PC of the month (December 2024)

In recent years, the Consumer Electronics Show (CES), traditionally held in early January in Las Vegas, has become a key platform for significant announcements in consumer electronics and PC components. This time, the "zero days" of CES 2025 already include major presentations from AMD, Intel, and NVIDIA, showcasing new products such as processors, graphics cards, and motherboards.

A lot of information about upcoming devices is already known, so by using the available data and analyzing current market offerings, we are crafting optimal gaming configurations across various price categories within the traditional "PC of the Month" feature.

Basic Gaming PC of the Month ($650–800)

Gaming Monitor for "Basic Gaming PC" by our partner MSI

MSI G271CQP E2 with a fast VA panel featuring a resolution of 2560×1440 and a response time of just 1 ms offers an impressive gaming experience for entry-level systems. This curved monitor (1000R) with a 27-inch diagonal has a refresh rate of 170 Hz.

Gaming Monitor MSI G271CQP E2

Processors

While awaiting potential platform changes for Intel in basic gaming configurations, we continue to rely on the Core i5-12400F. This 6-core, 12-thread model has proven itself well in this category, and its very attractive price (~$120) leaves little room for alternatives. With the emergence of affordable chips for LGA1851, an intriguing internal competition could arise – Core i5-12400F vs. Core Ultra 3 205/215.

Despite the potential of the new platform, practical advantages of 4P+4E chip configurations over 6/12 may not be as apparent. Additionally, at the initial stage of availability, the actual cost of new solutions is likely to be less favorable. Financial considerations for entry-level systems are typically decisive.

If you don’t want to wait for LGA1851 solutions, a motherboard on Intel B760 is necessary for a system using Core i5-12400F. Unfortunately, there’s not much good news here. Over the past few months, the most affordable models on this chipset have increased in price by $10–20. As a result, motherboards on this PCH with DDR5 memory are rarely under $110–120. The reason for this isn’t limited to the memory standard, as DDR4 versions have also risen to a similar price level.

MSI G271CQP E2 with a fast VA panel featuring a resolution of 2560×1440 and a response time of just 1 ms offers an impressive gaming experience for entry-level systems. This curved monitor (1000R) with a 27-inch diagonal has a refresh rate of 170 Hz.

Gaming Monitor MSI G271CQP E2

Processors

While awaiting potential platform changes for Intel in basic gaming configurations, we continue to rely on the Core i5-12400F. This 6-core, 12-thread model has proven itself well in this category, and its very attractive price (~$120) leaves little room for alternatives. With the emergence of affordable chips for LGA1851, an intriguing internal competition could arise – Core i5-12400F vs. Core Ultra 3 205/215.

Despite the potential of the new platform, practical advantages of 4P+4E chip configurations over 6/12 may not be as apparent. Additionally, at the initial stage of availability, the actual cost of new solutions is likely to be less favorable. Financial considerations for entry-level systems are typically decisive.

If you don’t want to wait for LGA1851 solutions, a motherboard on Intel B760 is necessary for a system using Core i5-12400F. Unfortunately, there’s not much good news here. Over the past few months, the most affordable models on this chipset have increased in price by $10–20. As a result, motherboards on this PCH with DDR5 memory are rarely under $110–120. The reason for this isn’t limited to the memory standard, as DDR4 versions have also risen to a similar price level.

Based on numerous independent reviews, the Intel Arc B580 12 GB indeed surpasses the aforementioned models in 1440p gaming, typically by 10–15%, as initially claimed by the developers. The significantly larger memory capacity plays a crucial role in achieving higher frames per second in scenarios where 8 GB is often insufficient. However, in cases of "heavy" projects, the increased memory may not always be sufficient to maintain high-quality visuals. This is especially true when GPUs can deliver, for example, average frame rates of 30 and 35 fps. In this context, the second GPU is technically 17% more powerful, but in reality, both will require compromises on visual quality settings.

Overall, having a 12 GB graphics card like the Intel Arc B580 for $249 is a significant milestone. There has been a noticeable gap in budget options for 1440p gaming. Intel Arc B580 clearly outperforms the GeForce RTX 3060 12 GB.

In addition to hardware improvements, Intel has introduced its next-generation smart scaling – XeSS 2 with frame generation and low latency features (Xe Low Latency). This move is clearly an attempt to compete with competitors' solutions like DLSS 3.0 and FSR 3.0/3.1. However, widespread adoption of XeSS 2 by developers still needs time. Initially, it has been integrated into F1 2024, with the list expected to expand to ten titles soon.

In the case of Intel Arc B580, a disappointing factor is its very limited selection of AIB partners, which hasn’t significantly changed since the first generation of Intel Arc. Current designs are offered by Acer, ASRock, Gunnir, Muxsun, Onix, and Sparkle. The absence of market leaders in discrete graphics cards undoubtedly limits the geographical reach of Intel’s new models. However, we remain hopeful to see official Intel graphics card sales in Ukraine this time.

Overall, the Intel Arc B580 12 GB has received positive reviews, presenting itself as competitive among current offerings. Thus, it’s not surprising that the first batches of these graphics cards have already been claimed. However, moving forward, it will have to compete with new-generation AMD/NVIDIA graphics adapters, making the power balance likely to shift.

With competitors also keenly interested in this segment, the competition will be intense. In AMD’s future lineup, an entry-level model with a 192-bit interface and 12 GB of memory, possibly named “8600 XT,” is expected. Although the branding might change, the increased bus width and memory capacity are unlikely to be abandoned.

Recent leaks suggest that the upcoming GeForce RTX 5060 will utilize the GB206 processor, equipped with a 128-bit interface and 8 GB of memory. However, the graphics card is expected to use GDDR7 memory chips with an effective clock speed of 28,000 MHz, resulting in a memory bandwidth of 448 GB/s – a 65% increase compared to the RTX 4060 (272 GB/s). Therefore, a substantial performance boost is anticipated here.

Whether the Intel Arc B580 retains its “+10%” lead to compete effectively with AMD/NVIDIA successors will be evident through direct comparisons. For now, the Intel Arc B580 is available, while competing models are still a few months away.

Returning to the Intel Arc B570 10 GB, this younger model uses the same BMG-G21 graphics processor but with fewer compute units (2304 vs. 2560), a 160-bit interface, and 10 GB of memory. We anticipate that this version will be around 15% slower than the Intel Arc B580, but its key selling point will remain the recommended price of $219. However, this remains a future prospect, especially for our market. Currently, the list of viable options remains unchanged – GeForce RTX 4060 8 GB, GeForce RTX 3060 12 GB, and Radeon RX 7600 8 GB.

Оптимальний ігровий ПК місяця ($1200–1400)

A gaming monitor for the "Optimal Gaming PC" from our partner MSI is a great addition to your gaming setup.

A 27-inch MSI MAG 275CQRF QD E2 gaming monitor with a fast Rapid VA panel, 2560×1440 resolution, Quantum Dot technology, 180Hz refresh rate, and a response time of just 0.5 ms ensures an optimal gaming experience and vivid, immersive visuals.

Processors

For optimal Intel configurations, we made a slight change. Instead of the Core i5-12600KF (6/12+4; 3.7/4.9 GHz + 2.8/3.6 GHz), we now prefer the Core i5-12600K with enabled integrated graphics. Recently, the cost of these versions has almost equalized (~$180), so the presence of iGPU, which was previously an additional $20–30, benefits the Core i5-12600K. Overall, this chip, introduced in October 2021, still performs confidently in mid-range gaming systems. Good IPC of productive cores, relatively high working frequencies, and the presence of four additional energy-efficient compute cores in case heavier projects require more resources than the 6/12 configuration.

By the end of the year, the price of Core i5-14400/14400F chips has noticeably dropped. The standard model can now be purchased for $200–210 instead of the recent $240–250, while the version with the deactivated iGPU is offered from $175. Thus, for those prioritizing the formal affiliation of the processor with the Core 14th generation, the choice becomes simpler. However, for higher performance at the same price, it is better to choose the Core i5-12600K, which, even architecturally, does not lag behind base Core i5 chips of the next generations.

Regarding genuinely next-generation chips, models within the LGA1851 platform are expected. We hope that the new Core Ultra 5 processors for this category will offer real advantages. Merely slight improvements in energy efficiency will not suffice. Current Intel chips already have acceptable power consumption during gaming, so more substantial arguments are needed, such as increased fps. Looking at the top Arrow Lake results, we anticipate that there could be issues with the more accessible 65-watt models. Faster integrated graphics and the presence of an NPU unit undoubtedly add appeal to new chips, but in gaming systems, high frame rates per second remain a priority. Therefore, we still anticipate thorough comparisons and a balanced approach in selecting the right solution. Mass Arrow Lake chips and motherboards based on more affordable chipsets will be presented at CES 2025, so we must be patient. Alternatively, for those not wanting to wait, we currently recommend the proven and predictable platform with "Core i5-12600K + Intel B760 + DDR5" at a reasonable cost.

