As part of the launch of the new Loihi 2 chip, built on a pre-production version of Intel’s 4 process node, the Intel Labs team behind its Neuromorphic efforts reached out for a chance to speak to Mike Davies, the Director of the project. Now it is perhaps no shock that Intel’s neuromorphic efforts have been on my radar for a number of years – as a new paradigm of computing compared to the traditional von Neumann architecture, and one that is meant to mimic brains and take advantages of such designs, if it works well then it has the potential to shake up specific areas of the industry, as well as Intel’s bottom line. Also, given that we’ve never really covered Neuromorphic computing in any serious detail here on AnandTech, it would be a great opportunity to get details on this area of research, as well as the newest hardware, direct from the source.
We’ve been keeping light tabs on Intel’s Neuromorphic efforts ever since it launched its first dedicated 14nm silicon for Neuromorphic Computing, called Loihi, back in early 2018. In an interview with Intel Lab’s Director Dr. Richard Uhlig back in March 2021, I asked about the development of the hardware, and when we might see a second generation. Today is that day, and the group is announcing Loihi 2, a substantial upgrade over the first generation that addresses a lot of the low-hanging fruit from the first design. What is perhaps just as interesting is the process node used: Intel is communicating that Loihi 2 is being built, in silicon today, using a pre-production version of Intel’s first EUV process node, Intel 4.
AMD has announced that it is aiming for a 30x increase in the energy efficiency of its server chips by the time we reach 2025.
In just a four-year span, a thirtyfold efficiency improvement is a seriously ambitious plan for its Epyc processors – the CPUs which go in servers – and AMD GPU accelerators which are used in AI applications and high-performance computing.
If you’re doubtful of AMD’s ability to achieve this, bear in mind that the company’s previous plan to hit 25x energy efficiency, kicked off in 2014, was realized in 2020. In fact, efficiency gains to the tune of over 31x were managed by AMD according to an independent assessment, but this time around, the aim is to score another similar energy efficiency victory in four years rather than six.
AMD admits that it’s not an easy goal to meet by 2025, and that “success is certainly not a forgone conclusion” by any means.
Just to underline what this would mean, a 30x increase with efficiency would be outpacing industry energy efficiency performance by some 150% if you compare to the previous five-year period, AMD believes. Take a little salt here, given that this is Team Red’s own workings and stat crunching – as ever with these kind of PR announcements – but it’s clear enough that the potential energy savings and positive environmental impact could be very substantial.
AMD observes that the possible savings could represent “billions of kilowatt hours of electricity” while “potentially reducing energy use per computation by 97% over the next five years”, no less.
Analysis: Epyc progress is vital for many reasons
This new goal for Epyc and AMD Instinct accelerators is clearly important for sustainability and data centers being more green-friendly in general at a point where climate change targets are becoming an increasing concern.
While this is heavyweight silicon AMD is talking about here, driving improvements in that sphere, and research and development therein, will doubtless have a trickle-down effect to other areas of Team Red’s CPU portfolio like consumer chips.
Indeed, in recent times we’ve already seen AMD take great strides in not just concentrating on upping the performance of its Ryzen CPUs, but also pushing forward with much better energy efficiency with impressive results. So there’s no reason to doubt that this should continue, too, as we plough on through to the midpoint of this decade.
AMD’s next-gen Navi 33 graphics card could sport 4,096 cores (stream processors) according to the latest from the GPU grapevine, which is fewer than the rumor mill previously theorized, plus it may not turn up until 2023 (we’ll come back to that point later).
This is speculation from one of the more regular Twitter hardware leakers, Greymon55, who updated their bio to indicate that figure of 4,096 cores – but with a question mark added, reminding us of how deep this is in rumor territory currently (keep that bagful of condiments handy).
The change is notable because as Wccftech points out, Greymon55 previously suggested that Navi 33 would have more cores than this: 5,120 to be precise. Hence some folks immediate thoughts on this fresh rumor are heading along the lines of Navi 33 – which could possibly be the GPU that powers the RX 7600 series, given that Navi 23 (current-gen) is the midrange RX 6600 – not being quite as powerful as originally anticipated.
Hold your stream processing horses before trying to draw even any vague conclusions here, though, as the overall performance level of Navi 33 is still predicted to exceed the current flagship RX 6900 XT by Greymon55. The leaker further tweets to clarify that Navi 33 was “designed to be better than the 6900 XT” and even if it doesn’t meet that design goal, its performance is still expected to be around the level of the 6900 XT.
In a nutshell, then, Greymon55 is suggesting that while Navi 33 will have fewer cores than they previously believed, what might be the midrange offering for RDNA 3 cards could still turn out to be more powerful than the RDNA 2 flagship – driven by what are presumably some seriously impressive architectural advances – or if not, it shouldn’t fall short and is expected to deliver around the same gaming oomph.
Really, though, it’s still very early in the overall timeframe for AMD’s next-gen graphics cards to be making such predictions. We’ve got to be very skeptical around all this, and even Greymon55 admits this in another tweet in which they say that as time goes on, the “specifications will become clearer” – and so obviously this speculation is in pretty muddy rumor terrain as it is.
Greymon55 further observes that AMD is “still planning” when it comes to next-gen GPUs, and that they won’t see the light of day until 2023, as mentioned at the outset of this story. Presumably early 2023, given that we were previously expecting the new graphics cards to launch at the end of 2022.
Back in July, we reported that EVGA was teasing its first AMD-based motherboard since the AM2+ days. Fast forward to now, and that dream of an EVGA model for AMD's Ryzen processors led by in-house engineer and extreme overclocker Vince 'KINGPIN' Lucido is now a reality. The EVGA X570 Dark benefits from a large premium 17-phase power delivery (14+2+1) designed for pushing Ryzen 5000 to its limits and support for up to DDR4-4800 out of the box across two memory slots to minimize latency.
