EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Etne on May 12, 2019, 07:06:27 am
-
Hello,
I'm starting electronic after programming since I'm 7 years old. I'm currently doing aerospace engineering studies with very little electronic courses so I decided to learn it by myself. I'm mostly interested in the embedded world and diy modular synth. I'm looking for an oscilloscope (and I know I'm not the first one to ask you this) for more or less 1000€. I'll buy it in June.
I want to be able to use my oscilloscope in my current projects of course but also the future ones. It has to be polyvalent, reliable and durable. I think I need 4 channels, serial/protocol decoding. I used in an association the Siglent SDS1202X-E which I really liked.
So far, I have found these scopes :
- Rigol MS05000 (https://www.batronix.com/shop/oscilloscopes/Rigol-MSO5072.html#yoReviews) which once hacked must be a very powerful scope for the price. But the software looks terrible and very slow from the teardown video. Does it get better after software upgrade?
- Siglent SDS1204X-E (https://www.batronix.com/shop/oscilloscopes/Siglent-SDS1204X-E.html) easy choice but it does everything I need I think. But maybe for my budget I can get better?
- Keysight DSOX1204A (https://fr.farnell.com/keysight-technologies/dsox1204a/dso-4-canaux-70mhz-1gsps-5ns/dp/2984794). I really like oscilloscopes from Keysight , they feel reliable and look really great. Is it easy to hack?
Thanks a lot for all of your advices.
-
SDS1204X-E vs SDS1202X-E, the 4ch model is more than 2x the scope.
Sure it has 2x 1GSa/s ADC's and 2x 14 MPts memory so looks like 2x the scope but it is somewhat more than that.
So we can use 2 channels and still maintain 1 GSa/s sampling and 14 Mpts memory.
Navigate and Search features plus options of 16ch MSO, AWG and WiFi connectivity, none of which are available in the SDS1202X-E.
Otherwise they are very very similar to use.
You can see how they perform at their rated frequencies and beyond in this thread:
https://www.eevblog.com/forum/testgear/dso-bandwidth-test-sds1104x-e-dsox1102g-to1104-gds1054b/ (https://www.eevblog.com/forum/testgear/dso-bandwidth-test-sds1104x-e-dsox1102g-to1104-gds1054b/)
-
I will get the -G version, the 4 channel, 70MHz version is great. AWG can't be hacked, unless you are willing to solder tons of parts.
BW can be hacked later if someone discovers a good method, or you are willing to ruin the warranty.
You can also legally enable 200MHz BW at additional cost in the future, so you are not facing a $2000 bill at once.
70MHz scope for $1000 is awful bang for buck; I would personally expect at least 200MHz bandwidth at $1000 total cost.
-
For that price I would definitely buy a Rohde & Schwarz or a Keysight.
Stay away from shitty brands like rigol and siglent.
-
Stay away from shitty brands like rigol and siglent.
Depending on what you are trying to achieve.
If you have 1 grand to blow on a 100MHz scope, sure.
If you only have 1 grand but you are after a hackable 350MHz/500MHz scope with all decoding features and super deep memory, a hacked Rigol is your only option.
Yeah but the OP expresses interest in 200 MHz DSO's which sorta sets his budget.
It might also appear that a deep memory is not required due to interest in the KS.
-
Thanks all for your answers!
For that price I would definitely buy a Rohde & Schwarz or a Keysight.
Well I would love a Rohde & Schwarz (RTB2004 looks amazing) but it's definitely out of my budget.
I will go for the Keysight, though I may be biased.
I will get the -G version, the 4 channel, 70MHz version is great. AWG can't be hacked, unless you are willing to solder tons of parts.
I like the idea of an oscilloscope that I can upgrade by software later, evolving with my needs. But the issue is that if I get the 70MHz (really expensive for basics capabilities) and I need to upgrade it to 200Mhz, I also need to buy the probes and this is also expensive. I really like this Keysight, I know I can trust it but it might be over budget : with the function generator it more than 1200€.
