The fact any modern computer chip works reliability is a pure miracle. The process variations are extreme, and you often end up with a lot of B-level engineers/technicians keeping things going. Having some experience in the semiconductor industry, it oftentimes felt like a lot of bubble gum and bailing wire was used to get the product out the door. Hats off to all the people keeping these systems alive and functioning.
I've worked on the construction of a large dam and I think that's the case for most modern technology. Obviously once a dam is up and running it's really stable compared to what goes on in a fab, but getting it built? That's a whole other story.
The math and engineering behind dam construction are well understood but actually getting them built in practice is a years long story of yak shaving, cat herding, and trying to overcome every little piece of BS nature has to throw at the project. Unique and predictable geological conditions, unknown underground water sources, unexpected soil composition changes, surprise fault lines. Then there's the logistical nightmare of actually moving all the equipment in and earth out, the weather and environmental factors that impede every action, humans ignoring safety altogether, and so on and on. All of this implemented by workers on the ground who just barely know what's going on (for no fault of their own).
I'm continually surprised that anything complicated ever gets built at all.
There must the a name for this phenomenon. Like, the more you know, the less faith you have in it actually working. I'm pretty sure everyone feels this way about their work. I just asked my partner if it's the same with her (non-tech) job and she said yes, she can't believe it works at all. Arthur C. Clarke's Travel by Wire comes to mind too.
I would say it doesn't just work. That's why being a programmer isn't a one-off job. You're still there to glue things back together when they inevitably break unexpectedly.
It's more like disregarding all of the advantages and only focusing on the negatives or when incidents happen.
People routinely clown on companies for downtime but do not celebrate sending multiple MB pictures and videos over cell networks in remote locations from a super computer in their pocket.
Even 95% reliability is relatively good for networks working across the globe relative to what we've had through most of history.
The average person easily plays into the trope that no one appreciates IT when it works, but readily has opinions when there are problems.
The funny thing is, anecdotally, I have never had a CPU fail on me. Memory, motherboards, PCIe cards, PSUs, hard drives, monitors, keyboards, mouse have failed but I have yet to loose a CPU or an SSD.
I’m at a point where I no longer call these things chips. I don’t know what they are, they have pins but to me the precision and machinery used is at the very least on a scale of atomic design. It should not be possible for automated machines to fabricate these things. Yet here we are.
AMD just delayed Ryzen 9000 by one and two weeks because of production issues. Including recalling samples sent to reviewers and already at stores.
We appreciate the excitement around Ryzen 9000 series processors. During final checks, we found the initial production units that were shipped to our channel partners did not meet our full quality expectations. Out of an abundance of caution and to maintain the highest quality experiences for every Ryzen user, we are working with our channel partners to replace the initial production units with fresh units.
As a result, there will be a short delay in retail availability. The Ryzen 7 9700X and Ryzen 5 9600X processors will now go on sale on August 8th and the Ryzen 9 9950X and Ryzen 9 9900X processors will go on sale on August 15th. We pride ourselves in providing a high-quality experience for every Ryzen user, and we look forward to our fans having a great experience with the new Ryzen 9000 series.
Jack Huynh, AMD SVP and GM of Computing and Graphics
That, or they're calling Intel's benchmark before microcode update bluff, and in either case they're upfront about the delay instead of clamming up for half a year while unrest mounts.
I'd wager it's not that easy to separate these two, especially when voltage is dynamically regulated.
If it's a Zeiss lens imperfection, a broken ASML machine, water or silicon impurity, etc, what matters to the customer is the final product you're buying.
The industry has always been kind of monolithic. I'm not sure you can count qualcomm as a competitor just yet. competing means being in the same market. Laptop, desktop and server have all been categories that havent seen anything other than x86 for a long time.
Is this Intel's marketing department doing damage control?
Intel's been a disgrace, and that's why these chips suck. Imagine having all of the money and hiring everyone in sight and pulling all sorts of self-serving, dirty tricks industry-wide yet still letting an underdog on the verge of bankruptcy (AMD) beat you to 7 nm and eat your lunch.
