Yuri "1usmus" Bubliy over the weekend released the latest version of DRAM Calculator for Ryzen, the definitive tool for overclocking and optimizing memory on your AMD Ryzen-powered PC. The tool lets you calculate the best possible memory-related settings specific to your machine. Version 1.7.1 introduces several new changes. For starters, it now has updated presets for three of the most popular DDR4 DRAM die types obtained from hundreds of hours of re-testing various brands of memory modules on the latest version of AMD AGESA. These affect everything from procODT to VDDG settings, and the introduction of CL15t, which stabilizes problems encounted in finding timings with CL14t.

The R-XMP button (which attempts to translate your DIMM's XMP to AMD-compatible settings), and "Calculate EXTREME" buttons have been removed, and in their place, the "Calculate FAST" button has been optimized for the best possible stable settings; and "Calculate SAFE" for reasonably fast yet safe settings. DRAM Calculator for Ryzen now has DIMM PCB revision awareness. Several popular memory vendors have multiplie revisions of PCBs for the same DIMMs (and same DRAM chips, which can affect tuning headroom). This provides an additional layer of accuracy in calculating timings. In addition, there are several UI changes, such as a "compare timings" button with added functionality. Preliminary support is also added for future "Zen 3" based processors. Grab the tool from the link below.

DOWNLOAD: DRAM Calculator for Ryzen v1.7.2 by 1usmusThe change-log follows.

Update 09:07 UTC: The DRAM Calculator for Ryzen has just been updated to v1.7.2. Detailed below.

The Ryzen 3 3100 is turning out to be a fun little toy for enthusiasts. PC enthusiast TSAIK succeeded in overclocking it to 5923 MHz under extreme cooling. The chip was fed 1.45 Volts, and put under liquid nitrogen cooling, to achieve the feat. An MSI MAG X570 Tomahawk motherboard and a single stick of 8 GB memory underclocked to DDR4-1600 made the rest of the critical hardware. The feat is the second highest OC record for a "Zen 2" powered processor, next only to TSAIK's own speed record with the flagship Ryzen 9 3950X, which was pushed to 6041 MHz. Find the HWBot submission for the Ryzen 3 3100 speed record here.

MSI in its weekly "MSI Insider" livestream shared fascinating insights into the way Intel appears to be sorting out its "Comet Lake" silicon across the various brand extensions of its 10th generation Core desktop processors. Its tech leads Eric Van Beurden and Michiel Berkhout spoke at length about MSI's own evaluation of the trays of Core i5-10600K/KF, i7-10700K/KF, and i9-10900K/KF chips it received (the only unlocked chips across the lineup), which they used as empirical evidence to suggest a model for Intel's binning.

MSI segregated the chips it received into three categories. Level A consists of chips that overclock higher than Intel's specifications (overclocking headroom higher than expected). Level B consists of those that overclock within Intel's specifications. Level C, on the other hand, have their overclocking headroom fall below Intel's specifications. It's important to note here that "Intel specification" doesn't mean "stock frequencies," it refers to the overclocking headroom Intel communicates to motherboard manufacturers, to give them an idea of the minimum board design requirements needed to guarantee overclocking within these specifications, for their Z490 motherboards. These are more of a guideline in nature, all three levels will overclock above stock frequencies.

Even as review embargoes remain on Intel's latest 10th Gen CPUs, benchmark scores that show what these 14 nm CPUS are capable of are already flooding the web. Case in point: a Cinebench 15 benchmark of Intel's unlocked Core i9-10900K running at an overclocked 5.4 GHz on all cores @ 1.35 V core. The 10-core CPU features a base clockspeed set at 3.7 GHz, so we're looking at a frequency increase of around 46%.

At those speeds, tested on an ASRock Phantom Gaming 4/AX motherboard and 16 GB of G.Skill DDR4-3200 MHz CL14 memory, the Intel Core i9-10900K managed to post a 3002 multi-core score. When at stock, it achieved a relatively paltry 2347 points. An AMD Ryzen 7 3800X CPU (8-core, 16-thread) typically scores around 2200 points, and an AMD Ryzen 9 3900X CPU (12-core, 24-thread) achieves a 3200 score. They do so at stock frequencies, though; and the Intel Core i9-10900K is sandwiched in-between those when it comes to core-count, but not on price: 10 Intel cores will set you back $488.00, while AMD's 8-core launched at $399 (and is now cheaper) and AMD's 12-core CPU launched for $499. Adding to the benchmarking caveat, the operating temperatures for this particular Core i9-10900K show 0º min and 69º max, so assuming the temperature report is correct, it's fair to say an air cooler wasn't used for this overclocking feat.

Historically, Intel has separated its processors and chipsets that accompany them to overclockable and non-overclockable ones. That means that only the "K" CPUs can be overclocked. With the latest generation, only some parts of the lineup are K CPUs, like the Core i9-10900K, i7-10700K, i5-10600K, etc. Those processors could only be overclocked one put in motherboards based on "Z" chipset, like Z390 and Z490. However, it seems like ASRock has developed a new technology that will overclock non-K CPUs on non-Z motherboards, which is quite impressive.

Called the Base Frequency Boost (BFB) technology, it will allow for overclocking the non-K processors on chipsets like B460 and H470. How will that work you might wonder? Well, ASRock will take the TDP of the CPUs and make it run in the PL1 mode, which increases the processor TDP form 65 W and turns it into a 125 W TDP beast. This will, of course, be user selective and case dependent, meaning that if your cooling system can not handle that much heat coming out from the overclocked processors, it is unlikely that they will reach the peak clocks ASRock can target. You can check out the slide below:

ASRock, proudly announces its latest range of motherboards featuring Intel 400-series chipsets and supporting the latest Intel 10th Generation processors with up to 10-cores using the LGA1200 socket.

Whatever the build requirements, with over 30 models available, ASRock has the right motherboard to meet your needs. This includes the flagship Z490 Taichi and Z490 PG Velocita for performance and overclocking enthusiasts, and extreme gamers; popular Z490 Steel Legend, Pro and Phantom Gaming series'; comprehensive mini-ITX options for SFF builds; plus H470, B460 mainstream and H410 entry options.

PowerColor today issued a press release where they present the world their new Red Devil and Red Dragon models of AMD's RX 5600 XT graphics cards. This comes after reports on AMD's move towards increasing TDP, memory and core clockspeeds on their new graphics card so as to better compete with NVIDIA's recently price-cut RTX 2060, which would make it a much better performer than AMD's RX 5600 XT at a slightly higher price ($279 vs $299).

Hence a reported strike back from AMD in increasing performance for their RX 5600 XT with increased power envelope (160 W over 150 W), faster memory (at 14 Gbps instead of the original 12 Gbps) and increased core clocks (1615 MHz gaming and 1750 MHz boost, versus 1375 MHz gaming and 1560 MHz on AMD's CES press-event slides). The change in configuration brought about changes in the card design, with the higher-powered Red Devil coming in with 1x 8-pin and 1x 6-pin power delivery inputs, instead of the firstly developed 8-pin only. The changes have been brought about by a BIOS change, and not all cards will ship with the new specifications. However, PowerColor said that a BIOS update will be made available for customers to get their RX 5600 XT on steroids. Of course, whether or not it should be the onus of users to do such an update (which may risk in bricking their graphics card) is another matter entirely. The press release follows.

Overclocking the memory clock of a GeForce RTX 2080 Ti to 2000 MHz (16 Gbps) isn't difficult, but most custom-design RTX 2080 Ti cards cap out at that, and so the enthusiasts over at TecLab took matters into their own hands by pulling off a daring memory chip replacement mod, by installing 16 Gbps-rated memory chips onto a Galax RTX 2080 Ti HOF graphics card. In a 16-minute video presentation, they detail the process of soldering a component as delicate and sensitive as GDDR6 memory chips, and 45 times over. The team had to sacrifice not one, but two Galax GeForce RTX 2080 Super graphics cards, which feature 16 Gbps-rated memory chips to support the SKU's 15.5 Gbps memory clock. A total of 33 manual solder operations had to be performed (removing the 15 stock chips from the RTX 2080 Ti, removing 11 chips from the two RTX 2080 Super cards, and soldering them onto the RTX 2080 Ti).

