Author Topic: New Agilent Scope--Unexpected deficiency  (Read 3173 times)

0 Members and 1 Guest are viewing this topic.

Offline The Electrician

  • Frequent Contributor
  • **
  • Posts: 547
  • Country: us
New Agilent Scope--Unexpected deficiency
« on: February 06, 2013, 10:44:43 pm »
I got an email from Agilent about their new 9000H series scope with 12 bit vertical sampling:

http://www.home.agilent.com/en/pd-2227294-pn-DSO9024H/high-definition-oscilloscope-250-mhz-4-channels?nid=-33085.1042675&cc=US&lc=eng

But check out the waveform update rate on page 13 of the data sheet:

http://cp.literature.agilent.com/litweb/pdf/5991-1520EN.pdf

Only 2k waveforms/sec?  After all the noise Agilent makes about the importance of this capability in a modern DSO?
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 7518
  • Country: us
    • SiliconValleyGarage
Re: New Agilent Scope--Unexpected deficiency
« Reply #1 on: February 06, 2013, 11:48:53 pm »
nah. it can do up to 250K waveforms per second.
see the full details. update rate changes depending on what mode you are in.
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline The Electrician

  • Frequent Contributor
  • **
  • Posts: 547
  • Country: us
Re: New Agilent Scope--Unexpected deficiency
« Reply #2 on: February 07, 2013, 12:08:47 am »
Page 23 of the data sheet shows that the 250k waveforms/sec only applies to segmented mode.  Look at the real time mode specs on that page.

Agilent makes much of the fact that the update rate of the 5000 and 6000 series scopes remains nearly constant at about 100k waveforms/sec no mater what mode you're in.  Why for this new and very capable scope would they not consider a high update rate in all modes worth doing?
« Last Edit: February 07, 2013, 12:12:01 am by The Electrician »
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 7518
  • Country: us
    • SiliconValleyGarage
Re: New Agilent Scope--Unexpected deficiency
« Reply #3 on: February 09, 2013, 04:18:41 pm »
Probably because of the 12 bit convertor and the low noise floor... Dunno. Ask them.
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Online jahonen

  • Super Contributor
  • ***
  • Posts: 1046
  • Country: fi
Re: New Agilent Scope--Unexpected deficiency
« Reply #4 on: February 09, 2013, 05:08:04 pm »
My guess is that scope is based on Windows PC platform, so no fancy ASIC graphics processor for super fast rendering to display, like in x/5000/6000 series.

Regards,
Janne
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 7518
  • Country: us
    • SiliconValleyGarage
Re: New Agilent Scope--Unexpected deficiency
« Reply #5 on: February 09, 2013, 06:29:09 pm »
My guess is that scope is based on Windows PC platform, so no fancy ASIC graphics processor for super fast rendering to display, like in x/5000/6000 series.

Regards,
Janne
windows has no impact in the 9x series. they use the same fancy asics. they have a special graphics card with dual ported memory. the scope asics write the traces directly into video ram bypassing even the pci bus. agilent has been doing this on all their windows scopes since the infiniium series was born.

Windows is just there for network/printing and GUI. 
Besides , the 2x 3x and 4x machines are also 'windows' based. It's running on an Stmicro Spear-600 cpu. Windows CE that is. The older 6x and 7x series are vxworks based.

in the 9x series they run windows XP or 7 on a simple Celeron at 1.2 Ghz. they don't need more than that as it's just there for GUI. no scope processing is done on the 'pc' side.

Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline Wuerstchenhund

  • Super Contributor
  • ***
  • Posts: 3040
  • Country: gb
Re: New Agilent Scope--Unexpected deficiency
« Reply #6 on: February 09, 2013, 08:51:04 pm »
My guess is that scope is based on Windows PC platform, so no fancy ASIC graphics processor for super fast rendering to display, like in x/5000/6000 series.

It most certainly has nothing to do with Windows, and besides, the rendering performance of those ASICs would have a hard time to keep up with GPUs under Windows that are more than a decade old.

1M wfms/s may be hot stuff on scopes, but it isn't for PC graphics under Windows or other operating systems.


 

Offline CarlG

  • Regular Contributor
  • *
  • Posts: 154
  • Country: se
Re: New Agilent Scope--Unexpected deficiency
« Reply #7 on: February 09, 2013, 09:51:50 pm »
Since the 9k-series is capable of 250k triggers/s in segmented memory mode, I guess it must be be graphics that's the bottleneck for the update rate. The user's manual states that the display is updated for every trigger in real time mode, which of course severly limits the capture rate.

This is in contrast to the x/6k/7k-series where, if I get it right, the acquired data is preprocessed and reduced in the Megazoom asic before transfer to the graphics processor at a much lower rate than the trigger rate.

My guess is that they only upgraded the front end, not the graphics data processing on the 9000H (compared to the std 9k-series). But that's not much of an excuse for such low update rate in a new scope. Maybe it's enough for marketing point of view.

free_electron, are you certain that the Megazoom (III?) asic is used in the 9000k-series? It would make sense since the 9k came after the 6k/7k models (if I remember correctly). But obviously there's some fundamental difference in how they put the rest together...
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf