The ongoing Chia coin craze has already increased prices of high-capacity HDDs and SSDs in retail quite significantly. Since demand for storage devices is not going to get any lower, Taiwan-based makers of SSDs are developing special-purpose SSDs for Chia mining that they plan to sell directly to miners. Furthermore, drive makers expect Chia to consume considerable 3D NAND production capacities in the coming months.
Adata, Apacer, Phison Electronics, and TeamGroup all reported significant increases of SSD orders in April when compared to March, according to DigiTimes. Adata said that demand for its SSDs increased by 400% – 500% sequentially last month, TeamGroup also saw an impressive increase and expressed optimism about the short-term Chia farming demand for SSDs. Other makers tend to agree that demand for SSDs will remain strong in the coming months.
In a bid to meet demand and offer farmers the components they need to make the best Chia plotting PC builds, Adata, Phison, and TeamGroup have started developing appropriate SSDs (or at least have created task forces).
Since the majority of SSD makers have enterprise-grade drives designed for write-intensive workloads in their lineups, developing high-endurance Chia-optimized storage solutions should not pose a significant challenge for their engineering teams. Essentially, they will have to develop high-endurance SSDs based on consumer-grade 3D NAND (with loads of over-provisioning space) that do not have enterprise features. TeamGroup has already announced its first SSDs for Chia mining and expects to ship them in high volumes in Q3 2021.
What is interesting is that Adata, Phison, and TeamGroup are looking at ways to ship their Chia farming SSDs directly to mining farms, so these drives may not end up in retail (just like we do not see many high-end HDDs and SSDs in retail). To some degree, this will ensure that large Chia farming operations will not get their drives from retail and inflate prices, which is good news. There is bad news too.
High-capacity high-endurance SSDs for Chia mining will naturally increase demand for 3D NAND memory in general. Phison already told DigiTimes that Chia SSDs would “consume considerably all the available capacity” at 3D NAND makers. To that end, demand for both 3D NAND and SSD controller are set to increase, which will cause price hikes.
While Intel’s revenue dropped in the first quarter, it is still the world’s largest supplier of chips by revenue, ahead of Samsung and TSMC, according to IC Insights. Intel’s arch-rival AMD significantly increased its sales in Q1 2021 and is now one of the world’s Top 15 semiconductor companies.
Demand for personal computers and chips grew significantly in the recent quarters, so it is not surprising that the sales of the top 15 semiconductor companies increased by 21% year-over-year in Q1 2021. Intel leads the market with $18.676 billion, yet its semiconductor sales were 4% lower than in the first quarter last year. Samsung followed Intel with $16.152 billion, whereas TSMC was No. 3 with $12.911 billion.
Samsung, which is the world’s largest supplier of 3D NAND and DRAM memory, has been challenging Intel for the top position on the market of chips for years. At times, Samsung can dethrone Intel when memory prices are on the rise, but it was not the case in Q1 2021.
(Image credit: IC Insights)
AMD and MediaTek were two new entrants to the top 15 list of semiconductor companies. AMD significantly increased shipments of its CPUs for servers and high-performance desktops, so its revenue in the first quarter totalled $3.445 billion, up 93% year-over-year. In Q1 2021 AMD was ranked 11th largest semiconductor company in the world. The company was ranked 18 in the first quarter 2020. MediaTek’s sales reached $3.849 billion, an increase of 90% year-over-year. AMD and MediaTek replaced Huawei’s HiSilicon and Sony’s chip division in Top 15.
The list of top 15 semiconductor sales leaders includes CPU, GPU, and SoC developers, makers of memory, suppliers of special-purpose chips, telecom giants, and contract makers of semiconductors. The list includes eight companies headquartered in the U.S., two firms from South Korea, Taiwan and Europe, and one from Japan.
Phison’s PS5018-E18 is a high-performance PCIe 4.0×4 NVMe SSD controller that has flooded the market in many of the newest and fastest-performing SSDs. It offers up very fast sequential performance but has been outshined in random responsiveness by top picks like Samsung’s 980 Pro and WD_Black SN850 due in part to Micron’s B27B 96-Layer TLC flash holding it back. Today, we’re taking a deep look at the company’s next iteration that gets a little help from Micron’s fastest flash yet.
While the first generation of E18-based NVMe SSDs used flash that interfaced with the controller flash at somewhat restricted speeds of 1,200 MTps, our second-gen sample is no longer shackled by this bottleneck. Now, featuring Micron’s latest B47R 176-layer TLC flash operating at speeds of up to 1,600 MTps, our new sample offers much more competitive performance. Still, this is only a preview, and while SSDs based on this flash due to hit the market soon, they will only hit the market after a few more firmware revisions.
Specifications
Product
500GB
1TB
2TB
Pricing
$99.99
$159.99
$319.99
Capacity (User / Raw)
500GB / 512GB
1000GB / 1024GB
2000GB / 2048GB
Form Factor
M.2 2280
M.2 2280
M.2 2280
Interface / Protocol
PCIe 4.0 x4 / NVMe 1.4
PCIe 4.0 x4 / NVMe 1.4
PCIe 4.0 x4 / NVMe 1.4
Controller
Phison PS5018-E18
Phison PS5018-E18
Phison PS5018-E18
DRAM
DDR4
DDR4
DDR4
Memory
Micron 176L TLC
Micron 176L TLC
Micron 176L TLC
Sequential Read
6,500 MBps
7,000 MBps
7,000 MBps
Sequential Write
2,850 MBps
5,500 MBps
6,850 MBps
Random Read
170,000 IOPS
350,000 IOPS
650,000 IOPS
Random Write
600,000 IOPS
700,000 IOPS
700,000 IOPS
Security
AES 256-bit encryption
AES 256-bit encryption
AES 256-bit encryption
Phison’s PS5018-E18 supports both TLC and QLC flash and can address capacities of up to 8TB, but most retail products will ship in the popular 1TB and 2TB capacities. In terms of performance specifications, Phison’s E18 hasn’t changed much besides the faster 1,600 MTps flash transfer rate. Sequential speeds are rated for up to 7.4/7.0 GBps read/write and in terms of random performance, and the company claims the SSD controller is capable of 1 million random read/write IOPS, given the proper flash and tuning.
Phison seems to be moving away from full dynamic SLC caching in the traditional sense. Instead, its SLC caching algorithms are adapting to better suit not only consumer workloads but heavy prosumer workloads, too. We will cover this more in-depth later in the article.
