Written by Hanners
Friday, 25 February 2011 08:57
You might have known it in a past life as LightPeak, but Intel's high speed I/O interface is now ready for prime time under the new name of Thunderbolt, complete with a deal with Apple to see it appear on their latest MacBook Pro parts before seeing any PC availability next year. Can Thunderbolt challenge USB and Serial ATA?
So here is the basic summary: Thunderbolt technology will combine both support for PCI Express based data transfers and DisplayPort technology for high resolution display outputs into a single controller and a single cable. That cable, as we saw yesterday, basically is identical in appearance to the mini-DisplayPort connection that we have seen on graphics cards for some time now. The connection will indeed run at 10 Gbps, making it about twice as fast (in theory) than USB 3.0, and will provide more than enough headroom for storage technologies and high speed data transfers into the future. Users like simple answers so Intel says that you can "transfer a full-length HD movie in less than 30 seconds" and "backup 1 year of continuous MP3 playback in just over 10 minutes." Pretty impressive stuff.
Technically, Thunderbolt is pretty interesting as both data and display signals will transfer over the same cable with arbitration from the controller chip. It is a bi-directional communication that actually provides 10 Gbps of bandwidth PER PORT. It remains backwards compatible with existing DisplayPort displays, but since those devices likely do NOT have female mini-DisplayPort connections on them, they will not facilitate the daisy-chaining required for connecting multiple devices to a single Thunderbolt connection. Unlike what some people worried, Thunderbolt will carry some amount of power over cable, though how much hasn't really been specified yet.
Technically, Thunderbolt is pretty interesting as both data and display signals will transfer over the same cable with arbitration from the controller chip. It is a bi-directional communication that actually provides 10 Gbps of bandwidth PER PORT. It remains backwards compatible with existing DisplayPort displays, but since those devices likely do NOT have female mini-DisplayPort connections on them, they will not facilitate the daisy-chaining required for connecting multiple devices to a single Thunderbolt connection. Unlike what some people worried, Thunderbolt will carry some amount of power over cable, though how much hasn't really been specified yet.
PC Perspective has a full explanation of the technology as it stands.
Written by Hanners
Friday, 25 February 2011 08:44
After seeing (and being rather impressed by) what the Pro version of OCZ's forthcoming Sandforce SF-2200 series controller-sporting SSDs were capable of the other week, now we can see what the consumer level Vertex 3 product will be all about via a number of previews.
If you pay attention to the table above you’ll notice that there are two specs for 4KB random write IOPS: burst and sustained. The burst value is for around 15 seconds of operation, the sustained is what happens when the firmware initiated performance cap kicks into action. By default the SF-2100/2200 drives have a cap of 20,000 IOPS for 4KB random writes. After a period of about 15 seconds, the max performance on these drives will drop to 20K. The SF-2500/2600 controllers are uncapped, max performance can remain at up to 60K IOPS.
The beta Vertex 3 review sample I have here today manages around 45K IOPS in our 4KB random write test. That test runs for 3 minutes straight so obviously the cap should’ve kicked in. However it didn’t.
I asked SandForce why this was. SandForce told me that the initial pre-release firmwares on the SF-2200 drives don’t have the cap enabled, but the final release will put the cap in place. I also asked SandForce if it was possible for one of its partners to ship with a special firmware build that didn’t have the cap in place. SandForce replied that anything was possible.
I asked OCZ if this meant the drive I was testing wasn’t representative of final, shipping performance. OCZ stated very clearly that performance will not change between the drive I have today and the drive that goes on sale in the next 2 months. To me this sounds like SF and OCZ have struck another exclusive firmware deal to ensure slightly higher performance on the Vertex 3 compared to a standard SF-2200 based drive.
The beta Vertex 3 review sample I have here today manages around 45K IOPS in our 4KB random write test. That test runs for 3 minutes straight so obviously the cap should’ve kicked in. However it didn’t.
I asked SandForce why this was. SandForce told me that the initial pre-release firmwares on the SF-2200 drives don’t have the cap enabled, but the final release will put the cap in place. I also asked SandForce if it was possible for one of its partners to ship with a special firmware build that didn’t have the cap in place. SandForce replied that anything was possible.
