Storage bridges come in many varieties within the internal and external market segments. On the external side, they usually have one or more downstream SATA ports. The most popular uplink port is some sort of USB connection. eSATA as an uplink interface is on the way out. High-end products have Thunderbolt support. Within the USB storage bridge market, device vendors have multiple opportunities to tune their product design for specific use-cases.

Today's review will take a look at two bridges from Inateck, the FE2010 and the UA1001. The common characteristic is that both of them present a SATA port at one end and a USB 3.0 Micro-B at the other. However, the use-cases are completely different. While the FE2010 is an enclosure for a 2.5" drive and can operate only via bus power, the UA1001 is a SATA adaptor meant for use with high-power SATA devices such as 3.5" drives and optical drives. The UA1001 can operate with just the bus power for 2.5" drives. For other use-cases, it utilizes a bundled 24W power adapter. We utilized it in the latter mode in our review of the Seagate BarraCuda Pro 10TB helium drive.

The FE2010 is a tool-less design, allowing part of the rear to slide off in order to insert a 2.5" drive into the enclosure's SATA port. A small foam pad is also included in order to ensure that internal drive is held snugly within the enclosure. The gallery below shows that the FE2010 has an explicitly on/off switch and an activity LED. Internally, the unit uses a JMicron JMS578 bridge chip with UASP support. The firmware is configured to power down the SATA drive after 30 minutes of inactivity. The enclosure itself is made of ABS plastic and looks sturdy. However, because of its lightweight nature, it does feel a bit flimsy in the hand. The UA1001 is meant for use by consumers wishing to swap out SATA drives often. To that end, it advertises hot-plug capability and features an exposed SATA port. The ASMedia ASM1153E bridge chip is used in this product.

Gallery: Inateck FE2010 and UA1001 Storage Bridges

Consumers need to keep the following aspects in mind for external storage devices / enclosures with a USB interface:

• Support for UASP (USB-attached SCSI protocol) for better performance (reduced protocol overhead and support for SATA Native Command Queueing (NCQ))
• Support for TRIM to ensure SSDs in the external enclosure can operate optimally in the long run
• Support for S.M.A.R.T passthrough to enable monitoring of the internal SATA device by the host OS

The table below presents the detailed specifications and miscellaneous aspects of the units and how they compare.

Our evaluation routine for storage bridges borrows heavily from the testing methodology for direct-attached storage devices. The testbed hardware is reused. CrystalDiskMark is used for a quick overview, as it helps determine availability of UASP support and provides some performance numbers under ideal scenarios. Real-world performance testing is done with our custom test suite involving robocopy benchmarks and PCMark 8's storage bench.

Performance Benchmarks

CrystalDiskMark uses four different access traces for reads and writes over a configurable region size. Two of the traces are sequential accesses, while two are 4K rando accesses. Internally, CrystalDiskMark uses the Microsoft DiskSpd storage testing tool. The 'Seq Q32T1' sequential traces use 128K block size with a queue depth of 32 from a single thread, while the '4K Q32T1' ones do random 4K accesses with the same queue and thread configurations. The plain 'Seq' traces use a 1MiB block size. The plain '4K' ones are similar to the '4K Q32T1' except that only a single queue and single thread are used.

Comparing the '4K Q32T1' and '4K' numbers can quickly tell us whether the storage device supports NCQ (native command queuing) / UASP (USB-attached SCSI protocol). If the numbers for the two access traces are in the same ballpark, NCQ / UASP is not supported. This assumes that the host port / drivers on the PC support UASP. We can see that the FE2010 performs way better than the UA1001 while using the same SSD. It also has UASP support (with NCQ). That said, the use-case for UA1001 is more ith 3.5" hard drives and other slower SATA devices. But, it is good to keep this mind while trying to transfer data to/from spare SSDs using the UA1001.

Storage Bridge Benchmarks - CrystalDiskMark
Inateck FE2010Inateck UA1001StarTech.com S251BU31C3CBSatechi B01FWT2N3K	Inateck UA1001StarTech.com S251BU31C3CBSatechi B01FWT2N3KInateck FE2010

Moving on to the real-world benchmarks, we first look at the results from our custom robocopy test. In this test, we transfer three folders with the following characteristics.

• Photos: 15.6 GB collection of 4320 photos (RAW as well as JPEGs) in 61 sub-folders
• Videos: 16.1 GB collection of 244 videos (MP4 as well as MOVs) in 6 sub-folders
• BR: 10.7 GB Blu-ray folder structure of the IDT Benchmark Blu-ray (the same that we use in our robocopy tests for NAS systems)

The test starts off with the Photos folder in a RAM drive in the testbed. robocopy is used with default arguments to mirror it onto the storage drive under test. The content on the RAM drive is then deleted. robocopy is again used to transfer the content, but, from the storage drive under test to the RAM drive. The first segment gives the write speed, while the second one gives the read speed for the storage device. The segments end with the purge of the contents from the storage device. This process is repeated thrice and the average of all the runs is recorded as the performance number. The same procedure is adopted for the Videos and the BR folders. The resuts roughly track what we observed in the CrystalDiskMark benchmark numbers. Readers interested in looking at all the graphs in one shot can choose the 'Expand All' option in the dropdown menu.

);Photos ReadPhotos WriteVideos ReadVideos WriteBlu-ray Folder ReadBlu-ray Folder WriteExpand All

High-performance external storage devices can also be used for editing multimedia files directly off the unit. They can also be used as OS-to-go boot drives. Evaluation of this aspect is done using PCMark 8's storage bench. The storage workload involves games as well as multimedia editing applications. The command line version allows us to cherry-pick storage traces to run on a target drive. We chose the following traces.

• Adobe Photoshop (Light)
• Adobe Photoshop (Heavy)
• Adobe After Effects
• Adobe Illustrator

Usually, PC Mark 8 reports time to complete the trace, but the detailed log report has the read and write bandwidth figures which we present in our performance tables. Note that the bandwidth number reported in the results don't involve idle time compression. Results might appear low, but that is part of the workload characteristic. Note that the same CPU is being used for all configurations. Therefore, comparing the numbers for each trace should be possible across different DAS units. Readers interested in looking at all the graphs in one shot can choose the 'Expand All' option in the dropdown menu.

);Adobe Photoshop Light ReadAdobe Photoshop Heavy ReadAdobe After Effects ReadAdobe Illustrator ReadAdobe Photoshop Light WriteAdobe Photoshop Heavy WriteAdobe After Effects WriteAdobe Illustrator WriteExpand All

Thermal Aspects and Power Consumption

The thermal design of external storage enclosures has now come into focus, as high-speed SSDs and interfaces such as USB 3.1 Gen 2 can easily drive up temperatures. This aspect is an important one, as the last thing that users want to see when copying over, say, 100 GB of data to the drive inside the enclosure, is the transfer rate going to USB 2.0 speeds. In order to identify the effectiveness with which the enclosure can take away heat from the internal drive, we instrumented our robocopy DAS benchmark suite to record various parameters while the robocopy process took place in the background. Internal temperatures can only be gathered for enclosures that support S.M.A.R.T passthrough. Readers can click on the graphs below to view the full-sized version. The UA1001 is an open-air design and relies just on the thermal design of the drive itself. However, the FE2010 puts the target drive in an enclosed space. In our evaluation, the temperature of the drive inside the FE2010 ended up just a little bit above the UA1001. However, neither bridge has problems with thermal throttling of the installed drive.

