Review, testing and overclocking of the Palit GeForce GTX 1660 Super StormX video card

Profitability of NVIDIA GTX 1660 SUPER

The current market price and mining profitability of the NVIDIA GTX 1660 SUPER graphics card fluctuates depending on the value of the cryptocurrency. The information in the plate is updated once a day.

View the full graphics card profitability table with the ability to sort by price, revenue, and ROI.

Packaging and equipment

In the general trends of most manufacturers, the box is noticeably larger than the size of the video card itself, at least vertically. Dimensions - 28.5 x 18.2 x 7.3 cm.

The design is bright, in the usual style of the MSI Gaming series. The main emphasis is placed on the lighting system and the efficient cooling system of Twin Frozr 7.

Beneath the thin cardboard outer box is a pull-out cardboard tray with a foam bed that protects the graphics card during transit.

There is no complete set as such. There is nothing except the usual MSI envelope with a minimal set of documentation.

Hashrate NVIDIA GTX 1660 SUPER

We have compiled a table of NVIDIA GTX 1660 SUPER hashrates for popular algorithms. If you are just choosing a video card for mining, and already know what coin you will mine, then you can estimate the approximate performance from the table. We also have an article about what hashrate is.

Appearance and Design

The video card looks nice. The stylish dark gray casing with ARGB lighting and fans decorated with MSI dragons, as well as a decorative plate on the back, will fit perfectly into modern gaming builds with an open view of the components.

The video card uses a dual-fan cooling system with a radiator elongated relative to the printed circuit board, so the total measured length is approximately 247 mm.

The backplate plays a decorative role and does not participate in heat dissipation. The surface is stylized to resemble the texture of a processed aluminum plate, which helps reduce the visibility of dust on it. A sticker with the model and serial number of the device is carefully glued in the corner next to the cutout that reveals the PCB model - 1V375-2.2.

The height of the cooling system is relatively small - two standard expansion slots.

The side of the plastic casing is decorated with a controllable backlight module and a dark gray GEFORCE GTX inscription.

The interface bar has four connectors for connecting image output devices: three DisplayPort 1.4 and one HDMI 2.0b. However, according to the specification, only three monitors can be connected simultaneously.

According to information from the manufacturer, the consumption of the video card is 125 W, so one additional eight-pin PCIe power connector is made. This solution is redundant; a six-pin connector would be sufficient.

Overclocking NVIDIA GTX 1660 SUPER for mining

We will overclock the NVIDIA GTX 1660 SUPER for mining using the MSI Afterburner program. You can find it in Yandex. The program interface is very simple, although a little colorful) To overclock the video card, we need two sliders:

• Core Clock – core overclocking, increases the core frequency;
• Memory Clock – memory overclocking, increases the memory frequency.

MSI Afterburner - overclocking video cards
You need to change the frequency values ​​a little at a time. Then test stability. If everything is good, then we increase the frequency further. If the video card starts to fail or the drivers crash, then we simply reduce the frequency and that’s it. There will be no harm. Next, we will look at overclocking options for popular alogorites.

Overclocking NVIDIA GTX 1660 SUPER for Ether

To overclock the NVIDIA GTX 1660 SUPER for mining ether or other coins using the DaggerHashimoto algorithm, you need to increase the memory frequency, and, on the contrary, lower the core to save energy because it does not affect production speed. Optimal overclocking parameters:

In progress

In rest mode, the video card does not stop the cooling fans. By default they rotate at a speed of about 1000 rpm.

The level of noise emitted is minimal; they are unlikely to be heard in a closed case. Taking into account active cooling, the GPU temperature at idle barely exceeds 30C.

Under gaming load on an open bench, the GPU warmed up to only 65C. At the same time, the fan rotation speed at peak increased up to 2000 rpm.

According to the given algorithm, the fan rotation speed increases to 76% of the maximum when the chip temperature reaches 63C. In our opinion, the efficiency of the cooler used is enough to reduce the speed without consequences, without fear of overheating the GPU.

To do this, you can use the same Thunder Master utility, where you can manually set a graph of the dependence of fan speed on processor temperature. If you wish, you can experiment here. The final numbers will depend on the conditions of use of the video card. It would be logical to choose an algorithm in which the GPU will not heat up above 70–75C. Here the results will depend on the general cooling conditions of the components inside the case. Again, on an open bench at a maximum of ~1600 rpm, the GPU temperature rose to 70–72C.

Thanks to the work of dynamic auto-overclocking GPU Boost, the graphics processor was accelerated to 1965 MHz under load. On average, operating frequencies were still lower - 1880–1920 MHz.

Backlight

Several zones are highlighted on the fan casing: four thin stripes along the edges of the fans, the bottom two of which will be clearly visible only when the video card is installed vertically, and a block with a logo on the side, clearly visible in a standard horizontal position.

Control and synchronization with other components is carried out through proprietary MSI Dragon Center software.

⇡#Test stand, testing methodology

In most test games, the average and minimum frame rates are derived from an array of individual frame rendering times, which is recorded by the built-in benchmark (or the OCAT utility, if it does not exist).

The average frame rate in the charts is the inverse of the average frame time. To estimate the minimum frame rate, the number of frames formed in each second of the test is calculated. From this array of numbers, the value corresponding to the 1st percentile of the distribution is selected.

Exceptions to this methodology are three games with a built-in benchmark - DiRT Rally 2.0, Far Cry 5 and Forza Horizon 4. The Far Cry 5 benchmark itself records the number of frames per second of the test in a file - the average FPS value and the 1st percentile are calculated based on from these numbers, and not by the rendering time of individual frames. The DiRT Rally 2.0 benchmark only records the average frame rate and the minimum frame rate calculated from the maximum frame time. Forza Horizon 4 does not output any performance parameters to a file and also reports average and minimum frame rate totals.

The power of video cards is recorded separately from the CPU and other PC components using a JUNTEK VAT-1050 ammeter. To simultaneously measure the current flowing through the additional power connectors and the motherboard slot, the video card is connected through a PCI Express x16 rigid riser, in which the power lines are broken and routed to a separate cable.

The test load for power and noise level tests is Crysis 3 at a resolution of 3840 × 2160 without full-screen anti-aliasing and maximum graphics quality settings, as well as the FurMark stress test with the most aggressive settings (resolution 3840 × 2160, MSAA 8x). All parameters are measured after the video card has warmed up, when the GPU temperature and clock frequencies stabilize.