Phison didn’t reveal anything about end-product endurance ratings with this new flash, but it may help improve endurance ratings over current-gen devices. The Phison E18 still features the company’s fourth-generation LDPC ECC and RAID ECC along with a DDR ECC engine and end-to-end data path protection to ensure your data is programmed and read reliably over the product’s useful lifespan. It also supports various security options such as Pyrite, AES 256, SHA 512, RSA 4096, and TCG Opal. Additionally, it comes with S.M.A.R.T. data reporting and Trim support and secure erase and crypto erase capability.
A Closer Look
The E18 interfaces with the host over a PCIe 4.0 x4 link and is NVM 1.4 compliant. Our 2TB sample comes in an M.2 2280 double-sided form factor, but smaller capacities come in single-sided form factors for broad compatibility with the latest ultra-thin mobile devices. This is in contrast to Samsung and WD M.2 SSDs, all of which come only in single-sided form factors.
Image 1 of 4
(Image credit: Tom’s Hardware)
Image 2 of 4
(Image credit: Tom’s Hardware)
Image 3 of 4
(Image credit: Tom’s Hardware)
Image 4 of 4
(Image credit: Tom’s Hardware)
The PS5018-E18 controller is an eight-channel, DRAM-based design. Our 2TB sample features two 1GB SK hynix DDR4 ICs, one on each side of the PCB. It leverages a Penta-core design with three single-core Cortex R5 CPUs handling the host/read/write tasks while an additional R5 CPU, in a lower-clocked dual-core configuration, acts as a co-processor. The E18’s primary cores are clocked much higher than the older E16’s cores, at 1 GHz versus 733 MHz.
The E18 comes with CoXPorcessor 2.0 technology, which offloads some of the firmware code from the primary cores to the dual-core R5 to ensure responsive QoS when hammered with heavy sustained workloads and aid power efficiency. One of the co-processor cores optimizes NAND die-queues to the flash while the other is for managing the DRAM/NAND tables.
Image 1 of 2
(Image credit: Tom’s Hardware)
Image 2 of 2
(Image credit: Tom’s Hardware)
The controller is manufactured on TSMC’s 12nm process technology and comes with multiple features to keep thermals under control while operating at high speeds. It features ASPM and ASPT support, can transition into the L1.2 low power state to sip just 3mW at idle, and can thermal throttle to preserve data integrity over performance.
Phison sent over our sample with a sleek-looking heat sink, but this heat sink does not necessarily indicate that retail models will require heat sinks to keep them cool under most consumer workloads. Although, when this controller is paired with 2TB of flash, it can gulp down over 8 watts of power under sustained load, which can create quite a bit of heat.
(Image credit: Tom’s Hardware)
Speaking of the flash, our 2TB sample comes with thirty-two 512 Gb dies of Micron’s new B47R 176L TLC, aligning with the controller’s native chip enable capability to optimize interleaving, and thus, performance. Micron’s 176L TLC offers a significant improvement over previous generations, and not just due to its high layer count.
This flash features the company’s new replacement-gate architecture that combines charge traps with CMOS-under array (CuA) technology, allowing for roughly a 30% smaller die size than its competitors. Enabled through multiple advancements in the new architecture and firmware support, operating interface speeds clock in at 1,600 MTps, which improves read and write speed by roughly 35% compared to the company’s previous-generation floating-gate 96L TLC.
Image 1 of 3
(Image credit: Tom’s Hardware)
Image 2 of 3
(Image credit: Tom’s Hardware)
Image 3 of 3
(Image credit: Tom’s Hardware)
Architecturally, Micron replaced the polysilicon control gates with metal, and it uses a different etching method compared to traditional NAND. This reduces resistance, thus allowing the program pulse to ramp up quickly, reducing programming complexities and overhead. The design reduces the electric field duration on the circuits since they can be programmed much more quickly. Micron also increased the etch diameter, allowing for more structural stability as the company ramps layer counts in future generations of the flash.
Additionally, in traditional NAND, cell-to-cell capacitive coupling issues limit performance, but by utilizing a nonconductive layer of silicon nitride (SiN) acting as a NAND storage cell, the replacement-gate design demonstrates almost no capacitance between cells.
These changes, along with other cell geometry adjustments, directly improve performance, endurance, conserve energy, and allow for increased storage capacity, especially as Micron continues to develop newer iterations of its flash.
If you’ve been following the SSD space at least a little bit, you definitely know Phison. An SSD controller vendor that started with low-performance budget chips, Phison has improved tremendously since and is now offering solutions that rival the fastest controllers available. Phison was founded in 2000 in Taiwan and is shipping hundreds of millions of controllers each year.
The Phison E18 controller is the company’s first PCI-Express Gen 4 controller. It has been used on drives like the Corsair MP600 Pro, Sabrent Rocket 4 Plus, and Addlink S95. In our MP600 Pro review, we found that Corsair’s fastest SSD delivers performance comparable to the Samsung 980 Pro and WD Black SN850.
All these Phison E18-based drives have one thing in common: They use 96-layer 3D TLC NAND flash from Micron, also known as B27B. These flash chips were released in 2018, so they aren’t exactly based on the latest technology. Especially compared to Samsung, this aging flash tech puts the Phison E18+Micron B27B combination at a small performance disadvantage. That’s why Phison has now qualified Micron’s 176-layer B47R TLC flash chips for the E18 controller. Technically, the controller is the same physical silicon as all changes are done in firmware, which helps keep manufacturing cost down.
Today, we are previewing the performance of this Phison E18+Micron 176-layer B47R flash by using a pre-production sample SSD provided directly by Phison. The drive uses not only B47R, but B47R Fortis Flash, which is the enterprise version of regular B47R. While rated for much higher endurance, it should be very similar otherwise.
Phison hasn’t indicated any pricing because it’s ultimately the SSD vendor’s decision. I have no knowledge of how expensive B47R is compared to B27B, but I doubt it’ll be cheaper. The controller itself shouldn’t be significantly different in pricing because it is the same silicon with firmware changes.
High demand for PCs and, importantly, desktop PCs has already resulted in increased graphics card, PSU, HDD, and SSD pricing, but apparently, now DRAM is getting even more expensive, too. According to a new report from Taiwan, memory prices are set for double-digit growth in the coming months.
Contract memory prices are projected to increase by 10% – 20%, DigiTimes reports (based on a Seeking Alpha story). According to the report, memory prices may rise by up to 25%, depending on the kit. Keeping in mind that demand is high and DRAM is transiting to DDR5 memory, the increase is not surprising.
Most enthusiasts who buy premium memory kits shouldn’t be particularly worried — high-end memory kits already carry a hefty price tag. Hence, a 10% increase in DRAM IC prices isn’t too extreme. However, every penny counts when it comes to mainstream desktop and laptop PCs, so this is where that 10% – 25% contract price hike gains importance. In fact, even a 2% to 5% increase in the bill-of-materials (BOM) can increase the price of a laptop by a significant sum.
