DSO2000X memory extension & bus decoding option

[kripton2035]

I dont want all protocol decoders for free, but come on i2c spi and rs232 are quite common today
dont tell me it's a huge effort in R&D departments to support that !
ok for more complex one like CAN LIN they can sell it apart but not for simple one. they should be with every MSO scope (not with DSO)

[Wuerstchenhund]

Sure, someone could sell a 300MHz scope with every feature and protocol decode under the sub for $500, and maybe even make some money from it.
But it won't be enough profit in order to do R&D for higher end products and stay in business.

That's not quite right. The low-end T&M market is a highly competitive and highly price-sensitive cut-throat market, and the profits are tiny unless you ship in big numbers. And even then it doesn't bring in enough dough to pay for any real R&D, and especially not for high-end products. If you're lucky it pays for product improvements or manufacturing/support cost reduction measures, but that's about it. This is the reason why Agilent has sold Rigol OEM stuff (apparently they now can make their own low-end stuff similarly cheap), and why LeCroy sells Siglent and Tek sells 10+ year old technology.

Manufacturers are not in this market because it rakes in huge profits, they are in because it increases brand awareness for their higher end products. And it's the higher end that offers huge profit margins (despite fewer sold products) and pays for most of the real cutting edge R&D costs.

Therefore no one is going to do it, it would be cutting the throats of the entire industry.

No, they won't do it because the current system of charging extra works extremely well. Why derail the gravy train?

The first company that does it will have a flurry of sales and them promptly go out of business, or just keep putting out mediocre products with no support etc.

If all this company makes is low-end stuff then yes, they perhaps would go bust because they selling the low-end scopes alone hardly makes them any money unless they can manufacture extremely cheap (as Chinese companies can do at the moment if they don't care for quality or working conditions) and sell a lot, and without a higher end that pays for R&D there would not be much money left.

It wouldn't be a a major problem for the big names, though, as they make most of their revenue from their higher end stuff anyways. Of course it would mean to voluntarily give up profits, which is a no-no in business, and therefore won't happen.

Big companies that put serious effort and money into developing new scopes and gear have to make enough profit back to stay in business to keep on doing that. So they do that by offering many price/performance options, and software options.

You seem to think that the price manufacturers charge for options somewhat reflects their own costs (i.e. development costs), but that is not the case. The price is set by what the manufacturer believes the market will pay, plain and simple. Of course there are some options that to develop have required noticeable investments, but these are generally found in high-end products and not in the low-end market (where the major R&D work for most options has already been paid by the high-end many years ago).

The only reasons why manufacturers get away with charging extra for even basic options on entry level scopes is because

• it's a very good way to squeeze additional money out of a market where profits are slim
• they can as their major competitors all do the same
• because they would look stupid if they charge enormous amounts of money for the software options for their high-end products when the cheap low-end variant comes with everything as standard (even when some of the high-end options were actually expensive to develop)

And I agree with kripton2035, these days basic stuff like I2C and RS232 decoding should really be free. But apparently the market bears that these options are charged extra, so it won't change any time soon.

[Baliszoft]

I am happy to see "evolving" the 2k series, though a bit pissed off at the same time. When i bought mine agilent came out with a free wave-gen and dvm license just a week after. Now i saw they are about it give away one free license (of your chioce) for new customers (country dependent). Is it only me who paid the full price for the basic model?! 

[marmad]

Is it only me who paid the full price for the basic model?! 

Welcome to the wonderful land of Earlyadoptionistan. 

Edit: The only advantage of being at the frontier, so to speak, of new products, is that sometimes you can exploit lax or not-yet-strictly-enforced rules - if you know what I mean. 

[Gunb]

I dont want all protocol decoders for free, but come on i2c spi and rs232 are quite common today
dont tell me it's a huge effort in R&D departments to support that !
ok for more complex one like CAN LIN they can sell it apart but not for simple one. they should be with every MSO scope (not with DSO)

Scopes are common for engineers for years - are they for free?

Are you working for free?

Do you like to work for a starvation wage?

Do you like to get support for a poduct?