Storage Enclosure Thermal Characteristics
Inateck FE2010Inateck UA1001StarTech.com S251BU31C3CBSatechi B01FWT2N3K	Inateck UA1001StarTech.com S251BU31C3CBSatechi B01FWT2N3KInateck FE2010

It is challenging to isolate the power consumption of the storage bridge alone while treating the unit as a black box. In order to study this aspect in a comparative manner, we use the same SSDs (Curcial MX200 500GB) in the units and process the same workloads on them (CrystalDiskMark 5.1.2's benchmark traces with a region size of 8GB and the number of repetitions set to 5). Plugable's USBC-TKEY power delivery sniffer was placed between the host PC and the storage bridge to record the power consumption. The average power consumption for each access trace was recorded. The pictures below present the numbers in a compact and easy to compare manner.

Power Consumption - CrystalDiskMark
Inateck FE2010Inateck UA1001StarTech.com S251BU31C3CBSatechi B01FWT2N3K	Inateck UA1001StarTech.com S251BU31C3CBSatechi B01FWT2N3KInateck FE2010

Any difference in power consumption for the same access trace between two different units is down to the storage bridge itself (since the drive used is the same in all cases). The UASP-enabled JMS578 consumes more power during traffic compared to the ASM1153E for most access traces. Only the 4K random accesses seems to stress the UA1001 more.

Miscellaneous Aspects and Concluding Remarks

Storage bridges that support UASP fully can translate the SCSI UNMAP command to TRIM commands for SSDs connected to the downstream port. Checking for TRIM support has been a bit tricky so far. CyberShadow's trimcheck is a quick tool to get the status of TRIM support. However, it presents a couple of challenges: it sometimes returns INDETERMINATE after processing, and, in case TRIM comes back as NOT WORKING or not kicked in yet, it is not clear whether the blame lies with the OS / file system or the storage controller / bridge chip or the SSD itself. In order to get a clear idea, our TRIM check routine adopts the following strategy:

1. Format the SSD in NTFS
2. Load the trimcheck program into it and execute
3. Use the PowerShell command Optimize-Volume -DriveLetter Z -ReTrim -Verbose (assuming that the drive connected to the storage bridge is mounted with the drive letter Z)
4. Re-execute trimcheck to determine status report

Conclusions can be made based on the results from the last two steps.

The FE2010 with the JMicron JMS578 bridge chip supports TRIM passthrough, as proved above.

On the other hand, the UA1001 with the ASM1153E bridge chip doesn't support TRIM passthrough - Windows reports that the optimization operation is not permitted by the hardware backing the volume.

Coming to the business end of the review, we find that the Inateck FE2010 and UA1001 target different use-cases. The UA1001 is the more versatile of the two (with support for a wide variety of SATA devices), but loses out to the FE2010 when it comes to features targeting SSDs (such as UASP support and TRIM passthrough). Both of them are reasonably priced - the UA1001 is currently $22 on Amazon, while the FE2010 is just $15. The UA1001 is priced higher due to the bundled power adapter.

The FE2010 can be recommended for its performance and features targeting SSDs. However, we did have some trouble with the initial firmware on the unit. While running CrystalDiskMark's access traces at higher queue depths, the unit would just freeze up. This could be reproduced only when the FE2010 was connected to a USB port with the Intel Alpine Ridge controller behind it. Though I was unable to reproduce the issue with any other USB port (even on the same PC), the problem had been reported by at least one other customer with a different board. Inateck's customer support delivered a firmware update after being able to reproduce the issue at their end. The new firmware managed to help the FE2010 complete our full benchmark routine without hanging. However, the unit still had uncomfortably long pauses in certain write segments of our robocopy benchmarks (drive activity pegged at 100%, but no data transfer at all in the Task Manager's Performance tab for the disk). The pause is reflected in the flat lines in the performance consistency graphs presented in the previous sub-section (the zero transfer rate is reflected in the comparison graphs for the average transfer rate, but not in the performance consistency graph itself). We have not had this issue with any other storage bridge using the same drive - so, our recommendation for the FE2010 is not without reservation. We strongly believe that the FE2010 firmware needs further improvement.

On the other hand, we have no hesitation in recommending the UA1001 despite its need for careful handing. The design, size and choice of build material makes it quite fragile. However, its versatility needs appreciation - it is a must-have gadget for all consumers who have to deal with multiple SATA devices and PCs in an efficient manner.

Cooler Master is a very familiar brand name amongst enthusiasts, with a virtually endless lineup of products ranging from PC cases and coolers to peripherals and accessories. In this review we are having a look at one of Cooler Master’s latest and most popular mechanical keyboards, the Master Keys Pro L White. 

Samsung was among the first television manufacturers to use quantum dot technology for its products, and this week the company has continued that trend by introducing the industry’s first curved monitors for gamers featuring quantum dots. The new 24” and 27” displays boast a wider color gamut and a very high contrast ratio, in addition to support for AMD’s FreeSync technology and a 144 Hz refresh rate.

Samsung’s CFG70 monitors are based on the company’s curved 8-bit VA panels with 1 ms moving picture response time (MPRT), as well as a 144 Hz refresh rate as well as the static contrast ratio to 3000:1, which is higher compared to many advanced displays on the market. As for brightness, the 24” CFG70 offers 350 nits, which is in line with other contemporary high-end displays. At present, Samsung does not share a lot of details about its CFG70 monitors, but we do know that the screens have 1800R curvature along with 178° viewing angles.

The purpose of using quantum dot technology is to expand color gamut of the display by increasing intensity of red and green wavelengths (you can read more here). Samsung confirms that in the case of the CFG70, quantum dots help to produce more accurate dark reds and greens, which increases the contrast. In addition, Samsung says that the use of quantum dot technology for the backlighting of the CFG70 has allowed the company to expand the color gamut to 125% of the sRGB color space.

However despite the monitors' larger-than-sRGB gamut, at this point Samsung is only confirming that the monitors' firmware supports the sRGB color space, listing nothing about the AdobeRGB or DCI-P3 color spaces. This is an important consideration due to the fact that it potentially limits the usefulness of having a color gamut over 100% of sRGB to begin with. With most monitor designs, manufacturers who offer larger gamuts also support color spaces that can use that gamut, which doesn't appear to be the case for Samsung.

The catch then is that having a gamut that exceeds a particular standard will not produce an accurate result when using that wider gamut. Case in point, sRGB content would end up oversaturated since it's meant for a smaller gamut, and the lack of support for larger color spaces makes it difficult to use the wider gamut with anything else. It is hard to believe that Samsung decided to develop a set of monitors that would produce incorrect colors out-of-the-box, so hopefully Samsung is offering a true 100% sRGB mode as well. This would negate the wider gamut of the monitors, but it would ensure their accuracy.

Moving on, the improved contrast ratio should give you the idea about the advantages that the quantum dot technology brings to the CFG70.

Besides being the first curved gaming monitors with quantum dots, the Samsung CFG70 are also among the first displays to use AMD’s FreeSync over HDMI technology (for maximum compatibility, the devices are also equipped with DisplayPort inputs). While we do not know the exact dynamic refresh ranges supported by the screens, 144 Hz maximum refresh rate implies on relatively wide dynamic ranges, more than enough to allow for AMD's low framerate compensation to work.

