While the new CAMM and LPCAMM memory modules for laptops have garnered a great deal of attention in recent months, it's not just the mobile side of the PC memory industry that is looking at changes. The desktop memory market is also coming due for some upgrades to further improve DIMM performance, in the form of a new DIMM variety called the Clocked Unbuffered DIMM (CUDIMM). And while this memory isn't in use quite yet, several memory vendors had their initial CUDIMM products on display at this year's Computex trade show, offering a glimpse into the future of desktop memory.

A variation on traditional Unbuffered DIMMs (UDIMMs), Clocked UDIMMs (and Clocked SODIMMs) have been created as another solution to the ongoing signal integrity challenges presented by DDR5 memory. DDR5 allows for rather speedy transfer rates with removable (and easily installed) DIMMs, but further performance increases are running up against the laws of physics when it comes to the electrical challenges of supporting memory on a stick – particularly with so many capacity/performance combinations like we see today. And while those challenges aren't insurmountable, if DDR5 (and eventually, DDR6) are to keep increasing in speed, some changes appear to be needed to produce more electrically robust DIMMs, which is giving rise to the CUDIMM.

Standardized by JEDEC earlier this year as JESD323, CUDIMMs tweak the traditional unbuffered DIMM by adding a clock driver (CKD) to the DIMM itself, with the tiny IC responsible for regenerating the clock signal driving the actual memory chips. By generating a clean clock locally on the DIMM (rather than directly using the clock from the CPU, as is the case today), CUDIMMs are designed to offer improved stability and reliability at high memory speeds, combating the electrical issues that would otherwise cause reliability issues at faster memory speeds. In other words, adding a clock driver is the key to keeping DDR5 operating reliably at high clockspeeds.

All told, JEDEC is proposing that CUDIMMs be used for DDR5-6400 speeds and higher, with the first version of the specification covering speeds up to DDR5-7200. The new DIMMs will also be drop-in compatible with existing platforms (at least on paper), using the same 288-pin connector as today's standard DDR5 UDIMM and allowing for a relatively smooth transition towards higher DDR5 clockspeeds.

In response to the increasing popularity of compact and portable gaming PCs in the past couple of years, we've seen several PC peripheral manufacturers release larger collections of SFX form factor cases and power supplies. Chief among these has been Cooler Master, where the prolific PC part vendor recently expanded their SFX series to include ATX 3.0-compliant PSUs. This has finally given the company a modern, miniature platform that can drive the latest NVIDIA video cards.

Today we're looking at Cooler Master's V850 SFX Gold, which represents a more budget-friendly option in the SFX power supply market compared to their previously reviewed V1100 SFX Platinum. The SFX Gold series includes four models, ranging from 550W to 850W, with the 850W unit that we are reviewing today being the most potent. This 850-Watt power supply carries an 80 Plus Gold certification and is designed for high-performance gaming and workstation builds. As the top-tier model in the SFX Gold series, it offers substantial power output for an SFX unit and maintains ATX 3.0 compliance, including a 12VHPWR connector in a compact SFX format.

Western Digital expanded its WD Blue NVMe family of budget SSDs this week with the introduction of the the WD Blue SN5000 series, an updated lineup of SSDs that, among other things, adds a 4 TB model. Being budget drives, the SN5000 series is not going to be a performance monster, but their combination of capacity, reliability, and relatively low price could make them popular both among casual buyers and among enthusiasts looking for relatively cheap bulk solid-state storage.

Western Digital's WD Blue SN500 NVMe drives come in an M.2-2280 form-factor and are based around an in-house WD controller (the company rarely discloses their codenames these days). WD's controller is a 4 channel, DRAMless design, which is what we typically see for budget and mainstream SSDs. Externally, the controller supports a PCIe 4.0 x4 connection, and per WD's specifications, even this budget drive should be fast enough to put the additional bandwidth of PCIe 4.0 to good use. All of the drives are single-sided, and are covered with a thin graphene heatspreader for heat dissipation.

More surprising here is WD's choice of NAND. According to the company, they are using a mix of different generations and different types of NAND, depending on the model. The 500GB/1TB/2TB models are using WD/Kioxia's older 112 layer BiCS 5 TLC NAND. Meanwhile the new 4TB capacity is being enabled with the company's newer BiCS 6 NAND – but the QLC variety. All drive capacities are being backed by the same controller, so there is a thread of commonality between them, but at a high level WD seems to be using the higher performance of BiCS 6 to offset the switch from TLC to QLC.

WD Blue SN5000 SSD Specifications
Capacity	500 GB	1 TB	2 TB	4 TB
Controller	WD In-House: 4 Channel, DRAMless
NAND Flash	WD BiCS 5 TLC			WD BiCS 6 QLC
Form-Factor, Interface	Single-Sided M.2-2280, PCIe 4.0 x4, NVMe
Sequential Read	5000 MB/s	5150 MB/s	5150 MB/s	5500 MB/s
Sequential Write	4000 MB/s	4900 MB/s	4850 MB/s	5000 MB/s
Random Read IOPS	460K	730K	650K	690K
Random Write IOPS	770K			800K
Peak Power	6.3W?
SLC Caching	Yes
TCG Pyrite Encryption	2.01
Warranty	5 Years
Write Endurance	300 TBW 0.33 DWPD	600 TBW 0.33 DWPD	900 TBW 0.24 DWPD	1200 TBW 0.16 DWPD
MSRP	$70	$80	$140	$280

When it comes to performance, the WD Blue SN5000 series drives are rated for read speeds between 5000 MB/sec and 5500 MB/sec depending on the capacity, while write speeds range from 4000 MB/sec to 5000 MB/sec. As for random performance, we are looking at up to 690K 4K IOPS random read speeds as well as up to 900K 4K IOPS random write speeds for the highest capacity model, while the lower-end 500 GB model is rated for 460K/770K 4K random read/write IOPS.

Overall, even with the 4TB model using QLC NAND, WD is touting it at offering better performance than any of the lower capacity models. We're accustomed to seeing QLC drives come in behind TLC drives in this respect, so that newer generation of NAND is doing a lot of heavy lifting to put it ahead of the other models.

This goes for write endurance as well; the 4TB QLC model has the highest endurance rating, at 1200 TB written, followed by 900, 600, and 300 TBW for the lower capacity models respectively. Depending on the specific drive model, this works out to between 300 to 600 drive writes in total, or around 0.164 drive writes per day, which is typical for drives in this class.

Overall, the new drive family supplants WD's previous generation of Blue drives, last year's SN580 series. Comparatively, the SN5000 drives are rated to offer better sequential and random drive performance at every tier. And the the tiny 250GB drive has been dropped entirely, making room for the new high-capacity 4TB model at the high-end while the 500GB drive model is the new entry-level capacity.

While the WD Blue SN5000 NVMe family seems to be well positioned to be a low-cost drives meant to compete aggressively on the pricing, for now WD's launch prices are a bit ambitious. The $70 price tag for the 500 GB version is pretty typical, while the $280 recommended price for a 4TB model puts it in a weird spot between a number of other 4TB drives. The tad slower Crucial P3 Plus 4TB can be found for $217, while WD's own considerably faster WD_Black SN850X 4TB is only $30 more, at $310. Ultimately, given that Western Digital's Blue drives are meant to be inexpensive SSDs, we expect retail drive prices to catch up with market realities shortly.

Thanks to the success of the burgeoning market for AI accelerators, NVIDIA has been on a tear this year. And the only place that’s even more apparent than the company’s rapidly growing revenues is in the company’s stock price and market capitalization. After breaking into the top 5 most valuable companies only earlier this year, NVIDIA has reached the apex of Wall Street, closing out today as the world’s most valuable company.

With a closing price of $135.58 on a day that saw NVIDIA’s stock pop up another 3.5%, NVIDIA has topped both Microsoft and Apple in valuation, reaching a market capitalization of $3.335 trillion. This follows a rapid rise in the company’s stock price, which has increased by 47% in the last month alone – particularly on the back of NVIDIA’s most recent estimates-beating earnings report – as well as a recent 10-for-1 stock split. And looking at the company’s performance over a longer time period, NVIDIA’s stock jumped a staggering 218% over the last year, or a mere 3,474% over the last 5 years.

NVIDIA’s ascension continues a trend over the last several years of tech companies all holding the top spots in the market capitalization rankings. Though this is the first time in quite a while that the traditional tech leaders of Apple and Microsoft have been pushed aside.

Market Capitalization Rankings
	Market Cap	Stock Price
NVIDIA	$3.335T	$135.58
Microsoft	$3.317T	$446.34
Apple	$3.285T	$214.29
Alphabet	$2.170T	$176.45
Amazon	$1.902T	$182.81

Driving the rapid growth of NVIDIA and its market capitalization has been demand for AI accelerators from NVIDIA, particularly the company’s server-grade H100, H200, and GH200 accelerators for AI training. As the demand for these products has spiked, NVIDIA has been scaling up accordingly, repeatedly beating market expectations for how many of the accelerators they can ship – and what price they can charge. And despite all that growth, orders for NVIDIA’s high-end accelerators are still backlogged, underscoring how NVIDIA still isn’t meeting the full demands of hyperscalers and other enterprises.

Consequently, NVIDIA’s stock price and market capitalization have been on a tear on the basis of these future expectations. With a price-to-earnings (P/E) ratio of 76.7 – more than twice that of Microsoft or Apple – NVIDIA is priced more like a start-up than a 30-year-old tech company. But then it goes without saying that most 30-year-old tech companies aren’t tripling their revenue in a single year, placing NVIDIA in a rather unique situation at this time.

Like the stock market itself, market capitalizations are highly volatile. And historically speaking, it’s far from guaranteed that NVIDIA will be able to hold the top spot for long, never mind day-to-day fluctuations. NVIDIA, Apple, and Microsoft’s valuations are all within $50 billion (1.%) of each other, so for the moment at least, it’s still a tight race between all three companies. But no matter what happens from here, NVIDIA gets the exceptionally rare claim of having been the most valuable company in the world at some point.