Nothing's for free, not even your own salary (if you get one). Neither the developer nor the support engineer, not the merchandiser.

It's not a question of R&D efforts, it's a question how a company earns money, stays alive by making profit and is able to offer people a job.
You pay for something, you'll get something - fair play in my eyes. Think about it, it could be your job that could be cut down to reduce costs.

[Gunb]

Sure, someone could sell a 300MHz scope with every feature and protocol decode under the sub for $500, and maybe even make some money from it.
But it won't be enough profit in order to do R&D for higher end products and stay in business.

That's not quite right.

It is absolutely right.

See it all: development, support (contact, spare parts sustainability), merchandising - and: an equitable salary for a well done job.

[kripton2035]

I'm not about the making money or not making money question.
I say that selling an MSO scope without any simple protocol is like selling a car with 3 wheels and saying
"well you can have a 4th wheel but I must sell it to you, but it's ok you can roll with only 3 wheels it will be ok"
for me it's the same
I'm ok they sell a DSO without any protocol analyser, but not a MSO, and UART i2c spi are the very basic protocol you can have with your already expensive MSO.

[Gunb]

I'm not about the making money or not making money question.
I say that selling an MSO scope without any simple protocol is like selling a car with 3 wheels and saying
"well you can have a 4th wheel but I must sell it to you, but it's ok you can roll with only 3 wheels it will be ok"
for me it's the same
I'm ok they sell a DSO without any protocol analyser, but not a MSO, and UART i2c spi are the very basic protocol you can have with your already expensive MSO.

The very basics of an MSO are 8/16 channels for measuring 8/16 channels in parallel, the same as 20 years ago. Analyzing UART, SPI or I2C is a quite different thing that does not even imply 8 channels, so it's OK to pay for that option extra.

My first digital analyzer could measure 16 channels in parallel, and could not be used for anything else. Then manufacturers moved that function into DSOs and named that MSO. Bus analysis is a feature that appeared during the last years, compared to MSOs that still a young feature. So they want to earn money with it, especially where scopes are much more powerful today and cheaper then 20 years ago.

Maybe they will include this feature for free in two/three years, now they want to earn money.

The comparison with the car is wrong. Worked for a few years in automotive and was involved in a few car projects.

Of course the car has got 4 wheels, also the scope is complete, but 15 years ago you had to pay extra money for the air condition, the ABS, the electric windows, and so on. Today in most cars these features are included, but you still have to pay for things like DSG, parking assistant....

With the scopes it absolutely the same: you get a complete device, but for extras as bus decoding you've pay.

Concludingly, the 4 wheels are there, the air condition also, but for navigation system you've to spend more money.


Rgds
Gunb

[kxenos]

IMO as long as the rules of the game are clear it's ok. I knew beforehand that I would have to pay for the protocols. And I can avoid this by buying a separate cheap LA to do this.
On the other hand, I find this strict corporate logic silly, because I believe that there are smarter ways to make more profit. For example think what would happen if Agilent decided to provide a way so that people could write scripts and upload them to their scopes to get whatever protocol analysis they want. If you wanted to buy an MSO wouldn't you choose to buy the one that has a community that develops free software for it? Agilent could blow the market by doing this and until the next vendor did it, it would be on top.
It's like arduino for atmel, the wavegen feature, the rigol hack, mdo of tek etc. And they still could make more money because then at some point you would need memory upgrade for supporting more protocols or they could charge the capability of having e.g more than 5 protocols loaded etc.
Which company provided seatbelts or diskbrakes for the 1st time? Doesn't people think of volvo's as secure cars even after 40 years? Progress is not linear. It comes in steps. You add a feature, you are the market leader for a while and then everyone else copies you but you're always the first who did it. And again from the start. Add free right mirror, add free second reverse gear light, add 1 airbag, add 2 airbags, add 8 airbags etc. 

[EEVblog]

You seem to think that the price manufacturers charge for options somewhat reflects their own costs (i.e. development costs), but that is not the case.

