Subject: Editorial | April 27, 2017 - 12:19 PM | Alex Lustenberg
Tagged: podcast, Win 3.11, ssd, riotoro, Optane Memory, Optane, Intel, GTX 1080Ti, fsp, evga, EK Supremacy, corsair, biostar, asus, video
PC Perspective Podcast #447 - 04/27/17
Join us for loads of Intel Optane, multiple water cooling parts, a Mini-ITX AM4 board, and more!
The URL for the podcast is: http://pcper.com/podcast - Share with your friends!
- iTunes - Subscribe to the podcast directly through the iTunes Store (audio only)
- Google Play - Subscribe to our audio podcast directly through Google Play!
- RSS - Subscribe through your regular RSS reader (audio only)
- MP3 - Direct download link to the MP3 file
Hosts: Jeremy Hellstrom, Allyn Malventano, Ken Addison, Morry Teitelman
Peanut Gallery: Alex Lustenberg
Program length: 1:50:22
Week in Review:
News items of interest:
Hardware/Software Picks of the Week
Subject: Storage | April 24, 2017 - 05:20 PM | Jeremy Hellstrom
Tagged: XPoint, srt, rst, Optane Memory, Optane, Intel, hybrid, CrossPoint, cache, 32GB, 16GB
At $44 for 16GB or $77 for a 32GB module Intel's Optane memory will cost you less in total for an M.2 SSD, though a significantly higher price per gigabyte. The catch is that you need to have a Kaby Lake Core system to be able to utilize Optane, which means you are unlikely to be using a HDD. Al's test show that Optane will also benefit a system using an SSD, reducing latency noticeably although not as significantly as with a HDD.
The Tech Report tested it differently, by sourcing a brand new desktop system with Kaby Lake Core APU that did not ship with an SSD. Once installed, the Optane drive enabled the system to outpace an affordable 480GB SSD in some scenarios; very impressive for a HDD. They also did peek at the difference Optane makes when paired with aforementioned affordable SSD in their full review.
"Intel's Optane Memory tech purports to offer most of the responsiveness of an SSD to systems whose primary storage device is a good old hard drive. We put a 32GB stick of Optane Memory to the test to see whether it lives up to Intel's claims."
Here are some more Storage reviews from around the web:
- Intel Optane Memory Review - 1.4GB/s Speed & 300K IOPS for $44 @ The SSD Review
- The Intel Optane Memory Module Review @ Hardware Canucks
- Kingston DCP1000 NVMe SSD Reaches 7GB/s @ Kitguru
- WD Blue 1,000 GiB SSD @ Hardware Secrets
- Synology DiskStation DS916+ 4-Bay NAS @ Kitguru
- Drobo 5N2 NAS @ Kitguru
- Kingston Ultimate GT 2TB Flash Drive @ The SSD Review
- Toshiba X300 6TB HDD @ Kitguru
Introduction, Specifications, and Requirements
Finally! Optane Memory sitting in our lab! Sure, it’s not the mighty P4800X we remotely tested over the past month, but this is right here, sitting on my desk. It’s shipping, too, meaning it could be sitting on your desk (or more importantly, in your PC) in just a matter of days.
The big deal about Optane is that it uses XPoint Memory, which has fast-as-lightning (faster, actually) response times of less than 10 microseconds. Compare this to the fastest modern NAND flash at ~90 microseconds, and the differences are going to add up fast. What’s wonderful about these response times is that they still hold true even when scaling an Optane product all the way down to just one or two dies of storage capacity. When you consider that managing fewer dies means less work for the controller, we can see latencies fall even further in some cases (as we will see later).
Introduction and Specifications
XPoint. Optane. QuantX. We've been hearing these terms thrown around for two years now. A form of 3D stackable non-volatile memory that promised 10x the density of DRAM and 1000x the speed and endurance of NAND. These were bold statements, and over the following months, we would see them misunderstood and misconstrued by many in the industry. These misconceptions were further amplified by some poor demo choices on the part of Intel (fortunately countered by some better choices made by Micron). Fortunately cooler heads prevailed as Jim Handy and other industry analysts helped explain that a 1000x improvement at the die level does not translate to the same improvement at the device level, especially when the first round of devices must comply with what will soon become a legacy method of connecting a persistent storage device to a PC.
