REVIEW - Owon SDS7102 - A look at the SDS series from Owon

[0xdeadbeef]

I'm not saying that USB scopes couldn't be priced lower (I think Velleman has made inroads in this direction), but I think you're a bit out of touch with the market. The Picoscope 2207 - which has the 100MHz bw and 1GS/s rate - plus a built-in AWG and function generator - lists for $740 on the Picoscope website.

And only 40k record length. This kinda destroys all possible benefits. A 2ch scope with such a tiny sample memory buffer doesn't look like a real bargain for this price.

The weird thing is that although I work with windows based touchscreen scopes almost every day and at the same time like my el cheapo USB logic analyzer, I still have strong feelings against an USB scope. Firstly, for daily work, it would be only acceptable with a 2nd screen. Maybe somebody comes up with a standalone unit that is using a fancy tablet/pad as control and display unit. Apart from this, the problem remains that USB based devices with vendor specific drivers will quickly become outdated as will the control software. You can use a standalone scope that is 20 or 30 year old and you will still be able to use the current scopes in ten years even if they're based on Windows XP or Windows 7 which will then look like DOS for us right now.
But you can kinda bet, that even big brands won't support you for ten years with USB drivers and software updates. Not talking of small companies that might not even exist any more in ten years. This is an acceptable risk for a logic analyzer for a few hundred €, but it's just a bad idea for scope that costs a few thousand €.

So honestly, I don't believe that USB scopes in their current form will ever take a big share of the market and USB3 won't change this in any way.

[free_electron]

Stuff like measuring in the sample data

Agilent measures in the sample data , not on screen. On live data the input attenuators and timebase are reprogrammed on the fly. They have a large offset range to do this. you can actually see a 1vpp sinewave superimposed on a 30 volt dc signal.. the input stage has a dac that feeds into a summing amplifier...

Gating for measurements

explain 'gating' please ... do you mean triggering or do you mean the capability to stop a recording when nothing is going on and then continue ? Agilent calls this 'segmented memory'
Some of their low end scope have that but it is an option you need ot pay for. it is not standard.

the Agilent DSX3000 has no button to switch between auto/normal trigger mode though this is something that you need all the time.

Hitting a button repeadedly cycles through modes on some agilent scopes. so if you press the 'trigger' button the trigger menu comes up , hit the same button again and it changes the most logical setting ( in this case trigger mode ).

Generally, I'm underwhelmed by the displayed precision for cursor positions and measurements. With 1GSa/s and above, it should be somehow possible to measure if a pulse is 100µs or 100.01µs. That's 10ns and with 1GSa/s 10 points in the sample data.

  the way you do this is by making a split view and zooming in on both edges. Rising edge in window 1 , falling in window 2. Then you get the required number of digits. if you zoom all the way out they faststep the cursor and turn off the trailing digits since they never change anyway... There is no point in showing 1.1000000000 . As you zoom in you get more trailing digits.

[free_electron]

I REALLY wish there was competition , i mean like okay 1Gigsamples ... What's the speed of a standard ADC converter in a off the shelf notebook you find now?
They have 24 bits 5gigsamples ADC's in there.
AND quad-core 1.4GHz 35W chips ... i mean what's stopping agilent to make cheaper stuff ? They have HP by their side and a fab of their own ( they make the megazoom asics so i assume they have a fab of theirs )

what have you been smoking ?

notebooks with 5 gigasample ADC's ? and 24 bit ? you have no idea what you are babbling about do you ?

Agilent has nothing to do with HP anymore. they long ago went separate ways. Neither Agilent nor HP have a waferfab (for digital monsters) at their disposal. The fabs went to Avago..
Besides those fabs are for optical things.

I know who fabs the agilent ASIC's ( can't tell you though ... NDA .. ). Those chips are extremely expensive due to the large amount of SRAM that is on board. The sram has to be able to run at full speed. Second problem is that they use gigantic internal pathways and have hardware assisted processing. The development costs of the Lynx ( Lynx is one of the megazoom Asic's ) are staggering. With the relatively small volumes they run a single chip runs well over a thousand of dollars.
The chipset in a 35000$ agilent scope costs them easily 15K$ to just fab alone ... An infinivision chip with 64 of 128 meg ram can handle 1 channel (so you need 4 of these beasts... in that 35K scope) add the multistage pipelined A/D needed per channel on there and you hit that 15K$ mark without problems...

