Nvidia RTX 2060 Super
|GPU Engine Specs||2176 NVIDIA CUDA® Cores|
|Memory Specs||14 Gbps Memory Speed|
|Technology Support||Ray Tracing Cores|
|Display Support||7680x4320 Maximum Digital Resolution|
The NVIDIA GeForce RTX 2060 SUPER is constructing the NVIDIA Turing architecture, which provides gamers and programmers with ultra-fast all-around performance and graphics.
It's also less costly and more available than the RTX 2070 was when it first came out, and the Nvidia GeForce RTX 2060 Super delivers nearly the same performance. We're left wondering why Nvidia didn't announce this card earlier.
NVIDIA's GeForce RTX 2060 SUPER supports DirectX 12 Ultimate and is a 12 nm process based on the TU106 graphics processor in its TU106-410-A1 version. All modern games will run on the GeForce RTX 2060 SUPER as a result of this.
Besides, the DirectX 12 Ultimate capability ensures that future video games can support hardware-raytracing, variable-rate shading, and other features. With a die area of 445 mm2 and 10,800 million transistors, the TU106 graphics processor is a big chip.
There are 272 tensor cores, which enable machine learning applications to run faster. There are also 34 raytracing acceleration cores on the card. The GeForce RTX 2060 SUPER and 8 GB GDDR6 memory connect through a 256-bit memory interface by NVIDIA. The GPU operates at a frequency of 1470 MHz, which increases to 1650 MHz, and the memory runs at 1750 MHz (14 Gbps effective).
The NVIDIA GeForce RTX 2060 SUPER is a dual-slot card. Remember here is that Nvidia says the RTX 2060 Super can compete with the original RTX 2070 in terms of functionality – it also outshines it in some of our synthetic workloads.
The Nvidia GeForce RTX 2060 Super boasts a slew of enhancements, particularly in terms of core numbers and VRAM. The Nvidia GeForce RTX 2060 Super has 2,176 CUDA cores and 1,920 on the RTX 2060.
They're much faster than before, with a boost clock of 1,650 MHz. These enhancements alone account for the increase in raw power to 7.2 TFLOPs. The graphics card now has 34 RT cores, up from 30 previously, enabling the RTX 2060 Super to offer 6 Giga Rays/s of ray tracing efficiency.
Furthermore, the Nvidia GeForce RTX 2060 Super has 8GB of GDDR6 VRAM, compared to 6GB on the original edition. Make the graphics card more future-proof, as well as make 1440p gaming more feasible.
All of this comes with an increase in power consumption, with the new Turing card drawing 175 Watts.
The output difference between the RTX 2060 Super and the original model is even more pronounced in sports. On the RTX 2060, we could get 58 frames per second in Metro Exodus, compared to 53 on the RTX 2060.
- Better value than other RTX cards
- More, faster memory than before
- Superb Founders Edition design
- DVI port for legacy monitors
- Didn't inherit the 2060 price
- Ray traced games picking up steam, but still relatively rare
- Thermal solution blows waste heat back into your case
NVIDIA GeForce GTX 1080
|GPU Engine Specs||2560 NVIDIA CUDA® Cores|
|Memory Specs||10 Gbps Memory Speed|
|Technology Support||Simultaneous Multi-Projection|
|Display Support||7680x4320@60Hz Maximum Digital Resolution|
The GeForce GTX 1080 is NVIDIA's award-winning Pascal architecture. Discover incredible gaming, performance, and energy efficiency. GTX 1080 is the most advanced gaming platform accessible.
NVIDIA's GeForce GTX 1080 is based on the GP104 graphics processor and built on a 16 nm process, supports DirectX 12 in its GP104-400-A1 version. All new systems will run on the GeForce GTX 1080 as a result of this.
The GeForce GTX 1080 is a pack with 8 GB of GDDR5X memory connected through a 256-bit memory interface. The GPU runs at 1607 MHz, which can be enhanced to 1733 MHz, while storage runs at 1251 MHz (10 Gbps effective).
This NVIDIA GeForce GTX 1080 is a dual-slot chip that uses one 8-pin power connector with a peak power draw of 180 W. 1x DVI, 1x HDMI, and 3x DisplayPort are among the display interfaces.
A revised mist chamber and vent for excellent, easy storage; a polygonal cloak; and a CPU cooler for heat dissipation from the graphics card's back surround the GeForce GTX 1080's powerful hardware. This high-end design allows for superior overclocking and whisper-quiet activity.
There are two billion more transistors in the GeForce GTX 1080. The GeForce GTX 980, the initial counterpart to the GeForce GTX 1080, had 5.2 billion transistors.
What's impressive is that, despite having so many more transistors, the GeForce GTX 1080's thermal design power (TDP) rating is just 15 watts higher than the GeForce GTX 980s. It consumes 180 watts, compared to 165 watts for the GeForce GTX 980.) What's more impressive is that the GeForce GTX 1080 should be significantly more powerful than the GeForce GTX 980 while requiring substantially less power.