DRAM quotes are not alone when it comes to price hikes, according to the report. Prices of NAND memory have also increased by 10% so far this year, and given the current Chia Coin farming craze, it’s easy to expect that trend to continue.
Gigabyte’s latest Aorus SSD promises to eliminate thermal throttling and deliver extremely fast read speeds of up to 7GBps. The drive is called the Aorus 7000s Prem., and it comes with a gigantic black heatsink measuring up to 1.76 inches (44.7mm) in height, to ensure the drive stays cool under long-duration reads and writes.
Image 1 of 2
(Image credit: Gigabyte)
Image 2 of 2
(Image credit: Gigabyte)
The 7000s Prem. is basically identical in most ways to the standard Aorus 7000s solid-state drive that we found competitive with the best SSDs when we reviewed it. Both feature a second-generation Phison E18 8 channel controller, AES-256 encryption support, 3D TLC NAND, and a DDR4 DRAM cache. But with the 7000s Prem., you’re getting that big beefy heatsink to improve performance.
The heatsink itself is called the M.2 Thermal Guard Xtreme and features a dual-heatpipe design, a nanocarbon coating, and an aluminum M.2 baseplate to help cool the bottom part of the drive. All in all, this heatsink is just as large as some aftermarket M.2 heatsinks you can add onto your current M.2 SSDs, so performance should be excellent.
(Image credit: Gigabyte)
With Gigabyte’s internal testing, the Aorus 7000s Prem. performed at its max read speed of up to 7GBps (7000MBps) consistently for over 8 hours of time, with no signs of slowing down. But just remember that this is Gigabyte’s internal testing and results could change if your chassis can’t supply enough airflow to the heatsink or you live in a hot ambient environment.
The biggest consideration when purchasing this SSD will be its size; many motherboards these days won’t have the necessary headroom for the SSD and its associated heatsink–at least not on all the M.2 slots. Many M.2 slots are located either behind the motherboard or right underneath your graphics card, so make sure you check the location of your PCIe 4.0 M.2 slot to see if you have enough clearance for the heatsink.
The 7000s Prem. will come in 1TB and 2TB flavors. Both capacities feature the same 7GB/s read speeds, however, the write speeds change from 5500MB/s on the 1TB to 6850MB/s on the 2TB model. Pricing and availability for the 7000 Prem. is unknown at this time. But given that the standard model sells for about $380, sans that honking heatsink, it’s a safe bet you’ll be looking at spending close to $500 for those precious sustained speeds.
Chipmakers might finally be cashing in on the global chip shortage. Times-Taipei News reported last week that since the second quarter of 2021, “more than 30 semiconductor companies have issued price adjustment letters, with product price increases ranging from 10% to 30%,” according to a translated version of the report.
Times-Taipei News said UMC, SMIC, and Power Semiconductor Manufacturing Co. are among the companies adjusting their prices. The reasons cited for those adjustments vary, but Rockchip reportedly said, “the cost of wafers, printed circuit boards, and packaging and testing have risen sharply to varying degrees.”
The price adjustments seem to vary by chip type. The report claimed that signal chain ICs saw price increases between 10% and 20%, for example, while the price of power management ICs was said to have risen between 10% and 30%. Some memory chips were also said to have risen in price by as much as 20%.
Those already steep increases might actually be tame compared to other adjustments. Times-Taipei News said that “the price of some IC products has soared dozens of times at present, which has caused difficulties in the production and operation of small and medium-sized enterprises,” per an “industry insider.”
There is at least one major exception to this trend: TSMC. According to the report, the company hasn’t issued any price increases, which means the world’s largest chipmaker is holding the line. Or at least it’s presenting that facade—its decision to end price discounts will have the same effect on the cost of its chips.
The effects of these price increases will probably still be felt throughout the industry, even without TSMC. Power management chips are nearly ubiquitous, for example, which means a wide variety of products could see their prices rise as well. That effect could also be compounded in products that rely on multiple types of chips.
DigiTimes also reported today that DRAM and NAND prices are expected to rise through the third quarter due to strong demand across various segments, panic buying spurred by the Chia cryptocurrency’s ascendancy, and limited supply of NAND flash device controllers leading to higher prices for those chips.
The report said that DRAM memory contract pricing rose “over 20%” in the second quarter, with “prices for PC DRAM chips hiking more than 25%,” per anonymous sources. DRAM contract prices are expected to rise an additional 10% to 20% in the third quarter, while NAND contract prices could rise by as much as 10%.
AMD CEO Lisa Su has made it clear she doesn’t think people should be too worried about the chip shortage. But at least in the short term, it’s going to be hard to find many products, from the PlayStation 5 to the latest graphics cards, and now it seems many of the devices that do manage to make it to consumers might cost more.
Adata SE900G is a speedy USB 3.2 Gen 2×2 portable SSD that features a dazzling RGB design but lacks the features and warranty support that we see with other SSDs.
For
+ Competitive performance
+ Stylish RGB lighting
+ Competitive pricing
Against
– 3-year warranty
– Lacks AES 256-bit encryption
– Lacks IP rating
– RGB lights aren’t controllable
Features and Specifications
RGB has made its way into everything these days: fans, cases, PSUs, motherboards, GPUs, RAM, and even M.2 and SATA internal SSDs. Now we can add portable SSDs to the list, too. Adata’s new SE900G is a speedy portable USB 3.2 Gen 2×2 SSD that is also quite the show stopper with its stylish and hypnotizing RGB shine.
The SE900G also comes with plenty of horsepower under the hood in the form of an M.2 SSD that is very similar to
the company’s controversial XPG SX8200 Pro NVMe SSD
. Like the SX8200 Pro, Adata says that the components inside this SSD can change with newer revisions. Adata guarantees that the drive will meet its performance and endurance specifications regardless of the chosen internal componentry.
Paired with a fast USB 20 Gbps interface, the SE900 flies past its 10 Gbps competition and comes with reasonable pricing, making it a great value for those looking for a flashy portable SSD.