(Carousel image courtesy MSN Money)

Just several years ago, we were impressed to witness a 2,000W power supply for mining systems at Computex. But with arrival or AMD's Ryzen Threadripper WX and Intel's Xeon W-3400-series – CPUs that can devour all the way to 900W under heavy loads – as well as NVIDIA's GeForce RTX 4090 graphics cards (or, well, H100 PCIe for AI), 2,000W PSUs are no longer entirely overkill. As new processors have raised the bar on power consumption, so has the bar been raised on PSUs themselves. as evidenced by this year's show, with several power supply manufacturers showing off new extreme PSU designs that go as high as 2,800W.

To that end, I will begin the story from the least 'extreme' PSU that I saw at Computex and this is Seasonic's Prime PX-2200 80+ Platinum-badged ATX 3 PSU. The unit handled a custom-built AMD Ryzen Threadripper Pro 7995WX-based PC with 256 GB DDR5-4800 memory and four Gigabyte GeForce RTX 4090 graphics cards without breaking a sweat at the trade show.

It is noteworthy that the Prime PX-2200 has only one 16-pin 12V-2x6 power connector for graphics cards, so Seasonic used adapters to power three out of four RTX 4090 boards.

Moving on to a more powerful PSU, which is Enermax's Platimax 2400W 80+ Platinum-rated ATX 3.1 power supply. This one has two 12V-2×6 auxiliary power connectors for graphics cards, which is more than enough for an extreme gaming system. Enermax did not showcase this one in action at Computex, perhaps, because it is almost half of a year away and yet has to pass all the necessary tests.

The Enermax Platimax 2400W is set to enter mass production this December and will be available for $499, according to the company.

But what is 2,400W, if you can get all 2,800W? Well, Super Flower's Leadex SF-2800F14HP 80+ Platinum-badged ATX 3.1 PSU can deliver 2,800W to a high-performance CPU and four graphics cards over five voltage rails. This power supply is well beyond what even a high-end desktop needs, and seems aimed more at workstation-style AI systems for research or on-prem inference. This one is expected to be available later this year for around $599, according to a representative at the booth.

It goes without saying that these extreme power supplies have a rather limited market. Not only because of the small base of systems that could ever need a PSU that could be measured as a multiple of horsepower (745W), but also because of outright mains compatibility – the standard US 15A/120V outlet can't drive much more than a 1600W PSU. So power supplies like these are only readily usable in Europe, and in locations in the US with non-standard outlets. Then again, if you need a 2,800W PSU, you can probably afford the electrician needed to get one wired up.

While the market for high-end CPU coolers has decidedly shifted towards closed-loop all-in-one liquid coolers over the last several years, air cooling remains alive and well. Even at the high-end, there are still further improvements and innovations being made, such as DeepCool's vapor chamber-based tower cooler, which was demonstrated at Computex.

Named the Assassin IV VC Vision, DeepCool's design is an advanced concept vehicle that equips a tower cooled with both a vapor chamber in the base as well as has an LCD pad on top for extra flourish. The vapor chamber is said to increase the cooling capacity by 20W, adding a bit more of an edge to an already very powerful tower cooler design.

While we expect this one to come to market eventually, don't be surprised if both vapor chamber and the screen to land on other products together or separately. For example, the massive DeepCool Assassin IV VC Vision has a more compact brother that has a screen and a vapor chamber.

The unit builds on top of the already monstrous DeepCool Assassin IV that comes with seven 0.6-mm heat pipes and can mount up to three 120/140mm fans, depending on installation (one is magnetically attached). With a weight of 1.575 kilograms – almost entirely copper and aluminum – this one is already good enough to cool down even the highest-performing CPUs.

DeepCool is currently trying to figure out recommended pricing for its Assassin IV VC Vision cooler, but the original Assassin IV costs $99.99, so expect the unit with a vapor chamber and a screen to build on top of that.

One of the more subtle trends at this year's Computex was that the majority of high-end motherboards now come equipped with a 5GbE network controller. At present, this might be considered a limited benefit as 5GbE and 10GbE switches and routers are still pretty expensive. But Realtek is planning on tackling the issue from both ends, as the company is preparing a hardware platform for sub-$100 5GbE switches. 

Realtek’s quad-port 5GbE switch platform consists of five key chips: one RTL9303 switch system-on-chip, and four RTL8251B 5GbE physical interfaces (PHYs). The chips are accompanied by various other components, such as power management ICs, but in general it relies solely on in-house developed components, which is why it can be made so cheap. 

RealTek's platform is meant for home and small offices, which is reflected in the switch's feature set. This is a simple, unmanaged switch with a handful of ports, making it ideal for linking up a few systems, while enterprise users will likely find it a bit too basic.

Segmenting the market like this is ultimately critical for bringing down the price of hardware. The bulk of 5GbE/10GbE switches on the market today are more enterprise-focused managed switches, which carry more features and a price premium to match. So developing a stripped-down platform for cheaper consumer switches is a huge development that should finally make it economical for consumers to adopt faster networking hardware, similar to 2.5GbE a few years ago.

At present, 2.5GbE switches are running at around $20/port, so RealTek's sub-$100 target for a 4-port switch aims to bring 5GbE in at just a slightly higher price tag of $25/port. Or, compared to the handful of unmanaged 10GbE switches on the market, which average $60/port, this will be less than half the price (though at half the bandwidth).

The proliferation of cheap 5GbE network switches will also mark a notable inflection point in Ethernet hardware design, as it's the fastest standard that is rated to work the ubiquitous Cat 5e cable. The NBASE-T standard was penned almost a decade ago in order to cover the missing-middle between 1GbE and 10GbE, while getting more bandwidth out of existing, widely-deployed Cat 5e cabling. So with the release of consumer 5GbE gear, the standard's goals are finally coming to fruition – though it does mean we're finally reaching the end of the road for the oldest network cabling still widely in use.

For now, Realtek is only talking about one customer offering a sub-$100 5GbE switch this September, but something tells me that other partners of the company will come up with similar devices soon enough. As a result of the competition, prices could get even lower, which is always good for buyers.

These benefits should also funnel into Wi-Fi 7 routers, to a limited extent. Wired backhaul speeds need to keep pace with ever-faster Wi-Fi standards in order to keep those newer radios fed, so it's not a coincidence that cheaper 5GbE is finally coming right as Wi-Fi 7 is taking off.

Overall, the company's 5GbE switch platform is part of a larger ecosystem of 5GbE hardware that Realtek was showing off at Computex. In addition to RTL8251B 5GbE PHYs and RTL9303 switch SoC, the company is also preparing their RTL8126 PCIe 3.1 network controller, and RTL8157 NIC for USB dongles.

Later this year Intel is set to introduce its Xeon 6-branded processors, codenamed Granite Rapids (6x00P) and Sierra Forest (6x00E). And with it will come a new slew of server motherboards and pre-built server platforms to go with it. On the latter note, this will be the first generation where Intel won't be offering any pre-builts of its own, after selling that business off to MiTAC last year.

To that end, MiTAC and its subsidiary Tyan were at this year's event to demonstrate what they've been up to since acquiring Intel's server business unit, as well as to show off the server platforms they're developing for the Xeon 6 family. Altogether, the companies had two server platforms on display – a compact 2S system, and a larger 2S system with significant expansion capabilities – as well as a pair of single-socket designs from Tyan.

The most basic platform that MiTAC had to show is their TX86-E7148 (Katmai Pass), a half-width 1U system that's the successor to Intel's D50DNP platform. Katmai Pass has two CPU sockets, supports up to 2 TB of DDR5-6400 RDIMMs over 16 slots (8 per CPU), and has two low-profile PCIe 5.0 x16 slots. Like its predecessor, this platform is aimed at mainstream servers that do not need a lot of storage or room to house bulky add-in cards like AI accelerators.

The company's other platform is TX77A-E7142 (Deer Creek Pass), a considerably more serious offering that replaces Intel's M50FCP platform. This board can house up to 4 TB of DDR5-6400 RDIMMs over 32 slots (16 per CPU with 2DPC), four PCIe 5.0 x16 slots, one PCIe 5.0 x8 slot, two OCP 3.0 slots, and 24 hot-swap U.2 bays. Deer Creek Pass can be used both for general-purpose workloads, high-performance storage, as well as workloads that require GPUs or other special-purpose accelerators.

Meanwhile Tyan had the single-socket Thunder CX GC73A-B5660 on display. That system supports up to 2 TB of DDR5-6400 memory over 16 RDIMMs and offers two PCIe 5.0 x16 slots, one PCIe 4.0 x4 M.2 slot, two OCP 3.0 slots, and 12 hot-swappable U.2 drive bays.

Finally, Tyan's Thunder HX S5662 is an HPC server board specifically designed to house multiple AI accelerators and other large PCIe cards. This board supports one Xeon 6 6700 processor, up to 1 TB of memory over eight DDR5-6400 RDIMMs, and has five tradiitonal PCIe 5.0 x16 slots as well as two PCIe 5.0 x2 M.2 slots for storage.

MiTAC is expected to start shipments of these new Xeon 6 motherboards in the coming months, as Intel rolls out its next-generation datacenter CPUs. Pricing of these platforms is unknown for now, but expect it to be comparable to existing servers.

Rounding out our Computex 2024 SSD controller news, let's talk about the final major controller vendor at the show: Silicon Motion. Silicon Motion has been somewhat late to the PCIe Gen5 SSD party, as its rival Phison has been offering their high-end PS5026-E26 controller for almost 18 months now. But as Silicon Motion gets ready to make the transition to PCIe 5.0, the company has a trump card up its sleeve: the SM2508 controller, an energy-efficient high-end SSD controller that is intended for lower-power drives.