You can see how they perform at their rated frequencies and beyond in this thread:
https://www.eevblog.com/forum/testgear/dso-bandwidth-test-sds1104x-e-dsox1102g-to1104-gds1054b/ (https://www.eevblog.com/forum/testgear/dso-bandwidth-test-sds1104x-e-dsox1102g-to1104-gds1054b/)
So it performs really great right? That may be the one I'll take. It has all functionnalities I need and more (I think) for a nice price. It's not as fancy as a Keysight or a Rohde & Schwarz but I'm a student :-//
Yeah but the OP expresses interest in 200 MHz DSO's which sorta sets his budget.
It might also appear that a deep memory is not required due to interest in the KS.
I don't know if I need a 200 MHz bandwidth, I just don't want to be restricted by a too small bandwidth later, so yes 200MHz felt like a natural choice, especially with my budget!
-
Hello,
I'm starting electronic after programming since I'm 7 years old. I'm currently doing aerospace engineering studies with very little electronic courses so I decided to learn it by myself. I'm mostly interested in the embedded world and diy modular synth. I'm looking for an oscilloscope (and I know I'm not the first one to ask you this) for more or less 1000€. I'll buy it in June.
I want to be able to use my oscilloscope in my current projects of course but also the future ones. It has to be polyvalent, reliable and durable. I think I need 4 channels, serial/protocol decoding. I used in an association the Siglent SDS1202X-E which I really liked.
So far, I have found these scopes :
- Rigol MS05000 (https://www.batronix.com/shop/oscilloscopes/Rigol-MSO5072.html#yoReviews) which once hacked must be a very powerful scope for the price. But the software looks terrible and very slow from the teardown video. Does it get better after software upgrade?
- Siglent SDS1204X-E (https://www.batronix.com/shop/oscilloscopes/Siglent-SDS1204X-E.html) easy choice but it does everything I need I think. But maybe for my budget I can get better?
- Keysight DSOX1204A (https://fr.farnell.com/keysight-technologies/dsox1204a/dso-4-canaux-70mhz-1gsps-5ns/dp/2984794). I really like oscilloscopes from Keysight , they feel reliable and look really great. Is it easy to hack?
Thanks a lot for all of your advices.
Add the GW Instek GDS-2074E to the list (hackable to 300MHz but limited to two channels at this bandwidth).
For the next couple of years I'd stay away from the Rigol MSO5000. There are too many bugs to fix in this scope.
Also buy a seperate function generator. It will be better for the same price.
-
If it's a tool that makes up its cost quickly, I won't worry about its price as long as I can afford it. If it's a toy (in my case, laser engraver, 3D printer, CNC, etc.), I go with the cheapest that actually work.
Here we aren't talking about price but rather how much scope you get for $1000. 70MHz is a toy, as soon as you look at any moderate speed digital signals you need as much bandwidth you can get if you want to see every overshoot, ringing, and other signal integrity problems.
-
- Rigol MS05000 (https://www.batronix.com/shop/oscilloscopes/Rigol-MSO5072.html#yoReviews) which once hacked must be a very powerful scope for the price. But the software looks terrible and very slow from the teardown video. Does it get better after software upgrade?
It already got a lot better and new features seem to be appearing, too.
Is it good enough for your strict UI standards? Who knows, but turning down a sub-$1000 350MHz touch-screen 'scope with four channels, signal generator and hundreds of megabytes of memory just because the touchscreen is a bit slow to move the channels around waves around seems silly to me. It's not a smartphone. Use the knobs like everybody else.
nb. Dave's video doesn't show what the touchscreen is most useful for because Dave's too old-school to do it. The touch screen is for menu selections, use it instead of the stupid twisty knob.
-
If it's a tool that makes up its cost quickly, I won't worry about its price as long as I can afford it. If it's a toy (in my case, laser engraver, 3D printer, CNC, etc.), I go with the cheapest that actually work.
