New 2GHz touchscreen scope from Tek, June 6th

[lukier]

You mean less expensive right Instead of 5 times the "real low-cost" market, only 3 times

Low cost for A-brand But I agree with nctnico that A-brands maybe should forget about the low end (i.e. 100-200 MHz and anything that can be made out of off the shelf FPGA+DDR+ADC) as competing with Rigol or Siglent is a race to the bottom. DSOX1000 however is not that bad, even cost wise when compared to buying e.g. non-hacked Rigol of the same bandwith with embedded decoders and AWG options. And it might be a strategy to get Tek out of the EDU market.

But I think here on the forum Tek got a bad rep not because they don't make $300 scopes, but because the mid-range (200Mhz-1GHz) offer is rather poor (slow scopes, expensive etc).  I guess that's the range that most people here that are above student budgets are interested in (semi-pro, small business etc). While Keysight has 2k/3k/4k series, R&S enters this market with RTB and LeCroy with WS3k. I doubt many people here are willing to spend more than $20k on a scope for their home hobby lab. Tek's rightfully innovative IsoVu probe alone is around $20k - that's why we don't have a 50 page thread about it on this forum.

[tautech]

But I think here on the forum Tek got a bad rep not because they don't make $300 scopes, .............

Actually they do, Tek just don't sell them at that price level. 

[KE5FX]

But I think here on the forum Tek got a bad rep not because they don't make $300 scopes, .............

Actually they do, Tek just don't sell them at that price level. 

Shots fired

[tautech]

But I think here on the forum Tek got a bad rep not because they don't make $300 scopes, .............

Actually they do, Tek just don't sell them at that price level. 

Shots fired

Didn't want it come across that way, only stated as fact.
Take a TDS2000, made in Asia to no better spec than many other entry level DSO's and sold for n times higher price. 

[nctnico]

The price isn't just hardware but also firmware which actually works.

[nctnico]

Don't you know that this forum is full of Tek bashers !

Not true. Just compare Tektronix oscilloscopes to their competition and you'll see the competition wins.

If they make something that is midrange then it is too slow even though they integrate a fully fledged spectrum analyser into it.

That full fledged spectrum analyser should be taken with a pinch of salt. The spectrum analysis features in the MDO series is more like an extensive FFT function compared to a real spectrum analyser when it comes to dynamic range. Also a scope which works slow is not very attractive to use.
BTW I own a Tektronix logic analyser which is a great instrument but the big let down is the Windows software. It is extremely slow for some reason. It never pushes the CPU usage to 100% even when working with an acquisition which is stored on a local hard drive.

[EEVblog]

But I think here on the forum Tek got a bad rep not because they don't make $300 scopes

But they DO make a $450 scope! And they are very rightly judged upon this scope (TBS1000), which is literally based on a 25(?) year old design with 2.5k of sample memory.
They can make the excuse that "the customers asked for it" all they want, it doesn't matter. You get judged on every new scope you bring to market.

[EEVblog]

In the US, we have an option in the tax code called Section 179 to allow certain amounts to be deducted as expenses 100% in the first year, but you still need to depreciate the asset value over the full term (i.e. 5 years).

In Australia you can now write off the entire cost of any asset $20k in the first year 
When they announced that, car companies started mysteriously selling utes with contract values of $19,999

[WackyGerman]

So maybe the new scope could be a 19999 $ model , who knows 

[boggis the cat]

But they DO make a $450 scope! And they are very rightly judged upon this scope (TBS1000), which is literally based on a 25(?) year old design with 2.5k of sample memory.
They can make the excuse that "the customers asked for it" all they want, it doesn't matter. You get judged on every new scope you bring to market.

Tektronix want to keep their brand in the education market, and they will get those low-end models for a lot less.  The alternative is to let Rigol or Siglent take that niche, then students are familiar with those brands and not Tektronix.

So I understand why they want to produce old technology cheaply for that market.  (It is at least Tektronix technology, rather than rebadging other products.)

