first time here. Today, I was looking on my motherboard/graphics card and I noticed tiny little pin width points of solder with nothing soldered to them. The distance between most of them are just right for a small rectangular capacitor or resistor, but I can't seem to think of why the manufacturers would put the solder there yet not do anything with it. Just curious, thanks.
There could be several reasons for having solder pads that aren't used:
- Test points - these are pads that are used to connect test equipment to the board to test it in the factory.
- Unneeded parts - the board was designed, then it was discovered that you could leave out some of the passive parts (resistors, capacitors, inductors) with out causing problems. Most often these are parts that were intended to reduce emmissions (radio frequency "noise" that can interfere with other devices.) Testing showed the whole board to be "clean" enough that they could be "dirtier" and still meet all requirements. Leaving out specific parts can reduce the price of building the board. They don't bother to change the board in such cases. Alternatively, the plans include extra parts to remove emmissions, and these are only used if the board isn't clean enough.
- The board can be used for multiple models. It has sub-sections that can be left off to produce a cheaper board with less functionality. You design one board that can be populated in different ways to produce different models. In these cases, there may also be parts whose only job is help the electronics detect which sections can be used. These may be as simple as a jumper or as complex as a read only memory device containing a list of usable sections.
There are probably other reasons as well. These are the ones that pop into mind.
It is not at all uncommon for parts to be in a schematic, yet not be fitted on a PCB; this is usually not because it was found to be unnecessary in many cases.
As an example, it is not uncommon to have extra positions for decoupling capacitors, just in case the total electrical noise in some part of the PCB is higher than expected.
It is also not unusual to have strapping option resistors for a device such as this one. This is where a pin (or group of pins) are tied either high or low (and in some cases floating) and will be sampled at reset to set certain functionality for the device. It is not uncommon for all possible options to be on the schematic (i.e. both a pullup and a pulldown are provisioned) but only a subset of those are actually fitted.
Many of those pads may also be test points for ICT.
In other cases, different variants of a PCB may utilise a different mix of components; this is alluded to in the comments. It means a manufacturer has a single base PCB and simply loads a different mix of components to have two different resulting items.
There are more reasons, but the above are certainly very common.
"Tiny little pin-width points of solder" might be plane interconnects. It's also common to plan for a lot of blocking capacitors on the supply lines and then figure out in testing that you'll get along with a small and/or strategically chosen subset just fine while using circuits from a particular supplier.
Then there are also testing pads for measurements (hard to get at signals otherwise on a modern PCB) and/or temporary test configurations (you actually solder "0 Ohm" "resistors" across them to activate) or for permanent configuration.