• Our new 3U PC is purpose-built for PCI expansion

    Updated , by Geoff Undrell

    After our announcement back in July that the new 3U PC was on its way, we can officially confirm that it is now available to purchase. We’re thrilled to finally welcome the unit into our product family, where it has taken its rightful place as the first of our 3U units.

    The perfect PC for PCI expansion

    3U PC 3 Dual PCIeWe’ve designed the 3U PC to be as versatile as possible, meaning it’s ideal for PCIe expansion. And because our units are used for all kinds of different applications across various industries, we’ve allowed for even greater flexibility by offering two variants of the unit, with horizontal and vertical PCIe slots respectively. This makes it so much easier to enjoy the benefit of multiple graphics (GFX) cards inside a single rack mount PC.

    Our horizontal model rises the PCI card through 90 degrees, enabling 2x consumer GFX cards 3U Redundant PSU 7 PCIto fit comfortably into the machine thanks to 5x full height single bays. Importantly, any GFX cards with power connectors built into the top edge, like NVIDIA GTX, can be used without compromising on the height of the unit – something which is typically difficult to achieve in 19” Rack units. By comparison, the vertical model of the 3U PC offers 7x full height single bays, allowing up to 3x professional GPUs to be fitted, whether Quadro or AMD.

    Here’s what you can expect from our latest PC:

    • 448(W)x457(D)x133(H)mm
    • 2x 3.5″ or up to 6x 2.5″ hot swappable SSD/HDD bays
    • Up to 7x PCIe x16 slots
    • Dual Socket AMD EPYC or Intel Xeon CPU Support
    • Rubber mounted internal cooling fan(s)
    • Optional 800W Redundant Power Supply
    • Optional water-cooling functionality

    Get it while it’s hot!

    We can’t wait to show you what our 3U PC can do. It’s got a load of PCIe potential to look forward to, so if you would like to find out how you can make the most out of your graphics cards by getting your hands on the new unit, get in touch with one of the team.

    Posted by: Geoff Undrell
    Posted in: News, Products

  • AMD Ryzen 9 3900X

    AMD’s 3900X CPU sets a new standard for performance

    Updated , by Geoff Undrell

    Titan chipmaker AMD’s 3rd generation Ryzen 9 3900X processor has got us very excited indeed. Since its official release on 24th July, the fast and furious CPU has already proven incredibly popular – so much so, in fact, that the overwhelming demand has meant that AMD is currently experiencing a chip shortage. It’s no wonder too – AMD is claiming that the 3900X is the world’s most advanced processor, with more speed, memory, and bandwidth than its predecessors, which is why we can’t wait to get it into our PCs.

    A powerful processor straight out of the box

    The 7nm desktop CPU is part of AMD’s Zen 2 architecture, and boasts 12 cores and 24 threads, seemingly leaps and bounds ahead of Intel’s flagship i9-9900K, which is still using a 14nm process and has 8 cores and 16 threads by comparison. And although the Ryzen 3900X’s boost clock speed is only 4.6GHz, a whole 0.4GHz less than the Intel chip, it seems that this isn’t a massive loss, as its performance is hugely impressive regardless.

    The revolutionary Ryzen chip promises some seriously awesome specs:

    • 12 cores, 24 threads
    • 4.6GHz Boost Clock
    • 3.8GHz Base Clock
    • 64MB L3 cache
    • 105W TDP
    • PCIe Gen 4 support

    The CPU that can handle anything you throw at it

    Thanks to its exceptional multi-core performance, the 3900X seamlessly handles multitasking, making it the ideal chip for demanding media applications such as video production that require a powerful, high-performing processor. It even works better when it’s working harder, needing less power when running intensive tasks than the i9-9900K. It does, though, typically consume more when left to its own devices.