Here we aren't talking about price but rather how much scope you get for $1000. 70MHz is a toy, as soon as you look at any moderate speed digital signals you need as much bandwidth you can get if you want to see every overshoot, ringing, and other signal integrity problems.
But these measurements are very specific. An older Agilent, Lecroy or Tektronix scope will do just fine and costs less than $1000. Better get a modern daily driver in the 100MHz to 200MHz range with lots of features.
Is it good enough for your strict UI standards? Who knows, but turning down a sub-$1000 350MHz touch-screen 'scope with four channels, signal generator and hundreds of megabytes of memory just because the touchscreen is a bit slow to move the channels around waves around seems silly to me.
It isn't silly. A slow/clumsy UI and bugs will make doing a measurement a slow and painful process. You want a tool to work for you and not the other way around. You won't need 350MHz and 100+ Mpts memory every day so trading the comfort you'll need every day for something you need once or twice per year is definitely not a good deal.
-
It isn't silly. A slow/clumsy UI and bugs will make doing a measurement a slow and painful process. You want a tool to work for you and not the other way around. You won't need 350MHz and 100+ Mpts memory every day so trading the comfort you'll need every day for something you need once or twice per year is definitely not a good deal.
How do people even manage without touch screen 'scopes? Moving your hand across to the knob and turning it is is such a pain compared to just being able to pinch-zoom. :-//
Seriously: This 'scope has the advantage of navigating the menus really quickly. Navigating menus with twisty knobs is a much bigger time waster than zooming with a dial.
Me? I'd settle for just the menus this time around. Get the zooming/panning next time.
(or maybe we shouldn't have the touch-screen menus just out of spite)
-
Id say Keysight too.
If you want a fast responsive scope that gets the job done reliably this is the way to go. The UI on these scopes goes back >20 years from back when they ware still called HP and had CRT screens. Its tried and tested to work. Also one of the rare scope breeds that feels just as responsive as a analog scope, any twiddling of controls seams to have an instant effect on the screen.
If later on you need a higher performance scope for >500MHz or something then just go buy an old scope off ebay and keep this little scope around. You will often not need the bandwidth.
I personally have a MSO6034A (Its kinda a old version of the MSOX3000 series) and its my goto scope for most tasks even tho i have a Keysight Infiniium MSO9000 with 4GHz of bandwith sitting right next to it.
Not that Rigol or Siglent are bad scopes, they are excellent value for money, but once you get over 1 grand there are better options.
-
Id say Keysight too.
If you want a fast responsive scope that gets the job done reliably this is the way to go.
So long as you don't want any memory.
(there's a reason they go faster - they don't have to process more than a tiny amount of sample memory)
-
If the specs and functionality aren't there it's moot no matter how good the UI is. I would personally not consider any scope below 100MHz at the $500 mark, and 200MHz at the $1000 mark, similar story for sample memory. I have used a rigol before and have no complaints about the UI, it gets the job done.
-
Well 1Mpts is not all that bad, sure more is nice but you don't get a whole lot of extra usability by going to 10Mpt and above, also here you get the max possible update rate with full memory enabled. Cases where you actually need more memory to get the job done are far and few.
This is one of the reasons i like my old MSO6000 so much. Turn it on, wait for the <10 second boot time and start poking around. No need to trade off memory versus update rate, you get all of it all the time. Its not about the performance or the features, its about quickly getting a look at the waveform you want to see.
But as usual it depends on the use case of the person buying the scope.
What works for me is having a zippy fast low performance scope for day to day use and the big 4Ghz 20GS/s 1Gpts boatanchor of a scope for when performance is needed. The low performance scope likely sees about 80% of total scope use just because it gets the job done faster and easier, i only go for the big one when i can't get the job done without the performance.
-
I'm failing to see how having (eg.) less than "analog" response time on the vertical position control would prevent a job from being done.