The EVGA X570 Dark, as expected, is primarily suited to extreme overclockers, with a variety of performance-enhancing features across the large E-ATX sized PCB. One of the most notable design characteristics includes a transposed AM4 CPU socket for better support when mounting an LN2 (liquid nitrogen) pot, and an empty CPU socket area to minimize risks when insulating for sub-ambient cooling. All the major power connectors are also at right angles, to be less of an issue when extreme overclocking for records.
EVGA uses a 16-phase power delivery organized into a 14+2 configuration, with fourteen premium 90 A power stages for the CPU section. It's a slightly different design to accommodate the transposed socket as it stretches around the bottom of the AM4 socket. It is using active VRM cooling with two cooling fans and uses a 10-layer PCB design. The 17th power stage is for the memory. Providing power to the CPU is a pair of 8-pin 12 V ATX CPU power inputs. Some of the most notable features of the EVGA X570 Dark include voltage monitoring points at the top of the board, as well as a variety of switches to enable/disable features that might be critical to sub-zero overclocking stability.
For end-users wanting to have a daily system, there are dual PCIe 4.0 x4 M.2 slots, eight SATA ports with support for RAID 0, 1, and 10 arrays, as well as 2.5 GbE wired and Wi-Fi 6 wireless networking. Looking at PCIe support, EVGA includes two full-length PCIe 4.0 slots operating at x16 and x8/x8, with a half-length PCIe 3.0 x4 slot. Other features include a Realtek ALC1220 HD audio codec backed by EVGA's NU audio solution, eight 4-pin cooling headers, a passively cooled chipset heatsink, and two USB 3.2 G2 Type-A, four USB 3.2 G1 Type-A, and one USB 3.2 G2 Type-C port on the rear panel.
The EVGA X570 Dark is currently available to purchase directly from the EVGA website for $690. This isn't a cheap motherboard, but overclocking-focused motherboards such as this come at a hefty premium. Although it has plenty of features for a daily Ryzen 5000 based system, the hope is that it provides the substance for extreme overclockers looking to push Ryzen 5000 silicon to its limits.
Our review of the EVGA X570 Dark will be coming very soon - it arrived a couple of days ago and is currently on our testbed!
G.Skill is one of the most recognizable computer memory producers on the market, so it's little wonder that it's preparing for the next generation of RAM (random-access memory).
DDR5 is predicted to become the new standard of RAM by as early as 2023, but previously announced kits from the likes of Samsung and TeamGroup haven't had the most inspiring designs for memory.
As G.Skill is especially favored by PC gamers owing to its creative designs, it's unsurprising to see what looks to be the familiar RGB lighting that features on popular products like the Trident Z range, with the video teaser even displaying the light in a 'Z' formation.
Are you ready for the next gen? 😎#gskill #DDR5 pic.twitter.com/MoqQiDSBQ7September 22, 2021
Given the video teaser is only 15 seconds long, we don't have much information specific to the G.Skill release to go on (including a lack of release date, price or other specifications), but we're going to assume a few things based on the wider market information we already have.
Firstly, given the letter 'Z' features in the video, and that the DDR4 Trident Z RGB series is easily one of the most recognizable products on the market (alongside other G.Skill memory kits like the Trident Z Royal), we're willing to bet the overall design will remain unchanged, or receive a minimal refresh to update the look along the same lines as the G.Skill Trident Royal Elite.
We also don't know the price, but TeamGroup suggests that 16GB will become the new standard for memory modules, pricing its own 32GB, 4,800 MHz kit at $399.99 (around £300, AU$550). As pricing for DDR4 4,000 MHz (32 GB) kits currently sits at around $299-$399, the cost of TeamGroup's kit feels rather reasonable for the going market rate, but there's nothing to say that G.Skill will follow suit.
In fact, it was also suggested by TeamGroup that DDR5 will be more expensive than the current DDR4 standard, but until we see more manufacturers list a variety of different kits, we can't get an accurate estimate for what the average market price will be.
Analysis: Should you buy DDR5 early?
It's early days for DDR5, despite the new generation of memory already having a number of advantages over DDR4. DDR5 can register speeds of up to 6.4Gbps, smashing DDR4’s potential rate of 3.2Gbps. You'll also be using less power, which is a great advantage for anyone looking to save on their energy bill.
Thing is, these kits are essentially unusable until DDR5-compatible motherboards are also released, so buying one will mean it has to sit in a box until other technology catches up.
It's possible that, as so few DDR5 kits will be available in the first few months after the release of compatible motherboards and processors, we could see similar high demand and low availability.
Both AMD and Intel aren't expected to launch DDR5 supporting ranges until sometime in 2022, but if you want to guarantee you'll get your hands on some, it might be best to buy yourself a kit early and just look at it lovingly until the rest of your build hits the market.
Best gaming PC: the best computers to get into PC gaming with
from TechRadar: computing components news https://ift.tt/3EUBR6T
via IFTTT
One of the best things about the CPU war between Intel and AMD is how it’s encouraging both companies to really up their games, and a new benchmark leak for Intel’s upcoming Core i9-12900K 12th generation Alder Lake processer suggests Team Blue could be on to a winner.
The benchmarks, which were posted on Twitter (see below) and reported by Toms Hardware, apparently show the Intel Core i9-12900K, paired with an Nvidia RTX 3080 graphics card, quite comprehensively beating the AMD Ryzen 9 5950X, AMD’s flagship consumer processor that will be the i9’s chief competitor, which is paired with an AMD Radeon RX 5700XT GPU.
A 12900K (at higher preset) + 3080 obliterating a 5950X + 5700XT. https://t.co/vDxKzxzbnK pic.twitter.com/a3lx1pMQb7September 20, 2021
See more
From this screenshot, it shows the Intel Core i9-12900K system scoring a lot higher than the AMD Ryzen 9 5950X. In other results posted online for the Ashes of The Singularity benchmark running at 1440p resolution, where both the Intel Core i9-12900K and AMD Ryzen 9 5950X were paired with an RTX 3080 GPU, the Intel setup scored 14,000, whereas the AMD setup scored 10,100 – a difference of around 38.6%.
