Slashdot Mirror

Whatever your feelings are about Intel's products, this product is cheaper than a similar product 1 year or 2 years ago, it's better, it's faster, it's less power hungry, it has more capacity.

Clearly the investment into the original batch of SSDs paid off and Intel was able to use the profits it made on them to invest into better technology. That is economics (as I said, some call it 'trickle down' economics), but really all real economics is 'trickle down' economics. The more profitable a company becomes, the more it can invest into its products and the lower the prices will be eventually, because more profits brings in more competition.

I bought a OCZ NIA to play with http://www.ocztechnology.com/nia-game-controller.html . Too impatient to learn how to use it, I shelved it. Friend from the Phillipines came to visit, he had no use of one hand and he tried it out. Within a day he had it doing a few basic things for him. He only got to use it for a weekend, and seemed to like it. Not sure if he picked up one when he went back to the Phillipines. I don't think they make them any more, but I'm sure you could find one some where.

Not the GP, but I suspect he was referring to the fact that the OCZ Colossus 1TB SSD is internally a RAID 0. If an enterprise is running these, they're running RAID0, but may not even realize it.

http://www.ocztechnology.com/ocz-colossus-lt-series-sata-ii-3-5-ssd.html

That said, it starts at $2500. The 600GB 15k SAS drives start at $650. $1200 (comparing 2 SAS to 1 SSD) buys you a lot of rack space and cooling. If you look at "Enterprise SSD" (whatever that means), you're looking at requiring a PCI-E card to get close to that density, and it's going to cost even more. 1.2TB run over $4k. At least you can get, what, maybe 3 PCI-E cards into a 2U box? Of course, you can fit 2 SAS drives per U pretty easliy.

Honestly, I think it's pretty hard to compare.

http://www.ocztechnology.com/aboutocz/press/2011/448 $500 and is "face meltingly fast".

Sounds like OCZ's IBIS just got standardized or copied.

http://www.ocztechnology.com/ocz-ibis-3-5-high-speed-data-link-ssd.html/

Even the best single SSDs come close to 2.5 Gbit, so to really justify Thunderbolt, you'd have to do RAID.

On SATA, SSD's have rapidly consumed whatever bandwidth is available. Before SATA 3.0, SSD's banged their head on SATA 2.0 limits, and now that SATA 3.0 is becoming widespread a new generation of SSD's is now starting to bang its head on SATA 3.0 limits.

This is not because SSD technology is getting faster. It is because the interface is getting faster. SSD technology is embarrassingly parallel and can easily go faster.

Make a faster "just plug it in" interface and SSD's will consume it. Some SSD companies have made their own custom interfaces for this very purpose, such as OCZ's HSDL.

tl;dr - You have screwed up cause and effect in your SSD example, proving that you are shouting from someone elses lawn right now.

Cutting eedge vaporware is still vaporware.

Well, that's awkward. Who wants to be the one to tell OCZ that the product that they have released to market is vaporware? (Shotgun not me.)

Similarly, you could fit twenty* or so of those Macbook Air "blade" SSDs in an appropriate enclosure, giving you 5TB of SSD in the same space of a 2TB spinning disk at current densities.

This would require a hell of a lot of cooling. Just because it's slim doesn't allow you to ignore heat. I think you might be able to get away with about 2-4x the density of the 2.5" drives, giving you about 2-3TB of space, but the engineering and testing of that would be expensive.

Ultimately, you will need to bypass the SATA cable as even single SSDs are close to the maximum throughput of SATA2, which means all it would take is roughly 4 SSDs to consistenly saturate a SATA3 bus. That's the reason storage providers have decided to go with PCIe boards ... anywhere you want to put a 3.5" drive, you will likely be able to put a card in instead.

Does that make me a fail or a win?... how about epicly in either direction?

512GB SSDs aren't a "future possibility"

1TB SSDs already exist

and find photo's here if you'd like to know more...

When Slashdot and Advertisements come together... Slashvertisements! You could have learned as much or more by reading the press release where it is revealed that "Enyo USB 3.0 Portable SSDs will begin shipping this now and will be available through OCZ's extensive worldwide channel." Thank goodness, I thought I would have to wait for the next now.

When will "then" be "now"?
soon...

