what an oscilloscope recommended for a woman passionate about electronics?

[FearTec]

Good luck and hope you choose a good one

Any chance you live close to https://www.youtube.com/c/AndreasSpiess/videos

[jonslab]

I would say go with the Rigol ds1202z-e

[tautech]

I don't understand why the producers play the hackers game; let's face it, if they wanted the Siglent engineers would not allow a lower model to be upgraded to the higher model with a hack; this policy makes no sense, it could be unfair to those who buy the superior model without the help of hackers ... 

The hobbyists don't care as they generally don't need the traceable calibration that accompanies these scopes on their Cal sheets whereas anyone in business designing/building a product for the consumer market need make damn certain their product is safe with instruments certified to a spec.
Some even argue if the SW is in a instrument that allows to perform better than its labeled model they have paid for the HW and SW so it's rightfully theirs to hack it or not.
Trouble is, there's a big community out there that lives for hacking gear and of any sort and even some businesses focus their trade on hacking, a well known one is those that tweak engine ECU's to get another level of performance that the manufacturer has deliberately detuned and sell for a lesser price.

We have a least a couple of real sharp members on EEVblog that have cracked most equipment discussed on this forum. Some have even written the online scripts to help with hacks......yes I watch for this sort of stuff. 

[jdutky]

I don't understand why the producers play the hackers game; let's face it, if they wanted the Siglent engineers would not allow a lower model to be upgraded to the higher model with a hack; this policy makes no sense, it could be unfair to those who buy the superior model without the help of hackers ... 

I assume it's a matter of cost on the low-end scopes. The cost savings of using all the same parts in two scopes with different rated bandwidths outweighs whatever they lose to people buying the lower spec model and hacking it. With the low end scopes you can bet that the profit margins are thin to begin with, so there's more pressure to cut costs, and only buying one bill of materials (BOM) for three models easily beats out three different BOMs at 1/3 or 1/4 the volume.

Just my guess.

Also, what tautech said about professionals needing taceable cal docs and being willing to pay makes a lot of sense. The pro-level scopes start at twice the price of the hobby scopes, and just go up from there, and anyone doing this as a business will be willing to pay those premiums because the high end scopes will pay for themselves in no time. Even the multi-hundred thousand dollar scopes will pay for themselves pretty quickly, because you've probably got a client that will pay, and you're putting it in front of an engineer whose annual salary is a fair fraction of the cost of such a scope (so making their job easier and faster saves you lots of money).

[jdutky]

I would say go with the Rigol ds1202z-e

I was leaning toward a Rigol for a while, but the SuperPhosphor feature of the Siglents sold me. All my experience is with "old" analog scopes, and the SuperPhosphor feature reproduces an important aspect of the analog scope experience. The reviews I've watched of the Rigols seemed to show that they didn't do the variable intensity thing very well, if at all.

Also, as I said, if I'm going to be splashing out money I see no reason not to get 4 channels.

The 1054z was the first digital scope that looked both capable and affordable to me, and was what I thought I was going upgrade to for quite a while. I'd be interested to hear your reasons for preferring the Rigol over the Siglent.

[CharlotteSwiss]

I don't understand why the producers play the hackers game; let's face it, if they wanted the Siglent engineers would not allow a lower model to be upgraded to the higher model with a hack; this policy makes no sense, it could be unfair to those who buy the superior model without the help of hackers ... 

The hobbyists don't care as they generally don't need the traceable calibration that accompanies these scopes on their Cal sheets whereas anyone in business designing/building a product for the consumer market need make damn certain their product is safe with instruments certified to a spec.
Some even argue if the SW is in a instrument that allows to perform better than its labeled model they have paid for the HW and SW so it's rightfully theirs to hack it or not.
Trouble is, there's a big community out there that lives for hacking gear and of any sort and even some businesses focus their trade on hacking, a well known one is those that tweak engine ECU's to get another level of performance that the manufacturer has deliberately detuned and sell for a lesser price.

We have a least a couple of real sharp members on EEVblog that have cracked most equipment discussed on this forum. Some have even written the online scripts to help with hacks......yes I watch for this sort of stuff. 