A PCIe Gen5 x4 controller, SMI's SM2508 boasts a very low active power consumption of 3.5W. Which, installed in an M.2 SSD, is intended to allow drive manufacturers to built 7 Watt (or lower) SSDs, according to the company. The SM2508 is made on TSMC's N6 process technology, which is a 7nm-class fabrication node with an optical shrink, which will make it one of the (if not the) most advanced controllers on the market once it ships. The relatively sophisticated node allows Silicon Motion to pack more cores and more features into its controller while keeping its power consumption in check – not to mention the chip size.

Silicon Motion NVMe Client SSD Controller Comparison
	SM2508	SM2264	SM2268XT2	SM2269XT
Market Segment	High-End		Mainstream
Manufacturing Process	6nm	12nm	12nm	12nm
CPU Cores	4x Cortex R8	4x Cortex R8	2x Cortex R8	2x Cortex R8
Error Correction	4K+ LDPC	4K LDPC	4K+ LDPC	4K LDPC
DRAM	DDR4, LPDDR4X	DDR4, LPDDR4X	No	No
Host Interface	PCIe 5.0 x4	PCIe 4.0 x4	PCIe 4.0 x4	PCIe 4.0 x4
NVMe Version	NVMe 2.0	NVMe 1.4	NVMe 2.0	NVMe 1.4
NAND Channels, Interface Speed	8 ch, 3600 MT/s	8 ch, 1600 MT/s	4 ch, 3600 MT/s	4 ch, 1600 MT/s
Sequential Read	14.5 GB/s	7.5 GB/s	7.4 GB/s	5.1 GB/s
Sequential Write	14 GB/s	7 GB/s	6.7 GB/s	4.8 GB/s
4KB Random Read IOPS	2500k	1300k	1200k	900k
4KB Random Write IOPS	2500k	1200k	1200k	900k

Under the hood, the front-end of the SM2508 SSD controller is based around four Arm Cortex-R8 cores, which will make it a notably powerful controller (Phison's E26, for reference, is two Arm Cortex R5 cores with an accelerator). This controller is meant to be used with DRAM, as well, and will support both DDR4 and LPDDR4.

The back-end of the chip offers eight NAND channels with 32 CE targets, supporting interface speeds up to 3600 MT/s. Altogether, SMI says that the controller can hit sequential read speeds of up to 14.5GB/sec and write speeds of up to 14 GB/s, while 4K random read and write performance will reach up to 2.5 million IOPS. And SMI's demo drive at Computex did even better than that, achieving a 14.9GB/sec read speed under CrystalDiskMark.

Silicon Motion and its partners are currently finalizing the firmware for the SM2508 and qualifying it with various manufacturers' TLC NAND. The chip also supports QLC NAND, though we'd be surprised to see anyone pair such a powerful controller with slower NAND so soon.

At Computex, Silicon Motion itself demonstrated an SM2508 drive in action at its booth. Various drive manufacturers also had SM2508-based drives on the floor as well, though only as static samples.

With the highly integrated nature of PCs these days, the Mini-ITX form-factor has become a very viable option for high-performance gaming system. With plenty of motherboards available for both AMD and Intel's latest platforms, equipping an ITX system with a CPU is easy enough. But the small size of the form factor means that housing an ultra-wide flagship-level graphics card remains an issue. For this year's Computex trade show, Antec was showing off their solution to the problem of giant video cards: the Performance 1M case, a Mini-ITX case with a separate chamber just for a large video card.

Antec's Performance 1M is a dual-chamber Mini-ITX chassis that splits off the video card from the rest of the system. The primary chamber houses a Mini-ITX motherboard and an SFX power supply, while the secondary chamber houses a huge graphics card. Notably, the two chambers are bridged using a PCIe riser cable, allowing the motherboard chamber to be far shorter than even a half-height PCIe card, while the video card chamber can hold a triple-slot video card parallel to the motherboard chamber, cutting down on wasted space. For Antec's Computex demo, the company had a triple-slot ASUS GeForce RTX 4090 installed to showcase how this worked.

Antec's decision to allocate most of the case's volume to th video card chamber does come with a slight catch, however: it doesn't leave much space for a CPU cooler in the motherboard chamber. The short motherboard chamber means that system builders will have to use a low-profile cooler; these are readily available, but it does leave less thermal headroom overall for high-end CPUs. So there are still trade-offs for being able to accommodate a high-end video card.

Otherwise, the Performance 1M chassis from Antec look very stylish and are traditionally well built from stainless steel and aluminum. The chassis can be equipped with two 120-mm or two 140-mm coolers for extra airflow (which will help performance), one 2.5-inch SATA SSD, and even some RGB bling. To make the PC built inside a Performance 1M case more convenient to use, there are two USB Type-C ports on the front.

Of course, housing a GeForce RTX 4090 graphics card in a Mini-ITX chassis has its peculiarities when it comes to dimensions and Antec's Performance 1M measures 424×175×260mm, which is pretty large. On the other hand, for an ultra-high-performance gaming system, this is pretty compact.

Antec's Performance 1M Mini-ITX chassis will be available in gunmetal gray and matte black colors sometimes later this year. Pricing is something that the company is thinking about now, though keep in mind that we are talking about unique premium products.

While Realtek is best known in the enthusiast space for for its peripheral controllers such as audio codecs and network controllers, the company also has a small-but-respectable SSD controller business that tends to fly under the radar due to its focus on entry-level and mainstream drives. But Realtek's stature in the SSD space is on the rise, as the company is not only planning new PCIe Gen5 SSD controllers, but also their first high-end, DRAM-equipped SSD controller.

For this year's Computex trade show, Realtek laid out a new SSD controller roadmap that calls for the company to release a trio of new SSD controllers over the next couple of years. First up is a new four-channel entry-level PCIe 4.0 controller, the RTS5776DL, which will be joined a bit later by a PCIe 5.0 variant, the RTS5781DL. But most interesting on Realtek's new roadmap is the final chip being planned: the eight-channel, DRAM-equipped RTS5782, which would be the company's first high-end SSD controller, capable of hitting sequential read rates as high as 14GB/second.

Realtek NVMe SSD Controller Comparison
	RTS5782	RTS5781DL	RTS5776DL	RTS5772DL	RTS5766DL
Market Segment	High-End	Mainstream			Entry-Level
Error Correction	4K LDPC			2K LDPC
DRAM	DDR4, LPDDR4(X)	No	No	No	No
Host Interface	PCIe 5.0 x4	PCIe 5.0 x4	PCIe 4.0 x4	PCIe 4.0 x4	PCIe 3.0 x4
NVMe Version	NVMe 2.0	NVMe 2.0	NVMe 2.0	NVMe 1.4	NVMe 1.4
NAND Channels, Interface Speed	8 ch, 3600 MT/s	4 ch, 3600 MT/s	4 ch, 3600 MT/s	8 ch, 1600 MT/s	4 ch, 1200 MT/s
Sequential Read	14 GB/s	10 GB/s	7.4 GB/s	6 GB/s	3.2 GB/s
Sequential Write	12 GB/s	10 GB/s	7.4 GB/s	6 GB/s	2.2 GB/s
4KB Random Read IOPS	2500k	1400k	1200k	-	-
4KB Random Write IOPS	2500k	1400k	1200k	-	-

Diving a bit deeper into Realtek's roadmap, the RTS5776DL is traditional DRAM-less PCIe Gen4 x4 controller with four NAND channels, and is aimed at entry-level drives. The controller's NAND support is quite modern, however, supporting the latest ONFI/Toggle standards, which will allow it to hit NAND transfer rates up to 3600 MT/second. Across four channels, that's enough bandwidth to fully saturate a PCIe 4.0 x4 connection with sequential read/writes, while random 4K IOPS can burst as high as 1.2 million. Engineering samples of the controller are set to emerge in Q4 2024, so do not expect actual drives based on this chip to emerge for sale until the second half of next year at the earliest.

Realtek is also preparing the RTS5781DL, which can be thought of as a PCIe 5.0 version of their DRAM-less controller platform. The RTS5781DL features the same four-channel NAND layout and 3600 MT/sec max NAND transfer rates, which with the additional bandwidth afforded by PCIe 5.0, shifts the performance bottleneck back to the NAND. Overall, Realtek expects drives with its first PCIe Gen5 SSD controller to hit sustained transfer rates up to 10GB/second, and up to 1.4 million random read/write IOPS. The company aims to make engineering samples of this controller available in the first quarter of 2025, so actual drives will hit the market in late 2025 at best.

Finally, the pinnacle of Realtek's roadmap will be its RTS5782, which is the company's first high-end SSD controller. Besides including support for on-drive DRAM (DDR4/LPDDR4X) for higher performance, the back-end of the RTS5782 will feature an eight NAND channel design that supports transfer rates up to 3600 MT/second. Drives built with the controller are expected to be able to sustain 14GB/second sequential read rates and 12GB/second sequential writes, while the 4K random IOPS performance jumps to 2.5 million 4K read and write IOPS.

Unfortunately, the RTS5782 is also the farthest out of the three controllers, as it's still in the planning stages at Realtek. Consequently, for the moment the company isn't offering any guidance on when the new controller will be ready. No doubt the company will have more to show off next year for Computex 2025.

Many companies showed off a range of concept designs during Computex 2024, but perhaps Noctua had one of the coolest. Noctua, primarily known for its air coolers, had a pumpless prototype that showcased Thermisiphon cooling. It looks like a conventional AIO cooler on the surface but is much more efficient, cooler, and quieter than traditional water coolers.

Touching quickly on Thermosiphon cooling, it uses the natural laws of thermodynamics to use heat (the CPU) as a source to turn the liquid to the top, evaporating through the system. This evaporated liquid rises and is pushed into the condenser (radiator), which, in the case of Noctua's cooler, is cooled by fans and turned back into liquid, which then flows back toward the CPU. Not only does Theromsiphon technology use gravity to help aid the flow of liquid, but it is also a highly efficient way of distributing heat as vaporization can absorb heat better, and it doesn't feature any moving parts in the mechanism, aside from the fans, of course.

