Author Topic: True analog scopes  (Read 50475 times)

0 Members and 1 Guest are viewing this topic.

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Re: True analog scopes
« Reply #200 on: December 18, 2022, 10:07:55 pm »
CROs have both "Normal" & "Auto" trigger mode.

When I am looking at various points in a DUT, I normally let the 'scope "free run" in Auto.

DSOs have the same, and a few more, eg. Single shot mode where you trigger once and it grabs the signal so you can look at it at in your own time, maybe even zoom in for a closer look... very useful, but CROs don't do it.
Not true. Storage CROs do exist. These can capture a (slow) waveform and keep it on the display. There just isn't any zooming in/out AFAIK. I have used these 30 years ago.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Re: True analog scopes
« Reply #201 on: December 18, 2022, 10:22:23 pm »
Frequently during a professional career people will have to make do with what's already available because
  • you need it now, not in a month's time
  • money has to be used for other necessary equipment
  • it won't pass the purchase request process
  • it is back at base, not here in the field
  • boss says you don't need it
  • there is nothing that is capable of doing the job
I've seen all of those, and had to find workarounds.

Might as well get engineers to think creatively from the outset. That's not true for technicians, of course - and vive la différence!
if your company cant afford $400 DSO, i suggest find another company because they surely cant make raise to your salary accordingly... i've also encountered such situation during my short time working as practising (profession) eng... but there is difference/limit between working to the limit and asking the impossible... and every workaround has its catches... be it money, space or time... you wont get anything far with $1 tools, let alone going into GHz region, unless you want to work full time as dumpster diver, but again, the catch is... time... one of it... concrete example will prove me wrong... ymmv.

I suspect there is a language divide here, because you have missed the point(s).
Mechatrommer got the point just fine and is absolutely right. If your employer can't afford good tools that make you do your job efficiently, then it is time to leave. On several occasions I even brought my own equipment along just to get the job done quickly. I want to make progress with the project at hand and not get stuck in endless tinkering & detours. Like your comment earlier about Rigol & Siglent: too many problems may linger so I won't be so quick to use test equipment from those brands in a professional environment (especially with less experienced engineers around). I've learned my lessons; low purchase price and TCO are not the same number. I'm getting paid to create a new product, not to build and/or debug test equipment.
« Last Edit: December 18, 2022, 10:31:38 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #202 on: December 18, 2022, 11:04:46 pm »
CROs have both "Normal" & "Auto" trigger mode.

When I am looking at various points in a DUT, I normally let the 'scope "free run" in Auto.

DSOs have the same, and a few more, eg. Single shot mode where you trigger once and it grabs the signal so you can look at it at in your own time, maybe even zoom in for a closer look... very useful, but CROs don't do it.
Not true. Storage CROs do exist. These can capture a (slow) waveform and keep it on the display. There just isn't any zooming in/out AFAIK. I have used these 30 years ago.

Some were used to capture fast difficult-to-repeat events, especially when measuring nuclear detonations ;)
« Last Edit: December 18, 2022, 11:10:02 pm by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #203 on: December 18, 2022, 11:08:41 pm »
Frequently during a professional career people will have to make do with what's already available because
  • you need it now, not in a month's time
  • money has to be used for other necessary equipment
  • it won't pass the purchase request process
  • it is back at base, not here in the field
  • boss says you don't need it
  • there is nothing that is capable of doing the job
I've seen all of those, and had to find workarounds.

Might as well get engineers to think creatively from the outset. That's not true for technicians, of course - and vive la différence!
if your company cant afford $400 DSO, i suggest find another company because they surely cant make raise to your salary accordingly... i've also encountered such situation during my short time working as practising (profession) eng... but there is difference/limit between working to the limit and asking the impossible... and every workaround has its catches... be it money, space or time... you wont get anything far with $1 tools, let alone going into GHz region, unless you want to work full time as dumpster diver, but again, the catch is... time... one of it... concrete example will prove me wrong... ymmv.

