ASUS ROG STRIX GTX 1660TI A6G GAMING Review – Specs | Test | Overclocking | Perfomance – Taking a break from the GTX 10 series generation in its mid-range, NVIDIA has launched the GTX 16 series models, which bring on board the new Turing architecture, although these without Tensor cores, unlike the RTX 20 series models .
To show the potential of these new NVIDIA models, ASUS has allowed us to use one of its copies for this series and not just any copy, but one of its most glamorous family in video cards, the 6GB ASUS ROG STRIX GTX 1660 Ti in the Advanced version.
Design of axial technology fans
The ROG Strix GeForce® GTX 1660 Ti is equipped with three powerful axial technology fans. Recently designed for the latest generation of high-end ROG graphics cards, these fans feature a smaller center for larger blades and an outside ring to increase air pressure in the direction of the heatsink.
Dual BIOS
The performance mode will allow enthusiasts to keep the card fan running all the time. Silent mode maintains the same power and top-end settings, but offers a less aggressive ventilation curve for quieter operation at medium temperatures. For even more customization and control, check out the Tweak II GPU software.
GDDR6 high performance memory subsystem
Turing is the first GPU architecture compatible with GDDR6 memory. GDDR6 is the next major advance in the design of DRAM GDDR high bandwidth memory. The GDDR6 memory interface circuits in Turing GPUs have been completely redesigned for speed, energy efficiency and noise reduction, achieving transfer speeds of 12 Gbps with an improved energy efficiency of 20% compared to the GDDR5X memory used in Pascal’s GPUs.
Specs
| GPU model | RADEON VII | NVIDIA RTX 2080 FE | NVIDIA GTX 1080Ti | NVIDIA RTX 2070 FE | NVIDIA RTX 2060 FE | NVIDIA GTX 1660 Ti | ASUS STRIX GTX 1660 Ti A6G GAMING | NVIDIA GTX 1060 |
|---|---|---|---|---|---|---|---|---|
| Fabrication process | 7 nm | 12 nm | 16nm | 12 nm | 12 nm | 12 nm | 12 nm | 16nm |
| Codename | Vega 20 | Turing | Pascal | Turing | Turing | Turing | Turing | Pascal |
| GPU Core | Vega20 | TU104-400 | GP102 | TU106-400A-A1 | TU106 | TU116 | TU116 | GP106 |
| CUDA Cores | 3840 | 2944 | 3584 | 2304 | 1920 | 1536 | 1536 | 1280 |
| Texture Units | 240 | 184 | 224 | 144 | 120 | 96 | 96 | 80 |
| Tensor Cores | 368 | 288 | 240 | |||||
| ROPs | 64 | 64 | 88 | 64 | 48 | 48 | 48 | 48 |
| Core / Boost Clock | 1400 MHz / 1750 MHz | 1515 MHz / 1800 MHz | 1481 MHz / 1582 MHz | 1410 MHz / 1710 MHz | 1365 MHz / 1680 MHz | 1500 MHz / 1770 MHz | 1500 MHz / 1800 MHz | 1506 MHz / 1708 MHz |
| Memory clock | 2 Gbps | 14 Gbps | 11 Gbps | 14 Gbps | 14 Gbps | 12 Gbps | 12 Gbps | 8 Gbps |
| Memory | 16 GB HBM2 | 8 GB GDDR6 | 11 GB GDDR5X | 8 GB GDDR6 | 6 GB GDDR6 | 6 GB GDDR6 | 6 GB GDDR6 | 6 GB GDDR5 |
| Memory Bus | 4096-bit | 256-bit | 352-bit | 256-bit | 192-bit | 192-bit | 192-bit | 192-bit |
| Power Connectors | 2x 8 pin | 1x 8 pin + 1x 6 pin | 1x 8 pin + 1x 6 pin | 1x 8 pin | 1x 8 pin | 1x 8 pin | 1x 8 pin | 1x 6 pin |
| TDP | 300 W | 225W | 250W | 175W | 160W | 120W | 120W | 120W |
| Price | MSRP: $ 699 | MSRP FE: $ 799 MSRP “NO FAITH”: $ 699 | MSRP: $ 699 | $ 490 | $ 350 | $ 280 | $ 200 |
First look.
For both the RTX 20 and GTX 16 versions, ASUS continues to maintain the same triple fan cooling system for the STRIX series.
Subsequently a smooth backplate, unlike the backplates with relief and with the “eye” of STRIX illuminated that has the RTX 20 models.
A video card with a length of almost 30 cm (29.8 cm) from the bracket to the rear edge of the cooling system, which exceeds 1 centimeter and fraction in relation to the PCB.
Auxiliary power is provided by an 8-pin connector.
Where is usually the bridge for SLI (technology that is not found in the GTX 1660 Ti), we find a BIOS selection switch (Yes, this model has dual BIOS), which shows
Video outputs consist of 2x HDMI and 2x DisplayPort
Under the cooling system, we find an aluminum block that covers the temperature sensitive areas, from memories to VRMs. We can see the space of a couple of missing memories, this PCB would be occupied in 8GB models too.