I keep thinking that in Siglent or Rigol etc, it is good that some models are hacked to the superior model, this is because many people are attracted to a model that you can make superior, and therefore they spend money on that manufacturer; let's take an example on the passband, for the engineers it would not be a problem to limit the instrument to that band, but orders from above prevent it ..
Hackers brag a lot, but they can only play because other people let them play.
 

[CharlotteSwiss]

I would say go with the Rigol ds1202z-e

as I wrote at the beginning of the discussion, this was also one of my possible choices; i chose siglent, now i don't remember based on what difference between the two.

I'm starting to suspect that Siglent and Rigol play calling oscilloscopes with roughly the same acronyms ...
 

[CharlotteSwiss]

I don't understand why the producers play the hackers game; let's face it, if they wanted the Siglent engineers would not allow a lower model to be upgraded to the higher model with a hack; this policy makes no sense, it could be unfair to those who buy the superior model without the help of hackers ... 

I assume it's a matter of cost on the low-end scopes. The cost savings of using all the same parts in two scopes with different rated bandwidths outweighs whatever they lose to people buying the lower spec model and hacking it. With the low end scopes you can bet that the profit margins are thin to begin with, so there's more pressure to cut costs, and only buying one bill of materials (BOM) for three models easily beats out three different BOMs at 1/3 or 1/4 the volume.

Just my guess.

Also, what tautech said about professionals needing taceable cal docs and being willing to pay makes a lot of sense. The pro-level scopes start at twice the price of the hobby scopes, and just go up from there, and anyone doing this as a business will be willing to pay those premiums because the high end scopes will pay for themselves in no time. Even the multi-hundred thousand dollar scopes will pay for themselves pretty quickly, because you've probably got a client that will pay, and you're putting it in front of an engineer whose annual salary is a fair fraction of the cost of such a scope (so making their job easier and faster saves you lots of money).

I agree with what I wrote, it reflects the truth; and allows hackers to stay in the evening... 

[CharlotteSwiss]

I would say go with the Rigol ds1202z-e

I was leaning toward a Rigol for a while, but the SuperPhosphor feature of the Siglents sold me. All my experience is with "old" analog scopes, and the SuperPhosphor feature reproduces an important aspect of the analog scope experience. The reviews I've watched of the Rigols seemed to show that they didn't do the variable intensity thing very well, if at all.

Also, as I said, if I'm going to be splashing out money I see no reason not to get 4 channels.

The 1054z was the first digital scope that looked both capable and affordable to me, and was what I thought I was going upgrade to for quite a while. I'd be interested to hear your reasons for preferring the Rigol over the Siglent.

now you reminded me of the considerations that I had made between the two 1202 Siglent and Rigol: rigol greater memory depth, but siglent better FFT for sure: I preferred Siglent, but Rigol is still on par I think ..
 

[borjam]

Hackers brag a lot, but they can only play because other people let them play.
 

Yes and no, it depends.

Security is hard. Very. So hacks are due to incompetent design.

That said, better security means higher development costs. What is the incentive to recover those costs?

You can divide customers roughly in two groups. I know there is a lot of shades of grey between them of course but it's irrelevant.

First, professional customers who know the specs they need (I need this bandwidth and I need to decode I2C and SPI). Many professional customers need guaranteed measurements as an error might even lead to damages.

Second, hobbyists. A hobbyist will try to get the maximum bang for the buck given their budget. Most hobbyists won't necessarily need a bandwidth of 200 MHz, maybe 50 MHz is enough. But a hobbyist is not going to buy an instrument for a certain project, as many professional customers may do. You don't know what you will be doing next year with the scope.

So, now imagine you are Siglent or Rigol. You must cover the development costs, of course manufacturing, and get some earnings.

Does the business work with professional customers paying full prices for upgrades and features while most hobbyists only buy the minimum package? Although it might look like you are losing money with hobbyists hacking their scopes, you are probably not. And those units you sell to hobbyists at the very least will help with the economies of scale. Some will even provide valuable feedback.