As with traditional AIO CPU coolers, Noctua's Thermosiphon design has a 240mm radiator with a pair of 120cm Noctua fans, which cools the vapor down and sends it back to the evaporator. While this isn't a new concept and vapor chamber cooling has been used for many years in CPUs and even laptop coolers, Thermosiphon is certainly a technology Noctua thinks it can leverage to ensure better cooling performance and robust reliability.

Speaking to Noctua at Computex 2024, they informed me of the benefits and pitfalls of the Thermosiphon technology. In terms of the pros, Thermosiphon cooling in itself has no moving parts, which not only improves reliability but also means there is less to go wrong. Another positive is that even the quietest AIO CPU coolers can be quite noisy, as the water pump that circulates the fluid can cause vibrations, which in turn makes noise. The biggest drawback of implementing Thermosiphon cooling for PCs is that conventional tubing can wear out over time. While adding metal tubing would last longer, it's not flexible, although Noctua said they have discussed this with system integrators.

Although Noctua's Thermosiphon is currently just a proof of concept and sis in the early development stage, there's no launch date or indication it will hit the market. It's an interesting design, and we hope it does eventually come to market.

The curtains are drawn and it’s almost showtime for Qualcomm and its Snapdragon X SoC team. After first detailing the SoC nearly 8 months ago at the company’s most recent Snapdragon Summit, and making numerous performance disclosures in the intervening months, the Snapdragon X Elite and Snapdragon X Plus launch is nearly upon us. The chips have already shipped to Qualcomm’s laptop partners, and the first laptops are set to ship next week.

In the last 8 months Qualcomm has made a lot of interesting claims for their high-performance Windows-on-Arm SoC – many of which will be put to the test in the coming weeks. But beyond all the performance claims and bluster amidst what is shaping up to be a highly competitive environment for PC CPUs, there’s an even more fundamental question about the Snapdragon X that we’ve been dying to get to: how does it work?

Ahead of next week’s launch, then, we’re finally getting the answer to that, as today Qualcomm is releasing their long-awaited architectural disclosure on the Snapdragon X SoC. This includes not only their new, custom Arm v8 “Oryon” CPU core, but also technical disclosures on their Adreno GPU, and the Hexagon NPU that backs their heavily-promoted AI capabilities. The company has made it clear in the past that the Snapdragon X is a serious, top-priority effort for the company – that they’re not just slapping together a Windows SoC from their existing IP blocks and calling it a day – so there’s a great deal of novel technology within the SoC.

And while we’re excited to look at it all, we’ll also be the first to admit that we’re the most excited to finally get to take a deep dive on Oryon, Qualcomm’s custom-built Arm CPU cores. The first new high-performance CPU design created from scratch in the last several years, the significance of Oryon cannot be overstated. Besides providing the basis of a new generation of Windows-on-Arm SoCs that Qualcomm hopes will vault them into contention in the Windows PC marketplace, Oryon will also be the basis of Qualcomm’s traditional Snapdragon mobile handset and tablet SoCs going forward.

So a great deal of the company’s hardware over the next few years is riding on this CPU architecture – and if all goes according to plan, there will be many more generations of Oryon to follow. One way or another, it’s going to set Qualcomm apart from its competitors in both the PC and mobile spaces, as it means Qualcomm is moving on from Arm’s reference designs, which by their very nature are accessible Qualcomm’s competition as well.

So without further ado, let’s dive in to Qualcomm’s Snapdragon X SoC architecture.

Samsung this week has unveiled its latest process technologies roadmap at the company's Samsung Foundry Forum (SFF) U.S. The new plan covers the evolution of Samsung's 2nm-class production nodes through 2027, including a process technology with a backside power delivery, re-emphasizing plans to bring out a 1.4nm-class node in 2027, and the introduction of a 'high value' 4nm-class manufacturing tech.

Samsung Foundry's key announcements for today are clearly focused on the its 2nm-class process technologies, which are set to enter production in 2025 and will span to 2027, when the company's 1.4-nm class production node is set to enter the scene. Samsung is also adding (or rather, renaming) another 2nm-class node to their roadmap with SF2, which was previously disclosed by Samsung as SF3P and aimed at high-performance devices.

"We have refined and improved the SF3P, resulting in what we now refer to as SF2," a Samsung spokesperson told AnandTech. "This enhanced node incorporates various process design improvements, delivering notable power, performance, and area (PPA) benefits."

Samsung Foundry for Leading-Edge Nodes Announced on June 12, 2024 Compiled by AnandTech
HVM Start		2023	2024	2025	2026	2027	2027
Process		SF3E	SF3	SF2 (aka SF3P)	SF2P/SF2X	SF2Z	SF1.4
FET		GAAFET
Power Delivery		Frontside				Backside (BSPDN)	?
EUV		0.33 NA EUV		?	?	?	?

This is another example of a rebranding of leading-edge fabrication nodes in the recent years by a major chipmaker. Samsung Foundry is not disclosing any specific PPA improvements SF3P has over SF2, and for now is only stating in high-level terms that it will be a better-performing node than the planned SF3P.

Meanwhile, this week's announcement also includes new information on Samsung's next batch of process nodes, which are planned for 2026 and 2027. In 2026 Samsung will have SF2P, a further refinement of SF2 which incorporates 'faster' yet less dense transistors. That will be followed up in 2027 with SF2Z, which adds backside power delivery to the mix for better and higher quality power delivery. In particular, Samsung is targetting voltate drop (aka IR drop) here, which is an ongoing concern in chip design.

Finally, SF1.4, a 1.4nm-class node, is on track for 2027 as well. Interestingly, however, it looks like it does not feature a backside power delivery. Which, per current roadmaps, would have Samsung as the only foundry not using BSPDN for their first 1.4nm/14Å-class node.

"We have optimized BSPDN and incorporated it for the first time in the SF2Z node we announced today," the spokesperson told us. "We will continue to refine this technology and apply it to future nodes, but we don't have a specific timeline to share at this time."

Chip Fab Roadmaps for Leading-Edge Nodes Data announced during conference calls, events, press briefings and press releases. Compiled by AnandTech
HVM Start		2023	2024	2025	2026	2027
Intel	Process	Intel 3	Intel 20A	Intel 18A	Intel 14A	Intel 10A
	FET	FinFET	RibbonFET (GAAFET)
	Power	Frontside	PowerVia (BSPDN)
	EUV	0.33 NA EUV			0.55 NA EUV + DSA
Samsung	Process	SF3E	SF3	SF2	SF2P/SF2X	SF2Z/SF1.4
	FET	GAAFET
	Power	Frontside				Backside/?
	EUV	0.33 NA EUV		?	?	?
TSMC	Process	N3E/N3P	N3S/N3X	N2	A16	A14 (?)
	FET	FinFET		GAAFET
	Power	Frontside			Super Power Rail (BSPDN)
	EUV	0.33 NA EUV				?

When compared to other contract fabs, Samsung's roadmap is now roughly in line with the rest in regards to 'nanometer' designations. Though absent further technical disclosures from Samsung, it remains unclear what the actual benefits will be for each node and how they compare to their predecessors, let alone how they'll compare to Intel Foundry and TSMC.

As Qualcomm's exclusivity for Arm-powered processors for Windows PCs is reportedly coming to its end, other chipmakers are getting ready to offer their Arm-based system-on-chips for Windows computers. And, according to a new report from Reuters, MediaTek will be among the companies jumping into the Windows-on-Arm field, with plans to launch their first PC processor late next year.

MediaTek's system-on-chip for Windows PCs will rely on Arm's 'ready-made designs,' according to Reuters. Which in turn hints that MediaTek would be using Arm's compute sub-system (CSS) for client PCs, a building block designed to significantly speed up development of SoCs.

With the vauge nature of the Reuters report, however, which version of Arm's IP MediaTek might be using remains unclear, and the answer to that will largely hinge on timing. Arm refreshes its client cores and IP offerings yearly – typically announcing them to the public in May – with finished chips rolling out as early as later in the year. So depending on just how late in the year MediaTek is planning to launch their chip, the company has a large enough window to potentially use either the current 2024 client designs, or next year's 2025 designs.

For reference, Arm's 2024 CSS for client systems is quite powerful on its own. It includes two ultra-high-performance Arm Cortex-X925 cores (each with up to 3MB L2 cache and clock speeds over 3.60 GHz, supporting SVE and SVE2), four high-performance Cortex-A725 cores, two energy-efficient Cortex-A520 cores, and an Immortalis-G925 graphics processor. And, of course, MediaTek has the expertise to skip Arm's CSS and build their own bespoke designs as well, if that's what they'd prefer.

Overall, the latest client designs from Arm can accommodate up to 14 CPU cores – Arm intentionally leaves headroom for designs to be scaled-up for laptops – which would make for quite a formidable chip. But the PC SoC market has no shortage of capable contenders with their own designs; besides Qualcomm's Snapdragon X processors, MediaTek would also be going up against the latest designs from Intel and AMD. All of whom are planning to make big plays for the mobile PC market in the next several months. So MediaTek will need to make a serious effort if their effort to jump into the PC SoC market are to succeed.

Since 2016, Microsoft has partnered with Qualcomm to bring Arm's processor architecture, which is widely used in smartphones, to Windows PCs. Qualcomm has an exclusive agreement to supply these chips for the next several months (the exact timing remains unclear), after which other designers like MediaTek can enter the market. Qualcomm, for its part, has benefited greatly from collaborating with Microsoft, so it will be interesting to see if Microsoft extends a similar hand out to other Arm chip makers.