The CFG70 displays currently do not have direct rivals: there are not of a lot of monitors that use quantum dots right now, and when it comes to gaming monitors with high refresh rates, Samsung is the first maker to use the tech. Moreover, the CFG70 are the first curved displays to feature quantum dots.

Samsung said that the CFG70 displays will be available worldwide, but did not mention their prices. Given the unique combination of characteristics, it is reasonable to expect Samsung to try to capitalize on it.

Gallery: Samsung CFG70: Curved 144 Hz Displays with Quantum Dot and AMD FreeSync

Seagate has formally introduced a new family of hard drives in the 2.5” form-factor. It is designed for laptops as well as external storage solutions. The new BarraCuda HDDs are based on 1 TB shingled magnetic recording platters and Seagate’s multi-tier caching technology. They enhance the maximum capacity of the company’s 2.5” HDDs to 5 TB - making the BarraCuda ST5000LM000 the world’s highest-capacity 2.5” hard drive.

The new Seagate BarraCuda 2.5” drives resemble the company’s Mobile HDDs introduced earlier this year and use a similar set of technologies: motors with 5400 RPM spindle speed, platters based on shingled magnetic recording technology with over 1300 Gb/in² areal density, and multi-tier caching. The 3 TB, 4 TB and 5 TB BarraCuda 2.5” HDDs that come with a 15 mm z-height are designed for external storage solutions because virtually no laptop can accommodate drives of that thickness. Meanwhile, the 7 mm z-height drives (500 GB, 1 TB and 2 TB) are aimed at mainstream laptops and SFF desktops that need a lot of storage space.

All the new BarraCuda 2.5” HDDs feature 128 MB of DRAM cache as well as multi-tier caching (MTC) technology, which is designed to hide peculiarities of SMR. Hard drives featuring shingled recording write new magnetic tracks that overlap part of the previously written tracks. This may slow down the writing process since the architecture requires HDDs to rewrite adjacent tracks after any writing operation. To “conceal” such peculiarities, Seagate does a number of tricks. Firstly, it organizes SMR tracks into bands in a bid to limit the amount of overwriting. Secondly, the MTC technology uses several bands of PMR tracks on the platters, around 1 GB of NAND flash cache as well as DRAM cache. When workloads generate relatively small amount of writes, the HDD writes data to NAND and/or to the PMR tracks at a predictable data rate. Then, during light workloads or idle time, the HDD transfers written data from the caches to SMR tracks, as described by Mark Re (CTO of Seagate) earlier this year.

To a large extent, Seagate’s multi-tier cache determines the performance of the company’s BarraCuda 2.5" and Mobile HDD drives. For example, when Seagate announced its Mobile HDD products earlier this year, the company declared maximum sustainable transfer rate of the HDD at 100 MB/s. However, the documentation was altered later in the summer to reflect a maximum transfer rate of 140 MB/s (possibly due to updated firmware, or a change in the performance measurement method). Now, the high-capacity BarraCuda 2.5” (3 TB, 4 TB and 5 TB) drives are rated for 130 MB/s, while the mainstream BarraCuda 2.5” (500 GB, 1 TB and 2 TB) are rated for 140 MB/s. This is still below the 145 – 169 MB/s offered by PMR-based Laptop HDDs from the company.

Seagate’s new BarraCuda 2.5”/7 mm HDDs for laptops should hit notebooks in the coming weeks as the company is already shipping them to its customers. Seagate does not currently offer SED and FIPS options for its BarraCuda family (those who need encryption can still get Seagate's Mobile HDDs based on SMR technology). The company plans to add these technologies to the BarraCuda arsenal in the future.

The BarraCuda 2.5”/15 mm drives for external storage solutions will be used inside Seagate’s own DAS devices such as the Backup Plus Portable Drive and the Expansion Portable Hard Drive products. The Seagate Backup Plus Plus Portable Drive 5 TB (STDR5000100) is due in early November and will cost around $150 - $160. This device will be the world's highest-capacity portable DAS and will not immediately have direct rivals in this particular form-factor. At present, the maximum-capacity 2.5” HDD offered by Western Digital is 4 TB with a 12.5 mm z-height. It is used inside a number of My Passport drives.

All the BarraCuda 2.5” drives are backed by a two-year warranty.

Seagate has officially introduced its fifth-generation solid-state hybrid drives (SSHDs) under the FireCuda brand name. They have a thinner form-factor compared to previous-gen solutions. The drives use Seagate’s 1 TB SMR platters as well as the company’s multi-tier cache technology. The FireCuda 2.5” are the first SSHDs to use shingled magnetic recording (SMR).

The Seagate FireCuda 2.5” family will offer 500 GB, 1 TB and 2 TB capacities and will be compatible with both desktops and thin laptops, thanks to its 7 mm z-height (down from 9.5 mm in case of the Laptop SSHD drives). To a large extent, the Seagate FireCuda SSHDs resemble the recently announced BarraCuda 2.5” HDDs: the drives are also based on the 1 TB SMR platters which have an areal density of more than 1.3 Tb/in² (note that the areal density of the FireCuda products is higher compared to that of BarraCuda devices). They feature 5400 RPM spindle speed and come with 128 MB of DRAM cache buffer. The maximum transfer rate and average latency for the BarraCuda 2.5” and FireCuda 2.5” are the same: 140 MB/s and 5.6 ms.

The differentiating aspect of the FireCuda SSHDs is their 8 GB NAND buffer. It caches data from frequently used sectors to enable faster boot times and shorter load times for certain applications. Therefore, when it comes to real-world performance, FireCuda 2.5” SSHDs should provide better performance compared to the BarraCuda 2.5” HDDs. However, it is unknown how the FireCuda 2.5” drive stack up against the Laptop SSHDs based on PMR platters. Seagate has not revealed whether the new FireCuda SSHDs use a new caching algorithm compared to the previous-gen hybrid hard drives, but this is certainly a possibility given the use of SMR platters. As before, the company says that the algorithm is continuously trying to optimize performance of FireCuda SSHDs.

Seagate has started to ship the FireCuda drives to its customers, but has not listed official prices or market availability dates. The latter depends on retailers and/or PC makers. Seagate tells us that the FireCuda 2.5" 2 TB will cost around $100, but the supply/demand situation is bound to affect that pricing. Meanwhile, Amazon offers the FireCuda 2.5" 1 TB for $70. All of the previous-gen Laptop SSHDs ended up in retail, so, it is a question of time before the whole FireCuda 2.5" family will be up for grabs. One of the advantages of Seagate’s hybrid drives is their five-year warranty, which is longer than that of typical HDDs.

Samsung this week announced its first LPDDR4 memory chips made using its 10nm-class DRAM fabrication technology. The new DRAM ICs feature the industry’s highest density of 16 Gb, are rated to run at 4266 MT/s data rate, and open the door to more mobile devices with 8 GB of DRAM.

Earlier this year Samsung started to produce DDR4 memory using its 10nm-class DRAM manufacturing process (which is believed to be 18 nm) and recently the firm began to use it to make LPDDR4 memory devices, just as it planned. The thinner fabrication technology allowed Samsung to increase capacity of a single LPDDR4 DRAM IC to 16 Gb (up from 12 Gb at 20nm introduced in August, 2015) while retaining a 4266 MT/s transfer rate.