The slower/cheaper machines use external ram modules.But the asics are still very expensive.

[marmad]

And only 40k record length. This kinda destroys all possible benefits. A 2ch scope with such a tiny sample memory buffer doesn't look like a real bargain for this price.

Hmm... no one said it was a bargain by any stretch of the imagination. In fact, I think alm and I agree that prices have to come down.

Apart from this, the problem remains that USB based devices with vendor specific drivers will quickly become outdated as will the control software. But you can kinda bet, that even big brands won't support you for ten years with USB drivers and software updates. Not talking of small companies that might not even exist any more in ten years. This is an acceptable risk for a logic analyzer for a few hundred €, but it's just a bad idea for scope that costs a few thousand €.

First, we're talking about the low-end of the market. How many people will be using their Rigol DS1052E in 20 years? Secondly, I'm still using PC tech-hardware that is more than 10 years old - and with the advent of USB, the only significant change in Windows drivers was going from 32 to 64 bit (which many people haven't even bothered to do yet). The driver problem was much more significant in the 90's then it has been in the last 10 years, so I don't agree with your assessment of the problem.

So honestly, I don't believe that USB scopes in their current form will ever take a big share of the market and USB3 won't change this in any way.

Again, 'a big share of the market' is not even close to what I wrote, which was, '...USB 3.0... might very well make USB scopes, for many people, the bang-for-buck match of...standalone DSOs.'

My argument was that USB 3.0 might help drive the price down on USB scopes for the low-end market. Of course, I could be wrong about this - we'll have to wait and see.

[mAJORD]

It may be usable on lower sampling rate, just like the long memory option on Rigol scopes, but at that point you're saving the relatively cheap DRAM, not the expensive RAM that can run at full speed. Not sure how much BOM costs this saves, especially if you factor in the premium for USB 3.0 transceivers.

Well, it seems pointless to argue this. You don't seem to feel that USB 3.0 scopes will make any difference in the market - I disagree. We'll just have to wait and see.

Sure, it has more features, though I'm not sure the Agilent is a fair comparison, since its banner spec is the fast update rate. My point was that this is the cheapest scope made by Picoscope (and I can't think of too many cheaper competitors) that exceeds the entry level scope which has been dominated by the Rigol DS1052E/1102E for the past five or so years. And you pay $600 extra for the lack of screen and controls.

I'm not saying that USB scopes couldn't be priced lower (I think Velleman has made inroads in this direction), but I think you're a bit out of touch with the market. The Picoscope 2207 - which has the 100MHz bw and 1GS/s rate - plus a built-in AWG and function generator - lists for $740 on the Picoscope website.

Also, I don't personally feel the lack of a tiny, old-fashioned 320x240 screen and some cheap controls matters if you get a more powerful tool with greater capabilities. It seems that at least some of the disdain for USB devices on this blog comes from 'traditionalist' engineers (and the like) who seem to believe that any piece of test equipment that isn't a standalone device is not 'serious' equipment. This whole attitude strikes me as rather hilarious. I no longer own any standalone media devices (television, dvr, stereo, etc) - everything in my home being driven from HTPCs. I expect my lab to keep moving in that direction as well.

I don't consider myself traditionalist at all (infact I'm always bitching about the lack of inovation in tech in general) but I have to strongly disagree here, in regards to people being stuck in there ways when it comes to stand alone equipment.

Whilst a USB scope has its place in controlled, or long term testing. For everything else, they'd be a pain in the arse. The ergonomics of it just aren't there, and nothing's going to change that in the forseabe future.

I don't think the HTPC analogy is entirely valid either. All that is really equivilent to is the increasing all-in-one functionality of DSO's.. i.e with a HTPC, you still sit in bed, or in your couch, and hold a remote.

