Keysight DSOX1202A specs (2020) are less impressive than HP54645D specs (2000)

[CDaniel]

This topic is stupid , what changed fundamentally in all the cars from 20 years ago ? Fly or have top speed of 350Km/h ?
Of course the new scope has more measurements , FFT , math and many more small improvements . If you want 1GHz bandwidth buy an expensive one . Some options will remain options regardless of time .

[pascal_sweden]

I was comparing the base model of the Keysight DSOX1202A with external trigger, without the options for which you have to pay extra!
This means a bandwidth of 100 MHz and a sample memory of 1 MB, exactly the same as 20 years ago.

Are you sure that the MegaZoom architecture in the Keysight DSOX1202A is MegaZoom IV?

What is the waveform update rate for the Keysight DSOX1202A?

I read that the HP54645D oscilloscope already had an update rate of 3 Million vectors per second.
If we can calculate the amount of vectors per second for the Keysight DSOX1202A, then we can compare them relatively.

Maybe the difference is not that big after all! And then we have the proof that the 20 years old architecture based on the Motorola 68000 was superior!

This would finally clarify the reason for my posting:

The base model of an oscilloscope manufacturer should be better in every single aspect than an oscilloscope which is 20 years old!

Anyhow, I give up now! There are too many people with flaps and without common sense here!

The bandwidth is the same as 20 years ago.
The sample memory is the same as 20 years ago.
The MegaZoom architecture is the same as 20 years ago.

1. wrong new one has 200MHz max bandwidth, and that is not for repetitive signals only but realtime. Huge difference.
2. Wrong, new one has 2Mpoint memory. But lame compared to competition, I agree.
3. Wrong. Old one had original Megazoom I, new one is Megazoom IV

[gslick]

Anyhow, I give up now!

Good.

There are too many people with flaps and without common sense here!

I only count one...

[TK]

I was comparing the base model of the Keysight DSOX1202A with external trigger, without the options for which you have to pay extra!
This means a bandwidth of 100 MHz and a sample memory of 1 MB, exactly the same as 20 years ago.

Are you sure that the MegaZoom architecture in the Keysight DSOX1202A is MegaZoom IV?

What is the waveform update rate for the Keysight DSOX1202A?

I read that the HP54645D oscilloscope already had an update rate of 3 Million vectors per second.
If we can calculate the amount of vectors per second for the Keysight DSOX1202A, then we can compare them relatively.

Maybe the difference is not that big after all! And then we have the proof that the 20 years old architecture based on the Motorola 68000 was superior!

This would finally clarify the reason for my posting:

The base model of an oscilloscope manufacturer should be better in every single aspect than an oscilloscope which is 20 years old!

Anyhow, I give up now! There are too many people with flaps and without common sense here!

The bandwidth is the same as 20 years ago.
The sample memory is the same as 20 years ago.
The MegaZoom architecture is the same as 20 years ago.

1. wrong new one has 200MHz max bandwidth, and that is not for repetitive signals only but realtime. Huge difference.
2. Wrong, new one has 2Mpoint memory. But lame compared to competition, I agree.
3. Wrong. Old one had original Megazoom I, new one is Megazoom IV

Sample memory is now 2Mpts, refresh rate is 200,000wfm/s.  Not even the current 3000T has 3Mwfm/s, it means something different.  On the modern KS, you can measure the wfm/s from the trigger output BNC.

[pascal_sweden]

The bandwidth is the same as 20 years ago.
The sample memory is the same as 20 years ago.
The MegaZoom architecture is the same as 20 years ago.

The only improvement is the increased sample rate and support for protocol decoding.

But where is the logic analyzer?

If one could properly implement a 16 channel logic analyzer on an oscilloscope with a Motorola 68000 microprocessor, why should one drop this after 20 years on an oscilloscope with a much more powerful microprocessor?

Regarding protocol decoding: I believe there were optional modules for protocol decoding already 20 years ago, that worked perfectly on the oscilloscope with a Motorola 68000 microprocessor.

You are right that the underlying microprocessor architecture is different:

The HP54645D architecture based on a Motorola 68000 microprocessor was super efficient!
The Keysight DSOX1202A architecture based on a much more powerful microprocessor is inefficient!

The oscilloscope from 20 years ago was designed by engineers that were much smarter than today's engineers!

Maybe the difference is not that big after all! And then we have the proof that the 20 years old architecture based on the Motorola 68000 was superior!

