According to the original reports, the situation was supposed to be that Hynix customers began to complain about the quality of some deliveries, and Hynix subsequently proceeded to download a total of 240,000 memory wafers.
However, the memory manufacturer objected to this report. He confirmed that downloading memory solves, but that the range is much smaller. At the same time, he was to initiate a police investigation into the “spread of gossip”.
240,000 thousand wafers would correspond to about 13% of Hynix’s total monthly production (it has a capacity of 1.8 million wafers per month) and almost 17% of Hynix’s monthly memory line production (these account for about 80% of the company’s total production capacity). Hynix claims that the actual scope is much smaller, on the other hand it did not disclose a specific value and provided only a rather vague comment:
|
We are currently negotiating a solution to a problem with a limited number of customers. Although it is too early to estimate potential losses, we do not expect them to be so significant, as defectivity is typical of quality control. — Hynix representative |
The company therefore does not know how significant the extent of the damage is, but at the same time it knows that it is not as it was published before the official statement. It also retracts the wafers because they are defective, but at the same time their defectivity does not deviate from the extent that otherwise passes quality control.
Whatever the situation, at least Hynix’s argument is not entirely convincing, and even if the number of defective wafers is an order of magnitude lower, it will still be another unpleasant blow to the IT world. In terms of supply, this is a de facto outage and the supply of memory is already on the edge, as all manufacturers are running at full capacity due to the extreme demand for hardware.
–