-
If you may consider second hand, with patience you'll find some decent HP/A/K, Tek, LeCroy and maybe R&S with serial decode at that price.
If you want more BW than 200MHz look for a used LeCroy 9354, 9374, 9384, LTxxx, LCxxx, DDA, you won't get serial decode with those models but for 1000$ you should be able to get a PC based serial decode solution as well as the DSO.
-
I'm failing to see how having (eg.) less than "analog" response time on the vertical position control would prevent a job from being done.
That is your problem right there! Try a few oscilloscopes and you'll notice there is so much more outside your Rigol box.
-
True, a slow unresponsive scope can still get the job done, but when you have used a lot of different scopes you really come to appreciate speed and ease of setup. The quicker you can set up the scope the less time it takes away from getting actual work done and the less often you decide to suffer trough a not quite optimally set up scope view that shows the waveform, but not very well, yet you don't feel like messing with the scope to get a better image so you just continue using it while potentially missing signal anomalies.
This sort of great UI experience is something you come into contact each time you use the scope. The >100MHz bandwith or >1Mpts only come useful occasionally in the average kind of scope use case. Heck i spend a lot of the time with the 20MHz BW filter enabled on my scopes because the extra bandwidth just makes the waveform fuzzyer when poking around low speed circuitry.
Scopes get expensive very quickly once you start going up in bandwidth, so for high performance stuff its much more cost effective to go look for a old used scope. They might be slow, clumsy and feature lacking but you can get your hands on 1GHz bandwith for the rare cases where you need it. The clumsily slow operation is not so important anymore because you are not going to be using it often.
Sure id love a scope that does both speed and performance, but something like a Keysight MSOX6004A is way out of my budget.
-
True, a slow unresponsive scope can still get the job done, but when you have used a lot of different scopes you really come to appreciate speed and ease of setup.
"Setup" is precisely where a touch screen comes into play - no more twiddly knob for making menu selections.
That is your problem right there! Try a few oscilloscopes and you'll notice there is so much more outside your Rigol box.
You mean like Siglents which use the twiddly knob to select measurements from a huge list instead of just having a row of buttons on the left?
-
Some additional questions:
1. Can someone give me a pratical example of the use of the 200 Mpts (of the Rigol)?
2. When the MSO5000 bugs are ironed out won't the Rigol be a extremely good bang-for-buck deal? I'm not a Rigol fanboy but it's the same company that created the DS1054Z... No offense to anyone!
3. Can you give examples of situations where this Rigol won't be doing what these Siglent, Keysight do? (no need to insist on the UI responsiveness, I've acked that one)
-
In the last month I tried the Rigol MSO5000, Rigol DS1054Z, Siglent SDS1104X-E and GWInstek GDS1054B, as well as my Keysight EDUX1002G and Micsig TO1104.
The Rigol MSO5000 seems impressive when hacked, but the LCD being dim, the UI lag and the decoding that was not fast enough (It certainly is SW based decoder), gave a lot of headache and had to return it. I don't think Rigol will be able to fix the UI lag on the MSO5000, because it is slower than the DS1054Z!!
The Keysight is very responsive, and hardware serial decoding. I recently tested SPI decoding using STM32 @ 180MHz with 22.5MHz SPI signal and the Keysight was able to capture all 35,000 single specific byte per second I was sending out of the STM32 board. The SDS1104X-E and GDS1054B (and I assume will be a similar rate on the MDO5000 and DS1054Z) was capturing a maximum 3,000 of the same byte per second (about 10% of Keysight). It doesn't mean that they cannot be used to debug effectively, you can still capture in the long memory and you will see all the bytes there, but if you are trying to catch a specific byte, you might miss it with the Siglent and Rigol.
For $499, the SDS1104X-E is a very good option.