When Slashdot and Advertisements come together... Slashvertisements! You could have learned as much or more by reading the press release where it is revealed that "Enyo USB 3.0 Portable SSDs will begin shipping this now and will be available through OCZ's extensive worldwide channel." Thank goodness, I thought I would have to wait for the next now. Also per the pr, "the Enyo features a sleek, anodized aluminum housing" ... the choice of words implies that it's a desktop SSD in a box. It would be nice to know which one, if so.

I have been running Win7 with an SSD for the last 8 months and I am never going back. You should actually read some recent literature on SSDs to answer this question. SSDs now advertise lifetimes comparable to or longer than magnetic disks, thanks to using very advanced firmware to do error correction and wear leveling. If there is an error in an SSD write, the drive will work around it and write to another sector (most drives have unreported capacity they use for this and wear leveling).

The OCZ Vertex II, currently an industry favorite, advertises 1.5-million-hour mean time before failure. That means that the drives are estimated to last 171 years of normal use before failing outright. A Western Digital Caviar Black, on the other hand, is rated for 300,000 load/unload cycles, meaning you can spin it up and down 300,000 times. If you spin up and down 5 times a day, it will last 164 years, not counting possible degradation of the magnetic disk coating or mechanical failure due to shock or temperature extremes.

Basically, SSDs are a pretty mature technology with reliability that meets or exceeds standard HDDs. In many applications, the vibration and temperature resistance of SSDs give them a significant advantage of HDDs. Just remember that you have to treat them differently from "dumb" HDDs--the SSD firmware automatically does all the defrag/wear leveling/error correction itself so trying to do that manually will kill the drive prematurely. File system alignment and RAID controller settings can be very important as well, so due diligence is required to use them properly.

Calling the article a 'press release' unfairly tarnishes OCZ. Their press release is still full of press release though:

http://www.ocztechnology.com/aboutocz/press/2010/362

I *love* this idea. Neural interfaces are still very much in their infancy with the best commercially available probably being the OCZ NIA and even it is mostly a glitchy gimmick at this point. But the standards they're considering for this X-Prize seem very high. Providing vision to the blind and being able to control virtual bodies both require an understanding of very intricate neural operations that we probably won't see for many years. Sure, there's been devices created that can sort of do these things already but not on levels that can significantly help the disabled. All the more reason that this prize is a good idea since it will hopefully draw interest to the appropriate fields of study. Unfortunately the kind of things that they're talking about will almost certainly require implants. Further advancement in non-implanted interfaces will undoubtedly be much slower so don't assume that you're going to be roaming the hills of Azeroth as if you were actually there any time in the next 20 years. But hey, hopefully I'm wrong.

The SAN argument is that your storage is so precious it must not be stranded. If you're paying $50K/TB with drives, controllers, FC switches, service, software, support, installation and all that jazz then that's absolutely true. If you're doing something like OpenFiler clusters on BackBlaze 90TB 5U Storage Pods for $90/TB and 720 TB/rack you have a different point of view. As for somebody showing up to replace a drive, I think I could ask Jimmy to put his jacket on and shuffle down to the server room to swap out a few failed drives every couple months - that's what hot and cold spares are for and he's just geeking on MyFace anyway. Low utilization? Use as much or as little as you like - at $90/TB we can afford to buy more. We can afford to overbuy our storage. We can afford to mirror our storage and back it up too. In practice the storage costs less than the meeting where we talk about where to put it or the guy that fills it. If you want to pay for the first tier OEM, it's available but costs 10x as much because first tier OEMs also sell SANs.

Openfiler does CIFS/NFS and offers iSCSI shared storage for Oracle, Exchange and SAP. If you need support, they offer it. OpenFiler is nowhere near the only option for this. If you want to pay license fees you could also just run Windows Server clustered. There are BSD options and others as well. Solaris and Open Solaris are well spoken of, and ZFS is popular, though there are some tradeoffs there. Nexenta is gaining ground. There's also Lustre, which HP uses in its large capacity filers. Since you're building your own solution you can use as much RAM for cache as you like - modern dual socket servers go up to 192GB per node but 48GB is the sweet spot.

Now that we've moved redundancy into the software and performance into the local storage architecture, moving storage to the edge is exactly what we want to do: put it where you need it and if you need a copy for data mining then mirror it to the mining storage cluster. We still need some good dedicated fiber links to do multisite synchronous replication for HA, but that's true of SAN solutions also. We're about 20 years past when we should have ubiquitous metro fiber connections, and that's annoying. Right now without the metro fiber the best solution is to use application redundancy: putting a database cluster member server in the DR site with local shared storage.