This would finally clarify the reason for my posting:

The base model of an oscilloscope manufacturer should be better in every single aspect than an oscilloscope which is 20 years old!

It should have a much higher bandwidth after 20 years!
Why isn't 350 MHz or 500 MHz standard today?

It should have a much deeper sample memory after 20 years!
Why isn't 200 MSamples standard today?

It should have a much higher waveform update rate after 20 years!
Why isn't 1 Million Waveforms per second standard today?

It should still have a logic analyzer after 20 years!
Why would a logic analyzer not be useful anymore today? Every user will need this some day!

[tv84]

This would finally clarify the reason for my posting:

The base model of an oscilloscope manufacturer should be better in every single aspect than an oscilloscope which is 20 years old!

Anyhow, I give up now! There are too many people with flaps and without common sense here!

Pascal,

I still haven't figured it out! 

Your OP was interrogative. As such, why don't you accept that others (with plenty of "common sense") have different opinions from the information/questions you posted?

If you aren't willing to accept their opinions why do you launch these type of open threads in the 1st place? Not everyone is willing to be  evangelized!

Now, excuse me, I'm going to the aviation forum to demand the same thing! I want to fly in a commercial SR-71, NOW!  (oops, it's not 20 years, it's 60 years old...)

[pascal_sweden]

I was hoping that more people here on this forum take my side, and agree that the vendor does it because of professional ignorance or pursuit of profit!

After 20 years of innovation, the bandwidth and the memory depth is the same, and the waveform update rate has not been significantly improved.

Only the sample rate has been improved, but that's about it!

In fact, the Logic Analyzer functionality has been removed. Really?

Check for yourself in the attached documents!

Note that the HP54645D oscilloscope was a real work horse, and could perform all this magic using an 8 MHz Motorola 68000 microprocessor, a 25 MHz Texas Instruments TMS320C50 DSP processor, and the MegaZoom architecture.

How is it possible that HP could make such a powerful oscilloscope back then with such a modest microprocessor and DSP processor?

There is a much more powerful CPU in today's Keysight DSOX1202A, still it does not perform any better whatsoever compared to its 20 years older brother!

Why does HP make an oscilloscope in 2020, that is very similar but in fact worse than an oscilloscope from 20 years ago?

Is this done on purpose to misguide and disrespect the intelligence of oscilloscope enthusiasts? Somehow I consider this a crime!

[2N3055]

Pascal,

you seem not to have enough knowledge on topic, to discuss this. Sorry.

For instance, old one claims 3 mega VECTORS. That means it was able to paint 3 milions interpolation segments on a screen. On most timebases that was same as 3 milion  POINTS per second. So at 500 horizontal points that sums up to 6000 triggers per second. Max.
Not 10000, not 100000, nor 1 milion screens per second. So you're wrong. And as I said 3000T series today has 1M triggers per second. Or in old parlance, that would be 600 milions of pixels/vectors. That is factor 200X better.

Analog bandwidth of input amplifiers is equally complicated and expensive as it was. Digital revolution don't mean squat there.

Short memory I give you. Ask Keysight why Rigol and Siglent can do it.

I have MSOX3000T and Zeroplus LAP-C, and Digital discovery, and Picoscope MSO. I don't use many channels very often. I don't repair old computers so much, nowadays most of the stuff is serial. PC based decoding is better for lot of decoding. MSO is nice but not necessary. 4ch scope does most of the job.

Of course manufacturers do it for profit. What else? And as long as they sell well, it means they're doing good job.
Industry doesn't require better. And Keysight and other A brands don't care much about cheap segment. Or your opinion as hobbyist.

[pascal_sweden]

I will stick with the HP54645D oscilloscope powered by the Motorola 68000.







For my next purchase I will definitely go for a Rigol or a Siglent oscilloscope.

These Chinese companies are very talented and build products for the masses not the classes!

In that respect they follow the Commodore philosophy! See attachment!

[pascal_sweden]

If you want to learn more about the system architecture in the HP54645D, check the April 1997 issue of the HP Journal.

https://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1997-04.pdf

[TK]

I like the 54622D. I think it has a 68030. Display is awesome

[tv84]

For my next purchase I will definitely go for a Rigol or a Siglent oscilloscope.

These Chinese companies are very talented and build products for the masses not the classes!

Very well said! 

[pascal_sweden]

HP54645D advertisement in the Nasa Tech Briefs Sept 1996 issue.

[pascal_sweden]

I am looking for the product presentation CD ROM on the HP54645D.

Does anyone have this CD ROM?