It's like software manufacturers offering student licenses for free or at a nominal cost. It's like Wolfram Research bundling a free Mathematica package with Raspberry Pis.

I don't think they care a lot about hobbyists hacking options. And of course they are not stupid and I am sure they read this forum. And I am sure that is the reason why they didn't bother to release several versions of the X-E scopes with different options. Let's sell it "factory hacked" and charge more to customers who need more powerful models.

[CharlotteSwiss]

beautiful analysis borjam, I agree in everything; But let's go back to the starting point: the instrument is hacked because the manufacturer allows it: I am thinking of the rigol 1154, how many users have bought it for the mere fact that you could have 100mhz on each channel simply by changing firmware? On the top ladder we have the CEO of the company, ordering his engineers to leave spaces for the hack; then below we have the hackers who unwittingly participate in the game, but play both for the company and for the hobbyist who will then change firmware.

If we go back to the days of the first satellite paytvs, it was enough to program a pic and you could see everything; then came the king of pays .. and the game ended almost immediately, and in any case it was no longer so simple: I don't want to decrease a Hacker, but as I said, he plays where he is allowed ...

[CharlotteSwiss]

borjam, after finishing the manual, I was reviewing all the menu items to see if I had described them all; Acquire menu- Page 2 / 2- Acq Mode, the choice is fast default and slow ..

I see you asked:
https://www.eevblog.com/forum/testgear/siglent-sds1202x-e-odd-behavior-and-questions/

Did you then understand what could be used to set slow? In the 1104 review I found: Fast is standard, use Slow only if for some reason you do not want to see multiple traces after a signal drop-out in normal trigger mode. Be aware that this will significantly
decrease waveform update rate.

But it's not clear to me ...
Do you know anything about it? thank you
 

[jdutky]

I keep thinking that in Siglent or Rigol etc, it is good that some models are hacked to the superior model, this is because many people are attracted to a model that you can make superior, and therefore they spend money on that manufacturer; let's take an example on the passband, for the engineers it would not be a problem to limit the instrument to that band, but orders from above prevent it ..
Hackers brag a lot, but they can only play because other people let them play.
 

Yes and no. The engineers certainly COULD design the different models to be physically limited and more difficult to hack, but that would come at increased cost of production, and the low end models are the least able to absorb any extra costs. Making four different models that are really identical hardware and software saves A TON of money in both design and manufacturing, and if that's enough saving to sell a 200MHz scope at a 50MHz price, then all the better. The company can still sell the scopes at the higher price, but what the buyer is paying for is the promise of performance, not the difference in hardware.

But even IF the engineers made some minor change to the hardware to limit the performance of the lower end models, for example by not including some components that would be required to meet the higher spec, and even IF that didn't increase the manufacturing cost, really enterprising customers could still go in a add the missing components, and we're back to square one. There's a point of diminishing returns, and it happens much faster with low-priced consumer equipment than with the premium priced professional gear.

But Rigol and Siglent DO make some of the differences un-hackable, where it makes sense: while the fundamental hardware of the 2 and 4 channel scopes may be basically identical, they go to the trouble of not populating the two of the channels in the 2 channel scope, because each pair of channels is a big part of the total cost of the scope. The ADC alone is probably $100 or more (I haven't checked on the price of good multi-GSps ADCs lately, but I'm sure that they're not cheap). Now Rigol and Siglent get some break on the price because they buy in quantity, but it's still a substantial fraction of the total cost, so it makes sense to do separate manufacturing for 2 and 4 channel scopes.

This is standard practice all over the industry. IBM is famous for this: At least since the late 60s they have shipped basically the same mainframes out to all their customers regardless of what specs the customer bought. They even have a clever marketing name for it, but basically, since all the mainframes come with the maximum hardware (memory and CPU) installed, and it's just limited by software, customers can just call their sales contact and have memory or CPUs added to the machine instantly. They can even rent extra memory and CPU power as needed. Again, it saves IBM a ton on manufacturing, and they still manage to extract the money for the higher spec machines from most, if not all, of their clients.

