If it's 0°F outside and 70°F inside and you have a single sheet of glass filling a hole in your wall, that single sheet of glass is going to have an inside surface temperature of just about 14°F. In those conditions it is almost a certainty that the sheet of glass is going to be covered with frost and ice if the house has anything approaching a "normal" level of indoor humidity. If that single sheet of glass isn't covered in frost and ice then it typically means that the hole-in-the-wall, despite being filled with that sheet of glass, is also likely to be leaking warm air and moisture so badly that the inside of the house is pretty much as dry as the proverbial bone.
Eventually people discovered that adding a second sheet of glass to the outside of the original sheet of glass was a very good idea because, among other reasons, it increased the inside surface temperature of the original sheet of glass to just about 43°F, a definite improvement over the single piece of glass. This was born the window / storm window combination.
Eventually (again) people discovered that they could get the same performance as the window / storm window combination in just the window alone if they stuck two sheets of glass next to one another separated by an airspace. Someone coined the term Insulating Glass Unit or IGU for this option.
To repeat, the dual pane, thermopane, IGU, and so on was developed to match the performance of a single pane with storm without the inconvenience of the storm window.
Eventually (once again) some really smart people discovered that placing a layer of microscopically thin metals on the surface of the glass improves the inside temperature even more. This metallic coating was called Low Emissivity or LowE, mostly because that is a description of what it does. As time went on these same and other really smart people developed all sorts of different LowE coatings for different applications in different environments. And somewhere in there someone discovered that filling that space between the two panes of glass (also called "lites") with an inert gas, rather than just air, resulted in even better thermal performance.
So while the original IGU was developed to equal the single pane window/storm combo, the addition of LowE coatings and inert gas to the airspace (argon, krypton, xenon, and anyone want to name the next gas in this sequence?), improved performance far beyond the original.
Insulation is all about air. Air insulates. Whether you fill a wall cavity with foam, cellulose, fiberglass, rock wool, old jeans, or whatever, it's trapped air that is providing the insulation. The better you trap the air the better the insulating performance of the insulation. If you want to degrade the ability of a material to provide insulation, then make the air move. When dealing with insulating ability static or dead air is good while moving air is bad. This is true inside your wall and this is also true in your window.
Your home loses heat thru your windows in three ways: Conduction, convection and radiation.
Imagine a sheet of Styrofoam filling the hole-in-the-wall rather than glass. If it's 70°F inside and 0°F outside then nature is going to be unhappy. Nature likes things in balance, nature doesn't like things to be different, so nature wants the temperature on both sides of that sheet of Styrofoam to be the same temperature.
As Gartheron pointed out so well, if you aren't constantly adding heat to the air on the warm side of the sheet of foam, eventually the two sides will reach equilibrium as the heat conducts thru the foam. R-value is resistance to heat flow. Basically the R-value tells you how fast the material will pass heat - not cold.
Radiation is what you feel when standing in the sun. Solar heat, as we feel it, is a combination of the longer wavelengths of visible light plus infrared. We don't feel UV or the narrow wavelengths of visible light as heat even though UVB is what causes "sunburn". Radiant energy results in about 60% of the heat loss thru a clear glass IGU.
Since air insulates then why doesn't a single pane and storm combination with a 4" airspace have better insulating performance than an IGU with only a 1/2" airspace? That's where convection comes in.
Convection occurs when you have warm air over here and cold air over there. That discrepancy really annoys nature so nature tells the warm air to go over and get cozy with the cold air until everything is equal. And believe it or not, that is the exact definition from thermodynamics 101.
Clear glass IGU filling a hole in the wall. Inner glass is facing a 70°F room and outer glass is facing 0°F outside air. The inside lite is going to be warmer than the outside lite. The air trapped between the two lites is going to be warmer where it contacts the inner lite because it is gaining heat from the warmer glass. Air against the outer lite is going to be colder because it is losing heat to the colder lite.
Basic physics, which everyone reading this already knows, warm air rises and cold air falls. Inside the IGU airspace the air touching the inner glass is warmer and it is rising while the air touching the outer lite is cooling so it is also falling. Air on the warm side rising and air on the cool side falling results in the formation of a convection loop inside our IGU airspace. The air is moving which is bad for insulating ability. We add argon, or krypton, or xenon and we slow the current. This improves window energy performance because we are transferring less warmth from the warm side to the cold side due to the convection loop, which also raises the inside glass temperature by a few more degrees.
So imagine a single pane window/storm combo. Wider airspace means much greater convection current than the narrow airspace in the IGU, ergo more heat loss, ergo less efficient despite having the wider space (more air), which results in comparable performance numbers with the clear glass IGU.
While trying to explain a few of the dynamics of window performance I think I really got off track from the original questions/concerns (there were already several really good replies to that in the thread of course).
Since this thing is really long and I am really tired I am stopping here, but if there is any interest in more in depth discussion of some of the stuff I mentioned previously or moving on into triple panes, and so on...more than happy to keep going.
Also since I am too tired to proofread this thing tonight I really hope that I haven't left any glaring errors.
Regards,