to distinguish by eye, they only differ by a very faintly printed part number, resulting in a simple assembly error that wasn't spotted at test/QC.

ISTR the encoders on my old MSO6034A had different colour shafts.
Not a big problem - if you happen to have an encoder or scrap front board  I'll change it but not worth swapping units.
Maybe it will become a rare collectible in time...

[bitseeker]

In Julian's recent video using the 1000X, some of the scope's interaction design seems inconsistent (weird autoscaling while changing values) and cumbersome (not enough configurability and granularity, especially in the frequency response analyzer).

The Bode plot is very coarse. Can the steps for the frequency sweep be made smaller without narrowing the span?

[fanOfeeDIY]

Someone was asking about pixel count, not sure if that query was ever answered.

Per a message back from Keysight, it's 480 x 800.

Scott

Thank you! Scott

I just updated the chart here.
https://docs.google.com/spreadsheets/d/1rtPqMAkNw2bSkqocuc7zcYkl57fsdPVlxzyqcJmbKIc/edit#gid=197124678

from the link here.
https://www.eevblog.com/forum/testgear/digital-oscilloscope-comparison-chart/

Reflecting the changes to other png files on the thread will be after finishing the columns of Function-Generator and External-IF.

[EEVblog]

In Julian's recent video using the 1000X, some of the scope's interaction design seems inconsistent (weird autoscaling while changing values) and cumbersome (not enough configurability and granularity, especially in the frequency response analyzer).

Yes, that is my complaint with it too. Fulfills it's educational role, but it's not a fully configurable replacement for a good analyser.

[EEVblog]

Maybe it will become a rare collectible in time...

Once they hit a certain collectible premium on Ebay you'll have people modding units to remove detents and selling fakes
Reminds me of the fake Red Dot HP35 Calculators

[mikeselectricstuff]

In Julian's recent video using the 1000X, some of the scope's interaction design seems inconsistent (weird autoscaling while changing values) and cumbersome (not enough configurability and granularity, especially in the frequency response analyzer).

Yes, that is my complaint with it too. Fulfills it's educational role, but it's not a fully configurable replacement for a good analyser.

Just tried it, and it does look pretty crusty- only 10 points on the graph makes it a bit of a gimmick.

The function in the PSU analysis on the MSOX3104T allows 50 points, which is rather better, but still only does the frequency in decades.

You can probably get more informative, and certainly faster displays by using a slow timebase, setting up the wavegen for frequency modulation on a ramp with 0% symmetry, and  triggering on the Wavegen modulation source
 

[mikeselectricstuff]

Comparison between 1000X and 3000X - can't see any reason to limit it to 10 points. 1000x also seems very slow to produce so few results.

[mikeselectricstuff]

While hunting for USB LAN interface, I came across this, and yes, it works!

[TheAmmoniacal]

While hunting for USB LAN interface, I came across this, and yes, it works!

So I guess it also saves to a USB memory drive? Like, what normal people use 

[Zbig]

At the risk of going OT, that's one slim floppy drive. I wonder how they managed to make it all (drive, eject mechanism, head actuator, etc.) so low profile. Someone took pride in designing it, seems almost not worth the effort for such an obsolete medium.

[GlowingGhoul]

While hunting for USB LAN interface, I came across this, and yes, it works!

I wouldn't trust my valuable data to 3.5" floppies. They haven't been in the market long enough to have a sufficient track record of reliability.

Stick to tried and tested tech. Luckily, there are MFM to USB interfaces that will allow me to hook up my Shugart 8" floppy drive to my MSOX-3000T.

[Keysight DanielBogdanoff]

The Bode plot is very coarse. Can the steps for the frequency sweep be made smaller without narrowing the span?

I'm pushing for more granularity. I'm being told the settings are like this to make sure the test doesn't take to long, but it seems technically feasible to add more flexibility. I'll be pushing for that (but this is not a commitment).

[TheAmmoniacal]

Silly argument. I can wait (maybe it could display an estimated time for running the test with the current settings?)

[mikeselectricstuff]

The Bode plot is very coarse. Can the steps for the frequency sweep be made smaller without narrowing the span?

I'm pushing for more granularity. I'm being told the settings are like this to make sure the test doesn't take to long, but it seems technically feasible to add more flexibility. I'll be pushing for that (but this is not a commitment).

The MSOX3104 lets you set the number of steps, so the user can tradeoff between time and detail.

[bitseeker]

The Bode plot is very coarse. Can the steps for the frequency sweep be made smaller without narrowing the span?

I'm pushing for more granularity. I'm being told the settings are like this to make sure the test doesn't take to long, but it seems technically feasible to add more flexibility. I'll be pushing for that (but this is not a commitment).

Thanks, Daniel. For fear that it would take too long is an odd reason to limit it so much. That's why progress bars and cancel buttons were invented. Can't wait? Cancel it and reduce the resolution. Anyway, hopefully they'll see the light.

[David Hess]

Or implement the bode plot using a step or impulse from the AWG or an external source.  If a step is used, then the input has to be differentiated before the FFT.  This does not yield the same dynamic range as a swept measurement but it can operate in real time.

The documentation says that the FFT will display the phase information but I am not sure if that is only in bode plot mode and unfortunately differentiation is missing from the math functions limiting its usefulness anyway.

[synthRodriguez]

Bummer, it's got a fan in it. 

Pretty loud too. My application is in a quiet, audio-based environment where I need to hear subtle changes in musical pitch.

The 200 Hz or so fan whine is going to be a real drag. 

Will probably tear into it to change the fan out to something quieter after the warranty runs out. Oh well.

[mikeselectricstuff]

Bummer, it's got a fan in it. 

Pretty loud too. My application is in a quiet, audio-based environment where I need to hear subtle changes in musical pitch.

The 200 Hz or so fan whine is going to be a real drag. 

Will probably tear into it to change the fan out to something quieter after the warranty runs out. Oh well.

Maybe do a water cooling mod 

[David Hess]

Pretty loud too. My application is in a quiet, audio-based environment where I need to hear subtle changes in musical pitch.

The old DSO that I use the most has a custom fan which is both large and looks like something to drive a wind tunnel but is practically silent.  A couple of my most used analog oscilloscopes are completely silent which can be a problem as well; I have left both on at one time or another without noticing for a week.

I do not know why so many test instruments now have such loud fans.  Maybe thermal engineering is considered too simple to be given enough consideration.

[znroot]

But the scope is 2GS/s when it's single channel or also when it is double channel?