Author Topic: Agilent Infiniium DSO8000A Hacking - anyone done this?  (Read 26703 times)

0 Members and 1 Guest are viewing this topic.

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3015
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #25 on: May 03, 2016, 07:48:44 pm »
Quote
As far as I know none of these scopes were bandwidth-upgradeable, so the older 54800 Series (which uses the same software) has probably only a few front ends which were artificially bw limited to create the various models.

There were various bandwidth options but I don't know what model(s) they were for. Some said 'enhanced bandwidth' and some seemed to be in xGHz steps but this might be to do with sample rates?.  IIRC the 80404B I briefly played with showed 4GHz 'Standard Bandwidth' on the GUI.
It was also remarkably flat to 4GHz when tested with a sig gen. Above 4GHz the response suddenly drops very sharply.
« Last Edit: May 03, 2016, 07:59:45 pm by G0HZU »
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3015
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #26 on: May 03, 2016, 09:50:53 pm »
I've managed to get close to what I want with the monitor displays...

I don't know how I managed it but it briefly 'worked' in high resolution on the external monitor. I got it to work up to the limit of the monitor by connecting to the other VGA port marked 'reserved'
and I messed with the BIOS to enable this connector. So it should now be the same as the one at work. But the one at work is limited to 640x480 in this mode?

Sadly, I got greedy and tried to get both displays running together. I've almost managed it but I can't get the VNA display on both at the same time. There's just too many things to mess with in the BIOS and the graphics adaptor. This adaptor is much more capable than the default PCI card. But it's fighting me and I can't quite get it to behave properly on both displays.

Maybe I should start a fresh thread on this in case anyone has been here already. It certainly isn't easy to set this up!

But the smith chart looks glorious in high res on the remote monitor!
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3015
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #27 on: May 04, 2016, 03:37:15 pm »
I solved the mystery wrt the E5071Bs at work. All of our 5071B VNAs have only one VGA port at the back. .
They run the mobo graphics and this has 4Mb of video memory. But the system clamps the settings to that of the LCD monitor, i.e. 640x480.

There's probably a workaround but I'm not too fussed. There's nothing about this in the E5071B manuals unless I've missed something. The E5071C VNAs already have a XGA display and they also support dual resolution so I think you can run the external monitor at 1920x1080 etc. But I've not tried.

My E5071B here at home has a mobo VGA port and a PCI VGA port so I can configure a crude fudge that lets me see HD via an external monitor.

Looking in the service manual my VNA has the same PCI graphics card as.....      .... the (low) infiniium scope range (small world)  ;D

Are the Infiniium scopes limited to the resolution of the LCD or can you do similar tricks as I have with my E5071B to get high resolution via an external monitor?


« Last Edit: May 04, 2016, 03:43:27 pm by G0HZU »
 

Offline andy2000

  • Frequent Contributor
  • **
  • Posts: 257
  • Country: us
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #28 on: May 04, 2016, 04:10:04 pm »
The board is a standard intel or AdLink mainboard with Agilent BIOS to enable support for the LVDS to PCIe adapter. Replacing the board means you need a microATX board with PCI that has support for an internal LVDS display and supports WinXP (so you're looking at an industrial mobo). It should also fit the ATX cut-out in the back of the chassis, unless you like to use a hacksaw to make it fit. But in general, yes, it could work.

The real question is why you would want to do that. The current board/CPU combo in the DSO8k/80k is already 64bit, but since the OS is 32bit XP and the drivers only work with 32bit XP there's really no use in having more than 4GB RAM, which is already supported by the current mainboard. Also, since the whole "scope" part is built on Agilent ASICs upgrading the CPU will not change or improve any scope functionality whatsoever. For running XP plus some kind of Office a fast Pentium4 processor and 2+GB RAM are more than adequate, especially when combined with a decent SATA SSD which improves on the boot times as much as possible (the scope app alone prevents this from ever becoming a 'quickstarter' like a scope built on an embedded platform).