I asked OCZ if this meant the drive I was testing wasn’t representative of final, shipping performance. OCZ stated very clearly that performance will not change between the drive I have today and the drive that goes on sale in the next 2 months. To me this sounds like SF and OCZ have struck another exclusive firmware deal to ensure slightly higher performance on the Vertex 3 compared to a standard SF-2200 based drive.
The new SandForce SF-2281 controller used in the Vertex 3 takes the technology of the original SF-1200 line of controllers, and expands it in line with the SATA 6Gbps standard, with SandForce claiming outlandishly fast speeds of up to 550MB/sec for sequential reads and 525MB/sec for sequential writes. These will be the absolutle best-case speeds, but they're still a huge step up from the 355MB/sec and 215MB/sec respective read and write speeds of the Crucial C300 256GB, which has been the fastest single-drive SSD on the market for the last nine months.
To accomplish this speed improvement, OCZ has implemented a number of improvements to the Vertex 3, not least the inclusion of support for the higher bandwidth ONFI (Open NAND Flash Interface) 2.1 memory and a SATA 6Gbps interface. The version we've been sent is the standard Vertex 3, rather than the Vertex 3 Pro previewed at Anandtech, and this means that it's fitted with Micron's new NAND flash memory too.
To accomplish this speed improvement, OCZ has implemented a number of improvements to the Vertex 3, not least the inclusion of support for the higher bandwidth ONFI (Open NAND Flash Interface) 2.1 memory and a SATA 6Gbps interface. The version we've been sent is the standard Vertex 3, rather than the Vertex 3 Pro previewed at Anandtech, and this means that it's fitted with Micron's new NAND flash memory too.
OCZ plans to initially introduce two solid state drives in the Vertex 3 line, a 120GB version and a 240GB version. The 120GB version is set to arrive with a $249 MSRP and the 240GB with a $499 MSRP. Looking at current pricing for the Vertex 2 (which has fluctuated significantly in recent weeks), it seems the Vertex 3 will carry a roughly 20% to 25% price premium when it arrives in a few weeks. Considering how strong the Vertex 3's performance was throughout testing and that it's likely to get better as OCZ has more time to work with and tweak the firmware on the drive, we can't help but be excited for this product, despite the price premium it will command.
If you've been waiting to make the move to a solid state drive, the arrival of the Vertex 3 (and a handful of other next-gen SATA III drives) is going to shake things up a bit. Pricing on many of the current generation offerings is sure to fall somewhat, which we're will make the proposition more attractive to many users. And if you've got a system that doesn't have SATA III ports, by all means, consider the upgrade. If you've got a system with native SATA III support though, the Vertex 3 is one heck of a performer and will be worth the additional investment, as the numbers have shown.
If you've been waiting to make the move to a solid state drive, the arrival of the Vertex 3 (and a handful of other next-gen SATA III drives) is going to shake things up a bit. Pricing on many of the current generation offerings is sure to fall somewhat, which we're will make the proposition more attractive to many users. And if you've got a system that doesn't have SATA III ports, by all means, consider the upgrade. If you've got a system with native SATA III support though, the Vertex 3 is one heck of a performer and will be worth the additional investment, as the numbers have shown.
Written by Hanners
Thursday, 24 February 2011 08:34
Intel's latest Sandy Bridge architecture rules the CPU roost right now despite suffering from a motherboard chipset issue, but does AMD have what it takes to challenge their huge rival? The company has used the recent ISSCC 2011 conference to publish more details about their next-generation Bulldozer architecture to give us some clues about what they have (with every pun intended) in the pipeline.
The more I read about Bulldozer, the more impressed I am becoming. AMD has barely been able to keep up with Intel for the past 6 years, which is disappointing considering the success they found with the original Athlon and then Athlon 64. Since Intel introduced the Core 2 series, AMD has only been able to design and sell processors that would sometimes achieve around 90% of the overall performance of Intel’s top end processors. AMD also suffered from die size and performance per watt disparities as compared to Intel’s very successful Core 2 and Core i7/i5/i3 processors. The latest generation of Sandy Bridge based units again exposed how lacking in overall design and performance AMD’s Phenom families of chips were.