The first product to use the 16 Gb ICs is Samsung’s 8 GB LPDDR4-4266 mobile DRAM package for smartphones, tablets, and other applications that can use LPDDR4. The device stacks four memory ICs and provides up to 34 GB/s of bandwidth when connected to an SoC using a 64-bit memory bus. The 8 GB DRAM package comes in a standard 15 mm x 15 mm x 1 mm form-factor, which is compatible with typical mobile devices, but Samsung can also make the package thinner than 1 mm to enable PoP stacking with a mobile application processor or a UFS NAND storage device.

Samsung has not revealed a lot about the cost efficiency or power consumption of the 16 Gb LPDDR4 ICs, nor have they discussed those details for the 8 GB LPDDR4 package either. What little Samsung has said is that the latter consumes approximately the same amount of power as a 4 GB LPDDR4-3200 device (four 8 Gb ICs) made using its 20 nm-class process technology. Taken at face value, one can extrapolate that the switch to the 10nm-class fabrication process allowed Samsung to double the capacity and increase performance by 33% at the same power. Unfortunately, we do not know anything about the geometry scaling of the new ICs relative to Samsung's older ICs, so it's hard to even guess how much Samsung's newest DDR4 costs to fab.

Samsung has not officially commented on when it plans to start commercial shipments of its 8 GB LPDDR4 packages, but it is reasonable to assume that the company will commence sales of such devices in the coming months, with actual products hitting the market in 2017.

Following suit as this fall's major games hit the (virtual) store shelves, NVIDIA has packaged another bundle of Game Ready updates for the latest releases. In the 375.57 WHQL update we get a few fixes, new game support, and some VR as well.

This time we have the first version out for driver release 375. In the list of fixes we start with Mirrors Edge Catalyst, which is no longer flicking in the Ansel UI when the image is moved with the mouse pointer on systems running a SLI GTX 980M setup. There is no longer get corruption from decals in Overwatch and there are also no longer issues changing the refresh rate on BenQ ZOWIE monitors, were before we saw cases of system crashes and signal loss to the monitor when changing the refresh rate. Lastly there is a fix for an issue with the GeForce GTX 650 causing dots on all the characters in GTA V.

As a new feature in this release NVIDIA improved the performance and quality of the "motion vectors provided by the Motion-Estimation-Only mode of the video encoder, specifically in stereo VR use cases". Though a bit of a mouthful for a description, this enhancement seems to apply to improving NVIDIA's recording quality with stereo VR games.

Moving on to the most interesting reason for updating our drivers, the Game Ready updates start with Battlefield 1, which came out this week. We can also find support for Sid Meier’s Civilization VI which was released in lockstep next to Battlefield 1. Titanfall 2 made the roster for this update though we won’t be seeing that game's release until next Friday on the 28th. We also have a new section in the release notes labeled Game Ready VR, which means that VR experiences join in on the party starting with the classic series Serious Sam in Serious Sam VR: The Last Hope and the incoming Eagle Flight VR From Ubisoft. We have also been given a new SLI profile for the MMORPG Lineage Eternal: Twilight Resistance.

Anyone interested can download the updated drivers through GeForce Experience or on the NVIDIA driver download page. More information on this update and further issues can be found in the 375.57 release notes.

PowerColor has introduced its first enclosure for external graphics cards, the Devil Box. The accessory enables gamers to connect desktop video cards to AIO, SFF or laptop PCs using the Thunderbolt 3 interface and comes with its own power supply. The Devil Box will be a limited edition product compatible with select graphics cards (due to driver and physical limitations) and, like other external GPU chassis we've seen thus far, is priced as a premium product. Nonetheless, for the time being, this box will be among a few TB3-based external enclosures capable of running desktop GPUs.

The PowerColor Devil Box supports both NVIDIA and AMD (XConnect) GPUs, and resembles Razer’s Core eGFX enclosure introduced earlier this year. The two boxes have slightly different dimensions, but both can accommodate a qualified double-wide PCIe x16 video card (which will operate in PCIe x4 mode) that is up to 12.2”/310 mm long. The design of the Devil Box (as well as the Core) allows operation of graphics adapters with different cooling systems (blower, open air), except hybrid (e.g., Radeon Fury X, Radeon R9 295X2) due to space constraints. To feed the GPUs, the Devil Box incorporates a 500 W PSU and the maximum GPU power draw is rated at 375 W (exactly the specs of the Core).

Next up is connectivity and this is where PowerColor’s Devil Box has an edge over Razer’s Core. Both enclosures support additional USB 3.0 receptacles to connect peripherals and a GbE port to enable high-speed wired Internet on ultra-thin laptops that do not feature GbE. However, the product from PowerColor can also host a 2.5” HDD or SSD (I suspect with the help of a USB-to-SATA bridge, to simplify the process), thus expanding storage capabilities of the host system.

When it comes to compatibility, PowerColor lists the latest AMD Radeon RX 400-series as well as NVIDIA’s GeForce GTX 10-series graphics adapters in addition to numerous previous-gen products. We cannot verify whether all of the GPUs listed by the manufacturer support plug’n’play properly, but they are guaranteed to work inside the Devil Box.

While the whole external GPU idea seems very plausible because all-in-one, small form-factor and mobile PCs are gaining popularity among gamers, eGFX chassis are still not mainstream. This is not exactly surprising given the fact that the eGFX hardware (Thunderbolt 3 with v16 or newer firmware) and software (Windows 10 with updates, drivers) were finalized only months ago and far not all PCs can properly support external graphics adapters. As a result, being aware of limited demand (because far not all TB3-enabled laptops are eGFX-certified), Power Color naturally does not want to produce a lot of Devil Boxes, which is why the enclosures will be available from select retailers and in select countries (see the table).

A good news is that PowerColor’s Devil Box will be more affordable than Razer’s Core and will sell for $379 or €419, depending on the market. The price of the enclosure is still rather high, but none the less a good improvement over where the Razer Core launched earlier this year.

Gallery: PowerColor Announces Devil Box: Thunderbolt 3 External GPU Enclosure

Despite the convenience of all-in-one liquid cooling solutions, hardcore enthusiasts still prefer customizable “open loop” systems, even with their overall complexity and cost. In this review we are testing our first “open” liquid cooling system, the Alphacool NexXxoS Cool Answer 360 DDC/XT kit, providing insight on the advantages, difficulties, performance and cost of advanced liquid cooling solutions beyond simple air and all-in-one liquid coolers.

Kicking off a busy day for the Internet of Things market, Intel this morning is announcing the Atom E3900 series. Based upon the company’s latest generation Goldmont Atom CPU core, the E3900 series will be Intel’s most serious and dedicated project yet for the IoT market.

We’ve talked about Intel’s IoT efforts off and on over the past couple of years. Having largely missed the boat on mobile, the company decided it wouldn’t miss IoT as well, and as a result they’ve been making significant investments into the IoT market, treating it as a fourth pillar of their business. Their efforts have not gone unrewarded, as IoT revenue has continued to quickly grow over the years, and as of Q3’16, IoT is now Intel’s third largest business by revenue, ahead of non-volatile memory and behind only client and data center revenue. Furthermore with 19% year-over-year growth, IoT still looks like a growing market for the company.

Overall the company offers a number of IoT products, ranging from the tiny Curie to the relatively powerful Atom E-series. However to date, the company has never built a dedicated, high-end IoT die. The Atom E3800 series were Bay Trail, and the recently announced Joule devices use Broxton, Intel’s canceled-for-mobile Atom SoC. However today that will be changing, as the launch of the Atom E3900 series brings with it Intel’s first custom silicon targeting the roughly 6W to 12W market of more powerful IoT devices.

