Today, we’re reviewing the NVIDIA Titan RTX for overclocking, gaming, thermal, and acoustic performance, looking at the first of two cards in the lab. We have a third card arriving to trade for one defective unit, working around the 1350MHz clock lock we discovered, but that won’t be until after this review goes live. The Titan RTX costs $2500, outbidding the RTX 2080 Ti by about 2x, but only enables an additional 4 streaming multiprocessors. With 4 more SMs and 256 more lanes, there’s not much performance to be gained in gaming scenarios. The big gains are in memory-bound applications, as the Titan RTX has 24GB of GDDR6, a marked climb from the 11GB on an RTX 2080 Ti.
An example of a use case could be machine learning or deep learning, or more traditionally, 3D graphics rendering. Some of our in-house Blender project files use so much VRAM that we have to render instead with the slower CPU (rather than CUDA acceleration), as we’ll run out of the 11GB framebuffer too quickly. The same is true for some of our Adobe Premiere video editing projects, where our graph overlays become so complex and high-resolution that they exceed the memory allowance of a 1080 Ti. We are not testing either of these use cases today, though, and are instead focusing our efforts on the gaming and enthusiast market. We know that this is also a big market, and plenty of people want to buy these cards simply because “it’s the best,” or because “most expensive = most best.” We’ll be looking at how much the difference really gets you, with particular interest in thermal performance pursuant to the removal of the blower cooler.
Finally, note that we were stuck at 1350MHz with one of our two samples, something that we’ve worked with NVIDIA to research. The company now has our defective card and has traded us with a working one. We bought the defective Titan RTX, so it was a “real” retail sample. We just wanted to help NVIDIA troubleshoot the issue, and so the company is now working with it.
We’ve covered one of Thermaltake’s Level 20 cases before, specifically the small form factor VT, which sought to bring the ultra-expensive Level 20 line down to ‘normal’ consumers. The Level 20 MT is a mid-tower in the same style, pairing rounded silver edges with flat tempered glass panels to equate “class,” or something, while overlooking some basic design concepts. We’re specifically reviewing the Level 20 MT ARGB, which comes with three 120mm ARGB intake fans at the front.
The front panel is restrictive, with tiny 1cm strips of mesh on either side of the glass section. Deceptively, these strips aren’t in the path of airflow and don’t act as filters. There isn’t any filtration at all in front of the fans, which instead pull air through the narrow gaps behind the edges of the front glass panel. There’s also a wider gap hidden at the bottom of the front panel, typical for cases with sealed designs like this.
The comments section of our Walmart case review and system review tell the story of what people think of Great Wall: everyone is expecting a fire, as the shell of the PSU is uninspiring, its rating sticker is lacking some metrics (maximum 12V capabilities, for example), and the brand isn’t familiar to a western audience. The funny thing is that this would be sort of similar to hearing “Asetek” for the first time, then making fun of it for being foreign to the market. Asetek supplies almost all of the closed-loop liquid coolers currently popular in North America, but never sticks its own branding on those. Great Wall is also a supplier, including being a supplier to brands viewed generally positively in the Western market.
To be fair, everything about the Great Wall 500W 80 Plus PSU does look like a cheap power supply – and it is cheap – but there’s nothing that should indicate this is an exploding power supply. Great Wall’s association with Walmart here is probably hurting their brand more than the inverse, funny enough, but we’ll be digging into that today.
We previously mentioned that Great Wall actually is a supplier and makes PSUs for Corsair, for instance, as discussed in our Walmart case review. It’s uncommon to find Great Wall PSUs unbranded, and this one didn’t even have the maximum 12V capabilities listed, so this unit did attract criticism from the community. What we’re here to do is test whether it’s deserving of that criticism, using our power supply testing setup to benchmark efficiency, ripple, and over-current protections.
The Intel i7-2600K is arguably one of the most iconic products released by Intel in the last decade, following-up the seminal Nehalem CPUs with major architectural improvements in Sandy Bridge. The 2600K was a soldered CPU with significant performance uplift over Nehalem 930s, and launched when AMD’s Phenom II X6 CPUs were already embattled with the previous Intel architecture. We revisited these CPUs last year, but wanted to come back around to the 2600K in 2018 to see if it’s finally time to upgrade for hangers-on to the once-king CPU.
Our original Intel i7-2600K revisit (2017) can be found here, but we had a completely different test methodology at the time. The data is not at all comparable.
