Silgent SDS1102X vs Owon MSO8102T

[__exit]

I am completely new to electronics and I am looking for an oscilloscope mostly for education, for small projects, house electronics, car electronics, some inspection and reverse engineering.
I plan on taking free online courses, doing beginner projects, following along with diligent's class and other examples online.

I want an oscilliscope that will cover most lab topics for courses as well as allow me to continue as a hobby and practical use. I am thinking that a higher end entry level scope is probably right for me so I can get prolonged use.

I have considered pc based vs non-pc based, after watching a eevblog video decided to go with standalone unit because quality is too variable in USB land and often the manufacturers listed ratings don't match product's ability.

I am aware of the following:
* 5-10x mhz = sampling otherwise marketing and not accurate
* higher mhz means wider range of circuits/technologies its useful for
* probes matter and can scale the ticks of the oscilloscope measurements / div
and some other begginer stuff that i've learned from watching eevblog videos and reading beginner guides.


I watched some videos about sds1000x series (sds1202x) video and rigol1054z videos from eevblog.
I am not sure if i need 4 channels or not and until I understand everything better I do not want to hack an oscilloscope because I want my warrantee, so that means a 50MHz instead of 100MHz. It seems to me like the siglent sds1102x is a better scope given this criteria; especially given SDS1012x currently has a free decoder option.

While I was looking around on Amazon I happened to notice the Owon MSO8102T (with logic analyzer, normally ~$800).

I bought both (SDS1102X and MSO8102T) @ ~$400 to lock in the price, but I don't know which I should keep.

Suggestions?


NOTE: I can't find a good breakdown / review of MSO8102T.

[tautech]

Suggestions?

Welcome to the forum.


SDS1102X+ (MSO model) has LA but the LA probe set and LA SW option is extra cost. These Plus models also have an inbuilt Function Gen free.
They're a bit dearer than the plain SDS1102X but you do get option to add the additional MSO functionality later if you need it.

Some screenshots of the MSO UI are here:
https://www.eevblog.com/forum/testgear/siglent-sds1000x-series-oscilloscopes/msg1068364/#msg1068364

[__exit]

From your response it seems like the Owon is not a good choice and that you suggest paying more money for the sds1102x+. I was trying to spend <$500 because this is a beginner scope. If its absolutely worth it, I can stretch my budget a bit.

Am I understanding your response?

[__exit]

Actually I think its more accurate to say - I want to spend close to the price of the Rigol 1054Z @ $399 and will expand the budget to get a better oscilloscope especially with holiday deals.

[tautech]

From your response it seems like the Owon is not a good choice and that you suggest paying more money for the sds1102x+. I was trying to spend <$500 because this is a beginner scope. If its absolutely worth it, I can stretch my budget a bit.

Am I understanding your response?

Not entirely.
I'm pointing out the MSO (X+) version in case you weren't aware of it.

Actually I think its more accurate to say - I want to spend close to the price of the Rigol 1054Z @ $399 and will expand the budget to get a better oscilloscope especially with holiday deals.

OK, you're budget constrained, just make sure the unit you buy has the Decode option for free as per the current promotion.

[rf-loop]

Siglent is real MSO.  Owon is not. It is LA  or  analog scope. But you can not use it mixed. I have not used this Owon model but I have never seen any evidence, any signs, that it can run LA and Analog signals simultaneously.  It is dual mode equipment, other mode is pure LA  alone. Other mode is Oscilloscope alone. No mixed, until some one show any evidence I'm wrong. I believe Owon show even one image in they AD's  where can see analog signal on the screen where is also digital LA running - if this Mixed mode is possible.

[__exit]

From your response it seems like the Owon is not a good choice and that you suggest paying more money for the sds1102x+. I was trying to spend <$500 because this is a beginner scope. If its absolutely worth it, I can stretch my budget a bit.

Am I understanding your response?

Not entirely.
I'm pointing out the MSO (X+) version in case you weren't aware of it.

Actually I think its more accurate to say - I want to spend close to the price of the Rigol 1054Z @ $399 and will expand the budget to get a better oscilloscope especially with holiday deals.

OK, you're budget constrained, just make sure the unit you buy has the Decode option for free as per the current promotion.