It also reminds me of the touchscreen vs mouse +keyb debate.. Are people traditionalists for using the 30y/o mouse still? or is it just that touchscreens are a pain in the arse?  (Case in point, I pulled my laptop out of my bag to reply to this, because, you guessed it, the phone s a PITA to type on).

What I would say instead,  for cheaper scopes at least, is to stop this stupid trend of fitting huge screens and 1million buttons on them and instead create better PC functionality (which is typically crap, thouhg I haven't used them all) . Then you have the best of both worlds. "traditonal" , simple front panel controls, + a small screen, for "hands on" bench work,  and all your fancy functionality, and screen real-estate on the PC for the tasks I listed earlier.

[0xdeadbeef]

Agilent measures in the sample data , not on screen.

I wish that was true, but unfortunately, it isn't. All the Agilent scopes I used up to now measured in the display data and changed the measurements when changing e.g. the zoom level in the already sampled data.
In the DSOX3000 manual, they also clearly state this:

"Measurements and math functions will be recalculated as you pan and zoom and turn channels on and off.
As you  zoom in and out on a signal using the horizontal scale knob and vertical volts/division knob,  you affect the resolution of the display. Because measurements and math functions are  performed on displayed data, you affect the resolution of functions and measurements."

Indeed e.g. Lecroy salesmen are aware of this and put their fingers in the wound. Of course, Agilent scopes have other shortcomings, e.g. the extremely lacking trigger possibilities, the crappy menu structure with the Auto/Normal triggering hidden somewhere in a submenu instead of having a button etc.

explain 'gating' please ... do you mean triggering or do you mean the capability to stop a recording when nothing is going on and then continue ? Agilent calls this 'segmented memory'
Some of their low end scope have that but it is an option you need ot pay for. it is not standard.

No, segmented memory is something completely different. Gating is an essential measurement feature which let's you define the part of the sample buffer or screen range, where the measurement should be done. If you want e.g. to measure the PWM duty cycle in a pulsed PWM, a good scope let's you set the gating cursors at the start and end of the pulse and then the DC is only measured within this window. While Tektronix, LeCroy and even Rigol offer this option, Agilent simply ignores it. Instead they recommend to zoom into the area so the zoom window is misused as gating area. The problem is however that you can't control the size of the zoom window exact enough for this to work.

Hitting a button repeadedly cycles through modes on some agilent scopes. so if you press the 'trigger' button the trigger menu comes up , hit the same button again and it changes the most logical setting ( in this case trigger mode ).

Even if this worked (I don't think it does for the MSO7000A we have at work), it's still a workaround and a very dumb design decision to leave this button away.

the way you do this is by making a split view and zooming in on both edges. Rising edge in window 1 , falling in window 2. Then you get the required number of digits. if you zoom all the way out they faststep the cursor and turn off the trailing digits since they never change anyway... There is no point in showing 1.1000000000 . As you zoom in you get more trailing digits.

Which just proves that it measured in the display data and the scope cripples the measured accuracy by weird implementation. Besides, if you zoom out again to make a screenshot of the signal and measurement, you get the imprecise value again. So no way to make a sensible screenshot for documentation. And yes, it makes a lot of sense to show 6 or more trailing digits if they are based on the actual sampling data. E.g. in a recent case, I needed to test if a period changed from 100µs to 100.01µs. Impossible with the Owon, pain in the ass with an Agilent MSO7000A, piece of cake with a Lecroy 6Zi. Hell, you can even make a statistical analysis in a gating area (including a histogram) and so detail analysis in WaveScan mode. This is a bit of luxury of course. But a full resolution (independent of zoom level) automatic measurement in a gated area should really be available in this price range. And AFAIK no single Agilent scope can do it.

[T4P]

I REALLY wish there was competition , i mean like okay 1Gigsamples ... What's the speed of a standard ADC converter in a off the shelf notebook you find now?
They have 24 bits 5gigsamples ADC's in there.
AND quad-core 1.4GHz 35W chips ... i mean what's stopping agilent to make cheaper stuff ? They have HP by their side and a fab of their own ( they make the megazoom asics so i assume they have a fab of theirs )

what have you been smoking ?

notebooks with 5 gigasample ADC's ? and 24 bit ? you have no idea what you are babbling about do you ?