The 2600K CPU was manufactured starting around 2009-2010, launching alongside the Intel Sandy Bridge 2nd Gen Core i-Series of CPUs. This launch followed Nehalem, which challenged the Phenom II X6’s appeal in a heated market. Sandy Bridge launched and has remained a well-respected, nearly idolized CPU since its launch. Intel made tremendous gains over Nehalem and hasn’t quite recaptured that level of per-core increase since. For everyone still on Sandy Bridge and the i7-2600K (or i7-2700K), we wanted to revisit the CPUs and benchmark them in 2018. These 2018 i7-2600K benchmarks compare against Ryzen (R7 2700 and others), the i7-8700K, and the i9-9900K, alongside several other CPUs. For anyone with a 2700K, it’s mostly the same thing, just 100MHz faster.
The AMD Athlon 200GE CPU enters our benchmarking charts today, but we’re reviewing this one with a twist: For this benchmark, we’re testing the CPU entirely as a CPU, ignoring its integrated graphics out of curiosity to see how the $55 part does when coupled with a discrete GPU. To further this, we overclocked this supposedly locked CPU to 3.9GHz using multiplier overclocking, which is disabled by AMD on most boards likely for product segmentation of future 200-series parts. In this instance, the 200GE at 3.9GHz posts significantly improved numbers over stock, making it a candidate to replace the retired price position once held by the Intel Pentium CPUs, at least, up until the 14nm shortage.
In the past, the Intel G3258 and successor G4560 stood as affordable options for ultra-budget builds that were still respectable at gaming tasks. The Pentium G5000 series – including the G5400 and G5600 (in this today’s benchmark) – has skyrocketed in price and dwindled in availability. The G5600 and G5400 alike are in the realm of $100, depending on when you check pricing, with the G5400 often ending up more expensive than the G5600. A lot of this is due to demand, but supply is also weak with the ongoing 14nm shortage. Intel is busy allocating that fab space to other products, minimizing the amount of Pentium G CPUs on market and allowing retailers control to boost prices and meet what demand will pay. This has left a large hole in the market of low-end CPU + low-end dGPU solutions, and that’s a hole which AMD may be able to fill with its Athlon 200GE solution, which had a launch MSRP of $55.
Unlike Ryzen proper chips, the 200GE includes an IGP (Vega graphics) that enables it as a fully standalone part once popped into a motherboard; however, we think its IGP is too weak for most of our normal testing, and we know it’d underperform versus the R3 2200G. The G4560-style market is one we like to look at, so we decided to test the 200GE as an ultra-budget replacement for coupling alongside a low-end dGPU, e.g. a GTX 1050 or RX 550/560. If the CPU holds up against our standardized test battery, it’ll work when coupled with a low-end GPU.
We already made known our feelings of Walmart’s complete system build quality, but now we’re beginning to delve into individual and isolated component quality. We’re starting with the case, with plans to move on to the PSU next.
The Walmart case isn’t presently available as a standalone product, but it is sourced from a common supplier (much akin to the Jonsbo relationship with Rosewill and others), and so could go mainstream should a manufacturer find it worthwhile. What we’re really doing is an academic exercise to evaluate the quality of this case, including thermal tests, ease-of-installation discussion, and noise testing. This can’t be purchased separately, primarily rendering this piece as a secondary look at the overall component quality and choice for Walmart’s Overpowered DTW gaming PCs (DTW1, DTW2, DTW3).
Today we’re reviewing the Intel i5-9600K CPU, a 6-core 8th-Generation refresh part that’s been badged as a 9000-series CPU. The 9600K is familiar to the 8600K, except soldered and boosted in frequency. The new part costs roughly $250 and runs at 3.7GHz base or 4.6GHz turbo, with an all-core closer to 4.3GHz, depending on turbo duration tables. When we last reviewed an i5 CPU, our conclusion was that the i7s made more sense for pure gaming builds, with the R5s undercutting Intel’s dominance in the mid-range. We’re revisiting the value proposition of Intel’s i5 lineup with the 9600K, having already reviewed the 9900K and, of course, the 8700K previously.
As a foreword, note that the R5 2600's current and maintaining price-point of $160 makes it a less direct comparison. The 2600X, which would perform about where an overclocked 2600 performs, is about $220. This is also cheaper, but still closer to compare. Even closer is the R7 2700, which is $250-$270, depending on sales. The 2700 maintains at about $270 when sales aren't active. The most fair comparison by price would be the 2700, then, not the by-name comparison with the R5 2600(X) CPUs.
