Why do chips not always "meet the grade"?

During manufacture, integrated circuits are tested at varying frequencies and temperatures to categorise them into speed grades. However, why don't all ICs come out the same and work the same? They all come from the same photolithographic mask, right? Am I missing something?

Modern ICs are REALLY small. Tolerances are huge during processes such as ion implantation and oxide growth. At such small sizes, these things can't be treated as anything but a probabilistic process. Lines also tend to be smeared due to the feature size being the minimum possible given the wavelength of light. When you get worst-case performance in a bunch of these different steps, then you get a non-functioning IC.

Companies don't design the IC so that it functions at the worst-case - it would be too costly. So instead they do Monte Carlo simulation of the manufacturing parameters, estimate a yield, and do testing after the fact.

Typical "design corners":

Source: What I remember from my IC manufacturing class.

• Added what I remember from my IC manufacturing class ;) Jan 11, 2011 at 0:56
• @tyblu: That graph means practically nothing to those of us who haven't taken an IC fab class (doesn't even seem familiar from VLSI) Jan 11, 2011 at 1:39
• @NickT, A manufacturing window within which transistors fall. Variations in the (at least) 2 doping stages make the 2 types of transistors either 'fast' or 'slow'. Will clarify edit when brain is less fuzzy. Jan 11, 2011 at 5:57

There isn't really a "worst case" - manufacturing IC's is a statistical process, there will be some (small) percentage of transistors that are a very long way from typical speed (imagine a statistical distribution curve drawn here). The distribution of speeds for the two types of transistors in CMOS (NMOS and PMOS) don't correlate well.

Thus, they pick those 4 corners: Fast/Fast, Slow/Slow, Fast/Slow, Slow/Fast out of their two transistor speed yield curves.

The further out from typical they choose to make the corners, the higher their yield, but the more difficult it is to design. If the corners are too far apart, design for operation at the 4 corners takes too much development time and can increase the die size. Increasing the die size will decrease yield.

The four corners often form more of a parallelogram than a rectangle. If both types of transistors, across one whole die, are all very fast, the part will probably work, maybe well beyond the rated speed, and same for both transistors getting slower - the part will work, but at a very slow speed. The difficult corners are the fast/slow and slow/fast.

The design is indeed fully simulated at typical/typical, and the four corners. Monte Carlo simulation is used to check some of the intermediate combinations.

To increase yield, they can sort the die into bins after manufacture and thus sell the slow/slow parts that would otherwise be thrown away, and sell the fast/fast parts for a premium.

And yes, some manufacturers will use fuses to restrict a part to a certain speed grade. Because manufacturing yield curves and market demand being uncorrelated functions, sometimes they have to down-grade parts.

Mostly what krapht said. I'd add that parts are getting so small that even without variation in geometry, the number of doping atoms per transistor are also becoming small, so simple statitical variation means some will be faster and some lower, and the gain also varies. You'll end up with some that either can't work, or can only work at say a slower clock rate. Also you might see some computers offered with say a three-core chip. Such a chip is probably a four-core chip, but a faulty core has been disconneted in postprocessing, and the chip sold as a cheep 3banger. I would add, nothing wrong with that, if that's the capability you need, then the manufacturers bad luck is our gain.

IC chips are just shrinked to the point where they "barely" work.

You can make chips on 3um process with nearly 100% good rate and very little variation, but it's just too expensive (and slow).

• 3um? I read that the newest Intel process is 32nm - Yes, nanometers! I understand they probably have a lot of scrap, and at $300 a pop (or$1000 for the extreme series) they can afford to work slow, but I'm not understanding your point here. Jan 12, 2011 at 23:53
• @reemrevnivek: The cost of producing 1 wafer's worth of chips is (roughly) constant. Yes, a cutting-edge 32 nm design has a lot of scrap. "The same" chip design printed 100 times larger (on a 3 um process) gives nearly 100% yield, but a given wafer will produce 1/10,000 as many chips, so the result is even more expensive per chip. BarsMonster seems to be pointing out that shrinking the design to some point in the middle -- such that most chips "just barely" work and quite a few don't work at all -- produces the most good chips per wafer, and so the lowest cost per chip. Apr 2, 2011 at 3:41

For analog functionality, sometimes the chip can be intentionally hobbled and the same die sold cheaper with not-so-good performance. Blowing on-chip fuses to reduce bias currents (for example) can reduce the overall performance of the chip.

For digital chips that are required to run at speeds approaching the capabilities of the manufacturing process, final test will weed out the best performers from the not-so-good performers and 'bin' them accordingly. Differences in speeds are usually caused by statistical variations in the manufacturing process.

• Without some evidence, I'd be hard pressed to believe that manufacturers would design in adjusting circuitry and take the time to blow fuses. I'd be much more inclined to blame plain old process variation than some sort of chip-hobbling conspiracy. Also, reducing bias currents is usually a good thing. Jan 12, 2011 at 23:55
• It's quick enough to blow a fuse in final test. As for using the same hardware for different performance grades intentionally, this happens, and not just with chips: technabob.com/blog/2010/12/27/… Jan 13, 2011 at 0:24
• Certainly back when Intel sold 386/486 chips with and without an FPU, they were from the same die: ones where the CPU worked and the FPU didn't would simply have the FPU disabled. Dec 21, 2012 at 9:43