What puzzles me a bit is the -A version with 12-bit (4096steps) vertical resolution and how they could keep up the claimed 75.000 wfps. It is a whole lot
more data to be crunched versus the traditional 8-bit (256steps) scopes. Also they are asking only like 150 Euros extra for the 12-bit version.
12Bit 1GSa/s ADC12Bit ADC. 1GSa/s in 8Bit mode, 500MSa/s in 12bit mode.
8bit 1GSa/s ADC
Both give same amount of samples out.
What need hanle more?
Of course 12 bit math is some amount more than 8bit math. How much continuous math you need do? (there in real time stream) With fastest wfm/s and or fastest horizontal speeds there is high amount of blind time for math and other handling.
If wfm/s with 20ns/div is 75000 (just imagined, not true) whole waveform time is around 200ns (if 10div)
and 13100 ns is just blind time for next 200ns capture. Acquire engine do not need do lot of math, it can do with side engine, depending design.
Today ADC's, FPGA's etc have developed forward very fast. Where is problem.
I predict that there is coming many other manufacturers who follow Owon.
We remember time when SDS was launched. 1GSa/s entry level scope where was 10M full speed sampling buffer for both channels. Front end design was amazing good with totally different than example market "leader" Rigol DS1052E at this time. At this time Owon also use new analog circuit front end and true one chip expensive ADC (original AD or its RuiFeng clone). and this time most manufacturers use example 10 cheap 100MSa/s ADC's more or less poorly interleaved and even more they use 40MHz classified cheap chips of this ADC what was designed to 100MHz.
Later they (looks like) nearly copy this Owon SDS front end principle to Rigol DS2000. (2/1GSa/s version)
Now Owon make XDS3000A versions what have 12bit ADC. ADC12bit A model and ADC8bit without A model is only difference. Rest of scope can be in HW same design. (but this is only guessing based to long time experience and knowledge with this kind of things and also experience with Owon, Rigol, Hantek, Siglent - and of course western companies equipments)
Bottle neck have been ADC if think resolution and sampling speed. It is not anymore so much if we talk in under 4-5GSa/s range. Other bottle neck is memory speed (and of course data handling speed between ADC and meomory specially for digital side trigger process when this principle is used. Older times it was nearly only possible in cheap scopes to use analog side trigger. As example Rigol DS1000E, Siglent SDS1000, Owon SDS series... etc. In these things world have changed really fast. Digital trigger was available only in advanced more expensive scopes.
At this time (I have heard from little birds) there is only A models available from Owon.
Personally I hope I get one XDS3kA soon, so I can check what is reality and what was only quessing.
I have waited long time manufacturers start making low noise analog front ends and more than 8 bit resolution to nearly entry level scopes. Lowest V/div in XDS is 1mV/div. I do not know if it is BW limit forced and/or if it is pixel zoomed from 2mV/div or what it is. If it is low noise native 1mV/div sensitivity it is ok for true 12bit scope. Siglent have launched SDS1000X serie where is 500uV/div native sensitivity (claimed by Siglent).
Entry level to lover middle level scopes have fast developed better. Perhaps soon we see next jump to up. Not sampling speed and memory speed but now also resolution start rise from tens of years "normal" 8bit to higher. It may happend fast. It also depends how components prices come down from sky.
But parts of about Owon XDS is now speculation until truth can see with real XDS3102A on the lab table.