Quartz clocks and watches are everywhere these days. They're accurate, relatively inexpensive and they will run for years without a battery change. At a minimum, even the least accurate quartz clocks are accurate to within less than one minute per month. But some are much better than that. While many of my Casio watches are said to have an accuracy of plus-or-minus 30 seconds per month or better, I had never formally tested them. In this video I test 56 Casio watches (and one Seiko watch), using only their own quartz accuracy. (I turned off the automatic reception on the watches which can otherwise set themselves to atomic time.) At the end of one month I knew how far they had drifted from the real "atomic time" that they all showed at the beginning of the month.
The watches I tested include some from the 1980s and from the 1990s. But most of them are less than 15 years old and the vast majority are less than 5 years old.
I tested 12 of the AWG-M100 watches, each using Casio's Module 5230. The best of them were less than 4 seconds fast at the end of the test and the "worst" were only off by 11-13 seconds.
I tested 7 variations of the WVA-M630 and WVA-M640, using Casio's Module 5161. The best were off by 7 seconds and the "worst" one was off by 15 seconds (just barely within specifications, since Casio says they should be accurate to plus-or-minus 15 seconds per month).
I tested 9 watches from the GWM-500 and GWM-530 model line, using Casio's Module 3405. The best of these was off by just 1.7 seconds after one month! The "worst" was 11.5 seconds off.
I have 6 G-Shock "Squares" with Multiband 6 "atomic time" reception, using Casio's Module 3159. These include different colors of the GW-M5610 along with the more expensive GW-S5600 and the top-of-the-line GW-5000. They performed well. The best ones were each off by only 3.5 seconds at the end of the month-long test and the "worst" was only off by 8 seconds.
I tested the GW-9110 (Module 3217) and it was off by 3.5 seconds. The similar GW-7900 (Module 3193) was off by 5 seconds.
My most basic Casio "Waveceptor" watches are both the resin and metal versions of the WV-58A (Module 3054), the metal and resin version of the WV-200A (Module 3140) and the WV-M60 (Module 3456). Casio says they should all be accurate to within 20 seconds per month. The "worst" of them was still only 13 seconds off. But the very best of them was only off by 1.25 seconds! And it's not even a very expensive watch!
But the most delightful news to report is that the very "cheapest" Casio watches in my collection performed so well. They are variations on the "Casio Royale" watch using Casio's Module 3198 and Module 3299. The "worst" of these was slow by 12 seconds at the end of one month, but still well within the margin of Casio's claims that they would be accurate to within 30 seconds. And the very best of them was the AE-1200WH-1A or the all-black, resin version of the "Casio Royale" using Module 3299. I bought it a few months ago for just $10 at a seasonal sale at Wal-Mart. And, at the end of one month, it was only off by 1.1666667 seconds! That's just 14 seconds per year!
It's important to note that no two watches will perform exactly as well as each other. Even if you had two of the same model watches made in the same factory on the same day by the same people, they aren't likely to be exactly as accurate as each other. But they will probably be within the published specifications, and that's all you should expect. I enjoyed seeing the results of this experiment but, as always, your experience may be different.
The bottom line is that, overall, these watches did very well.