Slashdot Mirror

CPU/GPU integration is for farmers, to paraphrase Seymour Cray.

CPU/GPU integration has much lower latency than discrete a GPU. The HSA based AMD chips pass data from the fast, single threaded, fast branching core to the massive array of relatively slow FPU units in a few nanoseconds.

Which is why HSA benchmarks seem to work so well

http://www.tomshardware.com/re...
http://wccftech.com/amd-kaveri...

If you want fast comptuting, low latency comms is where it's at :)

Good thing AMD never has such issues...

http://www.tomshardware.com/an...

As a note: I don't even use the onboard AHCI/SATA controller, as I have a dedicated PCI-E SSD, and this still prevented Windows from booting.

SSD is top spot for performance on devices & better filesystems exist for them (less overheads), ala -> http://www.tomshardware.com/re...

HOWEVER:

That's mainly about EXISTING FILESYSTEM's & their OVERHEADS (even though 'optimized' for SSD)!

There's filesystem designs that OMIT a LOT of the overheads those have that make that faster/lower latency/higher iops SSD do even BETTER!

(Since the filesystems coming soon don't HAVE to account for disk heads & sector/cluster overheads from legacy filesystems out there now)

Ala -> https://en.wikipedia.org/wiki/...

* There's SOME choices in that area iirc, from some of those listed there. There's probably other or possibly BETTER ones in the works too...

(Anyone knows or more or others? Let me know & "TIA", since I can always stand to learn a new thing too!)

APK

P.S.=> That's what we'll be seeing when they eventually hit "performance walls" on what they can do HARDWARE-SIDE on SSD's, ultimately imo & experience... they amp up the hardware to as far as they can go, THEN, the software-side catches up (mainly for "competitive purposes" really, to be "THAT MUCH BETTER" in company x, than company y their competition is)... apk

Fine, I don't know the reason for the use of the 780, maybe it's a GPU compute platform.

Assuming it's games though (and this'll probably be likewise for modelling/CAD), the 780 is still going to be overkill. The CPU is still going to be a major limiting factor, even if it's an i7 940XM. The 1x PCIe throughput you pull through the ExpressCard slot is also going to be a huge problem. Look at this performance dropoff with a 9800 GX2 when comparing x1 with x16. Then consider how much time has passed since the 9800 GX2 was top of the line and how much more memory bandwidth modern games are using.

The card is going to be hobbled by the slow CPU and tiny pipe, so again, if the purpose is just to get a working PC out of this, it would have been far more reasonable to just buy a new one. Hell, if you need to do some sort of hack, why not get the laptop's monitor/hdd/keyboard/etc. working with the new computer?

Intel's CPU will be an option, but surely you can get it with AMD as well:
http://www.tomshardware.com/ne...

Fury X beats Titan X at many games at 4k resolution and even more at 5k.
Fury X beats 980 Ti (pre-emptive release by nvidia, that anticipated Fury X) at 4k, whit the same recommended price.

Now, these boxes will have to of Furys.

FuryX also has a nice "FPS cap" feature, which allows it to drop frequency to save power when you are beyond reasonable FPS (i.e. 90+, actual number depends on your taste).

Had they chosen to not allow Intel's CPU it would cripple it, but with i7 option, it's a great product.

FX-8350 isn't that bad compared to i7 4790k, but not at gaming.

Anyway, Intel CPU inside it will be ONE OF THE OPTIONS. AMD CPU configurations will also be available

We have Quantum designs that feature both AMD and Intel processors, so we can fully address the entire market.

http://www.tomshardware.com/ne...

The problem is, following this logic they should have used Nvidia GPU parts as well. This showcases AMD's weaknesses more than anything else. Its confirmation of what everyone already knows, AMD cant make low heat parts.

The Fury X is quicker than the GTX 980 and in half of the games seem to be quicker than the Titan X it seems:
http://www.tomshardware.com/re...

So why the fuck would they use an Nvidia card if they got as quick card themselves?

I know it may not support feature level 12.1 of Direct X but that's it. One advantage is that it will allow you to get a cheaper FreeSync monitor.

1. If you're constantly upgrading you're doing it wrong. You should only buy a new GPU that moves you up at least 3 tiers.

2. I guess you don't play any MMOs such as WoW, any RTSs such as Starcraft, or use any mods for Skyrim, Minecraft, etc. Consoles aren't always even viable in some cases. If all you care about is dumb button mashers such as Diablo 3, or Destiny then sure, knock yourself out. Meanwhile some of us will be checking out the free Path of Exile and other PC only games.

