Products > Test Equipment
New Keysight 4 Channel Scope (1200 X-Series)
<< < (13/17) > >>
JPortici:
As other said deep memory is nice but IMHO overrated. The RTB slows down a bit at 1Mpts ;)
In a new asic i'd rather like to have less input noise and/or more resolution and/or increased VIDEO resolution to support a bigger display

That said, the 3000X is STILL an insanely good scope.
The amount of options and protocol decoders will be found only in scopes from the copetition that have another zero on the pricetag.
The update with the touch screen and zoom trigger and everything was very nice, it was done by someone young enough to understand that a touch screen is not a replacement for a single button mouse :palm:
The ARB generator that while it may not satisfy every user's needs it is perfectly integrated with the scope, another case where the user interface was took seriously and done right.
Berni:
Yes i also think they should up the screen resolution, im used to the  1024x768 of my MSO6000 and yet the X2000/X3000 despite having a much larger screen have less resolution.

The noise and resolution is really just ADC/AFE dependent so the price point of the scope puts some limitations on those.

What i would want instead are some of the Infiniium series features carried over to the low end scopes. Specifically i would want the FFT and High resolution modes. The FFT on those scopes is run like a spectrum analyzer where you just punch in the center,span,rbw freqency and the scope sets up the optimal acquisition settings to get that. Then the high resolution mode essentially makes the scope a lower bandwidth model using downsampling. In the case of the 9000 series this lets you turn a 20GS/s 4GHz 8bit scope into a 500MHz 12bit scope (or anything in between as you can select 9bit 10bit 11bit 12bit). This downsampling is done in hardware before the sample memory so it lets you capture for longer and doesn't cause any of the weird high res behavior that the X3000 scopes do (Where high res is done after it was captured and shifts about as you change the timebase). Sure its not a real 12bit scope but it really does bring the noise into 12bit territory with the flick of a button without any adverse side effects apart from less samplerate/bandwith. When you need your full bandwidth back just turn it back off.

That particular feature was mostly born out of Agilents kneejerk reaction to Lecroys 12bit scopes, but its genuinely a great feature that seamed to have been forgotten now that Keysight developed real 10bit ADCs for scopes.
Mr Nutts:

--- Quote from: Berni on March 11, 2019, 10:31:32 am ---Other scopes slow down when large memory is selected. The Keysight scopes don't even have a option to select less than max memory because the scope keeps running at top speed even when using all of it.

--- End quote ---


Does it really? As far as I know these Megazoom scopes use only a small part of the overall sample memory in a kind of bank switching mode where the scope switches between two small memory blocks (each of a size for enough data to fill the screen) and alternates between them, i.e. while one block is used for sampling the other block gets converted into display data. The only time it uses all available memory is when stop is pressed or when something like zoom is selected.
 
At least that’s what my Infinum DSO8064A seems to be doing, which also has Megazoom but much more memory (128M) ;)
Berni:
The ping-pong memory is used in most scopes out there, not only Keysight ones. Its just classical double buffering so that the scope can be capturing into one buffer while simultaneously drawing the other buffer to the screen.

The Keysight Infinnium scopes do have memory depth selection because they slow down a lot past 1Mpts due to the internal PC being involved in drawing the waveform rather than the ASIC doing all of it. For my particular MSO9000 scope they try to hide this double buffering by allowing the same max capture size for both run and single shot. Its only if you select the 1Gpts setting, then the scope will display a warning that the scope is now limited to single shot mode and that you need to go to 512Mpts for run mode to be available. If you don't have the option license for 1Gpts then you will never see that warning. So if you have a license for 256Mpts it will be 256 both in run or single short mode, but it still drops to 128 when you use more than 2 channels (Again typical scope behavior)

Tho so far i have never used that 1Gpts memory depth for anything useful. The scope is still reasonably responsive viewing a waveform at that depth, but the moment you start adding measurements or math channels to it the thing grinds to a halt, literally taking it over a minute in some cases to process the waveform and show results. But it is pretty cool that it can capture 1 full second of continuous trace at 1GS/s

The MSO8000 series still runs the old Infiniium software as far as i know, so it could be different. Still worth picking up a decent old Infiniium tho as a secondary scope, some can be had pretty cheap.

EDIT: But i do not recommend using a Agilent PC based scope as your main scope, they are kinda clumsy for doing quick poking around a board.
Mr Nutts:
My DSO8064A doesn’t seem to slow down with more than 1M. It gets slow if I select the full 128M but that’s to be expected ;)
 
The Infinum 8000 runs the old XP software but its architecture is different from the 548xx scopes as it doesn’t use two graphic cards with hardware overlay. Apparently the Infinum 8000 is much closer to the 9000 than to the previous scopes.
 
My day to day scope is a Lecroy LT574M but more and more often I use the DSO8064A. If only the tocuhs creen wasn’t that small, without a mouse it’s really hard to use, especially with big hands  :-[
 
 
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod