We often discussed the 1000 Hours Control tests, and their difference with the COSC.
I scanned the page of a 2002 Catalogue, where it is all explained.
Here it is:

Have a good read.
Nicolas
... is, that unlike the much-criticized COSC tests, the manufacturers rarely disclose the exact test criteria and limit values the watches/movements must not surpass. The text reproduced is typical for that.
"The measurement is based on tests performed in 6 different positions simulating conditions of regular wear." - what does this mean? Is the accuracy performance measured at full mainspring tension? At half? Is it repeated, how often, in which intervals? Or is the measurement based on an average value, like the COSC? What are the limit values, in all positions, and so on.
I, too, was quick to join the concert of critical voices regarding the COSC, until I realized that the only result of it is a complete loss of comparability regarding the performance of different manufacturers' products, and also a loss of impartiality. I now consider the manufacturers to be the wrong place to produce an objective evaluation of a watch's or movement's performance. Some kind of independent and objective authority has to test the watches. This needs not to be the COSC, necessarily. But until now, there is no real alternative.
Marcus
I will try to have more précisions on the 1000 Hours Control. As I said, this is an extract I found in a JLC Catalog.
Nonetheless, it is true that it would be better to have an external authority to do these tests. But if the outcome is less good than the COSC?
It is the story of the half empty, half full glass...
Best, Marcus.
Nicolas
What I am complaining about is the lack of transparency, when all manufacturers strictly keep to their own individual test procedures, which might - not necessarily have - be chosen to perfectly match the characteristics of their own products. However, we have to face the fact that the huge majority of customers appears not to be interested in too much transparency either.
Marcus
in the end the best kind of independent and objective authority is the customer and this is the ultimate alternative. Is it not? I mean, we should let the watchmakers do the tests they see fit for the purpose of their products, if the customer is satisfied with the result, it is fine for me. We are not talking here about "dangerous, possible life endangering" object. (although, a friend affirms that when pressing on the chronograph pusher of a certain watch brand something may suddenly hurt one's back side
) but this is still not life threatening
....I hope
If COSC is definitely not ideal nowadays, for many reasons. The main one, in my opinion, is that only the "uncased" movement is subjected to the test, which is not the case, excuse the pun! for the JLC 1000 Hours Control.
I think that the 1000 Hours Control which JLC puts its product through is a test with many merits, not the least of which is that they put their reputation on the line and in the end it gives plenty of satisfaction to the great majority of its customers.
Just my thoughts.
Cheers
Francois from Down Under
pages 187 & 188, where JLC defends the reason for the "Master Control" test, and it wasn't just about accuracy. So, while I agree on one hand with Marcus' analysis of "loss of comparability," JLC's reputation is at stake.
It is, also, why PP abandoned the Geneva Seal (which had nothing to do with accuracy until recently). This way they can say, "It's our reputation, and we stand behind it."
-Dean
more thing reading this article. That is, 800 rotations in two hours is not necessary the same as 400 rotations per hour!? Otherwise it would be stated
Joke aside, very interesting and from my experience, JLC "normal" wrist watches are definitely the most accurate.
Cheers
Francois from Down Under