Did I just suggest that PCIe 3.0 and the NVMe protocol - developed just for high-speed storage, is already legacy tech? Well, sorta.
That 'Future NVM' bar at the bottom of that chart there was a 2-year old prototype iteration of what is now Optane. Note that while NVMe was able to shrink down the yellow bar a bit, as you introduce faster and faster storage, the rest of the equation (meaning software, including the OS kernel) starts to have a larger and larger impact on limiting the ultimate speed of the device.
NAND Flash simplified schematic (via Wikipedia)
Before getting into the first retail product to push all of these links in the storage chain to the limit, let's explain how XPoint works and what makes it faster. Taking random writes as an example, NAND Flash (above) must program cells in pages and erase cells in blocks. As modern flash has increased in capacity, the sizes of those pages and blocks have scaled up roughly proportionally. At present day we are at pages >4KB and block sizes in the megabytes. When it comes to randomly writing to an already full section of flash, simply changing the contents of one byte on one page requires the clearing and rewriting of the entire block. The difference between what you wanted to write and what the flash had to rewrite to accomplish that operation is called the write amplification factor. It's something that must be dealt with when it comes to flash memory management, but for XPoint it is a completely different story:
XPoint is bit addressible. The 'cross' structure means you can select very small groups of data via Wordlines, with the ultimate selection resolving down to a single bit.
Since the programmed element effectively acts as a resistor, its output is read directly and quickly. Even better - none of that write amplification nonsense mentioned above applies here at all. There are no pages or blocks. If you want to write a byte, go ahead. Even better is that the bits can be changed regardless of their former state, meaning no erase or clear cycle must take place before writing - you just overwrite directly over what was previously stored. Is that 1000x faster / 1000x more write endurance than NAND thing starting to make more sense now?
Ok, with all of the background out of the way, let's get into the meat of the story. I present the P4800X:
Subject: Processors | April 19, 2017 - 08:00 PM | Scott Michaud
Tagged: skylake-x, ryzen, kaby lake x, Intel, Core, coffee lake, amd
According to DigiTimes, Intel is expecting to release several new processors earlier than they had originally planned. That said, there are two issues with this report. The first point, which should be expected, is that it compares internal dates that were never meant to be public. It is not like Intel has changed their advertised roadmap.
The second problem is that it’s somewhat contradicted by Intel’s earlier, public statements.
Their rumor claims that Intel will push up the launch of Basin Falls, which is Skylake-X, Kaby Lake-X, and X299, by about two months (around June). It also claims that Coffee Lake, which was originally scheduled for January 2018, will be released in August 2017. Both of these moves are being attributed to AMD’s new products.
The potential, somewhat, sort-of contradiction comes from a tweet that Intel made back in February. In it, they said that the 8th generation of Core processors are expected for 2H’17. This time frame doesn’t include January, although it only barely includes August, too. If Intel was always planning on launching Coffee Lake for the “back to school” season, then at least that half of DigiTimes’ story would be completely incorrect. On the other hand, if Intel’s tweet was talking about a sampling / paper launch in December, with volume shipment soon to follow, then DigiTimes would be fairly accurate.
We don’t know unless someone at Intel confirms either-or.
As for Skylake-X and Kaby Lake-X, it would be interesting to see them launch at Computex / E3. Previous rumors (also from DigiTimes) that place it in the Gamescom, which is a huge gaming conference in Cologne. Interestingly, this rumor claims that only the four-, six-, eight-, and ten-core models will arrive at the time, with a twelve-core model waiting until the whole line was supposed to launch.
This omission makes me wonder if, in fact, Intel are rushing the launch, but they realize that they cannot get enough good chips to fill out the top-end SKU. In that case, it would make sense to push the smaller and partially-disabled chips out the door, while banking the big chips that can run all twelve cores at a reasonable voltage for some clock rate.