The group detailed the process of removing the memory chips under hot air, giving the extracted chips fresh ball-grids, and placing the chips onto the RTX 2080 Ti PCB. No BIOS modding was required, as the RTX 2080 Ti card's video BIOS was able to auto-detect the chips and run them at 14 Gbps. From here on, manual overclocking easily runs the card at 2000 MHz (16 Gbps) memory, with overclocking headroom to spare. The memory clock could now be dialed all the way up to 2150 MHz (17.2 Gbps), something that's close to impossible with 14 Gbps chips. TecLab is calling their creation the world's first RTX 2080 Ti Super, which could very well be true. Last we heard, the RTX 2080 Ti Super could get more CUDA cores, and not just faster memory. Nevertheless, this mod blew our minds, and provides valuable pointers on how to solder dense BGA components without a multi million-dollar placer. We tip our hats to TecLab.Watch the TechLab video presentation here.

If you're one to pay more bucks for the same silicon in a bid to decrease risk of having a lower-performing overclocker than is possible with the latest AMD and Intel chips, this post is for you. Silicon Lottery has announced (absolutely expected) plans to bin AMD and Intel's latest high-performance processors starting this November.

AMD's Ryzen 9 3950X and Intel's Core i9 9900KS will be up for grabs in the website with guaranteed maximum clocks for you to peruse and then seat in your motherboard of choice. Just wait a little while longer for them to become available, since AMD's own Ryzen 9 3950X isn't yet available in the retail channel. Intel's own Core i9 9900KS has just been announced though, with availability just two days from now, on October 30th. So if you want to skip the hassle (or fun, as you see it) of finding just the right settings for your CPU of choice, keep refreshing Silicon Lottery's page. Availability is expected to be extremely limited on either part.

We've known for some time that AMD's mainstream-segment B550 chipset wouldn't bring all the bangs and whistles of its bigger, enthusiast-class cousin X570. For one, it wouldn't make sense to increase development and implementation costs of both the chipset and motherboards built for mainstream enthusiasts by adding PCIe 4.0 support and the more stringent signaling and power requirements the new standard entails. As such, B550 reportedly cuts down fully on PCIe 4.0 support, as well as on the latest USB standards, to offer a product that's sufficiently rounded up on I/O while offering overclocking support for users that demand it.

Reportedly, AMD's B550 will only support up to 2x USB 3.2 Gen2 devices, 6x USB 2.0, 4 + 4 SATA3 connections, and the interlink between the chipset and the CPU occurs via a 4x PCIe 3.0 interface, which means there's less bandwidth for communication between the CPU and the chipset than on X570 - not that that was a real problem on AMD's previous-gen Ryzen products, though, so that's more of a technicality at this point. Ryzen 3000 CPUs still offer 4x PCIe 4.0 ports, though, so these could be used for speeding up a PCIe 4.0 NVMe SSD, for instance. The launch of B550-bound products is expected towards October.

World famous overclocker Der8auer published his survey of boost clocks found on 3rd generation Ryzen CPUs. Collecting data from almost 3,000 entries from people around the world, he has found out that a majority of the 3000 series Ryzen CPUs are not hitting their advertised boost speeds. Perhaps one of the worst results from the entire survey are for the 12-core Ryzen 9 3900X, for which only 5.6% of entries reported have managed to reach the boost speeds AMD advertises. However, the situation is better for lower-end SKUs, with about half of the Ryzen 5 3600 results showing that their CPU is boosting correctly and within advertised numbers.

Der8auer carefully selected the results that went into the survey, where he discarded any numbers that used either specialized cooling like water chillers, Precision Boost Overdrive - PBO or the results which were submitted by "fanboys" who wanted to game the result. Testing was purely scientific using Cinebench R15 and clock speeds were recorded using HWinfo (which got recommendation from AMD), so he could get as precise data as possible.