I suspect there is a language divide here, because you have missed the point(s).
Mechatrommer got the point just fine and is absolutely right. If your employer can't afford good tools that make you do your job efficiently, then it is time to leave. On several occasions I even brought my own equipment along just to get the job done quickly. I want to make progress with the project at hand and not get stuck in endless tinkering & detours. Like your comment earlier about Rigol & Siglent: too many problems may linger so I won't be so quick to use test equipment from those brands in a professional environment (especially with less experienced engineers around). I've learned my lessons; low purchase price and TCO are not the same number. I'm getting paid to create a new product, not to build and/or debug test equipment.

I completely agree with you about it being a "time to leave" flag, but mechatrommer did miss the points I was making.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 7854
  • Country: au
Re: True analog scopes
« Reply #204 on: December 18, 2022, 11:33:46 pm »
CROs have both "Normal" & "Auto" trigger mode.
When I am looking at various points in a DUT, I normally let the 'scope "free run" in Auto.
and probably Wallace Gasiewicz's CRO doesnt have that feature, since from the way he stated, he never see "Normal" triggering effect in his CRO.

Very unlikely----I have never seen any CRO which has triggering at all, which doesn't offer both "normal" & "auto".

Very early, & a few later very cheap, 'scopes, had a "timebase" that was normally "free running" & used a "sync" control to "sort of" sync the timebase to the incoming signal.
He could, just possibly have one of those.
 

Online BillyO

  • Super Contributor
  • ***
  • Posts: 1848
  • Country: ca
Re: True analog scopes
« Reply #205 on: December 19, 2022, 03:06:55 am »
Here's a little use case for a CRO.

Just a few minutes ago I had a tiny epiphany about something I'm working on and I just wanted to check a signal.  No precise measurement, just wanted to verify it was there and looked like it should.  I went down to the lab, turned on the Tek 465, got a probe and attached it to the scope then to the DUT, twiddled the vertical and the horizontal, saw what I wanted to see, turned off my scope  and put the probe away.  All in about 1/2 the time it would take the SDS2104XP just to boot.
Bill  (Currently a Siglent fanboy)
--------------------------------------------------
 

Offline H713

  • Regular Contributor
  • *
  • Posts: 180
  • Country: us
Re: True analog scopes
« Reply #206 on: December 19, 2022, 05:31:56 am »
Everyone has a slightly different workflow and slightly different preferences when it comes to tools.

Some engineers, when setting up their lab, have both a DSO and an analog scope on their bench. Some will have multiple of each. Others will have just a DSO.

Some engineers will use a DSO for everything, while others will use an analog scope for some tasks and a digital scope for others. It's all an issue of personal preferences, workflow, bench space, etc.

This is no different from other industries. Some people still like to have a console in their recording studio, others do everything in-the-box. Some woodworkers use planes and chisels extensively, others choose to do everything with power tools. Neither way is "better", just different.

I think we can agree that only having an analog scope is a real limitation these days, but there are still valid reasons to have both - even if those reasons are as dumb as "adding some variety".


Why aren't they still made? Easy. DSOs with LCDs are objectively better for most applications, but more importantly, are far, far cheaper to produce and easier to make into a sexy form factor that is easy to sell. Doesn't mean there aren't valid reasons to use an analog scope, but it does mean that the market for new analog scopes is too small to justify the cost to produce them. CRTs are difficult and expensive to manufacture. I believe that there are some companies still making CRTs for maintaining legacy systems (military, aviation, etc) with long lifecycles, but we're talking very small numbers.
 
The following users thanked this post: robert.rozee

Offline Njk

  • Frequent Contributor
  • **
  • Posts: 348
  • Country: ru
Re: True analog scopes
« Reply #207 on: December 19, 2022, 08:26:38 am »
To finalize, current tech state of real time embedded real time has nothing to do with 8bit era.
Guys, take it easy. I recalled, when my daughter just graduated from the university (she decided to pursue an engineering career despite my objections), I asked her to pass my exam. Short time before I installed a new electricity meter that has IR interface, and I was interested in getting data from that meter in a high-tech style way. The protocol is not IrDA and not plain IEC 61107 (or how it is called now), it's a vendor-specific combination of the IrDA optical transceiver part with a pulse-based protocol like that used in remote controls.

