NEW 100MHz Keysight Scope Release on 1st March 2017

[porker1972]

£337!
 

[porker1972]

I was under the impression that all you have to do is to hit the Stop button on the InfiniiVision.

But the whole memory thing is not so important to me so I'm maybe the wrong person to discuss this.

Rather than stop, which leaves the last acquisition (with its memory halved from the ping-pong buffer) on the screen you press single and then it captures the next trigger with the full memory depth.

It is available, it's called segmented memory.


Yep, I actually did a quick video about this:

http://bit.ly/2bx0Jrr

What would be useful, especially for less frequent trigger events,  is a "repeated single" mode, which would be the equivalent of pressing "single" after each trigger. Maybe press and hold the single button to activate this mode?

[porker1972]

EDUX1000 I2C/SPI/RS232 Decoder: $190

Does that imply that the decodes for other models are more expensive ?
 
I think a problem they have is that they risk competing with their own 2000X series. Probably why there's no 4ch version. 
It's certainly a big improvement on the very tired Rigol badged model, and I'm sure it will sell because of Keysight's brand reputation. 
If they had included decodes and func gen as standard it could have been a much bigger deal - AFIAK nobody is yet including decodes as standard,  which in these days when pretty much everything has an MCU is just ridiculous.
I'm guessing the marketing suits thought the Tek TBS was real competiion and not a sad joke.

Do we know if the EDU will be on general sale, or just to educational establishments ?  I can sort of see that 100kpts isn't a big deal in education, but as we all know it's just hardware crippling, it still feels like a bit of a rip-off.

BTW another major manufacturer is releasing a new scope in about 2 weeks' time. Don't know how it's positioned market-wise but may know soon 
 

90% of the work we do is one channel, 2 channels won't matter for basic stuff, especially teaching. Funnily enough, the only time I worked with 4-channel scopes is when I worked at a company making RF amps, we still only used 1 channel but bought 4-channel scopes because we used to blow up a 50-Ohm input every week and it was cheaper to send one back for repair each month than once a fortnight!

It looks like the EDU model is marketing spin, just as Tek did with their -EDU model (2.5k memory!). Now that was a company selling on brand name, the product was 20 years old in a new box.

I spoke to a Keysight distributor this morning, he said education customers get a discount but they can sell both EDU and DSO models to anyone. They said they have to prove to Keysight it was an education site to offer the discount, or they don't get any support and basically make a loss on the sale. Volume discount to Universities would put this on par with all the cheap stuff from China but with software, extra features, warranty and support, it's no-brainer to go with a premium brand in such a case.

We had a couple of Agilent X2000s a few years ago, some additional features were unlocked with a (genuine) firmware upgrade a year or so later. Apparently there will be another firmware upgrade soon for the 2000 and the distributor is expecting more things to become unlocked as standard, so existing customers don't lose out. Maybe there's no need to hack it and invalidate the warranty after all.

[Andrey_irk]

AFIAK nobody is yet including decodes as standard,  which in these days when pretty much everything has an MCU is just ridiculous.

Pico: 1-Wire, ARINC 429, CAN, DCC, DMX512, FlexRay, Ethernet 10Base-T, USB 1.1, I²C, I²S, LIN, PS/2, SPI, SENT, UART/RS-232. Standard on all models. In fact I yesterday received 2408B and it's a little monster with its 128MB memory   

Yes, but apart from that its still an USB scope  so it can't compete with the Keysight in terms of waveform update rate.

[JPortici]

 
they uploaded the manual. good.
no real info on the decoders/triggers. bad

1) It was mentioned that some decode functions can be assigned to the external trigger input (which makes totally sense. why the hell don't others already do this?)
but which ones? only chip select as the manual would hint?

NOTE: The 2-channel DSOX1000-Series oscilloscopes support 3-wire SPI. The MOSI and MISO
signal settings are forced to be the same; essentially, you can probe one or the other

so EXT couldn't be i don't know... CLOCK so then you could do full duplex and limit packet with timeout. why? WHY?
(looking at you, 2 chan scope manufacturers. Probably a pico can.. i don't know, mine doesn't have an ext trigger)

2) Limitations? CANBUS baud rates? UART baud rates? Arbitrary? Fixed?

