I find this talk about fair and unfair comparisons strange, I'm not sure what some people are trying to say. Maybe we can clear this up a bit.
There doesn't seem to be, in principle, a reason why we can't compare anything if we provisionally define "compare" as examining and noting the differences and similarities between two objects. I can do this with ANY two objects, so the question is not whether such a comparison can be made but whether a comparison is useful. Comparisons, in the sense we are talking about here, are goal-oriented.
The goal is to give us some insight into the similarities and differences of video cards. Presumably, one kind of insight will be to allow me to determine which one to purchase. A comparison between a very old video card and a very new one can give us relevant information on the increase in computation power over time, price relative to historical power, heat, noise and other relevant metrics - the one thing that it wouldn't give us relevant information on is the question "which one should I buy?" since no one is going to buy the very old card.
Point so far: There are relevant and non-relevant comparisons relative to the questions we want answered.
What people are claiming is either:
1. A comparison between a 480 and a 5970 provides us with NO relevant information
OR
2. A comparison between a 480 and a 5970 provides us with relevant information
Now, if we take position #1 to be a universal, it is obviously false. A comparison between the two cards give us lots of currently relevant information. I won't enumerate some of those items since they should be obvious.
Position #1 does make sense as a non-universal claim and this is what most people are saying. The 5970 and 480 are in different price brackets and so there is no relevant comparison in order to provide information in answering the question "which one should I buy?" or some similar question. The assumption being that a consumer first decides on the $ amount they wish to spend and then looks for the best product at that price level. This obviously happens quite a bit. I budget many purchases based on a pre-set price limit relative to my means and don't bother looking at more expensive options.
Alternatively, if the ceiling is the limit in terms of price and all I want is the best thing going, then the comparison is relevant. I want to buy the fastest card and so I need to know at minimum that the 5970 is faster than the 480. The comparison is valid insofar as it directs my purchase properly. The higher price gives me good reason to assume the 5970 is faster, but before dropping $700 on it, I want hard proof: I may have a deep wallet, but I'm not stupid enough to buy a 700$ if the $500 card is faster for some reason.
Finally, I may have decided on a particular $ amount as a price limit but might be willing change my mind. I go to buy a product with the intention of spending 100$ but find out that I can get a significantly better product (in terms of some relevant feature or performance) for 50$ more. The question is "does the added incentive (feature, performance) justify my going over-budget?" In some cases the answer will be yes, in others, no. The only way for the consumer to decide is if the relevant information is provided in a direct comparison.
So neither position #1 nor #2 is completely accurate as a universal rule. There ARE situations that seem clearly relevant to having a comparison between the 480 and 5970 in order to direct me on relevant questions relating to which I should buy. That this is not true in all cases does not make it an "unfair" comparison.
The information should be out there for the people who would find it relevant.
Facebook
Twitter