So I asked her to design a PSK/FSK modem that convert that protocol to UART. To simplify the task, I bought a USB IrDA adapter, removed everything except the IR part and the USB bridge, and installed a cheapest Microchip PIC part between them (very basic 8-bit MCU with internal RC oscillator). She had just to write the "firmware".

For me, it was a simple task as I've done many things like that in late 80s. To show that it's doable, I'd implemented it over a couple of weekends. Worked just fine. BTW, analog scope was successfully used in the process. It's quite a real-time embedded application, isn't it?

Sadly, the exam was spectacularly failed. And the stuff becomes more and more complex over the years. I wonder who eventually will be able to do it with clear understanding of what he/she is doing.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #208 on: December 19, 2022, 09:58:59 am »
And the stuff becomes more and more complex over the years. I wonder who eventually will be able to do it with clear understanding of what he/she is doing.

That's an interesting topic.

Certainly I agree with you in that the sheer complexity of modern digital components, languages and libraries and frameworks means learning them is a significant cliff to climb. Too many people don't understand the limitations of their tools and/or don't look any deeper than the sales brochure or top-level description.

How can that be minimised? The best technique is a good education that is focussed on the underlying fundamentals of the topics that will remain valid as technology changes. It is only an exception person that can pick up such concepts solely "on the job"; I've come across one.

In contrast a bad education wastes too much time on the current fads which usually have a half-life of <5 years. Too many people have only a surface understanding of the tools and technology they are using. They can still be useful in many circumstances, but often they don't recognise how much they don't know.

On the other hand, I'm sure people were saying the same things 50 years ago about the technologies in use then. Certainly I remember doing practice maths exam papers from the early 50s, and thinking the questions were bloody hard!
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Re: True analog scopes
« Reply #209 on: December 19, 2022, 11:42:52 am »
In contrast a bad education wastes too much time on the current fads which usually have a half-life of <5 years. Too many people have only a surface understanding of the tools and technology they are using. They can still be useful in many circumstances, but often they don't recognise how much they don't know.

On the other hand, I'm sure people were saying the same things 50 years ago about the technologies in use then. Certainly I remember doing practice maths exam papers from the early 50s, and thinking the questions were bloody hard!
This is getting off-topic...
Old people do tend to get stuck in the past where 'everything was better'. But I have to agree that the quality of education has dropped significantly. A lot of people can get a bachelor's degree nowadays but only because the bar has been set to 50% to what it was -says- 30 years ago. My youngest son is studying software engineering at a bachelor's degree level. Much to my surprise they don't get any math lessons. When I bring it up he claims he'll just look it up on internet. Then I counter with: how do you know what to look for? I get no answer... I'm not going to claim I remember everything I was told during math lessons (I have two books covering all topics and those are 5cm thick each printed on really thin paper).

I also worked on software written by self-thaught programmers. In one case I asked the guy on how he got to a certain formula (partly implemented using ifs for different input ranges). His answer: I just fudged the numbers until the result looks OK. So I asked: and what if the input data is not what you expect, how to raise an error or make sure to show the result is obviously wrong? The guy answered with a shoulder shrug. FFS  :palm: Needless to say I replaced his crap with a single -continuous- formula that worked.
« Last Edit: December 19, 2022, 11:45:24 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #210 on: December 19, 2022, 12:16:53 pm »
In contrast a bad education wastes too much time on the current fads which usually have a half-life of <5 years. Too many people have only a surface understanding of the tools and technology they are using. They can still be useful in many circumstances, but often they don't recognise how much they don't know.

On the other hand, I'm sure people were saying the same things 50 years ago about the technologies in use then. Certainly I remember doing practice maths exam papers from the early 50s, and thinking the questions were bloody hard!
This is getting off-topic...
Old people do tend to get stuck in the past where 'everything was better'. But I have to agree that the quality of education has dropped significantly. A lot of people can get a bachelor's degree nowadays but only because the bar has been set to 50% to what it was -says- 30 years ago.