Ultimately, the market for Arm PC SoCs has the potential to get crowded quickly. According to previous reports from Reuters, both AMD and NVIDIA are also developing Arm-based chips for Windows. So if all of those projects come to fruition, there could potentially be several Arm SoCs available to PC manufacturers around the same time. All of which would be a massive change from the past 20 years of the PC, where Intel and AMD have been the entire market.

Both MediaTek and Microsoft have declined to comment on the ongoing developments, the news agency states.

The USB Implementers Forum (USB-IF) introduced USB4 version 2.0 in fall 2022, and it expects systems and devices with the tech to emerge later this year and into next year. These upcoming products will largely rely on Intel's Barlow Ridge controller, a full-featured Thunderbolt 5 controller that goes above and beond the baseline USB4 v2 spec. And though extremely capable, Intel's Thunderbolt controllers are also quite expensive, and Barlow Ridge isn't expected to be any different. Fortunately, for system and device vendors that just need a basic USB4 v2 solution, ASMedia is also working on its own USB4 v2 controller.

At Computex 2024, ASMedia demonstrated a prototype of its upcoming USB4 v2 physical interface (PHY), which will support USB4 v2's new Gen 4 (160Gbps) data rates and the associated PAM-3 signal encoding. The prototype was implemented using an FPGA, as the company yet has to tape out the completed controller.

Ultimately, the purpose of showing off a FPGA-based PHY at Computex was to allow ASMedia to demonstrate their current PHY design. With the shift to PAM-3 encoding for USB4 v2, ASMedia (and the rest of the USB ecosystem) must develop significantly more complex controllers – and there's no part of that more critical than a solid and reliable PHY design.

As part of their demonstration, ASMedia had a classic eye diagram display. The eye diagram demoed has a clear opening in the center, which is indicative of good signal integrity, as the larger the eye opening, the less distortion and noise in the signal. The horizontal width of the eye opening represents the time window in which the signal can be sampled correctly, so the relatively narrow horizontal spread of the eye opening suggests that there is minimal jitter, meaning the signal transitions are consistent and predictable. Finally, the vertical height of the eye opening indicates the signal amplitude and the rather tall eye opening suggests a higher signal-to-noise ratio (SNR), meaning that the signal is strong compared to any noise present.

ASMedia itself is one of the major suppliers for discrete USB controllers, so the availability of ASMedia's USB4 v2 chip is crucial for adoption of the standard in general. While Intel will spearhead the industry with their Barlow Ridge Thunderbolt 5/USB4 v2 controller, ASMedia's controller is poised to end up in a far larger range of devices. So the importance of the company's USB4 v2 PHY demo is hard to overstate.

Demos aside, ASMedia is hoping to tape the chip out soon. If all goes well, the company expects their first USB4 v2 controllers to hit the market some time in the second half of 2025.

One of Qualcomm's indisputable strengths are its 5G modems – something which even Apple has yet to successfully ween itself from. And while Qualcomm is not integrating a modem into its first-generation Oryon-based Snapdragon X chips, the company is still looking to leverage that technology advantage via discrete modems that can be installed in Snapdragon X laptops.

To date, Qualcomm has won 23 laptop designs with its Snapdragon X Elite SoCs, and all of the leading PC vendors have introduced systems based on Qualcomm's Snapdragon X Elite processors. However, only some of them will be equipped with modems, the company detailed at Computex 2024.

While the technical rationale for this is very straightforward (not every vendor wants to buy and dedicate the space to modems), it's still a bit of a surprise in as much as Qualcomm has traditionally heavily pushed laptop vendors to include their modems. But as the Snapdragon X has entered the picture, the joint Qualcomm/Microsoft always connected PC (ACPC) initiative is taking a back seat – meaning modems are no longer being pushed nearly as hard. In its place, the two companies have pivoted hard to equipping mainstream systems with the hardware needed for local AI processing (i.e. NPUs), and with it, Microsoft's Copilot+ PC branding.

Laptop manufacturers, in the meantime, are breathing a sigh of relief, as this switch to emphasizing AI comes at a much lower hardware cost, since vendors don't need to buy additional discrete hardware. Qualcomm for its part has never fully disclosed the full cost of including a Snapdragon modem with a laptop, but the total cost adds up quickly. Besides buying a discrete modem, device manufacturers also need to buy and integrate a 5G-capable radio frequency front end module (RF FEM), as well as the all-important antenna. And mmWave support of any kind can add another wrinkle, as multiple antennas at different orientations are needed to get the best results.

And while not said out-loud, Qualcomm's premium positioning strategy for 8cx-based laptops has not paid significant dividends. Snapdragon X laptops are being priced much more competitively, as Qualcomm is aiming to capture a meaningful share of the PC market – and high-cost features like modems would drive up the final price tag.

Still, virtually all Qualcomm representatives I talked to at Computex were happy to argue that an integrated modem is a huge benefit for their PCs, as they can get fast connectivity almost everywhere in the world instantly and not depend on Wi-Fi or even their smartphones. So the dream of widespread 5G-capable laptops is not dead at Qualcomm; it may just be delayed. In the meantime, for laptop buyers that do need or want a 5G modem, there will still be at least a few premium laptop models on store shelves with the necessary hardware.

Trade shows like Computex always bring out their fair share of oddities, and this year was no exception, with one of the highlights being a Corsair PC case with no fewer than 24 fans.

As one of a handful of companies offering really big desktop PC cases, Corsair was demonstrating its new creature: the 9000D Airflow. At 90 liters in volume – which is twice the size of regular PCs and 1.5x the volume of a typical car gas tank – the colossal case is bigger than ever. It's so big, in fact, that it can house two systems: a full-size ATX (or smaller) system, as well as a separate Mini-ITX system.

The most eye-catching aspect of this PC case (besides its large size, of course) is that it can house as many as 22 fans in addition to two liquid cooling systems. As the name of the 9000D Airflow implies, all of those fans are meant to create as much airflow as possible. And yet, because there are so many fans inside, they do not have to run at a high RPM to move the requisite amount of air, so the 9000D Airflow is quieter than its size otherwise lets on.

To simplify installation of all these fans, the chassis consists of adjustable mounting points on a sliding rail, making the case versatile for any build requirements. The 9000D includes two InfiniRail systems, one at the top (holding six fans) and one at the front, each capable of holding up to eight 120mm fans. Adding fans on the sides and rear increases the total to 24. For those using 140mm or 200mm fans, the InfiniRails can be adjusted by unscrewing and repositioning them based on marked guidelines, allowing for a customized setup despite fitting fewer larger fans. The flexibility of the InfiniRail system enables unique fan placement, enabling the freedom to tailor the cooling configuration to specific needs.

The case design also includes 30mm of clearance behind the motherboard for efficient cable management, making it well-suited for creating clean, organized, and powerful builds.

Besides its may fans, the 9000D Airflow also offers 11 drive bays, plenty of front I/O ports (four USB Type-A, two USB Type-C, audio connectors) with RGB lighting controlled through the iCue Link system. The spacious design allows for comprehensive component compatibility and expansion.

Corsair's Airflow 9000D will be available later this year.

Although consumer SSDs based on Phison's PS5026-E26 controller have been on the market for almost a year and a half now, the class-leading drives still carry a distinct price premium, and to some degree that's because it's still one of the only options for a PCIe 5.0 SSD. But it looks like the situation is going to change in the coming quarters, as SSDs based on Phison's PS5031-E31T controller are incoming, with at least one Phison customer demoing an E31T drive on the Computex show floor.

Phison's PS5031-E31T controller uses two Arm Cortex-R5 cores accelerated by the Andes N25 CoXProcessor, just like its bigger brother Phison PS5026-E26. But this is where their major hardware similarities seem to end. The new E31T controller is a DRAM-less controller with four NAND channels (16 CE targets) that is produced on one of TSMC's 7nm processes, whereas the E26 is an eight-channel controller made on TSMC's 12nm production node.

Besides cutting down on memory channels and the use of DRAM to drive down costs, the E31T also picks up a couple of new tricks by virtue of being nearly two years newer. In particular, the E31T sports Phison's 7th Generation LDPC error correction technology, as opposed to the E25's 5th-gen LDPC

Phison NVMe SSD Controller Comparison
	E31T	E27T	E21T	E26	E18
Market Segment	Mainstream Consumer			High-End Consumer
Manufacturing Process	7nm	12nm	12nm	12nm	12nm
CPU Cores	2x Cortex R5	1x Cortex R5	1x Cortex R5	2x Cortex R5	3x Cortex R5
Error Correction	7th Gen LDPC	5th Gen LDPC	4th Gen LDPC	5th Gen LDPC	4th Gen LDPC
DRAM	No	No	No	DDR4, LPDDR4	DDR4
Host Interface	PCIe 5.0 x4	PCIe 4.0 x4	PCIe 4.0 x4	PCIe 5.0 x4	PCIe 4.0 x4
NVMe Version	NVMe 2.0	NVMe 2.0	NVMe 1.4	NVMe 2.0	NVMe 1.4
NAND Channels, Interface Speed	4 ch, 3600 MT/s	4 ch, 3600 MT/s	4 ch, 1600 MT/s	8 ch, 2400 MT/s	8 ch, 1600 MT/s
Max Capacity	8 TB	8 TB	4 TB	8 TB	8 TB
Sequential Read	10.8 GB/s	7.4 GB/s	5.0 GB/s	14 GB/s	7.4 GB/s
Sequential Write	10.8 GB/s	6.7 GB/s	4.5 GB/s	11.8 GB/s	7.0 GB/s
4KB Random Read IOPS	1500k	1200k	780k	1500k	1000k
4KB Random Write IOPS	1500k	1200k	800k	2000k	1000k

Phison itself calls its E31T platform 'the first mainstream 10 GB/s platform,' which pretty much gives a performance indicator for some of the upcoming inexpensive PCIe Gen5 SSDs. As for random performance, we are talking about 1.5M IOPS per second, which is in line with performance offered by some of enterprise-grade PCIe Gen4 SSDs.