Before we all go off the deep end and claim that AMD will surpass Intel in overall performance, we need to calm down. We are getting perilously close to the very limits of IPC (instructions per clock) with current technology and designs. It appears to me that with the Bulldozer architecture, AMD should reach parity with Intel and their latest generation of CPUs when it comes to IPC per core. Where AMD could have an advantage over Intel are in several specific categories.
Before we all go off the deep end and claim that AMD will surpass Intel in overall performance, we need to calm down. We are getting perilously close to the very limits of IPC (instructions per clock) with current technology and designs. It appears to me that with the Bulldozer architecture, AMD should reach parity with Intel and their latest generation of CPUs when it comes to IPC per core. Where AMD could have an advantage over Intel are in several specific categories.
PC Perspective delves into the world of Bulldozer.
Written by Hanners
Wednesday, 23 February 2011 12:19
I guess this is the moment we get to relegate the PC Gaming Alliance to the annals of history...
One could argue that Valve software has become a de-facto platform champion both with its Steam marketplace and its relentless innovation when it comes to its own games. What can definitely be said is that the one group that should be championing the PC, the PC Gaming Alliance, is going backwards. In 2009 the group lost the biggest PC game developer/publisher Activision-Blizzard, and now it seems that both Microsoft and Nvidia have bid the alliance farewell.
This leaves the ‘promoter’ lineup for the industry body looking decidedly lame. Besides the high profile departures, AMD has dropped down to member. The list of promoters now reads Capcom, Dell, Epic Games, Intel, Razer and Sony DADC. That means that the list of supposed champions includes two cross platform publishers, one CPU builder, one peripheral maker, one system builder and a CD manufacturer.
This leaves the ‘promoter’ lineup for the industry body looking decidedly lame. Besides the high profile departures, AMD has dropped down to member. The list of promoters now reads Capcom, Dell, Epic Games, Intel, Razer and Sony DADC. That means that the list of supposed champions includes two cross platform publishers, one CPU builder, one peripheral maker, one system builder and a CD manufacturer.
PC Authority has the full story.
Written by Hanners
Tuesday, 22 February 2011 09:01
As manufacturing processes get smaller and smaller, so the technology required to create processors using those smaller process nodes becomes more complex. Enter Intel, who will be spending US $5 billion to create a fabrication plant in Arizona geared up for the creation of 14nm parts.
Intel Corp. on Friday announced plans to invest more than $5 billion to build a new chip manufacturing facility at its site in Chandler, Arizona. Although the company is looking forward 450mm chip fabs, the new factory will process 300mm wafers at least initially.
The new Arizona factory, designated Fab 42, will be the most advanced, high-volume semiconductor manufacturing facility in the world, according to Intel. Construction of the new fab is expected to begin in the middle of this year and is expected to be completed in 2013. The new factory will make chips using 14nm manufacturing process on 300mm wafers.
The new Arizona factory, designated Fab 42, will be the most advanced, high-volume semiconductor manufacturing facility in the world, according to Intel. Construction of the new fab is expected to begin in the middle of this year and is expected to be completed in 2013. The new factory will make chips using 14nm manufacturing process on 300mm wafers.
X-Bit Labs have the story.
Written by Hanners
Tuesday, 22 February 2011 08:38
We've been hearing a fair few unhappy rumblings of late surrounding OCZ's Vertex 2 drives, with the company introducing a new revision of the product featuring what is alleged to hold a small capacity and lower performance due to a switch to 25nm NAND flash memory. Is there any truth to this? Tom's Hardware investigates.
More recently, several Tom’s Hardware readers brought to my attention that OCZ had recently stopped shipping its Vertex 2 drives with 34 nm NAND flash from IM Flash Technologies (and 32 nm memory from Hynix), replacing them with 25 nm flash from the same Intel/Micron joint venture.
The switch-over completed in mid-January, according to OCZ, and it was largely low-key. That is to say, it wasn’t really picked up on until customers started seeing lower capacities and reporting reduced performance from certain models compared to the older drives.