In the context of AMD platforms, developers are preparing their 65-watt Ryzen 9000 processors. However, it's worth noting that this lineup already includes chips with such nominal TDP – Ryzen 5 9600X (6/12; 3.9/5.4 GHz) and Ryzen 7 9700X (8/16; 3.8/5.5 GHz). Shortly after the release of these models, AMD had to hastily revise their positioning and significantly simplify the process of increasing their TDP to 105W while maintaining warranty commitments. This led to results – a +15% increase in performance in multi-threaded tasks. However, even Ryzen 5 9600X with a retail price of ~$270 appears somewhat expensive for an optimal configuration. After all, we are talking about a 6-core processor. Therefore, we continue to recommend Ryzen 5 7600 as a base and anticipate more affordable models based on Zen 5 architecture. The most likely successor is expected to be Ryzen 5 9600, which is set to be unveiled in January. The frequency formula for the starting model is unlikely to differ significantly from that of Ryzen 4 9600X, though the recommended price is expected to be lower.

Simultaneously, with the expansion of the Ryzen 9000 lineup, developers will also introduce updated AMD B850/B840 chipsets and, accordingly, motherboards based on them. However, no functional revelations are expected here. The potential of current and upcoming Ryzen 7000/9000 chips can fully be realized on existing AMD B650 motherboards.

GPUs

For optimal configurations, we have long recommended GeForce RTX 4070 12GB graphics cards. Overall, this is a well-balanced model – a top choice for Full HD and a very decent solution for 1440p. At higher resolutions, it becomes challenging to utilize options with maximum quality and enable ray tracing, but simple adjustments typically allow for the desired results and a comfortable fps rate.

With the introduction of the GeForce RTX 4070 SUPER, the price of the base model has officially decreased, making it a very similar offering in terms of price/performance. However, recently, GeForce RTX 4070 models have seen a noticeable increase in price. While a few months ago, starting models were offered from ~$600, now they are closer to $640–650. In such conditions, the GeForce RTX 4070 SUPER 12GB (starting from $720) appears to be a slightly better value in terms of price/performance, offering around a 15% increase in power over the base model. If these expenses are acceptable, then super-modifications are worth considering. However, overall, such a price point goes beyond what is reasonable for a PC of this class.

Among alternatives, the Radeon RX 7800 XT 16GB remains a solid option. Models with decent cooling solutions can be purchased for ~$580–600. Although, it is worth noting a slight increase in the cost of these series models over the past month or two. On average, the Radeon RX 7800 XT lags behind the GeForce RTX 4070 SUPER by 5–10%, particularly in ray-tracing modes, where NVIDIA's GPU offers a significantly greater advantage.

Additionally, one could consider the Radeon RX 7900 GRE 16GB, but recently AMD discontinued models from this lineup. Moreover, they have virtually disappeared from sale in Ukraine.

If there is time, it may be worth waiting for the release of next-generation graphics from the key players in the market. NVIDIA is preparing to launch the GeForce RTX 5070. Unfortunately, the model in this series will again use a 192-bit bus and only receive 12GB of memory. However, the GDDR7 standard with an effective chip clock speed of 28,000 MHz will ensure a bandwidth of 672 GB/s – a third more than the RTX 4070 (504 GB/s). The GB205 graphics processor will have 6,144 cores, providing a quantitative advantage for the new solution. Naturally, the Blackwell architecture also promises additional improvements. The expected power consumption is 250 watts. However, the GeForce RTX 5070 is unlikely to be included in the first wave of RTX 50xx announcements. This adapter is more likely to be introduced in February-March.

At the upcoming CES 2025, AMD is also expected to unveil its new solutions based on the RDNA 4 architecture. However, fewer details are available on specific products at the moment. Just before the announcement, AMD seems to have decided to revise its familiar system of graphic adapter naming. According to the latest leaks, the model that was supposed to be named Radeon RX 8800 XT will be introduced as Radeon RX 9070 XT. The graphics card will feature the Navi 48 GPU with 4,096 cores, a 256-bit bus, 16GB GDDR6 (20,000 MHz), and power consumption of 260+ watts. While it’s too early to talk about expected performance, the name change potentially indicates that this could be a competitor to the RTX 5070/5070 Ti. With the official announcement approaching, we’ll soon learn more details.