Meanwhile, a Silicon Power ad at Computex indicates that that the company at least hopes to get to 12 GB/sec with its "US85" drive. Which at 4 NAND channels would require pairing up the controller with cutting-edge 3200 MT/sec NAND. It's a bit of an odd pairing given the mainstream, cost-conscious status of the E31T controller, but then the savings on the controller supporting DRAM can be invested back into the NAND on the drive itself. And more importantly, with 12 GB/sec reads and writes, Silicon Power's US85 SSD will be able to compete against earlier E26-based drives that are still being sold on the market, which brings this product to a whole new level. Though Silicon Power will have to be able to procure enough fast 3D NAND to meet demand.

Both Phison and Silicon Power expect E31T to start shipping in Q4, so by the start of next year, mainstream SSDs should be receiving a nice speed boost.

Update 06/13: SK hynix has sent a note to AnandTech clarifying that the company "plans to start mass production of GDDR7 in the fourth quarter of this year when the relevant market opens up." This article has been updated accordingly.

Being a major JEDEC memory standard, GDDR7 is slated to be produced by all three of the Big Three memory manufacturers. But it seems that not all three vendors will be kicking off mass production at the same time.

SK hynix was at this year's Computex trade show, showing off their full lineup of memory technologies – including, of course, GDDR7. SK hynix is the last of the major memory vendor's we've seen promoting their memory, and fittingly, they seem to be the last in terms of their mass production schedule. According to company representatives, the firm will kick off mass production of their GDDR7 chips in the last quarter of 2024.

Comparatively, the company's cross-town rival, Samsung, is already sampling memory with the goal of getting it out the door in 2024. And Micron has been rather gung ho about not only starting mass production this year, but starting it early enough that at least some of their customers will be able to ship finished products this year.

That said, it bears mentioning that with industry-standard memory technologies, mass production at one vendor does not indicate that another is late; it is just indicating that someone was first to validate with a partner and that partner plans to ship its product in 2024. And while mass production remains another 4+ months out, SK hynix does have sample chips for its partners to test right now, and the chips have been demonstrated at Computex.

As far as SK hynix's floor booth at Computex 2024 is concerned, the company had GDDR7 chips on display along with a table essentially summarizing the company's roadmap. For now, SK hynix is planning on both 16Gbit and 24Gbit chips, with data transfer rates of up to 40 GT/s. Though when SK hynix intends to launch their higher-end configurations remains to be seen. Both of the company's rivals are starting out with 16Gbit chips running at 32 GT/sec, so being the first to get a faster/larger chip out would be a feather in SK hynix's cap.

Enermax, established in 1990, is a renowned Taiwanese company in the PC hardware industry, particularly recognized for its innovative power supply units (PSUs), cooling solutions, and PC cases. Over the years, Enermax has built a reputation for engineering reliable, high-performance PSUs that primarily target enthusiasts and professional users. Their commitment to quality and technological advancement has kept them at the forefront of the industry, constantly evolving to meet the demands of the ever-changing tech landscape.

In this review, we are looking at the latest addition to Enermax's impressive PSU lineup: the Revolution D.F. 12 750W PSU. This is the second ATX 3.1-compliant power supply to arrive in our labs, and, broadly speaking, illustrates how we're approaching the inflection point for PSU vendors to update their designs for Intel's revised PSU standard.

As for the Revolution D.F. 12 itself, Enermax's new PSU pursues a balanced design, meeting modern gaming PCs mid-way with good conversion efficiency and an overall robust power delivery system. With features like fully modular cables with per-wire sleeving, a dynamic hybrid fan control for optimal cooling, and advanced topologies, the Revolution D.F. 12 750W is primed to deliver on both reliability and performance. We will delve into its specifications, build quality, and performance metrics to see if the new unit lives up to Enermax's esteemed legacy.

UPDATE 6/12: Micron notified us that it expects its HBM market share to rise to mid-20% in the middle of calendar 2025, not in the middle of fiscal 2025.

For Computex week, Micron was at the show in force in order to talk about its latest products across the memory spectrum. The biggest news for the memory company was that it has kicked-off sampling of it's next-gen GDDR7 memory, which is expected to start showing up in finished products later this year and was being demoed on the show floor. Meanwhile, the company is also eyeing taking a much larger piece of the other pillar of the high-performance memory market – High Bandwidth Memory – with aims of capturing around 25% of the premium HBM market.

GDDR7 to Hit the Market Later This Year

Micron's first GDDR7 chip is a 16 Gb memory device with a 32 GT/sec (32Gbps/pin) transfer rate, which is significantly faster than contemporary GDDR6/GDDR6X. As outlined with JEDEC's announcement of GDDR7 earlier this year, the latest iteration of the high-performance memory technology is slated to improve on both memory bandwidth and capacity, with bandwidths starting at 32 GT/sec and potentially climbing another 50% higher to 48 GT/sec by the time the technology reaches its apex. And while the first chips are starting off at the same 2GByte (16Gbit) capacity as today's GDDR6(X) chips, the standard itself defines capacities as high as 64Gbit.

Of particular note, GDDR7 brings with it the switch to PAM3 (3-state) signal encoding, moving from the industry's long-held NRZ (2-state) signaling. As Micron was responsible for the bespoke GDDR6X technology, which was the first major DRAM spec to use PAM signaling (in its case, 4-state PAM4), Micron reckons they have a leg-up with GDDR7 development, as they're already familiar with working with PAM.

The GDDR7 transition also brings with it a change in how chips are organized, with the standard 32-bit wide chip now split up into four 8-bit sub-channels. And, like most other contemporary memory standards, GDDR7 is adding on-die ECC support to hold the line on chip reliability (though as always, we should note that on-die ECC isn't meant to be a replacement for full, multi-chip ECC). The standard also implements some other RAS features such as error checking and scrubbing, which although are not germane to gaming, will be a big deal for compute/AI use cases.

The added complexity of GDDR7 means that the pin count is once again increasing as well, with the new standard adding a further 86 pins to accommodate the data transfer and power delivery changes, bringing it to a total of 266 pins. With that said, the actual package size is remaining unchanged from GDDR5/GDDR6, maintaining that familiar 14mm x 12mm package. Memory manufacturers are instead using smaller diameter balls, as well as decreasing the pitch between the individual solder balls – going from GDDR6's 0.75mm x 0.75mm pitch to a slightly shorter 0.75mm x 0.73mm pitch. This allows the same package to fit in another 5 rows of contacts.

As for Micron's own production plans, the company is using its latest 1-beta (1β) fabrication process. While the major memory manufacturers don't readily publish the physical parameters of their processes these days, Micron believes that they have the edge on density with 1β, and consequently will be producing the densest GDDR7 at launch. And, while more nebulous, the company company believes that 1β will give them an edge in power efficiency as well.

Micron says that the first devices incorporating GDDR7 will be available this year. And while video card vendors remain a major consumer of GDDR memory, in 2024 the AI accelerator market should not be overlooked. With AI accelerators still bottlenecked by memory capacity and bandwidth, GDDR7 is expected to pair very well with inference accelerators, which need a more cost-effective option than HBM.

Micron Hopes to Get to Mid-20% HBM Market Share with HBM3E

Speaking of HBM, Micron was the first company to formally announce its HBM3E memory last year, and it was among the first to start its volume shipments earlier this year. For now, Micron commands a 'mid-single digit' share of this lucrative market, but the company has said that it plans to rapidly gain share. If all goes well, by the middle of its calendar 2025 (i.e., the end of calendar Q1 2025) the company hopes to capture a mid-20% share of the HBM market.

"As we go into fiscal year 2025, we expect our share of HBM to be very similar to our overall share on general DRAM market," said Praveen Vaidyanathan, vice president and general manager of the Compute Products Group at Micron. "So, I would say mid-20%. […] We believe we have a very strong product as [we see] a lot of interest from various GPU and ASIC vendors, and continuing to engage with customers […] for the next, say 12 to 15 months."

When asked whether Micron can accelerate output of HBM3E at such a rapid pace in terms of manufacturing capacity, Vaidyanathan responded that the company has a roadmap for capacity expansion and that the company would meet the demand for its HBM3E products. 

During Computex 2024, ASRock held an event to unveil some of its upcoming X870E motherboards, designed for AMD's Zen 5-based Ryzen 9000 series processors. ASRock's announcement includes a pair of Taichi-branded boards, the X870E Taichi and the lighter X870E Taichi lite, which uses AMD's X870E (Promontory 21) chipset for AM5.

The current flagship model announced from ASRock's X870E line-up for Ryzen 9000 is the ASRock X870E Taichi. ASRock is advertising a large 27-phase power delivery through 110A SPS, suggesting this board is designed for overclockers and all-around power users. Two PCIe 5.0 x16 slots (operating in either x16/x0 or x8/x8) provide high-speed bandwidth for cutting-edge graphics cards and other devices. Meanwhile, ASRock has gone with 4 DIMM slots on this board, so system builders will be able to max out the board's memory capacity at the cost of bandwidth.

The storage offering is impressive; besides the obligatory PCIe Gen5 x4 M.2 slot (Blazing M.2), ASRock has outfit the board with another three PCIe Gen4 x4 (Hyper) M.2 slots. Also present are two USB4 Type-C ports for high-bandwidth external I/O, while networking support is a solid pairing of a discrete Wi-Fi 7 controller with a Realtek 5Gb Ethernet controller (and the first AM5 board we've come across with something faster than a 2.5GbE controller).

The audio setup includes a Realtek ALC4082 codec and ESS SABRE9218 DAC supporting high-fidelity sound. The BIOS flashback feature is also a nice touch, and we believe this should be a feature on all mid-range to high-end motherboards, which provides an easy way to update the firmware without installing a CPU. And, as no high-end board would be complete without it, ASRock has put RGB lighting on the X870E Taichi as well.

