Subject: General Tech, Storage | December 5, 2013 - 10:23 PM | Scott Michaud
Tagged: Intel, ssd
Computer storage website, Myce, got a hold of a few slides from Intel's SSD division. The semiconductor giant is expected to have (at least) nine active product lines with new SKUs apparently coexisting with certain older models. Two of the PCIe-based product lines, the P3700 series and the P3500 series, are expected to be available in capacities of up to 2TB. They will apparently be available in 2.5" form factor as well.
Image Credit: Myce
Intel has not produced the most mindblowing components over the last 3-4 years but, to my knowledge, they have been effective at wooing the enterprise customers. 2.8 GB/s reads and 1.7 GB/s writes at 450,000 IOPS for reading (150,000 IOPS for writes) seem pretty good, though. Combined with Intel's 5-year warranty and it will probably find its way into a few servers.
Each of the new products will be fabricated on the 20nm process (the older 910 Series and DC S3700 Series, both from 2012, will remain 25nm). Of course Intel has access to smaller processes at this point but, since these are enterprise products, it makes sense for them to use the more tried and true methods for the time being.
If you are interested in enterprise SSDs, keep an eye out in a couple of quarters. Maybe we will even see some stuff coming out of CES in a month.
Subject: General Tech, Graphics Cards, Processors | November 28, 2013 - 03:30 AM | Scott Michaud
Tagged: Intel, Xeon Phi, gpgpu
Intel was testing the waters with their Xeon Phi co-processor. Based on the architecture designed for the original Pentium processors, it was released in six products ranging from 57 to 61 cores and 6 to 16GB of RAM. This lead to double precision performance of between 1 and 1.2 TFLOPs. It was fabricated using their 22nm tri-gate technology. All of this was under the Knights Corner initiative.
In 2015, Intel plans to have Knights Landing ready for consumption. A modified Silvermont architecture will replace the many simple (basically 15 year-old) cores of the previous generation; up to 72 Silvermont-based cores (each with 4 threads) in fact. It will introduce the AVX-512 instruction set. AVX-512 allows applications to vectorize 8 64-bit (double-precision float or long integer) or 16 32-bit (single-precision float or standard integer) values.
In other words, packing a bunch of related problems into a single instruction.
The most interesting part? Two versions will be offered: Add-In Boards (AIBs) and a standalone CPU. It will not require a host CPU, because of its x86 heritage, if your application is entirely suited for an MIC architecture; unlike a Tesla, it is bootable with existing and common OSes. It can also be paired with standard Xeon processors if you would like a few strong threads with the 288 (72 x 4) the Xeon Phi provides.
And, while I doubt Intel would want to cut anyone else in, VR-Zone notes that this opens the door for AIB partners to make non-reference cards and manage some level of customer support. I'll believe a non-Intel branded AIB only when I see it.
Subject: General Tech | November 27, 2013 - 12:50 PM | Jeremy Hellstrom
Tagged: amd, Intel, arm, sales
Chips are hot this year, an increase in sales volume of 27% in Q1, 24% in Q2 and similar growth is expected over the coming year. Unfortunately for AMD and Intel most of these chips are in mobile devices, a market which neither company has leveraged successfully as of yet; PC chip sales have declined steadily over the previous quarters. The only good news is for AMD who managed to take a slightly larger share of this shrinking market. Both companies are going to have to become much more focussed on the ultra low voltage mobile market if they want to remain profitable, which means less development on high end desktop processors. Grab more market stats over at The Inquirer.
"PROCESSOR CHIP SALES will increase by almost quarter this year thanks to the growing demand for mobile devices such as smartphones and tablets, analyst outfit IHS has predicted."
