OK, now go license 64 cores of Oracle DB (for example) and get less performance than one core on a zEC12, as you say. I'll help you out: you'd probably pay about $1.5M in database software licensing plus $300K+ in annual maintenance for your 64 X86 cores versus $47K and $9.4K on a zEC12 core. And that's one cost factor among many, not the only one. So which server is "cheaper"? Is a bicycle cheaper than a truck? (Not an Olympic racing bicycle, probably.) It depends on what you're trying to do. Though I've noticed that the average Slashdot poster hasn't a freakin' clue about IT economics, sadly.

OK, here's a benchmark. You're welcome to try running an entire large bank (for example) on one server -- your choice. OK, two servers: I'll allow you one additional for off-site disaster recovery of all development, test, and production workloads, including concurrent batch and online, for all the bank's security zones. Choose wisely, Grasshopper.

Yes, everybody does that (out-of-order execution, pipelining, etc., etc.) And then...you still need to keep the CPU well fed to boost performance. Enormous 4-level caches help do that. Having a continuous 5.5 GHz clock speed is also quite helpful. So is having 101+ cores that can access the same cache rather than, say, 8 such cores. And at least a couple hundred (at least) other IBM performance tricks, many of which cost money to deliver and thus probably won't find their way into save-a-nickel parts of the market any time soon. It also very, very seriously helps when you design both hardware and software together, as the late great Steve Jobs (among others) reminded us all.

No, that's not a correct supposition -- quite the opposite, actually. All processors, including Intel X86, use microcode (or what IBM calls millicode) to a degree. IBM knows it well. After all, they invented microcode/millicode in the System/360 in 1965. But IBM uses microcode comparatively less nowadays than other processor architectures. The vast majority of zEC12 instructions are implemented entirely in hardware, including IEEE-754-2008 decimal floating point as an example. There's some really, really interesting new stuff in the instruction set, like the first transactional memory ("transaction execution facility") instructions in a commercial server, and some "feedback" instructions that can tell Java applications/the JVM how to dynamically tune itself in a live running environment. Very cutting edge -- so cutting edge I've got to crack open some engineering manuals to try to figure out what they've done, although they probably need to write those manuals.

Actually, L4 cache on this new IBM zEC12 is a minimum of 384MB up to 1.5GB per server in increments of 384MB. As you add cores the L4 is bumped up. IBM doubled the cache in only a 25 month product cycle. Bravo.

No, they aren't claiming that. Clock speed is still extremely important, though, and nobody else except IBM has figured out how to hit these high gigahertz numbers, much less within power and cooling constraints. What's all the more impressive is that IBM does it at mainframe service qualities, i.e. this machine runs continuously at 5.5 GHz without shutting off cores, without "burst" mode, and without weird/exotic stuff like cryogenics that might keep a chip running long enough for a screenshot. It's just balls out performance on every thread -- and there's a definitely a market for that. Nobody else is left doing this computer engineering, bless them. Also check their cache sizes (obscenely huge), out-of-order execution, pipelining, crypto and decimal floating point in every core, extremely complex instructions like transactional execution.... This z CPU is a gorgeous piece of engineering in every way. And no, you can't run an entire large bank (for example) on your laptop.

GiffGaff runs on the O2 network. They offer 500 MB for only £5 (or 1 GB for £7.50). You need to open the account with a minimum of £10 of credit. Just buy a GiffGaff (preferable) or O2 £10 card at any mobile top-up counter -- at petrol stations, post offices, off licences (convenience stores), etc. SIMs are free when mailed to a domestic U.K. address, but make sure to order a MicroSIM for an iPad. Activate online (via wifi). Smartphone rates are great, too.

By what measure is HP the "world's largest computer vendor"?

• Market capitalization? No, that's Apple.
• Software? No, that's Microsoft.
• Business software ("middleware")? No, that's IBM and then Oracle.
• Internet? No, that's Google.
• Mobile? No, that's Samsung (in units) and Apple (in profits).
• Servers? Depends on which quarter/year you check, but generally that's been IBM, especially in the more profitable high-end.
• Networking? No, that's Cisco.
• IT services? No, that's IBM.
• Business applications? No, that's SAP and Oracle.
• PC distribution? Yes, although Lenovo is now nipping at their heels.

HP is rather tiny now, especially in market capitalization terms (under $40B). For perspective, even Facebook, which has been battered, has a higher market capitalization. HP really needs to choose its battles wisely.

No, they're very different. They have entirely different instruction sets as one example. As another, the z196 is clocked at 5.2 GHz and POWER7 at 4.25 GHz (in the 795 Turbo). The z196 is quad-core while POWER7 is octo-core. POWER7 has about 1.2 billion transistors and z196 about 1.4 million. POWER7 has a private per-core L1 instruction cache of 32 KB; z196, 64 KB. POWER7's private per-core L1 data cache is 32 KB; z196, 128 KB. POWER7's private L2 cache is 256 KB; z196, 1.5 MB. The L3 cache design is different, too. And those are just a few examples.

None of the IBM z/Architecture microprocessors (or their ESA/390 and prior predecessors) are listed yet. So Stanford is only missing the highest clock speed CPU ever created in the entire history of computing to date -- the IBM z196 microprocessor. Which seems like a rather serious and obvious omission. Also a bit insulting, since IBM has been announcing their new z/Architecture microprocessor breakthroughs exclusively first at Stanford's own "Hot Chips" conference for several years now. (Ooops.)

LTO-6 is not available yet. When available, each cartridge will hold up to 3.2 TB uncompressed. (I can make up practically any figure I want for compressed data capacity. It just depends on the type of data and the compression algorithm.) That's not bad, but IBM's TS1140 tape drive has been available since June, 2011. It supports tape cartridges that each hold up to 4.0 TB uncompressed. The Oracle StorageTek T10000C tape drive supports cartridges that each hold up to 5 TB uncompressed. It depends on what you're doing -- raw storage capacity per cartridge is certainly not the only relevant specification -- but there are interesting choices.

IBM hit 5.2 GHz in 2010. That's with all cores active and constant operating speed, too.

I have looked.

High-frequency (program) traders make big investments in computing that is located as close as possible to trading floors. They employ extremely expensive programmers to write highly proprietary code to run on these machines located on prime real estate, in speed of light terms. That's exclusively a rich man's (or firm's) game. I'm hard pressed to think of any way in which such trading helps allocate capital materially better to the real economy, which is the only useful purpose of finance. It's a grand waste of talent and resources that could be put to much better use elsewhere in the economy.

Somehow, prior to high-frequency trading, the world had plenty of liquidity, extremely narrow spreads, and not any more volatility than today (and probably less). I think the (real) world will do just fine without HFT.