Any decent engineer will always be looking for the edge cases, thinking how to remove them, and jumping on any way possible to improve things. I'm frustrated where things haven't improved during my career!

In the UK in the 70s the "top" 8-10% went to university. The others became productive other ways.

Now the "top" 50% go to university. Since people haven't changed significantly in 50 years, if the percentage of 1st class degrees is the same then the boundary must be lower. (And in the UK there is currently angst that too high a proportion of people get 1st class degrees).

One of my heroes was Professor Eric Laithwaite at Imperial College. He used to set exams where one question was easy and sufficient get you a pass mark, several were more challenging and could get you a good degree, and one could not be answered adequately in the time available. He expected his undergraduate engineers to be able to determine which questions to avoid. If they couldn't, they wouldn't make good engineers anyway.

Quote
My youngest son is studying software engineering at a bachelor's degree level. Much to my surprise they don't get any math lessons. When I bring it up he claims he'll just look it up on internet. Then I counter with: how do you know what to look for? I get no answer... I'm not going to claim I remember everything I was told during math lessons (I have two books covering all topics and those are 5cm thick each printed on really thin paper).

Our 2nd year maths course was taught by someone from the maths faculty. When we showed our course notes to a maths student he said it was pretty much his entire year. I've no doubt that they went into it in more depth. The end of year exam rubric was "full marks may be obtained for answers to about six questions" :)


Quote
I also worked on software written by self-thaught programmers. In one case I asked the guy on how he got to a certain formula (partly implemented using ifs for different input ranges). His answer: I just fudged the numbers until the result looks OK. So I asked: and what if the input data is not what you expect, how to raise an error or make sure to show the result is obviously wrong? The guy answered with a shoulder shrug. FFS  :palm: Needless to say I replaced his crap with a single -continuous- formula that worked.

Ah yes. "It passes the unit tests therefore it works" belief. What frightened me was that "they" didn't see the flaw in that argument.
« Last Edit: December 19, 2022, 12:21:37 pm by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline Njk

  • Frequent Contributor
  • **
  • Posts: 348
  • Country: ru
Re: True analog scopes
« Reply #211 on: December 19, 2022, 01:32:20 pm »
How can that be minimised? The best technique is a good education that is focussed on the underlying fundamentals of the topics that will remain valid as technology changes. It is only an exception person that can pick up such concepts solely "on the job"; I've come across one.
I think it's typical quality vs. quantity problem. Yes, it was always like this. The billions of population does not mean that the percentage of individuals with autonomous surviving skills is now higher than at the pre-history time. Worse, more smart machinery effectively results in more dumb people.

Education is very important but the opportunity to use learned knowledge is decreasing as most people then have to do actually a blue-collar work where little of that knowledge is required.

For instance, previously I was designing my own SMPSs from discrete components, and a computer peripheral controllers with tens of LSI logic chips. Now it's no more needed, now it's actually an integration work that requires much less skills (left an FPGA alone). Later, when USB 2.0 just emerged and I was building a USB device, I wrote all the code from scratch. Now it's difficult to explain the reason because everyone knows that a drivers and libraries for all the classes are available. But it was not always like that and again, now much less level of understanding is required to put all the pre-built stuff together. It's something that is less and less of science and more and more of accounting.

Add to that the amount of "accounting" documents to read. Now it's typical to see a manuals of several thousand pages, not of hundred pages as before. As the speed of digesting an information remains constant, more time is required just to read that bs. You can't effectively master the HW without having to read the docs with due attention. Meanwhile, less and less time is available, because of increasing pressure from the market. I strongly suspect few developers did read that at all, because it's not necessary for a quick and dirty implementation.