Specifications
Product
SE900G 500GB
SE900G 1TB
SE900G 2TB
Pricing
$99.99
$159.99
$289.99
Capacity (User / Raw)
512GB / 512GB
1024GB / 1024GB
2048GB / 2048GB
Interface / Protocol
USB-C / USB 3.2 Gen 2×2
USB-C / USB 3.2 Gen 2×2
USB-C / USB 3.2 Gen 2×2
Included
USB Type-C & USB Type-C to USB Type-A
USB Type-C & USB Type-C to USB Type-A
USB Type-C & USB Type-C to USB Type-A
Sequential Read
2,000 MBps
2,000 MBps
2,000 MBps
Sequential Write
2,000 MBps
2,000 MBps
2,000 MBps
Interface Controller
ASMedia ASM2364
ASMedia ASM2364
ASMedia ASM2364
NAND Controller
SM2262EN
SM2262EN
SM2262EN
DRAM
DDR4
DDR4
DDR4
Storage Media
Micron 96L TLC
Micron 96L TLC
Micron 96L TLC
Power
Bus-powered
Bus-powered
Bus-powered
Dimensions (L x W x H)
110.8 x 66 x 16.5mm
110.8 x 66 x 16.5mm
110.8 x 66 x 16.5mm
Weight
160g
160g
160g
Part Number
ASE900G-512GU32G2-CBK
ASE900G-1TU32G2-CBK
ASE900G-2TU32G2-CBK
Warranty
3-Years
3-Years
3-Years
Adata’s SE900G is available in three capacities of 500GB, 1TB, and 2TB at reasonable pricing, given its sequential performance ratings of up to 2,000 MBps of read/write throughput.
Our 2TB sample carries the lowest price-per-GB of the lineup. At $0.14 per gigabyte, it undercuts many USB 20Gbps SSDs on the market and even USB 10Gbps SSDs like SanDisk’s Extreme v2. Unfortunately, Adata only backs the SE900G with a basic three-year warranty rather than the five-year warranty we typically see with most enthusiast-grade storage.
The SE900G also doesn’t feature an Ingress Protection rating like our best external SSDs, nor does it come with AES 256-bit hardware encryption support to keep your data secure. The SE900G does come with the standard support for S.M.A.R.T. data reporting, UASP, and Trim, though.
Accessories
(Image credit: Tom’s Hardware)
The SE900G comes with two USB cables — one 11.5-inch long USB Type-C cable for newer systems, and another USB Type-C to Type-A for compatibility with older ones.
A Closer Look
Image 1 of 2
(Image credit: Tom’s Hardware)
Image 2 of 2
(Image credit: Tom’s Hardware)
The SE900G is reasonably sized at 110.8 x 66 x 16.5mm, and it also has some heft to it. At 160 grams, the SE900G weighs two to three times more than many competing 5 Gbps and 10 Gbps portable SSDs. However, most of those drives also don’t run as fast as the SE900G and thus don’t need such robust thermal consideration. Adata’s SE900G features a large vented metal, heatsink-like back housing to provide adequate thermal dissipation under heavy use and provide some drop protection for the internals.
Image 1 of 2
(Image credit: Tom’s Hardware)
Image 2 of 2
(Image credit: Tom’s Hardware)
A black and clear plastic panel covers the top of the SSD, allowing the RGB lights to shine through. The RGB lighting also doubles as a power indicator, and there’s also an indicator light next to the USB Type-C port.
A white PCB with eight LEDs resides under the top panel and provides the light show. Unfortunately, you cannot control the lights. This PCB also contains the ASMedia ASM2364 USB 3.2 Gen 2×2 to PCIe 3.0 x4 NVMe bridge chip, M.2 slot, and supporting circuitry.
Image 1 of 2
(Image credit: Tom’s Hardware)
Image 2 of 2
(Image credit: Tom’s Hardware)
At the heart of the SSD is a drive very similar to the XPG SX8200 Pro — this variant is powered by Silicon Motion’s very responsive SM2262EN PCIe 3.0 x4 NVMe 1.3 SSD controller. This controller is a high-end eight-channel, DRAM-based design that leverages dual Arm Cortex R5 CPUs clocked at 625 MHz. This controller interfaces with 2GB of Samsung DDR4 DRAM at a clock speed of 700 MHz.
Image 1 of 3
(Image credit: Tom’s Hardware)
Image 2 of 3
(Image credit: Tom’s Hardware)
Image 3 of 3
(Image credit: Tom’s Hardware)
Our SE900G sample also shipped with Micron’s 512Gb B27A 96L TLC NAND flash. There are 32 dies distributed into the four NAND packages and interface with the controller at speeds of 650 MTps. Each die has a quad-plane architecture, meaning the controller can achieve even higher levels of parallelism than dual-plane flash, which equates to faster performance. It even features subdivisions of tile groupings for faster and more efficient random reads over competing flash, such as BiCS4.
The tiles are redundant latches grouped for small I/O (4K), while BiCS4 has other means such as SBL (shielded-bitline) current sense as opposed to ABL (all bitline) current sense. Also, unlike BiCS4, Micron’s 96L TLC takes advantage of the CuA (Circuitry Under the Array) architecture where the NAND cell arrays are placed on top of the periphery circuitry (decoders, sense amplifiers, timing circuitry, buffers, etc.) to help shrink the die as well as enable the use of the company’s unique tile grouping. Additionally, Micron claims its floating gate design gives it some inherent data retention benefit over competing charge trap flash, too.
It’s noteworthy that while these internal components came inside our review sample, Adata may change the DRAM, NAND, SSD controller, and/or bridge chip at any point in the production cycle for this drive. That means the components can vary over time. Adata guarantees that the drive will meet its performance and endurance specifications regardless of the chosen internal componentry.
Intel is about to release Optane H20 memory for laptop systems. Announced late last year, Intel’s Optane H20 memory aims to deliver the best of RAM and SSD storage in a single solution, accelerating loading times and data transfers.
Available with 512GB and 1TB capacities, Intel Optane H20 memory packs both QLC 3D NAND and Optane technologies into a module featuring an M.2 2280 form factor. Whether you choose a 512GB or a 1TB H20 memory module, both feature 32GB of Optane memory. Those interested in the Intel Optane H20 should know that this memory is only compatible with systems equipped with 11th Gen Intel Core processors.
Rated sequential read and write speeds are similar to a PCIe 3.0 SSD, featuring up to 3400MB/s reading speeds and up to 2100MB/s write speeds. Random 4K speeds vary between 65-390K IOPS while reading and 40-280K IOPS on writing. The drive’s endurance isn’t on par with the best PCIe 3.0/4.0 SSDs, but 185/370TBW should still enough for most users. The rated MTBF is set at about 1.6M hours, and all drives come with a 5-year warranty.
Intel has scheduled the release date of Optane H20 memory for June 20th, but pricing is still unknown.
KitGuru says: Are you thinking about acquiring a new laptop with Intel Optane H20 memory? What type of workloads would you use it for?