The GWInstek GDS1054B I tested was kind of slow when using 1Mpts and 10Mpts (cannot decode at this sample memory capacity), anyway decoding is unsupported and only available on hacked scopes. But you can enable segmented memory and search functions. But I did not like the way GWInstek uses the LCD display, it looks like a 10-year old LCD DSO, even when some characters are very well defined, but the traces, grid look very old compared to DS1054Z, SDS1104X-E or Keysight EDUX1002G.
The Keysight EDUX1002G lacks serial decode table or list, so you need to scroll the waveform to look for the information you need.
If I were looking for a scope today, I would probably buy the EDUX1002A for $390 and hack it to DSO1102G, it is not very difficult to solder the missing parts, and the latest hack from FERCSA enables lots of serial decode functions available on the DSOX3000A and you have 2 + 1 channels allowing to decode UART, i2c, I2S, SPI (3 wire - only MOSI or MISO at a time), and 2GSa/s, 1Mpts, segmented memory...
-
You mean like Siglents which use the twiddly knob to select measurements from a huge list instead of just having a row of buttons on the left?
I agree, the way you select measurements on Siglent is very inefficient, and the knob that always misses what I am trying to select...
-
I also like the DSO1204A/G, but I don't think a hack will be available anytime soon... it has a new Linux based software and the base price of $1000 makes it too expensive to experiment with hacking. The DSO1102A/G share the same code as the DSOX2000 and DOSX3000 series with extensive hacking history... That is why the latest software hack for the 1000X scope was so impressive in the number of features enabled by the hack.
-
You mean like Siglents which use the twiddly knob to select measurements from a huge list instead of just having a row of buttons on the left?
I agree, the way you select measurements on Siglent is very inefficient, and the knob that always misses what I am trying to select...
Which is why the seperate select button on the GW Instek scopes is so great. The combined rotary & pushbuttons are a nightmare to use.
-
- Rigol MS05000 (https://www.batronix.com/shop/oscilloscopes/Rigol-MSO5072.html#yoReviews) which once hacked must be a very powerful scope for the price. But the software looks terrible and very slow from the teardown video. Does it get better after software upgrade?
I got one since end of november 2018.
I couldn´t agree to those who call this thing slow or too buggy to work with it.
It got some bugs for sure, but it wouldn´t affect the daily work with it.
It got hardware based decoding, 8Gs/s, 200Mpt memorydepth, 1Mpt FFT, waste amount of math functions, waste amount of trigger functions, 350Mhz bandwith, HDMI output, etc, etc.
For sure a dim display but 9 inch tall, the lag of the touchscreen doesn´t worry me, I use the knobs.
I wouldn´t say this is an superior scope which "kill" everyone else on the market.
But for 1000 bucks new, there´s nothing on the market to compete with it.
Nothing.
-
It got hardware based decoding
I could not find any documentation that claims it has HW decoding, and based on my tests, it looks like it is implemented in software.
Too bad I returned mine, I would like to test and compare to the Keysight scope.
-
You mean like Siglents which use the twiddly knob to select measurements from a huge list instead of just having a row of buttons on the left?
I agree, the way you select measurements on Siglent is very inefficient, and the knob that always misses what I am trying to select...
Just like each scope has a different UI and need to used in its own way to get results the same applies to the universal adjustment control.
So you're having trouble selecting the option required, well use it in the correct way !
Place hand on top of scope and use thumb and forefinger to operate the control............there is a reason for where the multi-purpose control is positioned where it is.
Fingers should never leave the top of the scope while adjustments are being made, then precision control is quite possible and option selection is rarely missed.
-
You mean like Siglents which use the twiddly knob to select measurements from a huge list instead of just having a row of buttons on the left?
I agree, the way you select measurements on Siglent is very inefficient, and the knob that always misses what I am trying to select...
Just like each scope has a different UI and need to used in its own way to get results the same applies to the universal adjustment control.
So you're having trouble selecting the option required, well use it in the correct way !