[Zbig]

Man, this is anticlimactic. After Dave hyped this up hinting at the possibly game-changing pricing, turned into kind of a meh-fest after the actual prices have been revealed.

[MrW0lf]

Yes, but apart from that its still an USB scope  so it can't compete with the Keysight in terms of waveform update rate.

So youre disputing manufacturer claims then, based on what 
Most of Pico 2xxx series: 80,000
1000 X: > 50,000 wfm/s

[mikeselectricstuff]

AFIAK nobody is yet including decodes as standard,  which in these days when pretty much everything has an MCU is just ridiculous.

Pico: 1-Wire, ARINC 429, CAN, DCC, DMX512, FlexRay, Ethernet 10Base-T, USB 1.1, I²C, I²S, LIN, PS/2, SPI, SENT, UART/RS-232. Standard on all models. In fact I yesterday received 2408B and it's a little monster with its 128MB memory   

Yes, but apart from that its still an USB scope  so it can't compete with the Keysight in terms of waveform update rate.

You're assuming it's a dumb adc-only type USB scope, as opposed to a proper scope in a box with UI over USB.

[Andrey_irk]

Yes, but apart from that its still an USB scope  so it can't compete with the Keysight in terms of waveform update rate.

So youre disputing manufacturer claims then, based on what 
Most of Pico 2xxx series: 80,000
1000 X: > 50,000 wfm/s

Yes, but apart from that its still an USB scope  so it can't compete with the Keysight in terms of waveform update rate.

So youre disputing manufacturer claims then, based on what 
Most of Pico 2xxx series: 80,000
1000 X: > 50,000 wfm/s

Well, I may be wrong, but I see they claim "10 000 waveforms in a 6 ms burst typical", which makes more sense.
BTW, how can you transer so much data via USB? Tektronix says that to get the most out of your RSA306 apart from USB3.0 you have to have an SSD and Core I7. Maybe their software can do more staff simultaneously, but still...

http://www.saelig.com/product/picoscope-2408b.htm

link added

[nctnico]

BTW, how can you transer so much data via USB?

You don't! All the data is 'cooked' inside the FPGA and only what has to be drawn on the screen is transferred. For the screen you need 1000 points or so instead of the entire amount of data that has been recorded. Most DSOs works this way as well but have a more direct link to the FPGA.

[MrW0lf]

Well, I may be wrong, but I see they claim "10 000 waveforms in a 6 ms burst typical", which makes more sense.

https://www.picotech.com/oscilloscope/2000/picoscope-2000-specifications
Waveform buffers: 10,000
Maximum waveforms per second: 80,000
Max. waveforms in rapid trigger mode: 10,000
Trigger rearm time in rapid trigger mode: <1 us on fastest timebase

From that I understand it collects up to 10k wfm at max rate of 80k per sec, then sends to PC as batch.

However I never had to use this mode because it quite ok in normal mode due to some hardware compression they claim.

Guess every approach has pros and cons. Think doing extensive review on it in coming weeks. So far kicks like a mule: 1GSa/s at 5ms/div timebase  Also I "hacked" measurements system at it seemingly does 6-7digit horizontal accuracy in statistical measurements.

[Andrey_irk]

BTW, how can you transer so much data via USB?

You don't! All the data is 'cooked' inside the FPGA and only what has to be drawn on the screen is transferred. For the screen you need 1000 points or so instead of the entire amount of data that has been recorded. Most DSOs works this way as well but have a more direct link to the FPGA.

But I thought the idea of connecting a scope to a PC is not only in using it as a display, but in getting huge memory and decoding capabilities. And if you send only what is on the screen you can decode only that which is not much really.

[Jono427]

Page is up for the 1000x now:

http://www.keysight.com/en/pcx-2759552/infiniivision-1000-x-series-oscilloscopes

$448+ USD for the 50MHz ones

$647+ USA for 70,100MHz ones

Edit: Woops - some how missed a page and a half between yesterday and this morning...