The cooling system has 3 heat pipes that pass through the contact block to the GPU.
The PCB shows a low density of electrical components, the consumption of this GPU does not show to be so high as to add a large number of these elements even being a STRIX model, that is why some ITX versions of this GPU can be found in the market .
The backplate seems to provide mainly protection rather than refrigeration, since it does not show the presence of any thermal pad, unlike the previous block that has pads for what are the memories and VRM of the video card, the latter would deliver rigides and cooling.
The configuration of the power phases in this model is 6 + 2.
GDDR6 memory chips are manufactured by Micron with the 8XA77D9WCR model.
Testing and Methodology Platform.
| Test Platform | |
|---|---|
| Processor | – Intel Core i7 8086K |
| Motherboard | – ASUS ROG Maximus X Apex |
| Memories | – G. Skill TridentZ 2x8GB 3200 MHz |
| Refrigeration | – EK-XLC Predator 240 |
| Graphics card | – ASUS ROG STRIX GTX 1660Ti A6G GAMING |
| Power supply | – Corsair RM1000X |
| Storage | – SAMSUNG 960PRO SSD M.2 512GB |
| Monitor | – ASUS MG28UQ |
- Windows 10 Pro x64 operating system.
- The tests were performed in an environment with a temperature of approximately 25 ° C.
- The platform was used without a cabinet.
- The drivers used for the NVIDIA graphics cards: 425.31 WHQL
- The resolutions of the synthetic tests are the default ones for each of the benchmarks.
- The real test resolutions are 1920 × 1080 and 3820 × 2160 with all the maximum available graphics in each game.
Synthetic tests
Real tests
Overclocking.
For mid-range models, overclocking is a tool that delivers a great differentiation within the dense competition found in this segment. With this GTX 1660Ti model, we experiment with frequencies of 1600 MHz in the GPU and 1600 MHz in the memories, clearly not taking it to such an extreme scenario simply to see how it scales with these increases in its frequency.
The increase of these 100 MHz in both its memories and GPU, gave us an extra performance of 5%, the video card can easily cope with an increase in frequency over what was experienced in this section, did not present any instability and not even the use of extra voltage to the one you have by default. If the performance of this model easily achieves a satisfactory experience in large titles at resolutions 1080p including 2K with some slight changes in the configuration of the games, which is 5% extra (we could even say that up to 8% extra is possible to reach) , never are others.
Temperature.
Focusing on the results of maximum load, the temperatures slightly exceed 50ºC both in overclocking and in load without this and fans in automatic. In load, the fans automatically go around 45% of the maximum speed they have, so both the low GPU consumption and the robust cooling system deliver a video card with enviable attributes, allowing frequencies to be reached Under higher overclocking without problem.
Consumption.
The difference in the consumption of the video card between charging with and without overclocking is virtually imperceptible. The extra 5% of performance delivered with the overclock barely increases by 2W at maximum load, which shows that it has great potential to design models with greater performance from the factory over time thanks to overclocking.
Conclusion.
Performance by consumption is one of the great attributes that NVIDIA has managed to develop these latest GPUs, this has given it the possibility of designing very attractive models in the market and next to this give the assemblers the possibility of experimenting with new designs.
In what is design, despite creating certain differences such as axial fans with “joined blades” or the increase in the thickness of the heatsink, for what is grossly aesthetic remains the same model seen throughout the STRIX series GTX 10 series, which for some enthusiast ceases to be so attractive this, considering the changes that other video card manufacturers have experienced.
Clearly, consumption factors and low temperatures provide a favorable scenario for overclocking to bear great results, which clearly in this STRIX model we can find. ASUS in the overclocking discipline has always been among the first, stability under overclocking thanks to its robust electrical and cooling system give you the essential tools to stand out in this area, which for this GTX 1660 Ti model would achieve great results.
When we talk about price clearly ASUS is not our first options, in this manufacturer we have a series of elements of their technologies that undoubtedly give added value to their products, however when we look at the price comparison in our market for this GTX 1660 Ti model, we have ASUS far above models with more attractive qualities than those offered by this STRIX model, so the enthusiasm for acquiring this model is somewhat reduced.
It is very difficult not to recommend a model today, considering that manufacturers know very well the weaknesses and strengths of what today is to design a product. In what could affect in some cases is to deliver a product that its price is not agreed to the segment that you want to implement, but no doubt that scenario occurs very rarely.
Related posts:
Bitmain Antminer r4 Review – Benefits Profitability Payback Specification 
Nvidia GTX 1080 ti Hashrate – Review | Overclocking |Ethereum |Profit (sky high) 
Antminer S5 ASIC Hashrate – Review Profitability Payback Period 
Asic Antminer S1 Hashrate – Review | Test| Specs| Pros & Cons | Set-up| 
Miner Baikal Giant x11 – Hashrate|Profitability| Specs 
Radeon Hd 7970 Hashrate -Review|Profitability | Specs | Pros and Cons