Oh, and if you need a lot of IOPS then you choose the right motherboard and splurge on the 6TB of PCIe attached solid state storage per BackBlaze pod for over a million IOPs over 10Gig E. If you need high IOPS and big storage you can use adaptor brackets and 2.5" SSDs or mix in an array of The Collossus, though you're reaching for a $6K/TB price point there and cutting density in half but then the SSD performance SAN has an equal multiple and some serious capacity problems. If you go with the SSD drives you would want to cut down the SAS expanders to five drives per 4x SAS link because those bad boys can almost saturate a 3Gbps link while normal consumer SATA drives you can multiply 3:1.

If you're more compute focused then a BackBlaze node with fewer drives and a dual-quad motherboard with 4 GPGPUs is a better answer. At the high end you're paying almost as much for the network switches as you are for the media. If you're into the multipath SAS thing then buy 2x the controllers and buy the right backplanes for that - but

The SAN argument is that your storage is so precious it must not be stranded. If you're paying $50K/TB with drives, controllers, FC switches, service, software, support, installation and all that jazz then that's absolutely true. If you're doing something like OpenFiler clusters on BackBlaze 90TB 5U Storage Pods for $90/TB and 720 TB/rack you have a different point of view. As for somebody showing up to replace a drive, I think I could ask Jimmy to put his jacket on and shuffle down to the server room to swap out a few failed drives every couple months - that's what hot and cold spares are for and he's just geeking on MyFace anyway. Low utilization? Use as much or as little as you like - at $90/TB we can afford to buy more. We can afford to overbuy our storage. We can afford to mirror our storage and back it up too. In practice the storage costs less than the meeting where we talk about where to put it or the guy that fills it. If you want to pay for the first tier OEM, it's available but costs 10x as much because first tier OEMs also sell SANs.

Openfiler does CIFS/NFS and offers iSCSI shared storage for Oracle, Exchange and SAP. If you need support, they offer it. OpenFiler is nowhere near the only option for this. If you want to pay license fees you could also just run Windows Server clustered. There are BSD options and others as well. Solaris and Open Solaris are well spoken of, and ZFS is popular, though there are some tradeoffs there. Nexenta is gaining ground. There's also Lustre, which HP uses in its large capacity filers. Since you're building your own solution you can use as much RAM for cache as you like - modern dual socket servers go up to 192GB per node but 48GB is the sweet spot.

Now that we've moved redundancy into the software and performance into the local storage architecture, moving storage to the edge is exactly what we want to do: put it where you need it and if you need a copy for data mining then mirror it to the mining storage cluster. We still need some good dedicated fiber links to do multisite synchronous replication for HA, but that's true of SAN solutions also. We're about 20 years past when we should have ubiquitous metro fiber connections, and that's annoying. Right now without the metro fiber the best solution is to use application redundancy: putting a database cluster member server in the DR site with local shared storage.

Oh, and if you need a lot of IOPS then you choose the right motherboard and splurge on the 6TB of PCIe attached solid state storage per BackBlaze pod for over a million IOPs over 10Gig E. If you need high IOPS and big storage you can use adaptor brackets and 2.5" SSDs or mix in an array of The Collossus, though you're reaching for a $6K/TB price point there and cutting density in half but then the SSD performance SAN has an equal multiple and some serious capacity problems. If you go with the SSD drives you would want to cut down the SAS expanders to five drives per 4x SAS link because those bad boys can almost saturate a 3Gbps link while normal consumer SATA drives you can multiply 3:1.

If you're more compute focused then a BackBlaze node with fewer drives and a dual-quad motherboard with 4 GPGPUs is a better answer. At the high end you're paying almost as much for the network switches as you are for the media. If you're into the multipath SAS thing then buy 2x the controllers and buy the right backplanes for that - but

So don't buy a branded box.

Build your own PC. If a part is bad, you just RMA it. No bullshit, no having to send the whole thing in.
I know that companies like Asus have 3 year warranties on motherboards and video cards, AMD has a 3 year warranty on their processors, and companies like OCZ have Lifetime Warranties on RAM, SSD's, and PSU's.

Quit getting ripped off by the computer-in-a-box companies.

Controlling computers with our minds may sound like science fiction, but one Australian company claims to be able to let you do just that.