3. I've been building my custom gaming rigs since 1990. These days you can save a bundle by getting parts from NewEgg or MicroCenter. Order an i7, with a Hyper 212 EVO cooler, 16 GB RAM, a 256 GB SSD, and that $500 you spent on a GPU will give you a gaming rig that will last **years**.

4. I guess you don't care about framerate or resolution. Consoles can barely do 1080p @ 60 fps. Meanwhile over in PC land we're gaming at 120+ Hz with G-Sync / FreeSync Desktops resolutions are running at 2560x1440, the incorrectly labeled "2K" @ 3840x2160, or even higher such as Vanishing of Ethan Carter @ 8K.

Consoles are basically a 4 year old PC. Apples and Oranges. Different strokes for different folks.

@binarylarry: "It was great knowing you, Oculus. Hopefully Valve has a little more sense. They seem to have better hardware than Oculus."

'Engadget reports that Valve Software actually helped Oculus VR create the Crystal Cove VR headset that won so many awards at CES 2014 earlier this month ref.'

I usually use Tom's guide as a guide. Access with some kind of ad blocker though...
http://www.tomshardware.com/re...

Well, WD Red drives certainly have lower power consumption than the equivalent standard-consumer WD Green drives. See this test. The power use when idle is 3.9W for the WD30EFRX (the Red drive) and 6.1W for the WD30EZRX (the equivalent Green drive), and there's a similar difference when they're operating under different sorts of task loads. Keeping the heat down is important when you're building a RAID, so this should improve the reliability.

Other claims I've heard for Red drives are that (a) as the GP said, they're more vibration-resistant, or produce less vibration, and (b) they report a possible failure immediately as a failure so that a RAID controller can drop them from the array, whereas a standard consumer drive will keep trying to recover the data, since it's probably not in a RAID. I don't know how practical (a) is, but (b) seems like a sensible change to the firmware in this application.

So, I would expect the Red drives to be more reliable, in a large-scale RAID setup, than the Green drives. Do you actually have any test or reliability data that contradicts this?

Why not?

In AMT 7.0, Intel makes it possible to use a 3G cellular signal to send that remote kill command, greatly improving your chances of deactivating a stolen computer before it gives up any sensitive information. Administrators can use similar technology to reactivate the computer once it is recovered.

The BjÃntegaard metric is used to calculate the bitrate saving achieved by the test
encoders, based on the PSNR scores.

The problem with a lot of these studies is that the metrics don't always work that well. For example, look at the image comparison on pages 26, 27 and 28 in the NetVC presentation. The first codec on page 27 has a better PSNR score than Daala on page 28, yet to me the image compressed by Daala looks better and has more detail.

Daala's not ready yet but it's been proposed as the basis for the NetVC implementation. NetVC will probably end up being Daala merged with other contributions.

It could simply be that a few companies want to get more money from H.265.

I think it's this. The companies involved weren't satisfied with the licensing terms the MPEG LA had decided on so they formed a competing pool.

Whatever their motivations, it's exactly these licensing complications that make it clear that the best way forward for video on the Web is to develop a high quality, royalty-free codec that everyone will implement. A video codec that gets standardized through the Internet Engineering Task Force is more likely to the implemented by the likes of Apple and Microsoft than codecs developed internally at Google. NetVC aims to achieve that.

Firefox does support opportunistic encryption for HTTP now which allows self-signed certs (since it's not showing a lock icon or anything like that). Not sure if there's a way to pin it to require future visits to an opportunistically encrypted site also be encrypted and use the same cert (that would make it like SSH). Hopefully the other browsers will also add support for that and web servers will have a dead-simple configuration for it (since you don't care what the cert is, the server can just auto-generate it like an SSH server does).

This is why it's important to have royalty-free codecs for the web that everyone is free to implement. You can choose to do your own implementation of a given codec and take direct responsibility for the security of the implementation, or ship your preferred choice of third-party implementation directly integrated with your product without any patent licensing hassle. I just hope Opus audio and NetVC video become ubiquitous sooner rather than later.