No, I have never said that.
I have said exactly what you have said, it's simply a way for them to make the high profit margins required to keep themselves in business and making a sustained profit (the PRIMARY responsibility of a publically listed company)
It's impossible to account for real R&D costs like that within a large company and fairly spread it across the options, that's not ho the game works, ever.

The point here is that the money they make on the options allows them to invest in more R&D which means innovation in the industry, and eventually new technology filtering down to the lower end chain, like Agilent did originally with the 2000 and 3000 series, and what Rigol are doing now with the 2000 series for example.
You can argue all you want over what things should be included for free, that's beside the point.

The price is set by what the manufacturer believes the market will pay, plain and simple.

Yes, of course, that's big business 101.

[Hydrawerk]

https://www.home.agilent.com/agilent/editorial.jspx?cc=US&lc=eng&ckey=2308179&nid=-33028.0.00&id=2308179

Offer good only for residents of the the US and Canada.

Oh, why??

[EEVblog]

IMO as long as the rules of the game are clear it's ok. I knew beforehand that I would have to pay for the protocols. And I can avoid this by buying a separate cheap LA to do this.
On the other hand, I find this strict corporate logic silly, because I believe that there are smarter ways to make more profit. For example think what would happen if Agilent decided to provide a way so that people could write scripts and upload them to their scopes to get whatever protocol analysis they want.

That won't happen because that suddenly saddles the company with the support of all the shitty protocols that people write, and their customers download for free, and then they cop all the flack when the customers product fail. Think that won't happen? Think again. It is the #1 reason why a professional test equipment maker like Agilent will never do this, ever.

Fluke are the same for example. I asked them why they won't relase the protocol for their bluetooth wireless modules, and their response was exactly that. It's not for trade secret reasons, or sales, or whatever, it's to protect their reputation of providing safe and realiable test equipment. They will not trade that for some extra sales and kudus because they release they made their protocol open.

[marmad]

IMO as long as the rules of the game are clear it's ok. I knew beforehand that I would have to pay for the protocols. And I can avoid this by buying a separate cheap LA to do this.
On the other hand, I find this strict corporate logic silly, because I believe that there are smarter ways to make more profit. For example think what would happen if Agilent decided to provide a way so that people could write scripts and upload them to their scopes to get whatever protocol analysis they want. If you wanted to buy an MSO wouldn't you choose to buy the one that has a community that develops free software for it? Agilent could blow the market by doing this and until the next vendor did it, it would be on top.
It's like arduino for atmel, the wavegen feature, the rigol hack, mdo of tek etc. And they still could make more money because then at some point you would need memory upgrade for supporting more protocols or they could charge the capability of having e.g more than 5 protocols loaded etc.
Which company provided seatbelts or diskbrakes for the 1st time? Doesn't people think of volvo's as secure cars even after 40 years? Progress is not linear. It comes in steps. You add a feature, you are the market leader for a while and then everyone else copies you but you're always the first who did it. And again from the start. Add free right mirror, add free second reverse gear light, add 1 airbag, add 2 airbags, add 8 airbags etc. 

But you hit upon (at least part) of the problem early in your post - it's a MUCH different world than when Volvo introduced the seat belt- which, BTW, was ~65 years ago. We are in the throes of corporate capitalism - and the idea of slow, incremental change towards a long-term goal is rather an anachronism now.

I agree with you, Gunb, about equitable wages, sustainability, etc. but OTOH, it's no coincidence that the newly introduced protocols for the X2000 are split up and priced almost exactly the same - except a tiny bit cheaper - than those Rigol ones.

Let's face it - there are people smarter than us about these things - who work for corporations whos job it is to figure out exactly what and how to sustain profits, as Dave put it - because nowadays, if you're a corporation, the ability to meet or exceed Wall Street expectations is the bottom line.

[kxenos]

Dave has a very good point there. Reputation plays quite important role in this business on any other business for that matter. It's like android and the shitty apps that people blame the phone vendors for their phones being "slow". Plus Agilent's and Fluke's main business is not the hacker-maker market anyway - yet.