It's might not be worth it for the i915 based boards, but one of the main complaints I've heard about the older scopes in the line is that the interface is slow.  Assuming they have a PIII or earlier, upgrading the motherboard could help a lot.  That's assuming there's nothing special about the original motherboard and BIOS.

It might not work though.  I've tried to upgrade the i810 PIII in my TDS5450, but so for it's been unsuccessful.  Everything newer than a PIII board that I've tried (including a board that was used in later TDS5000 scopes) results in ACPI errors on PSOT followed by an ACPI related BSOD.  Apparently, Tek didn't did something not quite to ACPI/PCI standards on the interface board which must conflict with something on newer boards.  I'm going to give it one last try with either an i815 or an Athlon XP board. 
 

Offline WuerstchenhundTopic starter

  • Super Contributor
  • ***
  • Posts: 3088
  • Country: gb
  • Able to drop by occasionally only
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #29 on: May 04, 2016, 04:26:23 pm »
Maybe I should start a fresh thread on this in case anyone has been here already. It certainly isn't easy to set this up!

I definitely think this deserves a thread, although I doubt many people have been there before. An E5071 isn't exactly widespread amongst hobbyists  ;)

Quote
I solved the mystery wrt the E5071Bs at work. All of our 5071B VNAs have only one VGA port at the back. .
They run the mobo graphics and this has 4Mb of video memory. But the system clamps the settings to that of the LCD monitor, i.e. 640x480.

The 4MB is just initial (startup) memory, since this is shared memory graphics it can use part of system RAM as video RAM once the driver is loaded.

As to the resolution, I guess 640x480 is what the panel supports and if the monitor settings in Windows are setup correctly it won't offer any resolution higher than that for that display.

Quote
There's probably a workaround but I'm not too fussed. There's nothing about this in the E5071B manuals unless I've missed something. The E5071C VNAs already have a XGA display and they also support dual resolution so I think you can run the external monitor at 1920x1080 etc. But I've not tried.

I'm sure it will work.

Quote
My E5071B here at home has a mobo VGA port and a PCI VGA port so I can configure a crude fudge that lets me see HD via an external monitor.

Looking in the service manual my VNA has the same PCI graphics card as.....      .... the (low) infiniium scope range (small world)  ;D

Are the Infiniium scopes limited to the resolution of the LCD or can you do similar tricks as I have with my E5071B to get high resolution via an external monitor?

If this is really the same card as on the Infiniium 54800 Series then there isn't much you can do, as it comes with an antique Chips &Technology 6554x graphics chip which if I remember right can do VGA at 24bit and SVGA (800x600) at 16bit (or 8bit, depending on VRAM available). It was common graphics chip when DOS and Windows 3.11 were still a thing and Windows 95 was the next hot thing. I think the original chip was ISA or VLB but C&T came up with a PCI version which for some reason (well, it was cheap) found its way into a lot of embedded systems (it's also used in LeCroy's PowerPC based LC, WR LT and WP900 scopes).

What Agilent (actually HP) did for some on the 54800 Series (and I guess on the E5071A as well) is to use the graphic chip's hardware overlay function to combine Windows, the scope (or VNA) app, and the waveform display, which results in a range of limitations, including being stuck at basic VGA. I'm still not sure why they did go down that route, which seems a bit silly even back then. I mean, it's not that this C&T chip was some super-fast graphics accelerator back then, nor was the 54800 Series particularly fast.

One reason I've taken on the DSO8k is that it doesn't use that setup with a separate graphics card, but like any modern Windows scope uses the integrated GPU.
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3015
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #30 on: May 04, 2016, 04:50:41 pm »
Quote
If this is really the same card as on the Infiniium 54800 Series then there isn't much you can do, as it comes with an antique Chips &Technology 6554x graphics chip which if I remember right can do VGA at 24bit and SVGA (800x600) at 16bit (or 8bit, depending on VRAM available). It was common graphics chip when DOS and Windows 3.11 were still a thing and Windows 95 was the next hot thing. I think the original chip was ISA or VLB but C&T came up with a PCI version which for some reason (well, it was cheap) found its way into a lot of embedded systems (it's also used in LeCroy's PowerPC based LC, WR LT and WP900 scopes).