If so, that would, in fact, speak volumes about AMD’s roadmap (and Intel’s opinion of it).
Subject: General Tech | April 17, 2017 - 04:09 PM | Jeremy Hellstrom
Tagged: Intel, idf
It is not a post-PC world but it is now officially a post-IDF world as Intel announces they are too diverse to host such a PC-centric conference. It was 20 years ago today that the first IDF was held in Beijing and Intel announced some time ago the cancellation of this years event in China, however until today they had still planned to hold their scheduled event in San Francisco. The rationale offered is Intel's expansion into FPGAs, Optane storage, IoT devices, wireless communications and other fields pushes them beyond the scope traditionally represented at the IDF. Why cancellation of the event in preference of broadening the scope is not explained in their announcement. Ars Technica has related links here.
"While the company earlier said that it would not have a Chinese event, the San Francisco IDF was still being planned, albeit with a "new format," in the early months of 2017. It appears now that this 'new format' is in fact 'non-existence.'"
Here is some more Tech News from around the web:
- Leaked NSA point-and-pwn hack tools menace Win2k to Windows 8 @ The Register
- Spring tech gift guide: Fathers' Day, Easter and just because @ The Inquirer
- Microsoft raises pistol, pulls the trigger on Windows 7, 8 updates for new Intel, AMD chips @ The Register
- The F-Secure RADAR Advanced Security Scanner @ TechARP
- noblechairs EPIC Series Gaming Chair @ techPowerUp
Subject: General Tech | April 12, 2017 - 12:30 PM | Jeremy Hellstrom
Tagged: sueball, qualcomm, Intel, blackberry, apple
Ah, the old days of Microsoft versus governments, Apple and Samsung, Intel and AMD, SCO and ... well just about everyone; lately there has not been much legal vitriol in the headlines. This may change as Qualcomm is now involved in several suits which are likely to become high profile. First up is what may be the driving force behind their need to seek restitution from Apple; they were just ordered to pay Blackberry $815 million after that company was successful in their legal battle to dispute certain royalty payments. Even a company as large as Qualcomm is going to feel the pain from that.
On to the real upcoming battle royal, featuring Apple and Qualcomm over the iPhone 7. It seems that Apple has taken offence to Qualcomm's claims that Apple handicapped their chip to ensure it did not out perform the models which contained an Intel chipset. Qualcomm released a large rebuttal to the lawsuit which Apple launched this January which you can read about at Ars Technica, or follow the link to read the entire document. This may prove to be the next interesting technological legal battle, stay tuned for more.
"For example, Qualcomm’s technological contributions enable popular smartphone apps such as Uber, Snapchat, Spotify, Apple Music, Skype, Google Maps, and Pokémon GO, among others."
Here is some more Tech News from around the web:
- Microsoft’s Windows 10 Creators Update: Small Tweaks For a Big Difference @ Techgage
- Microsoft reduces Patch Tuesday to an incomprehensible mess @ The Inquirer
- Windows 10 Creators Update general rollout begins with a privacy dialogue @ The Register
- Scientists Prove Your Phone's PIN Can Be Stolen Using Its Gyroscope Data @ Slashdot
- PC sales slump might be over, unless it isn't @ The Inquirer
- Systems-on-a-chip are a huge, unaudited attack surface, says Project Zero's Wi‑Fi attack man @ The Register
- Still no flash in a flash as chip supplies remain fried @ The Register
- Canada Hid the Konami Code In Its Commemorative $10 Bill Launch @ Slashdot
Subject: General Tech | March 30, 2017 - 01:20 PM | Jeremy Hellstrom
Tagged: Intel, 14nm, 14 nm FinFET
At Intel's Technology and Manufacturing Day event in San Francisco there was a lot of talk about how Intel's 14nm process technology compares to the 16nm, 14nm, and 10nm offerings of their competitors. Investors and enthusiasts are curious if Intel can hold their lead in process tech as Samsung seems to be on track to release chips fabbed on 10nm process before Intel will. Intel rightly pointed out that not all process tech is measured the same way and that pitch measurements give only one part of the picture; meaning Samsung might not actually be smaller than them.