That’s a pretty huge difference in performance, and if the results are real (which is a big ‘if’ at the moment, as these results are for a CPU that’s not been released, and have not been verified), it means the Intel Core i9-12900K could crush the AMD Ryzen 9 5950X without breaking a sweat – which is good news for Intel, and bad news for AMD.
Another Twitter user also pointed out another Ashes of The Singularity benchmark leak that showed the Intel Core i9-12900K with an RTX 3080 GPU beating the AMD Ryzen 9 5950X with an RTX 3090 GPU. This is particularly interesting at it shows that even with the advantage of a much more powerful GPU, the AMD chip was unable to beat the Intel Core i9-12900K.
A 12900K (at higher preset) + 3080 obliterating a 5950X + 3090. #SiliconGang https://t.co/75BI7u0DIHSeptember 20, 2021
See more
A clear pattern is emerging, then, of the Intel Core i9-12900K easily outperforming the AMD Ryzen 9 5950X, which could give Intel a big boost in many people’s estimations, as AMD has recently been the flavor of the month for many people thanks to its popular Ryzen 5000 series of processors.
Analysis: Game over for AMD? Not quite
(Image credit: AMD; Intel)
If these benchmarks are accurate, then Intel could have a real hit on its hands with the Intel Core i9-12900K, and the rest of the upcoming Alder Lake lineup, and it could mean it once again has the best performing CPU for gamers.
This is the kind of positive news Intel needs, as it has been losing both market share and mindshare to AMD, after its rival had a series of well-received launches.
While this points to a resurgent Intel, it’s not all over for AMD just yet. Pricing will be key to Alder Lake’s success, and if Intel makes its CPUs too expensive, it could eliminate any potential performance benefits for many people – and AMD has traditionally been stronger when it comes to its price versus performance offering.
A recent price leak pointed to the Intel Core i9-12900K costing $610 (about £440, AU$830), which is certainly pricey for a CPU, and it’s an increase on the $549 (about £396, AU$748) suggested price for the Core i9-11900K it's replacing.
The AMD Ryzen 9 5950X, meanwhile, launched at $799 (around £620, AU$1,100), which is far higher. However, as it’s been out for a while, there have been some price cuts, though the global chip shortage and pent-up demand have meant prices have fallen slower than usual.
Those same factors could also mean that when the Intel Core i9-12900K launches, it will actually sell for higher than the price suggested in the recent leak.
This may lead to a situation where the Intel Core i9-12900K and AMD Ryzen 9 5950X will cost around the same amount. If that happens, then thanks to the performance advantage that the leaks are pointing to, Intel could really be putting the pressure on AMD with Alder Lake.
Let one of the best processors of 2021 do the heavy lifting whether for gaming, video editing, or multi-tasking at work. Sure, lesser chips might see you through the most basic tasks, but it’s only their most powerful and efficient counterparts that can really handle your more demanding computing needs (and stay cool under pressure).
As your computer’s brain, a CPU is responsible for executing all its commands, tasks and processes. So, going for a mediocre one and hoping for the best, especially if you expect it to perform intensive tasks, isn’t going to cut it – no matter if you’re building a new PC or upgrading your current one. And, thanks to Intel and AMD still battling it out in the CPU arena, we have a lot more excellent options at more affordable prices.
With chips like the Comet Lake-S and the Ryzen 5000 more powerful and more affordable than ever before, you no longer have to settle for less or the second best if you’re on a budget. To make it easier for you to choose, we found the best processors on the market for playing the best PC games, getting through your creative workloads, and more.
What's the best processor for gaming?
One of the best processors on the market today, the AMD Ryzen 7 5800X, combines excellent single-core performance and a massively improved multi-core one with its low power consumption and a fairly approachable price. And, it’s a strong option for gaming.
Scratch that: it’s arguably the best processor for gaming. It even beats out the Intel Core i9-10900K in Total War: Three Kingdoms, a game optimized for Intel hardware, by 7%. Based on AMD’s 7nm manufacturing process, its Core Die (CCD) design allows for one Core Complex per die. That means every Ryzen 7 5800X’s CCX has 8 cores, each of which direct access to 32MB of L3 cache, resulting in a breathtaking gaming performance.
The AMD Ryzen 9 5900X brings the biggest gen-on-gen jump in a single performance in years, making it a terrific upgrade. This latest release from AMD is not just a stronger processor across the board. It’s also an incredibly powerful processor for gaming and creative work full stop. The fact that you won’t need a new motherboard is just a nice perk.
The highest tier in the Ryzen 3000 series is the performance powerhouse you’re looking for, if you want something for heavily threaded computer work. Besides high-end gaming, the Ryzen 9 3950 also blasts through processing tasks. It’s expensive, but for a mainstream processor that can go toe-to-toe with HEDT processors, that’s hardly a surprise. And, it’s also well worth the price, if you need its level of performance. Just remember that it may take a bit to keep cool so be sure to follow AMD’s guidance.
With more threads than the Intel Core i5-9600K, this mid-range graphics card delivers impressive multi-threading performance. However, the AMD Ryzen 5 3600X doesn’t just stop there: it takes that budget-minded stage of performance to a new level, with its increased IPC (instructions per clock) performance and a higher clock speed while staying at the same price point. It also stays competitive in even the most intense single-threaded applications.
Budget-minded consumers who missed out on all the 3rd-generation Ryzen excitement the first time will appreciate AMD’s latest attempt in the entry-level sphere. There’s no denying that the AMD Ryzen 3 3100 is appealingly cheap, but it does so without compromising on performance. In fact, this chip delivers a performance that could almost rival that of processors double its price, without sucking up much power. And, it’s an excellent choice for budget gamers who are looking for something that can handle 1080p gaming.
Intel no longer has the monopoly on gaming CPUs. Rocking 8 cores and 16 threads, along with much stronger single-core performance, the AMD Ryzen 7 5800X is among the best CPUs for gaming – as well as less demanding creative work – right now. And it comes with a much more approachable price tag compared to most of Intel’s offerings, making it a much better value.