I think it is the same card as the 54800 series. It has the same part number and the image below is a picture I found online with the same Agilent part number. it looks the same as the outline drawings in the service manual and Control Panel gives the C&T 65550 number.

So this is what throttles the graphics capability.

But see below for a real live screenshot of my VNA taken with an external Dell 4:3 monitor in 1280x1024?  It is measuring S11 for a toroidal inductor. It looks really good compared to the internal VGA display :)
To see it sharply, right click on the text below the image and select 'open' to open it in the Windows Photo Viewer.

So can the Infiniium be hacked/fudged to do a similar thing? Could you run a second graphics adaptor card dedicated to higher resolution for an external display? I know very little about PC hardware/graphics but maybe the Infiniium can go hi res this way?






« Last Edit: May 04, 2016, 05:02:43 pm by G0HZU »
 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 5319
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #31 on: May 04, 2016, 06:51:26 pm »
On my 54831D I've tried a couple of things. The lack of memory on the video card meansmit can't support any resolution other than 640x480, and this card also uses hardware video overlaying for the waveform display, so is connected to another card which connects to the scope hardware.

There is a VGA port on the card, but this can only be used for duplicating the LCD contents, it will not extend the display. I use this the most, it's plugged into an LCD monitor at the soldering bench (which I also use with the trinocular microscope when making videos). For delicate browser probing, it's more convenient to have a screen at this level than having to comtinually glance up to the scope that's situated about 14" above the bench due to its depth.

You can also use the mobo's VGA to support up to 1280x1024 as an extended monitor, but not for the scope app as it doesn't support the video overlaying.

The same applies to use a DisplayLink adapter which also works surprisingly well for such an underpowered mobo. I should note that I also have a USB 2.0 PCI card installed for this as the mobo is only USB 1.0.

 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3015
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #32 on: May 04, 2016, 11:36:19 pm »
Quote
You can also use the mobo's VGA to support up to 1280x1024 as an extended monitor, but not for the scope app as it doesn't support the video overlaying.
My VNA appears to be limited to 640x480 if I try all the usual things. If I enable the mobo VGA port it can be configured as an extended 1280x1024 monitor and I can drag/swap common windows icons and window boxes between screens with this monitor in 1280x1024.

But the VNA app itself is stuck at 640x480 and it can't be dragged or moved. I can make it appear on the extended desktop if I make the external monitor the 'primary' display and the VNA app then uses up 640x480 pixels (in one corner of the monitor) but it can't be dragged/moved.

But I found clues in the program code that indicate it can be unlocked from this. However, it takes a certain sequence of user inputs to achieve it. I can't get it to run like this each time it switches on unless I perform a certain sequence of actions and this provides a special gateway into this hi res mode from first bootup. So I have to do this sequence EVERY time I want to use the external monitor in hi res mode  :(

I doubt anyone would find it easily unless they knew how to look for it because the system appears to stubbornly limit the VNA app to 640x480 pixels. So it either runs full screen 640x480 on the internal LCD or stubbornly sits in a 640x480 corner of the 1280x1024 external monitor if I play with the display adaptor options in Control Panel. 
Quote
I use this the most, it's plugged into an LCD monitor at the soldering bench (which I also use with the trinocular microscope when making videos). For delicate browser probing, it's more convenient to have a screen at this level than having to comtinually glance up to the scope that's situated about 14" above the bench due to its depth.

Yes, I am now going to look for a small hi res portable monitor to use down at eye level with the VNA.

I need to read up a bit on hardware 'video overlaying' as I know nothing about this subject. It looks like my VNA doesn't need to do this because it has a simple/slow display requirement. Such a shame that Agilent didn't make it easier to use a hi res monitor with it.
« Last Edit: May 04, 2016, 11:44:06 pm by G0HZU »
 

Offline WuerstchenhundTopic starter

  • Super Contributor
  • ***
  • Posts: 3088
  • Country: gb
  • Able to drop by occasionally only
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #33 on: May 05, 2016, 10:56:50 am »
My VNA appears to be limited to 640x480 if I try all the usual things. If I enable the mobo VGA port it can be configured as an extended 1280x1024 monitor and I can drag/swap common windows icons and window boxes between screens with this monitor in 1280x1024.

