This is completely true in that one benchmark. That one benchmark is the only one I was interested in, because I was discussing the validity of the benchmark conclusions, not their truth. In general, given a proposition P, it shows the conclusion Q. (If P, then Q.) For this type of logical argument, if P is true, then Q is true. If P is not true, we know nothing about Q. Tom presented data, and then made a conslusion. If we say that "P" means "DDR produces better benchmark result data" and "Q" means "DDR is faster", then this is a valid argument.
We do not yet know if "P" is true, but we can already see that the form of the logic is valid. If "DDR produces better benchmark result data", then "DDR is faster". I will not argue with this. It's totally valid. This however, says nothing about whether or not P is in fact true. We must go to the table of data he presents in that benchmark. Looking over the table we see that 100MHz DDR performs the same as 100MHz SDRAM. Hmmmm. We see he has numbers for 133MHz DDR, but no numbers for 133MHZ SDRAM, so we cannot say anything about DDR in that case. Results are inconclusive there.
Based on the data, we cannot say he showed "P" to be true. I am also not saying I believe "P" to be false. I in fact believe "P" actually to be true, but we must go only on the data shown in Tom's table for his benchmark. So whereas we cannot show "P" to be true (we cannot show that DDR produces better benchmark result data) for that benchmark, we cannot show "Q" to be true (we cannot say that DDR is faster.)
However, Tom jumps right to saying that "Q" is true (DDR is faster.) He does this without citing any supporting evidence. He certainly can't use "P" as evidence. The data in his tables doesn't support it. Therefore, his argument is logically invalid. Because his argument is invalid, his conclusion means nothing. It may be true, it may be false. But, it's not supported anywhere in his benchmark or argument.
Later, in other benchmarks at other sites, we find out that DDR is faster than SDRAM through some good data and valid arguments. This does not change the fact that Tom's logic was flawed. The flaw in his logic for this benchmark is what I was presenting as evidence that raw data should be preferred over merely taking a reviewer's conclusions at face value.
-Raystonn
-- The center of your digital world --
We do not yet know if "P" is true, but we can already see that the form of the logic is valid. If "DDR produces better benchmark result data", then "DDR is faster". I will not argue with this. It's totally valid. This however, says nothing about whether or not P is in fact true. We must go to the table of data he presents in that benchmark. Looking over the table we see that 100MHz DDR performs the same as 100MHz SDRAM. Hmmmm. We see he has numbers for 133MHz DDR, but no numbers for 133MHZ SDRAM, so we cannot say anything about DDR in that case. Results are inconclusive there.
Based on the data, we cannot say he showed "P" to be true. I am also not saying I believe "P" to be false. I in fact believe "P" actually to be true, but we must go only on the data shown in Tom's table for his benchmark. So whereas we cannot show "P" to be true (we cannot show that DDR produces better benchmark result data) for that benchmark, we cannot show "Q" to be true (we cannot say that DDR is faster.)
However, Tom jumps right to saying that "Q" is true (DDR is faster.) He does this without citing any supporting evidence. He certainly can't use "P" as evidence. The data in his tables doesn't support it. Therefore, his argument is logically invalid. Because his argument is invalid, his conclusion means nothing. It may be true, it may be false. But, it's not supported anywhere in his benchmark or argument.
Later, in other benchmarks at other sites, we find out that DDR is faster than SDRAM through some good data and valid arguments. This does not change the fact that Tom's logic was flawed. The flaw in his logic for this benchmark is what I was presenting as evidence that raw data should be preferred over merely taking a reviewer's conclusions at face value.
-Raystonn
-- The center of your digital world --
Your comparison of the 100 mhz FSB of DDR and SDR have another explanation. At 100 FSB there is not enough bandwith available to the processor itself to take advantage of the increased performance of DDR RAM. That does not mean that DDR has no benfits period, only that at 100 mhz fsb it definatly is not needed. But we all already know this and seen this demonstrated when RDRAM and the p-3 were coupled.
Facebook
Twitter
Instagram