I completely disagree. The majority of the HPC realm still uses Nvidia only because they know CUDA and not because of any technological advantage. AMD has held the line at not allowing sloppy programming methods into their OpenCL compiler and that has held back a lot of HPC users from jumping ship. You can even see this in many complaints from open source projects, like Blender, where they refuse to produce proper multi-threaded code and rely heavily on the CUDA compiler to do the work for them.

The rest of your complaints, "shitty drivers", "piss-poor memory handling" and "worse performance per watt" are also bogus. I own or manage machines using a large number of Nvidia and AMD video cards, and have seen as many driver issues between the two that neither has come out worse. This is a typical fanboy stereotype that keeps being repeated with no real fact behind it.

Your second complaint is seen a lot in programming forums, but I have never seen anyone do a proper write up of any memory issues with any of AMD's generations and most of the conversations lead me to believe it was an issue of the programmer's personal preference not wanting to learn a second platform with less market share than an actual technical issue. Most of these issues would be alleviate if the programmer would just use a common optimized library and stop trying to redo the work themselves.

Lastly, AMD's offerings have historically produce more performance per watt and their latest offerings continue that trend. This, besides the bit shift ability you mention, is also one reason why AMD was used for Bitcoin mining and supercomputers.

http://www.tomshardware.com/re...
http://www.green500.org/news/g...

Now, my latest personal computer has an Nvidia GeForce GTX 980 inside because I more often need to fix CUDA code and noticed some of the games I wanted to play ran better on it (again, from the game designer's preference and not a technical merit). I personally own eight other video cards across AMD, Nvidia and Matrox (who use AMD GPUs these days) and three generations for testing.

And I am only sticking up for AMD because I admire their push to get people to code for multi-core better. Nvidia has been too conciliatory in the last six years in that respect, which is fine for their revenue stream and market share but not a good thing in the big picture for the broader computer industry. Since Moore's law has begun to slow, we are going to need a massive shift to multi-core optimized applications and we need programmers ready for that day.

AMD seems to be ready with the tough love to get everyone there while Nvidia keeps enabling bad behaviors.

I wouldn't be making this comment if the OP didn't mention loving Valve. I think it's absolutely lovely that Valve has done this, but I kinda hate Valve.

Valve seems to have problems internally that's lead me to believe that it's not worth hanging hopes that they'll usher in a true golden age era of PC gaming.

First, there seems to be some bizarre drama going on inside of Valve, as evidenced here. The flat structure isn't as idyllic as once thought.

Then there's problems inside of their online market place. Shit just doesn't work. Valve doing a bad job policing Greenlight. I'm not even going to bring up Hatred.

And of course, where the bloody hell is Half Life 3? Or the steambox? Or a stable release ready version of steamOS?

Valve can't be all things to all people and try to spread itself as thin as possible. It winds up doing nothing well and it's all starting to fall apart. I just hope they let us know what happened to Barney Calhoun before the company shits the bed.

I remember seeing a Conner hard drive like that many years ago. Tom's Hardware discusses it.

> One thing I've tended to wonder, why have a single read-write needle on conventional drives (especially in multi-platter situations).

It was tried like 30 years ago.
Extra complexity leads to increased costs and more failures.

Here's one where the cheating was exposed when leaked benches for new Macs surfaced before Cinebench had been updated to take them into account.
http://www.tomshardware.com/re...

Those scores require a bit of context, though. The 32-bit build of Geekbench uses x87 code, for starters, so it isn’t optimized for any of the other instruction set extensions that Westmere-EP or Ivy Bridge-EP support. Getting close to Apple’s claim of doubled floating-point performance requires software compiled with the AVX flag. John Poole, the founder of Geekbench, posted several other reasons why the next-gen and previous-gen Mac Pros might be separated by such a narrow margin.

The leaked result was run using the free 32-bit build of Geekbench on a pre-release build of OS X Mavericks. Switching over to the paid 64-bit build of the benchmark adds SSE support, though that’s still a pre-Pentium 4 extension. Tab between the 32- and 64-bit runs on Xeon X5675-based systems and you’ll find that the SSE-capable build averages 14%-better performance.

Curious as to how the very same 12-core Xeon E5-2687 V2 compared in Windows, I ran my own test on a 64-bit build of Geekbench and scored in excess of 30,000 points—more than 25% faster than the leaked number. The individual sub-tests showed both Xeon E5-based platforms trading blows in the integer and floating-point components, but clearly a more real-world comparison was needed in order to establish the new Xeon’s performance in a workstation environment. Fortunately, I have the upcoming Xeon E5-2697 V2, the upcoming Core i7-4960X, an existing eight-core Xeon E5-2687W, and a Core i7-3970X.

This kind of thing isn't exactly a revelation. Benchmarks have been tainted by Intel and the ICC for ages. The real problem is that a lot of actual software is as well, so in the end the artificially-gimped performance reflected in the benchmarks translates to actual usage. Even among fairly-compiled programs Intel's parts typically maintain an IPC advantage, but it's no where near to a degree that would justify the cost difference. Add in nVidia's moneyhatting and gameworks bullshit, and you've got AMD taking it from both ends. This sort of thing should piss you off regardless of what brand you prefer because it stifles competition, increases prices, and retards progress.

There are some results from LibreOffice Calc at the bottom of this page.

other articles are claiming eDP 1.4a can support 8K (actually 7680 x 4320) resolution at 60 Hz http://www.tomshardware.com/ne...

Don't worry, they already have some nice features:
http://www.tomshardware.com/ne...

Kind regards,
Roel

"the current ATI hardware is a complete non-starter. There's really no level at all where it can be justified."

That's simply not true unless you're talking about edge cases like gaming on Linux.

Tom's hardware has the latest Jan update for graphics comparisons. The R7 260x is best bang for buck but the 750ti is not quite as good value but as you don't like ATI it may be a better choice. All really comes down to why you are upgrading and what you play though.
http://www.tomshardware.com/re...

It's worth doing some reading, to understand the differences between the switch types. Here's a good description of three of the switches. You likely don't want the really loud ones - I recently bought a keyboard using Cherry Brown, which are tactile, but a bit quieter - it's still loud enough that my officemates had to get used to it, but at least they didn't kill me.

A lot of the sound from the mechanical keyboards with non-clicky switches like the Cherry reds and browns is from the keys bottoming out. You can add rubber o-rings to the keycaps to get rid of that bottoming out "clack".

It's worth doing some reading, to understand the differences between the switch types. Here's a good description of three of the switches. You likely don't want the really loud ones - I recently bought a keyboard using Cherry Brown, which are tactile, but a bit quieter - it's still loud enough that my officemates had to get used to it, but at least they didn't kill me.

And while Intel will "begin offering 3D NAND drives in the second half of next year" Samsung has been doing just that for a few months. For insgtance, here a review from last June: Samsung 850 Pro SSD Review: 3D Vertical NAND Hits Desktop Storage. But, then again, since when has IT World needed any facts? ;-)

RT.

Also - couldn't you actually just sign the drivers that are needed for trim? What prevents that?

As the author of the popular "trim enabler" software (which patches the original apple drivers and so causes the original drivers to fail the kext signing check) puts it:

"all of Apple’s AHCI SATA drivers are closed source and undocumented, which makes it impossible for me to create my own Trim driver and get it signed."

Which is also the reason why there are no trim drivers available from hardware manufacturers like Samsung, etc. No access to Apple's driver documentation - no signed trim drivers.

So some tiny Austrian company manages to do something Samsung can't - http://www.tomshardware.com/ne...

BTW, it would be kind of awesome if the computer hardware testing sites incorporated sound tests into their general testing of stuff.

You mean like this:
  Tom's Hardware: Sapphire's Vapor-X R9 290X 8GB - Temperature, Noise And Power.

Actually I continuously get frustrated by "enthusiast" computer sites reviews who seems to being entirely lacking in technical knowledge when it comes to anything beyond quoting the manufacturers press material. Half of them might as well have a companion site reviewing shoes and fashion tends given their display of technical ignorance.

People are leaving Firefox because for a long time Chrome was flat out better (not counting add-ons) - faster, more stable. Firefox has been kicking ass in the last few browser comparisons at Tom's Hardware ( http://www.tomshardware.com/re... ) but I think public perception hasn't caught up.

And Mozilla is probably happy to take Google's money from the Google Ad Network instead of direct grants, if that's what it takes to keep the Mozilla Foundation open. What they can't do is survive on end user donations. The number of major open source projects with as many developers as Firefox that survive on that model is near zero.

Which, of course, means the devices have to be identified by the system and a driver has to exist for whatever they identify as.

I was relatively sure that part of the problem was that the system's USB controllers could also be coopted by plugging in a bad USB device, whether the system recognized said device or not. At that point, as long as your system recognizes any USB devices, you're toast. Correct me if I'm oversimplifying the problem.

