Is a Rigol MSO5000 overkill for a hobbyist?

[skander36]

Hi Fungus!
Rigol 5k does not have this setting. Also the others scopes RTB2k, Keysight 2k or Micsig.
It is not like Siglent. I had Siglent 1202X-E and I remember.

[skander36]

Neither RTB2k don't do any better. I'm sure that there is an engineering or mathematical explanation but I can't provide it

HiRes is not active on the RTB2000, it is displayed in gray font, and not in white font. I suspect there are not enough samples available.

Peter

It is possible. I did not realized the difference. Any advice about setup?

[2N3055]

I honestly think HR on scopes is overrated.. How many times you have to observe a perfectly repetitive signal upon which to oversample?
Many times you don't have a precise idea about what you are going to observe, and turning HR on could mask the real stuff you want to see...

HiRes (ERES) is not same as Averaging mode...

[balnazzar]

I honestly think HR on scopes is overrated.. How many times you have to observe a perfectly repetitive signal upon which to oversample?
Many times you don't have a precise idea about what you are going to observe, and turning HR on could mask the real stuff you want to see...

HiRes (ERES) is not same as Averaging mode...

AFAIK the HR modes are more or less implemented as follows (but please correct me if I'm wrong):

Pre-Filter --> Over-Sample --> Average --> Data-Thinning --> Filter

[balnazzar]

That's to say, it's more or less a way to perform averaging in a more sophisticated and less simplistic way.

Bottom line, HR modes are not going to be real substitute for proper high bit count FE/ADCs.

[tautech]

The 10bit R&S, no wonder, "wins"...

SDS2000X Plus in 10 bit mode can challenge it and at much less cost.....remember you had one of those. 

That's why I explicitly requested the same signal captured with the 2x+

Sorry sold out or otherwise I would oblige. 
Only the X-E like yours and a SDS6000A in stock until more arrives.
BTW, do you want a proper chat about SDS5kX, I have some comments that might be of interest. Email me.

I will, as soon as I know more about the fate of that scope. I'll go to the lab on wednesday. But you know, it's not my scope, and it's outside of the price range I can personally afford. But I'll ask.

As for the 2x+, I think Martin has it, and will test with that signal.

Had, now has SDS2104X HD, cough cough now SDS2504X HD !

[skander36]

HiRes is only activated from 100 µs/DIV and larger.

Peter

Thanks! That work.
That means you can't use it to measure high frequency signals. I can't even trigger for single capture at this signal level combined with 100us timebase . LE - After some fine tunning on trigger I was able to do a single capture.

Rigol 5K HiRes does not have this limit.

Robert

[2N3055]

I honestly think HR on scopes is overrated.. How many times you have to observe a perfectly repetitive signal upon which to oversample?
Many times you don't have a precise idea about what you are going to observe, and turning HR on could mask the real stuff you want to see...

HiRes (ERES) is not same as Averaging mode...

AFAIK the HR modes are more or less implemented as follows (but please correct me if I'm wrong):

Pre-Filter --> Over-Sample --> Average --> Data-Thinning --> Filter

Not really.
That doesn't explain what is being processed and how.
In scope parlance Averaging is averaging same data points from different trigger events. Good only for repetitive signals. It does not lowpass filter signal.
Hires is a running average lowpass filter. It trading temporal resolution for amplitude one.
ERES is alternative implementation of lowpass filter with better impulse response..
There is no "data thinning"(we call it decimation) as a separate step anywhere, nor Pre-filter ..
There is also no oversampling but we simply sample as a scope normally would and than apply transformation.
On some scopes (LeCroy, some Siglent) ERES is available as a math function, so you retain original data and can apply it after the fact.

Search internet for ERES whitepapers by Lecroy. They explain it better than I ever could.

[2N3055]

That's to say, it's more or less a way to perform averaging in a more sophisticated and less simplistic way.

Bottom line, HR modes are not going to be real substitute for proper high bit count FE/ADCs.

First sentence no, we cannot say that. See my other response.

Bold sentence absolutely correct, HiRes or ERES are no substitute for low noise, high resolution front end.
BUT they are VERY useful supplementary tool to use when signal observed permits it and calls for it.

[balnazzar]

HiRes is only activated from 100 µs/DIV and larger.

Peter

Thanks! That work.
That means you can't use it to measure high frequency signals. I can't even trigger for single capture at this signal level combined with 100us timebase ...
Rigol 5K HiRes does not have this limit.

Robert

Thanks for all these tests you are making. Also, that seems to be quite an effective HR implementation, although I don't like that horizontal limitation.

Would you please feed a truly arbitrary waveform of your choice to the scope and see how its HR mode fares with it? Thanks!

[Fungus]

I honestly think HR on scopes is overrated.. How many times you have to observe a perfectly repetitive signal upon which to oversample?

You're thinking about "averaging".  Hires is different.

[balnazzar]

I honestly think HR on scopes is overrated.. How many times you have to observe a perfectly repetitive signal upon which to oversample?
Many times you don't have a precise idea about what you are going to observe, and turning HR on could mask the real stuff you want to see...

HiRes (ERES) is not same as Averaging mode...