But the VNA app itself is stuck at 640x480 and it can't be dragged or moved. I can make it appear on the extended desktop if I make the external monitor the 'primary' display and the VNA app then uses up 640x480 pixels (in one corner of the monitor) but it can't be dragged/moved.

That's overlay. As the name suggests it superimposes the scope/VNA app onto the Windows desktop. It's a technique that was used for video playback back then as most PC graphics processors of those days were unable to maintain a fluid playback rate for raw video in a somewhat decent resolution, so the video was rendered on the CPU and superimposed (overlaid) on top of the screen content.

Quote
But I found clues in the program code that indicate it can be unlocked from this. However, it takes a certain sequence of user inputs to achieve it. I can't get it to run like this each time it switches on unless I perform a certain sequence of actions and this provides a special gateway into this hi res mode from first bootup. So I have to do this sequence EVERY time I want to use the external monitor in hi res mode  :(

I doubt anyone would find it easily unless they knew how to look for it because the system appears to stubbornly limit the VNA app to 640x480 pixels. So it either runs full screen 640x480 on the internal LCD or stubbornly sits in a 640x480 corner of the 1280x1024 external monitor if I play with the display adaptor options in Control Panel.

On the Infiniiums, the scope app (which is the same for the old 54800 scopes with the dual gfx card setup and the newer DSO8k with single graphics) seems to detect whether it is running on the old architecture or the newer one, and switches from hardware to software overlays (done in the graphics processor). If the VNA app on the A models is the same as on the B or C models then it might be the same with the VNA app, i.e. if it finds better graphics hardware it might work at higher resolutions.

I'll do a few tests with the DSO8k and the scope app over the weekend.

Quote
Yes, I am now going to look for a small hi res portable monitor to use down at eye level with the VNA.

There are a few things you can try. One would be to remove the PCI graphics card and use the on-board graphics in the intel chipset with an external monitor connected to the mainboard VGA port. Then set the resolution to something higher than 640x480 and start the VNA app.

If that works then all you need to do is to find a PCI graphics card with internal LVDS interface for better internal display. There are quite a few cards for the embedded market available.

Quote
I need to read up a bit on hardware 'video overlaying' as I know nothing about this subject. It looks like my VNA doesn't need to do this because it has a simple/slow display requirement. Such a shame that Agilent didn't make it easier to use a hi res monitor with it.

I really like HPAK instruments but in my opinion the architecture of the old Infiniiums was completely brain dead. I can't remember what Tek did back then with their TDS scopes (but I don't think their scope suffered from similar limitations), but the Windows based LeCroy scopes from the same era pushed much higher update rates via the standard mainboard graphics, with the scope app running as a normal Windows app that could be run fullscreen (at any resolution) or windowed, and if windowed dragged around like every other Windows program.

It took a very long time uptil Agilent finally came up with a decent Windows scope (DSO9k/90k). Which is a bit surprising, considering that their non-Windows scopes were very good.
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3015
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #34 on: May 05, 2016, 03:40:25 pm »
I'll have another play over the weekend to see what can be improved. But I can get it to run in 1280x1024 fullscreen at high resolution at the moment if I mess about with the settings each time I restart the VNA.

Quote
As far as I know none of these scopes were bandwidth-upgradeable, so the older 54800 Series (which uses the same software) has probably only a few front ends which were artificially bw limited to create the various models.
Getting back to the Infiniium scopes, I can now definitely confirm that it's possible to upgrade the BW on the 80000 series because I was granted access to the 80404B again briefly.

As supplied it displayed 'Standard 4GHz BW' on the screen and this was confirmed again with a decent sig gen. I then tried creating various licence files and managed to up it to 8GHz in that it displayed 'Standard 8GHz BW' on the screen.

