4 ms·
This will make it harder for them to invest in production technology, which will make it harder for them to catch up to TSMC. It might be the only move they can
by danpalmer 5y ago
This will make it harder for them to invest in production technology, which will make it harder for them to catch up to TSMC. It might be the only move they can make, but that doesn't make it a great one.
- tomalpha 5y agoIs this not just the efficient-market at work? Simplistically: Intel's chips aren't as good as AMDs so it has to drop prices. And for future investment Intel still has ~ $24 billion cash on hand as of June 2021 [0] [0] https://www.macrotrends.net/stocks/charts/INTC/intel/cash-on-hand https://www.macrotrends.net/stocks/charts/INTC/intel/cash-on...
- scandinavian 5y ago> Simplistically: Intel's chips aren't as good as AMDs so it has to drop prices. Shouldn't you replace AMD with TSMC in that sentence, unless you meant design instead of chips? AMD doesn't manufacture chips.
- rcthompson 5y agoBy "AMD's chips", they clearly meant chips marketed and sold under AMD's brand. If you're trying to make the point that TSMC deserves the real credit for competing with Intel, then just say that.
- scandinavian 5y agoYou don't think it's important to make the distinction between chip design and manufacturing? Intel has clearly failed with regards to manufacturing new nodes, but is the chip design really that bad when they could compete for a long time with a large node disadvantage?
- pdimitar 5y agoThey almost never competed in the dictionary sense of the word, historically. They have plenty of shady tactics to gain more market share. Assuming that the "free market" works -- or that it even exists -- is way too charitable and optimistic of you.
- tomalpha 5y agoIt's an interesting point, but I buy from Intel or AMD - I don't buy from TSMC. Intel and AMD supply me the product and set the pricing. As a simple consumer, I perhaps don't know about their upstream suppliers (granted the HN crowd will absolutely know...).
- Retric 5y agoIf you own a cellphone or console your likely buying from TSMC. If you’re buying AMD then you are buying TSMC. Intel is unusual in that they still manufacture their high end chips in house, cutting edge fabs are simply mind boggling expensive. So basically everyone else outsources and if your outsourcing high end chips you might as well buy from the bets if you can.
- adrian_b 5y agoIt is true that the main reason why the AMD chips are better than the Intel chips is that the TSMC 7 nm manufacturing process is significantly better than the Intel process used for the Ice Lake Server chips. Nevertheless, the AMD designers must be praised for making the right design choices year after year for the last half of decade, which were needed to fully exploit the characteristics of the modern CMOS processes. On the other hand the Intel designers appear to have lived in a fantasy land, where they had absolutely no idea about how their future manufacturing processes will behave, even if in their case the required information should have come from another division of the same company, not from different foundry companies, like in the case of AMD. Once again, Intel was not able to switch in time their style of design, to be in sync with the advance of CMOS technology. During 2003 - 2008, Intel needed 5 years to follow AMD and switch to CPUs with integrated memory controllers and now, during 2016 - 2021, Intel required again 5 years to follow AMD in the transition to the use of multiple interconnected chiplets instead of large monolithic chips.
- danpalmer 5y agoTSMC are spending about that every year for the next 3 years on production improvements. Chip fabrication is so expensive to develop, I'm concern that $24bn is nowhere near enough to build a 5nm process.
- pbalau 5y agoIntel has 24bn available in cash. One could assume Intel also has $FOO x 24bn available in loaning power.
- MangoCoffee 5y agomoney can't buy node process. Intel struggle with 10nm for so long and 7nm is delayed while TSMC is now 5nm production ready and developing 3nm.
- the-dude 5y agoIsn't a new fab around $20bn?
- aeyes 5y agoYou usually don't buy the construction of a new fab with cash.
- humps 5y agoThis move is to attract AMD customers back to Intel, so while in the short term it could hurt revenue, longer term it may mean increased profits and therefore offer more room to invest. There's also the potential for increased sales at the lower pricing, which will still have a profit attached. So I doubt that overall this will have much impact on investment.
- tehbeard 5y agoIs a discounted chip price going to sway people enough to offset the hotter core (therefore pricier in terms of power and cooling), and limited (in comparison to EPYC) IO?
- dtech 5y agoI don't think money is the problem with Intel's troubles.
- adwn 5y agoA lack of capital is hardly the reason for Intel falling behind TSMC. If it were, they wouldn't have lost their lead in the first place, and TSMC wouldn't have been able to overtake them.
- jychang 5y agoIntel has $24.8 billion in cash. https://www.macrotrends.net/stocks/charts/INTC/intel/cash-on-hand https://www.macrotrends.net/stocks/charts/INTC/intel/cash-on... Intel dropping their prices and thus revenue temporarily should not affect their ability to compete at all. They're not THAT badly mismanaged to the point they're out of cash.
- joakleaf 5y ago... and dropping prices doesn't necessarily meaning dropping revenue or even profit. Intel's per chip profit may drop, but if they sell more because of lower prices, they may actually increase their overall profit. It is really hard to tell without knowing Intel's current profit margin and the increase in number of chips sold from this maneuver (if any).
- starfallg 5y agoIntel drank the management consultant kool-aid like many large pharmaceuticals corps, relying more on financial engineering than their research pipeline to compete. The flip side is that they have lots of money to splash around. Companies like Pfizer for example.
- christophilus 5y agoDid they? I thought they made a heavy bet that hasn’t paid off (and maybe never will).
- cisvolk1016 5y agoSo if Intel drops $0.8B on a risktaker learning from what I. B. M. did to make Watson and not make those mistakes. What new thing could come from that? A new category product is what Intel should look for around the corner(s).
- starfallg 5y agoYou're right for the 10nm (7nm) process development. They focused on shrinking the wrong parts and ended up with an inferior product. Instead of changing direction, they doubled down.