Firstly, there is no real reason for "min" value as it is always zero.
Of course there is a reason for min value. It's not always zero.
You are not thinking this through. If for example you have constant traffic between 2 and 10 mbps, then the min value for that traffic at that time will not be zero but 2mbps.
I haven't seen such chart in winbox and on purpose I tested constant 20Mbps (it of course varied between 19-20) traffic for over half hour to verify, if only max value is adjusted or if also min value is adjusted. I would not write such claim without checking it first.
If you have different expecience, please let me know in which window you observed it. also let me know if there were some special conditions. Also screenshot would be lovely.
here is detailed description:
btest, TCP, both, local TX 30M, remote TX 20M. After few secs, without stopping, change local TX to 22M and hit "run" again. Thanks to this, you will notice scaling change, once the first few seconds are discarded from memory and that means there are no more "zero" values displayed on the chart.
Based on your theory, min value (lowest pixel of the chart) should be approximately 20M while max value (highest pixel of the chart) should be 22M. But that is not what we observe:
2018-07-21_1043.png
As you can see, traffic has no drops and is constantly on TX22M/RX20M with slight variations. It was running like this for almost half hour (since I replied till this edit)
Based on your description, scaling of chart should ajdust the way that we will see almost no red (RX) line (just a couple of pixels on bottom) because min value of chart will be close to 20M while blue lines (TX) lines will fill almost whole area, because they are close to 22M. And that is clearly not happening.
Actually, if you download the image, zoom it in and start counting pixels, you will find out that difference between blue and red lines is about 6 pixels. That means 6px represents 2M difference between our RX and TX. 1px therefore represents 2M/6=333k difference approximately. Whole height of blue lines is around 73px which turns into 73*333k=24M Whole height of red lines is about 67px which turns into 67*333k=22M There is clearly some imprecision but in overal, you get the picture, right? Lowest (min) point of chart is definitely zero. Always. It does not adjust similarly to Highest (max) point.
feel free to bring some opposite proof
I am always happy for constructive discussion. But stating claims without proof is just waste of time.
You do not have the required permissions to view the files attached to this post.