I only had a 6GHz sig gen available but it was flat to 6GHz so the BW has definitely gone up. I'll bring across a 20GHz generator next time and see how high it can go in BW in terms of options and also with the sig gen test.

But as 'standard' it rolled off sharply immediately at 4GHz but with the 8GHz BW option listed it was dead flat to 6GHz at least.

I can also confirm that my 'bad boy' licence works where it's possible to unlock lots of options (sometimes all of them?) with a single generic licence that targets all Infiniium scopes. i.e. it isn't locked to a serial number and it isn't locked to a specific model or sub range. I think it would work on every Infiniium scope that uses that licensing scheme. On this 80404B it released a whole load of options (too many to count) and the initial splash screen grows considerably in size to accommodate the options.  Obviously I'm not going to release this global bad boy licence into the wild as it would probably work on their current models and not just the old obsolete scopes. But this does look to be a significant vulnerability in their licensing code?
« Last Edit: May 05, 2016, 03:47:24 pm by G0HZU »
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3015
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #35 on: May 05, 2016, 10:51:27 pm »
Quote
It took a very long time uptil Agilent finally came up with a decent Windows scope (DSO9k/90k)

What's the typical price range of these when used? I'm assuming that the 90k series is going to be extremely expensive but what about the 9k models? I like the big display and it isn't a deep body scope like the 80404B. Also it has 1M ohm input Z for use up to a few hundred MHz. So this scope could make a decent general purpose DSO for me as long as it can run the 89600 SW efficiently and it doesn't make too much fan noise. I'm hoping it won't be as noisy as the big old HP54540C here. I suspect that the 4000 series might be a better choice for my needs because I just want to be able to grab wideband signals with it via 89600 and also it would be nice if it could be used as a general purpose scope with a nice big display.

The other option is to buy an older Infiniium (even one with a dog slow UI) and only use it with the 89600 SW and then sling it into storage when not being used for this mode. This is bound to be much cheaper.
 

Offline WuerstchenhundTopic starter

  • Super Contributor
  • ***
  • Posts: 3088
  • Country: gb
  • Able to drop by occasionally only
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #36 on: May 06, 2016, 05:46:55 am »
Quote
As far as I know none of these scopes were bandwidth-upgradeable, so the older 54800 Series (which uses the same software) has probably only a few front ends which were artificially bw limited to create the various models.

Getting back to the Infiniium scopes, I can now definitely confirm that it's possible to upgrade the BW on the 80000 series because I was granted access to the 80404B again briefly.

As supplied it displayed 'Standard 4GHz BW' on the screen and this was confirmed again with a decent sig gen. I then tried creating various licence files and managed to up it to 8GHz in that it displayed 'Standard 8GHz BW' on the screen.

I only had a 6GHz sig gen available but it was flat to 6GHz so the BW has definitely gone up. I'll bring across a 20GHz generator next time and see how high it can go in BW in terms of options and also with the sig gen test.

But as 'standard' it rolled off sharply immediately at 4GHz but with the 8GHz BW option listed it was dead flat to 6GHz at least.

I guess they didn't design an individual front end for each bandwidth step but instead built one or two frontends (I guess 8GHz and 13GHz) and used bw limiting to come up with the other steps.

I don't think it was meant to be used as bw upgrades as you can do on newer scopes, at least I'm not aware it was ever offered as an end user option.

Quote
I can also confirm that my 'bad boy' licence works where it's possible to unlock lots of options (sometimes all of them?) with a single generic licence that targets all Infiniium scopes. i.e. it isn't locked to a serial number and it isn't locked to a specific model or sub range. I think it would work on every Infiniium scope that uses that licensing scheme. On this 80404B it released a whole load of options (too many to count) and the initial splash screen grows considerably in size to accommodate the options.  Obviously I'm not going to release this global bad boy licence into the wild as it would probably work on their current models and not just the old obsolete scopes. But this does look to be a significant vulnerability in their licensing code?