AFAIK the HR modes are more or less implemented as follows (but please correct me if I'm wrong):

Pre-Filter --> Over-Sample --> Average --> Data-Thinning --> Filter

Not really.
That doesn't explain what is being processed and how.
In scope parlance Averaging is averaging same data points from different trigger events. Good only for repetitive signals. It does not lowpass filter signal.
Hires is a running average lowpass filter. It trading temporal resolution for amplitude one.
ERES is alternative implementation of lowpass filter with better impulse response..
There is no "data thinning"(we call it decimation) as a separate step anywhere, nor Pre-filter ..
There is also no oversampling but we simply sample as a scope normally would and than apply transformation.
On some scopes (LeCroy, some Siglent) ERES is available as a math function, so you retain original data and can apply it after the fact.

Search internet for ERES whitepapers by Lecroy. They explain it better than I ever could.

Thanks for the detailed explanation. I didn't know that stuff.

But the fact that it's a running avg doesn't cause the HR-processed signal to be kind of 'late' with respect to the real signal? (I'm talking about HiRes, not Siglent ERES)

[2N3055]

HiRes is only activated from 100 µs/DIV and larger.

Peter

Thanks! That work.
That means you can't use it to measure high frequency signals. I can't even trigger for single capture at this signal level combined with 100us timebase ...
Rigol 5K HiRes does not have this limit.

Robert

Robert,
see my previous response.
But shortly, Hires and ERES are lowpass filters by definition. So when you enable them you, by definition loose higher frequencies.
How much depends on sampling rate and length of convolution kernel in a filter. Some scopes specify improvement in virtual bits of resolution improvements. Some don't let you choose "how much" hires you get..

[2N3055]

I honestly think HR on scopes is overrated.. How many times you have to observe a perfectly repetitive signal upon which to oversample?
Many times you don't have a precise idea about what you are going to observe, and turning HR on could mask the real stuff you want to see...

HiRes (ERES) is not same as Averaging mode...

AFAIK the HR modes are more or less implemented as follows (but please correct me if I'm wrong):

Pre-Filter --> Over-Sample --> Average --> Data-Thinning --> Filter

Not really.
That doesn't explain what is being processed and how.
In scope parlance Averaging is averaging same data points from different trigger events. Good only for repetitive signals. It does not lowpass filter signal.
Hires is a running average lowpass filter. It trading temporal resolution for amplitude one.
ERES is alternative implementation of lowpass filter with better impulse response..
There is no "data thinning"(we call it decimation) as a separate step anywhere, nor Pre-filter ..
There is also no oversampling but we simply sample as a scope normally would and than apply transformation.
On some scopes (LeCroy, some Siglent) ERES is available as a math function, so you retain original data and can apply it after the fact.

Search internet for ERES whitepapers by Lecroy. They explain it better than I ever could.

Thanks for the detailed explanation. I didn't know that stuff.

But the fact that it's a running avg doesn't cause the HR-processed signal to be kind of 'late' with respect to the real signal? (I'm talking about HiRes, not Siglent ERES)

Late according to what reference frame?

[tautech]

(I'm talking about HiRes, not Siglent ERES)

Well Siglent do both in the DSO range.

[balnazzar]

Ok, with the Siggy 1104x-e, BW limitation always enabled.

These are 3 groups of 3 measurements. From top to bottom:

The groups are:
1. Sine
2. Staircase
3. Arbitrary EEG

Each groups is:
1. No ERES
2. ERES +0.5 bits
3. ERES +3 bits (maximum)

[balnazzar]

HiRes is only activated from 100 µs/DIV and larger.

Peter

Thanks! That work.
That means you can't use it to measure high frequency signals. I can't even trigger for single capture at this signal level combined with 100us timebase ...
Rigol 5K HiRes does not have this limit.

Robert

Robert,
see my previous response.
But shortly, Hires and ERES are lowpass filters by definition. So when you enable them you, by definition loose higher frequencies.
How much depends on sampling rate and length of convolution kernel in a filter. Some scopes specify improvement in virtual bits of resolution improvements. Some don't let you choose "how much" hires you get..

And indeed my measurements prove that. No staircase at 3 bits ERES.
Very different EEG at 3 bits.
At least the Siggy lets you select the whole range 0.5,1,2,3 bits.

[balnazzar]

Late according to what reference frame?

The real signal's one.

[balnazzar]

I'd appreciate a lot if you could make the same measurements with your instruments, when and if you have time to waste.

[skander36]

I can't even trigger for single capture at this signal level combined with 100us timebase ...

Here is a single trigger with the smallest signal where the frequency counter still shows correctly. The signal is from the internal generator with external 30 dB attenuator.

Peter

This is 10 Khz. My tests was about 8 MHz. But now I have succeed to single capture also at this freq.
Thanx!

[2N3055]

Late according to what reference frame?

The real signal's one.

Put your thinking hat on.
A question: what does scope care about that?

Scope has one reference frame: a trigger point.

Signal you see on a screen is delayed 10s of ms before you see it.
Scope can (and does) correct any timing of any transformation before it shows it on a screen..
So any filter, ERES or whatever will be shown in proper place in time.

[balnazzar]

Scope has one reference frame: a trigger point.

True. I was forgetting that.

[trp806mo]

SDS2104X in 10 bit BW20 MHz, one screenshot is "in live", the other in oneshot

[trp806mo]

same but in 8 Bits "live" & single

[trp806mo]

and my old TDS3034 (1mV is the lowest sensibility)