AMD launched its RX 580 2048 silently in China a few months ago, and in doing so damaged its brand credibility by rebranding the RX 570 as an RX 580. The point of having those two, distinct names is that they represent different products. The RX 580 2048 has 2048 FPUs (or streaming processors), which happens to be exactly what the RX 570 has. The RX 580 is also a few MHz higher in clock, which is fully attainable with an overclocked RX 570. Working with GamersNexus contacts in Taiwan, who then worked with contacts in China, we managed to obtain this China-only product launch so we could take a closer look at why, exactly, AMD thinks an RX 570 Ti deserves the name “RX 580.”
Taking an existing product with a relatively good reputation and rebuilding it as a worse product isn’t new. Don’t get us wrong: The RX 570, which is what the RX 580 2048 is, is a reasonably good card, especially with its new prices of roughly $150 (Newegg) to $180 elsewhere. That said, an RX 580 2048 is, by definition, not an RX 580. That’s lying. It is an RX 570, or maybe an RX 575, if AMD thinks that a 40MHz clock difference deserves a new SKU. AMD is pulling the same deceitful trick that NVIDIA pulled with its GT 1030 DDR4 card. It’s disgraceful, misleading, and predatory of consumers who may otherwise not understand the significance of the suffix “2048.” If they’re looking for an RX 580, they’re still finding one – except it isn’t one, and to brand the RX 580 2048 as an RX 580 is disgraceful.
We have a separate video scheduled to hit our channel with a tear-down of the card, in case you’re curious about build quality. Today, we’re using the DATALAND RX 580 2048 as our vessel for testing AMD’s new GPU. Keep in mind that, for all our scorn toward the GPU, DATALAND is somewhat unfortunately the host. DATALAND didn’t make the GPU – they just put it on the PCB and under the cooler (which is actually not bad). It also appears that DATALAND (迪兰) works alongside TUL, the parent company to PowerColor.
We paid about $180 USD for this card, which puts it around where some RX 570s sell for (though others are available for ~$150). Keep in mind that pricing in China will be a bit higher than the US, on average.
We thought we were getting the DTW3 – Walmart’s new $2100 gaming PC – but the company instead shipped its $1400 model while still charging us an extra $700 for parts we didn’t receive. What we ended up with was a GTX 1070, an i7-8700, an H310 motherboard with half the bus speed of any other chipset, and 16GB of 2400MHz RAM for nearing $2300 (after taxes and shipping).
What a rip-off.
But we knew it’d be a rip-off when we placed the order, we just didn’t know it’d be a rip-off of such unchallenged proportions. Even if we assume that our receipt of a SKU $700 down-ticket was an honest mistake – and Walmart has agreed to replace it (after they get it back, so a 2-week window) – it’s still just an awful selection of components. The video below shows our genuine first reactions to this product, the Overpowered DTW3 by Walmart (by eSports Arena, by someone else), but the article will really dig in deep. Continue reading (or watch below) for more information.
The Intel i9-9980XE is a revamped i9-7980XE with solder and higher out-of-box clocks. It’s also got much higher out-of-the-box thermals as compared to a delidded 7980XE, as you’ll see in our testing, and is disappointingly limited in its overclocking headroom when using practical cooling solutions. The 9980XE should effectively be a higher clocked 7980XE with a better stock cooling interface and could be a good candidate for future streams where we RIP YouTube personalities. That is, it would be with chilled water on top of it, whereas the 7980XE has more thermal headroom out of the delid tool. Regardless, we have full benchmarks of this new CPU, including perspectives from both the enthusiast overclocker’s viewpoint and the professional user’s viewpoint. Testing includes overclocking, thermals, Photoshop, Premiere, Blender, gaming, power, and more.
Differences between the 7980XE and 9980XE are relatively minimal when compared to launches with new architectures. The 9980XE functionally is a 7980XE, it’s just soldered and faster – a pre-overclock, more or less. We immediately ran into overclocking limitations on the X299 DARK and Gigabyte Gaming 9 motherboards alike, the former of which has been used by our team to claim (fleeting) TimeSpy world records. These limitations stemmed from a lack of thermal headroom, something our delidded 7980XE doesn’t face to the same degree.
The X299 DARK was used for overclocking tests and the Gigabyte Gaming 9 was used for 'stock' tests, although its MCE toggle apparently does nothing. We used the latest BIOS for each motherboard. Additional test methodology information is in our 9900K review.
We moderate comments on a ~24~48 hour cycle. There will be some delay after submitting a comment.