Episode Summary
Executive Summary: The episode centers on Apple’s 2018 MacBook Pro controversy: keyboard reliability, keycap/membrane design, the i9 throttling problem, and Apple’s software fix. The hosts also unpack Intel’s outdated CPU/TDP assumptions, the utility of Apple’s patch, the return of Intel Power Gadget, and broader concerns about Intel’s roadmap versus Apple’s possible future processor alternatives. The back half shifts into personal tech regrets and an extended, playful food discussion.
Main Topics: MacBook Pro keyboard design and repairability: The hosts discuss Apple’s revised butterfly keyboard, noting that keycaps are now easier to replace and that the membrane appears intended to block debris, not just reduce noise. They debate whether this is a meaningful fix or another band-aid on a fundamentally fragile design. Apple’s thermal throttling bug and software fix: They analyze Apple’s official explanation for the 2018 MacBook Pro performance issue, concluding the problem was a firmware/software bug in thermal management rather than a VRM hardware failure. They note the patch improved benchmark results and smoothed out clock behavior. Intel CPUs, TDP, AVX, and changing performance assumptions: The hosts revisit their understanding of base clock speeds and TDP, learning that modern Intel CPUs can run below base clock for idle savings or AVX workloads, and that TDP is more of an average/expected ceiling than a hard limit. They emphasize that Intel’s 14nm stagnation complicates laptop design. Intel Power Gadget and suspicion around the data source: They discuss Intel’s Power Gadget utility disappearing and returning, and why that fueled conspiracy theories. The conclusion is that the tool’s return and Apple’s patch suggest a mundane technical issue rather than a deliberate cover-up. Intel’s strategic decline and future CPU options for Apple: Using a Stratechery article as a springboard, they explore how Intel’s process delays and inability to deliver 10nm chips weaken its long-standing advantage. This revives the idea of ARM Macs or Apple considering AMD/other options if Intel can’t keep pace. Listener questions: display scaling, regrets, pasta, and crash plan: The show ends with assorted Q&A: non-native Retina scaling preferences, tech opinions the hosts regret, family pasta sauce and meatball traditions, and an update on CrashPlan usage. These segments serve as lighter, personal counterpoint to the hardware-heavy main discussion.
Key Arguments: Apple’s membrane under the keyboard is likely primarily for debris exclusion, not acoustics; Apple’s own support docs and iFixit teardowns support that interpretation. The keyboard improvement is real but incomplete: easier keycap repair helps, yet the design may still trap debris and remain vulnerable to failure. The i9 throttling story was more complex than a simple “CPU should never drop below base clock” rule; modern Intel behavior includes idle downclocking, AVX-induced reductions, and TDP definitions that are more nuanced than commonly assumed. Apple’s firmware update fixed the MacBook Pro performance issue by adjusting thermal management behavior across the whole system, not by changing a VRM limit. The original Reddit VRM theory was an impressive armchair diagnosis, but Apple’s own statement and the after-the-fact benchmarking point to a software/firmware bug instead of a hardware bottleneck. Intel’s prolonged 14nm era and failed 10nm roadmap are eroding its historic advantage, making Mac hardware planning harder and increasing the attractiveness of ARM or other alternatives over time. Many consumer tech judgments age poorly because they assume today’s priorities (specs, size, price, input design) will remain stable, when in reality preferences and use patterns shift.
Data Points: MacBook Pro thermal fix impact: 2018 Core i9 after patch: 72 minutes; before patch: 80 minutes - Macworld’s preliminary 4K Premiere test showed the Apple update improved performance. Comparison machine runtime: 2017 2.9 GHz Core i7: 90 minutes - Used as the older baseline in the same Macworld benchmark comparison. Repair extension duration: 4 years - Mentioned as the kind of coverage Apple’s older keyboard models effectively had via repair extension programs. Intel TDP examples: 45 watts, often exceeded to 65 watts; sometimes up to 100 watts - Discussed as the actual behavior of Coffee Lake processors under turbo/load, versus simplified assumptions about TDP. Intel Power Gadget utility update: Version 3 - The tool reappeared with added support/features, including Windows 10 support and multi-socket power estimation. USB-C hub availability estimate: 2020 - The hosts discuss an anonymous report that Intel’s hub chip might enable better USB-C hubs by then, with products not likely until around 2020. Apple CPU roadmap pressure: 2019 and beyond - Intel’s inability to deliver 10nm parts in time complicates planning for future Macs and related chip decisions. CrashPlan backup usage: Still in use - John says he remains on CrashPlan for a small business/media backup setup, despite quirks in the client. Meatball size: 3.5 ounces - A listener-reported “correct” weight for the family meatballs.
Pivotal Quotes: "a missing digital key in the firmware that impacts the thermal management system and could drive clock speeds down under heavy thermal loads" — Apple (as quoted in transcript): Apple’s official explanation for the MacBook Pro performance bug. "The correct solution to this problem is to design a keyboard that either dust can get in and out freely... or design a keyboard that dust can't get into it all." — Marco: Argument that Apple’s membrane is only a partial fix and not the ideal long-term design. "I think the jury is still out on this." — John: His assessment of whether the new MacBook Pro keyboard is an acceptable solution overall.
Implications: Listeners should expect incremental Apple fixes, not instant perfection. The episode suggests the MacBook Pro issue was software-fixable, but also that Intel’s roadmap uncertainty may force bigger platform changes, potentially making ARM Macs or other CPU shifts more plausible.