(sorry, that was a bit long winded, but I hope you get the idea)

[CharlotteSwiss]

(sorry, that was a bit long winded, but I hope you get the idea)

you were very clear, I had never thought about this aspect of industrial production in fact it is a reasoning that does not make a turn; after all I imagine that at the beginning of a new series of models there is a lot of study time on the possible earnings: you work for that.
thanks 

[jdutky]

(sorry, that was a bit long winded, but I hope you get the idea)

you were very clear, I had never thought about this aspect of industrial production in fact it is a reasoning that does not make a turn; after all I imagine that at the beginning of a new series of models there is a lot of study time on the possible earnings: you work for that.
thanks 

I've wondered about this too, though. My main experience with oscilloscopes is a couple decades out of date, and at the time there was basically no such thing as a low-cost scope. Companies like Tektronix built devices on or beyond the bleeding edge, and poured ever last bit of technical know-how into the process. This was great, if you were in the market for top-of-the-line test and measurement equipment, and you had the resources of a large company behind you, but it also meant that many of the parts in that equipment may have been custom made for or by the manufacturer, which meant it would be tough to repair, and it cost a lot from the outset.

We aren't in that world (for the most part) any more, now even comparatively high spec scopes are being built with COTS parts (but a fair amount of institutional design and manufacturing skill). I was shocked a couple years ago when I found out that I could get a 200 MHz scope for under $10,000. Now one can be had for under $1000! Oh! What brave new world...

[CharlotteSwiss]

200mhz as the 1202 also costs less than $ 500 it seems to me 

[borjam]

Yes and no. The engineers certainly COULD design the different models to be physically limited and more difficult to hack, but that would come at increased cost of production, and the low end models are the least able to absorb any extra costs. Making four different models that are really identical hardware and software saves A TON of money in both design and manufacturing, and if that's enough saving to sell a 200MHz scope at a 50MHz price, then all the better. The company can still sell the scopes at the higher price, but what the buyer is paying for is the promise of performance, not the difference in hardware.

The plot thickens, even.

Imagine something challenging to manufacture such as a complex IC (a microprocessor). Processes are not perfect and you get a variable yeld (percentage of valid units). Also, maybe not all of the valid units offer maximum performance due to process variability. But if you sell the same component in several grades you can test which ones are the best and assign them to the higher class, selling the rest as the basic ones.

This happened years ago with Intel processors, for example. They sold several versions of the same processor, valid for different clock speeds. Let's simplify with two: 75 MHz and 100 MHz.

As far as I know both came from the same wafers. They tested the chips in order to select which ones were 100 MHz capable. But it got a bit more complicated If they had produced, say, 100,000 processors and they had an order for 10,000 100 MHz units they would only test 10,000 of them for 100 MHz performance, not bothering about the rest. The 100 MHz test added to the total manufacturing costs and it was pointless to test units you were going to mark as 75 MHz for 100 MHz compatibility.

And it got yet more complicated With time manufacturing process can improve. So, if at the beginning of the manufacturing period you get, say, a 70% yield with 10% capable of the highest performance, maybe in a year you improve the yield to 80% with, say, 50% or more of those 100 MHz capable. But if you are going to sell only 20% of those as the 100 MHz version, turns out many of the low end version units are actually 100 MHz capable. Overclocking was born

With these scopes the same could happen. These units at 200 MHz are not that challenging to manufacture. But for more complex equipment I wouldn't be surprised if hacked a low end version didn't work perfectly as a higher class model.

So, again, a professional customer who requires dependable measurements will need to pay for the high end model instead of hacking.

[borjam]

I see you asked:
https://www.eevblog.com/forum/testgear/siglent-sds1202x-e-odd-behavior-and-questions/

Did you then understand what could be used to set slow? In the 1104 review I found: Fast is standard, use Slow only if for some reason you do not want to see multiple traces after a signal drop-out in normal trigger mode. Be aware that this will significantly
decrease waveform update rate.

But it's not clear to me ...
Do you know anything about it? thank you
 

I am not an expert on oscilloscope design, surely there are much more knowledgeable people here. So, take my answer with a grain of salt.

How many frames per second can your eyes perceive? Certainly not the thousands of waveforms per second it can capture. So, in order to present useful information the scope does some processing to somewhat summarize information on lots of waveforms into a single frame. That is the fast mode.