So I am sort-of budget constrained. Its my first everything and I have no real experience yet in electronics, so I artificially budgeted myself to a beginner scope that will have practical applications. I don't understand the difference between the 1102X and the 1102X+ models to make a call between them. My simple view of the world involved higher bandwidth generally means better as long as the sample rate supports it. Also having a warranty sounds awesome, so i wanted to avoid hacking, so i was steering of the Rigol DS1054Z.

I just purchased a saelae logic 8 to fill my logic analyzer curiosity from some help i received in the #eevblog IRC channel.

I also purchased a Rigol DS1054Z and I am in progress of purchasing a SDS1102X so I can compare them side by side. I've watched where the final choice is possibly DS1054Z depending on the reason you want one. Im totally new so I don't actually know what I will need it for. From that video it seems like the SDS1202X is the rival of the higher quality Rigol 2000 series so i made the assuption that the 100MHz was also higher quality. I know the siglent is double the wfms of the rigol after hacking (siglent can go up to 400000 for easier capture of anomalies?). Also it has a better screen. The rigol after hacking (not 100% opposed to it) has deeper memory depth and will be 100MHz for about $25 cheaper with current deals.

So to me, the choice between the rigol and the siglent came down to "Do I need 4 channels?" and "Do I want to hack my scope?" and "Should i just buy higher quality because its $25 dollars away?". The answers were in order, "not sure, "probably not unless i have to", and "screw it, $25 bucks...". So I arrived at the SDS1012X.

After talking in the IRC channel last night, I arrived at the conclusion that I need to research a bit more and just make a call, so I am just buying both because either way I will have a scope that I can deal with.

So between the SDS1102X and the X+, it seems like its just the waveform generator @ 25MHz. Is a 25MHz generator worth the extra money? I have an idea of what they do but don't know that I will need it. Right now I have an arduino, some schematics, now a logic analyzer, and some project ideas. Will I use a AWG?

I am trying to use this oscilloscope to help me debug electronics and further learning. Also to understand the domain of parameters to learn how electronics work (instead of taking university classes @ n x $1000).

[__exit]

Siglent is real MSO.  Owon is not. It is LA  or  analog scope. But you can not use it mixed. I have not used this Owon model but I have never seen any evidence, any signs, that it can run LA and Analog signals simultaneously.  It is dual mode equipment, other mode is pure LA  alone. Other mode is Oscilloscope alone. No mixed, until some one show any evidence I'm wrong. I believe Owon show even one image in they AD's  where can see analog signal on the screen where is also digital LA running - if this Mixed mode is possible.

Owon scratched off the list. I also got some help in IRC last night where they suggested just getting a standalone USB logic analyzer. I liked that approach even though I couldn't see them side by side. This post helped be move past the Owon.

[tautech]

So between the SDS1102X and the X+, it seems like its just the waveform generator @ 25MHz. Is a 25MHz generator worth the extra money? I have an idea of what they do but don't know that I will need it. Right now I have an arduino, some schematics, now a logic analyzer, and some project ideas. Will I use a AWG?

As you have an saelae logic 8 now forget the X+ and stick with an X. Should you need a AWG later you have saved a few $ and can address that later if and when needed.

After talking in the IRC channel last night, I arrived at the conclusion that I need to research a bit more and just make a call, so I am just buying both because either way I will have a scope that I can deal with.

There's some interesting discussion here and further on that's worth studying:
https://www.eevblog.com/forum/testgear/rigol-ds1054z-bandwidth/msg1098270/#msg1098270

[MrWolf]

Here are some tests on horizontal system:
https://www.eevblog.com/forum/testgear/testing-dso-auto-measurements-accuracy-across-timebases/
Rigol is basically 1k memory scope for many intents ans purposes. Siglent said to have 70k...

[__exit]

Here are some tests on horizontal system:
https://www.eevblog.com/forum/testgear/testing-dso-auto-measurements-accuracy-across-timebases/
Rigol is basically 1k memory scope for many intents ans purposes. Siglent said to have 70k...

sorry, im still quite new and i am trying to decipher what the test results mean. They seem to be measured in time (ns and us - nano, micro?) and then against a time base (what time is shown in capture?). Is the plot the time it takes to display a Rise / Period?

If so, then faster (less) is better?

It also seems like you are looking for an indication of which memory "zone" is used, primary or secondary. Is this similar to tiered performance memories like cpu cache? Or are you testing about the tradeoffs made when the available memory shrinks because memory is allocated in zones and as different functions of the oscilloscope use different amounts of memory, forcing the a reallocation of memory zones that has a negative effect on performance?

