Keysight to launch new oscilloscope

[Stranger_danger]

Looks like Keysight is launching a new S-Series class product... from the looks of the image.

https://connectlp.keysight.com/New-Scope-Coming-Soon

[jjoonathan]

Ooh, shiny! Although I have to say the red-on-black color theme starts looking a bit sinister when they take it this far. In any case, you know what this means: it's baseless speculation time 

For some background research let's look at their current offerings and what they've been up to recently in other price tiers:

Infiniium S: 8GHz, 10GS/s * 10bit (can gang 2 channels)
UXR: 100GHz, 4 (InP sampling) * 4 (SiGe Sampling) * 16GS/s * 10bit


Alright, now I'm going to put those specs in a crystal ball and swish them around a bit. Behold, my prediction:

Infiniium T: 13 GHz, 16GS/s * 10bit (can gang 4 channels) -- Comes in any color, as long as it's black!

[Venturi962]

Series name could be MXR?

Two instances of this string in the page source: 'FY20_WW_WW_Digital_MXR-Oscilloscope-NPI-Teaser_CMP'

[jjoonathan]

I bet you're right. "MXR" sort of suggests that it's only a single tier below "UXR," though, so maybe this is more of a V/Z series replacement than an S series replacement? Perhaps the S series replacement will be "EXR."

[voltsandjolts]

Hmmm, how many Keysight shills have posted here?

[jjoonathan]

[CRTbrain]

Keysight did have a fairly large population of members of the Church of Jesus Christ of latter-day saints in their oscilloscope development/marketing.  Interesting to see this graphic.

[CRTbrain]

Tek also had a large LDS population.

[Electro Detective]

That page looks like an invitation to adware

too many questions like what Country and Confirm = outta there 

Looks like ol Luc and his team may have had a hand in the graphix perhaps,
to pump business along and maybe attract a few souls more,
to up the numbers back at Fire n Brimstone HQ 

So who's going to buy these things anyway, now that the corona fear industry has rooted businesses, R+D and education spend ? 

[DaJMasta]

I'm hoping for a somewhat overdue replacement for the DSOX4000/3000 series.  Decent scopes, but it's been like 10 years and it seems like the leading edge, the high end, and the budget market segments have already seen their updates.

[nfmax]

As the UXR uses an enhanced, faster version of the S-series ADC (and lots of them), I would expect that this new scope will use the same ADC: probably either a single or a pair, giving 10Ghz bandwidth on two or four channels respectively. It probably replaces the S series.

[Eric_S]

Ooh, shiny! Although I have to say the red-on-black color theme starts looking a bit sinister when they take it this far. In any case, you know what this means: it's baseless speculation time 

For some background research let's look at their current offerings and what they've been up to recently in other price tiers:

Infiniium S: 8GHz, 10GS/s * 10bit (can gang 2 channels)
UXR: 100GHz, 4 (InP sampling) * 4 (SiGe Sampling) * 16GS/s * 10bit


Alright, now I'm going to put those specs in a crystal ball and swish them around a bit. Behold, my prediction:

Infiniium T: 13 GHz, 16GS/s * 10bit (can gang 4 channels) -- Comes in any color, as long as it's black!

Seems reasonable.

Motor and drive testing seems to attract some money (I'm looking at you, LeCroy and Tek), with scopes in the same price bracket. So there may be an 8 channel version?

[MadTux]

Who cares? Can't afford one anyways and if I had one, it would be about as useful as an old Tek to me, except I would probably fry the inputs quite quickly with the work I usually do. Nice 20k$ or so doorstep, then.

Not to mention the price of those special probes. I rather get a used 10yo rigol for cheap, once the market is saturated. Once it's broken, into the dumpster it goes, similar to what would happen to a Keysight one, since repairability is about the same, only the difference in price is like 100x

[filssavi]

Yes a response to lecroy HDO4000/6000 and tektronix series 5 Would be very nice, since right now when we buy a scope (for power electronics) however the specs for I saw in this post are very far from what Would be necessary:

- up to least 6 channels, even better 8
- low-ish bandwidth (up to 1/2 GHz) more is not really needed and jacks up the price needlessly
- decently fast FFT with good UI
- optically isolated high CMRR probes (at least 120 dB possibly more) for high side gate drivers and such

Again if it is anything like what they have in the current lineup they are completely useless

[Mr. Scram]

Who cares? Can't afford one anyways and if I had one, it would be about as useful as an old Tek to me, except I would probably fry the inputs quite quickly with the work I usually do. Nice 20k$ or so doorstep, then.

Not to mention the price of those special probes. I rather get a used 10yo rigol for cheap, once the market is saturated. Once it's broken, into the dumpster it goes, similar to what would happen to a Keysight one, since repairability is about the same, only the difference in price is like 100x

Those Teks weren't any cheaper in their day. Without new test gear there is no old test gear. Who cares you can't afford it?

[MadTux]

Those Teks weren't any cheaper in their day. Without new test gear there is no old test gear. Who cares you can't afford it?

Things back then were build well, designed for reliability, had schematics and firmware published or EPROMs could be dumped and average joe could repair it with commonly available parts. That's why they are still around.

Stuff nowadays will be broken beyond repair, if it even gets anywhere near as old as these classic Teks before hitting the dumpster.
If I want crap that can't be repaired, I get some junk from china. At least it's at a price where I could throw it into the dumpster without much regret once it's broken. And a new one most likely costs less than a repair evaluation at KS.

