Shout out to the Itanium sales forecast: https://upload.wikimedia.org/wikipedia/commons/8/88/Itanium_...
(This version of the graph is pretty old, but it's enough to get the flavor. The rate of new installations is still increasing exponentially, and the IEA continues to predict that it'll level off any day now...)
(It’s hard to harvest more power from a star than a Dyson sphere is capable of)
So, yeah, in a few years they'll be right. Even if for just a short time while the rest of the economy grows to keep up with the change.
an AMD alternative wasn't even a blip on the radar
I'm well aware of Opteron's impact. In fact, the event when that info was related to me, was partly held for me to scare the hell out of Intel sales folks. But 64-bit wasn't really part of the equation. Long time ago and not really disposed to dig into timelines. But multi-core was an issue for Intel before they were forced to respond with Yamhill to AMD's 64-bit extensions to x86.
That's one way to explain it. Alternatively, one might say that FSB-based Netburst servers would not benefit much from multi-core because the architecture (and especially FSB) has hit its limitation. Arguably, Intel had no competitive product on the mass server market until 2006 and Core-based Xeon 5100 introduction. Only enormous market inertia has kept them afloat.
> In the 64-bit server space, which is really what's relevant to this discussion, AMD was pretty much not part of the discussion until Dell (might have been Compaq at the time) and Sun picked them up as a supplier in the fairly late 2000s.
That was one relatively small (servers number-wise) segment of the market. Introduction of Opteron servers and Windows Server 2003 64-bit has created a new segment of mass 64-bit servers which have very quickly taken over entire 32-bit (at that time) mass server market. That was the real market that Intel wanted for themselves with introduction of proprietary Itanium but failed to acquire it because of the compatibility issue. High-end mainframe-adjacent market segment indeed belonged to Itanium for many years after, but that wasn't the goal of Itanium. Intel wanted to be a monopoly on the entire PC&server market with no cross-licensing agreements but failed and had to cross-license AMD64 instead.
I don’t know what factors would make IEA underestimate solar adoption.
The IEA is an energy industry group from back in the days where "energy" primarily meant fossil fuels (i.e. the 1970s), and they've never entirely gotten away from that mentality.
Remember all the conspiracy theories about how someone invented a free energy machine and the government had to cover it up? Well they're actually true - with the caveat that the free energy machine only works in direct sunlight.
Something I ponder from time to time, while trying to figure out how to be less of a cynic and more of a leader.
Explain Amazon, Uber, Spotify, Tesla, and other publicly listed businesses that had low or even negative profit margins for many years.
The idea that Wall Street only rewards short term profit margins is laughable considering who is at the top of the market cap rankings.
It reminds me of a meeting long ago where the marketing team reported that oil was going to hit $400/bbl and that this would be great for business. I literally laughed out loud. At that price, gasoline would be about $18/gal and no one could afford to move anything except by ox cart.
Just for some rough math here - I’m currently paying around $1.20/L for gas, and crude oil cost is roughly half of that, so if crude went up by 6x, I’d be looking at $5/L for gas. Gas is currently about 20% of my per-km cost of driving, so that price increase at the pump would increase my per-km cost by about 60%.
FWIW that’s roughly the same per-km cost increase that people have voluntarily taken on over the past decade in North America by buying more expensive cars.
(Though this does apply to personal transportation only, the math on e.g. transport trucks is different)
Not to mention less efficient cars.
I drive electric so like to imagine myself sheltered from gas price increases but I know grocery costs would explode
I googled for a couple sources on the breakdown of the price of gasoline, and they seemed to be in agreement that the raw cost of crude is somewhere around half. (And broke refining out separately.)
I'm sure it's not perfect, but it seems fairly reasonable. (And it can be off by quite a lot and still not make a huge difference to the cost-per-km of driving.)
Look at gas prices in your area. Look at the price of crude. Divide.
How could you possibly not be able to estimate the fraction?
And yeah ideally you use an average number over some months and you sample the crude earlier than the gas but those are minor tweaks.