Ultimately, as ASRock's high-end X870E board, the X870E Taichi comes with pretty much every last cutting-edge technology that ASRock can fit on the board.

Comparatively, the ASRock X870E Taichi Lite is a more streamlined and functional version of the X870E Taichi. The Lite retaining all of the latter's key features, including the 27-phase power delivery with 110A smart power stages, dual PCIe 5.0 x16 slots operating at x16 or x8/x8, four DDR5 DIMM slots, and four M.2 slots (1x Gen5 + 3x Gen4). The only significant difference is aesthetics: the Taichi Lite features a simpler silver-themed design without the RGB lighting, while the standard Taichi has a more intricate gold-accented and fanciful aesthetics.

In terms of availability, ASRock is not disclosing a release date for the board at the show. And, checking around with other tech journalists, Andreas Schilling from HawrdwareLUXX has heard that X870E and X870 motherboards aren't expected to be available in time for the Ryzen 9000 series launch. We will investigate this and contact the motherboard vendors to confirm the situation. Though as X870E/X870 boards barely differ from the current crop of X670E/B650E boards to begin with, the Ryzen 9000 series won't be fazed by a lack of slightly newer motherboards.

With the rise of the handheld gaming PC market, we've seen PC vendors and their partners toy with a number of tricks and tweaks to improve improve framerates in games, with some of their latest efforts on display at this year's Computex trade show. Perhaps the most interesting find thus far comes from ADATA sub-brand XPG, who is demoing their prototype "Nia" handheld PC, which uses eye tracking and dynamic foveated rendering to further improve their rendering performance.

For those unfamiliar, dynamic foveated rendering is a graphics technique that is sometimes used to boost performance in virtual reality (VR) and augmented reality (AR) applications by taking advantage of how human vision works. Typically, humans can only perceive detailed imagery in the relatively small central area of our vision called the fovea, while our peripheral vision is much less detailed. Dynamic foveated rendering, in turn, exploits this by using real-time eye tracking to determine where the user is looking, and then rendering just that area in high/full resolution, while rendering the peripheral areas in lower resolution. The net result is that only a fraction of the screen is rendered at full detail, which cuts down on the total amount of rendering work required and boosting framerates on performance-limited devices.

As stated before, this technology is sometimes used in high-end AR/VR headsets, where high resolution displays are placed mere inches from one's face. This ends up being an ideal use case for the technique, since at those distances, only a small fraction of the screen is within the fovea.

Using dynamic foveated rendering for a handheld, on the other hand, is a more novel application. All of the same visual principles apply, but the resolutions at play are lower, and the screen is farther from the users' eyes. This makes a handheld device a less ideal use case, at least on paper, as a larger portion of the screen is going to be in the fovea, and thus will need to be rendered at full resolution. None the less, it will be interesting to see how XPG's efforts pan out, and if dynamic foveated rendering is beneficial enough for handheld PCs. As we sometimes see with trade show demos, not everything makes it out of the prototype stage.

According to a press release put out by ADATA ahead of the trade show, the eye tracking technology is being provided by AMD collaborator Eyeware. Notably, their software-based approach runs on top of standard webcams, rather than requiring IR cameras. So the camera hardware itself should be pretty straight-forward.

Foveated rendering aside, XPG is making sure that the Nia won't be a one-trick pony. The handheld's other major claim to fame is its hardware swappability. The prototype handheld not only features a removable M.2-2230 SSD, but the company is also taking advantage of the recently-introduced LPCAMM2 memory module standard to introduce removable DRAM. Via a hatch in the back of the handheld, device owners would be able to swap out LPCAMM2 LPDDR5X modules for higher capacity versions. This would give the handheld an additional degree of future-proofness over current handhelds, which use non-replaceable soldered-down memory.

Rounding out the package, the current prototype is based on an AMD's Zen 4 Phoenix APU, which is used across both of the company's current mobile lines (Ryzen Mobile 7000/8000 and Ryzen Z1). Meanwhile, the unit's display is adjustable, allowing it to be angled away from the body of the handheld.

Assuming all goes well with the prototype, XPG aims to release a finished product in 2025.

Ultra-high performance memory modules are a staple of of Computex, and it looks like this year G.Skill is showing off the highest performance dual-channel memory module kit to date. The company is demonstrating a DDR5 kit capable of 10,600 MT/s data transfer rate, which is a considerably higher speed compared to memory modules available today.

The dual-channel kit that G.Skill is demonstrating is a 32 GB Trident Z5 RGB kit that uses cherry-picked DDR5 memory devices and which can work in a DDR5-10600 mode with CL56 62-62-126 timings at voltages that are way higher than standard. The demoed DIMMs are running the whole day in a fairly warm room, though it does not really run demanding applications or stress tests.

Traditionally, memory module makers like G.Skill use Intel processors to demonstrate their highest-performing kits. But with the DDR5-10600 kit, the company uses AMD's Ryzen 5 8500G processor, which is a monolithic Zen 4-based APU with integrated graphics that's normally sold for budget systems. The motherboard is a high-end Asus ROG Crosshair X670E Gene and the APU is cooled down using a custom liquid cooling system The Asus ROG Crosshair X670E Gene motherboard has only two memory slots, which greatly helps to enable high data transfer rates, so it is a very good fit for the DDR5-10600 dual-channel kit.

Though I have sincere doubts that someone is going to use an ultra-expensive DDR5-10600 memory kit and related gate with this inexpensive processor, it is interesting (and unexpected) to see an AMD APU as a good fit to demonstrate performance potential of G.Skill's upcoming modules.

Speaking of availability of G.Skill's DDR5-10600 memory, it does not look like this kit is around the corner. The fastest DDR5 kit that G.Skill has today is its DDR5-8400 offering, so the DDR5-10600 will come to market a few speed bins later as G.Skill certainly needs to test the kit with various CPUs and ensure its stability. 

One other thing to keep in mind is that both AMD and Intel are about to release new desktop processors this year, with the Ryzen 9000-series and Arrow Lake processors respectively. So G.Skill will undoubtedly focus on tuning its DDR5-10600 and other high-end kits primarily with those new CPUs.

During the Intel keynote hosted by CEO Pat Gelsinger, he gave the world a glimpse into the Intel Client roadmap until 2026. Meteor Lake launched last year on that roadmap, and Lunar Lake, which we dived into yesterday as Intel disclosed technical details about the upcoming platform. Pat also presented a wafer on stage, Panther Lake, and he gave some additional information about Intel's forthcoming Panther Lake platform, which is expected in 2025.

We covered Intel's initial announcement about the Panther Lake platform last year. It is set to be Intel's first client platform using its Intel 18A node. Aside from once again affirming that things are on track for a 2026 launch, Pat Gelsinger, Intel's CEO, also confirmed that they will be powering on the first 18A wafer for Panther Lake as early as next week.

Intel CPU Architecture Generations
	Alder/Raptor Lake	Meteor Lake	Lunar Lake	Arrow Lake	Panther Lake
P-Core Architecture	Golden Cove/ Raptor Cove	Redwood Cove	Lion Cove	Lion Cove	Cougar Cove?
E-Core Architecture	Gracemont	Crestmont	Skymont	Crestmont?	Darkmont?
GPU Architecture	Xe-LP	Xe-LPG	Xe2	Xe2?	?
NPU Architecture	N/A	NPU 3720	NPU 4	?	?
Active Tiles	1 (Monolithic)	4	2	4?	?
Manufacturing Processes	Intel 7	Intel 4 + TSMC N6 + TSMC N5	TSMC N3B + TSMC N6	Intel 20A + More	Intel 18A + ?
Segment	Mobile + Desktop	Mobile	LP Mobile	HP Mobile + Desktop	Mobile?
Release Date (OEM)	Q4'2021	Q4'2023	Q3'2024	Q4'2024	2025

One element to consider from last year is that Lunar Lake is built using TSMC, with the Lunar Lake compute tile with Xe2-LPG graphics on TSMC N3B, and the I/O tile on TSMC N6. Pat confirmed on stage that Panther Lake will be on Intel 18A. Still, he didn't confirm whether the chip will be made purely at Intel, or a mix between Intel and external foundries (ala Meteor Lake). Intel has also yet to confirm the CPU cores to be used, but from what our sources tell us, it sounds like it will be the new Cougar Cove and Darkmont cores.

As we head into the second half of 2024 and after Lunar Lake launches, Intel may divulge more information, including the architectural advancements Panther Lake is expected to bring. Until then, we will have to wait and see.

In recent months, Frore Systems has been turning heads with their fanless solid-state air cooler technology. The AirJet, as it's come to be called, was previously shown off at CES this year; and for Computex, the company is back with a fresh round of demonstrations.

For the show, Frore has a number of demonstrations running in a fairly large showroom. The company is looking to address a wide range of products, from tablets to notebooks to small PCs, as well as embedded tablets. But there were two showcases in particular that caught my immediate attention: a Samsung Galaxy Book with and without Frore's AirJet, and an 8 TB Sabrent SSD in an external enclosure.

The Samsung Galaxy Book 2 Pro is an ultra-thin notebook that is normally cooled by a fan that, as argued by Flore, does not do its job properly. According to the company, the stock laptop only has enough cooling capacity to sustain 12W heat/power before it hits Tmax, whereas a retrofitted version with Frore's AirJet installed allows it to hold steady-state operation at 16W – and consequently delivering higher performance. In terms of Cinebench R23 multi-threaded results, we are talking about 5330 points for the modded notebook, versus 4255 for the off-the-shelf Galaxy Book 2 Pro.