I am still reading through trying to understand whats you mean by 1K mem and 70K. I believe you are saying that the Rigol behaves as if it has 1K of memory while the siglent behaves as if it has 70K while they both claim much more available memory.

Maybe i'm off?

[__exit]

After talking in the IRC channel last night, I arrived at the conclusion that I need to research a bit more and just make a call, so I am just buying both because either way I will have a scope that I can deal with.

There's some interesting discussion here and further on that's worth studying:
https://www.eevblog.com/forum/testgear/rigol-ds1054z-bandwidth/msg1098270/#msg1098270

Are you are pointing out that Rigol uses a smoothing function that heavily approximates actual results and that siglent's smoothing function is closer to the real sample set? Given that you are a distributor and seem to believe in siglent products I assume this is your position as well as rf-kill's initial point in the post.

I haven't read past the first page because I have to look up every concept in wikipedia and extrapolate, but am I in the right ballpark?

I am going to keep reading but want to make sure I am making progress in the right direction. Also its nice to interact with intelligent people , this is a cool forum.

[tautech]

After talking in the IRC channel last night, I arrived at the conclusion that I need to research a bit more and just make a call, so I am just buying both because either way I will have a scope that I can deal with.

There's some interesting discussion here and further on that's worth studying:
https://www.eevblog.com/forum/testgear/rigol-ds1054z-bandwidth/msg1098270/#msg1098270

Are you are pointing out that Rigol uses a smoothing function that heavily approximates actual results and that siglent's smoothing function is closer to the real sample set? Given that you are a distributor and seem to believe in siglent products I assume this is your position as well as rf-kill's initial point in the post.

I haven't read past the first page because I have to look up every concept in wikipedia and extrapolate, but am I in the right ballpark?

I am going to keep reading but want to make sure I am making progress in the right direction. Also its nice to interact with intelligent people , this is a cool forum.

Don't mock rf-loop, he's one of a few very knowledgeable members on this forum and a fellow Siglent and OWON distributor.
He's been a position to buy and test some Rigol products extensively and then post the facts here on the forum.

If you choose to ignore his findings, that's entirely your prerogative.
If you want to study more of what he has found use the Search button with a brand/model # and specify his forum name in the member box.

By all means keep asking questions. 
Good luck with your swat.


 

[__exit]

After talking in the IRC channel last night, I arrived at the conclusion that I need to research a bit more and just make a call, so I am just buying both because either way I will have a scope that I can deal with.

There's some interesting discussion here and further on that's worth studying:
https://www.eevblog.com/forum/testgear/rigol-ds1054z-bandwidth/msg1098270/#msg1098270

Are you are pointing out that Rigol uses a smoothing function that heavily approximates actual results and that siglent's smoothing function is closer to the real sample set? Given that you are a distributor and seem to believe in siglent products I assume this is your position as well as rf-kill's initial point in the post.

I haven't read past the first page because I have to look up every concept in wikipedia and extrapolate, but am I in the right ballpark?

I am going to keep reading but want to make sure I am making progress in the right direction. Also its nice to interact with intelligent people , this is a cool forum.

Don't mock rf-loop, he's one of a few very knowledgeable members on this forum and a fellow Siglent and OWON distributor.
He's been a position to buy and test some Rigol products extensively and then post the facts here on the forum.

If you choose to ignore his findings, that's entirely your prerogative.
If you want to study more of what he has found use the Search button with a brand/model # and specify his forum name in the member box.

By all means keep asking questions. 
Good luck with your swat.

Not mocking him. rfkill is a kernel module that has been haunting me on my raspberry pi recently. Just a swap. rf-loop seems like a badass, very informative posts. Just a beyond my grasp currently.

What I was trying to understand in the prior post was:

"Are you are pointing out that Rigol uses a smoothing function that heavily approximates actual results and that siglent's smoothing function is closer to the real sample set?" - this is what I believe i read.

"Given that you seem to believe in siglent products I assume this is your position as well as rf-loop's initial point in the post." - this is supporting information for my conclusion

I removed the distributor comment which I was using as more supporting information and corrected my speilling of rf-loop (s/rf-kill/rf-loop/g).

rfkill / raspberry pi 3 reference: https://www.raspberrypi.org/forums/viewtopic.php?f=28&t=138858

I guess this is the internet, so its probably not uncommon for people to ask for help and then be jackasses.