Am I missing something here? OCZ's Neural Impulse Actuator, a similar product, has been out since for about a year.

Sure has, but NIA consists of like, three electrodes, and is fucking terrible at everything. This is supposedly better, but I'll wait for some more reliable reviews before I put down any cash.

To quote their PDF "There is also the issue that brain research has advanced but we are not too concerned with the dissection of the signals into specific origins of neuronal / muscular groups, rather, as mentioned above, it is the synergism of all physiological activities that is the critical signal we are trying to read and understand to the point where we can use it." http://www.ocztechnology.com/files/misc_products/NIA_complete_English.pdf also "The new headbands we are using embrace sensors based on carbon nanofibers that are about 100 times more sensitive than older technology. This allows for a much wider dynamic range and also to pick up signals that were completely masked in previous approaches and that makes the use much more intuitive and easier."

Controlling computers with our minds may sound like science fiction, but one Australian company claims to be able to let you do just that.

Am I missing something here? OCZ's Neural Impulse Actuator, a similar product, has been out since for about a year.

Then you could use "thought macros" to control wearable computers.

What, like this product?

Actually, not only Intel is working on that kind of stuff. I got one of those and it is working (well, kind of ;-) ).

Heh... Well...even those can be, after a fashion DIY put together. They're all made from some of the same names you're familiar with in the desktop space- you can buy motherboards, CPUs, memory, hard disks, DVDs, etc. for Laptops and put 'em together. It's not as easily done or readily available as the desktop stuff is right now, but it can be done.

Here's a starting point for some:

http://www.ocztechnology.com/products/diy_notebooks/

As an interesting note, the new line of Patriot SSD come very close to the 300MB/s speed, clocking in 280MB/s in reads.

Yep, the next step up seems to be PCI express cards directly, even with SATA3 on the horizon it's not moving fast enough. For example the OCZ Z-Drive. These are basicly just internally RAID'd SSDs but a preview of what's to come.

As someone who's been managing RSI for some time, and still needs to be careful to avoid overdoing it, I'd be very happy for a way to supplement keyboarding and mousing with even limited additional input methods, preferably methods which used a different paradigm altogether.

I've been checking out neural impulse actuators, like the one by OCZ, but it looks like they only provide 2-3 buttons, need recalibrating every time, and are only really supported for gaming. Does anyone know of similarly commercially available hardware? I'm aware of research systems which can control a mouse this way noninvasively, but surely it's time they came out of the labs.

I'm also curious about the long-term effects of devices which detect muscle action. People who migrate to voice recognition can damage their voice from the new strain. Would your face start creasing or cramping after a long time using a device which relies on facial muscles? It seems like some form of non-muscular neural interface is the way to go.

Actually, the OCZ Vertex http://www.ocztechnology.com/products/flash_drives/ocz_vertex_series_sata_ii_2_5-ssd/ can sustain about 230MB/s. This is equal to about 2.3 Gb/sec. Allowing for the rate at which SSD technology seems to be changing, I'd say that this standard is just in the nick of time. Chances are that the next thing in high-end, consumer SSDs will saturate a SATA link. If this standard doesn't get pushed out soon, drive manufacturers will be doing ugly, proprietary, OS specific hacks to support multiple SATA links to a single device. In addition, lots of people are packaging multiple physical drives into a single SSD with an internal RAID-0 controller. These are definitely being(or soon going to be) held back by the 3BG/s SATA link.

Really? That's extremely strange considering that the operating system and browser which I am using to type this message are running off of a 16GB OCZ Rally2 which I am holding in my hand. Here, I'll read what it says on the body, again ...

Yep: It still says OCZ Rally 2 16GB.

According to their official website they don't make any 16GB drives in the Rally 2 series.

But the site does kind of suck so it might be in the deprecated model line.

It's true that TRIM is very much in it's infancy, but the clouds aren't as dark in SSD's future as they once were (even as little as a month ago). Many, many, many companies see SSD's as the future of storage and I'm inclined to agree with them. With that kind of muscle propelling development and increased consumer interest fuelling funding, the landscape is and will continue to change very rapidly.

My own take on things, FWIW, is that tapes will go the way of the floppy and spinning disks will become near-term storage for most enterprises. Look for SSD's to become mainstream SAN devices, especially as hardware manufacturers remove the driver/OS hurdle and present SAN devices as "just another disk". And as SSD's mature, I expect the performance gap between DRAM and SSD's to shrink quite a lot.