Agilent has nothing to do with HP anymore. they long ago went separate ways. Neither Agilent nor HP have a waferfab (for digital monsters) at their disposal. The fabs went to Avago..
Besides those fabs are for optical things.

I know who fabs the agilent ASIC's ( can't tell you though ... NDA .. ). Those chips are extremely expensive due to the large amount of SRAM that is on board. The sram has to be able to run at full speed. Second problem is that they use gigantic internal pathways and have hardware assisted processing. The development costs of the Lynx ( Lynx is one of the megazoom Asic's ) are staggering. With the relatively small volumes they run a single chip runs well over a thousand of dollars.
The chipset in a 35000$ agilent scope costs them easily 15K$ to just fab alone ... An infinivision chip with 64 of 128 meg ram can handle 1 channel (so you need 4 of these beasts... in that 35K scope) add the multistage pipelined A/D needed per channel on there and you hit that 15K$ mark without problems...

The slower/cheaper machines use external ram modules.But the asics are still very expensive.

K, i stand corrected i have been smoking weed.
What i meant was DAC's which are a different area and are not that fast but more so around a capability of god knows how much pixels and 3x8bit
But yeah i know agilent went their own way and HP are not with them anymore and i am aware HP obviously can't fab anything but anyway i am very wrong that they have a fab, i apologize about that. I understand the need for NDA 
But we're not talking 35k scopes here, only 2k scopes

[akis]

http://focus.ti.com/paramsearch/docs/parametricsearch.tsp?family=analog&familyId=390&uiTemplateId=NODE_STRY_PGE_T

The cheapest 1GSa/s chip is like $170 - other chips go up to many hundreds of $. There is one chip that costs $9,000 ...

[jcbottorff]

I'd have to agree with marmad that USB 3.0 will allow attractive low price designs. It should allow streaming a couple hundred MBytes/sec.

I actually think PCIe based scopes are also ripe for massive improvements in price/performance. A PCIe 2.0 x16 slot can stream a few gigabytes/sec (my bland video card can transfer 5+ GBytes/sec from RAM). It should be possible to stream the sample data directly into GPU memory and then then have like 300 GPU cores processing the sample data. A cheap GPU has a gigabyte of RAM, and 300 GPU cores might be enough to process ALL the samples (i.e. like zero dead time), in real-time, for an intensity graded display. This will take a bunch of high performance software though, and the ability of oscilloscope companies to write good software has been mixed. I could actually imagine an open source GPU accelerated based oscilloscope, with the ADC card coming from somebody else, or multiple ADC card brands that all work with one piece of software. Notice that companies that make video cameras often don't make video editing software, and I could imagine a similar hardware/software split happening in test equipment. I'm probably biased though, I'm a server system software engineer, not an oscilloscope hardware designer. Still, for $150, I can now buy a TFLOP of computing power that pops into a $500 computer. A lot of GPU power is being built into the CPU now too. It would certainly be worth while for somebody to figure out what a modern GPU could do in terms of processing a sample stream. Like maybe 300 cores is enough to do software based triggering (i.e. finding cyclic patterns in the samples), making the ADC card even simpler. Being heavily software based, you could change the features by writing code. If it were open source, YOU could change the features, or you could pay for commercial software and let software folks change the features. So you need an analog front end, 1 to 4 ADC chips, an FPGA for PCIe bus interfacing, software, an appropriate computer with appropriate GPU.

[alm]

What do you expect your average PC to do with this hundreds of MB/s? You can't store it on disk, RAM will be filled up in seconds, and the CPU is probably quickly going to run out of steam if you want to do any kind of complex math at that speed. There's a reason why everybody uses FPGAs or ASICs in real scopes.

PCIe has a much faster interface, but again, what are you going to do with that data? Can you do DMA transfers directly into the GPU memory, or does it have to involve the main memory (my guess would be the latter)? PCIe 2.0 and low level drivers will be obsolete much quicker than even USB devices. PCIe excludes the use of laptops, so portability will be worse than stand-alone scopes. Even if this is feasible, then it probably needs some serious engineering effort from a company like Lecroy. Don't expect the big scope vendors to open up the market and make it more accessible to competitors, they like their expensive, monolithic and proprietary boxes. These are essentially Windows PCs with a bunch of scope hardware to do all the sampling and data transfer stuff.