[Andrey_irk]

A good review with pointing out all the advantages and limitations of the thing would be great.

[EEVblog]

Man, this is anticlimactic. After Dave hyped this up hinting at the possibly game-changing pricing, turned into kind of a meh-fest after the actual prices have been revealed.

The original pricing I was told they were shooting at seemed like a winner. Lower starting price and cheaper, almost no-brainer options.
But it seems that somewhere along the line someone decided it needed to go blow-for-blow with the Tek TBS1000
I had no idea the low end would be an EDU crippled model.

[EEVblog]

Please no more talk on the Picoscope here, take it to another thread.

[Zbig]

The original pricing I was told they were shooting at seemed like a winner. Lower starting price and cheaper, almost no-brainer options.
But it seems that somewhere along the line someone decided it needed to go blow-for-blow with the Tek TBS1000
I had no idea the low end would be an EDU crippled model.

Yeah, I realize you were surprised too, sorry if I made it sound otherwise.

[S13]

Hmm yeah, after seeing the teardown video and pricing i no longer think this will be a game changing scope like its predecessor 2000X and 3000X series were at their launch 6 yrs back.
Dont get me wrong, its still a good scope with nice features and something i would love to have sitting on my bench, but value-for-money-wise its now more on par with other existing low budget scopes.

[nctnico]

BTW, how can you transer so much data via USB?

You don't! All the data is 'cooked' inside the FPGA and only what has to be drawn on the screen is transferred. For the screen you need 1000 points or so instead of the entire amount of data that has been recorded. Most DSOs works this way as well but have a more direct link to the FPGA.

But I thought the idea of connecting a scope to a PC is not only in using it as a display, but in getting huge memory and decoding capabilities. And if you send only what is on the screen you can decode only that which is not much really.

You can either do decoding in the FPGA or send all the data to the PC but then again decoding doesn't need to be fast. One way doesn't necessarily rule out the other way. Looking at numbers which change so quick you can't read them isn't very useful.

[ebastler]

Man, this is anticlimactic. After Dave hyped this up hinting at the possibly game-changing pricing, turned into kind of a meh-fest after the actual prices have been revealed.

The original pricing I was told they were shooting at seemed like a winner. Lower starting price and cheaper, almost no-brainer options.
But it seems that somewhere along the line someone decided it needed to go blow-for-blow with the Tek TBS1000
I had no idea the low end would be an EDU crippled model.

Never mind. After the first pricing information leaked through, I think we had pretty good price estimates. (See pages 3 and 4 of this thread.) And the existence and somewhat crippled specs of the EDU model were clear as well, as the spec sheet had been published at the same time. So there shouldn't be much disappointment now, as the official prices are nicely in line with what we had discussed four weeks ago.

[Fgrir]

Looking at numbers which change so quick you can't read them isn't very useful.

It is sometimes nice to be able to trigger off the decoded content though, which is only possible with real-time decoding.

[nctnico]

Looking at numbers which change so quick you can't read them isn't very useful.

It is sometimes nice to be able to trigger off the decoded content though, which is only possible with real-time decoding.

Which is easely done in an FPGA.

[mikeselectricstuff]

Looking at numbers which change so quick you can't read them isn't very useful.

It is sometimes nice to be able to trigger off the decoded content though, which is only possible with real-time decoding.

Which is easely done in an FPGA.

I suspect that in many implementations, the only major thing the decode shares with the trigger is the setup User interface. Maybe less so for more complex protocols.

[ebastler]

I agree with Mike. Nowadays protocol decoding should be standard on an oscilloscope.

Yes. And they should all be sold at the 50MHz price tag, but with 100 MHz enabled. 

Obviously the options are a way to show a lower sticker price, but still sell the product line at a profit. I agree that eventually manufacturers will begin to include some options with the base product, as they already do in occasional time-limited offers today. But that will mean that either the average prices of scopes will have to go up, or manufacturing costs need to be brought further down (without passing those savings on to the sticker price).

[tautech]

But the rust option is free.