Solid-state drives have a number of advantages when compared to hard drives, which include performance, dimensions, and reliability. Yet, for quite a while, HDDs offered a better balance between capacity, performance, and cost, which is why they outsold SSDs in terms of unit sales. Things have certainly changed for client PCs as 60% of new computers sold in Q1 2021 used SSDs instead of HDDs. That said, it’s not surprising that SSDs outsold HDDs almost 3:2 in the first quarter in terms of unit sales as, in 2020, SSDs outsold hard drives (by units not GBs), by 28 perecent.
Unit Sales: SSDs Win 3:2
Three makers of hard drives shipped as many as 64.17 million HDDs in Q1 2021, according to Trendfocus. Meanwhile, less than a dozen SSD suppliers, including those featured in our list of best SSDs, shipped 99.438 million solid-state drives in the first quarter, the same company claims (via StorageNewsletter).
(Image credit: Trendfocus/StorageNewsletter)
Keeping in mind that many modern notebooks cannot accommodate a hard drive (and many desktops are shipped with an SSD by default), it is not particularly surprising that sales of SSDs are high. Furthermore, nowadays users want their PCs to be very responsive and that more or less requires an SSD. All in all, the majority of new PCs use SSDs as boot drives, some are also equipped with hard drives and much fewer use HDDs as boot drives.
Exabyte Sales: HDDs Win 4.5:1
But while many modern PCs do not host a lot of data, NAS, on-prem servers, and cloud datacenters do and this is where high-capacity NAS and nearline HDDs come into play. These hard drives can store up to 18TB of data and an average capacity of a 3.5-inch enterprise/nearline HDD is about 12TB these days nowadays. Thus, HDD sales in terms of exabytes vastly exceed those of SSDs (288.3EB vs 61.5EB).
Meanwhile, it should be noted that the vast majority of datacenters use SSDs for caching and HDDs for bulk storage, so it is impossible to build a datacenter purely based on solid-state storage (3D NAND) or hard drives.
Anyhow, as far as exabytes shipments are concerned, HDDs win. Total capacity of hard drives shipped in the first quarter 2021 was 288.28 EB, whereas SSDs sold in Q1 could store ‘only’ 66 EB s of data.
Since adoption of SSDs both by clients and servers is increasing, dollar sales of solid-state drives are strong too. Research and Markets values SSD market in 2020 at $34.86 billion and forecasts that it will total $80.34 billion by 2026. To put the numbers into context, Gartner estimated sales of HDDs to reach $20.7 billion in 2020 and expected them to grow to $22.6 billion in 2022.
Samsung Leads the Pack
(Image credit: Trendfocus/StorageNewsletter)
When it comes to SSD market frontrunners, Samsung is an indisputable champion both in terms of unit and exabytes shipments. Samsung sold its HDD division to Seagate in 2011, a rather surprising move then. Yet, the rationale behind the move has always been there for the company that is the No. 1 supplier of NAND flash memory. Today, the move looks obvious.
Right now, Samsung challenges other SSD makers both in terms of unit (a 25.3% market share) and exabyte (a 34.3% chunk of the market) shipments. Such results are logical to expect as the company sells loads of drives to PC OEMs, and high-capacity drives to server makers and cloud giants.
Still, not everything is rosy for the SSD market in general and Samsung in particular due to shortage of SSD controllers. The company had to shut down its chip manufacturing facility that produces its SSD and NAND controllers in Austin, Texas, earlier this year, which forced it to consider outsourcing of such components. Potentially, shortage of may affect sales of SSDs by Samsung and other companies.
“Shortages of controllers and other NAND sub-components are causing supply chain uncertainty, putting upwards pressure on ASPs,” said Walt Coon, VP of NAND and Memory Research at Yole Développement. “The recent shutdown of Samsung’s manufacturing facility in Austin, Texas, USA, which manufactures NAND controllers for its SSDs, further amplifies this situation and will likely accelerate the NAND pricing recovery, particularly in the PC SSD and mobile markets, where impacts from the controller shortages are most pronounced.”
Storage Bosses Still Lead the Game
(Image credit: Trendfocus/StorageNewsletter)
Western Digital follows Samsung in terms of SSD units (18.2%) and capacity (15.8%) share to a large degree because it sells loads of drives for applications previously served by HDDs and (perhaps we are speculating here) mission-critical hard drives supplied by Western Digital, HGST (as well as Hitachi and IBM before that).
The number three SSD supplier is Kioxia (formerly Toshiba Memory) with a 13.3% unit market share and a 9.4% exabyte market share, according to TrendFocus. Kioxia has inherited many shipment contracts (particularly in the business/mission-critical space) from Toshiba. Kioxia’s unit shipments (a 13.3% market share) are way lower when compared to those of its partner Western Digital (to some degree because the company is more aimed at the spot 3D NAND and retail SSD markets).
Being aimed primarily at high-capacity server and workstation applications, Intel is the number three SSD supplier in terms of capacity with an 11.5% market share, but when it comes to unit sales, Intel controls only 5% of the market. This situation is not particularly unexpected as Intel has always positioned its storage business as a part of its datacenter platform division, which is why the company has always been focused on high-capacity NAND ICs (unlike its former partner Micron) for advanced server-grade SSDs.
Speaking of Micron, its SSD unit market share is at an 8.4%, whereas its exabytes share is at 7.9%, which is an indicator that the company is balancing between the client and enterprise. SK Hynix also ships quite a lot of consumer drives (an 11.8% market share), but quite some higher-end enterprise-grade SSDs (as its exabytes share is 9.1%).
Seagate is perhaps one exception — among the historical storage bosses — that controls a 0.7% of the exabyte SSD market and only 0.3% of unit shipments. The company serves its loyal clientele and has yet to gain significant share in the SSD market.
Branded Client SSDs
One interesting thing about the SSD market is that while there are loads of consumer-oriented brands that sell flash-powered drives, they do not control a significant part of the market either in terms of units nor in terms of exabytes, according to Trendfocus.
Companies like Kingston, Lite-On, and a number of others make it to the headlines, yet in terms of volume, they control about 18% of the market, a significant, but not a definitive chunk. In terms of exabytes, their share is about 11.3%, which is quite high considering the fact that most of their drives are aimed at client PCs.
Summary
Client storage is going solid state in terms of unit shipments due to performance, dimensions, and power reasons. Datacenters continue to adopt SSDs for caching as well as business and mission-critical applications.
Being the largest supplier of 3D NAND (V-NAND in Samsung’s nomenclature), Samsung continues to be the leading supplier of SSDs both in terms of volumes and in terms of capacity shipments. Meanwhile, shortage of SSD controllers may have an impact on the company’s SSD sales.
Based on current trends, SSDs are set to continue taking unit market share from HDDs. Yet hard drives are not set to give up bulk storage.