The GeForce GTX 1080 has a base clock speed of 1.6GHz, with a boost clock of 1.73GHz in some circumstances. We should expect the cards to go even higher than this, as Nvidia is often cautious when stating the limit of these baseline clocks.
Though DisplayPort 1.2 can accommodate a 4K (3,840 x 2,160) monitor at 60Hz, version 1.3 increases the channel's bandwidth enough to support a 4K monitor at 120Hz, and version 1.4 increases it even further to support nascent 8K at 60Hz.
Nvidia shot scenes at 20x magnification in a preview for the press before the GTX 1080's launch, and the resulting file was 3GB in size and 46,000 pixels wide, or "46K" to use the standard terminology.
- Outrageous performance leap over GTX 980
- Excellent performance
- Hugely power efficient
- HDMI 2.0b, DisplayPort 1.4
- Fans don't turn off in idle
- Doesn't blow away Radeon cards in heavily AMD-optimized games
XFX RX 5700 XT Raw II 8GB
|Base Frequency||1465 MHz|
|Boost Frequency||Up to 1725 MHz|
|Memory Speed||14 Gbps|
|Max Memory Size||8 GB|
The XFX RX 5700 is a fantastic twist at its price point, taking 1440p gaming at Ultra settings to the masses. It's without any second thought one of the best graphics cards available today.
In our view, the best graphics card would provide outstanding performance at a fair price. You can have all the features you want, but anything else will fall apart if you don't have the crucial price-to-performance ratio. Fortunately, the XFX RX 5700 follows this rule, providing decent output at an affordable price.
Even with all the settings switched on, you'll be able to enjoy all of the latest PC games at 1440p with this graphics card. At the same price point, the RX 5700 offers around 2GB more VRAM than the Nvidia GeForce RTX 2060, making the pricing structure even sweeter.
This built-for-gaming graphics is based on an architecture that boosts performance-per-clock by 1.25x right out of the gate, instantly extracting more performance from the graphics card at stock settings. When you consider the higher core counts of these GPUs, it's no surprise that the AMD Radeon RX 5700 outperforms anything like the Radeon RX 580.
The addition of PCIe 4.0 is another major update to these graphics cards. This will only process on AMD Ryzen 3rd Generation systems with an X570 motherboard, but the increased bandwidth can boost graphics efficiency.
If your favorite game doesn't have this new contrast sharpening technology, don't worry: Radeon Image Sharpening, or RIS, is available in almost every game.
Additionally, AMD provides its Radeon Anti-Lag program for any esports gamers out there, which will allow the CPU and GPU to function in sync, reducing input lag.
In 3DMark TimeSpy extreme, for example, the AMD Radeon RX 5700 blew the Nvidia GeForce RTX 2060 out of the water, scoring 3,753 versus 3,413 for Team Green. And it didn't stop there: the 5700 scored 5,355 points in Firestrike Ultra, compared to 4,352 for the RTX 2060. When it comes to synthetics output in this price range, the Radeon RX 5700 comes out on top.
This Graphics card even applies to real sports. The AMD Radeon RX 5700 achieved a rock-solid 60 frames per second in Metro Exodus at 1080p Ultra, compared to the RTX 2060s 53 frames per second. The AMD Radeon RX 5700 is a no-brainer with that kind of in-game result at this price point.
The RX 5700 is an excellent choice if you want to enjoy the latest PC games at 1440p while still saving money. It won't be able to run your games at 4K 60 fps, but that's not what it's designed for. The AMD Radeon RX 5700 is one of the best AMD graphics cards on the market today.
- Cheaper than the competition
- PCIe 4.0 support makes it future proof
- Best-in-class power efficiency
- Excellent gaming performance
- No real-time ray tracing
- Blower-style cooler has higher temps than Nvidia's FE design
XFX RX 5700 Xt
|Base Frequency||1605 MHz|
|Boost Frequency||Up to 1905 MHz|
|Memory Speed||14 Gbps|
|Supported Rendering Format||4K H264 Decode|
The XFX RX 5700 Xt must have a winning combination of gaming success and a long list of impressive features. The AMD Radeon RX 5700 XT remains among the best graphics cards today, even though Big Navi is set to release on October 28 and boasts a menu that could offer twice the capacity.
The Radeon RX 5700 XT is the higher of the two latest graphics cards based on a 7-nanometer (7nm) manufacturing process that aims to dampen Nvidia's dominance in the midrange sector.
After running it through our benchmarks, we can see that this card is up to the task. It's a little hotter than its Nvidia equivalents (the RTX 2060 and RTX 2070 Supers), and early overclocking results aren't lovely.
First and foremost, there's the issue of control. AMD cards have long been the more sensitive, more profound, and power-hungrier cards at any given level made by the two major GPU manufacturers, but the Navi cards show gains here. In contrast, the AMD Radeon RX 5700 XT consumes only 225 watts of power at peak performance, compared to 295 watts for its predecessor card, the Radeon RX Vega 64 from 2017.