Mid-range buyers will appreciate the fact that they actually don’t have to shell out quite a bit of cash for Intel’s hyper-threading technology. With Intel’s Core i5-10600K, among Intel’s latest chips to hit the streets in 2021, it’s possible to have it for much less than $500. This chip may lack PCIe 4.0 support and has a higher power consumption, but it makes up for those in spades. Besides hyper-threading, this processor also delivers superb multi-core performance as well as improved single-core performance. What’s more is that it’s got great thermals to keep that pesky heat down.
Launched alongside the even more potent Ryzen Threadripper 3970X, the AMD Ryzen Threadripper 3960X may have the same core count as its predecessor. However, it comes with a brand-new architecture that delivers performance gains as well as PCIe 4.0, making it among the best processors in the Threadripper arena. The 3960X delivers dramatically improved single-threaded performance and has successfully reduced its predecessors’ idiosyncrasies that affect their own performance. It may come with a higher price tag and require the TRX40 motherboard – not to mention, a powerful cooler – but it’s certainly worth the fuss if you can maximize its skills to your advantage.
Whether you’re looking for a processor for your media center or for your desktop PC, the AMD Athlon 300GE is among the best processors you can find on the market today in terms of value. It’s reliable and surprisingly fast for what it can do. It’s never going to deliver a breathtaking performance, but for a processor in the budget space, it’s definitely going to impress you.
Lower performance in applications than alternatives
With a noticeable, if small, performance advantage over its predecessor, AMD’s Ryzen 5 3400G is a capable option for your home theater PC and budget gaming computer. That means that it’s among best processors for folks who don’t quite take gaming as seriously as the big gamers, touting a solid 720 and a fairly decent 1080p gaming performance.
from TechRadar: computing components news https://ift.tt/3nBzMnF
via IFTTT
AMD has expanded its offer to give away free copies of Far Cry 6 and Resident Evil Village to include its flagship GPU, the Radeon RX 6900 XT.
Previously, this ‘Raise the Game’ bundle was only available to people who bought both a Ryzen CPU along with an RX 6000 GPU together in a qualifying system. However, now those buying just an AMD graphics card can grab both of these high-profile games for free, at least if they purchase an RX 6900 XT as mentioned.
There is a slight caveat, namely that buyers of both a CPU and GPU also get a three-month trial of Xbox Game Pass, whereas that additional incentive doesn’t apply with the 6900 XT offer. Just something to bear in mind.
The bundle is now available with any RX 6900 XT card, including third-party or reference models, as Tom’s Hardware reports, and the offer runs from now until December 31, 2021. It is, however, only available to buyers in the US and Western Europe at participating retailers.
Analysis: The Big Navi GPU with the big price tag…
That said, it's slightly disappointing that the only GPU which applies is the most expensive Big Navi product, as it’s truly eye-wateringly expensive (a quick scan of online retailers indicates current pricing starting at around $1,700 in the US, or £1,500 in the UK, for those models which are actually in stock, of course).
It’s also worth bearing in mind that Far Cry 6 comes complete with a major boost from AMD’s FidelityFX Super Resolution, and according to the firm’s own testing, frame rate leaps of over 40% are possible with FSR tech. Pretty cool indeed, and with the power of the 6900 XT, the game should run very impressively.
To date, most of the new AI hardware entering the market has been a ‘purchase necessary’ involvement. For any business looking to go down the route of using specialized AI hardware, they need to get hold of a test system, see how easy it is to migrate their workflow, then compute the cost/work/future of going down that route, if feasible. Most AI startups are flush with VC funding that they’re willing to put the leg work in for it, hoping to snag a big customer at some point to make that business profitable. One simple answer would be to offer the hardware in the cloud, but it takes a lot for a Cloud Service Provider (CSP) to bite and offer that hardware as an option to their customers. Today’s announcement between Cerebras and Cirrascale is that as a CSP, Cirrascale will begin to offer wafer-scale instances based on Cerebras’ WSE2.
The AMD vs Intel battle still persists, with no one manufacturer coming out the one true winner. That’s great news for us consumers. This intense rivalry continues to gift us the best processor for gaming, creative workloads, and casual use, giving us a whole lot more options to choose from at every price point.
This also means, however, that choosing between AMD and Intel is much harder. Luckily, you don’t necessarily have to choose one over the other. Both have their own strengths and weaknesses, as well as excellent CPUs on offer catering to every need and budget. Still, as this is strictly an Intel vs AMD faceoff, it's important to know where each manufacturer's strengths lie, especially in terms of price, overclocking abilities, graphics, and component variety. That way, you can choose wisely.
So, let’s dive right in. Whether you’re building a computer or simply upgrading the gaming PC you already have, we’ll help you make sense of it all by comparing the two and sorting out how the AMD Ryzen 5000 series chips (and their predecessors) are different from Intel’s latest Tiger Lake and Rocket Lake chips (and their 10th-generation counterparts). We’ll also discuss where each brand is headed in the coming years.
If you want a breakdown of the AMD vs Intel rivalry, you’ve come to the right place. Dig right in before you make a decision.
(Image credit: TechRadar)
AMD vs Intel: price
In the past, if you were looking for a decent CPU with a budget-friendly price, your go-to choice was AMD. However, with their newest generation of Ryzen CPUs, AMD has been on par or even surpassed Intel components on price. The AMD Ryzen 9 3950X, for instance, is easily the brand's most expensive unit, retailing for around $750 (£580, AU$1130). However, the sticker shock you feel backs up some impressive technology.
The Ryzen 9 3950X boasts 16 cores and 32 threads, exceeding the previous flagship, the Ryzen 9 3900X. And it doesn't just have more cores and threads than its predecessor, the Ryzen 9 3950X has more cache memory and faster processing speeds. It also has the ability to store up to four different profiles for different performance and overclocking settings, so no matter if you're a casual or hardcore gamer or hobbyist artist and animator, you'll always have the best performance you can get out of your CPU.