But sometimes you might prefer less processing at the cost of losing information. There is always the risk of extra processing somewhat inventing signal characteristics, like the same dots/lines dilemma. So you can disable fast mode. It would be especially useful for captures in which you are hunting for a strange signal after a trigger and you don't want it "mixed" with more waveform captures that don't show the event you are searching for.

Siglent has done something good here (as they have done with the "dots" mode) allowing the user to disable some post processing. Rigol, in comparison, is more opaque in that regard on the DS1000Z (no idea about other series).

[jdutky]

The plot thickens, even.

I wanted to talk about this, but I thought my post was already more than long enough. I expect that there is some of that going on with the Rigol and Siglent scopes as well, but I think that the bigger factor at the low end is just down to volume and reducing manufacturing costs (which, I suppose, it what it really came down to for Intel in the example you gave, but let's not get distracted). I suspect, but have not got the evidence to prove, that the ability to hack the low end units to higher performance actually increases sales of the entire model line: it's a feature for a segment of the customer base who would not, or could not, pay for the certified 200MHz scopes, and it entices them to buy one of the lower spec units, which increases the volume of production, which decreases the total cost of manufacture, and everyone is happy!

Long story short, I don't think that any of Siglent or Rigol executives or shareholders are crying about people hacking the 100MHz models scopes to operate at 200Mhz. In fact I'll bet that they just roll around in piles of money and laugh and laugh, and I respect that.

[CharlotteSwiss]

Imagine something challenging to manufacture such as a complex IC (a microprocessor). Processes are not perfect and you get a variable yeld (percentage of valid units). Also, maybe not all of the valid units offer maximum performance due to process variability. But if you sell the same component in several grades you can test which ones are the best and assign them to the higher class, selling the rest as the basic ones.

This happened years ago with Intel processors, for example. They sold several versions of the same processor, valid for different clock speeds. Let's simplify with two: 75 MHz and 100 MHz.

As far as I know both came from the same wafers. They tested the chips in order to select which ones were 100 MHz capable. But it got a bit more complicated If they had produced, say, 100,000 processors and they had an order for 10,000 100 MHz units they would only test 10,000 of them for 100 MHz performance, not bothering about the rest. The 100 MHz test added to the total manufacturing costs and it was pointless to test units you were going to mark as 75 MHz for 100 MHz compatibility.

And it got yet more complicated With time manufacturing process can improve. So, if at the beginning of the manufacturing period you get, say, a 70% yield with 10% capable of the highest performance, maybe in a year you improve the yield to 80% with, say, 50% or more of those 100 MHz capable. But if you are going to sell only 20% of those as the 100 MHz version, turns out many of the low end version units are actually 100 MHz capable. Overclocking was born

interesting news on intel processors; it is obvious that the production in large numbers is obtained through large calculations of economic return for the company; many engineers will not be happy with their imposed work, but if the company decides so they must execute the order

[CharlotteSwiss]

The plot thickens, even.

I wanted to talk about this, but I thought my post was already more than long enough. I expect that there is some of that going on with the Rigol and Siglent scopes as well, but I think that the bigger factor at the low end is just down to volume and reducing manufacturing costs (which, I suppose, it what it really came down to for Intel in the example you gave, but let's not get distracted). I suspect, but have not got the evidence to prove, that the ability to hack the low end units to higher performance actually increases sales of the entire model line: it's a feature for a segment of the customer base who would not, or could not, pay for the certified 200MHz scopes, and it entices them to buy one of the lower spec units, which increases the volume of production, which decreases the total cost of manufacture, and everyone is happy!

Long story short, I don't think that any of Siglent or Rigol executives or shareholders are crying about people hacking the 100MHz models scopes to operate at 200Mhz. In fact I'll bet that they just roll around in piles of money and laugh and laugh, and I respect that.