AMD claims that its latest RDNA-based cards will provide 1.5 times the output per watt of cards based on the GCN architecture. Another impressive AMD claim: The 14nm Vega 10 chips take up 495mm2 of space on a die, while the first round of Navi cards, based on the latest 7nm manufacturing process.
The RX 5700 XT is quieter than the RTX 2060 Super, but it consumes about 25W more power at the outlet and runs about 11C hotter. It's not the end of the planet, but by contrast, Nvidia comes out on top.
The RX 5700 XT can comfortably accommodate 1080p ultra and even 1440p ultra. There are a few games where you'll have to lower the settings to get a steady 60fps at 1440p (e.g., Assassin's Creed Odyssey), but for the most part, the 5700 XT is capable of 60fps, and in some cases, even 100fps or more. When you pair the RX 5700 XT with a FreeSync monitor, any game that can reach 50 frames per second or higher will function properly.
The other value is for a brand new PC, which is where the RX 5700 XT shines. The RX 5700 XT ranks first in our device value charts for all three monetary markets I tested, only ahead of the 2070 Super, RX 5700, and 2060 Super. Outside of the budget domain, I'd strongly advocate going for a high-end GPU like AMD's RX 5700 XT or one of Nvidia's new RTX Super cards if you're building or planning to create a new gaming PC sometime soon.
The Radeon RX 5700 XT from AMD is a beautiful addition. You won't get fancy ray tracing support, but if you've been looking for a new graphics card that doesn't need Nvidia support, this is the best AMD has to give, and it's well worth exploring.
- Great mid-range 1440p graphics performance
- Easily faster than RTX 2060 Super
- Good design
- No idle fan stop
- Poor overclocking potential
MSI GeForce GTX 1050 TI
|NVIDIA CUDA||Cores 768|
|Frame Buffer||4 GB GDDR5|
|Memory Speed||7 Gbps|
With the launch of a more budget-oriented piece - the GTX 1050 Ti - Nvidia's Pascal architecture continues its top-down steamrolling of the graphics product stack.
Not only does it easily outperform the previous-generation GTX 950, but it also saves some power over the low-powered Maxwell card, which didn't need any additional PCI Express power. The GTX 1050 Ti matches the performance of the GTX 960 while also providing class-leading quality.
The GTX 1050 Ti will bring you to 1080p60 in most games with some tuning, but this GPU class is more about matching ballpark console performance with better graphics and more stable frame rates.
The GTX 1050 Ti is a fully-loaded variant of the GP107 with 768 CUDA core processors (equivalent to the last-gen GTX 950) and 7gbps GDDR5 memory fed by a 128-bit memory interface. But the real show-stealer here is that the card itself is minimal - depending on the manufacturer - and needs no additional PCI Express power in standard configurations.
This is critical. Importantly, as with the GTX 750 Ti and an inadequate quantity of GTX 950s before it, this opens the door to a robust GPU rise that will fit into the vast bulk of PCs while requiring no additional PSU energy.
All runs on the meager 75W supplied by the motherboard slot, and we were blown away by how much output we were able to squeeze out of such a small power supply.
One thing that instantly points out is that the boost clock spec is 300MHz lower than what we've come to expect from the Pascal line, owing to the low 75W TDP. However, we're pleased to say that in practice, the spec is exceptionally conservative.
The Zotac board we received happily boosts to 1650MHz without breaking a sweat right out of the package and overclocking taps out at 1850MHz without hitting the hard 75W cap. As a result, the GTX 1050 Ti outperforms the last-gen GTX 960 right out of the box, and its overclocking potential may surpass the 1850MHz we achieved with this Zotac model, assuming you get a GTX 1050 Ti with an additional six-pin PCI Express power input.
MSI has also guaranteed that the GTX 1050 Ti has sufficient VRAM to complete the task. We've been pleasantly impressed by how well the older GTX 950 and GTX 960 cards have done with just two gigs of onboard memory.
The excellent efficiency is beneficial in other areas as well. The 768-core Pascal-based GTX 1050 Ti easily outperforms the 768-core Maxwell-based GTX 950.
The GTX 1050 Ti's distinct features pay off handsomely. It's a tiny card that runs almost silently, and the fact that it runs so cool - even with a very simple heat sink and fan - means it can happily sit in a tiny PC with little or no risk of exceeding its thermal limits.
In terms of power consumption, the GTX 1050 Ti looks fine.
One thing to keep in mind is that the latest GP107 found in the GTX 1050 Ti uses a different manufacturing technology than the other Pascal processors, namely Samsung's 14nm FinFET process 16nmFF method used in the others. On the other hand, the high overclocking results show that Pascal scales well across competing data collected.
- Extremely efficient
- Excellent price
- Didn't found any