Don't forget the Neural Impulse Actuator from OCZ.

Levitating a ball by concentrating? This article is a bit behind. Please refer to: http://www.ocztechnology.com/products/ocz_peripherals/nia-neural_impulse_actuator And for an earlier poster, no these things will never work straight out the box (not for years anyway), because just like teaching your brain to control your muscles, teaching your brain to control the computer is also a long learning process. It requires mental states and processes that, in some cases, are completely foreign to our brains. Prothetics are a little different because the brain already knows how to control an arm, so by teaching the computer exactly what signal to expect for any given motion, when you get into things like controling additional "appendages" beyond your regular biological nerves, you are, in effect, teaching your brain to use a whole brand new appandage. Now, some of these are very simple - such as the levitating ball if you concentrate trick. It will simply read the brain-wave intensity and if strong enough, the ball will rise. But when you start getting into things even as simple as the OCZ NIA above, where only 3 different brain-waves or "fingers" they call them are picked up by the device, the task for your brain is signifiantly increased. Its never as easy as just WANTing to move forward - the brain wave associated with controling the device has nothing to do with your desire to move forward - at least not right away. Only with long long practice and calibration does your brain adapt and learn and able to actually control a device. Its almost impossible to explain in words the state of mind required for something like this to work. The closest I can come is - for those who have ever practiced mediation or a martial art - the empty zen state one can sometimes achieve with these practices. Thinking nothing, blank. You then throw yourself into the game and watch as your character moves around completely randomly, or so it would seem at first. After a few hours, you begin to see that the motion isn't entirely random. And after many many more hours the subconsious connection between a certain state of mind and the resulting action on the computer screen begins to solidify. You practice long enough, and controling the machine WILL be as easy as walking or lifting your arm - you never even have to think about it - just WANT to do it. "The views expressed here are mine and do not reflect the official opinion of my employer or the organization through which the Internet was accessed."

Levitating a ball by concentrating? This article is a bit behind. Please refer to:

http://www.ocztechnology.com/products/ocz_peripherals/nia-neural_impulse_actuator

And for an earlier poster, no these things will never work straight out the box (not for years anyway), because just like teaching your brain to control your muscles, teaching your brain to control the computer is also a long learning process. It requires mental states and processes that, in some cases, are completely foreign to our brains. Prothetics are a little different because the brain already knows how to control an arm, so by teaching the computer exactly what signal to expect for any given motion, when you get into things like controling additional "appendages" beyond your regular biological nerves, you are, in effect, teaching your brain to use a whole brand new appandage. Now, some of these are very simple - such as the levitating ball if you concentrate trick. It will simply read the brain-wave intensity and if strong enough, the ball will rise. But when you start getting into things even as simple as the OCZ NIA above, where only 3 different brain-waves or "fingers" they call them are picked up by the device, the task for your brain is signifiantly increased. Its never as easy as just WANTing to move forward - the brain wave associated with controling the device has nothing to do with your desire to move forward - at least not right away. Only with long long practice and calibration does your brain adapt and learn and able to actually control a device. Its almost impossible to explain in words the state of mind required for something like this to work. The closest I can come is - for those who have ever practiced mediation or a martial art - the empty zen state one can sometimes achieve with these practices. Thinking nothing, blank. You then throw yourself into the game and watch as your character moves around completely randomly, or so it would seem at first. After a few hours, you begin to see that the motion isn't entirely random. And after many many more hours the subconsious connection between a certain state of mind and the resulting action on the computer screen begins to solidify. You practice long enough, and controling the machine WILL be as easy as walking or lifting your arm - you never even have to think about it - just WANT to do it.

"The views expressed here are mine and do not reflect the official opinion of my employer or the organization through which the Internet was accessed."

A couple of companies (http://www.emotiv.com and http://www.ocztechnology.com/) are already working on products utilizing direct "mind control" style interfaces (previously posted on ./ here: http://hardware.slashdot.org/article.pl?sid=08/03/22/138201 among others).

Interestingly both of those products also utilize facial expression recognition to supplement the basic "mind reading" done by the probes attached to your head!

I'd like to see where this technology goes outside of the gaming industry, far better to be able to use it to control a full computer UI.