On the much cheaper end, AMD just rolled out the AMD Ryzen 3 3300X and AMD Ryzen 3 3100, it’s new budget champions. These cost $120 (about £100, AU$190) and $99 (about £79, AU$150) respectively while delivering truly impressive power, proving once again that AMD is king when it comes to affordability and multi-threaded performances.
When it comes to Intel, Coffee Lake Refresh processors are probably what most people are familiar with. The latest generation of Coffee Lake Refresh processors from Intel includes the i9-9900K which retails for around $550 (£420, AU$830) and gives you 8 cores to work with as well as native integrated graphics.
However, the new Comet Lake-S desktop processors have just been unveiled and are well on their way, led by the Core i9-10900K. At $488 (about £395/A$765), this high-specced chip boasts 10 cores, 20 threads and a boost of 5.3GHz. Another notable chip in this new line is the Core i7-10700K, which offers 8 cores and 16 threads with a 5.1GHz turbo for just $374 (about £300, AU$571).
Comet Lake’s mobile processors, on the other hand, have been filtering out for months now, with the Core i7-10710U leading them. These chips start at $281 (about £215, AU$405) and cost up to $443 (about £340, AU$635). Their gaming laptop counterparts, the Comet Lake-H processors, have also been trickling out, with the Core i9-10980HK leading the charge at $583 (about £470, AU$910).
If you're in the market for a really high-end CPU and have super deep pockets, the third generation of AMD Threadripper units is the best choice for professional 3D modeling artists, animators, filmmakers, and data scientists. The Threadripper 3990X retails for a whopping $3,900 (£3,030, AU$5,890), putting it well out of range of casual PC builders and average PC gamers. This CPU is built with 64 cores and 128 threads to give you plenty of power to render 3D models and rip through complex mathematical models to get the most out of your workday.
(Image credit: Intel)
AMD vs Intel: performance
So you've set yourself a budget for a new CPU, but you still have a ton of options when it comes to performance. On the whole, AMD and Intel Processors have been on a pretty even keel when it comes to overall performance. Between the two, it all comes down to whether you need to multitask well or want to play games at their highest settings.
If you're looking at buying an AMD processor, be advised that very few of their available CPUs feature integrated graphics. Those that do are referred to as an Accelerated Processing Unit. The ultra-low budget AMD Athlon 240GE retails around $80 (£62, AU$120) and features Radeon Vega 3 integrated graphics. This makes it perfect for low- to mid-grade gaming as well as video streaming for high quality graphics rendering at a low price.
Now that AMD has rolled out its budget gaming champions, the AMD Ryzen 3 3300X and AMD Ryzen 3 3100, you’ve got a few more options here. These two might set you back a bit more, but not by much, while offering a much better performance. The Ryzen 3 3300X, for example, delivers performance that until now was only available on mid-range or high-end hardware. Meanwhile, the Ryzen 3 3100 is simply excellent for 1080p gaming.
If you're into higher-end gaming, however, you'll have to pair a Ryzen 7 or 9 CPU with a dedicated GPU to take your game to the next level.
For day-to-day work and multitasking, the AMD Ryzen line is also a safe bet, as they offer the most PCIe lanes so you can use more solid-state drives for super-fast computer start up and file recall. Just keep in mind that AMD processors tend to run hotter than their Intel counterparts, so you'll need to consider either a supplementary fan or liquid cooling system for your new CPU.
With Intel, on the other hand, each chip has on-die integrated Intel HD or Iris graphics, so you can play most mainstream games or stream quality video right out of the box, no matter what CPU you pick. However, like their AMD cousins, if you want to play more graphically demanding games, you'll need to choose a companion GPU.
But with the newest Coffee and Ice Lake processors, each CPU will beat out AMD Ryzen and Threadripper units on core-by-core performance - though that gap is minimal. The late 2020 introduction of Tiger Lake could see even more integrated graphic rendering ability for a better streaming or gaming experience right out of the box. Intel has also heavily hinted at plans to release their own dedicated Intel Xe GPU in 2020.
(Image credit: Future)
AMD vs Intel: specs
We've reviewed both the Ryzen 9 3950X and the Intel i9-9900K to give you more insight on each processor's capabilities, performance, and price. Both processors give you plenty of power, but each has their own pros and cons.
As mentioned before, the Ryzen 9 3950X has 16 cores and 32 threads. This gives you all the power you need and then some to tackle everyday multitasking and general workloads in an office setting. It also has enough juice to give you great frame rates in both full HD and 4K gaming settings so you don't have to deal with terrible amounts of lag or screen tearing. The entire Ryzen 3000 series are all fairly evenly-matched when it comes to frame rates and multitasking abilities, so it all comes down to how many cores and threads you'll need.
The Ryzen 9 3950X features dual channel memory support and 64MB of cache. This ensures faster recall of your frequently-used files and programs. With a base clock speed of 3.5 GHz and a Max Boost Clock of 4.7 GHz, you'll be able to tackle just about any game or work task at blazing speeds.
The Intel i9-9900K has half the number of cores and threads as the Ryzen 9 3950X, but it makes up for some of that with slightly stronger single core performance. The i9-9900K has a base speed of 3.6GHz and a Turbo clock of a whopping 5GHz. It also uses just 95 watts of power compared to the Ryzen 9's 105 watts - though you are getting around half the total performance.
With Intel's integrated graphics, you'll get both full HD and 4K video and graphical support right out of the box. You'll not only get a great picture for both streaming video and playing the latest games, you'll also get awesome frame rates as well, preventing lag and screen tearing.
(Image credit: Intel)
AMD vs Intel: technical and customer support
AMD has an entire webpage dedicated to customer and technical support of their CPUs. On this page, you can download the latest drivers for integrated Radeon graphics processors or GPUs. You can also check up on your product's warranty, download full spec sheets, and ask other AMD users questions on a dedicated forum. If you have a problem with a specific unit, you can use a drop-down menu to select your CPU to be directed to a page of driver download links and a customer support page for more in-depth troubleshooting.
