RTB2000, which has only 64 gradations, does not have such "steps". And it has significantly less noise due to the 10bit ADC.
SDS2000X+ is a very good scope for the money. But the intensity grade are not implemented well enough. Why claim otherwise? Let the manufacturer fix it, because this is only software problem.
Yes and no...
I was talking about front end (analog amplifiers before A/D converters) noise.
Despite 10 bit converter RTB2000 doesn't have less noise, but it has more resolution.
SDS2000X+ has low noise, combined with 8 bit resolution. Front end is quiet enough that you can see quantization steps on A/D converter.
RTB2000 has 4x more vertical points, and you cannot see quantization effects, because front end is noisier than A/D..
I'm not saying this is for sure a reason for this kind of rendering. But sure it can be. I don't have inside information and didn't design those scopes.
And not defending anybody, simply stating facts that it might not be that simple as it looks at first sight.
R&S is rendering to screen using some kind of pixel antialiasing algorithm. That is why waveforms look so smooth. Siglent doesn't do that. They scale and plot actual data points.
What is better? In the end it is philosophical (and aesthetical) question. One is "prettier" and looks "analog like". Other way might be considered true, and mathematically correct.
If you pull data from buffer, they will both give you clean data without any tampering.
I don't mind Siglents plotting. I know sampling can create artefacts (like moiré effect here) and don't care. I pretty much expect it..
If Siglent wants to spend development time on more eye pleasing CRT emulation, it is their choice.
I personally thing they should invest all the time they have enhancing and adding to analytic capabilities, which is where this scope shines in its price range.