Progressive Gaming PC of the Month($1800+)

Gaming Monitor for the "Progressive Gaming PC" from our partner MSI

MSI MAG 321UP QD-OLED – a 32-inch gaming monitor with next-generation QD-OLED panel, resolution of 3840×2160, refresh rate of 165 Hz, and a response time of just 0.03 ms (G-t-G). An ultimate display with expanded color coverage (DCI-P3 99%) and color accuracy Delta E ≤2, along with certifications Clear MR9000 and DisplayHDR True Black 400, and HDMI 2.1 port with VRR and ALLM.

Gaming Monitor MSI MAG 321UP QD-OLED

Processors: For a progressive gaming system on Intel, we have two equivalent platforms with their own advantages and disadvantages. Official shipments of Core Ultra 5 245K/KF (6P+8E; 4.2/5.2 GHz; 3.6/4.6 GHz) started in Ukraine in December, allowing us to consider building a PC on LGA1851. The chip with integrated graphics is offered at ~$360, while the version without iGPU starts at ~$340. More interesting is the modified version with integrated graphics, which now uses Arc Xe-LPG architecture, enhancing the video core capabilities. A 2–2.5 times performance increase during a generation shift is quite rare, which is notable.

The integrated graphics capabilities in the context of a gaming system might be a secondary consideration, but in practice, Arrow Lake features are indeed impressive.

For a PC with the Core Ultra 5 245K processor, a motherboard on Intel Z890 will be required. The cheapest models on this chipset start at ~$220–230, but to have at least a few options to choose from, one should consider around $240–250. Soon, we expect more affordable motherboards on Intel B860, but, as with previous-generation platforms, models on such PCH won’t allow for CPU overclocking. While not a necessity, it is still a useful bonus, especially considering the high-performance cores of Core Ultra 5 245K can be overclocked up to 5.4–5.5 GHz under full load. Overall, to increase the appeal of the new platform, Intel should enable independent overclocking for all chips under LGA1851, or at least not restrict these capabilities on motherboards with Intel B860. The portion of users actively overclocking CPUs isn’t large, but for budget-conscious enthusiasts, it remains an entertaining addition.

A second viable option for progressive Intel configurations remains the combination “Core i5-14600K + Intel Z790.” Despite the brief period of generosity where the processor was priced at $250, the Core i5-14600K in box version can now be freely purchased for around $270. This is still a reasonable price for this CPU.

When directly comparing Core Ultra 5 245K and Core i5-14600K, Intel’s new processor generally offers a slight advantage in productivity tasks. Architectural improvements and increased IPC compensate for the lack of Hyper-Threading technology, ensuring good multi-threaded performance even in Arrow Lake. However, there are some downsides. Initially, new Core Ultra processors do not excel in gaming performance. On average, there is even a slight regression in performance compared to Raptor Lake Refresh models. The difference usually remains within a few percent, but in certain cases, it is more substantial. Though there are counterexamples as well.

Intel officially acknowledged that the launch of Core Ultra processors did not go “according to plan,” and the benchmarks during reviews showed lower results than anticipated. The company is working on improving the situation, but more effort will likely be needed. On the bright side, recent patches, like Cyberpunk 2077’s patch 2.2, managed to boost Arrow Lake’s performance by ~20%.

Ultimately, the choice of a progressive platform depends on whether one wishes to embrace the new platform, relying on further refinement of the problematic aspects, or prefers a stable configuration offered by Core i5-14600K within LGA1700.

Arguments for Core Ultra 5 245K include a better overall IPC, significantly more powerful integrated graphics, the presence of a hardware NPU block, slightly lower power consumption and heat levels, 20 PCI-E 5.0 lanes (for SSD connections without splitting PCI-E x16 5.0 links), and the ability to work with the fastest memory modules. However, the importance of these factors in choosing a gaming platform depends on individual priorities and needs.

Core i5-14600K offers a similar or even slightly higher level of gaming performance at a lower cost. It remains a compelling argument. Not groundbreaking, but effective.