Intel's official site also has a dedicated page for technical support when you have trouble with your new or existing CPU. You'll be able to browse a variety of blog posts that answer frequently asked questions, view spec sheets, download drivers, and access the support community forums if your question isn't answered by the FAQ. Intel also offers live phone or chat support if you need help walking through a solution.
(Image credit: AMD)
AMD vs Intel: future speculation
AMD has had a massively successful past few years with their Ryzen and Threadripper lines of CPUs. With twice the number of cores and threads as their counterparts, they've given Intel something to worry about and proven their worth as a reliable choice for PC components. Intel, on the other hand, has seen some embarrassing failures such as their lackluster release (or lack thereof) of Cannon Lake. However, they seem to have taken their knocks and learned a few lessons to apply to future releases.
Intel has announced plans for future releases of their Tiger Lake and Comet Lake-S lines of processors for laptops and desktops, respectively. The new CPU line will most likely try to stay abreast of AMD's Ryzen and Threadripper products. However, it's unlikely that Intel will be able to top AMD any time soon, as even Intel CFO George Davis has admitted that Team Blue won't reach parity with AMD's 7nm manufacturing process until 2021.
Future generations of AMD's Ryzen processors will most likely continue to give you more cores and threads for faster and more efficient multitasking. Whenever AMD Ryzen 4000 processors for desktop make their way to market - which should be some time this year - the shift to a more efficient 7nm+ manufacturing process should see further boosts to IPC (instructions per clock) performance along with power efficiency. Threadripper's future seems to stay on track to offer powerful CPU options to industry professionals for 3D modeling and animation or data science work.
Interested in graphics cards? Check out the Nvidia vs AMD smackdown
from TechRadar: computing components news https://ift.tt/3hTMNYY
via IFTTT
Whether overclocking the CPU is your cup of morning tea/coffee/something stronger or not, there are specific motherboards built by professional overclockers and engineers designed to squeeze out as much performance as possible. For example, while the regular everyday PC user might groan at the thought of spending $500 on a Z590 motherboard with two memory slots, boards such as the ASRock Z590 OC Formula pay homage to the art of overclocking. As such, legendary overclocker and ASRock employee Nick Shih has overseen the design and creation of one of the best series of motherboards for overclocking. The OC Formula has been ASRock's brand for their pinnicle performance motherboards for generations - it seemed dead as we hand't seen it in a while, but it as come back with some blazing style. The latest iteration has had a facelift and comes with a wave of features - but can ASRock strike the right balance between enthusiast and conventional? Let's find out in our review of the Z590 OC Formula.
An unidentified Intel Xe graphics card which has just been spotted in a leaked benchmark looks like it’s a laptop version of Intel’s Arc Alchemist discrete GPU, teasing the possibility that we might see the beefiest flagship solution in gaming notebooks next year perhaps.
PC Gamer spotted the Geekbench 5 result as flagged up on Twitter by a less familiar source for us, though still one which has provided previous spillage, namely Benchleaks, so plenty of salt here as ever – but it’s a very interesting sighting.
As you can see, the GPU is only named as an Intel Xe discrete graphics card running in a Tiger Lake laptop, but it has 512 EUs (Execution Units) meaning it certainly isn’t a current Xe laptop card, and that EU count corresponds with Alchemist, and the top-end configuration of Intel’s incoming Arc gaming GPUs. It’s shown with a clock speed of 1.8GHz, though that could be pushed higher with the release version.
Due to that, the result of 34,360 for the OpenCL score isn’t something we can draw any real conclusions from, save that it seems to be in the right ballpark for such an early pre-release product. Besides, Geekbench is far from the best place to be evaluating a GPU’s performance, anyway (there’s much better benchmarking software out there for that job).
Analysis: Great news for gaming on the go?
Obviously we need to be careful here, but this certainly looks like a tantalizing hint that before long we’ll have some seriously powerful new gaming laptops. If a notebook can pack an Intel Alder Lake CPU – with its major design innovations on the mobile front, with the hybrid architecture offering clear advantages for laptops – alongside a flagship Alchemist GPU, the potential performance leap in gaming during 2022 (maybe relatively early in 2022) is something to get excited about that.
Think of it this way: we’ve got what could hopefully be a seriously powerful Arc GPU driving frame rates, coupled with all the battery benefits Alder Lake promises to bring to laptops, and that should make for not just gaming notebooks with more grunt, but models which can last for a bit longer on the move too.
Nvidia could be releasing next-gen (presumably RTX 4000) graphics cards in October of next year – and maybe in direct competition with AMD’s RDNA 3 GPUs (plus RTX 3000 Super variants might be unleashed at the beginning of 2022, as a stopgap).
This is the latest from the rumor mill (via VideoCardz) going by a couple of prominent hardware leakers, so exercise a ton of caution, but this is what Greymon55 and RedGamingTech believe to be the case.
Let’s start with Greymon55 who tweeted a reply to a PC builder who wanted to know how long they might have to wait for next-gen GPUs, and whether those cards might be just around the corner.
Next-gen gpu will be in October next year, about a year from now.September 7, 2021
See more
Interestingly Greymon55 has a firm release date in mind, naming October rather than saying Q4, before clarifying that they aren’t talking about Nvidia specifically, but just next-gen graphics cards in general. As someone tweets in response, that seems to imply the possible scenario that both AMD and Nvidia’s next range of GPUs will both come out in October 2022, but Greymon55 doesn’t confirm or deny – which might suggest agreement (or not – we’ll go into that in more depth later).
Greymon also just tweeted to expect RTX 3000 Super laptop cards "early next year", backing up a piece of (somewhat older) speculation from YouTube-based leaker RedGamingTech about Super GPUs being readied for a theoretical launch at the start of 2022 at CES. Those would be mobile parts, and the pepped-up laptop GPUs may only offer a modest increase of 5% to 10% in performance compared to the existing (non-Super) mobile GPUs, RedGamingTech has heard.