I doubt it works beyond the DSO80/80k, the DSO9k/90k use a different software which is newer and very likely relies on a much newer (and less vulnerable) version of FlexLM. Also, even if it did, no business would use it to create licenses as Keysight keeps a register of which S/N comes with what option, and they would certainly put up a stink if a business sends in kit with 'self-created' option codes for calibration or repair.  ;)

Quote
It took a very long time uptil Agilent finally came up with a decent Windows scope (DSO9k/90k)

What's the typical price range of these when used? I'm assuming that the 90k series is going to be extremely expensive but what about the 9k models?

The lower bandwidth models (i.e. DSO9204A) start at somewhere around $8k for the older versions running XP (which often comes with the mobo which can't be upgraded to Win7), and noticeably more ($10k+) for the later models. In the UK you pretty much end up paying the same figure in GBP.

Quote
I like the big display and it isn't a deep body scope like the 80404B. Also it has 1M ohm input Z for use up to a few hundred MHz. So this scope could make a decent general purpose DSO for me as long as it can run the 89600 SW efficiently and it doesn't make too much fan noise. I'm hoping it won't be as noisy as the big old HP54540C here.

The noise level should be comparable to the DSO80k you tried, so it's not exactly quiet but no screamer either. The DSO9kA is a nice scope, although the UI on the XP models is pretty much the same as on the DSO8k/80k and 54800 Series. The display has the same resolution (XGA) as the DSO8k/80k, but at least the big screen means touch starts to make sense and you can work without a mouse.

Also, the waveform rates are pretty low.

Quote
I suspect that the 4000 series might be a better choice for my needs because I just want to be able to grab wideband signals with it via 89600 and also it would be nice if it could be used as a general purpose scope with a nice big display.

Probably, but that depends on your bandwidth needs and how much data you want to grab (the InfiniVision scopes like the DSOX4k only have 4M memory which in most modes gets also halfed so you might end up with 2M only).

Quote
The other option is to buy an older Infiniium (even one with a dog slow UI) and only use it with the 89600 SW and then sling it into storage when not being used for this mode. This is bound to be much cheaper.

It is. Again, it depends on what bandwidth you need but if you go that route then I'd go for a DSO8k/80k instead of the old 54800 Series, even more so if you want to run the 89600 software on the scope. The P4 in the DSO8k/80k is a lot faster than the old P3 in the 54800 Series, plus you can get some decent amount of RAM in the scope.

Not sure if that's a viable alternative to you, but there's also the R&S RTE/RTO Series and the R&S VSE (Vector Signal Explorer) software  ;)
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3015
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #37 on: May 06, 2016, 11:59:25 am »
Quote
I guess they didn't design an individual front end for each bandwidth step but instead built one or two frontends (I guess 8GHz and 13GHz) and used bw limiting to come up with the other steps.
I don't think it was meant to be used as bw upgrades as you can do on newer scopes, at least I'm not aware it was ever offered as an end user option.

I had access to the 80404B again and brought some different licence files along. The best one achieved an onscreen message of '12GHz Standard Bandwidth' and this was confirmed with an old HP83752A 20GHz synthesised source.

This old sweeper has a very accurate ALC system and is flat to 20GHz. The updated 80404B showed a small amount of rolloff by 12GHz but this was partly in the 18GHz rated cable. By 13GHz it drops suddenly.

There may be more to come, I only have brief access to the scope and I would ideally need to spend time with it and spend more time fishing for stuff in the program code. but a 4GHz to 12GHz BW upgrade is already impressive :)

I'm becoming drawn to this scope because of the sheer grunt it offers but it isn't really much use as a general purpose scope as the inputs are fixed at 50 ohm. I'm not sure how to check how much memory I can unlock in it but I think I will need quite a bit of memory if I want to capture wide BWs. My E4406A can capture a ~10MHz BW for only a tiny fraction of a second when used with the 89600 SW.
 

Offline WuerstchenhundTopic starter

  • Super Contributor
  • ***
  • Posts: 3088
  • Country: gb
  • Able to drop by occasionally only
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #38 on: May 06, 2016, 12:23:40 pm »
I had access to the 80404B again and brought some different licence files along. The best one achieved an onscreen message of '12GHz Standard Bandwidth' and this was confirmed with an old HP83752A 20GHz synthesised source.