Here is some more Tech News from around the web:
- Open-Source NVIDIA Driver Works On Some GeForce 700 GPUs, Fails On Others @ Phoronix
- Windows CE On A Raspberry Pi @ Hack a Day
- CIOs, IT chiefs: ARRGH! What do you MEAN, HR just bought 400 iPads and didn't tell us @ The Register
- Netgear R6250 Dual Band Gigabit Smart WiFi Router @ eTeknix
- Rosewill RPLC-200PKIT Powerline Adapter Kit @ Benchmark Reviews
Subject: General Tech | November 22, 2013 - 12:29 PM | Jeremy Hellstrom
Tagged: Supercomputing Conference, Intel, SoC, Near Memory, knights landing
Intel spilled more beans about the new Near Memory architecture that will be accompanying their new Xeon release. The memory will be stacked directly onto the CPU giving much quicker access than you would normally see from DDR3 which has to travel over the motherboard. They have not disclosed expected speeds, which could be up to what we see in current CPU caches only in much larger sizes. This is not quite a Xeon SoC but in the presentation The Register heard of Intel's plans to incorporate optical fabrics and switches onto the CPUs as well with size being the only limit. Perhaps they do have a leg to stand on when they claim the return to power of homogeneous computing.
"According to an EE Times report, Intel's Rajeeb Hazra, a VP and general manager of its data centre group, said Intel would customise high-end Xeon processors and Xeon Phi co-processors by closely integrating memory, both by adding memory dies to a processor package and, at a later date, integrating layers of memory dies into the processor along with optical fabrics and switches."
Here is some more Tech News from around the web:
- Forget terabit Ethernet, the next step is 400 gig – if we can afford the R&D @ The Register
- Migrating from Windows XP – Time to move on... @ The Register
- Microsoft to launch new version of Xbox One in 1Q14 @ DigiTimes
- Bitcoin Mining with a GPU is Still Viable - Using Litecoin @ hardCOREware
- ASUS RT-AC68U & PCE-AC68 Wireless AC Review @ Hardware Canucks
Subject: General Tech | November 20, 2013 - 12:53 PM | Jeremy Hellstrom
Tagged: Xeon Phi, knights landing, Intel, 14nm
Intel has been talking up the Xeon Phi, first of the Knight's Landing chips which shall arrive in the not too distant future. This new architecture is touted to bring a return of homogeneous systems architecture which will perform parallel processing on its many cores, currently 61 is the number being tossed around, at a level of performance that will exceed the GPU accelerated heterogeneous architecture being pushed by AMD and NVIDIA. Whether this is true or not remains to be seen but many server builders may prefer the familiar CPU only architecture and as at least some of the Phi's will be available in rack mounted form and not just addin cards they may choose Intel out of habit. You can also read about Micron's Automata Processor which The Register reports can outperform a 48-chip cluster of Intel Xeon 5650s in certain scenarios.
"From Intel's point of view, today's hottest trend in high-performance computing – GPU acceleration – is just a phase, one that will be superseded by the advent of many-core CPUs, beginning with Chipzilla's next-generation Xeon Phi, codenamed "Knights Landing"."
Here is some more Tech News from around the web:
- AMD makes more profit than Sony on the PS4 @ The Inquirer
- Decades ago, computing was saved by CMOS. Today, no hero is in sight @ The Register
- Who wants 10TB of FREE cloud storage? Hands down if China is a deal breaker? @ The Register
- Benchmarking Amazon's New EC2 "C3" Instance Types @ Phoronix
- What took you 2 years? LSI finally rolls out next-gen SandForce kit @ The Register
- Beginners Guide: How To Install / Remove an Intel Socket LGA2011 CPU @ PCSTATS
- Hardware.Info UK Awards 2013: vote and win a Zotac GeForce GTX 780 AMP! Edition
Subject: General Tech, Chipsets, Storage | November 12, 2013 - 04:37 PM | Scott Michaud
Tagged: Intel, 9-series, SATA Express
Intel is preparing to launch several processors next year. For back-to-school, Haswell will return with new SKUs and a new 9-series chipset; in the holiday season, Haswell-E will arrive for high-end (high wattage) enthusiasts on the X99 chipset; and, just before 2015, Broadwell-K will be available for the mainstream 9-series desktop.
The specification, which more than triples SATA 6Gbps's "up-to 600MB/s" bandwidth rating, will not be validated for Intel 9 Series chipsets. Intel was originally rumored to be its launch partner. The host connector accepts connections from both SATA (up to two per host connector) and PCIe-based (one device, up to two lanes) hard drives. Two PCIe lanes provides 2GB/s of bandwidth.