RedGamingTech says that his source for the mobile graphics card info observes that there may be Super GPUs on desktop, too, but that they similarly won’t offer much of a big jump in performance.
With third-party graphics card manufacturers yet to be briefed on the rumored Super desktop models, these can be expected to pitch up later down the line (if they turn up at all), given that Nvidia won’t want to run them too close to next-gen cards – maybe debuting in October as mentioned – for obvious reasons.
Another interesting snippet to note from RedGamingTech: both AMD and Nvidia are ‘nervous’ about Intel’s debut in the desktop GPU market, given that Team Blue has supposedly thrown a lot of money at R&D in pushing forward with Intel Arc cards. That’s certainly believable and a comforting rumor to hear – the more competition in the GPU arena, the better, for sure.
Analysis: Timing is a tricky beast to call
Greymon has previously indicated that Nvidia ‘Lovelace’ GPUs (RTX 4000) would be likely to arrive no earlier than the end of 2022. Arguably, October is Q4 which is of course the final quarter, so still sort of the end of next year, but it feels like the leaker is now more optimistic about a potential arrival timeframe of slightly sooner than anticipated before.
Except, of course, for the crucial fact that Greymon doesn’t clarify that it will be Team Green which gets its next-gen GPUs out next October, and as we touched on above, a fellow Twitter denizen raises the possibility of both an Nvidia and AMD graphics card launch in October, which Greymon doesn’t rule out (or confirm). So, that could be a possibility, or maybe the leaker is hinting that in fact AMD may be coming with next-gen (RDNA 3) cards in October and not Nvidia (as opposed to as well as Nvidia). AMD has previously been rumored to be looking at a Q4 launch in 2022, after all, so that still fits.
Don’t forget, though, that another rumor peddler, Kopite7kimi on Twitter, believes that Lovelace cards will turn up not in late 2022, but a ‘little bit earlier’, which would seem to fit with an October release date for Team Green.
Of course, before we get carried away with any of this, or the possibility of a head-to-head next-gen GPU deathmatch, we must remember that at this point, trying to nail exact launch dates requires extreme caution as these plans can slide anyway (any launch timescale can). What we can take from this, though, and other recent chattering from the grapevine, is that it seems like we can be hopeful that we’ll see some kind of next-gen GPU at least a while before the end of 2022 rolls around.
As for the Super spins on RTX 3000 GPUs, if RedGamingTech is anything like near the money with that guess of maybe as little as 5% performance difference with the revamp, that’s obviously not much to get excited over. Still, it’s certainly a rumor to keep an eye on going forward.
Before you give up altogether and spend money on a new computer or gaming PC, check your current PC’s memory first if its performance is beginning to lag. If it’s having trouble keeping up with the more intensive tasks like playing PC games or video editing, it could just be that you need to upgrade to the best RAM.
Whether you primarily use your computer for gaming, creating content, or simply getting through your daily workload, having enough fast memory is as integral as a robust graphics card and a powerful processor. So, if your computer is experiencing slowdowns, it could just be a matter of getting a faster memory or increasing its amount – these days, 8GB is the minimum while 16GB is the ideal for the more demanding tasks.
Upgrading your PC’s memory could be the perfect solution to your computer blues, which means you won’t have to spend a boatload of money on a pricey GPU or CPU (or on an entirely new computer). Bring the one you already have back to life with one of the best RAM we’ve picked below. And, be sure to check out our price comparison tool as you peruse so you can find the best RAM deals available as well.
Corsair Vengeance LED is just as fast and responsive as it is aesthetic. (Image Credit: Corsair)
Corsair is one of the most trusted names when it comes to the best RAM on the market. Its Vengeance series, especially, has something for everyone with its LED DDR4 offerings. On top of giving PC builders that extra touch of vibrant lighting, this RAM has robust heat spreaders that maximise cooling for higher overclocking and maximum performance. Corsair’s Vengeance LED DDR4 series also features CL16 latency and whopping 3,466 MHz speeds, so it’s just as fast and responsive as it is aesthetic.
It’s hard to argue with the G. Skill TridentZ RGB as the best RGB RAM. (Image Credit: G.Skill)
At this point, everyone kind of knows about G. Skill and its Trident Z RGB series of RAM. This is some of the best RAM not just because it’s fast, but the top of every DIMM features a full-spectrum rainbow wave light bar that you can use to match your RGB lighting across your entire system. Trident Z RGB DDR4 RAM features CAS latencies between 14 and 19 which is awesome enough, but with speeds up to 4,266, it’s almost perfect. No matter your aesthetic, it’s hard to argue with the G. Skill TridentZ RGB as the best RGB RAM.
Kingston HyperX Predator is particularly tuned for speed and extreme-performance. (Image Credit: Kingston)
3. Kingston HyperX Predator
Best DDR3 RAM
Type: DDR4 | Capacity: 8GB, 16GB | Speed: up to 4600MHz
Great price
Excellent overclocking potential
RGB doesn't always work
Pricier than others
Kingston’s high-performance HyperX has some of the best DDR3 RAM on the market and the Predator models is particularly tuned for speed and extreme-performance that’s further expandable with XMP profiles. The Predator DDR3 series achieves CL9 to CL11 latencies and speeds between 1866MHz to 2666MHz.
Kingston HyperX Fury can have a huge impact on the performance of your rig. (Image Credit: Kingston)
4. Kingston HyperX Fury
Best budget RAM
Type: DDR4 | Capacity: 4GB, 8GB, 16GB, 32GB | Speed: up to 3733MHz
Affordable at lower speeds
Strong performance
A bit slow
3733MHz speed is pricey
This Kingston HyperX Fury is auto overclocked memory that won’t empty your wallet. This smart DDR3 or DDR4 RAM auto-detects system components to overclock to the highest speeds possible, optimizing performance for all of Intel’s latest chipsets. It comes with latencies between CL14 and CL16, and has speeds between 2,133 and 2,666MHz. While it’s inexpensive, it really can have a huge impact on the performance of your rig.