Like this? http://www.ocztechnology.com/products/flash_drives/ocz_throttle_esata_flash_drive http://www.newegg.com/Product/Product.aspx?Item=N82E16820227390&Tpk=N82E16820227390 It's about twice the cost of most USB flash drives the same size and it still requires a USB connection to power the device on most computers. You get to carry around a cable to use it... Awesome!

Most USB flash drives are very slow, and rely on heavy cacheing to make them usable. That doesn't help when you need to write large amounts of data (ie. in an apt-get update/install).

Some flash drives that advertise themselves as 10-15MiB/sec write capable peak out at only 1 or 2, and even less with small-block random I/O (since the erase-write cycle operates on relatively large blocks).

Several vendors make specialized flash drives that are somewhat more expensive (ie. 20-50% over average), but perform much better.

One is here: OCZ Turbo USB 2.0

whether laptop OR desktop, and the largest commercial SSD currently is a mere 128 GB.

Wrong

250GB OCZ SSD

I'm surprised no one else has mentioned the OCZ Neural Impulse Actuator yet.

Its essentially the same thing, but with a USB interface for PCs.

http://www.ocztechnology.com/products/ocz_peripherals/nia-neural_impulse_actuator

Price is a little steep for a first gen product, but it has been quite widely reviewed and said to actually work.

So, why would you pay $4 per GB for this when you can get a 1TB drive for around $140ish or so? Practically 0 seek time AND ~120MB/sec reads and ~90MB/sec writes. Hence WD upping the RPM's to 20K. SSD's, while pretty much in infancy for the consumer market, are already the fastest thing out there. It won't be long until they catch up on capacity.

Now, the only thing that stopped me from picking one of these bad boys up was checking out their support forums. It looks like these things have some pretty serious problems for at least some chipsets. While I realize that support forums represent the voiced minority, just running through those posts show some major issues at least with certain system combinations. Not to mention, these SSD's are pretty new, yet OCZ just announced a new rev for the lineup, now complete with a USB interface built into the drive to allow for firmware upgrades? I know this is bleeding edge stuff, but wow.

Anyway, I really wanted to upgrade with one of these because the hard drive is the slowest component in the system usually and I didn't see the VelociRaptor as a big enough upgrade for the $$. However, after all these reported problems, I think I'll wait to see how things are in 6 months or so.

NIA

OCZ Core series drives come to that mark. RSP of $479 for the 128GB model. See their press release, or a news message

Source? The retarded "Overclocker 3d" (rolleyes) site /. chooses to link to for every story about this thing is dead, but Anandtech says it reads both brain waves and facial movements. OCZ's website says it uses electro-myogram, electro-encephalogram, and electro-oculogram -- electrical patterns from the face, brain, and eyes.

Hardly, OCZ has already released their neural impulse actuator, which allows gamers to map neural impulses to keys that would be used in gameplay. (ie. WADS) It's not vaporware, it's already here and on shelves (or will be very shortly.)

http://www.ocztechnology.com/aboutocz/press/2008/273

Okay I have to admit this is beginning to fall into the "I will believe it when I see it" category. Yes yes I know it's POSSIBLE, and definitely DOABLE, but it's never been done and honestly there'd be a HUGE ... uh yea... H U G E ! run on these things if they work as advertised. That said I hit every link I could find to get more info.

So far I'm coming up dry. The article /. links to says OCZ's going to be shipping the product "next week"... the article is dated March 1 '08, that means ... March 7 shipping to stores. I can't find reference to this product in ANY RETAILER MAJOR OR MINOR. There is no preorder option because as far as I can tell according to the retailers this product does not exist.

I am concerned about a hoax here.

Gamestop: Nada. Nothing OCZ NIA or OCZ Neural.
Amazon: Nothing useful. Some irrelevant books when searching for OCZ Nia. Less relevant results when refining the search.
OCZ Technologies: Web site makes exactly one mention http://www.ocztechnology.com/aboutocz/press/2008/262 scroll down.
-- Device not listed under products.
Fry's Electronics: Zip.

Looks to me like they aren't shipping anything to anyone retail. I'm begging someone to prove me wrong here, because it doesn't look like we'll be getting this technology in our hands inside the next couple weeks... or even the next few months. Signed: I want to be wrong. File this under "if_it_sounds_too_good_to_be_true..."

OCZ and Corsair make some high performance USB drives. Corsair's red Voyager GT (Not the standard blue one) gets over 20MB/sec write. Of course, "reasonably priced" may be a stretch... they're about double the price of much slower drives.