So what does an IoT-centric Atom look like? By and large, it’s Broxton and more. At its core we’re looking at 2 or 4 Goldmont CPU cores, paired with 12 or 18 EU configurations of Intel’s Gen9 iGPU. However this is where the similarities stop. Once we get past the CPU and GPU, Intel has added new features specifically for IoT in some areas, and in other areas they’ve gone and reworked the design entirely to meet specific physical and technical needs of the IoT market.

The big changes here are focused on security, determinism, and networking. Security is self-evident: Intel’s customers need to be able to build devices that will go out into the field and be hardened against attackers. Bits and pieces of this are inerieted from Intel’s existing Trusted Execution Technology, while other pieces, such as boot time measuring, are new. The latter is particularly interesting, as Intel is measuring the boot time of a system as a canary for if it’s been compromised. If the boot time suddenly and unexpectedly changes, then there’s a good chance the firmware and/or OS has been replaced.

Meanwhile also new to the E3900 series is what Intel is calling their Time Coordinated Computing Technology.  This high precision timing mechanism is to allow multiple Atom E3900s to be tightly synchronized, down to 1 microsecond. We’re told that this feature is of particular interest to Intel’s manufacturing partners, as a means to improve accuracy and synchronization between devices on a manufacturing line.

Finally, a bit more nebulous, the E3900 die also includes some changes specifically to improve determinism. What Intel is pitching here isn’t hard determinism, but a higher level of determinism for devices that need better guarantees about how soon an action will be completed. Apparently a big part of implementing this is at the cache level, with Intel noting that polling loops in particular were greatly impacted by this change.

All told, Intel will be offering 3 SKUs of the E3900, ranging from 6.5W to 12W. As relatively high power processors these aren’t meant for wearables and such, but rather primarily devices on mains power where additional intelligence is needed. In Intel terminology, the E3900 is focused on “edge” devices as opposed to “core” devices. The idea being that Intel wants to move out data processing to the edge of an IoT network – into sensors and such devices – as opposed to having to use a dumb sensor that sends data back for processing.

Intel’s 3 big markets here are the video/sensor, industrial, and automotive market. The first and last in particular are areas that the previous E3800 couldn’t readily compete in, due to a lack of processing power for image processing and video encoding. Thanks in big part to the Gen9 GPU and some E3900-specific Image Processing Unit (IPU) changes – the chip can support 15 1080p30 video inputs – Intel can now go after these markets. And that may be the biggest part of this story here for Intel: they haven’t had a part like the E3900 before. The Bay Trail based E3800 had a decent enough CPU, but it’s the E3900 where GPU computing and computer vision become viable, and this is the cornerstone of a lot of new functionality.

Of course, hardware is only part of the picture. Along with the E3900 itself, Intel will also be shipping a number of software libraries to help developers take better advantage of the hardware, and really, bootstrap the whole process. A good deal of this is on the image processing side, providing functions that tap into the hardware’s new image processing capabilities.

Finally, in the first-half of next year, the E3900 series will be joined by the A3900 series. This is an automotive-specific SKU that is rated for higher operational temperatures; 110C versus 85C for the E3900. As we mentioned before automotive is a big part of Intel’s efforts here, and that means matching automotive tolerances. Given the performance of these chips, we don’t get the impression that Intel’s entering the fully autonomous car market right now, but they are hoping to go after some lower hanging fruit with driver assistance and in-car entertainment systems.

Xiaomi has added two new phones to its lineup: the Mi Note 2 and the Mi MIX. Both are flagship phones with large screens that emphasize performance and design; however, despite some overlap in criteria and internal hardware, these are two very distinct devices.

Like its predecessor, the Mi Note, the Mi Note 2 has an aluminum frame and a glass back with curved edges, but instead of the previous design’s flat front, the Mi Note 2 uses a flexible OLED display to allow curved edges to the front too, design features Samsung’s recent Galaxy phones also employ. Another obvious change from the previous model is the addition of a capacitive fingerprint sensor. The pill-shaped sensor, which sits below the screen and is flanked by capacitive navigation buttons, keeps the lower bezel relatively slim, giving the Mi Note 2 a screen-to-body ratio of 77.2%. The rear-facing camera sits flush with the shiny and smooth glass on the back and includes a slightly raised and polished ring to protect the lens.

The design is generally symmetric and balanced: The centered earpiece has the sensors and front-facing camera to either side, and the centered USB Type-C port on the bottom has matching arrays of holes for the downward-firing speaker and microphone. The offset 3.5mm headphone jack on the top edge and the dual-color LED flash on the back are the only features to break symmetry.

The Mi Note 2’s 5.7-inch OLED display delivers excellent black levels and covers 110% of the NTSC color gamut, according to Xiaomi, which basically means it should deliver highly saturated colors. For people who prefer more accurate colors, the phone provides an sRGB mode and a control for adjusting the display’s color temperature. Unfortunately, its 1920x1080 resolution is less than ideal for such a large display.

Inside the Mi Note 2 is Qualcomm’s latest Snapdragon 821 SoC, which includes four of its custom Kryo CPU cores and an Adreno 530 GPU, and a sufficiently large 4070mAh non-accessible battery that supports Qualcomm’s Quick Charge 3.0 fast charging technology. It also comes with either 4GB of RAM and 64GB of internal UFS 2.0 storage or 6GB of RAM and 128GB of storage, but there’s no support for microSD cards.

The 8MP front-facing camera includes autofocus and Xiaomi’s third-generation beauty mode processing. Around back is a 22.5MP Sony IMX318 Exmor RS sensor, which is among a new generation of sensors that use smaller 1.0µm pixels. Despite the reduction in full-well capacity, Sony claims its new sensor matches the image quality of its previous IMX230 Exmor RS sensor that uses larger 1.12µm pixels. Stacked above the sensor is a lens array with six elements and an f/2.0 aperture.

The Mi Note 2’s rear camera supports phase detect autofocus (PDAF) to improve focus speed. There’s no optical image stabilization (OIS), which could negatively impact low-light still image performance, but the camera does employ 3-axis electronic image stabilization (EIS) when shooting videos, even in 4K.

The Mi Note 2 comes in two different colors—Piano Black and Glacier Silver—and three different versions—two which will only be available in China and a global version that supports 6 modes and 37 frequency bands. The Mi Note 2 will be available in China starting November 1st for ¥2799 (4GB RAM / 64GB NAND) or ¥3299 (6GB RAM / 128GB NAND). The global version will cost ¥3499 (6GB RAM / 128GB NAND).

The larger Mi MIX was jointly developed with prolific designer Philippe Starck and features an all-ceramic body with ceramic buttons. The front is dominated by a 6.4-inch display with virtually no bezels along the top or sides, giving the phone a 91.3% screen-to-body ratio. The 5MP front-facing camera is located in the lower-right corner of the slim lower bezel, while the ultrasonic (instead of infrared) proximity sensor is located behind the screen. There are no physical or capacitive buttons in the lower bezel either, because it uses onscreen navigation controls.

Eliminating the upper bezel meant eliminating the traditional earpiece speaker too. In its place, the Mi MIX uses a piezoelectric-driven cantilevered beam that produces sound by vibrating the phone’s ceramic frame instead of a speaker driver. What’s not clear, though, is if people nearby will also be able to hear the conversation, or if the sound is directed only towards the user.