Corsair Dominator Platinum RGB is the best RAM around in 2019. (Image Credit: Corsair)
If you want the best of the best, and you don’t care about pricing, Corsair’s Dominator Platinum memory has always been at the top of the list. However, in 2019, it’s been refreshed to include Corsair’s new Capellix RGB LEDs, leading to a much richer and more colorful design. It’s the same high-end RAM we know and love, with speeds up to 3,600 MHz, but the better RGB make the Corsair Dominator Platinum RGB the best RAM around in 2020.
With speeds of up to 3,733MHz and CL15–19 latencies, HyperX Fury RGB 3733MHz is simply fast.(Image credit: HyperX)
6. HyperX Fury RGB 3733MHz
Best high frequency RAM
Type: DDR4 | Capacity: 4GB to 32GB single modules, up to 128GB kit configurations | Speed: 3733MHz
Great RGB
Fast
Choppy RGB lighting
Gigabyte and MSI RGB support doesn't always work
The HyperX Fury RGB 3733MHz is not just pretty with its RGB design. With speeds from 2400MHz to 3,733MHz as well as CL15 to CL19 latencies, it’s just as fast as it is stunning, making it well-deserving of the best high frequency RAM accolade. What’s more is that it’s fitted with predefined Intel Extreme Memory Profiles for maximum performance as well as yperX’s patent-pending infrared sync technology. For an even more immersive gaming experience, you can daisy chain this RAM to sync with multiple other RGB devices. This is some of the best RAM to invest in right now.
If you need a lot of RAM without taking up too many DIMM slots, G.Skill Trident Z RGB DC is for you. (Image Credit: G.Skill)
Sometimes, especially when you’re looking for the best gaming components, ‘go big or go home’ is the best advice. And, when you want more RAM than you’ll know what to do with, you’re going to want to go with the G.Skill Trident Z RGB DC memory. The DC stands for double capacity, as in, 32GB per stick. Now, this is definitely not the fastest memory in the world, right now it’s only available in up to 3,200 MHz, but if you need a lot of RAM without taking up too many DIMM slots (like if you have a big CPU cooler or a Mini-ITX board), you can’t go wrong with the G.Skill TridentZ RGB DC.
Adata Spectrix D80 is great for your all-out gaming rig. (Image Credit: Adata)
8. Adata Spectrix D80
Best gaming RAM
Type: DDR4 | Capacity: 16GB (2x8GB), 32 (2x16GB) | Speed: up to 4,133MHz
Great looks
Performs very well in four-DIMM configurations
Expensive
Mediocre two-DIMM performance
If you’re building an all-out gaming rig, you’ll want the flashiest hardware on offer – who doesn’t want a gaming rig that lights up the entire room? If this sounds appealing, the Adata Spectrix D80 is the RAM to go for. Not only is it available in frequencies up to 5,000MHz, but it’s also liquid cooled, meaning you don’t have to worry about it overheating. The Adata Spectrix D80 is going to be the best RAM for gaming in an RGB-lit cavern.
The TridentZ Royal memory gets our vote as the best RGB RAM on the market.(Image credit: G.Skill )
Finding RAM that fits in perfectly with your light up gaming rig is easy with G.Skill. In fact, its TridentZ Royal memory gets our vote as the best RGB RAM on the market right now, thanks to its unique crown jewel design. That means that it’s fitted with a crystalline light bar that basically scatters the eight customizable RGB colors for lighting that dazzles and sparkles. The RAM itself is polished with aluminum heatspreaders of gold or silver colors to keep the whole thing cool. Of course, it’s not all looks – the G.Skill TridentZ Royal also delivers great reliability and performance.
Corsair Vengeance LPX is essential for PC builds with massive CPU coolers. (Image Credit: Corsair)
Corsair’s Vengeance LPX RAM is made for users who want the maximum performance with minimum footprint. This low-profile RAM is essential for PC builds with massive CPU coolers. Despite being so low-profile, they still feature an eight-layer heat spreader to cool while overclocking. While its thin design doesn’t allow for LEDs, but you can get it in three colors: black, red or blue.
G.Skill Mac RAM has memory upgrades for Macs. (Image Credit: G.Skill)
11. G.Skill Mac RAM
Best Mac RAM
Type: DDR3 | Capacity: 4GB (2x2GB) | Speed: up to 1066MHz
Affordable alternative
Best value
May not be compatible with much older Macs
Slow for power users
G.Skill isn’t just dedicated to laptop and desktop RAM, but it also has memory upgrades for Macs. You see, Apple charges inordinate amounts for RAM upgrades, so going with a third party kit can save a ton of money. Well, as long as your Mac is old enough to still have upgradeable RAM. These SO-DIMM kits have a CAS latency between 9 and 11 and feature speeds between 1,333MHz and 1,600MHz. The best bang for your buck seems to be the G.Skill DDR3-1,333 for Mac.
Crucial Ballistix Sport boasts efficient battery life and delivers fast speeds for multi-tasking. (Image Credit: Crucial)
12. Crucial Ballistix Sport
Best laptop RAM
Type: DDR4 | Capacity: 32GB (2x16GB) | Speed: up to 3200 MHz
Low profile
Great value for gaming
Doesn't alway work with AMD Ryzen
Nothing extraordinary
Crucial is by far the best RAM for laptop systems. This manufacturer has designed its Crucial Ballistix Sport SODIMMs for efficient battery life while still delivering fast speeds for multi-tasking. It’s ideal for anyone wanting to speed up the best laptops, and this memory likely offers higher memory speed than anything coming stock in your laptop. Just make sure your laptop actually lets you upgrade the RAM – repairability is falling out of fashion these days.
If you want a low profile, low power consumption RAM with automatic overclocking prowess, Silicon’s Power Zenith PowerX is an excellent option. It boasts an aluminum build and a design that effectively manages and minimizes heat when working overtime. Considering its performance, that low profile for more compact setups is even more impressive. And, it’s just the ticket for modders and overclockers.
from TechRadar: computing components news https://ift.tt/36O1WWh
via IFTTT
_575px.jpg)
