Also, this is an ancient web site, built back before everyone had to have glitzy junk. So it has a little whimsy. Did you notice how fast it loaded?
Perhaps the moral here is that "newer" code was so obfuscated (minified; framework-infested) that I wasn't able to read it, so I only ever ended up looking at very dated HTML to fish examples out of. (Figuring out how to phrase something to Google can sometimes be slower and/or harder than simply looking in existing source code where you know you'll find something...)
FWIW, I started tinkering with HTML circa 2003 and have slowly (not especially concertedly) studied it over the years, which is my excuse^Hdefense. HTML5/CSS3 aren't especially intimidating to me (thanks to browser compatibility :D) and I could probably photocopy 60-80% of the designs I see here given significant time (this likely describes a lot of people here), but I don't really actively develop for the Web, so little oversights like this happen. Goes and reads how to include a CSS file again
I guess it's a good thing that I don't do much frontend coding anymore!
I think I followed a similar path to what you did, albeit 7 years later. IMO frontend dev is fine, given the time that seems to be required in order to stumble on these "things we didn't know we needed to know", whether by concerted tuition/learning or simple wandering.
Would it be better without an animation? Possibly.
But given the current state of the web I'd ve fine with a few extra "i hate animations"-animations if they replaced sites that spy on me, only work in Chrome and advertises dating sites I don't need.
Edit: good -> not too bad
His work is an even-handed and well-presented coverage of modern architecture. I'm perennially amazed by the way that people will regularly spout off about what might or might not be expensive or cheap on modern IA when you could just go look it up in Agner's work.
I think "unbelievably bad" is unfair; specifically, compared to what? If you think Intel docs are bad you must not have read many equivalent docs for other processors. We worked on a range of processors for the Hyperscan project before version 4.x, and while Intel docs leave a lot to be desired, you should see the other guys.
That said, there are a lot of things that need to be fixed.
L/T numbers are always idiosyncratic as there are a bunch of ways of getting them and none perfectly predict behavior for all instructions and all cases.
L/T numbers in the ORM frequently 'go missing' which isn't great.
The ORM is a giant 'additive song' that gets a new chapter with each new extension, and tries to provide optimization direction for over 10 years worth of processors. There's still Pentium 4 material in there!
There's somewhat of a culture of secrecy about the architecture that Intel would actually benefit from relaxing. It's clear that other vendors know how to get close to IA in performance or even pass it on some metrics. So I think it would benefit Intel a great deal to help people build better mental models of how their processors work.
I mean, the image is just totally unreadable. https://imgur.com/a/jqxpqAf
This is especially true in light of comparisons to other vendors; I'm not aware of any up-to-date machine-readable information that's the equivalent for Appendix C for any other vendor, and historically (pre Intel acquisition) we had trouble finding this information out at all for some of our platforms to which we had Hyperscan ports. It was either out of date, not available, or behind a paywall ("show you've bought our SDK to get basic processor docs").
>Regality theory is a theory saying that people show a preference for strong leadership in times of war or collective danger, but a preference for an egalitarian political system in times of peace and safety. These psychological preferences in individuals are reflected in the political structure and culture of the society. A society in danger will develop in the direction called regal, which includes strong nationalism, discipline, strict religiosity, patriarchy, strict sexual morals, and perfectionist art. A society in peace will develop in the opposite direction called kungic, which includes egalitarianism and tolerance. This book is both theoretical and experimental. The theoretical explanation of the regal-versus-kungic dimension is based on evolutionary psychology and human ecology. Contributions from the social sciences and the humanities are added to further analyze historical examples of regal and kungic developments. The theory is tested on data from both contemporary and ancient societies. These tests confirm the predictions of the theory.
Seriously, any practitioner should be reading Fog. Of course, they should read the Intel + AMD optimization manuals, Dr. Bandwidth, Developer Zone, the patents, but they definitely should be reading Fog.
> Register read stalls has been a serious, and often neglected, bottleneck in previous processors since the Pentium Pro. All Intel processors based on the P6 microarchitecture and its successors, the Pentium M, Core and Nehalem microarchitectures have a limitation of two or three reads from the permanent register file per clock cycle.
> This bottleneck has now finally been removed in the Sandy Bridge and Ivy Bridge. In my experiments, I have found no practical limit to the number of register reads.
This (and the preceding related discussion) gives you a ton of insight into the limitations Intel engineers were facing and the trade-offs they made. Here, ports on a register file have a cost--they require transistors, may limit cycle time, etc. Intel realized that in many cases, the operands to an instruction would be available on the bypass network, and you could get away with having fewer ports on the register file than you'd theoretically need given the number of functional units.
The webpage is awesome in my opinion. It's functional, simple and with a good amount of humor :) It a good example of a content-first approach.
Just goes to show that there's always an idiot out there with a bikeshedding comment.