[akis]

I'd have to agree with marmad that USB 3.0 will allow attractive low price designs. It should allow streaming a couple hundred MBytes/sec.

I actually think PCIe based scopes are also ripe for massive improvements in price/performance. A PCIe 2.0 x16 slot can stream a few gigabytes/sec (my bland video card can transfer 5+ GBytes/sec from RAM). It should be possible to stream the sample data directly into GPU memory and then then have like 300 GPU cores processing the sample data. A cheap GPU has a gigabyte of RAM, and 300 GPU cores might be enough to process ALL the samples (i.e. like zero dead time), in real-time, for an intensity graded display. This will take a bunch of high performance software though, and the ability of oscilloscope companies to write good software has been mixed. I could actually imagine an open source GPU accelerated based oscilloscope, with the ADC card coming from somebody else, or multiple ADC card brands that all work with one piece of software. Notice that companies that make video cameras often don't make video editing software, and I could imagine a similar hardware/software split happening in test equipment. I'm probably biased though, I'm a server system software engineer, not an oscilloscope hardware designer. Still, for $150, I can now buy a TFLOP of computing power that pops into a $500 computer. A lot of GPU power is being built into the CPU now too. It would certainly be worth while for somebody to figure out what a modern GPU could do in terms of processing a sample stream. Like maybe 300 cores is enough to do software based triggering (i.e. finding cyclic patterns in the samples), making the ADC card even simpler. Being heavily software based, you could change the features by writing code. If it were open source, YOU could change the features, or you could pay for commercial software and let software folks change the features. So you need an analog front end, 1 to 4 ADC chips, an FPGA for PCIe bus interfacing, software, an appropriate computer with appropriate GPU.

A cheap $150 GPU transfers 128 GB/sec to and from its on-board 1GB RAM. CPUs have onboard RAM of 3MB, 6MB or more, and can access it at clock cycle speeds (if you were to make a RISC chip with just the onboard cache and strip the onboard memory controller, parallel bus and cache controllers etc).

I think the only limiting technology is the ADC chip which from what I saw on the TI web page, very expensive. I am sure the technology is there but not the volumes of chips sold for CPUs and GPUs to price these chips at a few dollars each rather than many hundreds of dollars.

[marmad]

What do you expect your average PC to do with this hundreds of MB/s?

To get an idea about this, you should be looking at the Velleman scopes. They are well-designed and built, but have modest bandwidths and are fairly inexpensive. They use the data being transferred to the PC for doing things like real-time Bode plotting.  I've read of many people buying their scopes for this feature alone - even with the low bandwidths - since it isn't available on much more expensive scopes (or you have to write or assemble the code yourself with Mathlab, Labview, etc).

I was never suggesting that USB 3.0 was going to revolutionize the oscilloscope industry. My only point was that I believe the faster transfer rates were going to bring higher bandwidths and faster sampling rates to lower cost USB scopes like the Velleman scopes - just as it will to low cost USB logic analyzers - how could it not? And for more prospective oscilloscope buyers, if the PC software is good and feature-laden (and I believe it is for both Picoscope and Velleman) - and the bandwidth is sufficient for their use - and the price is attractive (Velleman levels - not Picoscope levels) - they will chose that option.

[krenzo]

...GPU...GPU...GPU...GPU...

Can we stop making stuff up?  This is almost as bad as the thought that notebooks comes with gigahertz ADCs.  Go read up on GPUs before you declare they they should come anywhere near an oscilloscope.  GPUs are designed for massively parallel operations (operating on millions of individual pixels at the same time).  Each individual core by itself is slow.  Sampling a waveform is a serial operation.  Triggering is a serial operation.  You would be using the wrong tool for the job.