Fast-forward to mid-February. Sensing increased discontent over the community’s reaction, OCZ released a notice announcing its industry-first transition to 25 nm NAND adoption. The news purported that the smaller lithography would translate to more affordable performance-oriented SSDs. Indeed, the last time I looked at Vertex 2 pricing was mid-January, and the 120 GB drive was selling for $250. Today you can find the "exact same" model for $230.
What wasn’t addressed in OCZ’s release was whether the $230 drive, with its fancy 25 nm IMFT-based NAND actually is any slower or smaller than its predecessor. Naturally, I wanted to dig. After all, the complaint I received over and over again from Tom’s Hardware readers was that they were buying specific models and getting less usable capacity and unexpected performance results.
The switch-over completed in mid-January, according to OCZ, and it was largely low-key. That is to say, it wasn’t really picked up on until customers started seeing lower capacities and reporting reduced performance from certain models compared to the older drives.
Fast-forward to mid-February. Sensing increased discontent over the community’s reaction, OCZ released a notice announcing its industry-first transition to 25 nm NAND adoption. The news purported that the smaller lithography would translate to more affordable performance-oriented SSDs. Indeed, the last time I looked at Vertex 2 pricing was mid-January, and the 120 GB drive was selling for $250. Today you can find the "exact same" model for $230.
What wasn’t addressed in OCZ’s release was whether the $230 drive, with its fancy 25 nm IMFT-based NAND actually is any slower or smaller than its predecessor. Naturally, I wanted to dig. After all, the complaint I received over and over again from Tom’s Hardware readers was that they were buying specific models and getting less usable capacity and unexpected performance results.
Read their findings in full over here.
Written by Hanners
Monday, 21 February 2011 17:01
I have to confess that I still can't see the lure of tablet systems such as the iPad at all personally, but it's clearly a huge growth market right now, with everyone forced to play catch-up to Apple's concept. Google's plans along these lines will come full to fruition with the release of the "Honeycomb" revision of their mobile device OS; Anandtech explore what it's all about.
Perhaps the biggest change in Honeycomb is that the Dalvik VM environment is now multi-core aware. This means the OS can now leverage the processing power and potential power savings of multi-core SoCs to speed things up and offer enhanced battery life. In fact, even single threaded apps can see speed improvements in Honeycomb. While running in a dual-core environment, the OS automatically lets the bytecode run on one core and the Dalvik garbage collector run on the other. Apps specifically written to support symmetric multiprocessing will see even larger performance gains. With companies like ARM (Cortex A15) and Qualcomm (Snapdragon “Krait”) already announcing their next-gen dual/quad-core SoC architectures, it will be exciting to see what kind of mobile performance we have in store for the future!
Check out the full preview over here.
Written by Hanners
Monday, 21 February 2011 16:38
Remember when Half-Life 2's source code got stolen? Ever wonder how or why it happened? Then you'll want to read this story.
At 6am on 7th May 2004, Axel Gembe awoke in the small German town of Schönau im Schwarzwald to find his bed surrounded by police officers. Automatic weapons were pointing at his head and the words "Get out of bed. Do not touch the keyboard" were ringing in his ears.
Gembe knew why they were there. But, bleary-eyed, he asked anyway.
"You are being charged with hacking into Valve Corporation's network, stealing the videogame Half-Life 2, leaking it onto the internet and causing damages in excess of $250 million," came the reply. "Get dressed."
Gembe knew why they were there. But, bleary-eyed, he asked anyway.
"You are being charged with hacking into Valve Corporation's network, stealing the videogame Half-Life 2, leaking it onto the internet and causing damages in excess of $250 million," came the reply. "Get dressed."
EuroGamer has this fascinating tale in full.
Written by Hanners
Thursday, 17 February 2011 08:53
In recent times, Sandforce's SF-1000 series controllers have ruled the roost in terms of performance, boasting a home in numerous SSDs from a number of manufacturers. Can their next-generation controllers mimic and improve on this notable success? A few sites have received early, pre-production professional level OCZ Vertex 3 Pro SSDs which are powered by a Sandforce SF-2500 part, potentially giving us an early feel for what they might have to offer.