A circular camera sits flush with the ceramic back, stacked above a circular fingerprint sensor. A USB Type-C port sits centered on the bottom edge flanked by a symmetric array of small holes hiding a downward firing speaker and microphone. A 3.5mm headphone jack is on the top edge.

Its 6.4-inch IPS LCD display has an odd 17:9 aspect ratio that’s supposed to make room for the onscreen navigation controls while still providing a traditional 16:9 viewable area. Its 2040x1080 resolution seems low for such a large display, but it still provides a decent 362 PPI pixel density. Xiaomi claims a peak brightness of 500 nits and 94% coverage of the NTSC color gamut, making this a wide-gamut panel with more saturated colors. Like the Mi Note 2, the Mi MIX also provides a control for adjusting the display’s color temperature and a native sRGB mode.

Its internal hardware is also similar to the Mi Note 2’s, packing in an impressive list of hardware. The key differences are a larger 4400mAh battery and twice the internal UFS 2.0 NAND.

The Mi MIX will be available exclusively in China beginning November 4th and will come in two different versions. The standard version will cost ¥3499 with 4GB of RAM and 128GB of internal storage, while a special limited edition version, which comes with 6GB of RAM, 256GB of internal storage, and 18K gold-plated accents around the rear camera and fingerprint sensor will cost ¥3999.

We're readying up for a Live Blog of the Microsoft event in NYC. Panos Panay is expected to take to the stage, with Major Nelson from the Xbox team in tow (we believe) for some new announcements. The presentation is set to start at 10am ET, and we'll fire up the Live Blog as it starts!

As part of the now annual Microsoft Surface event, Panos Panay announced the next member of the Surface family, the Surface Studio. The Studio is ultimately a prosumer all-in-one device promising more functionality and versatility than any other desktop all-in-one PC by allowing the device to also turn a desk into a studio.

Front and center in what makes the Studio impressive is the size of the display: a 28-inch thin-bezel LCD display with a 3:2 aspect ratio, coming in at a 4500x3000 resolution and 192 pixels per inch. By contrast to 4K, this is 13.5 million pixels compared to 8.3 million in UHD, and Microsoft is promoting True Scale with the studio such that two A4 pieces of paper can be rendered side by side at full resolution and at a higher DPI than most standard office printers. The display is 12.5mm thin, with Microsoft redesigning the LCD stack to ensure a slim profile.

The display connects to the base via a specialist hinge, featuring 80 machined parts on each side for what Microsoft calls a ‘Zero Gravity Hinge’. This allows the display to be moved seamlessly and for any plausible angle, as well as taking on extra weight in studio mode. The display has two buttons on the right-hand side for power and volume. On the top of the display is the Windows Hello-enabled camera, with a 5.0 MP element capable of 1080p video (we assume 30 FPS). The Studio supports the Surface Pen, which can attach to the side of the display.

For color reproduction, Microsoft is advertising the display as supporting both DCI-P3 and sRGB with a simple toggle on the Windows sidebar to switch between the two. While Microsoft says that the displays are calibrated for both, this has fundamental issues with color reproduction.

In the base is a set of arguably last-generation specifications: 6^th generation (Skylake) Intel Core i5 and Core i7 processor options (probably 65W desktop parts?) paired with up to 32GB of DDR4 memory (probably DDR4-2133). This comes with a NVIDIA GTX 965M 2GB for two of the three options, and a GTX 980M 4GB on the high-end model. Connectivity comes via USB 3.0, rather than USB 3.1/Thunderbolt. Storage is labeled as ‘1TB or 2TB Rapid Hybrid Drive’ options, which in the presentation looked like an M.2 drive but as yet it has been unstated if this is SATA or PCIe (or if a Rapid Hybrid Drive actually means an SSHD).

Connectivity comes via four USB 3.0 ports, a full-size SD card reader, a mini DisplayPort output and a 3.5mm headset jack. WiFi is provided by an 802.11ac unit, although Microsoft does not say which one (I’d hazard a guess and say Intel’s AC8260 2x2 solution). The unit also supports Xbox Wireless, allowing for Xbox controllers to also be connected for gaming.

The whole unit weighs in at 21 lbs (9.5 kg), and Microsoft has stated that it will be available only in limited quantities during Q4, with the official release date as 15^th December. Current configurations available will be:

$2999 : Intel Core i5 (Skylake), 8 GB DDR4, 1TB, GTX 965M 2GB
$3499 : Intel Core i7 (Skylake), 16 GB DDR4, 1TB, GTX 965M 2GB
$4199 : Intel Core i7 (Skylake), 32 GB DDR4, 2TB, GTX 980M 4GB

Windows 10 Pro is included with a 30-day Office trial.

Gallery: Microsoft Announces the Surface Studio: 28-inch AIO with Touch, Pen, 4500x3000, Haswell, GTX 980M

Edit: Originally this piece was posted with the incorrect Intel Generation code name in the title. It should read 'Skylake', not 'Haswell'. The piece has been edited to clarify.

Seagate on Tuesday announced its new generation of hard drives with 15K spindle speed and up to 900 GB capacity. The new HDDs use NAND caching to boost response times, offer additional SED and SED-FIPS security features at no extra cost and boast with enhanced reliability for 24x7 workloads.

The new Seagate Enterprise Performance 15K v6 HDDs come in 300, 600 and 900 GB configurations. Just like their predecessors, they use a dual port SAS 12 Gbps interface as well as a 2.5”/15 mm form-factor. The new Enterprise Performance 15K v6 hard drives have single or dual level caches: either a 256 MB of DRAM cache only or a 256 MB of DRAM cache and a 16 GB of NAND flash cache (4Kn/512E models only). The latter is used for caching of frequently used “hot” data to maximize read performance and reduce latencies (Seagate offers similar capability with its consumer-grade FireCuda and other SSHDs). It is noteworthy that Seagate reduced the amount of NAND flash compared to previous-generation 15K HDDs, but the company seems to believe that its improved caching algorithms will ensure that the new drives are faster than their predecessors.

When compared to predecessors, the new sixth-generation 15K HDDs are rated for a 27% increase in sequential read write speed: up to 315 MB/s. In addition, the drives also promise 100% faster random write performance. Seagate claims that its advanced caching algorithms promote hot data three times faster when compared to 512N drives without NAND, but it does not reveal exact performance numbers for its NAND cache. When it comes to reliability, Seagate declares 2 million hours MTBF and an annualized failure rate of 0.44%, which is in line with the predecessors.

To simplify inventory management, Seagate will offer Seagate Enterprise Performance 15K v6 hard drives with its FastFormat feature, which allows formatting the drives to either 512 emulation or 4K native formats, thus reducing the number of models the manufacturer has to produce and its resellers have to keep in stock.

Nowadays 15K HDDs are no longer the fastest storage devices, even in RAID environments - modern high-end enterprise-class SSDs usually offer considerably higher random and sequential performance. Nonetheless, 10K and 15K hard drives for mission-critical applications are still used by multiple datacenters worldwide. Moreover, some say that there are deployments that use both mission-critical SSDs as well as fast mission-critical HDDs, which suggests that the new drives will have some demand. In fact, Seagate claims that the total available market for such HDDs is around six million units per quarter, but admits that sales of 15K drives have been declining recently. Still, the company itself sells roughly three million of such drives every three months. Given the very long life cycles of mission-critical HDDs, it looks like development of this 15K generation makes sense for Seagate.