Not a good environment for individual's personal skills development. Various "turnkey solutions" that every monkey can use are killing that very efficiently.
Few are lucky, but typically - want an interesting work, enjoy a low income.
« Last Edit: December 19, 2022, 02:06:41 pm by Njk »
 

Offline H713

  • Regular Contributor
  • *
  • Posts: 180
  • Country: us
Re: True analog scopes
« Reply #212 on: December 19, 2022, 01:39:53 pm »
Not sure how true this story is (came from a prof who was something of a nut), but it does relate to education.

There was an old Hitachi analog scope in the lab - something really crusty like 20 MHz. Nobody used it, but it was chained to a cart that was more valuable than the scope itself. One student spent some time with an Arduino and two DACs, and when thanksgiving break came, was the last person to leave the lab.

When everyone came back after thanksgiving, there was a dick and balls burned into the CRT.

Suffice it to say, the cart it was sitting on was liberated.
 

Offline Njk

  • Frequent Contributor
  • **
  • Posts: 348
  • Country: ru
Re: True analog scopes
« Reply #213 on: December 19, 2022, 03:11:16 pm »
BTW I'd obtained my first scope when I was a high school student. Found it on the local junk yard. It was so big and heavy I was barely able to lift it. With the help of my friend, we moved the treasure to my home. It was of very true analog type, with big CRT, no transistors inside, only the tubes. To my surprise, it was in working condition. I spent more than a year learning how a tube amplifier circuits work and what are the effects from changing the values of res and caps. Excellent learning tool.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #214 on: December 19, 2022, 03:12:50 pm »
How can that be minimised? The best technique is a good education that is focussed on the underlying fundamentals of the topics that will remain valid as technology changes. It is only an exception person that can pick up such concepts solely "on the job"; I've come across one.
I think it's typical quality vs. quantity problem. Yes, it was always like this. The billions of population does not mean that the percentage of individuals with autonomous surviving skills is now higher than at the pre-history time. Worse, more smart machinery effectively results in more dumb people.

Who knows, but the best back then were just as relatively inventive as the best now. Classic idiocy is "I don't know how to build pyramids, therefore they didn't, therefore it must have been aliens", and many varieties of that theme

Quote
...
Now it's difficult to explain the reason because everyone knows that a drivers and libraries for all the classes are available. But it was not always like that and again, now much less level of understanding is required to put all the pre-built stuff together.

Problems arise when the libraries are subtly faulty, possibly because the implementer wasn't up to the job or because they didn't document their assumptions about how it should not be used.

Quote
Add to that the amount of "accounting" documents to read. Now it's typical to see a manuals of several thousand pages, not of hundred pages as before. As the speed of digesting an information remains constant, more time is required just to read that bs. You can't effectively master the HW without having to read the docs with due attention. Meanwhile, less and less time is available, because of increasing pressure from the market. I strongly suspect few developers did read that at all, because it's not necessary for a quick and dirty implementation.

I recently had a very pleasant surprise in that regard.

I started to us a novel multi-core processor (up to 4000MIPSX) which is like an FPGA in that it allows hard real-time guarantees to be made without measuring and hoping. The RTOS is to all intents and purposes implemented in hardware.

The language and processor descriptions were notably short and very easy to digest. The subtleties were explained clearly with pseudo-code examples. There were no "here there be dragons" exceptions and errata. All that documentation was <200 pages, and fun to read.

I had zero surprises: everything just worked as expected, first time. The first kick-the-tyres speed tests were up and running within a day of receiving the hardware and firing up the tools. Stunning.

Then I added other things like a front panel and peeked/poked comms over a USB link, and everything continued to work as expected and at the same speed.

Absolutely no unexpected behaviour; everything just "worked as it said on the tin". What more could I want!
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #215 on: December 19, 2022, 03:15:36 pm »
BTW I'd obtained my first scope when I was a high school student. Found it on the local junk yard. It was so big and heavy I was barely able to lift it. With the help of my friend, we moved the treasure to my home. It was of very true analog type, with big CRT, no transistors inside, only the tubes. To my surprise, it was in working condition. I spent more than a year learning how a tube amplifier circuits work and what are the effects from changing the values of res and caps. Excellent learning tool.