As for USB 3, I don't see that having any effect on the scope market.  There's no real incentive to use it.  If it let you do gigabit transfers, then I could see companies jumping on it because then all the scope would need to do is digitize the signal and send it to the computer.  You would cut out a lot of parts and could put all of your cost into using a fast ADC.  All a USB scope does is cut out the need for a screen.  There's no other benefit to building one as a company, and most users see the lack of a screen as a negative.

[marmad]

If it let you do gigabit transfers, then I could see companies jumping on it because then all the scope would need to do is digitize the signal and send it to the computer.

Um... I think you mean gigaBYTE transfers, don't you? Because It DOES let you do gigabit transfers - 5Gb/s max. In terms of transferring captured data, probably something closer to half that (2.5Gb/s) when you consider overhead. That means it's theoretically feasible to build a USB 3.0 scope with a 300MS/s rate (30MHz bandwidth?) that sends the data directly to the PC.

There's no other benefit to building one as a company, and most users see the lack of a screen as a negative.

This belies the fact that there are already dozens of USB scopes - introduced, for the most part, after the 2.0 spec. was adopted. And again, I personally don't see the lack of a tiny 6" screen as a negative if I'm looking at my real-time waveforms on a 24" monitor. But, hey, that's just me.

[Bored@Work]

and most users see the lack of a screen as a negative.

That and the lack of real knobs. And typically the lack of isolation.

This belies the fact that there are already dozens of USB scopes - introduced, for the most part, after the 2.0 spec. was adopted.

And most disappeared from the market faster than they were introduced, are rubbish overpriced toys or ended up in a market niche.

if I'm looking at my real-time waveforms on a 24" monitor.

There is no such thing like looking at realtime waveforms beyond a few Hz. Your eyes aren't fast enough to follow a high frequency signal, your oscilloscope updates its screen much slower than a high frequency signal comes in, and your oscilloscope has a deadtime where it doesn't even look at the incoming signal at all.

And 24"? Well, the key in seeing what you want is not looking at endless rubbish, but learning to use the trigger to get what you want to see.

[T4P]

If it let you do gigabit transfers, then I could see companies jumping on it because then all the scope would need to do is digitize the signal and send it to the computer.

Um... I think you mean gigaBYTE transfers, don't you? Because It DOES let you do gigabit transfers - 5Gb/s max. In terms of transferring captured data, probably something closer to half that (2.5Gb/s) when you consider overhead. That means it's theoretically feasible to build a USB 3.0 scope with a 300MS/s rate (30MHz bandwidth?) that sends the data directly to the PC.

There's no other benefit to building one as a company, and most users see the lack of a screen as a negative.

This belies the fact that there are already dozens of USB scopes - introduced, for the most part, after the 2.0 spec. was adopted. And again, I personally don't see the lack of a tiny 6" screen as a negative if I'm looking at my real-time waveforms on a 24" monitor. But, hey, that's just me.

I agree with you, 5gigabit max but i have no knowledge of data transmission so i cannot comment further.
As for because i'm using a 22 incher i find it massively easier to track measurements down i mean like you get lots of clarity and resolution and if the screen goes i only need to replace the screen not the entire scope, but i would prefer the MFCTRS to place a big hunkin high speed sram and only send data not all at a time

BAW, erm, let's put something into perspective... would you rather view all 4 channels(eg) on a 8" or a 24" ? It's easier on the eyes

[hesam.moshiri]

Recently I just got 7102V and I'm satisfied. Till now I've not seen any of the previous problems like backlight, or Channel one noise yet ..., By the way I'll check more. Firmware version is 2.6.1

[rf-loop]

Recently I just got 7102V and I'm satisfied. Till now I've not seen any of the previous problems like backlight, or Channel one noise yet ..., By the way I'll check more. Firmware version is 2.6.1

AFAIK it is functionally same as 2.5.1

(it is also oscilloscope product version, not exactly FW version.)

Owon display show "Version" not Firmware version if think features and user interface (UI)

Example, if something change in HW it may also affect this version number  but with user eyes it is same. (service need know) Sometimes there may be manufacuring reasons for do some small  component level changes but for specs or UI it do not mean anything. For end-user it is same value.

It is littlebit "confusing" but it is Owon way to do it and that's it.