The problem with Tektronix, IMO, is having quite variable 'user experience' in their pricey models.  Some are responsive and accurate, others are really painful to use.  You really need to check them out in person rather than by the spec sheet.  Keysight seem to do better on that count, and in many cases offer better value on top of that.

*cough* Danaher kiss of death *cough*

[snoopy]

Not true. Just compare Tektronix oscilloscopes to their competition and you'll see the competition wins.

So are you saying that is true of every scope they make ?

[jjoonathan]

Not true. Just compare Tektronix oscilloscopes to their competition and you'll see the competition wins.

So are you saying that is true of every scope they make ?

No, the fact that their UIs tend to clunk a bit means that you are allowed to ignore all of the things they do well in comparisons, which leads to the competition always winning. Them's the rules.

[KE5FX]

Not true. Just compare Tektronix oscilloscopes to their competition and you'll see the competition wins.

So are you saying that is true of every scope they make ?

No, the fact that their UIs tend to clunk a bit means that you are allowed to ignore all of the things they do well in comparisons, which leads to the competition always winning. Them's the rules.

A clunky or slow UI suggests that the people building the scope either didn't have a lot of experience with scopes in general, didn't have the budget or time to do the job right, or just plain didn't care. 

The first rule of Test Equipment Club is you eat your own dog food.

[Muxr]

A clunky or slow UI suggests that the people building the scope either didn't have a lot of experience with scopes in general, didn't have the budget or time to do the job right, or just plain didn't care. 

The first rule of Test Equipment Club is you eat your own dog food.

That is true, and it's probably the reason why the old analog Tek scopes were so great, because the guys that used them designed them. I'd venture a guess that the UI and software is probably done by software peeps who don't use a scope on a daily basis.

[timb]

Not true. Just compare Tektronix oscilloscopes to their competition and you'll see the competition wins.

So are you saying that is true of every scope they make ?

No, the fact that their UIs tend to clunk a bit means that you are allowed to ignore all of the things they do well in comparisons, which leads to the competition always winning. Them's the rules.

A clunky or slow UI suggests that the people building the scope either didn't have a lot of experience with scopes in general, didn't have the budget or time to do the job right, or just plain didn't care. 

The first rule of Test Equipment Club is you eat your own dog food.

AFAIK, it's only the MDO3000 that's laggy and only when you're using the spectrum analyzer portion. Yes, it could have used a beefier CPU, but it is what it is at this point. The DPO/MSO2000, 4000 and 5000 and are completely responsive. (Well, waveform update rate can slow down on the 2000 series when using FFT, but the controls are still perfectly responsive).

[nctnico]

AFAIK protocol decoding is also extremely slow on these scopes.

[timb]

AFAIK protocol decoding is also extremely slow on these scopes.

I don't know about the other series, but protocol decoding is basically real time on my MSO2024B. No lag at all and moving between decoded packets with the next/previous search buttons is pretty much instant. That performance keeps up even while simultaneously decoding a SPI and 16 ch parallel bus, along with three analog channels.

I can't image the 4000 or 5000 series would be *worse* considering they have much beefier processors.

[nctnico]

Also with deep memory (20Mpts) enabled?

[Keysight DanielBogdanoff]

AFAIK protocol decoding is also extremely slow on these scopes.

I don't know about the other series, but protocol decoding is basically real time on my MSO2024B. No lag at all and moving between decoded packets with the next/previous search buttons is pretty much instant. That performance keeps up even while simultaneously decoding a SPI and 16 ch parallel bus, along with three analog channels.

I can't image the 4000 or 5000 series would be *worse* considering they have much beefier processors.

I don't want to Tek bash, but I'd be shocked to see one of those scopes not slow down while decoding with 3 analog and 16 digital channels running (especially with memory turned up). UI responsiveness can be hard to characterize, but update rate tells the story there. For single-shot or slower devices I'm sure it does fine, but in a "real time" situation I have trouble using the 4000 and below scopes. I don't have much stick time with scopes faster than that, so I don't have first-hand experience there.