It seems like the real benefit is to allow internal drives be connected with PCIe speeds through a ribbon-cable. Currently Intel has not given a reason to pass on the standard.
Subject: General Tech | November 4, 2013 - 01:28 PM | Jeremy Hellstrom
Tagged: Intel, linux, open source, Broadwell
Over the weekend 62 patches to the Linux kernel were released, enabling Broadwell GPU support well ahead of the processors scheduled release date. Not only is this great news for open source enthusiasts who appreciate it when large companies like Intel release detailed driver code but also means that Broadwell should function well with Linux on its release date. Phoronix also reports that more code is scheduled to arrive this week to enable other features which are unique to Broadwell, keep your eyes peeled for any specifications we can infer from the code as it becomes available
"While Intel's Broadwell processors won't be launching until 2014 as the successor to Haswell, this weekend the initial open-source Linux GPU kernel driver was published ahead of the Linux 3.13 kernel merge window. The changes are massive and it's looking like the Broadwell graphics improvements will be astonishing and provide significant improvements over Haswell and earlier generations of Intel graphics."
Here is some more Tech News from around the web:
- Netflix starts 4K TV trial ahead of 2014 rollout @ The Inquirer
- Linux 3.12 Released, Linus Proposes Bug Fix-Only 4.0 @ Slashdot
- Google announces another partially fixed security flaw @ The Inquirer
- Google teaches Chrome Canary to sing when it sniffs dodgy downloads @ The Register
- Ding-dong! Who's at the door now with a big wad of cash, BlackBerry? @ The Register
- TRENDnet AC1750 Dual Band Wireless Router @ NikKTech
Subject: Processors, Mobile | October 29, 2013 - 12:24 PM | Ryan Shrout
Tagged: techcon, Intel, arm techcon, arm, Altera, 14nm
In February of this year Intel and Altera announced that they would be partnering to build Altera FPGAs using the upcoming Intel 14nm tri-gate process technology. The deal was important for the industry as it marked one of the first times Intel has shared its process technology with another processor company. Seen as the company's most valuable asset, the decision to outsource work in the Intel fabrication facilities could have drastic ramifications for Intel's computing divisions and the industry as a whole. This seems to back up the speculation that Intel is having a hard time keeping their Fabs at anywhere near 100% utilization with only in-house designs.
Today though, news is coming out that Altera is going to be included ARM-based processing cores, specifically those based on the ARMv8 64-bit architecture. Starting in 2014 Altera's high-end Stratix 10 FPGA that uses four ARM Cortex-A53 cores will be produced by Intel fabs.
The deal may give Intel pause about its outsourcing strategy. To date the chip giant has experimented with offering its leading-edge fab processes as foundry services to a handful of chip designers, Altera being one of its largest planned customers to date.
Altera believes that by combing the ARMv8 A53 cores and Intel's 14nm tri-gate transistors they will be able to provide FPGA performance that is "two times the core performance" of current high-end 28nm options.
While this news might upset some people internally at Intel's architecture divisions, the news couldn't be better for ARM. Intel is universally recognized as being the process technology leader, generally a full process node ahead of the competition from TSMC and GlobalFoundries. I already learned yesterday that many of ARM's partners are skipping the 20nm technology from non-Intel foundries and instead are looking towards the 14/16nm FinFET transitions coming in late 2014.
ARM has been working with essentially every major foundry in the business EXCEPT Intel and many viewed Intel's chances of taking over the mobile/tablet/phone space as dependent on its process technology advantage. But if Intel continues to open up its facilities to the highest bidders, even if those customers are building ARM-based designs, then it could drastically improve the outlook for ARM's many partners.
UPDATE (7:57pm): After further talks with various parties there are a few clarifications that I wanted to make sure were added to our story. First, Altera's FPGAs are primarly focused on the markets of communication, industrial, military, etc. They are not really used as application processors and thus are not going to directly compete with Intel's processors in the phone/tablet space. It remains to be seen if Intel will open its foundries to a directly competing product but for now this announcement regarding the upcoming Stratix 10 FPGA on Intel's 14nm tri-gate is an interesting progression.
