1. EM does not own a FAB
Maybe EM does not 'own' the bricks and mortar and all the machinery, but it clearly has the ability to access time not just in one FAB but in three simultaneously. If that does not give them credibility I suggest you go down to three of your own local FABs and ask if you can use their equipment. If I'm not mistaken AP's previous FAB vendors (SONY, TowerJAZZ) all provided time on their lines when it suited them. And when it no longer suited them and there were competing commercial demands on their time AKP was well and truly "FABless", leaving them unable to advance the necessary slow and sequential development work for their chip. That was precisely why AKP buddied up with EM. EM was eager to work with AKP and saw the significance of what they were doing. Irrespective on any other considerations concerning EM being 'Chinese', that counts, and has counted for a lot of the progress that AKP has been able to make over the past 5 years since 2018. However the CIDM model works, it clearly has worked in AKP's favour. It changes the meaning the word 'owning' has when applied to 'owning a FAB'. The most advanced enterprises in the West are all doing similar things I believe. All based on collaboration between those with different capabilities to address common aims and opportunities.
2. Problems at all 3 fabs at the same time
Damned bad luck all round! But it does happen. Making 3D MEMS is a lot different from making 2D microprocessors etc that most FABs are engaged with. 3D MEMS is harder. They have different kinds of problems, and the larger the wafer the more those problems are likely to become evident. It would appear to be clear (to me) that had there not been a problem with the 6" wafer at the time that there was (November 28), AKP would have probably have had chips to demo to holders and customers by end of Dec23 - as previously announced. Damned bad luck for us all!
I am interested that both the 8" FABs had a problem more or less at the same stage of progress. That suggests it was the same problem and the same cause. That also suggests that the same thing would have been developing on 6" wafers, but did not give rise to the same level of problems until the wafer surface area got bigger. Finding a solution to this will be valuable for all dimensions of wafer. My understanding of this is that stress is always a problem on all wafers but is not usually severe enough to interfere with the completion of the wafer work. Commensurately all FABs are aware of the problem and its management, and they are used to coming up with engineering solutions that fix the issue. One could presume that this was a 'first time through' for the 8" wafers in both Fabs. This is indicated to be the case. However, clearly the 'process of record', the (POR), the step by step recipe for the device manufacture, has been able to be transferred to these additional FABs such that they could start work and get right through to the later stages of production before striking the problem. I gather this is not particularly unusual or surprising. There is, as they say, a first time for everything, and the first time through is the time one discovers unknown unknowns.
3. Pushed off the 6" FAB line in favour of another customer's job
So I remain confident that for both the 6" and the 8" wafers, production will be got back onto track in short order. How short the piece of string is depends on how quickly it can be solved. That does not include the fact that the 6" FAB undoubtedly has other work to pursue, and that such work would be scheduled into the FABs calendar under a contractual commercial agreement. The fact that EM has jumped to respond by acquiring additional equipment that would minimise the bottlenecks that result in the conflicts of job scheduling is very encouraging, and indicative of their commitment to getting AKP's job done as soon as possible - or sooner if money is spent to permit it.
One could assume that the 8" FABs would have expected to have finished wafers at 'around the same times' as the 6" FAB. That may be optimistic, but presumably was part of the reasoning that AKP had in mind when no information was provided to the market.
Personally I think AKP should have warned us that this previously announced Nov/Dec timeline may not be met, as this would lead to holder expectations being dashed. But I think it would have been better to be told that early, when it was a distinct possibility, than long after the fact, when holders felt somewhat betrayed and frustrated at being treated like the proverbial mushrooms. I imagine this let down has been part of the ASX's request for information. Clearly as we have seen the SP drop by almost 15% as holders respond to both the ASX challenge and to AKP's response, this has clearly had a material effect on the SP.
"... C. Listing Rule 3.1, which requires a listed entity to immediately give ASX any information concerning it that a reasonable person would expect to have a material effect on the price or value of the entity’s securities. ..."
And where to from here?
I am of the opinion that these technical problems will be solved just as soon as humanly possible. That will likely as not mean that chips will be available for demos April-ish. Just in time for the AGM.
I sincerely hope that before hitting the SELL button out of frustration and pique at the company as a means of venting their indignation at being under-informed in an insufficiently timely manner, that they will note that only 0.1% of all shares have traded at these low prices. This too shall pass.
Expand
(20min delay)