Looking at current retail prices for Ryzen 7 7800X3D and Ryzen 7 9800X3D, we continue to rely on the capabilities of Ryzen 7 9700X (8/16; 3.8/5.5 GHz). Yes, undoubtedly, models with 3D V-Cache are the best options for gaming systems, but $700–850 for an 8-core CPU is well beyond rational purchasing territory. Of course, if there’s a desire/need to get the best, there’s little room for argument here.

In January next year, AMD is set to expand the Ryzen 9000X3D lineup by introducing the 12-core Ryzen 9 9900X3D and the 16-core Ryzen 9 9950X3D. Initially, it was expected that developers would surprise us with the inclusion of additional third-level cache for both cores with computing power, bringing the total L3 capacity to a staggering 192 MB for desktop chips. However, according to recent updates, as in past cases, the 3D V-Cache will only be present on one of the two CCDs.

Ryzen 7 9800X3D will likely remain the fastest gaming processor due to the absence of additional delays in inter-core communication. However, future 12/16-core chips based on Zen 5 will once again elevate the bar for multi-threaded tasks. Utilizing second-generation 3D V-Cache, developers have managed to significantly boost CPU clock speeds with increased L3 cache. It will be interesting to see the actual prices of these new models. Clearly, relying solely on recommended values is not sensible. Even as a starting point.

Graphics Cards. The graphical position for progressive gaming systems is currently filled by GeForce RTX 4070 Ti SUPER 16 GB and Radeon RX 7900 XT 20 GB. These are the entry-level models for this category, which, by the way, have also increased in price. Is additional strengthening needed? It’s never a bad idea, but the question really comes down to cost.

As mentioned earlier, NVIDIA is preparing powerful new GPU solutions based on the Blackwell architecture, which will be showcased on the eve of CES 2025. The presentation is scheduled for January 6. The new lineup will be led by the GeForce RTX 5090 with a monstrous GB202 chip, featuring a die area of 744 mm² and comprising 21,760 cores. Additionally, the GPU will utilize a 512-bit memory bus and GDDR7 memory totaling 32 GB with a bandwidth of 1792 GB/s. Leaks have already surfaced with images of the reference PCB, which will use a 14-layer design. The power delivery subsystem will have a 29-phase configuration (16+6+7), with a total power consumption of 600 W. Naturally, PCI-E 5.0 x16 will be utilized. It seems we are gradually being prepared for the notion that the suggested price for the GeForce RTX 5090 32 GB will significantly exceed $2000.

Given its specifications and potential cost, the GeForce RTX 5090 will nominally be the flagship of the gaming lineup. In reality, it will likely focus on AI models, while simpler solutions will be provided for gamers. For example, the GeForce RTX 5080, which is effectively "half" of the RTX 5090. The GB203 chip features 10,752 cores, a 256-bit memory bus, and 16 GB of memory. Interestingly, in this case, the GDDR7 chips will operate at effective speeds of 30,000 MHz, resulting in a bandwidth of 960 GB/s. The power consumption will be 400 W. According to preliminary information, the GeForce RTX 5080 will be the first to hit the market, followed by the top-tier model.

Another interesting first-wave Blackwell GPU is the GeForce RTX 5070 Ti. It will also use the GB203 processor but with 8960 active cores. The adapter will maintain a 256-bit memory bus and will come equipped with 16 GB of GDDR7 (896 GB/s). Maximum power consumption is expected to be 300 W. The GeForce RTX 5070 Ti looks like a potentially attractive option for upgrading a progressive configuration. However, we will wait for confirmation of specifications, pricing, and actual performance benchmarks.

Alongside the new GeForce RTX 50 series, NVIDIA is likely to introduce the next generation of DLSS, along with a fundamentally new technology called "Neural Rendering." We will learn more details very soon.

AMD is likely not aiming to compete directly with NVIDIA's top-tier solutions, instead focusing on more affordable graphics cards. As mentioned, the company will soon unveil the Radeon RX 9070 XT 16 GB, which seems poised to lead the new GPU lineup at its initial stage. While specific capabilities are yet to be detailed, previous information on RDNA 4.0 models points to significant improvements in ray tracing performance.

Alongside the new hardware, developers are preparing the next generation of AMD FidelityFX Super Resolution. This likely means the introduction of FSR 4.0 with more advanced AI algorithms. As a result, alongside technical advancements, there is a growing emphasis on technological competition, increasingly incorporating elements of artificial intelligence. The rapid implementation of these developments is impressive, and the future holds even more exciting innovations.