    Coming to G2 products soon

    We’re hoping to see stock restraints for the Ryzen 3900X subside by the 1st week of October, so are very much looking forward to seeing what they can do when paired with our line-up of Rack and Mini PCs. If you would like to find out more about how you can get your hands on AMD’s exciting 3rd gen chip within our machines, please get in touch with the team.

    Posted by: Geoff Undrell
    Posted in: News

  • Power over Ethernet NUC

    Power over Ethernet: The perfect rack mount solution

    Updated , by Geoff Undrell

    As specialists in the design and manufacture of PCs for challenging environments, we understand how difficult many PC installations can be, sometimes stopping a project in its tracks before it’s even begun. And it’s not simply a case of negotiating considerations such as case size and weight – the tidiness of your comms cabinet as well as the setup of a reliable Wi-Fi connection can also prove obstructive depending on your unique circumstances. We believe that a big part of the solution in the future will be making the most of the possibilities presented by Power over Ethernet.

    We’ve recently seen an overwhelming demand from our customers for products that capitalise on Power over Ethernet (PoE), and are therefore delighted to announce that we’ve designed a device to help you get the most out of PoE within selected models of our product range.

    Why Power over Ethernet?

    Power over Ethernet is the ideal solution for complicated PC installations. Seamless yet incredibly powerful, PoE enables both power and data to be transmitted through a single Ethernet cable. For this reason, it’s the perfect fit for challenging and remote installations, particularly in locations where a mains power cable would prove impractical. With less cables to contend with, it makes mounting your G2 units easier and ensures your comms cabinet remains uncluttered. Most helpfully of all, as the Ethernet cable required to run our PoE hardware is usually already installed in many office buildings, PC installation is instantly more cost-effective. And, thanks to its transmission via a wired Ethernet connection, your data stays far safer and connectivity is more stable. More and more often, PoE-powered products are proving themselves as the best possible choice for tricky PC installations. That’s why we’ve introduced PoE support to some of our most popular products.

    Nothing but Ethernet

    Available across both our 1U NUC and Bantam PCs, but soon to extend to the rest of our product line-up, our bespoke Power over Ethernet device enables you to take advantage of all four PoE standards:

    • IEEE 802.3af-2003, Type 1
    • IEEE 802.3at-2009, Type 2 or PoE+
    • IEEE 802.3bt-2018, Type 3
    • IEEE 802.3bt-2018, Type 4

    With all of this captured in a single PC, you could therefore not only benefit from PoE+, which can support devices with up to 30W of power such as VoIP phones, but PoE Types 3 and 4, which support information kiosks, video conferencing, high-performing digital signage, and much more. Conformity with all the prevailing standards means our products offer lots of flexibility when designing solutions using PoE.

    However, what’s important to keep in mind when considering PoE is that it is a developing technology. It’s not possible to use it to simply power any legacy PC, so it’s critical to understand what you’ll be investing in, as well as the practical and commercial implications, before you decide to make the change.

    Our range of PoE-powered PCs were designed specifically with complex installations and all of the necessary considerations in mind, so you’re free to enjoy all of the opportunities Power over Ethernet represents.

    If you would like to learn more about how you can harness the power of PoE in one of our products, or else find out the best way to migrate to PoE-powered hardware, please get in touch with one of the team.

    Posted by: Geoff Undrell
    Posted in: Development Projects, Products, Technical Articles

  • AV Awards Finalist

    AV Awards finalist for the second year running!

    Updated , by Geoff Undrell

    We’re absolutely delighted to have been shortlisted for the AV Awards 2019’s coveted Manufacturer of the Year award! This is our second year running as a finalist, after our 2U Nano was a contender for Production Product of the Year in 2018. It’s brilliant to once again be recognised for the innovation, dedication, and ingenuity that goes into creating our PCs, especially when the other ‘shortlisters’ are such esteemed names in the industry.