Per Corsair "Ram Guy" forums:

"The average read and write for a
  4 G GT FV 25.5mbs Wite and 34 Mbs Read
  Non GT 4G FV is 2.2mbs Write and 19.2mbs Read."

(Reference: http://www.asktheramguy.com/v3/showthread.php?t=65150&highlight=voyager+speed)

Corsair appears to be holding to that, offering replacements for a few that report 22MB/sec reads.

Product links:

Corsair Voyager GT 8GB:

http://www.newegg.com/Product/Product.aspx?Item=N82E16820233054

OCZ Rally 2 Dual channel advertises 8MB/sec+ writes:

http://www.ocztechnology.com/products/flash_drives/ocz_rally2_usb_2_0_dual_channel_flash_memory_drive

Actually, it's a question that has already been answered.

Not even close.

USB2 is theoretically 480Mbps, but in reality it maxes out around 260Mbps. (33MB/sec)
http://www.barefeats.com/usb2.html (read the "tested on Windows" part)
Plus, if you look into it further, using USB devices saps CPU power. USB2 is a polled I/O bus meaning that the CPU must constantly do work monitoring each device for activity. Even empty USB connectors require constant CPU effort to scan for device insertion.

Next, 7200RPM drives max out around 60MB/sec. We really don't have a reason to be using anything past ATA/66 if you have just one drive on each channel. I mean, look at the max speed for the 160MB Seagate desktop drive and the fastest 7200rpm notebook drive ever. Not 60MB/sec.
http://www.notebookreview.com/default.asp?newsID=2 840

Finally, the fastest class of USB2 flash drives is the dual channel ones. Essentially RAID stripe across two chips. It puts out data at 28MB/sec. http://www.ocztechnology.com/products/flash_drives /ocz_rally_high_performance_usb_2_0-dual_channel-f lash_memory_drive

Flash drives have the advantage in seek time, but the majority of time you're working with RAM on disk, it's going to be pages, which the bandwidth can easily negate the seek time advantage.

I think most here knew that this was always going to be a stupid vanity platform, almost as stupid as water-cooled memory modules. Now, the only thing more sad and stupid than a vanity platform, is one where the vanity isn't even there.

This should have ended as abandoned concept art in a drawer.

(PS. My current gaming rig is AMD X2-based, but if they don't have the performance/$ then they won't get in on the next upgrade)

DDr2 ram isn't being used because of it's abysmal timings. http://www.ocztechnology.com/aboutocz/press/2005/1 48
compare that to the 2-3-2-2 timings one can get on DDR modules.

The reason why there is such a huge discrepency between performance between some review sites and others is that some sites are using abysmal 3-3-3-3 timings DDR memory for the FX-60 while others are using the better timed DDR chips. For gaming there is a HUGE advantage to having 2-3-2-2 timings because the entire content of the ram can be dumped almost twice as often as 4-4-4 timed DDR2, which because of it's better frequencies can pump more data at a slower rate.

Mind you AMD will need DDR2 support in the future, unless they somehow decided GDDR3 was better, because in about a few years DDR2 modules will be coming down to the 2-2-2 timings level, and will blow away the standard ddr modules. i mean technically if you look at video cards with ddr2 and ddr3 memory there is no engineering reason why someone couldn't make a ddr2 or ddr3 memory that worked awesome today, but there is plenty of 'marketing' reasons why they nead to have a 'clear' roadmap into the future.

DDR memory still has a lot of years of life left in it if you get the good timings stuff, like ocz or patriot. too bad ddr2 is 240 pin and ddr1 is 184 pin, so one can't make them pin compatable.. and no doubht ddr3 and ddr4 won't be pin compatable when they come out either.

ah well, tought to say, but if i was at AMD and trying to think of a way to 'counter' the DDR2 solution intel is using i'd instead opt for the simplified GDDR3 as main system memory. At least i'd consider the viability of doing so. the high end memory card market overnight decided to drop agp support and ddr2 support and go all pci-e with gddr3, because they were simpler more elgeant and properly working designs. agp is, was and remains a kludge to work around a problem that a better solution hadn't been thought of and ddr2 is full of legacy design needs from it's legacy heritage too.

Anyways, I'd rather see an AMD system (on 65 nm core) with GDDR3 modules than DDR-II modules.