Place hand on top of scope and use thumb and forefinger to operate the control............there is a reason for where the multi-purpose control is positioned where it is.
Fingers should never leave the top of the scope while adjustments are being made, then precision control is quite possible and option selection is rarely missed.
:clap: You are describing exactly how I ended up using the knob after a couple of days of frustration... but the Keysight has the same multi use knob and it almost never selects the wrong option. You don't need to be doing the strange 3/4-finger stuff with the knob, you just use 2 and it works just fine
-
You mean like Siglents which use the twiddly knob to select measurements from a huge list instead of just having a row of buttons on the left?
I agree, the way you select measurements on Siglent is very inefficient, and the knob that always misses what I am trying to select...
Which is why the seperate select button on the GW Instek scopes is so great. The combined rotary & pushbuttons are a nightmare to use.
Yes, it is a great solution and I like it a lot.
-
I could not find any documentation that claims it has HW decoding
First you can see it in the phoenix architecture, second by the measurements I´ve taken some times ago ( will look and post it tomorrow, was in the eev-blog topic) .
My impression after nearly 6 months working with it, most in the company :
Raw diamond, needs to be polished by firmware upgrades.
If they (rigol) takes it serious, everyone will be smiling.
-
You mean like Siglents which use the twiddly knob to select measurements from a huge list instead of just having a row of buttons on the left?
I agree, the way you select measurements on Siglent is very inefficient, and the knob that always misses what I am trying to select...
Just like each scope has a different UI and need to used in its own way to get results the same applies to the universal adjustment control.
So you're having trouble selecting the option required, well use it in the correct way !
Place hand on top of scope and use thumb and forefinger to operate the control............there is a reason for where the multi-purpose control is positioned where it is.
Fingers should never leave the top of the scope while adjustments are being made, then precision control is quite possible and option selection is rarely missed.
Is there any specific tune you should whistle while doing this? And what brand and type of tin foil hat to wear to make this process even easier?
More seriously: I used to own a high end Agilent scope and that suffered from exactly the same problems with the combined rotary/push button. Setting up decoding on it was extremely tedious.
-
Some additional questions:
1. Can someone give me a pratical example of the use of the 200 Mpts (of the Rigol)?
2. When the MSO5000 bugs are ironed out won't the Rigol be a extremely good bang-for-buck deal? I'm not a Rigol fanboy but it's the same company that created the DS1054Z... No offense to anyone!
3. Can you give examples of situations where this Rigol won't be doing what these Siglent, Keysight do? (no need to insist on the UI responsiveness, I've acked that one)
1. it means 50 Mpoints per ch. At 2Gs/s that is 25ms worth of data in full bandwidth and timing resolution.
2. I think it will be best for embedded dev once they finish the development. But they have lots of work to do.
3. No, it will be superior, except for low level signals. Siglent and Keysight have better frontend sensitivity. Siglent is quite good at that. Rigol is better bandwidth wise. Also, MSO5000 is proper MSO. Keysight is not MSO at all, and Siglent has MSO addon but it is not properly integrated.
Like Martin says, hacked, it has potential to become more legendary than the little Z ..
-
I could not find any documentation that claims it has HW decoding
First you can see it in the phoenix architecture, second by the measurements I´ve taken some times ago ( will look and post it tomorrow, was in the eev-blog topic) .
My impression after nearly 6 months working with it, most in the company :
Raw diamond, needs to be polished by firmware upgrades.
If they (rigol) takes it serious, everyone will be smiling.
I used an STM32 to generate 18MHz SPI signal (generating 11,000 repetitions of a specific byte sequence per second) and measured the trigger output on the Keysight EDUX1002G, Siglent SDS1104X-E and GWInstek GDS1054B. The Keysight can trigger all 11,000 repetitions per second (I updated the test gig to go up to 180MHz clock and 22.5MHz SPI signal generating 35,000 repetitions, and Keysight was able to trigger on all 35,000). The Siglent and GWInstek were able to trigger max 3,000 of the 11,000.