Blending Optane memory and QLC flash, Intel’s Optane Memory H20 is an innovative M.2 NVMe SSD that delivers a unique caching experience. It is ultra-responsive to most consumer workloads, especially repetitive tasks.
For
+ Optane caching improves system responsiveness
+ Rivals high-end NVMe SSDs in light and mixed workloads
Against
– Low endurance
– Optane caching not beneficial in all workloads
– Limited to specific systems and 1TB maximum capacity
– Slow sustained write performance after the SLC cache fills
– Lacks AES 256-bit encryption support
Features and Specifications
Not too long ago, Intel killed off most of its client Optane products to focus on one — the Optane Memory H20. The H20 marries the company’s latest QLC flash with Optane Memory to provide fast performance for most client workloads, but it can fall short in large sequential transfers. It’s a clever blend of innovation and technology, bringing improvements over the Optane Memory H10, but it still isn’t quite something that has fully won us over for day-to-day use. It also isn’t available for stand-alone purchases at retail, so it doesn’t make our list of Best SSDs.
Data caching isn’t anything new. For years, Intel has accelerated performance through data caching via various implementations. The company even has competition in the space from software vendors like Enmotus. My first experience with this technology started with the company’s Smart Response Technology a decade ago, which allowed you to use an SSD to cache data from an HDD for faster retrieval. But that was just the beginning.
Intel further refined the technology to leverage its very own Optane Memory to offer unparalleled response times compared to traditional SSDs. The company released products like the M10; low-density, fast access Optane SSDs for use with supported systems. Shortly after, Intel progressed to new designs.
Two years ago, the company released the H10, a dual-controlled, hybrid SSD that was very unique and unlike anything we’ve seen before. The idea was simple: combine the best of both worlds onto a single, slim M.2 SSD stick – high-density NAND flash for capacity along with bleeding-edge Optane Memory for speed. However, our initial impression of the NVMe SSD was rather underwhelming compared to the best SSDs available.
Since then, the company has tweaked and tuned the Rapid Storage Technology caching software with multiple optimizations and improvements. Concurrently, the company focused on advancing the H-series hardware, too. Today, we analyze the latest version of the software in use with the H10’s successor, the H20. Leveraging essentially what is an Intel SSD 670p in conjunction with a newer Optane controller, and of course, that sweet, sweet Optane Memory, Intel’s Optane Memory H20 is quite similar to the H10, only improved.
When Optane caching is enabled, the SSD aggregates the performance of both storage mediums for fast peak performance and fast access times. Intel’s H20 is an OEM-only-oriented product, however, meaning that it is highly unlikely you would ever find this SSD on sale in retail outlets. Still, they may trickle down to eBay or similar marketplaces, so purchasers of systems containing this unique SSD can upgrade their capacity.
Specifications
Product
H20 512GB
H20 1TB
Capacity (User / Raw)
512GB / 512GB
1024GB / 1024GB
Form Factor
M.2 2280 S3
M.2 2280 S3
Interface / Protocol
PCIe 3.0 x4 / NVMe 1.3
PCIe 3.0 x4 / NVMe 1.3
Optane Controller
SLMXT
SLMXT
Optane Media
1st Gen 3D Xpoint
1st Gen 3D Xpoint
Optane Capacity
32GB
32GB
NAND Controller
SM2265
SM2265
DRAM
DDR3
DDR3
NAND Flash
Intel 144L QLC
Intel 144L QLC
Sequential Read
3,300 MBps
3,300 MBps
Sequential Write
2,100 MBps
2,100 MBps
Random Read
65,000 IOPS
65,000 IOPS
Random Write
40,000 IOPS
40,000 IOPS
Security
Pyrite 2.0
Pyrite 2.0
Endurance (TBW)
185 TB
370 TB
Intel’s H20 comes in limited capacities of just 512GB and 1TB, and both models come equipped with 32GB of Optane Memory. Intel rates both for the same up to 3.3/2.1 GBps read/write speeds and up to 65,000/40,000 random read/write IOPS at a queue depth (QD) of 1. While the sequential figures aren’t groundbreaking, no flash-based SSD comes close to delivering the same random IOPS performance at low QDs.
As it is an OEM-only product, the pricing for the device is not clear, but the company states it will ship in June for PCs priced at roughly $800 and up. Additionally, the H20 comes with tight hardware requirements. The H20 is only compatible with 11th-Gen Intel Core series processors and Intel 500-series chipsets or newer. Also, you will need Windows, Intel’s RST Driver 18.1 or newer, and there is no planned retroactive support for previous-gen systems.
The H20 supports Pyrite 2.0 security but lacks AES 256-bit hardware-accelerated encryption. It also supports S.M.A.R.T. data reporting, Trim, and is rated to consume as little as 35mW at idle to reduce power consumption and heat generation.
A Closer Look
Image 1 of 3
(Image credit: Tom’s Hardware)
Image 2 of 3
(Image credit: Tom’s Hardware)
Image 3 of 3
(Image credit: Tom’s Hardware)
Intel’s H20 comes in an M.2 2280 single-sided form factor. The drive interfaces with the host over four PCIe 3.0 lanes and communicates via the NVMe 1.3 protocol. The H20 takes advantage of the same components found in the company’s 670p and a newer and faster Optane controller. As with the H10, the H20’s SSD controllers are given two lanes each, meaning that sequential performance is limited if you don’t enable Optane Memory acceleration.
Image 1 of 3
(Image credit: Tom’s Hardware)
Image 2 of 3
(Image credit: Tom’s Hardware)
Image 3 of 3
(Image credit: Tom’s Hardware)
Silicon Motion developed the NAND controller specifically for Intel and the company’s 144L QLC. The SM2265 is a dual-core, four-channel NVMe controller that interfaces with the flash at fast speeds of up to 1,200 MTps, roughly double the speed of the previous-gen flash. The drive does have DRAM, but very little – our 1TB sample contains only 256MB of DDR3.
Intel’s 144L QLC uses a floating gate design with three 48-layer decks stacked atop each other. Each deck can operate as SLC or QLC. Each deck can also be erased without disturbing the data on other decks, which helps reduce latency spikes caused by garbage collection. It also has four planes for handling parallel data operations and a few new reading and writing techniques to improve responsiveness.
Image 1 of 2
(Image credit: Tom’s Hardware)
Image 2 of 2
(Image credit: Tom’s Hardware)
Intel made revisions under the hood of the Optane Media controller, too. It features both performance and power management improvements that reduce overall power consumption, something that is needed when handling power-hungry Optane media on an M.2 form factor in its own right, and even more important when combining it with the secondary storage components. The Optane Memory is still first-gen media, however.