This old sweeper has a very accurate ALC system and is flat to 20GHz. The updated 80404B showed a small amount of rolloff by 12GHz but this was partly in the 18GHz rated cable. By 13GHz it drops suddenly.

There may be more to come, I only have brief access to the scope and I would ideally need to spend time with it and spend more time fishing for stuff in the program code. but a 4GHz to 12GHz BW upgrade is already impressive :)

It is. So it looks like the DSO80k is all the same hardware.

Quote
I'm becoming drawn to this scope because of the sheer grunt it offers but it isn't really much use as a general purpose scope as the inputs are fixed at 50 ohm.

True, but that is true for any scope over a few GHz of bw. And there's a high impedance adapter that somewhat overcomes the problem (although the max voltages remain limited).

Quote
I'm not sure how to check how much memory I can unlock in it but I think I will need quite a bit of memory if I want to capture wide BWs. My E4406A can capture a ~10MHz BW for only a tiny fraction of a second when used with the 89600 SW.

You can check the available memory in the acquisition settings, there's a way to manually set memory. But don't forget that as soon as the sample rate is 4GSa/s or more the sample memory shrinks to 2MB per channel, no matter what option is installed. This is a limitation of the slow architecture. Which means for long captures you're limited to 2GSa/s so essentially around 900MHz useable real-time bandwidth.

Which is actually worse than the DSO8000A, my DSO8064A has 128M even at 4GSa/s, which easily covers the true 1.27GHz bandwidth I have.

I guess the question is what you want to capture, i.e. the raw signal or a down-mixed IF (in which case 900MHz or so should be plenty).
« Last Edit: May 06, 2016, 12:31:18 pm by Wuerstchenhund »
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3015
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #39 on: May 06, 2016, 12:43:52 pm »
I'll be downconverting to an IF and I probably only need 200MHz BW at the IF for most tasks in the immediate future. So the 80404B is overkill even in 4GHz guise. I think I need a scope with 500MHz BW minimum but would prefer 1GHz BW.

I like the look of the DSO9104H but need to do some more research on it to look for showstoppers in the specs etc. I wouldn't need to mess with BW hacks etc with this scope.

Quote
So it looks like the DSO80k is all the same hardware.
It's worth me emphasising that that I'm a complete novice wrt these scopes and I just did a quick test with the 20GHz sweep generator. However, it does look like it's possible to unlock 'something' that extends the BW quite a bit (I'm guessing it's about 13GHz) and brings up the 12GHz BW logo. It might not mean the scope is now the direct equivalent of a 12GHz variant of this scope. I'm really just tinkering and observing/measuring :)
« Last Edit: May 06, 2016, 12:48:19 pm by G0HZU »
 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 5319
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #40 on: May 06, 2016, 04:40:27 pm »
I'll be downconverting to an IF and I probably only need 200MHz BW at the IF for most tasks in the immediate future. So the 80404B is overkill even in 4GHz guise. I think I need a scope with 500MHz BW minimum but would prefer 1GHz BW.

I like the look of the DSO9104H but need to do some more research on it to look for showstoppers in the specs etc. I wouldn't need to mess with BW hacks etc with this scope.

Take a look at this thread https://www.eevblog.com/forum/testgear/exploring-the-secrets-of-agilent-9000h-series-scopes/
 

Offline WuerstchenhundTopic starter

  • Super Contributor
  • ***
  • Posts: 3088
  • Country: gb
  • Able to drop by occasionally only
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #41 on: May 31, 2016, 04:31:37 pm »
I had to shift my focus to some more important stuff buy yesterday I managed to spent some time with the DSO8064A. I upgraded the memory to 3GB and the CPU from the bog-standard intel Celeron D 340J (2.93GHz, FSB533, 256k cache) to a 3.4GHz Pentium4 650 with FSB800 and 2MB cache.

I also enabled HyperThreading in the BIOS and the scope seems to work fine.