Any working scope is better than no scope. A good engineer will work out how to use whatever tools are available.

Those that say something to the effect that "only the latest tool should be considered" are doing budding engineers a disservice.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: robert.rozee

Offline MegaVolt

  • Frequent Contributor
  • **
  • Posts: 931
  • Country: by
Re: True analog scopes
« Reply #216 on: December 19, 2022, 03:39:48 pm »
I started to us a novel multi-core processor (up to 4000MIPSX) which is like an FPGA in that it allows hard real-time guarantees to be made without measuring and hoping. The RTOS is to all intents and purposes implemented in hardware.
It is very interesting! Can you name him?
 

Offline Njk

  • Frequent Contributor
  • **
  • Posts: 348
  • Country: ru
Re: True analog scopes
« Reply #217 on: December 19, 2022, 03:53:20 pm »

Problems arise when the libraries are subtly faulty, possibly because the implementer wasn't up to the job or because they didn't document their assumptions about how it should not be used.
And even more problems when it's actually not an option to don't use the library. Because of one reason or another.

Quote
I started to us a novel multi-core processor (up to 4000MIPSX) which is like an FPGA in that it allows hard real-time guarantees to be made without measuring and hoping. The RTOS is to all intents and purposes implemented in hardware.
Intel 432 project resurrection?


Quote
Absolutely no unexpected behaviour; everything just "worked as it said on the tin". What more could I want!
That looks very suspiciously. Where's it came from? A reasonable question at present times. Perhaps it's designed to phone home on every occasion...
« Last Edit: December 19, 2022, 04:02:58 pm by Njk »
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #218 on: December 19, 2022, 04:22:46 pm »
I started to us a novel multi-core processor (up to 4000MIPSX) which is like an FPGA in that it allows hard real-time guarantees to be made without measuring and hoping. The RTOS is to all intents and purposes implemented in hardware.
It is very interesting! Can you name him?

Search this forum for my posts on the XMOS xCORE processors running xC.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: MegaVolt

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #219 on: December 19, 2022, 04:29:59 pm »

Problems arise when the libraries are subtly faulty, possibly because the implementer wasn't up to the job or because they didn't document their assumptions about how it should not be used.
And even more problems when it's actually not an option to don't use the library. Because of one reason or another.

Urg! Sometimes the best approach is to say "no, not me, get someone else in another company to do it" :) Knowing when to walk away is a key skill - as per Eric Laithwaite's exam questions.

Quote
Quote
I started to us a novel multi-core processor (up to 4000MIPSX) which is like an FPGA in that it allows hard real-time guarantees to be made without measuring and hoping. The RTOS is to all intents and purposes implemented in hardware.
Intel 432 project resurrection?

<deity> no!

Transputer (with a sprinkling of Sun's Niagara T chips) and more modern hardware, Occam/CSP, more pragmatism, and carefully limited objectives. It is hard realtime embedded, not general purpose.

CSP/Occam concepts/constructs are making their way into modern languages, e.g. Rust and Go.
« Last Edit: December 19, 2022, 04:34:07 pm by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline Njk

  • Frequent Contributor
  • **
  • Posts: 348
  • Country: ru
Re: True analog scopes
« Reply #220 on: December 19, 2022, 05:09:16 pm »
Transputer (with a sprinkling of Sun's Niagara T chips) and more modern hardware, Occam/CSP, more pragmatism, and carefully limited objectives. It is hard realtime embedded, not general purpose.

CSP/Occam concepts/constructs are making their way into modern languages, e.g. Rust and Go.
Interesting. I've a book Transputer Applications, edited by Gordon Harp from Royal Signals and Radar Establishment. It was published in 1989. That time, there was some hype about that. Transputers, Occam2, TDS, etc. I was thinking it's long time over
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: True analog scopes
« Reply #221 on: December 19, 2022, 06:06:42 pm »
Transputer (with a sprinkling of Sun's Niagara T chips) and more modern hardware, Occam/CSP, more pragmatism, and carefully limited objectives. It is hard realtime embedded, not general purpose.

