Q9 - Computer scientist: For several decades

[Laura Damone]

What does the Question Stem tell us?
"properly inferred" = Must Be True

Break down the Stimulus:
The first line tells us that the number of transistors on a microchip has been doubling every 18mo, and that "hence" the computing speed has also been doubling. This implies a conditional relationship between the number of transistors and the computing speed of a chip. The next line tells us that since the mid-90's, doubling the speed has been accompanied by a doubling of the cost of production: another conditional relationship.

Any prephrase?
Although there is only 1 conditional logic indicator word in this stimulus ("each"), there seems to be a transitive property at work here: 2X Trans --> 2X Speed --> Cost Up

Correct answer:
D

Answer choice analysis:
A) A classic Inference question trap answer. Just because that's one way to do it, doesn't mean it's the only way.

B) We have no reason to believe that's true. We know the cost of production doubled. Why would that imply the cost wasn't passed to the consumer?

C) Just because cost of production has been doubling alongside speed doesn't mean that no one has been trying to control cost of production. Maybe they've tried and failed, or maybe this is the result of those efforts. We just don't know.

D) Correct! If we know each doubling of the number of transistors leads to a doubling of computing speed, and each doubling of computing speed since the mid 90's leads to a doubling in cost of production, we can infer that each doubling of the number of transistors leads to a doubling in cost of production. Don't be mislead by the order of the concepts in D--it can look like a reversal if you're not careful. Upon close inspection, however, you can see the indicator word "each" which functions like "all" and introduces the sufficient condition, so it is the increase in transistors that guarantees an increase in cost, not vice versa.

E) No way. There is nothing that would allow us to infer something's future likelihood. This is one of the less common LSAT fallacies, but not so uncommon that it doesn't bear mentioning: the past doesn't predict the future.

Takeaway/Pattern: If a Must Be True question has any conditional indicators, dig deeper and see if you can spot any classic conditional logic patterns. Conditional relationships are all about guarantees, and these can be expressed without the classic indicator words.

#officialexplanation

[jardinsouslapluie5]

I can't visualize the stimulus.
Could you explain, please?
Thank you.

[timmydoeslsat]

The first thing we want to do is identify the question stem, which is a must be true. A very high standard to meet obviously.

So for many decades the number of transistors doubled on new computer microchips, which had the effect of doubling the computer speed basically every 18 months.

But, we know that from the mid-90's into the next decade, we have had a doubling in the cost to produce the microchip that doubled in speed.

Answer choices:

A) We know doubling the transistors can double the speed, but we do not know that this is the only effective way. Classic wrong answer for a must be true question stem.
B) We do not know if there was in fact an increase in the retail cost of this chip. We know about the production of the microchip, and since it doubled, if anything this would seem to be very much a pipe dream.
C) Last several decades? Computer scientists did not focus on the cost of production? No way can we infer this.
D) This is literally what the last sentence says except it plugs in the transistor idea for computing speed, but we were given this lock-step union in the premise (first sentence).
E) Unlikely? Ever? No way on a must be true question. Who knows what the future holds.

[ohthatpatrick]

Informative AND snarky! My favorite combination.

[a3friedm]

This question got on my nervous durring my practice test this morning. Answer choice (D) seemed too good to be true in that it basically restating the premises, so I wound up waisting an extra thirty seconds trying to find a way out of it.