I  have also shortly  looked version 2.7.
 I have not find any functional or stability difference if look with user eyes.  But in service level it mean "something".
It works just same as 2.5.1 or 2.5.1.6

(2.5.1.x.x  is latest software shared by Owon as update)
--------------

Special CH1 noise issue you can not find becouse there is not it!
This issue was solved last year! And it was only in some manufacture lots.

TFT back light control have generated some amount of noise to channels in some manufacuring lot and it is solved also.  If someone get this from some seller stock, there is solution and Owon will help to remove this problem.

I will ask Owon: Why make changes to good product. Please do not make changes if all is ok.
Good example was this original TFT back light control circuit. It was good. It works ok. It can adjust also light from very dim to full. Not only numbers but also real light intensity.
What was problem/reason for make next, next and next changes. Nothing. Total lack of experience?  If designer can not read some components datasheets right or understand these,   what he is doing in design team? It is best to send him out for play mahjong and not gambling with designs. 

[akis]

Could someone please tell me is there a latest release/model/version of the SDS1702 to look for when buying? I would like to avoid buying some old stock, if at all possible.

[aghp]

Could someone please tell me is there a latest release/model/version of the SDS1702 to look for when buying? I would like to avoid buying some old stock, if at all possible.

My opinion is that there is not any seller who sell this OLD model what is OP rewieved in this topic.

If you want highest BW try buy 2.5.x.x version.

If you want later version with reduced BW buy new 2.6.x.x  --->
(yes these MAYBE can modify IF really want higher analog front end BW becouse... but it is also good to understand what all things it means if BW is more wide.)

Of course reduced BW have less noise.

This is also explained in many papers published by Hewlett-Packard and later Agilent.

There is no need worry what version you buy. All these are good today  and well better than most of others in this class if look signal quality.

How it can buy TV or computer if do not know that power supply is rev b.2.1 and not crap b.2.2. No one tell you what is inside and what version things there is and still model is exactly same. How it can buy? What makes oscilloscope different.
There is specifications. Do not believe you get anything more.

Do we really know and realize this or same kind of sentence what reads nearly every place.:

Well known Fxxxx:

"NOTE: The design of this instrument is subject to continuous development and
improvement. Consequently, this instrument may incorporate minor
changes in detail from the information provided in this manual."

And it really mean it in all meanings what it may mean.

How I can buy Fxxxx product?

<kind smile>

[akis]

Hey thanks. I presume 1.5.x and 1.6.x is the firmware version that can be seen on the screen so I can ask the supplier to send me a "1.6.x" version?

All I wanted to know is that I am not buying some old stock that, as someone said above, is not made anymore, receives no more software updates and has a ton of bugs that have been fixed in the later versions.

[T4P]

The newer ones come with LAN right ?

[aghp]

The newer ones come with LAN right ?

It is factory default, not option. It have been around half year. (still some shop want more money if you want LAN as option. (If someone do not want LAN, I do not know how they do..  )
 

[akis]

When we say LAN option does this mean an ethernet port like a printer so that a PC may connect to it ?

[aghp]

When we say LAN option does this mean an ethernet port like a printer so that a PC may connect to it ?

LAN port is just normal (but slow) ethernet and physically it have "RJ45" connector.

Yes you can connect all  xx number of classroom scopes to one hub and then teacher can read all scopes or show example one selected scope on the big classroom   screen.

Or you can install one oscilloscope to some system in China or neighbour home and read/watch  it in Zimbabwe or in you car over internet connection. (pity it is only for read oscilloscope)

If you can read User manual maybe it help something.

If need just connect one Oscilloscope to PC on the same or near table there is not any advantage with LAN. (also USB is faster)
From Oscilloscope you can not "send" anything to ethernet or USB (I mean example image to printer.  It is just for READ oscilloscope and anything else.

PC software is tiny and poor. It is made just only for reading data from scope. It can poll data continuously and save it to your disk. (screen captures, vector data or measurements) Also it can translate vector data to other format example to CSV for using with third party softwares.  It is NOT as USB oscilloscope software what you can use from PC and command oscilloscope and chhange settings. It is just "read scope and save data to PC" (and convert .bin vector data files to .csv.)