CSP/Occam concepts/constructs are making their way into modern languages, e.g. Rust and Go.
Interesting. I've a book Transputer Applications, edited by Gordon Harp from Royal Signals and Radar Establishment. It was published in 1989. That time, there was some hype about that. Transputers, Occam2, TDS, etc. I was thinking it's long time over

The concepts have kept reappearing, if you know what you are looking for. Some have appeared in Texas Instruments DSP chips, some in libraries for mainstream languages, etc.

Now that single cores have hit a performance barrier, it is no longer to wait 18 months for performance issues to "disappear". Clearly the future is based around multicore processors, multiple threads, distributed memory, and distributed computing. Unfortunately the languages and concepts that implicitly presume single core computation haven't kept up. Yes, they battle valiantly, adding sticking plaster here and there, and will be around as long a COBOL. But new fundamental concepts are necessary.

The XMOS hardware/software ecosystem has tackled those limitatations and problems head-on without compromise amd without being unnecessarily hobbled by backward compatibility. They aren't sufficient for the future: we need more radical advances. But they do indicate the benefits that can be gained.
« Last Edit: December 19, 2022, 06:09:15 pm by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29809
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: True analog scopes
« Reply #222 on: December 19, 2022, 07:51:56 pm »
Any working scope is better than no scope.
It is however when you need to invest in a working scope one need consider if what's available is a worthy investment. Old stuff is just that, old and nearing the end of that bell curve of reliability where the novice especially isn't well served by some unreliable junk that needs a duplicate to help fix it.
Often that novice, a new Uni student is more focussed on projects and studies rather than need fix some old boat anchor. Here in NZ there is not the resources of old scopes in working condition and why would a budding engineering student even want to use one when at Uni they are to use better, much better.


Quote
Those that say something to the effect that "only the latest tool should be considered" are doing budding engineers a disservice.
Those that haven't experienced modern equipment are excused for their ignorance as no longer is a scope just a scope but a LA, FRA, SA, Protocol analyser, Logger, FG and a remote capture instrument and all in a single small box at a single new instrument cost. One investment = most of an analysis lab.

How can even the experienced engineer not be seduced by all this capability or is it they are still living in another century ? Maybe all this technology in a single box blinds them.  :-//
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: True analog scopes
« Reply #223 on: December 19, 2022, 08:23:44 pm »
Those that haven't experienced modern equipment are excused for their ignorance as no longer is a scope just a scope but a LA, FRA, SA, Protocol analyser, Logger, FG and a remote capture instrument and all in a single small box at a single new instrument cost. One investment = most of an analysis lab.

How can even the experienced engineer not be seduced by all this capability or is it they are still living in another century ? Maybe all this technology in a single box blinds them.  :-//

I prefer single purpose standalone instruments in almost every case. I have yet to be impressed with the experience of "Swiss army knife" style test equipment, too often it tries to do everything and ends up doing nothing particularly well.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29809
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: True analog scopes
« Reply #224 on: December 19, 2022, 08:37:52 pm »
Those that haven't experienced modern equipment are excused for their ignorance as no longer is a scope just a scope but a LA, FRA, SA, Protocol analyser, Logger, FG and a remote capture instrument and all in a single small box at a single new instrument cost. One investment = most of an analysis lab.

How can even the experienced engineer not be seduced by all this capability or is it they are still living in another century ? Maybe all this technology in a single box blinds them.  :-//

I prefer single purpose standalone instruments in almost every case. I have yet to be impressed with the experience of "Swiss army knife" style test equipment, too often it tries to do everything and ends up doing nothing particularly well.
Agreed, especially when just 10 years back this multi capability was in its infancy and in those days I too thought a scope was just a scope and even instruments of that time were quite basic compared with what you can get for the same price today.
Particularly good performance implies an instrument was designed for specific tasks however the modern instrument with multiple capability is a compromise in design but still does tasks adequately well.
Pick your poison and let that define your budget.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf