Posts by Ammaross Danan 1042 publicly visible posts • joined 23 Sep 2009
"let customers have access to purchases they have made outside the App Store." - NYT
Which is obviously stating that if you purchased content outside of the App Store (think of your eBooks with Kindle) you won't be able to access them on your iDevice. Likely, there will be some way to tag in-app content that was purchased through the App Store, but since I don't develope for iOS, I can't fathom what that would be.
Dear Customer;
We regret to inform you that we, Amazon, as well as many other publishers, will no longer be supplying or supporting apps for the Apple iOS platform. We have removed our apps due to a 30% commission required by Apple which we would have no choice but to pass on to you, the Customer, for continuing use on their platform. However, we fully support apps on other popular platforms such as Windows and AndroidOS. Below are links to our website for information regarding our Kindle reader as well as the Archos 101 tablet.
Thank you for your understanding.
http://www.amazon.com/Kindle-Wireless-Reader-Wifi-Graphite/dp/B002Y27P3M
http://www.amazon.com/ARCHOS-101-Internet-Tablet-16GB/dp/B00422TXZA/
Yeah, fake letter, but honestly, would have a serious impact. Why? Likely people have an iPad because there was no alternative at the time. Perhaps they already had an iDevice and wanted consistancy? I think the obvious solution to this 30% tax levy is to pass a margin of that on to the consumer using the iDevice. Perhaps tag which books have or haven't had the AppleTax paid on them, and give end-users the option to pay the extra $$ to view the material on their iDevice reader. Of course, there'd be a little question-mark people can touch to read why they're being charged (again/more) to view the item on the iDevice, at which point this tax levy can be explained. Such negative publicity, as well as being spoon-fed an alternative, is content-provider's best hope, excepting charging EVERYONE just a little bit more to help even the playing field.
Just wait until they start charging you for viewing photos not taken with the crappy built-in camera. Perhaps not allowing you to import MP3s into iTunes because they may not be DRMed. May even start blocking video players because you didn't buy a digital license with the DVD through Apple.
If you've read any previous comments from me on the matter, I've stated it was just a matter of time before they started charging royalties for presenting content on their devices.
"In all honesty, if you were looking after a 3rd party application that had a database, you would probably do the same as the schemas arent often published."
100% agree with you there. There are many applications that use some obscure DB system, or worse yet, use a common DB but don't tell you the password(s) to manage it. All in the name of database integrity and proprietary systems of course. This will prevent BackupExec and the like from sinking its hooks in.
"lots of perfectly acceptable ways of getting a consistent backup, from using an applications backup API to using a snapshot system that is application aware (such as VSS on Windows)."
I doubt VSS is Sybase-aware. Sorry. And as for backup APIs in the software, you neglected to read the author's next sentence. He mentions using the software's internal backup functions as an option.
"AIMstor will give CDP combined with snapshot instances so you've got the frequent crash consistent images (ie every minutes) combined with less frequent recovery point which are fully consistent"
Such CDP systems would only benefit DBs if they grab the transaction log. Then you can have the DB roll back the log to a consistant state. The CDP system isn't, in itself, magical enough to do this on its own.
""that is why an RTO of zero is science fiction."
Really? Isnt this what mirroring is for? Or there are some products that do real time replication."
Such a fail comment. In the event of a database foobar, that corruption is automagically replicated to your mirror. Same goes for corrupting (modifying/deleting) files. The only time mirroring to a hotsite is useful is for system failure. Server bursts into flames? No problem, we have a hotsite. Directory tree got deleted? You lose your zero RTO due to having to restore files.
"Once Steve Jobs goes away, which is probably not far away..."
Steve has been with Apple for a long time. Bill has already moved on from Microsoft. Surely Steve is wanting to retire soon as well. It's obvious that's what he's referring to, not some impending death of Steve. Granted, his timing made it rather easy to mistake it as referring to health issues, but it's honestly and obviously a "you can't be in the game forever" comment.
I agree about skimping on the keyboards. I think it's utter BS that there's NOT ONE ergo keyboard in the lot. Check the Microsoft Natural Ergonomic Desktop 7000 (can be had for roughly the cost of the Apple keyboard) and it even comes with a mouse! I think the keyboards without a numbers keypad should have been docked points too. Especially if you're going for a "business"-type look'n'feel (yes, I'm looking at the Apple keyboard, among others in this faux review). Now, if you really want to quibble over not having a numpad, you can check out the MS Wireless Entertainment Desktop: the receiver is a USB hub, keyboard and mouse RECHARGE with the dock, backlit keys, power indicator, comes with a mouse, slightly ergo keys...only thing not to love is the MS sticker on the top (and the price), but should have been considered. Utter trash of a review IMHO.
I think this guy spent a bender last weekend watching Star Trek: NG and Final Cut (for those of you not in the know, Final Cut is a SciFi about a tech that embeds a recorder in the brain, at birth, that records everything the person sees and hears from their visual perspective [eyeballs] and is used to make a memorial video about their life after they die).
The problem with entirely vocal access to computers is privacy. If you've got a nifty in-eyewear screen, ear piece, cam, etc. and you're on a subway wanting to check your bank records... "Computer, open my bank account and display last week's transactions." Yeah, smart move. Perhaps "Computer, pull up that Scottish sheep porn from yesterday." Or, "Computer, open my LibDem activist website." What would be better than a vocal interaction would be to control the computer with a thought. Currently, we have the archaic move-the-mouse hat-tricks, but hopefully they'll work on a more memory-print level or the like. Probably not to the extent of The Matrix, but at least the computer may understand you're thinking about your money and will pull up a HUD with your current funds. Perhaps a bit of look-at-icon-and-think-"click" magic too for the things the computer hasn't been taught to recognize yet (brain imprinting is like speech recognition; it needs to be trained).
"Spending $10 billion2 to buy 4 per cent of notebook storage market share, or $2 billion in revenue, is not viable"
This statement is the one I take issue with. Why? It doesn't seem to factor in retail value. With 2.5" spindles going for $0.10/GB, and SSDs going for $1.90/GB, That's a 1900% increase! [(0.1*(1900*0.01)) if you want to do the math] Their $2bn is the capitalization of spindles by storage capacity sold. If laptops were stuffed with similarly-cheap spindle drives, then yes, $2bn would be an accurate guess. However, SSD-equiped laptops wouldn't be bottom-barrel laptops, but performance-oriented laptops (dual drive perhaps?) and netbooks with next-to-no capacity (16GB perhaps?). The performance or midrange laptops would likely be (optionally) stuffed with 64-120GB drives in lieu of 500GB counterparts. Each one of these supplantings would, at the least, cost 2x the drive they're replacing (if you do the math, that's a 64GB(ish) in place of those 500GB drives, and a 16GB in the netbook sub-200GB single-platter replacements), which would net them a minimum of $4bn by supplanting $2bn of normal platter data. If these numbers are profit margins vs cost, rather than actual sales, then the numbers swing even more favorably toward NAND. So, 2.5yrs for a $10bn fab, then ice cream and cake afterwards? Sounds like a good investment....oh wait? Isn't that why so many manufacturers are jumping the bandwagon and DOUBLING (nearly) the NAND output in 1 year? Yeah, thought so.
Win7 with 6GB or RAM at home loads faster than XP with 2GB of RAM at work...not surprising one bit. Besides the Win7 prefetch and caching enhancements, I wouldn't be surprised if your work PC runs a lower-performance "cheap" HDD as opposed to a decent performer in your home PC. With your specs listed, I'd be surprised if you weren't at least using a WD Black at home.
Either way, if your home PC was WinXP as well, I'd consider it a fair assessment, but it's not. Which means FF probably tweaked their loadtimes to target Win7 boxes over XP (likely GUI enhancements/calls).
"...may allow offenders to adapt or restrict their behaviour to conduct which falls short of our prosecution threshold"
I think El Reg has simply misunderstood the statement, and simply not asked more precisely. The authorities seem to be stating that they have a "grey" area where those on the border of violating my not be prosecuted, but there's a certain threshold that one can pass (perhaps actual cutting in BDSM depicted, as opposed to simple restraints?) at which point they'll aggressively persue charges, but for simple restraints they wouldn't.... This line is likely what they are saying they won't disclose, rather than details of what constitutes violation of the (proposed) law.
"To reach the same capacity as a 3TB, 3.5-inch... 7K3000 drive, customers would have to purchase three 1TB 2.5-inch 7,200 RPM drives, which will consume up to 170 per cent higher watts to power the drives and will occupy three times the storage array slots“. It says there is a 32 per cent reduction in watts/GB versus Hitachi’s prior generation 2TB A7K2000 drives."
So....get 1x3.5" with 3TB, 3x2.5" 1TBs (which should have no comparison btw, since they use different bays), or simply get 2x3.5"s of the "prior generation" to get 4TB instead of just 3.... Sure, you lose a bit on the power savings and the 6gbps SATA/SAS, but I'm sure those drives just became cheaper....
I'm in agreement too. There's just some things a few quick hand-scrawlings and drawings are perfect for. To do this on a computer, you need to load the doc in some editor that can support such things, save it, pass it on to whomever needs it, and hope your scrawlings are still legible and intact. Not to mention the pen interface, touchscreen, or crappy mouse-drawing one would need in addition to that software support....
"but my netbook, using hte same OS, is noticeably slower when "basckground processing""
Atom CPUs are rather nerfed, low-power chips as compared to even moderately low-end desktop CPUs. Hz for Hz the Atom will lose. Why? Power conservation.
These dual-core CPUs promise better performance and lower battery life, not only because they're dual-core, but because they're newer. Better architecture and smaller die size has been what makes "Sandy Bridge" processors better than last-year's Nehalem chips. Hz for Hz, they perform better. The die size contributes to lower power requirements. The extra core in the A9s is power-gated anyway, so when it isn't in use, it draws NO power. New power management features like this promise to deliver better battery life, hand in hand with more capability. As more work is done on phones these days, better CPU capability will always be welcomed. If you don't believe me, use Android on a 500-750Mhz instead of the 1GHz chips.
"As if they don't already confuse them with countless add-ons, up-sales, superfluous insurance, crapware unbundling services."
Windows come bundled since it makes the computer reseller's (Dell, HP, et al) life easier. Bulk licensing from MS, and they don't have to have *nix techies on staff to do the minor troubleshooting prior to returning the device for repair (which they do under warranty).
In addition to that, the resellers get quite the pretty penny from "McAfee" (Intel now...), Symantec, CrapWareUSA/UK, and its ilk for pre-loading their crapware on your shiney new computer. They won't sell you Linux since they can't load the crapware on it. They have to make money on their miniscule margins somehow.
As for those buying a computer, but wanting to put "another OS" on it, it's likely that these people already know enough to build the kit themselves. Demanding a refund would only be useful in Corporate environments where the IT staff simply reimage the machines with their Volume License prebuild or other similar situations. This is more likely than the occasional user who has purchased a copy of Windows off-the-shelf and plans to bin their old machine and move the license. This said, you're only likely to get ~$10 back anyway, as bulk licensing is fairly cheap for OEMs.
Very good point! The Cloud people are great in that they provide backups (hopefully) for you. Now, just what point in time you want them to restore from? Good luck. User X deleted a NEEDED file middle of last month... you use it to generate month-end stuff perhaps. Will CloudX be able to restore it? Do they keep month-ends? Do they keep your required 7-year backups? Perhaps. Back it up locally you say? Why? You have the cloud, remember?!
Now serious stuff: I've used several backup softwares, such as ArcServe (ick) or BackupExec (tape libraries, with severe issues with overwrite...very disheartening to check on a backup only to realize it was aborted due to trying to write over a previous backup or write to a tape not in its library (clean tape even). Anyway, compound that with the crap speed of the LTO drive we had and ick.
Solution? Moved to disk-based backups. External 1TB drives cost about as much as our tapes did, perhaps slightly more. Benefit? Faster backup times. Perhaps even better resiliance. Definately better random-file recovery. As for backup software? Ever try restoring a file from an version of ArcServe? Does it even run on WinSrv08? What if you misplaced the license key or the software? Install disk damaged? Granted these are worst-case ideas, but you can't expect your install disk of BackupExec9 to be available when an earthquake causes a box to land on your stack of disks.... Ever try restoring a backup from a different/newer version of the software? Doesn't always work unfortunately. I tend to agree with the copy *.* P:/ solution mentioned earlier, however it is still a crap idea. Robocopy or SyncToy would have been a better suggestion. Not to mention the potential need to encrypt any data that goes off-site. This is where backup software starts to look better.
Really, it all comes down to need. Do you need to have your data back online in under 5 minutes? Are your backup windows starting to converge? I ran into a problem a while back of having our week-end full backups approach 72hrs runtime (yes, it was across a T1, not on-site, but the data needs to go to where the IT people are most of the time), which starts to conflict with monday morning, as you can imagine. A switch to disk, and changing the backup software/method cut that time considerably (would you believe down to 15min?). As for Continuous Data Protection, it's a great idea. Recovering a file from 5 minutes ago is quite useful, as is having a historical archive for the past few days/weeks. The backup system I have in place now uses file-level deduplication between backup sets, which isn't as good a block/bit level, but still allows for hourly/dailies for the past couple of weeks, and weeklies and monthlies spanning back a good year; all accessible at a moments notice, rather than having to phone the cloud or run to the safety-deposit box for tapes.
"AMD, you have to find out which generation of chipset and such you need to make it work"
While it is true that different chipset generations offer varying capabilities for the motherboard (number of SATA ports, USB3 support, PCIe lanes, etc) the socket is really a lot like Intel's. You have a AM3 CPU? Buy a mobo with an AM3 socket. Already got an AMD mobo? AM2?, AM2+? Your AM3 CPU will work in it. The new CPUs are socket downgradable. Of course, don't expect you'll get full AM3+ capability out of an AM2 motherboard, but in 6 months when all the SATA ports are SATA3, and most of the USB is USB3, or perhaps when PCIe v3.0 comes out, you can worry about upgrading to that new AM3 socket to match your processor's capability.
On the flipside of the coin, the new Sandy Bridge dropped one pin down to 1155, making the new chip incompatible with the previous 1156 boards (even though ASRock has a board that sidesteps this minor setback). In Q4 when the X67 chipset comes out, that will have LGA-2011, so another socket again.... Likely they'll make another new socket with their next "Tick" as well.
That said, AMD has a good example of how backwards socket compatibility is achieveable, however, as new features become available (such as DDR3) it becomes harder NOT to change sockets. I applaud AMD in this regard in at least maintaining that backward compability. However, Intel chips do outperform AMD offerings in the top 25% performance market, even though the 1090T is a mighty tempting offering, except that the i7-2600K smokes it once OCed to the 4.4+GHz range. AMD needs to strike back with tangible performance increases, not core count. They have 16-core chips. Good stop there for now and work on the microarchitecture.
"The cloud means consumers are using remote storage from service providers more and more, so that they need fewer gigabytes or terabytes of external storage."
I don't know about you, but it takes FOREVER to upload/download a few gigabytes worth of pictures from a cloud service. Perhaps when 100mbps fiber runs to everyone's home will cloud storage be viable for the common use. However, I tend not to use it for anything larger than documents and spreadsheets.
That said, The standard hard drives computers come with now (500GB) practically eliminate a home user's desire to buy an external drive. Before, they only had 120GB if that, so they needed to buy externals to store more of the family pictures/movies/porn. Home users tend not to consider backups, and are unlikely to buy an external soley for backups, which is why cloud services are a benefit to the casual word-doc writer.
As SSDs gain prominence, PCs likely will start shipping with SSDs instead of HDDs in netbooks, laptops, perhaps desktops. However, "home user" won't know the difference and will simply go for the big numbers. I think, quite literally, joe-uzer sees "500GB" and thinks that's RAM, and instantly believes that's better than the 120GB SSD-equiped computer. Even though they likely won't even use a quarter of the SSD, they still buy the 500GB machine. This is why SSDs will have a hard time penetrating the retail market. My hope is that Win8 comes out with their own brand of ZFS storage tiering so manufacturers can pop a 60GB SSD and a 1TB HDD into a computer and just let Windows decide where to stick the OS, Programs, and word docs. Home users wouldn't know the difference between C: and D:, and so the OS should just make it a transparent thing for "Simple Mode" (think of Simple Network Sharing vs "advanced").
Wind Farms have been under fire by many since their inception due to the unrealistic predicted gains. For instance (under the old model even!) it was predicted one would have to blanket the state of Texas in wind turbines to power the USA. Of course, that's assuming the wind is actually blowing....In northern Scotland, this might be likely, but in many places it will take a wide variety of methods to provide a base load and peak load. I, for one, would dump wave generators in ocean currents, wind farms in usually-windy spots, lease windfarm land for livestock grazing, and provide base load with nuclear power. Build a plant by my city, I don't care. I'd suggest more hydro-electric, but we have a dam in almost every conceivable spot as it is....at least without the environmentalists going militia on us...
Of course, I like the point of view in my development to add in /basic/ things such as this as part of the standard package, and if they specifically request to leave the passwords plaintext, I'll have that written out in the final hand-over documents that we both sign.
Granted, that's just my opinion, but leaving plaintext passwords is tantamount to purposely not putting SQL injection prevention measures.
The sad thing is, just why they assume the resolution will be 2048x1536:
File name conventions for images for the new iPad2 vs the original iPad ask that the iPad2 images have an X2 in the filename so the iOS can automatically display the proper images for the new device (likely due to a resolution bump). Now, X2 could be gen2, but very likely could be "twice as big" x2 multiplier... Guess which one makes the best news story? Even if it is x2, the 2048x1536 guessed resolution is taking the current 1024x768 and multiplying BOTH dimensions: 1024x2 = 2048, and 768x2 = 1536 and boom: x4 screen. LIKELY what they're using the x2 for (if it is indeed indicating a multiplier) is a rounded PIXEL count. Current iPad displays are 786k pixels, so 2x786k = 1.57MP. Granted, reversing that down to a 1920 screen leaves only 819 for the x9 side instead of 1080, this discrepancy is likely just because it's easier to say x2 than x2.63671875.
'Nuf said.
Blizz isn't stopping/destroying anything. They've simply claimed that something in the vid infringes their copyright (likely the StarCraft name). Not saying the character creation, nor HUD, etc doesn't whole-sale knock-off WoW, but still....
Anyway, a mod to SC2 likely won't be as full-featured as a real W/GoS MMO even if it runs a pure, reskinned WoW-engine. Granted, it would simply cannibalize the existing WoW base, but it may also draw a crowd from other sci-fi MMOs like Star Wars/Trek.
Sorry, but you can't exploit flash (let alone just the flash H.264 decode on these procs) simply by trying to chuck a malformed flash file at the IP. Flash is not a server (some might argue this) on the box. It does not have open ports listening for connections, let alone listening for flash files to then try to display.
/Fail for complete lack of understanding of network communication.
One must remember, these figures are from Symantec, and thus, it means that this distribution is based on THEIR software DETECTING the infection on the computer. So, only people who have Symantec installed (and have their phone-home-stats bit being allowed...) are in the mashup. Now, considering the number of OSX users running Symantec AV, having 16% of infections is a VERY concerning thing. If the virus survived a Linux reboot, I'd express the same concerns with their (non-)figures. Not that they'd stray from their ClamAV or the like anyway...
It is striking that only 7% of Vista/Win7 machines were infected though. I guess the numpties haven't bothered buying a new computer in a while. How many unwashed mass members do you know that would be bothered to buy Win7 and install it on their current computer anyway?
/paris, because even for the elites, protection is needed
"Add to that the fact that some vendors (*cough*adobe*cough*apple*cough*) don't distinguish between security patches and bloated new versions"
Actually, you're incorrect. Reader 9 does not list Reader 10 as an "update." You'll update to 9.4.2 and no further. You have to download 10 seperately if you want the new version.
As for not using a program while the patch is /downloading/, yes, fail on the software! At least the Firefox update is quick....
The problem is, while some point-releases do offer patch-updates, I can think of several programs that when they're "updated" they require you to buy the new version. Ever "update" from PowerDVD 8 to 9? Linux OSes manage this easily because the upgrades are free. The best you'll find in Windows pay-for software is a "lifetime license." That aside, freebie addons, such as Flash Player, iTunes, Quicktime, Firefox, etc should have a package-manager system they "register" with for free updates, rather than the whole "check for updates every time you start" crap or even worse, the adobe/apple update notifiers.
As a side thought, I wonder if this "research" classified Internet Explorer as a "third party" program, as opposed to an Operating System component, or if IE8 was considered an OS "update." If it was classified as a "third-party" app, like it should be, then I can more than believe the numbers toward programs being the bane of existance rather than the OS. It's been a while since we've seen a Code Red-type virus that propogates on its own rather than require the (stupid) user to hit "yes" 3 times.
The laptop in question is a personal laptop it seems. Blocking USB devices from being used on hopsital computers only goes so far. We use several USB-based devices, and it is trival to unplug on, pop on your thumb drive, and copy data. We have policies in place that ban carrying data offsite, period, unless authorized by IT. This way, we can ensure encrypted transport and end-point use. We've even taken to encrypting personal laptops as an extension of services to help ensure data will be safe in the event that it needs to be taken home for any reason. With the IT staff signing off on data-carryout, we also have records (general as they may be) of who carried data off premisis and when.
Good security is a high inconvenience for sure. However, whole-disk encryption, VPNs, and a few other measures help keep the security high (enough) without imposing the inconvenience of the paranoid.
"Wow hold the horses there, what Facebook can upload my details simply from someone (who is a member of Facebook) who happens to have my phone number? Please explain what Facebook can upload, as this is news to me? ... what can they upload?"
Easy. You flesh out your "friends" missing contact info in your Facebook iPhone contacts app. This information, for your convenience and "data-recoverability" is stored on Facebook servers so you have access to your "additional" information whether your on your phone app or the web. Thus, Facebook now has your address and/or phone number, birth date, etc. all of which you falsified or left out intentially in the first place. Quite simple. What's worse, unlike being tagged in a photo, you don't get notified if someone filled out contact info on you.
"@"vast majority of users simply agreeing with everything they're asked"
I see the Gene Pool shallow end is causing trouble again."
While you can sideline this as a "shallow gene pool" issue, it is not. Unfortunately for many, humans are a trusting lot. Just look at punters on Ebay or craigslist. The truth of it is, they're installing the app because they want what the app supposedly does. Does it matter if they have to hit a "Allow" button? Nope. The programs should all be blocked from having access and leave it to the "shallow pool" people to figure out how to enable it post-install. Might work a bit better. Of course, there's always vetting, as mentioned. Finding a balance between helping the most amount of people and inconvenience level is what many companies are struggling with. Just think of UAC on WinVista.
Yes, and it is expensive. It would take roughly 20 years of leccy generation to pay off the solar panels and infrastructure for your own home. (depending on use, quality of materials [and panels], and whether you can sell your surplus back). I'm a fairly heavy electronics user, so likely is more expensive for me though...in a northern clime. Perhaps the southwestern deserts of the USA would be better off.
All software needs patches. Fortunately MS is willing to provide them (sometimes with new/extra features) for free.
I can't say (like some others) that I've had no problems with Win7, but then again, I'm not a normal Win7 "print the pictures and surf the web" user. I'm just glad I can use Eyefinity to mirror my primary desktop to my HDTV and have a second monitor extend my desktop on the side (a feature WinXP couldn't manage, since it had to either extend or clone, no compromises).
Now if I just didn't have to use WinXP Mode to run Civ4....
Open for local intranet use, sure, that's obvious. But "...and likely also remotely." fails. Remote access to a company's servers usually involves a VPN connection to the firewall. If they did (unlikely) just stick this machine in the DMZ or outside the firewall, then yes, they should be sacked. Assuming that is what they did? No, I don't see it. Even a radiology clinic with no IT staff would still be behind a DSL/cable modem/router with a built-in firewall and the machine given a 192.168.0.0/16 address. at the very least. So, the "breach" was likely port-forwards or other security slight. Granted, since they had a CoD server running, that means they had port-forward capability on the firewall/modem/etc or the machine WAS in the DMZ....still, fail for assuming and not thinking it through.
There are plenty of USB Network Printeservers out there for 50$ or so. Be careful, because many cheapo printers require the computer to run the printing so the printserver would be inadequate (think of WinTel modems of yester-year and why they wouldn't work in Linux).
However, the Belkin F5L009 seems a much better solution than this crappy hardware. 5 USB ports and they can be independantly used by network PCs as if the USB device had been plugged directly into the computer. I actually use one of these Belkins for USB modems for a couple VMs. Gotta love being able to virtualize a server that requires a modem, and still retain VMotion and the like. :)
The graphs are a torture synthetic benchmark. They dump random(ish) data sequentially to the hard drive with no TRIM or the like, nor breathing room for the Garbage Collector to kick in. This is NOT how a normal computer operates. This benchmark is to show the actual physical performance of the drive, not use in real-world scenarios. If you stuffed this Vertex 2 behind a benchmark of random read/write blends of 70/30% with a queue depth of 4 (or ideally 32), you'll start to see the drive breath a little. Also, this is continuous hard drive hammering. While I'm typing this, unless I'm transcoding video or the like, my hard drive likely isn't even in use, thus the garbage collector could potentially run and clean the drive a bit.
The real apples to oranges parallel in all of this is this DDRdrive uses DRAM and NAND, whereas the Vertex2 drives actually have no drive cache AT ALL, let alone the multiple GB of cache these DDRdrives have.
HDMI cables are like network cables. Cat5 and Cat5e are different. As is Cat6. What is the difference? Bandwidth (the MHz frequency range the cable can handle). Other differences include wire twist methods, conductor quality, and end termination. All of these components help determine what standard the cable can be rated for. Cat5 quality and twist prevents it from handling Gigabit frequencies properly, and likely your NIC will limit you to 100Mbit. The frequency (bandwidth) range required exceeds the cable's ability. HDMI cables have similar issues, where they're rated for HDMI 1.3b, 1.4a, etc. Granted, you can get a cable that, based on that hard-kink you put in it to sit your TV or Blu-Ray flush to the wall, will compromise the cable's already-mediocre build and cause it to down-grade the signal to the point of only being able to pass 720p. Cable length is important also, since 1080p frequencies across a poor-quality cable will likely require a shorter length, just like Cat5e is not recommended for longer than 100 meter runs (even though it may actually work "well" for 150m in your situtation, lucky you).
With all this in mind, if you're buying a 6ft HDMI cable, likely ANY cable you get will run at it's RATED spec (1.3b cables have no hope of running 3D Blu-Ray, for instance. That's what 1.4 cables are for), unless your cable is defective (or you broke it). Defective rates or User Error is outside the scope of this retort.
So, for those that missed the point, here's a summary:
HDMI is a digital spec. It will auto-negotiate the best quality the cable can handle in the given situation. If you or your cable is a numpty, you may only get 720p when you were hoping for 3D Blu-Ray. Read the cable's spec. Most are likely 1.3b, which can handle 1080p, but not 3D Blu-Ray. No, your HDMI cable from 1.1 days won't handle 1080p.
The "it's digital" people don't acknowledge the auto-negotiate side of the equation, which is what the "high-quality cable is important" crowd is likely seeing.
"It certainly doesn't need 64 bit architectural address space... now or in the next ten years"
Not in 10 years, no. But by 19 January 2038, yes.
"ARM doesn't greatly need more memory"
Akin to the "640k memory" quoted already.... IF ARM is to make desktop or (especially) server chips, it will NEED to handle plenty of memory. 4GB is a good start, but x64 is a less-complicated necessity to handle >3.25GB.
And as for your code-density comment, the executable code size isn't the problem. Running exe + dlls (windows) or libs (linux) amounts to a very small footprint indeed, compared to what actually eats up the RAM: multimedia. Images (bitmaps), audio, etc. These are things that can't be blitted in a compressed form, and would be too CPU-intensive to keep in a compressed form 100% in RAM and decompressed each time the screen area needs refreshing (think of your wallpaper for instance).
"When do I get a job in Cambridge?"
Never. Lack of understanding.
@Chris Harden:
The feat of putting the shuttle into space with no pre-existing implementations to work from is laudable. That's a victory in itself. However, the shuttle and method is a failure. The costs to make, launch, and maintain the shuttle+supports is grossly expensive: as proven by "commercial" solutions.
The other sore spot is your comment "pull off a 100% sucess rate." I do not believe Challenger nor Columbia can qualify as "100% success rate." Granted, failures happen. It doesn't make them less tragic, but they can't be dismissed and have the project claimed 100% success rate either.
"Start padding out game data with garbage, duplicate data, etc so games fill a 25Gb, 50Gb or 66Gb disc. (Sony has a format in the works for 66Gb)"
Garbage data is easily compressible (unless it's highly random, non-repeating strings of bits) or is simple enough to create a "PS3 ISO compressor" that simply reverses whatever method they use to detect and skip over the garbage data and remove it, then on-the-fly dump the "non-garbage-padded" ISO to a bzip or the like compressed file. On the receiver's end, they would feed the compressed file back into the ISO compressor tool to reinflate it and add the garbage back in. All in all, unless it's real data, it can be compressed (usually). The compressor might even strip the encryption on the files (if needed) to make the actual data more compressible and reapply the encryption on the "inflate" side, since the key is known.
As for things like "Add a challenge / response during PSN sign on. e.g. asks firmware to checksum some arbitrary range of bytes in its fw or memory to proceed. Wrong answer gets flagged." it would be simple enough to have your new "rogue" code respond with an authentic checksum response or the like. Basically, if the response is generated console-side, it can be duplicated. As well as "audits" being defeated.
But you recognize this by stating: "At the end of the day none of these things are foolproof." but fail to mention that it is fairly trivial to the point of worthless to attempt to incorporate the measures.
"Mind you, I'm not sure what kind of "snazzy media processing" folk really need these days"
Need I remind you that your desktop/laptop is what converts original "HD" content to content your phone can play? And, unless I'm mistaken, your phone doesn't have a 1920x1080 display. It likely doesn't even have a 720p (1280x720) display. Therefore, your "HD" content is likely "HD" compared to the old 4:3 content. But, to transcode video/audio to a decent-quality, lower res, higher compression file (read: stuff a 6GB DVD movie onto a 1GB or less file next to the mp3/mp4s on your phone) takes a bit of compute power. Even the new Sandy Bridge can only do 220fps, which for a 24fps vid is only 9sec of video per second. Fanless ARM chips can manage this without draining your battery (much) due to in-chip video decoding, just like what the recent crop of CPUs/GPUs have. Next time you want to foam at the mouth that your ARM (iPhone likely?) CPU is better than nVidia/Intel/AMD hardware, try transcoding Iron Man 2 on your phone (if you can) and see how long it takes. Be sure to plug it in and put it in the icebox first.
/needed to be said.
"He should have sacrificed his political career and stood up to American and said "We used to be great. Now we suck. It's time you stopped waiting for the White House to fix your problems for you. We can't. If another politician tells you otherwise, he's lying or doesn't understand how it works. We can help, we can try to motivate, but IT IS YOU that has to fix the problem. If you can't find a job doing what you do, find a job digging a ditch. If it's so far beneath you that you'd rather sit on a couch and wallow in your sorrow as opposed to earning your on living, YOU ARE THE PROBLEM!"."
Living in an area with lots of government-assisted housing, I see this all the time. There's plenty of people just lounging around milking money from "the system." Why? Because at a certain income level, one can make more money from the government than by flipping burgers at two jobs. All for doing nothing. Not only that, but if you do take that burger job or ditch digging job, you lose your government incentives and actually take a "pay cut" as it were. The government social system is encouraging "bottom-dwellers" to remain there. So, although I fully agree with the above quote, the current edu-work fail is very intricately linked with many other "problems." Most of it comes down to "if there's an easy way out, the majority will take it." This is the failure. A society where it's acceptible to "dishonor your family" by being lazy. Perhaps that is why the Chinese are winning: they have a strong self-worth and even that is moot against your family's honor.
That is all.
The problem with the hybrid drives is they're trying too little. 4GB is a great way to boost, say, OS boot times. However, a 16-32GB cache would be much more useful. And include a tool that you could, in OS, "add file to drive cache." Would be a no-brainer to let the OS get auto-cached and simply add/remove files of your Game-Of-The-Day (think the big MPQ files from Blizzard perhaps?).
Actually, it is in repurposing and shoe-horning *nix and such onto initially unsupported hardware that some flaws in the system are found, be they compatibility modules, drivers, API calls, etc. Heck, just running the OS on a setup with some shortcoming may reveal a weakness (lack of function, not security) of the OS. I'm quite sure they have to troll through the original driver code to see what can be copy/pasted for their port as well, which means: code review. Just reading through the code doesn't solve problems; working with it and modifying it solves problems.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