That's why I don't care.

[Electro Detective]

MadTux certainly isn't alone in that logic and thrift mindset     

I'd still like to bump the previous question:
So who's going to buy these 'not so repairable' future dumpster candidates in the high $k anyway,
now that the corona fear industry has rooted businesses, R+D and education spend ? 

[wizard69]

I'm imagining a scope that doesn't charge you extra for built in features or simple software options.

[stafil]

I'm imagining a scope that doesn't charge you extra for built in features or simple software options.

Not going to happen. That's their bread and butter. Their real target are not individuals, but enterprises/companies, which don't have problem with the current scheme.

For individuals there is always hacking.

[filssavi]

MadTux certainly isn't alone in that logic and thrift mindset     

I'd still like to bump the previous question:
So who's going to buy these 'not so repairable' future dumpster candidates in the high $k anyway,
now that the corona fear industry has rooted businesses, R+D and education spend ? 

Regular companies will buy them, what needs to be kept into mind is that the for them the scope is a fundamental tool that makes money, if we assume the scope will last for 10 years (which might be even conservatives). A 10/15k equates to a 1000 to 1500 bucks a year, that is just not that much compared to wages

As for the repairability, the majority of the companies will either ship it back to the manufacturer if it is new, if it is old however it gets replaced

[TheUnnamedNewbie]

Makes perfect sense that these are not repairable on component level. It is just not feasible to do it because stuff is too small. Can't make it bigger, because then it gets more expensive and performance goes down. Integrate more and more is the reason that a 'cheap' scope (compared to the whole line-up) now can do better than the fastest scopes could 30-40 years ago.

Just look at how much time guys like Shahriar from TSP spend on repairing such a device. Consider the cost of such a repair engineer and his tooling. It is cheaper and faster for everyone to just have module-level repair.

Nobody who buys the high-end stuff cares about if it lasts 40 instead of 10 years. In fact, I would be annoyed if they made my scope more expensive just so it can go on for 40 years instead of 10 years. In 10 years, it is likely this scope won't be making me any money anymore, and I will be getting a new one anyways. Over-engineering is just as bad as under-engineering.


Same with the licensed upgrades. I don't wanna pay for the R&D time of some PCIe or whatever verification toolbox if I don't need it.

 

I bet you're right. "MXR" sort of suggests that it's only a single tier below "UXR," though, so maybe this is more of a V/Z series replacement than an S series replacement? Perhaps the S series replacement will be "EXR."

Isn't the UXR kinda the replacement for the Z-series?

[Mr. Scram]

Things back then were build well, designed for reliability, had schematics and firmware published or EPROMs could be dumped and average joe could repair it with commonly available parts. That's why they are still around.

Stuff nowadays will be broken beyond repair, if it even gets anywhere near as old as these classic Teks before hitting the dumpster.
If I want crap that can't be repaired, I get some junk from china. At least it's at a price where I could throw it into the dumpster without much regret once it's broken. And a new one most likely costs less than a repair evaluation at KS.

That's why I don't care.

Things used to be better in the old days. Old men shaking fists at clouds. We can't even have asbestos for breakfast any more. We know the drill. They made plenty of crap back then. There's a lot of survivorship bias and it's good you can repair things as it seems to be required quite a lot. I'm glad people are preserving history a bit but I'm also glad the status quo is advancing. Keysight's 2000X series has likely been a catalyst for all the cheap and very capable oscilloscopes available today. These capabilities were laughable dreams not too long ago for any regular mortal. Modern hardware is more capable and orders of magnitude cheaper than old boat anchors were in their day. Move over, progress is coming through. I don't mind the classics and enjoy using them and tinkering with them as much as the next guy but clinging to them or putting them on a pedestal while hissing at progress is pathetic.

[MadTux]

Sure, there is some survivorship bias involved (for example, most HP 8xxx pulse generators were garbage IMO) but with the throwaway instruments they make nowadays, hardly anything will be left in 40 years.

Either because configuration from PLD or microcontroller EPROMs leaks empty and no way to recover. Or it looses calibration and no public info on how to calibrate. Without schematics, even something simple can fail and you can't repair, because you don't realize it's broken, without the insights a good schematic provides. So why no schematics, service and calibration info now, if I pay 10x the price of a chinese instrument?

The better scopes/logic analyzers back then were all modular, Tek 7k, 11k, HP 54750..., makes an instrument much more versatile and it's simple to replace the frontend, once something breaks down. Makes diagnostics simpler as well, than the 1board garbage we have nowadays. Why not today, make 1 good instrument with different plugins instead of 10, each artificially crippled in a different way? And split things up onto different boards, so things can be tested by swapping them. Once a moderately complex board is identified as broken, parts level repair usually isn't that hard, in case the manufacturer would provide the parts.

[SilverSolder]

[...]  if I pay 10x the price of a chinese instrument?  [...]

Surely that is the key to the mystery...   If a throwaway instrument is sold at throwaway prices, presumably few would mind.

The day will come when you can subscribe to an oscilloscope with fixed monthly payments, just like we see subscription models for everything else nowadays...

[Keysight DanielBogdanoff]

Ooh, shiny! Although I have to say the red-on-black color theme starts looking a bit sinister when they take it this far. In any case, you know what this means: it's baseless speculation time 

We're taking bets over here.