Firewire is still fairly widely used in media production, and the devices using it include cameras, control boards, and DAT decks, which do get passed around.

Yes it is, although disappearing rapidly, at least from the set of cameras I looked at last year.

And without USB, where do you think you'd plug that thumb drive?

I could see thumb drives going to a ROM only version in the near future. That solves the trust issue with at least thumb drives. I know the arguments against ROMs, but if no one will use your drive because of a virus threat, well, then you have no market.

Drop 2 sets of file transfers through the disks on that one hub, and see what happens.

Okay, so you're doing two copy operations and are surprised when seek times slow them by more than 50%?

I have about 10 disks hooked up and was copying files between 3 sets (3 full speed copy operations, including 2 SSDs) with each disk capable of +100MB/s on large file sequential read/write speeds.

Are you truly that dense? It is 1 pair of drives per copy operation, going through 1 USB controller. Yes, the USB connection will slow down. In your case, I'll also bet your SSDs are on SATA connections, those are 1:1 internal, you mention that yourself.

I'll lay it out specifically:

• internal sata SSD 1 (Samsung 840) copy operation to USB hosted disk 1, controller 1.
• internal sata WD Caviar Black to external USB hosted disk 2, controller 1.
• Internal sata Samsung Spinpoint to external USB hosted disk 2, controller 2.

All externals are newer WD or Samsung disks. 2 of those disks have eSata connectors on them. I've hooked them up as stated, and via eSata to split the data transfers across multiple busses. The throughput increased dramatically in the eSata configuration, leaving only the USB bus as the sucky culprit. Maybe it is the super cheap controllers, but USB has never come close to FW transfer speeds under anything but ideal 1 device per controller scenarios for anyone I know.

I expected the Nexus 6 to have a microSD

Why would you believe that?

Apple taught everyone that it's much more profitable to included cheap, on-board memory at a premium price. Also, Google's insistence on the cloud was made clear when Matias Duarte said explandable storage was going the way of the dinosaur.

Windows 8.1 sends my every search query to Microsoft if I don't block them by IP at the DNS, router, and hosts file levels.

To configure Smart Search, you need to visit PC Settings, the new Metro-based replacement for Control Panel, and navigate to Search and Apps, and then Search.

Use Bing to Search Online. Enabled by default, this option determines whether Bing-driven web results appear in the Search results page. If you set this to Off, you will no longer see these results (and will only see Everywhere, Settings, and Files as options in the Search pane).

Your Search Experience. This option---available only when Use Bing to Search Online is set to On---determines whether Bing personalizes its search results for you and for your location. If you're going to leave Bing searching enabled, I recommend leaving this on its default: Get Personalized Results From Bing That Use My Location.

My advice? Leave it alone and give it a shot. But if you do end up wanting to turn off the Bing web integration, that's how you do so.

Windows 8.1 Tip: Configure Smart Search

It regularly disables my wireless card so that it can reset it and verify my connection by reestablishing the link with Microsoft's privacy-invading servers.

On occasions, the system is programmed to turn off the Wi-Fi adapter, when idle. This might be the reason for your spoiled Wireless connection. Troubleshoot the situation by deactivating this feature of Windows 8.1 and see if it works out.

Press Windows key + W on your keyboard to initiate Start search.
Type Network and Sharing Center in the search box and hit Enter to open its window.
In this window, choose your Wi-Fi network and the Wi-Fi Status screen will appear.
Click the Properties button near the lower left corner to open another window.
In Wi-Fi Properties window, click on the button titled Configure. Go to the Power Management tab; uncheck the following option and click OK button.
Allow the computer to turn off this device to save power.

If the problem persists, replace the adapter.

Windows 8.1 has a kind of crash I've never seen in any Windows version until this one: memory management.

The reason you've never seen this crash before is because it is also most likely a hardware error. MEMORY_MANAGEMENT error in Windows 8.1

In twenty years as a Windows home user, truly bizarre and outrageous behavior has always come down to a hardware problem --- sometimes an easy fix like resetting a chip or board, sometimes a warning that the system is EOL Time to pull the plug.

Now let's mention the one and only discussion we've seen about Windows 10 having a keylogger embedded in it while overlooking that random forum posters have said that it's because the OS is in beta but Microsoft has never confirmed that the keylogger would be removed.

There is no need to read the random forum post. Privacy Statements for Windows Technical PreviewThe Win 10 preview explicitly targets the enthusiast and the IT Pro. It is not an open public Beta as that term is generally understood.

https://en.wikipedia.org/wiki/Gigabit_Ethernet
Hardwire - CAT 5 cable is not likely to work consistently, Cat 6 and Cat 7 cable is recommended, shorter is better, but 100m is possible.
Fiber to your PC also works, but the fiber modem cards are pricey.

https://en.wikipedia.org/wiki/Gi-Fi
https://en.wikipedia.org/wiki/Gi-Fi
https://en.wikipedia.org/wiki/Wireless_Gigabit_Alliance
https://en.wikipedia.org/wiki/Wireless_Gigabit_Alliance
http://www.tomshardware.com/reviews/wi-fi-802.11ac-router,3386.html
http://www.tomshardware.com/reviews/wi-fi-802.11ac-router,3386.html

These new WIFI devices are fast, but much slow than a cable connection.
However unless you are trying to store point cloud images in the cloud it is unlikely you will tax a 802.11ac wifi link.

Point cloud cameras and the related infrastructure are expensive, both in data size and $$
leica-geosystems P20 $115,000.00.

Reminds me of an old browser memory comparison - http://www.tomshardware.com/re...

Opera looks pretty terrible, until you realize what was going on - for example, it installs more feature complete than these other browsers (fewer extensions needed), by default pre-fetched links much more aggressively, and held the complete history/state of closed tabs in a 'trash bin' ~ perhaps a rarely used feature, but supremely handy when you really need it. Similarly, IE looks pretty good until you realize that this is because it's hiding RAM usage in system threads, and is a disaster once you use it for webpage viewing. Safari, on the other hand, shows double the next browsers RAM usage at rest for all the shims to get it going on Windows; but it's rather svelte under load otherwise. Firefox seems to have figured out the impossible, using very little ram for many-tabs - but after the first or second 30 second interval, it's wrote the session out to the disk, and lagged when switching to these tabs.

Unless you're short on RAM, it really doesn't matter on modern browsers, all the major players are pretty decent programs - it's not like one of them is an iTunes-esque hot mess. Responsiveness & compatibility can be broadbrushed, but otherwise the importance of features, available extensions/configurability, site compatibility, mix/match of java and page rendering engines, and UX preferences will vary somewhat-to-wildly between individuals. Use the browser that best matches for you and the sites you frequent. Or install a bunch of them, they can co-exist. Personally, I cannot stand the look and feel of Chrome they are all sort of converging towards (I prefer to use older Opera builds in the 11.xx-12.xx range). YMMV.

I think the reason the specs aren't increasing much is because the pace of hardware improvements isn't moving as fast as it used to. Nowadays, you pick up an i7 and 16Gb of RAM, your favorite video card, toss an SSD in there and you've basically hit the limit.

All we're getting these days is more cores as the whole gigahertz wars ended 10 years ago.

Actually, the performance difference between an i3 and an i7 is negligible.
http://www.tomshardware.com/re...

You get 80% of the performance of a $570 i7 if you buy a $125 i3. For $250 you could get a dual processor i3 that was faster than an i7

User experience is probably pretty key; people will buy the cheapest SD cards they can find (despite some people buying their iPhones up-front, most get them 'free' with plans and like the 'cheap' route). Without some sort of quality control on the cards you could get some pretty dodgy performance. It seems Tom's Hardware did some performance testing of them a while ago, with the slowest random write being 25x slower than the fastest and the slowest random read being 4x slower than the fastest. Those are some pretty large differences.

User experience is probably pretty key; people will buy the cheapest SD cards they can find (despite some people buying their iPhones up-front, most get them 'free' with plans and like the 'cheap' route). Without some sort of quality control on the cards you could get some pretty dodgy performance. It seems Tom's Hardware did some performance testing of them a while ago, with the slowest random write being 25x slower than the fastest and the slowest random read being 4x slower than the fastest. Those are some pretty large differences.

I know there's some humor intended here, but heck i figure it's worth mentioning that Dual/Quad core atom's rival if not beat the pants off P4 processors.
I mean, just look at This.

That has been done and abandoned. HPT (head-per-track) drives were popular way back, but were a bear to keep aligned.

Hmm... That's not what I heard.