Everything I've described up to this point applies to both the previous generation (SF-1200/1500) and the new generation (SF-2200/2500) of SandForce controllers. Now let's go over what's new:
1) Toggle Mode & ONFI 2 NAND support. Higher bandwidth NAND interfaces mean we should see much better performance without any architectural changes.
2) To accommodate the higher bandwidth NAND SandForce increased the size of on-chip memories and buffers as well as doubled the number of NAND die that can be active at one time. Finally there's native 6Gbps support to remove any interface bottlenecks. Both 1 & 2 will manifest as much higher read/write speed.
3) Better encryption. This is more of an enterprise feature but the SF-2000 controllers support AES-256 encryption across the drive (and double encryption to support different encryption keys for separate address ranges on the drive).
4) Better ECC. NAND densities and defect rates are going up, program/erase cycles are going down. The SF-2000 as a result has an improved ECC engine.
1) Toggle Mode & ONFI 2 NAND support. Higher bandwidth NAND interfaces mean we should see much better performance without any architectural changes.
2) To accommodate the higher bandwidth NAND SandForce increased the size of on-chip memories and buffers as well as doubled the number of NAND die that can be active at one time. Finally there's native 6Gbps support to remove any interface bottlenecks. Both 1 & 2 will manifest as much higher read/write speed.
3) Better encryption. This is more of an enterprise feature but the SF-2000 controllers support AES-256 encryption across the drive (and double encryption to support different encryption keys for separate address ranges on the drive).
4) Better ECC. NAND densities and defect rates are going up, program/erase cycles are going down. The SF-2000 as a result has an improved ECC engine.
At the heart of the upcoming Vertex 3 Pro SSD is a brand new controller from SandForce, the SF-2582. The SF-2582 is a member of the upcoming SandForce SF-2000 family of SSD processors, which was announced a few months back. The new SF-2000 SSD processor family builds upon the success of the current SF-1500 / SF-1200 series by adding 6Gb/s SATA III interface support, integrating a more powerful encryption engine, and adding support for new flash memory types. While new, the SF-2000 family does, however, leverage technologies from the previous generation. As such, we'd recommend checking out some of our previous SandForce coverage for some back-story on SandForce and what makes their SSD controllers so sought after. In our look at the OCZ Vertex Limited Edition, we go into detail on DuraWrite and RAISE and some of SandForce's other proprietary technologies. We highly recommend checking that story out.
As a professional level product, as well as by offering the best performance we have seen from any SSD to date, the Vertex 3 Pro costs a premium. Below are the prices set forth by OCZ in the included documentation, though they are cautious enough to warn of potential pricing changes before full product availability.
- 100G @ $525 ($5.25 / GB)
- 200G @ $775 ($3.87 / GB)
- 400G @ $1350 ($3.37 / GB)
- 100G @ $525 ($5.25 / GB)
- 200G @ $775 ($3.87 / GB)
- 400G @ $1350 ($3.37 / GB)
Written by Hanners
Thursday, 17 February 2011 08:39
It's amazing that we're still having to talk about GPU physics and who should support what in 2011 after it's been such a hot topic for years, but such discussions show no sign of ending for now. Along those lines, AMD's Manju Hegde (formerly of AGEIA) shares his current views on the GPU physics ecosystem.
In terms of gaming physics, the OpenCL offering Hegde talks about concerns a new project using Bullet Physics, and this is a strategy that AMD appears to be taking very seriously if you look at the guys AMD's hiring to join its physics team. 'We have Erwin Coumans as part of our team,' notes Hegde, 'who used to be at Sony and he's the guy who's brought the Bullet Physics engine to the industry, which is also used in cinematic effects.
'Also, we now have Takahiro Harada at AMD - he used to be a professor in Japan and then joined Havok and he's been here for two months. As you can see, we do have a physics strategy, which is built on this core team, and we're augmenting this with additional people too.'
'Also, we now have Takahiro Harada at AMD - he used to be a professor in Japan and then joined Havok and he's been here for two months. As you can see, we do have a physics strategy, which is built on this core team, and we're augmenting this with additional people too.'
bit-tech has the discussion in full.
More Articles...
Page 8 of 81