The potential use cases for Frore Systems's AirJet solid-state cooling technology do not end with CPUs, either. As mentioned previously, the company is also demonstrating the AirJet Mini on Sabrent 8TB SSDs in Orico external enclosures, showcasing the advantage of the silent active cooler over passive cooling. The passively-cooled drive reached 62°C and leveled out at 1,320 MB/s due to thermal throttling. In contrast, the AirJet-cooled drive maintained a temperature of 42°C and achieved a considerably higher performance of 3,016 MB/s.

According to Frore, this significant improvement in both temperature and performance has already led to one major external SSD vendor adopting AirJet technology to improve the performance of their drives. Unfortunately, Frore isn't naming any names, only stating that it's a "big name."

Now, Frore's AirJet Mini and Mini Slim coolers can dissipate up to 5W of power each, and can be combined in to larger blocks of up to 5 coolers (we are talking about announced solutions, technologically scaling could he higher, but this is an entirely different conversation). So the technology does have some scalability limitations that makes it best-suited for lower-power devices. None the less, removing 25W of thermal energy from a modern laptop without a fan can make a huge difference in the performance of these normally passively-cooled devices.

Of course, the main goal for these Computex demos is far more than just showing off AirJets to the public; what Frore would really like to do is to land a deal for its solid-state cooling solution with a major PC vendor (e.g., Apple, Samsung, etc.). Though to do that, Frore has to pass qualification tests and ensure availability of its products, which is something the company says it's currently working on. Meanwhile, from performance point of view, especially given their dimensions, AirJets look very impressive.

Intel this morning is lifting the lid on some of the finer architectural and technical details about its upcoming Lunar Lake SoC – the chip that will be the next generation of Core Ultra mobile processors. Once again holding one of their increasingly regular Tech Tour events for media and analysts, Intel this time set up shop in Taipei just before the beginning of Computex 2024. During the Tech Tour, Intel disclosed numerous facets of Lunar Lake, including their new P-Core design codenamed Lion Cove and a new wave of E-cores that are a bit more like Meteor Lake's pioneering Low Power Island E-Cores. Also disclosed was the Intel NPU 4, which Intel claims delivers up to 48 TOPS, surpassing Microsoft's Copilot+ requirements for the new age of AI PCs.

Intel's Lunar Lake represents a strategic evolution in their mobile SoC lineup, building on their Meteor Lake launch last year, focusing on enhancing power efficiency and optimizing performance across the board. Lunar Lake dynamically allocates tasks to efficient cores (E-cores) or performance cores (P-cores) based on workload demands by leveraging advanced scheduling mechanisms, which are assigned to ensure optimal power usage and performance. Still, once again, Intel Thread Director, along with Windows 11, plays a pivotal role in this process, guiding the OS scheduler to make real-time adjustments that balance efficiency with computational power depending on the intensity of the workload.

Closing out the last of the major PC-focused keynotes at Computex 2024 this evening, we have Intel. The long-reigning leader of the PC CPU market, Intel is in the middle of executing its plans to get back on track on both the manufacturing and chip design aspects of the business. Tonight’s keynote, being helmed by the highly-animated Pat Gelsinger, is titled “Bringing AI Everywhere.” And, like so many other Computex presentations and announcements this week, AI hardware is going to play a big part, as Intel outlines a full stack of products for client and server computing.

Of the four great PC chip vendors at the show, Intel has been the most up-front about what to expect from their hour-long presentation. The company’s Computex 2024 page already outlines their four major topics: AI PCs, Xeon 6 Processors, Gaudi AI accelerators, and Intel’s OpenVINO software ecosystem.

On the consumer hardware front, the company set the table with a significant teaser earlier this month about their forthcoming mobile PC SoC, Lunar Lake. The next generation of Core Ultra processors, Intel is touting significant energy efficiency gains for the mobile-focused chip, with new architectures driving their Performance and Efficiency CPU cores, Xe2 GPU, and a much faster 45+ TOPS (INT8) NPU. While the Lunar Lake announcement is coming relatively soon after the Meteor Lake launch, Intel has made it clear that it’s not going to hold back on shipping future products; they are looking to make up for lost time. Still, Lunar Lake devices are not expected to hit retail shelves until Q4 of this year, so this announcement is coming months in advance of the hardware itself.

On the server front, Intel has been publicly prepping for the launch of a new generation of Xeons with the Xeon 6 platform. The most notable part of this being the release of the company’s first Efficiency-core Xeon, Sierra Forest. Sierra Forest is set to be the first Xeon 6 chip out the door this year, and will offer up to 288 E cores on a single chip, allowing Intel to tap into the many (many) core CPU markets that AMD and Arm-based rivals have been unopposed at thus far.

Finally, the company has fully pivoted its server AI accelerator strategy to its Gaudi accelerators. Gaudi 3 was introduced back in April, and while it isn’t expected to go toe-to-toe with NVIDIA’s top accelerators in every workload, Intel is betting that they can beat NVIDIA on critical workloads, all while undercutting them significantly in pricing. The first Gaudi 3 parts are set to be released in the second half of this year, so hopefully we’ll be hearing a bit more about Intel’s plans as part of their keynote.

As always, the AnandTech crew is live and in person to catch this final Computex keynote. So please come join us at 8:00pm PT / 11:00pm ET / 03:00 UTC to get all the details.

With AMD’s Zen 5 CPU architecture only a month away from its first product releases, the new CPU architecture was placed front and center for AMD’s prime Computex 2024 keynote. Outlining how Zen 5 will lead to improved products across AMD’s entire portfolio, the company laid out their product plans for the full triad: mobile, desktop, and servers. And while server chips will be the last parts to be released, AMD also saved the best for last by showcasing a 192 core EPYC “Turin” chip.

Turin is the catch-all codename for AMD’s Zen 5-based EPYC server processors – what will presumably be the EPYC 9005 series. The company has previously disclosed the name in earnings calls and other investor functions, outlining that the chip was already sampling to customers and that the silicon was “looking great.”

The Computex reveal, in turn, is the first time that the silicon has been shown off to the public. And with it, we’ve received the first official confirmation of the chip’s specifications. With SKUs up to 192 CPU cores, it’s going to be a monster of an x86 CPU.

AMD EPYC CPU Generations
AnandTech	EPYC 5th Gen (Turin, Z5c)	EPYC 9704 (Bergamo)	EPYC 9004 (Genoa)	EPYC 7003 (Milan)
CPU Architecture	Zen 5c	Zen 4c	Zen 4	Zen 3
Max CPU Cores	192	128	96	64
Memory Channels	12 x DDR5	12 x DDR5	12 x DDR5	8 x DDR4
PCIe Lanes	128 x 5.0	128 x 5.0	128 x 5.0	128 x 4.0
L3 Cache	?	256MB	384MB	256MB
Max TDP	360W?	360W	400W	280W
Socket	SP5	SP5	SP5	SP3
Manufacturing Process	CCD: TSMC N3 IOD:TSMC N6	CCD: TSMC N5 IOD: TSMC N6	CCD: TSMC N5 IOD: TSMC N6	CCD: TSMC N7 IOD: GloFo 14nm
Release Date	H2'2024	06/2023	11/2022	03/2021

Though only a brief tease, AMD’s Turin showcase did confirm a few, long-suspected details about the platform. AMD will once again be using their socket SP5 platform for Turin processors, which means the chips are drop-in compatible with EPYC 9004 Genoa (and Bergamo). The reuse of SP5 means that customers and server vendors can immediately swap out chips without having to build/deploy whole new systems. It also means that Turin will have the same base memory and I/O options as the EPYC 9004 series: 12 channels of DDR5 memory, and 128 PCIe 5.0 lanes.

In terms of power consumption, existing SP5 processors top out at 400 Watts, and we’d expect the same for these new, socket-compatible chips.

As for the Turin chip itself, while AMD is not going into further detail on its configuration, all signs point to this being a Zen 5c configuration – that is, built using CCDs designed around AMD’s compact Zen 5 core configuration. This would make the Turin chip on display the successor to Bergamo (EPYC 9704), which was AMD’s first compact core server processor, using Zen 4c cores. AMD’s compact CPU cores generally trade off per-core performance in favor of allowing more CPU cores overall, with lower clockspeed limits (by design) and less cache memory throughout the chip.

According to AMD, the CCDs on this chip were fabbed on a 3nm process (undoubtedly TSMC’s), with AMD apparently looking to take advantage of the densest process available in order to maximize the number of CPU cores the can place on a single chip. Even then, the CCDs featured here are quite sizable, and while we’re waiting for official die size numbers, it would come as no surprise if Zen 5’s higher transistor count more than offset the space savings of moving to 3nm. Still, AMD has been able to squeeze 12 CCDs on to the chip – 4 more than Bergamo – which is what’s allowing them to offer 192 CPU cores instead of 128 as in the last generation.

Meanwhile, the IOD is confirmed to be produced on 6nm. Judging from that fact, the pictures, and what AMD’s doing with their Zen 5 desktop products, there is a very good chance that AMD is using either the same or a very similar IOD as on Genoa/Bergamo. Which goes hand-in-hand with the socket/platform at the other end of the chip staying the same.

AMD’s brief teaser did not discuss at all any other Turin configurations. So there is nothing else official to share about Turin chips built using full-sized Zen 5 CPU cores. With that said, we know that the full-fat cores going into the Ryzen 9000 desktop series pack 8 cores to a CCD and are being fabbed on a 4nm process – not 3nm – so that strongly implies that EPYC Zen 5 CCDs will be the same. Which, if that pans out, means that Turin chips using high performance cores will max out at 96 cores, the same as Genoa.

Hardware configurations aside, AMD also showcased a couple of benchmarks, pitting the new EPYC chips against Intel’s Xeons. As you’d expect in a keynote teaser, AMD was winning handily. Though it is interesting to note that the chips benchmarked were all 128 core Turins, rather than on the 192 core model being shown off today.

AMD will be shipping EPYC Turin in the second half of this year. More details on the chips and configurations will follow once AMD gets closer to the EPYC launch.