Taobao has a 64GB SSD @ RMB200 (US$32.47) before shipping. That's 50 cents per GiB for you, with a SM2246EN Silicon Motion controller and Intel 25m IMFT NAND flash. According to Tom's review on this controller, this can hold its own against Samsung 840 EVO.

Hasn't this idea been debunked many times? Oh yeah, it has, at this point it's only fools and trolls who say it. If you can't imagine spending that much on one computer, even a Windows PC with power beyond anything you'll ever actually need, I suspect you must have stopped buying computers shortly after that first one. I've worked for businesses that don't mind dropping $10,000 on a single system, and that's not even the upper limit. That doesn't mean everyone's going to go out and spend that much or would even notice any benefit from doing so, but if you know you're a criminal stealing it anyway, you may as well go all out.

As always, it's more complicated than that.

The zealous ARMists are mostly those people who bought AMD during the speed-wars of the 90's. Intel won, and they can't stand that. ARM is their next hope for trying to get revenge on Intel (even though Intel is more than capable of optimizing ARM designs and fabricating them at half the size of any competitor). The minor fact that Intel's Atom line had surpassed ARM at every benchmark except the "power consumption when doing nothing" in 2012 does not in any way alter their firmly held belief that ARM is infinitely efficient and the future of computing.

For a person owning a single workstation power consumptions means little.

Even if you are an intel fanboi, look at the single core performance, in a comparable test.
http://www.tomshardware.com/ch...

I don't thing the newer chips will be much faster at multi-core raytracing either.

Annual global IP traffic will pass the zettabyte threshold by the end of 2015 ...

So, at the end of 2015 it'll take ~17432 hardware AES engines just to decrypt the traffic*, ignore all the possible overhead of hashing, public/private key computation, etc. So, perhaps you're right on that front. Realistically, I don't see them actually managing anything close to 2GB/s sustained as once you start including any actual storage for that sort of data, the I/O delays are so substantial.

*This benchmark implies closer to 6GB/s in software/hardware, but it's hard to know from a benchmark how the real world results would be given how many times sha hashing or other steps might need to be taken. In fact the results imply that the first step would be to involve much more hashing into the protocol (with several iterations, several recomputations steps midstream, etc) That's one reason I included the caveat of scaling the problem to counter Murphy's Law as a protocol with a static iteration count that used an "effiicient" amount of CPU time for a 1994 computer * the world population's traffic is much less of an insurmountable obstacle today. If the protocol encouraged scaling the problem up, then at least the problem could be made several hundred times harder to crack, although I'll admit that eventually the new protocol would be done in hardware and eliminate most/all the gains. It's more of a rat race than a final solution. :/

Not that I'm aware of... you need a PC, and nVidia is pretty much the only game in town too (apparently it's possible with AMD but painful and inferior). Myself, I have it hooked up to a home theatre with 720p projector (sadly, 1080p 3d gaming is not possible yet... we need HDMI 1.4b/2 devices for the necessary bandwidth).

The interesting thing about 3D gaming is that not all games work equally well. The 3D stuff generally works fairly decently, what it mainly comes down to is the 2D elements. For instance you'll get a menu in front of the 3D scene, but it's depth isn't set correctly so your brain will be convinced it should be behind the scene... really messes with your eyes in an unpleasant way. :)

So with most games it's somewhat "accidental" how well they work with 3D (or don't). A relative handful actually take the effort to address such issues during development and get fully "3D Vision Ready" certified. nVidia keeps a big list of games (and their 3D playability) here. Fortunately with some games, mods can fix the 3D deficiencies.

At recommended viewing distances, 4K resolution is difficult for most of the population to detect a difference in.

Um... just no... that is completely and totally false, I wish people would stop repeating that nonsense... Maybe YOUR eyes suck and you can't see a difference, but put them side-by-side, sitting 6 to 10 feet away, the difference is clear and obvious to most people...

I speak from experience...

The problem with 4k monitors is that they have slow refresh rates (30hz?), slow response time, and all the usual non-IPS problems like poor viewing angle and color. None of which matters terribly for programming (save response time which might make scrolling a bit blurry.)

More wrong information. 60hz 4k panels are out now, and they don't have poor viewing angle or color. You simply need DisplayPort to get 60hz (which anyone buying such a monitor today should have).

http://www.anandtech.com/show/...

http://www.tomshardware.com/re...

60hz, IPS viewing angels, just crazy expensive at $3,500 (actually below $3K now, give it a few years to get cheap).