[tautech]

Your understanding based on your assumptions are correct.

But at the end of the day your decision must be the best fit for you (specs and $)....one of my mottos. 

Just another thing I'll add in case it hadn't occurred to you; take note of the dates of any reviews as well as content in any threads on instruments you're looking at, as much of what you'll find is out of date quite quickly. Any issue that's 6 months old has likely been fixed so if you find something you're not sure of just ask either here or in the appropriate thread.

[nctnico]

Here are some tests on horizontal system:
https://www.eevblog.com/forum/testgear/testing-dso-auto-measurements-accuracy-across-timebases/
Rigol is basically 1k memory scope for many intents ans purposes. Siglent said to have 70k...

sorry, im still quite new and i am trying to decipher what the test results mean. They seem to be measured in time (ns and us - nano, micro?) and then against a time base (what time is shown in capture?). Is the plot the time it takes to display a Rise / Period?

If so, then faster (less) is better?

It also seems like you are looking for an indication of which memory "zone" is used, primary or secondary. Is this similar to tiered performance memories like cpu cache? Or are you testing about the tradeoffs made when the available memory shrinks because memory is allocated in zones and as different functions of the oscilloscope use different amounts of memory, forcing the a reallocation of memory zones that has a negative effect on performance?

I am still reading through trying to understand whats you mean by 1K mem and 70K. I believe you are saying that the Rigol behaves as if it has 1K of memory while the siglent behaves as if it has 70K while they both claim much more available memory.

Maybe i'm off?

The whole discussion where MrWolf is pointing to is about whether an oscilloscope uses screen data or memory data for doing calculations on waveforms. On oscilloscopes with a lot of memory and a slow processor it will take a long time to do a calculation on all the memory data and it is debateble if it makes sense to do a calculation on a waveform you can't see the shape of and/or using too much data (at some point you won't get any extra precission and/or resolution from adding extra samples). The wrong conclusion from that discussion is that only really cheap scopes use screen data; all oscilloscopes from Keysight use screen data!

It seems Siglent choose to do a best effort and use a larger portion of the memory but due to lack of processing power they had to limit the amount of data they use for  calculations. On my GW Instek GDS2000E series scope (which has a multi core 1GHz ARM Cortex CPU) the data used for the calculations is selectable between cursors, screen and all data (up to 10MPts). I don't see (much) use for using 10Mpts but I guess they thought 'we have the processing power available so let's leave it up to the user'.

I see you also got the Rigol DS1054Z (just hack it!). I'd compare it very closely because having 4 channels is a very usefull feature. Personally I wouldn't even consider an oscilloscope with less than 4 channels.

[MrWolf]

The wrong conclusion from that discussion is that only really cheap scopes use screen data; all oscilloscopes from Keysight use screen data!

Actually the conclusion is that if you use too small dataset for calculations you will get substandard results. Also there are two ways of creating "screen buffer". Thru processing real data into smaller set, or just taking some unmodified points from main record. In last case you can re-gain precision with heavy averaging etc. In first case data goes thru two stages of processing and will be corrupted to almost unusable state. Some examples:
10k "single buffer" scope will perform worse than 100k/1M "dual buffer".
10k "single buffer" will perform better than 1k/1M "dual buffer" etc.
This goes for automated horizontal calculations (rise, period, freq, duty and many other),
not manual fiddling with cursors in zoom mode etc.
So it's mainly about buffer size, not dual vs single buffer. BUT seems that "dual buffer" scopes have no data about "screen buffer" size in datasheet, so you may fall into trap buying one thinking it will provide substantial performance improvement compared to your current scope while it might be much worse in reality.

I will update my thread sometime to make this more clear. Found some new tests to demonstrate it. For example 50k "single buffer" scope can calculate pretty accurate values for frequencies differing up to 3 orders of magnitude (CH1 50Hz, CH2 50kHz for example). Doubt that 1k one will do very well...

[__exit]

I ended up cancelling my order for the Rigol and Saleae and went all in on the Siglent as MSO. This is primarily going to be used for education and I figured a waveform generator would be required from some experiments. Especially after looking at the abilities of the diligent analog discovery unit. I think 2 channels will be enough (i hope*).

I want to thank everyone in here for being so helpful and providing so much useful information. And I can't wait to fully grok the information in these forums! I love how technical it is.

I will see you all again when I have my next set of problems .