The intercooler on the car above looks to be relatively inefficient which certainly explains the size. I can visually see the individual crimps in the external channels which means fin density is fairly poor. In addition the internal channels appear to be rather large. What this means is that for a given "B" (Height) measurement you have few internal channels than a top dollar core. This means less flow for a given frontal area. None of this matters as that core, even as inefficient as it is, will probably cool down around 300hp, but it does illustrate how companies base the HP claims only on size.
A few things about cores themselves. On a cross flow design you have three measurements. "A" is width, "B" is height and "C" is length. This may not be industry standard but it is how they are used to at the shop.
The width (A) is important that it not be so large that airflow stalls somewhere halfway through the core, resulting in lower efficiency at the backside of the core. An example of this would be a 6" "A" measurement on a street car. In addition, as the air passes over the fins it heats up, so by the time it gets to the back inch or so, it's not cooling anything anymore. The largest "A" measurement you would want for a street car is probably 4-4.5". I'd say optimally for cost and efficiency, 3-3.5" is ideal in most cases.
The "B" measurement basically determines your frontal area and is usually restricted by the space available. Remember that intercoolers don't have high efficiency as radiators. The air passing over them is what counts, so an external channel hidden or blocked off behind a bumper beam is a hot channel. When I build my custom cores, I usually try to put external channels on the top and bottom plates so that every internal channel has two external channels to cool it. This measurement is probably the most important as it dictates the overall number of external channels.
The length "C" is an interesting one because of the way people look at it. They always assume longer is better. That is not a pun. Longer is better to a point, also not a pun. Lets say I have a 24 inch long core and my overall efficiency rating sits around 90%. By increasing the length another 4 inches, you won't see a proportional increase in efficiency. It might be closer to an additional 1%. So for an increase in length of some 15-20% you get a net result of 1% efficiency. However, if you were to take that 24 inch core and drop it down to 18", the next lowest common size, you might see a drop off in efficiency of 5-15%, throwing a number out here but you get the idea. Length is not proportional to efficiency.
The measurements of a core are not just guesswork. Larger is better to a point. At some point all three measurements can begin to hurt you.
The intercooler above does have what I would call optimized inlet placement, however the upturn at the inlet side is a bit of a power robber. It would be better to make the inlet turn an immediate 90 degrees(Mandrel) out of the end tank rather than drop down and around. Diagonally opposed inlet/outlet placement is best on cross flow intercoolers with no internal baffle to redirect flow. With an internal baffle the diagonal placement is less important.
To illustrate the point about the efficiency model of fin density, baffling and turbulators, I have a core which I would say is probably equal to or more efficient that the core posted above. I cannot see all aspects of the above core but if the fin density, channel size is an indication of the rest of the design....
Then this core...
...is more than likely more efficient and is 3.5 x 6 x 18. That is the qualitative difference between Ebay/GhettoChrome cores and top quality cores such as API. It is literally that much.
/brox