The Tech Report were present at that meeting and have written up an in depth look at what Intel means when they dispute the competitions claims, as well as their rationale behind their belief that the 14nm node still has a lot of life left in it.
"As process sizes grow smaller and smaller, Intel believes that the true characteristics of those technology advances are being clouded by an over-reliance on a single nanometer figure. At its Technology and Manufacturing Day this week, the company defended its process leadership and proposed fresh metrics that could more accurately describe what a given process is capable of."
Here is some more Tech News from around the web:
- Scientists Discover Way To Transmit Taste of Lemonade Over Internet @ Slashdot
- There's a Samsung Galaxy S8 Microsoft Edition, for some reason @ The Inquirer
- 'Trash-80' escapes the dustbin of history with new TRS-80 emulator @ The Register
- Beyond Zelda: The first month of Switch games acts as a promising crystal ball @ Ars Technica
- ZX Spectrum Vega Plus backers complain of months-long refund delays @ The Register
- Microsoft wants screaming Windows fans, not just users @ The Register
- GDC 2017 and NVIDIA Editor's Day Coverage @ Neoseeker
Subject: Processors | March 28, 2017 - 11:48 AM | Morry Teitelman
Tagged: FinalWire, aida64, ryzen, amd, Intel
Courtesy of FinalWire
Today, FinalWire Ltd. announced the release of version 5.90 of their diagnostic and benchmarking tool, AIDA64. This new version updates their Extreme Edition, Engineer Edition, and Business Edition of the software, available here.
The latest version of AIDA64 has been optimized to work with AMD's Ryzen "Summit Ridge" and Intel's "Apollo Lake" processors, as well as updated to work with Microsoft's Windows 10 Creators Update release. The benchmarks and performance tests housed within AIDA64 have been updated for the Ryzen processor to utilize the VX2, FMA3, AES-NI and SHA instruction sets.
New features include:
- AVX2 and FMA accelerated 64-bit benchmarks for AMD Ryzen Summit Ridge processors
- Microsoft Windows 10 Creators Update support
- Optimized 64-bit benchmarks for Intel Apollo Lake SoC
- Improved support for Intel Cannonlake, Coffee Lake, Denverton, Kaby Lake-X, Skylake-X CPUs
- Preliminary support for AMD Zen server processors
- Preliminary support for Intel Gemini Lake SoC and Knights Mill HPC CPU
- NZXT Kraken X52 sensor support
- Socket AM4 motherboards support
- Improved support for Intel B250, H270, Q270 and Z270 chipset based motherboards
- EastRising ER-OLEDM032 (SSD1322) OLED support
- SMBIOS 3.1.1 support
- Crucial M600, Crucial MX300, Intel Pro 5400s, SanDisk Plus, WD Blue SSD support
- Improved support for Samsung NVMe SSDs
- Advanced support for HighPoint RocketRAID 27xx RAID controllers
- GPU details for nVIDIA GeForce GTX 1080 Ti, Quadro GP100, Tesla P6
Software updates new to this release (since AIDA64 v5.00):
- AVX and FMA accelerated FP32 and FP64 ray tracing benchmarks
- Vulkan graphics accelerator diagnostics
- RemoteSensor smartphone and tablet LCD integration
- Logitech Arx Control smartphone and tablet LCD integration
- Microsoft Windows 10 TH2 (November Update) support
- Proper DPI scaling to better support high-resolution LCD and OLED displays
- AVX and FMA accelerated 64-bit benchmarks for AMD A-Series Bristol Ridge and Carrizo APUs
- AVX2 and FMA accelerated 64-bit benchmarks for Intel Broadwell, Kaby Lake and Skylake CPUs
- AVX and SSE accelerated 64-bit benchmarks for AMD Nolan APU
- Optimized 64-bit benchmarks for Intel Braswell and Cherry Trail processors
- Advanced SMART disk health monitoring
- Hot Keys to switch LCD pages, start or stop logging, show or hide SensorPanel
- Corsair K65, K70, K95, Corsair Strafe, Logitech G13, G19, G19s, G910, Razer Chroma RGB LED keyboard support
- Corsair, Logitech, Razer RGB LED mouse support
- Corsair and Razer RGB LED mousepad support