Subject: General Tech, Processors | October 28, 2013 - 07:21 PM | Scott Michaud
Tagged: Intel, Haswell-E, Broadwell-K, Broadwell
Ivy Bridge-E was confirmed for this holiday season and Haswell-E was proclaimed to follow in Holiday 2014 bringing good tidings of comfort and joy (and DDR4). Broadwell, the Haswell architecture transitioned to a 14nm process technology, was expected to be delayed until at least 2015 because it was not on any roadmap before that.
Image credit: VR-Zone China
Until recently when something called "Broadwell-K" popped up slated for late Holiday 2014.
VR-Zone China, the site which broke this story (machine translated), cautiously assumes Broadwell-K signifies the platform will first arrive for the mainstream enthusiast. This would align with Intel's current "K" branding of unlocked processors and make sense to be introduced for the Consumer product segment without a Business offering.
If true, which seems likely, the question then becomes why. So let us speculate!
One possible (but almost definitely incorrect) reason is that Intel was able to get the overclocking challenges at 22nm solved and, thus, they want to build hype over what the enthusiasts can accomplish. Josh Walrath, our monitor of the fabrication industry's pulse at PC Perspective, did not bother entertaining the idea. His experiences suggest 14nm and 22nm are "not so different".
But, in the same discussion, Ryan wondered if Intel just could not get power low enough to release anything besides the upper mainstream parts. Rather than delay further, release the parts as they can fit in whatever TDP their market demands. Josh believes that is "as good [of a theory] as any". This also seems like a very reasonable possibility to me, too.
Two other theories: yields are sufficient for the "K" market (but nowhere else) or that Intel could be throwing a bone to the mid-range (lower than Haswell-E) enthusiast by letting them lead. It could also be almost any combination of the above or more.
Or, of course, Broadwell-K could refer to something completely arbitrary. At this point, no-one knows but anyone can guess.
So then, your turn? Comments await.
The Really Good Times are Over
We really do not realize how good we had it. Sure, we could apply that to budget surpluses and the time before the rise of global terrorism, but in this case I am talking about the predictable advancement of graphics due to both design expertise and improvements in process technology. Moore’s law has been exceptionally kind to graphics. We can look back and when we plot the course of these graphics companies, they have actually outstripped Moore in terms of transistor density from generation to generation. Most of this is due to better tools and the expertise gained in what is still a fairly new endeavor as compared to CPUs (the first true 3D accelerators were released in the 1993/94 timeframe).
The complexity of a modern 3D chip is truly mind-boggling. To get a good idea of where we came from, we must look back at the first generations of products that we could actually purchase. The original 3Dfx Voodoo Graphics was comprised of a raster chip and a texture chip, each contained approximately 1 million transistors (give or take) and were made on a then available .5 micron process (we shall call it 500 nm from here on out to give a sense of perspective with modern process technology). The chips were clocked between 47 and 50 MHz (though often could be clocked up to 57 MHz by going into the init file and putting in “SET SST_GRXCLK=57”… btw, SST stood for Sellers/Smith/Tarolli, the founders of 3Dfx). This revolutionary graphics card at the time could push out 47 to 50 megapixels and had 4 MB of VRAM and was released in the beginning of 1996.
My first 3D graphics card was the Orchid Righteous 3D. Voodoo Graphics was really the first successful consumer 3D graphics card. Yes, there were others before it, but Voodoo Graphics had the largest impact of them all.
In 1998 3Dfx released the Voodoo 2, and it was a significant jump in complexity from the original. These chips were fabricated on a 350 nm process. There were three chips to each card, one of which was the raster chip and the other two were texture chips. At the top end of the product stack was the 12 MB cards. The raster chip had 4 MB of VRAM available to it while each texture chip had 4 MB of VRAM for texture storage. Not only did this product double performance from the Voodoo Graphics, it was able to run in single card configurations at 800x600 (as compared to the max 640x480 of the Voodoo Graphics). This is the same time as when NVIDIA started to become a very aggressive competitor with the Riva TnT and ATI was about to ship the Rage 128.