I will re-test the MDO5074 in about 1-2 weeks with the same test setup and update here the results.
-
You mean like Siglents which use the twiddly knob to select measurements from a huge list instead of just having a row of buttons on the left?
I agree, the way you select measurements on Siglent is very inefficient, and the knob that always misses what I am trying to select...
Just like each scope has a different UI and need to used in its own way to get results the same applies to the universal adjustment control.
So you're having trouble selecting the option required, well use it in the correct way !
Place hand on top of scope and use thumb and forefinger to operate the control............there is a reason for where the multi-purpose control is positioned where it is.
Fingers should never leave the top of the scope while adjustments are being made, then precision control is quite possible and option selection is rarely missed.
:clap: You are describing exactly how I ended up using the knob after a couple of days of frustration...
Glad you learnt a Siglent UI trick on your own. ;)
But seriously, I hear you.....it could be improved. A bigger knob would certainly ease selection as would the addition of an indented encoder but are they 100% necessary for full control of the selection control, no.
but the Keysight has the same multi use knob and it almost never selects the wrong option. You don't need to be doing the strange 3/4-finger stuff with the knob, you just use 2 and it works just fine
Funny this ^ is the same result I have with the Siglent encoder.
I guess it's just what you're used to.
-
You mean like Siglents which use the twiddly knob to select measurements from a huge list instead of just having a row of buttons on the left?
I agree, the way you select measurements on Siglent is very inefficient, and the knob that always misses what I am trying to select...
Just like each scope has a different UI and need to used in its own way to get results the same applies to the universal adjustment control.
So you're having trouble selecting the option required, well use it in the correct way !
Place hand on top of scope and use thumb and forefinger to operate the control............there is a reason for where the multi-purpose control is positioned where it is.
Fingers should never leave the top of the scope while adjustments are being made, then precision control is quite possible and option selection is rarely missed.
:clap: You are describing exactly how I ended up using the knob after a couple of days of frustration...
Glad you learnt a Siglent UI trick on your own. ;)
But seriously, I hear you.....it could be improved. A bigger knob would certainly ease selection as would the addition of an indented encoder but are they 100% necessary for full control of the selection control, no.
but the Keysight has the same multi use knob and it almost never selects the wrong option. You don't need to be doing the strange 3/4-finger stuff with the knob, you just use 2 and it works just fine
Funny this ^ is the same result I have with the Siglent encoder.
I guess it's just what you're used to.
Actually I just checked with my Keysight and Siglent side by side, and I am doing the same trick... 2 fingers on the knob and the remaining fingers on the top of the scope... but on the Siglent I cannot let my 2 fingers go off the knob before pushing it, it means that I am pushing with my 2 fingers on the side of the knob, but on the Keysight I can let the 2 fingers go off the knob and push with the thumb at the end of the knob. I think it is just a sensitivity issue that Siglent needs to tune
-
On the Keysight, the dented knobs have a softer dent and on the Siglent they have a harder dent... but it is the opposite on the small knobs without dents... they turn very smoothly and easily on the Keysight but on the Siglent are stiffer and harder to turn.
-
I could not find any documentation that claims it has HW decoding
First you can see it in the phoenix architecture, second by the measurements I´ve taken some times ago ( will look and post it tomorrow, was in the eev-blog topic) .
My impression after nearly 6 months working with it, most in the company :
Raw diamond, needs to be polished by firmware upgrades.
If they (rigol) takes it serious, everyone will be smiling.
I think the Phoenix chipset only covers the front end and what Rigol calls SPU is probably the ZYNQ-7000 and then the 2 ARM-9 cores... and the advanced triggering is on the SPU, so software based.
-
Touchscreen doesn't always make it faster and easier to set things up.