    G2 at Judging Day

    For the very first time this year, our submission involved more than a written account of what we do and how we do it. In addition to this, we were selected from more than 70 entries in our category and invited to attend a Judging Day in Canary Wharf. Here, our Managing Director, Chris Hollidge was asked to give a presentation and further demonstrate our credentials as a worthy winner, taking questions from a panel of industry experts along the way. His efforts must have impressed as we’re now one of ten finalists in the running for Manufacturer of the Year.

    It would mean the world to win

    It’s a massive honour for us to simply be shortlisted for this award, particularly in a category with such outstanding nominees, and if we were to win it would be an incredible achievement. We’re really looking forward to the awards night itself, which will take place at Battersea Evolution in London this November, and cannot wait to toast the winner regardless of who takes away the trophy. Best of luck to all finalists – we’re elated to join you!

    Posted by: Geoff Undrell
    Posted in: News

  • PCIe lanes

    A simple guide to understanding PCIe lanes

    Updated , by Geoff Undrell

    For the sake of achieving a more cost-effective solution, it is often the case that the number of PCIe lanes available in given processors is largely overestimated. Importantly, the maximum number of available PCIe lanes varies with different series of chip – for example, Intel’s Desktop ‘i’ CPUs have a maximum of 16 PCIe lanes – so ultimately, the number of PCIe lanes on offer across the board on consumer grade CPUs just isn’t enough. Offering only 16 to 20 lanes at the very most, this would typically only enable you to connect only two graphics cards, which for larger, more complex applications doesn’t cut it.

    The problem with PLX switchers

    Most of the time, when an application requires more than one graphics card, users look for motherboards with lots of PCIe slots, but in these cases manufacturers will turn to a PLX switcher. PLX switchers supposedly resolve the issue by multiplexing the PCIe lanes, essentially switching lanes on and off as needed depending on where a device is detected, so you can use more. This can practically double the number of lanes ‘available’, which sounds ideal. The problem with this is that by constantly switching lanes, a bottleneck is created, which when running at high performance can slow things down considerably. So, while a PLX switcher may allow you to use more lanes, you won’t truly achieve the performance you need from your expansion cards.

    The downside is, without a switch, every device needs its own dedicated connection to the CPU, so it’s easy to run out of available lanes very quickly. So, what’s the solution?

    More lanes, more performance

    The simple answer is that you need more lanes. It’s easy to settle for a lower grade CPU to keep costs down, but if it isn’t achieving everything required of it, there’s really no use in having it at all. By comparison, Intel’s Xeon and X299 workstation processors have a far larger number of PCIe lanes, which can generally be anywhere from 28 to 48 lanes – a vast improvement from consumer CPUs.

    We recently demonstrated this to one of our customers, who needed 20 lanes of data to fit two high-performing graphics cards, as well as a 10GB Ethernet port on their motherboard. We rebuilt the unit to allow for their requirements which has now enabled them to plug in the two cards they needed ready for their application.

    Do your research

    When it comes to PCIe lanes, we would really recommend doing some thorough research into the CPUs you’re looking to invest in to get the most you possibly can out of them. This is especially important due to the fact that some companies now have a tendency to combine both the CPU and chipset PCIe lane counts to promote the highest number, which can be misleading. Just remember to be aware of what you’re buying into.

    If you would like to better understand how to get the most out of your PCIe lanes, or need advice choosing the right chip for your machine get in touch with one of the team.

    Posted by: Geoff Undrell
    Posted in: Technical Articles

  • 3U PC front

    Get ready to meet our brand-new unit: the 3U PC

    Updated , by Geoff Undrell

    We’re excited to officially reveal that we have a brand-new rackmount unit in the works – the 3U PC. The very first 3U computer to join our ever-expanding line-up of Mini and Rack PCs, we’ve created the 3U PC in response to an increasing demand for multi GPU configurations.

    Why are multiple GPU slots so hard to come by?