Exact pricing of Seagate’s sixth-generation Enterprise Performance 15K HDDs will depend on volumes and contracts. Interested parties should get in contact directly with their local Seagate distribution partners.

Gallery: Seagate Introduces New Generation of Enterprise Performance 15K HDDs with NAND Caching

EIZO has introduced two new displays designed for specialized and business environments that require multi-monitor setups. The new FlexScan EV2451 and EV2456 are equipped with ultra-thin bezels for their standout feature, and are IPS panels with 178° viewing angles as well as four display inputs to maximize compatibility.

The FlexScan EV2451 and EV2456 monitors have generally typical specifications for today’s office displays: 1920×1080 and 1920×1200 resolution (respectively), 250 and 350 nits brightness (respectively), a 60 Hz refresh rate and a rated 1000:1 contrast ratio. The monitors can use DisplayPort, HDMI, DVI-D or D-Sub to connect to host PCs such that they can be used with new and legacy computers. In addition, they support EIZO’s EcoView Optimizer 2 feature, which dynamically adjusts backlight brightness in accordance with environment brightness to reduce power consumption.

The key features of the two monitors are their ultra-thin bezel sizes: 1 mm on the sides and top as well as 4.6 mm on the bottom (keep in mind that that the monitors also have black borders between bezels and screens, hence, we cannot call them completely borderless). Such thin bezels mean that EIZO is aiming these displays for multi-monitor setups used in trading and control rooms. Moreover, such bezels make the FlexScan EV2451 and EV2456 monitors plausible candidates for gamers, who might use multiple monitors to play titles like flight simulators.

To maximize flexibility for multi-display environments, both monitors also use new stands with height-adjustment range of 158.6 mm (EV2456) and 172.7 mm (EV2451) as well as 40° (5° down, 35° up) tilt, 344° swivel, and 90° pivot for viewing in portrait mode. The stands can be easily removed using their quick-release buttons, and the screens can be installed onto a VESA-compatible mounting device. Finally, EIZO also supplies special Screen InStyle software that lets users adjust the color, circadian dimming and power settings on multiple FlexScan displays at once. This includes flicker free modes, blue-light reduction modes, and paper-reading modes.

EIZO plans to start shipments of its new monitors shortly, their prices will naturally vary by country. The predecessor of the FlexScan EV2456 (the EV2455) has an MSRP of $569 in the U.S., whereas the predecessor of the FlexScan EV2451 (the EV2450) costs $399 in the U.S. As for warranty, the FlexScan monitors come with a five-year warranty with a six-month zero bright sub-pixel guarantee.

Sometimes there is no substitute for performance. Most of the laptop market is focusing on thin and light designs, with companies attempting to outdo each other by shaving a millimeter or two off of their laptop z-height compared to the competition. But in the Desktop Replacement (DTR) category, there are no such concessions. Clevo is one of the few laptop makers that is in the DTR market, and thanks to the assistance of Mythlogic, we have the Mythlogic Phobos 8716 DTR for review today. As a Clevo, the model would be P870DM2.

We're here ready to go for another set of Apple Announcements!

Today LG and Apple made a joint product announcement of sorts. Apple announced their new MacBook Pro, which uses four Thunderbolt 3 ports for IO. During the launch event, Apple also noted that LG would be launching a new pair of monitors to go along with the MacBook Pro. These new monitors are branded as LG UltraFine, and LG is advertising them as monitors designed to seamlessly integrate with macOS and both new models of the MacBook Pro. I've included the known specifications for both monitors below:

The first LG UltraFine display is a 21.5" 4K model. Unlike most "4K" displays which really use the 3840x2160 resolution defined in the UHD spec, this is a 4096 x 2304 display that actually has over four thousand pixels on its longer axis. The 27" model is a 5K panel with a resolution of 5120 x 2880. Both monitors connect over a single cord. In the case of the 4K model it can be driven using a single DisplayPort 1.2 stream over a normal USB-C port. The 5K monitor requires two DisplayPort 1.2 streams and so it is a tiled display. However, using Thunderbolt 3 for the link avoids the dual-cable solution used on existing 5K monitors. These solutions also allow for data and power transfer, and so the monitors can actually charge a connected laptop while also providing three 5Gbps USB-C ports and stereo speakers.

The panel characteristics for both displays are similar. Both have a peak brightness of 500 nits, and they use the P3 color gamut, although it's almost certain that they retain the standard 2.2 gamma and white point of D65 from existing displays. The one key difference to note is that the 21.5" model is only an 8bit display, with support for 16.7 million colors, while the 27" model is a 10bit display supporting 1.07 billion colors. The 27" model also has a camera and a built in microphone for audio and video calls.

At this point you may have noticed that all of these specs are the same as Apple's iMacs. LG Display makes the IPS displays in the iMacs, so it's very likely if not certain that these are the same panels. Given that these monitors are targeted at users of the new MacBook Pro, it's also very likely that they do not have any sort of sRGB fallback mode, as such an option is unnecessary on macOS where there is end to end color management in any AppKit app and even in most apps that don't use AppKit. This is important to keep in mind for anyone interested in buying these for a Windows computer. The state of color management on Windows means that in nearly every application you'll be stuck with oversaturated colors and no workaround, so I wouldn't recommend it even if it can work.

While I'm on the topic of Windows support, I will say that from a technical perspective, the displays should work with a Windows computer. However, product listings for the monitor specify that you need a Mac with USB-C for the 4K model, and a Mac with Thunderbolt 3 for the 5K model. This limits the 4K model to Apple's MacBook, and the new MacBook Pros, with the 5K model only working on the latter. This also ties back to the integration with macOS. Most monitors use physical buttons for controlling brightness and volume, but the LG UltraFine displays use the integrated controls that macOS provides, meaning that it can respond to the volume and brightness keys on Mac keyboards. It doesn't look like the monitors have any buttons to fall back on, so this may be another situation where Windows users should be wary, and it definitely explains why the monitors are being marketed specifically for their integration with macOS.

Both of LG's new UltraFine monitors have already shown up online, although you can't buy them directly from Apple's online store as of right now even if you'd like to purchase one alongside a new MacBook Pro. The listings do tell us that the 4K model is priced at $699.95 in the US, and the 5K model is $1299.95. According to LG, the 4K model will be available next month, and the 5K model will launch in early December.

Much ink has been spilt over the last year on the subject of the MacBook Pro. In short, Apple hasn’t given its professional laptop lineup a real overhaul in some time. Only the smaller 13” model ever got Intel’s 5^th gen Broadwell CPU – the 15” model is even further back, on 4^th gen Haswell – and otherwise the laptops haven’t had major feature updates since 2013. As a result, Apple has been due as anyone can be to update their laptops. And at a special event held at their campus this morning, they have done just that, launching the 4^th generation of the MacBook Pro.

For better or worse the new 4^th generation family doesn’t have a catchy name like Retina to help set it apart from the prior models. But make no mistake, Apple has been busy on giving their languishing laptops a much-needed facelift.

All-told, Apple has introduced 3 new MacBook Pro models: a basic 13” version, and then two full-featured versions at 13” and 15” respectively. All three models are form the same mold, so to speak. All of them get the same new unibody design, the same wide color gamut (P3) display, and the same upgraded internal hardware composed of Intel’s 6^th generation Skylake CPUs, Thunderbolt 3, and new SSDs. What separates the basic 13” from the other two models, besides cost, is that it doesn’t include Apple’s new multi-touch Touch Bar (more on this in a sec).