B) They shouldn't correlate by a particularly large amount in a competitive environment. For an approximation as rough as "half" and assuming no other changes it's not a big deal.
https://bitmason.blogspot.com/2024/02/the-sinking-of-itanic-...
https://www.sigmicro.org/media/oralhistories/colwell.pdf
'And I finally put my hand up and said I just could not see how you're proposing to get to those kind of performance levels. And he said well we've got a simulation, and I thought Ah, ok. That shut me up for a little bit, but then something occurred to me and I interrupted him again. I said, wait I am sorry to derail this meeting. But how would you use a simulator if you don't have a compiler? He said, well that's true we don't have a compiler yet, so I hand assembled my simulations. I asked "How did you do thousands of line of code that way?" He said “No, I did 30 lines of code”. Flabbergasted, I said, "You're predicting the entire future of this architecture on 30 lines of hand generated code?" [chuckle], I said it just like that, I did not mean to be insulting but I was just thunderstruck. Andy Grove piped up and said "we are not here right now to reconsider the future of this effort, so let’s move on".'
Even HP dropped it eventually. And the former CEO of Intel (who was CTO during much of the time Itanium was active) said in a trade press interview that he wished they had just done a more enterprisey Xeon--which happened eventually anyway.
One really interesting related angle is the rise of open source software in business IT which was happening contemporaneously. X86 compatibility mattered so much back then because people had tons of code they couldn’t easily modify whereas later switches like Apple’s PPC-x86 or x86-ARM and Microsoft’s recent ARM attempts seem to be a lot smoother because almost everyone is relying on many of the same open source libraries and compilers. I think Itanium would still have struggled to realize much of its peak performance but at least you wouldn’t have had so many frictional costs simply getting code to run correctly.
Time and again, I run into professionals who claim X, only to find out that the assertion was based only upon the flimsiest interpretation of what it took to accomplish the assertion. If I had to be less charitable, then I’d say fraudulent interpretations.
Promo Packet Princesses are especially prone to getting caught out doing this. And as the above story illustrates, you better catch and tear down these “interpretations” as the risks to the enterprise they are, well before they obtain visible executive sponsorship, or the political waters gets choppy.
IMHE, if you catch these in time, then estimate the risk along with a solution, it usually defuses them and “prices” their proposals more at a “market clearing rate” of the actual risk. They’re usually hoping to pass the hot potato to the poor suckers forced to handle sustaining work streams on their “brilliant vision” before anyone notices the emperor has no clothes.
I’d love to hear others’ experiences around this and how they defused the risk time bombs.
It’s comforting to know that massively strategic decisions based on very little information that may not even be correct are made in other organizations and not just mine.
https://www.tomshardware.com/pc-components/cpus/former-intel...
One Dell has an early 64-bit mainboard but only a 32-bit CPU in that socket, just fine for Windows XP and will also run W10 32-bit (slowly), mainly dual booting to Debian i386 now since it retired from office work. Puts out so much heat I would imagine there is a lot of bypassed silicon on the chip drawing power but not helping process. IIRC a 64-bit CPU for that socket was known to exist but was more or less "unobtanium".
Then a trusty HP tower with the Pentium D, which was supposedly a "double" with two x86 arch patterns on the same chip. This one runs everything x86 or AMD64, up until W11 24H2 where the roadblocks are unsurmountable.
It was targeted at DSL modems, and I think the platform has faded and is now somewhat obscure.
https://royalsociety.org/people/sophie-wilson-12544/
https://old.hotchips.org/wp-content/uploads/hc_archives/hc14...
Intel has plenty of engineering talent, if the bean counters, politicians and board would just get out of the way they would come back. But instead you see patently stupid/poor execution like then still ongoing avx512 saga. Lakefield, is a prime example of WTFism showing up publicly. The lack of internal leadership is written as loud as possible on a product where no one had the political power to force the smaller core to emulate avx512 during the development cycle, or NAK a product where the two cores couldn't even execute the same instructions. Its an engineering POC probably being shopped to apple or someone else considering an arm big.little without understanding how to actually implement it in a meaningful way. Compared with the AMD approach which seems to even best the arm big.little by simply using the same cores process optimized differently to the same effect without having to deal with the problems of optimizing software for two different microarch.