According to a Reuters report, a group of four U.S. senators might unveil their $52-billion microelectronics industry backing proposal on Friday or early next week. The chip industry support plan is set to include funding of R&D and production and the establishment of national programs. The $52 billion five-year plan pales in comparison to South Korea’s intention to support its semiconductor industry with $450 billion over the next 10 years. Still, the real question is if we can compare the plans directly.
$52 Billion U.S. Semiconductor Bill
Various companies in the U.S. design the vast majority of chips used globally, yet only 12% of semiconductors are produced on American soil. The microelectronics industry funding bill is designed to at least partially change that and bring more chip production back to the USA.
(Image credit: GlobalFoundries)
The U.S. semiconductor industry funding law is developed by senators Mark Kelly, John Cornyn, Mark Warner, and Tom Cotton. According to a draft document seen by Reuters, the plan is expected to include $39 billion in production and R&D incentives and $10.5 billion to implement various national programs over a five-year period. Among the government-supported R&D initiatives are the National Semiconductor Technology Center and National Advanced Packaging Manufacturing Program.
The chip industry supporting bill will be a part of a bigger $110 billion effort to fund U.S. technology research in a bid to better compete with China. The same bill is also set to include semiconductor requirements of this year’s National Defense Authorization Act, which will obviously somewhat help the American microelectronics industry, too.
“There is an urgent need for our economic and national security to provide funding to swiftly implement these critical programs,” the summary of the bill seen by Reuters reportedly reads. “The Chinese Communist Party is aggressively investing over $150 billion in semiconductor manufacturing so they can control this key technology.”
Typically, when a semiconductor company builds a fab in the U.S. and Europe, it gets incentives from federal and local governments. As a result, expect states to provide incentives to companies like Intel, so in the end, these chipmakers will get considerably more than $39 billion. Yet, even when federal and state chip funding initiatives are combined, the total sum will be considerably lower than South Korea’s planned spending on its microelectronics industry.
$450 Billion South Korean Chip Plan
Earlier this week South Korea unveiled an intention to support its local chip industry with $450 billion over the next 10 years.
(Image credit: Panasonic)
Historically, automotive and chemical/petrochemical products were South Korea’s main export items. Today, semiconductor sales account for around 14.6% of Korea’s exports and represent the largest exports category. Furthermore, since chips enable the vast majority of today’s products, from mice to TVs and from smartphones to vehicles, the importance of the semiconductor industry is hard to overestimate, especially for South Korea.
There is a catch, though. Samsung and SK Hynix — Korea’s largest chip companies — mostly export commodity DRAM and 3D NAND memory chips, not advanced logic chips, like CPUs and SoCs. Samsung, of course, has leading-edge process technologies, yet it produces chips developed by others and in different parts of the world. In fact, there aren’t many designers of sophisticated processors in South Korea – even parts of Samsung’s Exynos SoCs are designed in the USA.
The multifaceted $450 billion plan is set to change this. In a bid to revamp its domestic semiconductor industry, South Korea intends to help train 36,000 engineers between 2022 and 2031 and contribute $133 billion (1.5 trillion won) to semiconductor R&D programs. Furthermore, the country will help chip designers, manufacturers, and suppliers with tax breaks, lower interest rates, eased regulations, and strengthened infrastructure (which includes ensuring adequate water and power supply for chip makers), reports Bloomberg. South Korea has already gained endorsements from ASML and LAM Research that announced plans to expand their presence in the country.
In general, South Korea wants to build a vertically integrated semiconductor industry that will include leading-edge R&D operations, the development of world-class processors, and the production of chips using the most advanced fabrication technologies.
Different Goals
While the U.S. and South Korean semiconductor industry supporting plans are both unveiled in May and are aimed at the same industry, they are designed for different purposes, and to that end should not be compared directly as they pursue different goals.
(Image credit: TSMC)
American companies like AMD and Nvidia are focused on developing advanced logic chips that they can sell for thousands of U.S. dollars per unit and earn great margins. Some other U.S.-based companies like Apple and Tesla design chips so they could add unique capabilities to their products, such as smartphones, PCs, and automobiles. Manufacturing in the USA is something that these companies might prefer, yet their business efforts are focused on the development of products (which includes hardware and software) that they actually sell, not on the production of microelectronics.
For the U.S. as a country, manufacturing advanced semiconductors at home makes a lot of sense both from commercial and national security points of view. The USA already has the world’s leading developers of CPUs, GPUs, SoCs, and process technologies. Hence, the $52 billion bill is mostly designed to support domestic manufacturing, perhaps not the most lucrative (yet crucially important) portion of the global semiconductor supply chain. Considering that companies like TSMC and Samsung spend around $30 billion a year on capital expenditures (i.e., building new fabs and upgrading existing ones), $39 billion planned to help U.S. chip production over a five-year period is not much, though.
South Korea is the world’s leading producer of commodity memory chips. In contrast, it is not known for leading-edge microprocessors and system-on-chips. Refocusing the industry from commodities to premium products requires investments in education, the attraction of foreign companies and talent, and stimulating local chip designers, all while helping Samsung and SK Hynix to remain the world’s leading makers of memory and logic chips.
South Korea’s plan is much more ambitious and much more complex than the one U.S. senators are about to announce, which is why it costs a lot more money.
There are two big questions here. Can the U.S. remain on top of the global semiconductor supply chain now that other countries are pouring in hundreds of billions in their microelectronics industries? Can South Korea buy its way into the major league with money? Only time will tell.
Samsung spent tens of billions of dollars annually on its various chip businesses for several years now. On Thursday the company announced that its renewed semiconductor investment plan includes spending $150 billion on R&D and CapEx for its LSI and foundry businesses by 2030.
$15 Billion per Year
About two years ago Samsung announced plans to spend $115 billion through 2030 (i.e., $9.51 billion a year) on its LSI and foundry businesses, which includes R&D of new process technologies and materials, chip design, and actual fabs. $9.51 billion a year is a lot of money, but from now on the company is committed to spending $15 billion a year on its LSI and semiconductor production operations. The increase is motivated by the growing demand for chips and semiconductor fabrication services driven by megatrends like 5G, AI, and HPC.
Samsung Electronics has been outspending all other semiconductor producers in the world in recent years. Just last year it invested around $28.1 billion in production facilities that are used by its Samsung Foundry and memory (3D NAND, DRAM) businesses. With an increased commitment to foundry business, Samsung will almost certainly increase the general CapEx of its semiconductor division to well beyond $30 billion a year.
“The entire semiconductor industry is facing a watershed moment and now is the time to chart out a plan for long-term strategy and investment,” said Dr. Kinam Kim, Vice Chairman and Head of Device Solutions Division at Samsung Electronics.