    The fundamental issue with consumer grade GPUs, while cost effective, is that they come built with power connectors on their top edge. This means they are generally very tall, and difficult to fit into more compact, space-sensitive PCs, restricting the number of graphics cards that can be deployed in a single unit. By comparison, the power connectors on ‘professional’ cards can be found on the back of the GPU rather than the top edge, thereby only affecting case depth, which is a much easier challenge to negotiate.

    Enter the 3U PC

    Our new 3U PC was designed to alleviate the issues associated with multiple GPU slots. As the need for PCs with a smaller height but sufficient room for more expansion cards continues to grow, our 3U case allows professional GPUs to be mounted vertically, making it possible for customers to add several to their unit.

    Available in two variants, the 3U PC takes both pro and consumer cards into consideration. The first variant will offer 7x full height single bays able to fit up to 3x dual-width NVIDIA RTX or AMD professional graphics cards. The second variant rises the bays through 90 degrees, offering 5x full height single bays able to fit up to 2x dual-width NVIDIA RTX/GTX consumer graphics cards.

    3U PC rear alternate variant3U PC rear variant

    But that’s not all there is to get excited about – the 3U PC is bursting with other brilliant features:

    • 425(W) x 547(D) x 133(H) mm
    • 4x 3.5” or 8x 2.5” hot swap bays
    • Dual Socket AMD EPYC or Intel Xeon CPU Support
    • Water-cooling functionality (optional)
    • Redundant power supply (optional)
    • DVD/Blu-ray Bay (Optional)

    Proving three’s not a crowd

    We can confirm that the 3U PC has now been fully designed and is currently awaiting initial production samples, but we don’t anticipate it’ll be very long before the new unit joins our family of compact and rackmount PCs. We anticipate availability in around 5 weeks’ time, so watch this space for any updates on how you can get your hands on one.

    If you would like to learn more about the 3U PC, or any of our other products, please get in touch with one of the team.

    Posted by: Geoff Undrell
    Posted in: News, Products

  • Intel i9-9900KS

    Meet Intel’s powerhouse processor, the i9-9900KS

    Updated , by Geoff Undrell

    Officially announced at Computex 2019, Intel recently unveiled its latest 9th gen processor, the i9-9900KS, and boy, is it a powerhouse. The new chip’s capabilities are already blowing its impressive predecessor, the i9-9900K, out of the water, given the fact that it’s optimised to run every one of its 8 cores at a Turbo frequency of 5GHz straight out of the box. By comparison, Intel’s current flagship i9-9900K can only hit 5GHz on a single core with a Turbo speed that’s a whole 300MHz slower.

    What’s the difference?

    Beyond the startling ability to achieve 5GHz all-core Turbo, with claims that this speed can be reached consistently in any scenario, there isn’t an overwhelming difference between the i9-9900KS and the i9-9900K. A “special edition” variant, the i9-9900KS is set to deliver 5GHz during both single and multi-core workloads, and has a base frequency of 4GHz in contrast to the i9-9900K’s 3.6GHz, but still uses a 14nm++ process as well as the same silicon, only better binned. Maintaining both 16MB of L3 cache and UHD 630 graphics, the new CPU is also ready for the same motherboards as its older cousin, but does include a small firmware update.

    Still speculating

    Intel surprisingly didn’t offer much more information at Computex than had already been previously revealed about the i9-9900KS. We’re still yet to know a definitive price point or its thermal design power (TDP), although based on the performance and power consumption of the i9-9900K, we’re pretty certain that the chip will get seriously hot and require a lot of power. Following our successful implementation of water-cooling functionality in a recent customer project that involved overclocking the i9-9900K, we anticipate that similar measures will need to be employed to ensure the i9-9900KS doesn’t cause your machine to overheat.

    The i9-9900KS is due to launch in Q4 of this year, so we should see more information about it emerging over the next few months. In the meantime, if you would like to learn more about the CPU or when this will become available in our line-up of products, please speak to one of the team.

    Posted by: Geoff Undrell
    Posted in: News