Scope memory depth and low frequencies

[JSE009]

I am looking for a new scope (<€1.500,=) and the use is especially measurement of audio signals but I use more and more uControllers (I2C etc.).
So for audio it is the low frequency spectrum and for uController the more higher frequency spectrum but I think a scope with 70Mhz should be more than enough.
Now I was thinking “What about memory depth of a scope and measurement of frequencies?”

When looking around you see several scopes with a wide range of memory depths and what wonders me is that sometimes to low-end market
scopes have more deep memory then high-end market scopes.
Now it could be that high-end brand wants to enter the lower market and designed scopes with less deep memory to be price competitive?

- Rigol DS1000Z series with 12Mpts and 1GSa/s
- Siglent SDS2202X-E series with 28Mpts and 2GSa/s
- Keysight DSOX1000X series with  1Mpts and 2GSa/s
- Rigol MSO5000 series with 100Mpts and 8GSa/s
- R&S RTC1000 series with 2Mpts and 2GSa/s

All as standard but most are upgradable (options)

When browsing white papers, you can find: Record length = sample rate × acquisition time

When I look at this formula the use of more memory depth and high sample rate for especially low audio frequencies is essential?
So a scope with for example 1M memory and a sample rate of 1GSa/s (take the Keysight DSOX1000 series) is not a smart combo because the sample rate will drop very fast when
measuring low frequencies and zoom into this?
Am I right in this or am I missing something? 

Happy when specialist share thoughts on this 

[apblog]

you are correct, but it's not that big of an issue for a single capture at a given sweep rate.

However, a small memory size dramatically limits your ability to zoom in.  For working with digital  circuits, being able to capture a long train of pulses and then zoom in to decode them is very convenient. 

I have a keysight 3000T with 4MB memory and even that is limiting.

Also, be sure to look at the midrange Instek scopes.  If I recall, there is one in particular that has an excellent FFT function, which is very useful for audio work.  My impression is that they are less buggy than Rigol or Siglent.

[ADT123]

When looking at audio signals you dont want to be sampling at 1GS/s - assuming  20kHz max frequency then there will be little to gain by sampling faster than 100ks/s (5 x oversampling).  If your scope has say 10M memory then it can capture audio for 100 seconds and you can zoom in to see full details.

When looking at your 70MHz signal you will be sampling probably at 500MS/s so you can capture for 0.02 seconds (10/500).

Both may well be OK so you have a wide choice of scopes but if you want to measure a couple of seconds of audio on one channel and the 70Mhz signal on another you will need a lot of memory - say 1GS (1 billion samples).  That will really limit your choice of scope.

Another factor to bear in mind is that for audio signals you may well want a scope with >8 bits resolution to ensure you have a good dynamic range in scope and spectrum mode.   Most scopes are 8 bit but quite a few are 10 or 12 bit.  Higher resolution comes at a higher price.

[JSE009]

I also thought about using a scope wit more then 8bits resolution and when browing the web I found the Owon XDS3064AE:
4 channel DSO
60 MHz bandwidth
1 GSa/s sample rate
14-Bit resolution
40 MSa memory depth

Normaly I would not look directly for an Owon because in the past there quality was not that good, but when looking at the price (€600,=) for this scope it could be a nice asset on the workbench
measuring low frequency signals?
I think that a brand like Owon will improve in quality over the years?

[rf-loop]

I also thought about using a scope wit more then 8bits resolution and when browing the web I found the Owon XDS3064AE:
4 channel DSO
60 MHz bandwidth
1 GSa/s sample rate
14-Bit resolution
40 MSa memory depth

Normaly I would not look directly for an Owon because in the past there quality was not that good, but when looking at the price (€600,=) for this scope it could be a nice asset on the workbench
measuring low frequency signals?
I think that a brand like Owon will improve in quality over the years?

Owon HW building quality is not at all as bad as many may think (partially due to some misleadings and partially due to some mistake Owon made in history and how they handle it).
I have some real life experience with Owon and I must say that mostly they have been amazing reliable overall related to price. (my experience is mainly from xxx numer of older SDS7k series)

But then, data sheets, specially Owon data sheets,  need read carefully! 

For Audio and other LF things this do not matter so much but 14 bit is available only with max 100MSa/s.
If look audio etc and want FFT, it use only 8k data points.
But it have also some nice things, like example digital filters and really good TFT screen in this price group  (As always with Owon, of course because it is part of Lilliput company who knows something about displays... ), battery option,
I do not have any kind of idea how its 8, 12 and 14 bit modes real ENOB is. In any case, very far from the "resolution bits" (as always and not only Owon).
I have bit older one XDS3102A (8/12bit model). (here in forum can find some images)

I hope that I will not be punished for hanging out, even though I will not bark and shout Owon to a completely worthless trash.
Advantages and disadvantages are different settings for different users and needs.