[afho]

Is there a chance that Agilent will make USB decode to the 3000X series!? The 4000X series has this option (as we can see in Daves review) an is based on the same hardware.

Please Agilent, give us USB decoding on the 3000X series ...

[Gunb]

Just contact Agilent and ask their support.

[mikeselectricstuff]

That won't happen because that suddenly saddles the company with the support of all the shitty protocols that people write, and their customers download for free, and then they cop all the flack when the customers product fail. Think that won't happen? Think again. It is the #1 reason why a professional test equipment maker like Agilent will never do this, ever.

Actually there is a precedent for this - back in the days of the 1650/51 logic analysers, HP sold a software package that allowed users to write their own disassembler  (Or inverse assembler as they called it) to run on the logic analyser.
Of course that was pre-internet so chances are anything a customer developed would probably not have been distributed much.
 
Protocol decode (efficiently) is so tightly bound to the hardware that it's hard to see how you could implement a user-definable system.
And you do always have the option of post-capture analysis by downloading the captured traces over ethernet - not quite as convenient, but for the sort of jobs where you really need protocol decode, probably not a huge inconvenience.

[nctnico]

That won't happen because that suddenly saddles the company with the support of all the shitty protocols that people write, and their customers download for free, and then they cop all the flack when the customers product fail. Think that won't happen? Think again. It is the #1 reason why a professional test equipment maker like Agilent will never do this, ever.

Actually there is a precedent for this - back in the days of the 1650/51 logic analysers, HP sold a software package that allowed users to write their own disassembler  (Or inverse assembler as they called it) to run on the logic analyser.
Of course that was pre-internet so chances are anything a customer developed would probably not have been distributed much.
 
Protocol decode (efficiently) is so tightly bound to the hardware that it's hard to see how you could implement a user-definable system.

AFAIK the older Tektronix logic analysers (TLA700 series) had user definable protocol decoders but that worked so bad they removed it in later software versions. However some people did some reverse engineering on the assembler support packages which allowed me to write my own I2C and SPI decoders for the TLA700 software in C++.

[EEVblog]

And I forgot to mention that the protocol decoding in the Agilent is done in hardware, inside the ASIC, as well as functionality in the software.
Even if it was an FPGA, that's not the same ball-game when it comes to letting users write their own or play around with it.

Dave.

[Gunb]

Honestly I'm glad that not everything is open source.

Implementing firmware on embedded systems, also for embedded linux. Don't want to see hobby programmers writing firmware for professional measurement equipment and would like to see support from the manufacturers. By the way, especially manufacturers with self-made ASICs would have to reveal there intellectual properties - wouldn't do that either.

Manufacturers would be blamed for buggy firmware, not good for them.

On the other hand: there are cheap China scopes with open firmware - where are the programmers implementing the first bus analysis? Or are they still available? Instead of implementig Ping Pong for these scopes (which I had for my ZX81 ) it would make more sense to implement such useful applications. So, I would like to see first that people are able do implement decoders on these scope before asking for open source on Agilents, Teks,....


Kind rgds
Gunb

[hammil]

hobby programmers writing firmware for professional measurement equipment and would like to see support from the manufacturers ... Manufacturers would be blamed for buggy firmware

I.. don't think that happens... And even if people are silly enough to expect manufacturer support, what of it? Not like they're going to get it. It's about choice, I guess... manufacturer support, or hacking/customisation. And only people who have no idea what they're doing would blame a manufacturer for problems in third party firmware.

On the other hand: there are cheap China scopes with open firmware - where are the programmers implementing the first bus analysis? Or are they still available? Instead of implementig Ping Pong for these scopes (which I had for my ZX81 ) it would make more sense to implement such useful applications.

The DS1052 isn't open source by any means. In fact it's pretty closed... there's a single window for modification (the firmware upgrades) and a single way of working out how to program the thing (the serial console). Rigol provides absolutely no documentation or support regarding firmware modification. It's quite incredible that people have managed to change anything at all - in fact this whole idea has only just come to fruition; barely in the experimental stage, let alone ready for user distribution. So yeah, at first, people are going to implement Pong and little messages on screen. Eventually, when everything is better documented, we'll see some useful modifications come about.