Users that don't want to play the silicon lottery game have been using services that offer pre-binned and pre-overclocked chips for a while now. Silicon Lottery is one of the most well known players in this game, but German retailer Caseking is now offering the same for AMD's latest Ryzen 3000 processors. AMD's work on automatic overclocking and boost clocks for their Ryzen chips has rendered manual overclocking almost (read: almost) obsolete, and in some cases it may even be detrimental to the CPU's performance to set a manual overclock that overrides AMD's boost clock algorithm. This is because AMD's boost increases speed on a single core, with subsequent cores being clocked slightly lower according to their capabilities. In effect, this means that manually overclocking all cores to, say, 4.0 GHz can sometimes render lower performance in particular tasks, since the all-core overclock is, by necessity, handicapped by the least-overclockable core.

Caseking's offerings have been pre-overclocked, and are guaranteed to hit stable overclocks at the claimed frequency, thus saving users from getting a "bad" overclocker CPU from AMD. Caseking's offerings have been tested by their own King Mod team and overclocking superstar Roman "der8auer" Hartung, with Prime95 26.6 software being used to test the overclocked chips' stability with a FFT length of 1344 for at least one hour. This practice is backed by a two-year limited warranty on the CPU. Sadly, most CPUs are out of stock at the moment, so keep on checking availability, unless one of the offerings is exactly up your alley.

GIGABYTE has prepared its Gaming OC variant of AMD's Radeon RX 5700 XT graphics card. The card features a PCB that is very similar to the reference model with 8+6 pin power connectors. However, it features an updated triple-fan cooler design in a 2.5 slot thick body and standard set of I/O, meaning that three DisplayPorts and one HDMI port are present. The card also has applied factory overclock, but exact details of frequency and card's availability are unknown so far. The AORUS model is yet to be released, nonetheless, we can hope to see it soon as well.

Silicon Lottery, a company specializing in the process called binning which involves testing of CPUs for particular features (overclocking potential in this case), has released its portfolio of 3rd generation of Ryzen CPUs. As of now, they are offering only Ryzen 7 and Ryzen 9 models, covering Ryzen 7 3700X, 3800X and Ryzen 9 3900X. Ryzen 9 3950X is said to be introduced in September and that is the date Silicon Lottery will reveal the information about overclocking potential of that model and frequencies they have achieved. Mid range Ryzen 5 models should be added at later date as well.

GIGABYTE today officially launched its latest addition to the Aorus series of motherboards, made for the new generation of Ryzen 3000 series processors. The "Master" as it is called, is an impressive feat of engineering designed to handle even the most power-hungry Ryzen CPUs like the 16 core Ryzen 9 AMD recently showcased.

For starters, the board is featuring twice the amount of copper wires usually needed to implement a PCIe connection, which means less information loss on PCB. It has a 14 direct phases of Infineon digital IR 3556 PowIRstage MOSFETs VRMs that are capable of delivering 50A each, which means that the VRM is capable of delivering up to 700A of current, providing additional headroom for CPU overclock. To handle the large amount of VRMs effectively, the board is equipped with beefy heatsinks and a heat pipe that has direct contact with VRMs. Sandwiches between the heatsink and the board is a new generation of thermal pads designed by LAIRD, with 1.5 mm thickness and 5 W/mK thermal conductivity.

Intel has launched a software overclocking utility tool for their latest 9th Gen, unlocked CPUs that promises to take the guesswork and BIOS delving out of the overclocking equation. The Intel Performance Maximizer tries to do exactly what the name implies by automagically overclocking your unlocked (read, K-suffix CPU for unlocked multipliers) with no further user intervention needed. Intel describe this tool as the one that "(...) makes it easier than ever to dynamically custom-tune an unlocked Intel processor based upon its individual performance potential."

Intel says they will offer you a one-time, free CPU replacement in addition to your usual 3-year warranty on a company CPU, should anything wrong arise during this method of overclocking. Intel say's this is a way to maximize your CPU performance on a per-sample basis, so results may vary according to your CPU's thermal properties (as we know, some CPUs perform and overclock better than others due to slight variations in the manufacturing process). But if you don't want to get inside your BIOS for a dirty, hands-on approach, you can always use Intel's software, which has been released at a very conspicuous time indeed, considering AMD's Ryzen 3000 series release.

