If you recall our last story about Intel and the Ivy Bridge non-delay, we mentioned that it happened due to a large customer order. Things get interesting when you stop and think about why they would need to make such a change at the last minute.
That customer is Apple of course, and along with a few others, suddenly decided they needed more GPU power. This may seem a bit odd when you think about it, Apple is going to be using Nvidia GPUs in most of the MacBooks as we exclusively told you several months ago. All was fine and dandy, we actually understand why this decision was made, Nvidia’s Kepler is probably the better GPU architecture for this round. The only question was that given the company’s process woes, could they make enough to supply Apple?
Dark clouds started to gather on the Nvidia Q4/2012 conference call last month, and a rather nervous sounding Jen-Hsun admitted that 28nm was going horribly for the company. This completely validated what we have been saying for months, SemiAccurate moles have been giving a diametrically opposing version of events from the official “unicorns and rainbows” version. Needless to say, we were really curious about how this supply shortage would play out in Cupertino, or who else would get the short end to satisfy the notoriously supply-chain related humor impaired Apple.
Well, the answer was given by Apple itself a few weeks ago. Just after TSMC had their mysterious production line stoppage, Apple changed their orders at Intel. What exactly changed? Apple upped their SKUs from parts bearing awful Intel GPUs to variants with more of those awful shaders. Since those Ivy Bridge CPUs are going in to laptops that have a GPU, upping the shader count from 6 to 16 should be a waste, they will never be turned on.
If they are going to be turned on, that would mean that the discrete GPU in those machines is either going to be much higher spec’d, or it won’t be there. Since Nvidia can’t supply enough small GPUs, what do you think the odds of them supplying the same number of larger and lower yielding ones are? There goes that option, leaving only one possibility, the next gen low and mid-range MacBooks are not going to have a GPU, only a GT2 Ivy Bridge.
That is exactly what SemiAccurate moles are telling us is going on. Nvidia can’t supply, so Apple threw them out on their proverbial magical experience. This doesn’t mean that Nvidia is completely out at Apple, the Intel GPUs are too awful to satisfy the higher end laptops, so there will need to be something in those. What that something is, we don’t definitively know yet, but the possibilities are vanishingly small.
Given how late it is in the game, and how long it takes to retool a laptop’s physical parts, we don’t think Apple will deviate from the current plan at the higher end of the lineup. We also doubt that AMD will get any business out of it, but you never know. Until the complete stoppage at TSMC, AMD was not having yield issues like Nvidia, but that doesn’t mean wafer supply was enough to supply their existing demands, much less take on Apple as a customer at the last minute.
Our analysis indicates that the lower end MacBooks will simply do without a GPU, the higher end parts will remain unchanged, and the middle ground will have some models with and some without a GPU instead of almost all with a discrete Nvidia GPU. Those without will make a much larger portion of the mix than they would have at this time last month, if there were any at all. Since the laptops are not launched yet, and specs have not leaked, presumably no one outside of Apple and Nvidia will know the difference.
For Nvidia, this hard fought win at Apple just went up in smoke for the reasons we have been warning readers about since last summer. The largest single order for the GPU maker this year was scaled way way back, but we can’t say the exact percentage. By the time there are new bids for the next generation laptops, Haswell will essentially kill off the segment that Apple would have used, meaning this market is dead forever. It may sound dramatic, but this is the end of the mid-range GPU segment as a standalone part. This most lucrative slice of the market is now on its last legs.S|A
Latest posts by Charlie Demerjian (see all)
- Thing go bump(gate) in the night for Nvidia’s GP100 Pascal GPU - May 3, 2016
- Cavium’s Octeon TX blends compute and packet moving - May 2, 2016
- Qualcomm releases Zeroth API to developers - May 2, 2016
- Another detail about Qualcomm server SoCs revealed - Apr 27, 2016
- AMD finally really honestly launches the dual Fury - Apr 26, 2016