[w2aew]

For what it's worth - I use my MDO4000C nearly everyday, and don't find it to be slow, sluggish, clunky, etc.  You'll see the MDO in dozens of my YouTube videos, and I doubt that any of them show any sluggish behavior. 

Now, can you make it get sluggish? - sure you can.  Because the scope gives you the ability to directly control the waveform record length *AND* the fact that waveform processing (such as waveform measurements, Search & Mark, etc.) operate on the full waveform record and not a decimated set of display samples - you can make the scope "sluggish" by increasing the record length and turning an operation that pounds through the entire record (like Search & Mark).  Similarly, asking for a very narrow RBW on the spectrum analyzer will result in a huge FFT vector length which takes time for the embedded processor to crunch.  I do wish that the UI buttons were handled in a separate processing thread, this would help the UI responsiveness in this corner case.

But for everyday operation and use cases, I find the scope very responsive and lively.

[snoopy]

AFAIK protocol decoding is also extremely slow on these scopes.

I don't know about the other series, but protocol decoding is basically real time on my MSO2024B. No lag at all and moving between decoded packets with the next/previous search buttons is pretty much instant. That performance keeps up even while simultaneously decoding a SPI and 16 ch parallel bus, along with three analog channels.

I can't image the 4000 or 5000 series would be *worse* considering they have much beefier processors.

I don't want to Tek bash, but I'd be shocked to see one of those scopes not slow down while decoding with 3 analog and 16 digital channels running (especially with memory turned up). UI responsiveness can be hard to characterize, but update rate tells the story there. For single-shot or slower devices I'm sure it does fine, but in a "real time" situation I have trouble using the 4000 and below scopes. I don't have much stick time with scopes faster than that, so I don't have first-hand experience there.

I'm curious. Can your scope search through the whole acquisition record for a certain decode value on say an SPI bus or do we only get to see what is on the screen ?

[rsjsouza]

I do wish that the UI buttons were handled in a separate processing thread, this would help the UI responsiveness in this corner case.

Alan, in my experience that is probably already the case - a separate thread for the UI is not necessarily an advantage in a single processor architecture if other processing tasks are given a higher priority and do not "yield" to lesser threads. The solution would be either to adjust the priorities or have a second (or dual core) processor on the system.

[TestAment]

My 1st post on here having stumbled across the forum while searching for any bulletins by third party companies on the upcoming Tek 2G.
Basic info provided by a 'friend on the inside'.
Its an 8 channel scope with higher ADC resolution, as you increase the digital channels in use, you sacrifice the number of analog channels you use. Up to 125M record length shared across all channels - interleave 2x 62.5M channels. So advertised 10GS/s sample rates only apply for useless periods of time (as usual).
Usual Tek only probes with moulding around each input.
OK to use but RRP is high (no surprise). Will discount high to make the deal look better i.e. Half Price! on a unit twice the value it should be
---- I am just a messenger ---- Cease Fire! ----

[electrolust]

Its an 8 channel scope with higher ADC resolution, as you increase the digital channels in use, you sacrifice the number of analog channels you use.

That was going to be my guess, argh should have posted it!

Based on not seeing a dedicated digital port.  Even in the dramatically shaded photos you can see abscence of a digital input port.  That is, I'm guessing that each analog port is 4 or maybe 8 digitals, you decide based on whether you plug in a probe or a pod.  Not a bad setup.

[Keysight DanielBogdanoff]

I'm curious. Can your scope search through the whole acquisition record for a certain decode value on say an SPI bus or do we only get to see what is on the screen ?

Haha, I don't want to derail the thread debating this, but our scopes can actually search for packet info off screen as long as those packets were decoded at some point and are in the lister. So, you zoom out, capture & decode, and search through acquired data while having the scope scaled to only show one packet horizontally.