Not sure how much that improves scope performance (unlike LeCroy scopes which do pretty much everything in the x86 CPU these Agilent scopes use proprietary ASICs for all waveform calculations) but at least it should improve the general performance of Windows applications, including the VSA software.

The next thing to try is if the scope can be upgraded to Windows 7.
 

Offline Mrt12

  • Regular Contributor
  • *
  • Posts: 132
  • Country: ch
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #42 on: February 14, 2017, 08:54:05 pm »
Hey @G0HZU,

I wonder how it works to add options. I have a 54831B which I upgraded to a 54832B. I am interested in the serial protocol analysis option and the 128M memory option. How can that be done?
 

Offline Aiy

  • Regular Contributor
  • *
  • Posts: 68
  • Country: se
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #43 on: November 23, 2017, 11:24:46 pm »
I tried booting up a virtualbox with tinyxp on it but it wont run the agscope.exe application.. no error message, nothing...
Wanted to try some options on my scope ...

G0HZU: do you have to use a real computer for some reason ?
 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 5319
  • Country: gb
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #44 on: November 24, 2017, 06:20:40 am »
In my experience, the build is extremely picky. It will look for certain specific PCI hardware and there is a licensing component. I’ve not been able to build a working scope from scratch, but I have been able to upgrade the OS on an existing working scope, not that that gave any benefits. I have also managed to change the motherboard but it needs a fair bit of work to get the device drivers installed properly and software working again afterwards.

That was after many days of investigation and dozens of OS installs.
 

Offline Aiy

  • Regular Contributor
  • *
  • Posts: 68
  • Country: se
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #45 on: November 25, 2017, 09:11:16 am »
In my experience, the build is extremely picky. It will look for certain specific PCI hardware and there is a licensing component. I’ve not been able to build a working scope from scratch, but I have been able to upgrade the OS on an existing working scope, not that that gave any benefits. I have also managed to change the motherboard but it needs a fair bit of work to get the device drivers installed properly and software working again afterwards.

That was after many days of investigation and dozens of OS installs.

Understood. I wanted to get some extra options enabled on my scope (memory and spi/i2c decoders perhaps)

Do the windows XP based scope software run on the tinyXP / microXP dists or is that a waste of time?

Cheers
 

Offline TRN

  • Supporter
  • ****
  • Posts: 127
  • Country: pt
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #46 on: March 14, 2018, 04:50:31 pm »
Has anybody managed to install the Infiniium 4.20 System Software (9000 infiniium-firmware) on a infiniium 8000 oscilloscope?

thanks
 

Offline rubidium

  • Regular Contributor
  • *
  • Posts: 102
  • Country: us
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #47 on: July 01, 2018, 11:11:35 pm »
I am about to come into possession of a DSO-8104A, and wondered if someone would be kind enough to share the methods for unlocking various options. PM me is you prefer.
Thanks,
Jim
 

Offline schenkmi72

  • Contributor
  • Posts: 24
  • Country: ch
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #48 on: June 16, 2019, 07:25:42 am »
In a few days I got a DSO8064A. From this thread it looks like we may up-hack this kind of DSO to enable the full equipment sample memory and for more features. Maybe even the bandwidth can be increased to 1GHz. Is anyone willing to help me doing this task ? This would be absolutely awesome! PM me.
Cheers
 

Offline Mr Nutts

  • Regular Contributor
  • *
  • Posts: 142
  • Country: us
Re: Agilent Infiniium DSO8000A Hacking - anyone done this?
« Reply #49 on: June 16, 2019, 01:27:33 pm »
Frankly I don't know why the software for these scopes has not been hacked  :-//

The whole system is based on a very old version of FlexLM license manager which is used in a lot of other applications and for which there are many hacks available ;)

For those other apps there often is a license keygen available to to create license files yourself. I'm sure this could be done for the Infinum software on these scopes as well :)

You could also patch the executable to enable options but some options like power analysis or VSA rely on separate executables which will fail to work :(

The best hacking method would be to develop a keygen which creates valid license files, something G0HZU has already done so it is definitely possible  :-+

The question is if our resident hackers are up for it  :box:
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf