Mark77

Its certainly possible that Google has limited connectivity to New Zealand and has chosen not to locally cache the 1080p versions of the videos. So when you play them back, they might have to be transferred from other regions. If there is any disruption to this process, your playback will be impaired. That is why I suggested that you make sure that you eliminate your own hardware as that problem, and try to watch them on an internet connection at a school/library/coffee shop/restaurant, etc., to see if the results are different at 1080p resolutions. Just to eliminate your hardware and Google from the equation. "Speed tests" are pretty useless typically.

Lots of ISPs will give you a new IP if you change your computer's (or gateway's) MAC address. This can be accomplished with a single command in Linux, or through a GUI with dd-wrt and presumably similar 'consumer' gateways.

New Zealand, eh? Might be more of a Google thing. Do you know if playback works with other Internet connections on your island/country?

All you should need to know is your professor's public key. Do you have a public key for the professor? I'm thinking a site like "startssl" would generate a public key (as well as a private key) for yourself, but in order to communicate with the professor, you would need his public key installed into your system accordingly. That is probably why you are receiving the rejection notice.

Weird... Yeah the hard drives in those old printers were mainly for things such as font and form storage. Today we don't think anything of sending 100mb to a printer over Ethernet or USB for a print job, but "back then", printers, particularly in institutions, were often wired with RS-232 or other very slow connections. So being able to render pages without transferring the fonts and forms associated with a page was a speed-up in some limited applications. But probably more hyped than actually used.

Some older printers had hard drives. For example, Lexmark Optra R/R+/L etc. had an option for 100mb hard drive (yes, 100mb, as in, 1/10th of 1gb) drive to store fonts and forms. Cost was $700 apparently in 1996 or so. http://www.walshcomptech.com/ohlandl/printer/optra_r+.html#Hard_Drive I also remember a hard drive option of some sort for the Canon CLC-series of colour laser copiers, circa 1991-1992 or so. The hard drive was associated with the image processor capability. Cost of a half decent setup was $60-$80k in 1992 or so plus consumables.

Yeah definitely. However, RAID-5 sucks and good use cases are really few and far between. RAID-6, if you absolutely have to, with decent hardware, UPS, battery backup, etc. But otherwise most people are better off with normal redundant mirrors (ie: RAID-1). If you configure a 2-drive RAID-1 as a RAID-10 on Linux md, it has the read performance of a RAID-0.

FlightSim X is pretty tame. Now P3D, I hear, is a crazily multi-threaded resource hog for which you might want to seek out those dual 8-core CPU machines that people can hack together for $500-$600 in parts (+ RAM and a case that takes the E-ATX boards).

Yeah if its not a common laptop, then good luck. I'd set your searches along the lines of generic things that could be done to Toshiba laptops. Maybe there's something applicable or some pads on the PCB that could be shorted, etc.

A degree in CS can make you a lot of money if you go to a top school and are a relatively top student. It can make a lot of money if you're lucky too. But if you graduate somewhere in the middle, even the upper middle, the problem is that you'll be considered 'overqualified' for very mundane jobs like tech support. But under-experienced and under-skilled for the top end jobs. In a nutshell, its a big gamble. If you're 15, already doing Java code and understanding big-O notation, binary trees, graphs, and all that, then by all means, go do CS at a good school, and you'll be making $200k/year at 30. Okay, that $200k/year will probably be in a place where a house costs $1M (not that you'll have much time at home to enjoy it!). But its still good money for a 30-year-old. But if that's not you, there are definitely some other great pathways into the IT professions which are still quite useful and don't involve university. For instance, employers for "network engineering" positions generally place more value on vendor (ie: Cisco, Microsoft) tech certificates such as CCNA, CCIE, than they do in Computer/Electrical Engineering degrees. And plenty of people find jobs on the help desks, and eventually move into areas such as project management, earning decent money. The big elephant in the room these days is that the employers have brought close to a million IT workers from foreign countries (read: India) where even $60k/year is considered an extravagant salary, on guest worker visas. Displacing local workers. So at some point, if your job could be done offshore, or by someone from a foreign land for cheaper, chances are, its going to be under pressure.

Throw the name of your laptop and "locked BIOS" into Google. What laptop are you using anyways? If the laptops are similar enough architecturally, in a pinch, you might be able to remote its SSD and install it in the other laptop. I know I can do this with my 2016-built laptop and my 2007-built laptop (it has to do a few things with drivers, but otherwise, Windows 10 handles it very gracefully). Is Bitlocker or a TPM involved? If so, hopefully you saved your recovery keys somewhere.

As the above poster indicated, in a nutshell, you bought the wrong CPU's. *However*, the CPU compatability list indicates that the CPU you bought should be compatible, at least in a single CPU configuration. However, you will need to flash to BIOS version P1.4 or above: http://www.asrockrack.com/general/productdetail.asp?Model=EP2C602-4L/D16#CPU So this might be your issue. Fortunately there's a socketed EPROM on the board, so you might be able to flash it externally. Alternatively, obtain a supported Sandy Bridge E5-based CPU (see the above link) and fire it up with that.

Bzip2 seems to be popular for distributing Linux software these days. But seriously, get a bigger HDD if you're running into these issues.

If you want to help him remotely, a "Q" chipset board would be something I'd suggest, and figure out how to get Intel AMT working, particularly the remote KVM feature which is embedded into it. Forget the HDD, go with a SSD, even a cheapie 128gb one. 16gb, fair enough. Although SSD equipped machines tend to be far more graceful in dealing with lower RAM situations because of their sheer speed on swap. What current machine does he have that he's stuffed 12gb into? Perhaps look into SSD'ing that, and maybe upping the RAM instead.

Every so often, I've had Linux get into a state where its still sort of responsive, but you can't get into it with conventional ssh. So a conventional watchdog timer (or even a hardware-based one) might not trip the machine into a reboot. Also, if you ever want to do something like upgrade an OS or kernel remotely, every so often, something bad will happen for which the only extrication from such would be either a console visit (not too desirable, even if you're close to the machine), or the use of some sort of out of band management solution like IPMI.

Those Asus dongles are for boards that very specifically allow for that as an add-on. Have you considered using a "Q" motherboard that supports Intel AMT with a vPro chip? The price is barely much higher than the H or B motherboards. But seriously, proper Supermicro boards that support IPMI aren't that expensive these days. What's your time and the hassle of getting to your machines worth?

Strange. Can you swap the storage from one machine to another and see if the problem exists? Maybe the NIC is fried? Do you have any verification that the NIC works at all?

Can you buy HGST?

http://www.intel.com/content/www/us/en/processors/core/core-i7-lga-2011-datasheet-vol-1.html Have at it. Pin groups 26 through 45 are power or ground.

It could be built as both (so the vendor doesn't have to develop 3 unique boards). And one is just selectively disabled either permanently (much like Intel turns off cores), or by the board asserting a pin to flip it from utilizing one architecture to another. Yeah FPGA's are embedded in all sorts of communications equipment, for various functions that in the past would have been dealt with using custom ASICs with long development cycles and no in-service customizability. Their use as business application software coprocessors is a very rare use of them. Obviously not zero, but we're talking only thousands/tens of thousands of units compared to millions of FPGAs shipped each year. In such applications, the micro may very well be mostly disabled and its really immaterial whether its ARM or x86. In a typical FPGA embedded application, an embedded micro cuts down the overall BOM and reduces system cost.

For some applications, it may be possible to put both an x86 and an ARM core on the same die, and selectively disable one or the other. For self-driving cars, for example, they are likely to require redundant CPU's with different software architectures performing redundant calculations to meet the system reliability requirements. So one chip could boot up as an ARM chip, and run ARM code. The other, identical chip, comes up in x86 mode, and runs x86 code. And a third chip comes up randomly assigned to one or the other. A comparator would compare the results of the 3 chips and trigger an alarm if there was a discrepancy.

True, but on a typical CPU, most of the pins are for power or for ground. Simply to get the 100-200A that the chip actually requires from the motherboard's PSU, into the chip itself. I was questioning whether or not all those pins were required, as opposed to just one giant pad for Ground, and another for Power. Ground, for instance, could even be the other side of the package, ie: the heatsink side of the chip. If 2000 pins are power/ground, you could save 1000 pins if you just turned the heatsink into a giant ground.

Maybe in that very specific realm, but there are plenty of communications applications, in particular, for which trying to "code" them in OpenCL or whatever would be absurd. We're talking embedded systems here, not just specialized coprocessing of banking applications.

2000+ pins are either Vss or GND I bet.

LOL. Do you even know what a HDL is?