Eliovp, who describes himself on GitHub as a Belgian [crypto] mining enthusiast, created what could go down as the best thing that happened to AMD Radeon users all decade. The AMD Memory Tweak Tool is a Windows and Linux based GUI utility that lets you not just overclock AMD Radeon graphics card memory on the fly, but also lets you tweak its memory timings. Most timings apply live, while your machine is running within Windows/Linux GUI, some require memory retraining via a reboot, which means they can't be changed at this time, because rebooting reverts the timings to default. The author is trying to figure out a way to run memory training at runtime, which would let you change those timings, too, in the future. While you're at it, the tool also lets you play with GPU core frequency and fan-control.

The AMD Memory Tweak tool supports both Windows and Linux (GUI), and works with all recent AMD Radeon GPUs with GDDR5 and HBM2 memory types. It requires Radeon Software Adrenalin 19.4.1 or later in case of Windows, or amdgpu-pro ROCM to be actively handling the GPU in case of Linux. The Linux version further has some dependencies, such as pciutils-dev, libpci-dev, build-essential, and git. The source-code for the utility is up on GitHub for you to inspect and test.

DOWNLOAD: AMD Memory Tweak Tool by Eliovp

TechPowerUp today released GPU-Z 2.20.0 as a very quick follow-up to version 2.19.0 from a few hours ago. We have come across a few critical bugs with the older version that needed immediate fixing. To begin with, your overclock getting reset on NVIDIA graphics cards with Boost when using GPU-Z, has been fixed. A crash noticed on machines with NVIDIA "Pascal" GPUs with no driver loaded, has also been fixed. Crashes noticed on Apple machines (i.e. Windows on Apple) with AMD Radeon Vega 12 GPU have been fixed. We touched up the memory bus-width read-out to show "bit" instead of "Bit," while we were at it. Grab the download from the link below.

DOWNLOAD: TechPowerUp GPU-Z 2.20.0The change-log follows.

The upcoming EVGA Z390 DARK is designed from the ground up to be the world's best overclocking motherboard. Need proof? The EVGA Z390 DARK was recently used by Finland Overclocker, Juhani Luumi AKA LUUMI to set a brand new CINEBENCH 8-Core CPU World Record. Armed with an Intel Core i9-9900K, and EVGA Z390 DARK Motherboard and Liquid Nitrogen Cooling, LUUMI was able to clock his CPU to nearly 7GHz and achieve a score of 3142 cb. A new CINEBENCH 8-Core CPU World Record!

Gigabyte is now readying a new hybrid cooled graphics card, based on NVIDIA's GeForce RTX 2080. Dubbed the Aorus Xtreme WaterForce (GV-N2080AORUSX W-8GC). It uses an all-in-one liquid cooler based design that makes use of a 240 mm radiator. The copper baseplate is quite large and receives heat from the cards VRM via a copper heat pipe, thus cooling all major components. The fans are of course RGB LED-equipped, which goes with the illuminated translucent wing on the shroud. Overall it is an attractive card if you have the extra space for its cooling solution.

This model isn't just a standard offering with a fancy cooler, Gigabyte has equipped it with a sizable overclock, pushing the card to 1890 MHz the same core clocks as the Aorus Xtreme air and the waterblock equipped WaterForce WB. Memory speeds see a speed bump as well of 140 MHz which pushes the overall memory clock speed up to 14140 MHz. The board design is also based on the Aorus Xtreme and Aorus Xtreme Waterforce WB cards and comes with a 12+2 phase design and dual 8-pin power connectors. Display connectivity is also the same as the other models and includes; 3x DisplayPort, 3x HDMI, and a single USB TypeC.

BIOS modder Vipeax has released a special patched version of NVFlash (version 5.527.0), the utility that allows you to extract and flash the video BIOS of your NVIDIA GeForce graphics card. This special version lets you to bypass NVIDIA restrictions and flash GeForce RTX 20-series Founders Edition (FE) graphics cards with BIOS ROMs of custom-design graphics cards. The official versions of NVFlash that support "Turing" GPUs report a "board ID mismatch" error when trying to do this, and an additional CLI parameter that made it ignore this warning, was removed by NVIDIA, effectively walling off Founders Edition cards from BIOS cross-flashing. You still can't flash the card with a BIOS you modified, because of NVIDIA's digital-signature restriction that has been in place since "Pascal," however, this new change could come handy if you want to flash your FE card with the BIOS of a custom-design card that is largely based on NVIDIA's reference-design PCB.