- AlphaCool Heatmaster II, Aquaduct, Aquaero, AquaStream XT, AquaStream Ultimate, Farbwerk, MPS, NZXT GRID+ V2, PowerAdjust 2, PowerAdjust 3 sensor devices support
- Improved Corsair Link sensor support
- NZXT Kraken water cooling sensor support
- Corsair AXi, Corsair HXi, Corsair RMi, Enermax Digifanless, Thermaltake DPS-G power supply unit sensor support
- Support for Gravitech, LCD Smartie Hardware, Leo Bodnar, Modding-FAQ, Noteu, Odospace, Saitek Pro Flight Instrument Panel, Saitek X52 Pro, UCSD LCD devices
- Portrait mode support for AlphaCool and Samsung SPF LCDs
- System certificates information
- Advanced support for Adaptec and Marvell RAID controllers
AIDA64 is developed by FinalWire Ltd., headquartered in Budapest, Hungary. The company’s founding members are veteran software developers who have worked together on programming system utilities for more than two decades. Currently, they have ten products in their portfolio, all based on the award-winning AIDA technology: AIDA64 Extreme, AIDA64 Engineer, AIDA64 Network Audit, AIDA64 Business and AIDA64 for Android,, iOS, Sailfish OS, Tizen, Ubuntu Touch and Windows Phone. For more information, visit www.aida64.com.
Subject: Storage | March 27, 2017 - 12:16 PM | Allyn Malventano
Tagged: XPoint, Optane Memory, Optane, M.2, Intel, cache, 3D XPoint
We are just about to hit two years since Intel and Micron jointly launched 3D XPoint, and there have certainly been a lot of stories about it since. Intel officially launched the P4800X last week, and this week they are officially launching Optane Memory. The base level information about Optane Memory is mostly unchanged, however, we do have a slide deck we are allowed to pick from to point out some of the things we can look forward to once the new tech starts hitting devices you can own.
Alright, so this is Optane Memory in a nutshell. Put some XPoint memory on an M.2 form factor device, leverage Intel's SRT caching tech, and you get a 16GB or 32GB cache laid over your system's primary HDD.
To help explain what good Optane can do for typical desktop workloads, first we need to dig into Queue Depths a bit. Above are some examples of the typical QD various desktop applications run at. This data is from direct IO trace captures of systems in actual use. Now that we've established that the majority of desktop workloads operate at very low Queue Depths (<= 4), lets see where Optane performance falls relative to other storage technologies:
There's a bit to digest in this chart, but let me walk you through it. The ranges tapering off show the percentage of IOs falling at the various Queue Depths, while the green, red, and orange lines ramping up to higher IOPS (right axis) show relative SSD performance at those same Queue Depths. The key to Optane's performance benefit here is that it can ramp up to full performance at very low QD's, while the other NAND-based parts require significantly higher parallel requests to achieve full rated performance. This is what will ultimately lead to a much snappier responsiveness for, well, just about anything hitting the storage. Fun fact - there is actually a HDD on that chart. It's the yellow line that you might have mistook as the horizontal axis :).
As you can see, we have a few integrators on board already. Official support requires a 270 series motherboard and Kaby Lake CPU, but it is possible that motherboard makers could backport the required NVMe v1.1 and Intel RST 15.5 requirements into older systems.
For those curious, if caching is the only way power users will be able to go with Optane, that's not the case. Atop that pyramid there sits an 'Intel Optane SSD', which should basically be a consumer version of the P4800X. It is sure to be an incredibly fast SSD, but that performance will most definitely come at a price!
We should be testing Optane Memory shortly and will finally have some publishable results of this new tech as soon as we can!