For photographs, jpeg has really been optimized without reducing compatibility, and also in another less compatible way (incompatible viewers can display it without erroring out, but the colors are wrong) and there's such an encoder in the JPEG XL repo.

After selecting, each SIM slot is subject to inter freq / inter RAT reselection / handover.

Both are controlled by messages received from the tower (e.g. on 4GLTE, for reselection, System Information messages), though there is an additional constraint: what's supported by/enabled in the phone.

Perhaps one SIM slot was in the connected state and the other was in the idle state at one point. So the reselection logic applied for one and the handover logic applied for the other. There is for example a problem called ping pong handover. Once a phone is switched to a different frequency or RAT, the tower may have the phone be sort of stuck in the new frequency, until the conditions of the previous RAT or frequency improve substantially, in order to prevent the phone being like a ping pong ball between the two. This frees resources that would otherwise be spent on repeated handover-related messages.

Each frequency has its own signal strength (free space path loss, transmit power, one frequency might be on one tower and another might be on another, etc).

Signal strength is like the loudness of music being heard. It's possible for music to be quiet but otherwise excellent, or loud but low-quality. However, if it is too quiet, then the "music" becomes almost unintelligible, which the offseted bars should still be able to indicate.

In Wi-Fi, 6GHz and 5GHz are often used instead of 2.4GHz. 2.4GHz would likely win in signal strength. Yet, the others are used anyway, because the others are good for other reasons. However, if range ( ...or compatibility) is critical, then 2.4GHz is used.

Similarly, in cellular, there is a lower frequency e.g. band 8/12/14/17/20/28/71 and a higher frequency e.g. band 1/3/7/30/38/40/41/66/77/78. (Less basically, it can be more granular.)

So this sequence of events is possible: Tower switches the phone to a higher frequency -> speed increases but the signal strength reduces (confusing, but at least doesn't seem bad if there are 3 or 4 bars.) A switch to a lower frequency normally occurs instead if the high frequency signal is weak.

Cellular can be slow due to interference (maybe more common than signal strength issues; the metric to use instead might be SNR/SINR), congestion (maybe more common than signal strength issues; the metric to use instead is confusing, maybe the CFI value (if automatically changed) or RSRQ with a high SNR/SINR might rule it out), the speed of the rest of the network (the metric to use might be RSRQ during a download with a high SNR/SINR), data plan (the metric to use instead might be RSRQ during a download with a high SNR or SINR/QCI (requires interpretation)), and the width of it (the metric to use is BW). So it's confusing, and not exactly that full bars are always better.

For 2G, with each nearby cell (coverage area) basically getting its own channels, signal strength might've been more important, though interference was there somewhat (so there was MAIO planning etc.)

But aside from speed, there's the battery to consider. If the signal strength from the tower to the phone is "satisfactory", it's implied that so is the signal from the phone to the tower, so the phone will have to have an elevated transmit power.

Let's say with certain current there is 0.687V. If two diodes are connected in series i.e. (point a) → diode 1 → diode 2 → (point b), and each has a 0.687V voltage across, that's 0.687 + 0.687 = 1.374 V between point b and point a.

For the solar diodes, the "current" depends on how "strong" the sun is. If at a certain "current", across each diode is 0.687V, you'll need 216 diodes in series (between two points) to get 0.687×216≈148.4 V.

If there is 0.002 V increase per diode per 1°C decrease, with 216 diodes, that's a 0.002×216=0.432V increase per °C decrease, so with a 4°C decrease it exceeds the MPPT's limit.

Another thing about solar that differs from how diodes are "normally" used in circuits is that the "true" voltage depends on the max current achievable with the how bright the sun is right now, instead of the "true" current. When the "true" current is 0 A, the voltage across each diode might be 0.687 V. When the "true" current is 0.5 A, maybe 0.65 V. 1A, maybe 0.6V. 2A, maybe 0.3V. Try to get more "true" current, the "true" voltage drops. Try to get more "true" voltage, the "true" current drops. Power is voltage × current so when full speed charging, the MPPT uses an algorithm to find the (possibly) best minimum (not maximum) input voltage based on the temperature etc and trades voltage for current at the output. If there is no minimum input voltage restriction, solar will follow the battery's terminal voltage + cable drop instead, and instead of something like 111V, there could for example be a ~4x less powerful 25.5V (if the battery is a "24 V") with just 10% more current.

At the MPPTed min input voltage for full speed charging, maybe 111V, all might seem well even with low temperatures, but when the battery is full and there is approximately nothing using electricity from solar, the real input current will be ~0 A, so there will no voltage "sag", so the solar will realize the full voltage corresponding to the temperature and the illumination, potentially >150 V...