If you’ve been watching Apple over the years, then you know that the company has a (sometimes unhealthy) obsession with reducing their products’ thickness, and the 4^th gen MacBook Pro follows that obsession to a T. The new family of laptops is thinner than ever, with Apple taking them from 18mm to 14.9mm and 15.5mm for the 13” and 15” models respectively. Altogether, the total volume is down 23% for the 13” model and 20% for the 15” model.

Apple is still using their now tried-and-true aluminum unibody design for the family, so if you’re familiar with the current color-coated MacBook, then you know what to expect here. Though the event was very brief on how Apple was able to shave off so much volume compared to the last generation, it sounds like they’ve gone with a thinner cooling system, and in the case of the 15” model, AMD’s low-profile Polaris 11 GPU.

Under the hood, this change in volume comes without a significant change in official TDPs (so all factors held equal, Apple would be removing the same amount of heat in a thinner design). Apple has finally made the jump across the entire family to Intel’s 14nm Skylake processors. The basic 13” model uses a 15W CPU, the higher-end 13” model uses a 28W CPU, and the 15” model uses a 45W CPU, the latter two being the same TDPs as the last generation. Apple isn’t talking up CPU performance too much here, but given how long they’ve been on Haswell on the 15” model in particular, we should see a good performance bump.

Meanwhile, in a change from the 3^rd generation design, all 15” 4^th gen MacBook Pros come with discrete Radeon Pro graphics. Apple has tapped AMD’s new Polaris 11 GPU, in part for the low z-height it offers, which replaces the 4 year old Cape Verde GPU in the previous MacBook Pro. Apple is offering 3 dGPU configurations at increasingly higher performance. From what AMD has told me, these chips are all in the ballpark of 35W. Relative to the previous dGPU models – or even more, the previous model that only offered an integrated Intel GPU – GPU performance should be greatly improved over the last generation. Over 2x for the top-tier Radeon Pro 460 is a very reasonable assumption right now, though one that testing will need to confirm.

On the SSD side of matters, Apple has quickly talked up the fact that the new MacBook Pros use a new SSD. Apple traditionally uses multiple vendors, but the quoted 3GB/sec read speeds are consistent with the latest generation of drives out of Samsung and Toshiba. So is the fact that Apple is now offering SSDs up to 2TB in size.

More visible is the change to the display. Apple hasn’t increased their resolutions at all – so these are the same 2880x1800 & 2560x1600 resolution configurations at before – but they have upgraded what these displays can do. All of the MacBook Pros now support the wider P3 color gamut that Apple has been championing for the last year and already offers on the iMac, iPhone 7, and iPad Pro. Meaning that the new laptops now support a wider color gamut and can display material mastered for P3 (or photos taken from the iPhone 7) at their full gamut. The new displays are also said to be quite a bit brighter than before; Apple is throwing around 500 nits, or 67% brighter.

And, perhaps most interestingly, Apple has confirmed that the new displays support “variable refresh rate” technology. I need to dig into this more, but if this is an implementation of AMD’s FreeSync tech or something similar, then it would potentially be a nice improvement in perceived screen smoothness.  Apple’s also touting that the new display consumes 30% less power, but it’s not clear how much of that is due to variable (e.g. lower) refresh rates and how much of that is from panel and backlighting improvements.

Apple has also reworked the keyboard and trackpad, in part to integrate the latest technologies and to achieve their thinness goals. The new MacBook Pros utilize a newer revision of the butterfly switch first introduced on the MacBook keyboard. This means we’re talking about a shorter travel distance than the prior generation MacBook Pros – something that may not be popular in all corners – but at the same time we’re told that these aren’t the same identical switches as on the MacBook, and that there have been some changes to improve the feel. Meanwhile the Force Touch trackpad has been enlarged; it now goes from the bottom edge of the laptop to the keyboard, the full area that Apple could hope to give it.  This is a 100% area increase for the 15” model, and a 46% increase for the 13” model.

Apple has also overhauled their input/output options on the new MacBook Pros; quite radically in fact. Apple has doubled-down on Thunderbolt 3/USB Type-C. All three models contain solely either 2 (13” basic) or 4 (all other models) TB3 ports (which also double as USB-C 3.1 Gen 2 ports), and then a 3.5mm audio jack. This means the MagSafe port is gone, as are the USB Type-A ports and the Thunderbolt 2 ports. In this sense it’s very close to what has happened with the MacBook, except the MacBook Pro gets a larger number of ports. TB3 is enabled through the use of Intel’s Alpine Ridge controller; the 2 port versions have a single controller, and the 4 port versions have 2 controllers (since Intel doesn’t make a 4 port version). Furthermore all 2 or 4 ports can be used for charging or display outputs as necessary, so they are all equals in functionality.

Last, but certainly not least however, is perhaps the marquee feature of the 4^th generation MacBook Pros: the Touch Bar. A bit of a poorly kept secret in the run-up to today’s launch, the Touch Bar is a (presumably OLED) capacitive multi-touch display that replaces the row previously occupied by the function keys.  The idea, in a nutshell, is to make a dynamic row of virtual buttons that can be whatever an application (or user) needs them to be, bringing with it all of the advantages of multi-touch that Apple has developed over the years for their iOS devices. Note that the Touch Bar is only available on the higher-end MacBook Pros; the basic 13” model forgoes the Touch Bar in favor of a traditional row of function keys, presumably for cost reasons.

Rounding out the feature set, the Touch Bar also includes a Touch ID fingerprint sensor on its right side, and the necessary controller – what Apple calls the T1 – to handle Touch ID and the Secure Enclave duties. This sensor means that virtually everything that can be done with Touch ID on iOS devices can be extended to macOS: one-touch login, Apple Pay confirmation, etc. You can even use it to do things iOS can’t do, such as swap between users. Taking a page from those iOS devices, the Touch ID sensor is the one physical button that’s part of the Touch Bar; it is both a fingerprint scanner and the device’s physical power button.

Judging from the presentation alone, the Touch Bar is going to be divisive among users. Replacing physical buttons with a capacitive strip is a major change; if you’re used to feeling buttons, there’s nothing here to feel. On the other hand, Apple is convinced that the function keys are underutilized right now, and that space can be put to better use. Even with the track pad already offering Force Touch, the Touch Bar has the advantage of being able to display what button it is, and behave separately from the trackpad. As someone who admittedly is still old school enough to configure the function keys to work as actual function keys – and not the screen/sound/playback controls Apple defaults to – I’m definitely going to have to sit down with the new Touch Bar to get an idea of what the transition is like. It’s a bold idea for Apple; but bold ideas can be both good and bad.

Wrapping things up, the new MacBook Pros are available for order today. The basic 13” model (the one without the Touch Bar and only 2 TB3 ports) will retail for $1499 and is shipping immediately. The other two models will be at $1799 and $2399 for the 13” and 15” models respectively, and will not be shipping for another 2-3 weeks.

Finally, Apple has also illustrated how the overall MacBook family of laptops is organized now that they’ve added the new MacBook Pros. The 11” MacBook Air has been removed from sale entirely, while the 13” model is still for sale, but hasn’t received any update. Given that the new MacBook Pros are thinner than the MacBook Air 13”, it seems likely that it’s going to eventually be phased out entirely. Consequently the new, modern MacBook lineup consists of 4 laptops: the 12” MacBook, the basic 13” MacBook Pro, the advanced 13” MacBook Pro, and the 15” MacBook Pro.