I'm pretty sure, before the model hits the market, Siglent executives already have the sales numbers on the table:
-mod.xxxx with possible hack will sell tot units
The party of smiles will therefore be attended by executives, hackers and end customers of low cost products

[CharlotteSwiss]

I see you asked:
https://www.eevblog.com/forum/testgear/siglent-sds1202x-e-odd-behavior-and-questions/

Did you then understand what could be used to set slow? In the 1104 review I found: Fast is standard, use Slow only if for some reason you do not want to see multiple traces after a signal drop-out in normal trigger mode. Be aware that this will significantly
decrease waveform update rate.

But it's not clear to me ...
Do you know anything about it? thank you
 

I am not an expert on oscilloscope design, surely there are much more knowledgeable people here. So, take my answer with a grain of salt.

How many frames per second can your eyes perceive? Certainly not the thousands of waveforms per second it can capture. So, in order to present useful information the scope does some processing to somewhat summarize information on lots of waveforms into a single frame. That is the fast mode.

But sometimes you might prefer less processing at the cost of losing information. There is always the risk of extra processing somewhat inventing signal characteristics, like the same dots/lines dilemma. So you can disable fast mode. It would be especially useful for captures in which you are hunting for a strange signal after a trigger and you don't want it "mixed" with more waveform captures that don't show the event you are searching for.

Siglent has done something good here (as they have done with the "dots" mode) allowing the user to disable some post processing. Rigol, in comparison, is more opaque in that regard on the DS1000Z (no idea about other series).

So in simple terms, with fast (default) to create a single frame to show to the display, the oscilloscope captures many frames and merges them into one; if we set Slow instead, will the oscilloscope acquire a single frame to compose the single frame to show on the display? (in practice it does not average the calculations to be proposed as with fast?)

It is not a problem if we are not experts in some topic, the speech is that the Siglent manual (but in the google searches) give us results on certain settings, so I treat them in this discussion 

it may also be that many settings were not present with the first firmware, therefore absent in the manual; if I were CEO Siglent, I would order from my engineers: if you make an addition to the menus with a new firmware, together with the firmware download also attach the new pdf manual where you will have added the new function ...

[borjam]

I'm pretty sure, before the model hits the market, Siglent executives already have the sales numbers on the table:
-mod.xxxx with possible hack will sell tot units
The party of smiles will therefore be attended by executives, hackers and end customers of low cost products

Don´t get me wrong, however. Some people tend to justify software piracy by claiming that it is always beneficial to the manufacturer.

It may be beneficial to some, but not to all of them. And I've seen briliant software companies go down because everyone pirated their products.

[borjam]

So in simple terms, with fast (default) to create a single frame to show to the display, the oscilloscope captures many frames and merges them into one; if we set Slow instead, will the oscilloscope acquire a single frame to compose the single frame to show on the display? (in practice it does not average the calculations to be proposed as with fast?)

I am not sure how many waveforms, I haven't checked. But yes, maybe one. It makes sense after all.

it may also be that many settings were not present with the first firmware, therefore absent in the manual; if I were CEO Siglent, I would order from my engineers: if you make an addition to the menus with a new firmware, together with the firmware download also attach the new pdf manual where you will have added the new function ...

Writing proper documentation is extremely expensive. Moreover, for a newcomer to the market it is much harder. Have a look at the outstanding documents you can download from the traditional instrument manufacturers. But remember that many of those documents have been refined for 40 years or longer!

That is actually the worst limitation for new Chinese manufacturers. It is hard to hire good technical writers. And it takes a lot of work to distill the knowledge of your engineers into a usable document. Moreover, their engineers are themselves relative newcomers to the field while the old brands employ old people who have seen a lot!

So many times in my life I have wondered "Have the design team actually used one of these (whatever, oscilloscope, etc) before designing one?" And it's not infrequent to think that when using a Siglent or Rigol product. Of course they have improved a lot if only by reading curses on the forum!

[CharlotteSwiss]

Don´t get me wrong, however. Some people tend to justify software piracy by claiming that it is always beneficial to the manufacturer.

It may be beneficial to some, but not to all of them. And I've seen briliant software companies go down because everyone pirated their products.

obvious that here we were talking about hackers, who still bring the same money to producers, you have to buy an oscilloscope anyway; software piracy is a problem