Samsung P3: A Fab for DRAM and Logic
In an interesting turn of events, Samsung also announced that it had begun construction of its new P3 fab near Pyeongtaek, South Korea, that will be used to produce both DRAM and logic chips using process technologies that rely on extreme ultraviolet (EUV) lithography. Normally, Samsung uses different fabs to make DRAM and logic chips, but it looks like with next-generation EUV tools it makes sense to use one fab for both types of products.
Samsung’s P3 is expected to be completed in the second half of 2022. Initially, it will produce chips using Samsung’s 14 nm technology for DRAMs as well as 5 nm node for SoCs. Looks like leading-edge fabrication technologies, including 4nm and 3nm, will continue to be initially used at Samsung Foundry’s V1 fab near Hwaseong, South Korea.
“For the memory business, where Samsung has maintained its undisputed leadership position, the company will continue to make preemptive investments to lead the industry,” said Dr. Kinam Kim.
Taiwan started a rotational electricity blackout across the country on Thursday following a major incident at a coal-fired power plant in southern Taiwan that caused its shutdown. The outages have not affected TSMC, the world’s largest contract maker of chips, other foundries, and DRAM makers. However, chip packaging houses had to halt operations briefly.
Taiwan’s high-tech manufacturing, which serves the lion’s share of the world’s need for chips, is crucially dependent on electricity supply. According to Reuters, Taiwan Semiconductor Manufacturing Co. admitted that some of its facilities suffered a ‘brief power dip,’ but power had been restored. According to DigiTimes, United Microelectronics Corp.’s fabs located in southern Taiwan also suffered a voltage drop, yet this did not interrupt production.
Taiwan is home to several major contract makers of semiconductors, including TSMC, UMC, Vanguard International Semiconductor (VIS), and Powerchip Semiconductor Manufacturing Corp (PSMC). At this point, they control 56% of the global contract chip production, so any issue they experience can become global.
Semiconductor production facilities usually run on diesel generators during brief power dips, but these cannot feed the power hungry fabs for long. Neither TSMC nor UMC had to shut down any fab because of the controlled blackout. TSMC said that it had seen “no major impact so far.” Just like the top contract chipmakers, smaller contract producers of semiconductors in Taiwan, namely VIS and PSMC, did not run into any problems, according to reports from DigiTimes and TrendForce.
There are also multiple DRAM fabs owned by Micron and some independent manufacturers like Nanya, accounting for about 21% of the global output. NAND production is not Taiwan’s strong side, but some makers produce specialty flash memory. Fortunately, production at all DRAM and NAND fabs was not interrupted, yet some companies confirmed voltage drops.
Several companies, including AU Optronics and Innolux, produce LCD panels in Taiwan. Like semiconductor fabs, LCD production facilities cannot be shut down even for a short time without any impact. Therefore, blackouts can certainly affect supplies. The good news is that neither AUO nor Innolux had to halt operations anywhere in Taiwan because of the blackouts.
ASE Technology, the world’s largest outsourced semiconductor assembly and test provider, confirmed that its operations had been affected. ChipMOS Technologies also informed that its gold bumping lines were impacted by the blackouts but did not elaborate.
According to a statement by the Ministry of Science and Technology cited by Bloomberg, all three major science parks in Taiwan were at least partly affected by the blackouts. Obviously, Southern Taiwan Science Park (STSP) was affected the most. Meanwhile, the administration of STSP said that 7 out of 40 companies located in the park were affected by the power blackout.
A major failure at the coal and gas-fired Hsinta Power Plant in Kaohsiung that required its shutdown occurred at 2:47 PM local time (2:47 AM ET). At around 3 PM, Taiwan residents received an emergency alert about the incident, reports Taiwan News. According to Reuters, the notice said that the country did not have enough electricity capacity in the power grid and that the government began a rotational electricity blackout across the country.
Taipower said that the rotational outages would continue until at least 4:40 PM and that six million homes were hit by blackouts, though this seems to be an underestimate as Taipei went black at 3:15 PM.
Phison expects SSD pricing to increase by at least 10% in the coming months due to rising demand driven by Chia cryptocurrency mining coupled with supply constraints. Furthermore, the company expects a currently short supply of SSD controllers to persist through 2022 and 2023.
Phison sells SSDs and other NAND flash-based storage devices powered by its own controllers to branded drive suppliers and PC makers. The company has a vast portfolio of products that use NAND memory from different manufacturers, so it has business relations with foundries (which produce its controllers), NAND makers, component suppliers, and OEMs. Microsoft is one of Phison’s major customers — the company uses a Phison-based SSD in its latest Xbox game consoles.
Unofficial reports say that Phison and other Taiwan-based designers of SSD controllers increased prices of their chips in the first quarter because of high demand amid production capacity constraints at foundries. Phison generated $460 million in revenue in Q1 2021 and earned $60.6 million in profits during the quarter, reports DigiTimes. The company is optimistic about its business in Q2 2021 as the demand for storage is increasing.
KS Pua, chairman of Phison, reportedly said that NAND flash makers were set to increase memory pricing again, by approximately 10% in Q3 2021, because of high demand driven by the growing PC market and the rise of Chia cryptocurrency mining. Higher flash pricing and the higher price for controllers will inevitably make SSDs more expensive in the coming months.
Unfortunately, it is hard to guess how significantly SSD prices might increase. Still, traditionally the prices of entry-level models get substantially higher as they are sold with razor-thin margins, and there is no way for suppliers to keep their prices stable, even at the cost of their own profits. Meanwhile, as Chia farmers prefer higher-end SSDs with better endurance, rapidly growing demand for such drives may also affect their prices.
SSD controller pricing will remain high in 2022 and 2023 due to constraints of production capacities at foundries, according to the chairman of Phison. Companies like TSMC, UMC, GlobalFoundries, and others are expanding their mature node production capacity to help address the shortfall, but the process will take some time.
We use cookies on our website to give you the most relevant experience. By clicking “Accept”, you consent to the use of ALL the cookies.
This website uses cookies to improve your experience while you navigate through the website. Out of these, the cookies that are categorized as necessary are stored on your browser as they are essential for the working of basic functionalities of the website. We also use third-party cookies that help us analyze and understand how you use this website. These cookies will be stored in your browser only with your consent. You also have the option to opt-out of these cookies. But opting out of some of these cookies may affect your browsing experience.
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.
Any cookies that may not be particularly necessary for the website to function and is used specifically to collect user personal data via analytics, ads, other embedded contents are termed as non-necessary cookies. It is mandatory to procure user consent prior to running these cookies on your website.