I had the Keysight MSO9000 with a touchscreen for a good few years now and most of the time ended up using a mouse instead. They did revamp the UI to be more touch friendly and it does help, but its still not quite there because its still a pretty Windows-ey UI Also since scopes tend to live at the back of the bench (so that you can work in front of it) means that you end up extending your arm all the way to reach the scope. After a while your arm really gets tired and you can't use the "lean one finger on top of the scope" trick i tend to use with most scopes because you can't reach a lot of the touchscreen buttons. Not saying touch is bad, but the Infiniium series is not a very good implementation of it (But still good enough to be usable)
But using a knob to scroll trough a long list is surely worse right?
Well.. not all knobs are created equal. Keysight did a particularly good job at knobs. They never miss steps and have really really good velocity control to them. You can easily click trough one increment at a time or flick the knob and jump 1000 increments pretty predictably. Combine this with the result showing up pretty much on the next LCD refresh cycle means you never overshoot any settings and can quickly jump trough long lists when you know where things are.
Latency is the mortal enemy of encoder knobs. My dads 2016 Volvo V60 has a particularly bad case of it for example. You get trough the countless menus on the infotainment trough a encoder knob (but luckily still has plenty of dedicated buttons unlike more modern touchscreen crap) yet the knob has about 300ms of latency between the knob and seeing it happen on the screen. This makes it feel horribly clumsy to use for me. I constantly overshoot menu items even tho the encoder actually has really nice detents. But it does queue UI input so when you memorize a path trough menus you can actually push buttons faster than the screen updates and it will catch up after you.
Oh and i have also used a Micsig touchscreen scope quite a bit, and overall didn't really like it because of how the menus work and its a bit sluggish. But compared to other portable/near-handheld scopes its pretty awesome.
-
I could not find any documentation that claims it has HW decoding
First you can see it in the phoenix architecture, second by the measurements I´ve taken some times ago ( will look and post it tomorrow, was in the eev-blog topic) .
https://www.eevblog.com/forum/blog/new-rigol-scope/msg2034745/#msg2034745 (https://www.eevblog.com/forum/blog/new-rigol-scope/msg2034745/#msg2034745)
-
I could not find any documentation that claims it has HW decoding
First you can see it in the phoenix architecture, second by the measurements I´ve taken some times ago ( will look and post it tomorrow, was in the eev-blog topic) .
https://www.eevblog.com/forum/blog/new-rigol-scope/msg2034745/#msg2034745 (https://www.eevblog.com/forum/blog/new-rigol-scope/msg2034745/#msg2034745)
If you take the 50ms timebase, stop or do single capture... and zoom in... can it decode correctly? If yes, then it is software based and decoding from the sample memory.
Another test... can you do a single capture without decode and then enable decode? Does it decode correctly? Then it is software based and decoding after the fact.
-
Hi,
Will do it in a few days, on work - at home, I don´t have a SPI "Source".
The MSO5000 uses the Ultravision II platform:
(https://s18.directupload.net/images/190514/temp/g7c9reew.jpg) (https://www.directupload.net/file/d/5452/g7c9reew_jpg.htm)
-
Rigol Ultravision existed for ever, and I assume it is a combination of CPU + FPGA. From the picture you posted, I assume Ultravision II is the combination of Spartan-6 FPGA + 2 core ARM-9 + ZINQ-7000... it is not very clear from the available information, what is what... My interpretation is that Phoenix is a chipset for acquisition and the rest is done in FPGA + Softcore + CPU + software
http://www.saelig.com/supplier/Rigol/ultra_vision_070214p.pdf (http://www.saelig.com/supplier/Rigol/ultra_vision_070214p.pdf)
-
The picture shows "Hardware Decode" in the SPU Block, the phoenix chipset means only the frontend, I guess.
I´m in good contact with rigol.eu, I simply ask them for it.
Edit:
Found the pic in much better resolution:
(https://s18.directupload.net/images/190514/temp/d6lmssib.jpg) (https://www.directupload.net/file/d/5452/d6lmssib_jpg.htm)