[EEVblog]

I.. don't think that happens... And even if people are silly enough to expect manufacturer support, what of it? Not like they're going to get it. It's about choice, I guess... manufacturer support, or hacking/customisation. And only people who have no idea what they're doing would blame a manufacturer for problems in third party firmware.

But with the communications revolution came the ability for a single disgruntled person to sew the seed to publicly damaging the reputation of a company, rightly or wrongly. Companies have to be more careful these days with stuff like this.

[marmad]

But with the communications revolution came the ability for a single disgruntled person to sew the seed to publicly damaging the reputation of a company, rightly or wrongly. Companies have to be more careful these days with stuff like this.

Not that I care about spelling errors, Dave, but the image of someone sewing seeds give me a smile. 

[Gunb]

I.. don't think that happens... And even if people are silly enough to expect manufacturer support, what of it? Not like they're going to get it. It's about choice, I guess... manufacturer support, or hacking/customisation.

Don't know which manufacturer you mean, but I've always got support from manufacturers (Hameg, Rigol, Agilent) when I needed their advice or I've found bugs.
Anyway, it's the good right that companies protect their intellectual property and to decide themselves how to deal with there code.

And only people who have no idea what they're doing would blame a manufacturer for problems in third party firmware.

That's your opinion, man 

The DS1052 isn't open source by any means. In fact it's pretty closed... there's a single window for modification (the firmware upgrades) and a single way of working out how to program the thing (the serial console). Rigol provides absolutely no documentation or support regarding firmware modification. It's quite incredible that people have managed to change anything at all - in fact this whole idea has only just come to fruition; barely in the experimental stage, let alone ready for user distribution. So yeah, at first, people are going to implement Pong and little messages on screen. Eventually, when everything is better documented, we'll see some useful modifications come about.

Didn't mean the DS1052.

I would recommend to develop an open source scope from scratch, then we'll see how much work it really means to reach the manufacturers level and if it is really only a money-making option to implement proper bus decoding. There are powerful FPGAs for less money today, also raspberry pi offers hardware accelerated graphics.

If I would develop a scope, I would rather want to earn money with it than to give it away for free - my opinion.


However, have a nice day 

[hammil]

Don't know which manufacturer you mean, but I've always got support from manufacturers (Hameg, Rigol, Agilent) when I needed their advice or I've found bugs.

I was referring specifically to your suggestion that "hobby programmers writing firmware for professional measurement equipment and would like to see support from the manufacturers". However if it's the manufacturer's own firmware, then they should of course offer support.

Didn't mean the DS1052.

Oh, alright. It just looked that way when you mentioned Pong - since someone very recently got Pong to work on that scope (you should check it out - it looks awesome. I'm going to see if I can contribute to that project)

If I would develop a scope, I would rather want to earn money with it than to give it away for free - my opinion.

Sure. That's why a lot of people do... Regarding open-source hardware and software, there's a whole bunch of different factors. Ubuntu, for example, is developed by a commercial entity and given away for free, because they sell support services for it, and make a lot of money that way.

I could, for example, develop some hardware, release the schematics with a noncommercial licence (CC BY-NC-SA for example) so that people could either build their own, or buy a premade unit which I would profit from. Even if someone copied it, they wouldn't be able to sell units for much less than I could, and it wouldn't be legal either. This doesn't work if you have huge R+D costs though...  So.. -shrug- It really depends. I understand why not everything is open-source, but I strongly support the movement and applaud those who apply it to their products.

Regarding hacking/tinkering - it's my belief that if you buy a product, it is yours to do with as you wish. If companies want to cripple their devices so they can charge the consumer a fortune for 'upgrades' in order to offset their R+D cost, fine by me. But they shouldn't be surprised or offended if people find a way to do it themselves. The reason bus decoding costs so much as an upgrade isn't because it's a lot of work, because it really isn't. It's because it's so useful. They already have everything developed, so they're charging as much as people are willing to pay. It's all just marketing.

However, have a nice day 

You too, mate