PC enthusiasts look to flash their Founders Edition cards with BIOS ROMs of custom-design graphics cards by other NVIDIA add-in card partners, mainly to increase power limits that allow the GPU to sustain boost frequencies better, and increase overclocking headroom. As an obligatory word of caution, use of NVFlash isn't covered by product warranties, and you use it at your own risk, especially when cross-flashing between cards that might have subtle differences. We manually checked the modified executable (not just Virustotal) and it doesn't contain any malware.DOWNLOAD: NVIDIA NVFlash with Board ID Mismatch Disabled

The first official data we have received about the performance of the new Intel processors are not exactly spectacular. The Core i9-9900K has aroused considerable controversy due to the unfair Principled Technologies test bench. The results have been reviewed and confirm that the performance gain is debatable, but independent analyses have yet to appear for Core i9-9900K, Core i7-9700K and Core i5-9600K processors.

Today we have some info about one of them: a new video in China shows a Core i5-9600K being benchmarked with a MSI Z390 MEG Godlike motherboard with 16GB of DDR4 memory and a Silver Arrow Extreme cooler from Thermalright. We don't have game benchmarks, but at least we have some Cinebench results both with the processor working with its 3.7 GHz base clock and overcloked to 5.2 GHz. That process was done without problems despite using an air cooler.

MSI is working on its next-generation flagship graphics card, the GeForce RTX 2080 Ti Lightning Z (going by previous naming conventions for the Lightning). The card will be positioned above its current Gaming X Trio, as well as the liquid-cooled SeaHawk series, and it's possible there could be a sub-variant with factory-fitted full-coverage water-block. As with previous generation cards that bear the name, the card could have the highest factory overclock in MSI's product stack, the strongest VRM setup that's voltmod-friendly, and a gargantuan cooling solution for its air-cooled sub-variant. There's no word on when MSI could roll the beast out. MSI teased the card with a final slide in one of its media events in China, which depicts lightning bolts captioned "Coming Soon."

Ukrainian PC enthusiast and software developer 1usmus today released DRAM Calculator for Ryzen version 1.4.0. This utility was formerly known as "Ryzen DRAM Calculator," which has since been voluntarily renamed by the author in the interest of avoiding any future trademark conflict with AMD, or giving users the impression that the software has been made by AMD. The change in name doesn't change the fact this could be your go-to app to figure out the best memory settings for your AMD Ryzen-powered machine.

PC enthusiasts usually only remember 4 or 5 DRAM timing settings besides DRAM clock and voltage, letting the motherboard BIOS figure out the rest of the stable values, which could often be looser than needed. DRAM Calculator for Ryzen figures out nearly every under-the-hood timing, voltage, clock-speed, and other setting needed to make the most out of your memory overclock. You can also make the app work out "safe," "stable," and "extreme" variations of its own calculations. Version 1.4.0 isn't just a name-change for the application. It introduces a large number of critical updates to the app that improve accuracy and functionality.

DOWNLOAD: DRAM Calculator for Ryzen v1.4.0The change-log follows.

Intel took their opportunity to add a golden sheen to their new 9th Gen CPUs by going as far as showcasing their extreme overclocking capabilities right from the launch event. partering with world-renowned overclocker Splave, Intel showcased how a Core i9-9900K can withstand up to 1.7V Vcore, with a little aid from yours truly liquid nitrogen.

That scorchingly high Vcore paired with the sub-zero temperatures of LN2 allowed for all CPU cores to be overclocked up to 6.9 GHz, with Splave saying some samples could reach 7.1 GHz across all cores. Intel then went on to show some of the increased performance benchmarks - which, as you'd expect, don't showcase a linear performance improvement with increased frequency. Still, it's an impressive pure voltage and frequency feat. Splave went on to say that these new 9th Gen Intel processors can now achieve some 5.3 GHz under watercooling and at a much more mundane 1.4V Vcore. We'll all be able to test that for ourselves soon now won't we?