EVGA once again ditches fancy RGB lighting and trim to deliver gamers the best performance and temperatures possible. In this review we will be taking a look at the EVGA RTX 2070 XC Ultra and EVGA RTX 2070 XC Gaming to see which of these premium binned GPUs is right for you.
- MSRP: $599.99 (XC Ultra), $579.99 (XC Gaming)
- CUDA Cores: 2304
- Boost Clock: 1725 MHz (XC Ultra), 1710 MHz (XC Gaming)
- Texture Fill Rate: 248.4GT (XC Ultra), 246.2GT/s (XC Gaming)
- Memory: 8192 MB, 256 bit GDDR6, 7000 Mhz (14,000 effective)
- Cooling: iCX2 Cooling, Dual Asynchronous Fans
- RGB: Logo
- Interface: PCIe 3.0
- Ports: 3x DisplayPort, 1x HDMI, 1x USB-C (DisplayPort 1.4, HDMI 2.0b)
- Max Monitors: 4
- Max Refresh Rate: 240Hz
- Max Resolution: 7680x4320
- Dimensions (XC Gaming): H4.5” (114.3mm) L10.6” (269.2mm) - Dual Slot
- Dimensions (XC Ultra): H4.5” (114.3mm) L10.6” (269.2mm) - 2.75 Slot
- Power: 1x 8-pin, 1x 6-pin
Test System: CPU: Ryzen 2600x @ 4.5 Ghz, Cooler: CoolerMaster ML360R, Motherboard: MSI x470 Gaming M7 AC (Bios Version: 7B77v13), RAM: 32GB HyperX Predator RGB 3200 (15, 16, 17, 19), OS Drive: Kingston A1000, Game Drive: HyperX Fury RGB SSD, GPU: Testing Variable, PSU: 800W ThermalTake Riing PSU
EVGA has long been known for putting functionality over aesthetics, a trend that remains consistent with the release of the RTX 2070 XC Ultra and XC Gaming GPUs that lack flashy RGB lighting and stylized parts. The premium feel of the XC line starts right with the box - a thick hard cardboard the slides out into two separate pieces, revealing a GPU crammed into some nice foam.
Once out of the box it quickly becomes apparent how massive the XC Ultra is compared to the XC Gaming - or any other graphics card for that matter. Sporting almost 3 entire expansion slots of space (a hair over 2” total) the XC Ultra, and the XC Gaming, will have three display ports, one HDMI and USB-C for VirtualLink standard with VR Headsets. Both XC cards have a nice, premium feeling black metallic backplate that’s slightly rough to the touch and a distinct lack of RGB lighting - only the logo houses the LEDs which can be changed with the new Precision X1 software. Lastly both cards require an 8-pin and 6-pin power connection, but the 6-pin is the same size as the 8-pin with two of the pins blocked off.
The main difference between the two cards is the radiator size and a slightly lower boost clock on the XC Gaming. Whether the larger radiator contributes to a higher overclock remains to be seen, though in theory more temperature overhead should mean better clocks, as some unwanted software issues prevented me from doing full overclock testing in time for the article. The larger radiator did result in overall lower temperatures, however, and may all be better suited for those that live in warmer climates compared to the slimmer XC Gaming.
Before we get into the results of the testing I want to clear the air about Battlefield V. In a VERY recent patch we witnessed the first real time ray-tracing implementation in a released game. All testing was done for the EVGA cards by the time this patch came out so it will not be included in this review. That being said I do plan on doing testing across the RTX cards for Battlefield V ray-tracing performance, and hopefully other games if it gets implemented in time. In addition to the lack of Battlefield V benchmarks I chose NOT to do any 4k resolution testing. While some people may use an RTX 2070 for 4k gaming, the demographic is aimed squarely at 1440p and ultra-high refresh rate 1080p gamers. 4k gaming on an RTX 2070 requires a lot of graphical sacrifices to maintain an acceptable 60 frame per second in AAA games and doesn’t quite make it to the 1080ti performance which is widely accepted as the standard minimum power level for 4k gaming.
We will be pitting the 2070s against each other in this review and while you may not see much difference in frame rate between the two the key data point for comparison is temperatures, given the radiator discrepancies between the XC Ultra and XC Gaming. The 1080ti will make an appearance for comparisons at 1440p as the majority of games will bottleneck on the CPU at 1080p. Nvidia’s 1070Ti Founder’s Edition, a card previously targeted for 1440p users, will be used as a point of comparison as well.
The last thing I want to talk about is binning. Many people have suspected in the past that NVIDIA saved the best GPUs for the FE cards, a practice called binning, which has now be confirmed though it’s not just for FE cards. This means that NVIDIA is selling higher-quality 2070 chips to partners for use in their more expensive RTX 2070 models. So while the base $499 model of a card is a 2070 just like the $599 version, the $599 version will likely have the higher quality GPU which leads to better overall boost performance and overclocking headroom. While we don’t typically dismantle GPUs at MMORPG.com, Gamer’s Nexus did confirm that he XC Ultra uses these binned GPUs.
When directly comparing the XC Ultra and the XC Gaming, it’s clear that both provide roughly the same level of performance - easily within the margin of error. In our testing the XC Ultra did consistently outperform the XC Gaming by a marginal amount which I tend to contribute to the larger radiator. The additional temperature head-room theoretically equates to less thermal throttling on the clock, leading to a slightly higher average frame rate, and combined with the extra 15Mhz of boost potential I believe that’s what we’re seeing occur.
The EVGA 1080ti FTW3 still dominates the RTX 2070’s and 1070Ti, which is to be expected, with the exception of FFXIV which seems to be CPU throttling even at 1440p. Though this isn’t unheard of for MMOs it was still interesting to see it unfold during testing.
In the realm of 1080p gaming the XC Ultra dn XC Gaming will both sport massive frame rates for the common AAA titles. As long as the game is graphically intense enough and properly utilized the graphics card the RTX 2070s make a great choice for high refresh rate gaming this resolution. As I stated before it’s important to note that performance between the XC Ultra and XC Gaming is nearly identical, though that difference widens when we look at the temperature data points.
We included the temperature findings for the 1080Ti FTW3 and the 1070Ti FE for consistency, but most of the attention should be drawn to the orange (XC Ultra) and blue (XC Gaming). The XC Ultra’s additional radiator gives an edge up on the XC Gaming, providing lower all around temperatures to the gamer. The additional temperature headroom can theoretically mean better average frame rates by staving off thermal throttling just a little longer in addition to more thermal headroom when overclocking and pushing more voltage (and heat) into the GPU. All this isn’t to say the XC Gaming has bad temperature performance. Both cards maintain temps very well but we would likely see larger differences in warmer climates or cases with poor airflow.
The EVGA XC Ultra and XC Gaming are both solid cards for their price bracket. The $20 additional price of the XC Ultra buys you a marginally higher boost clock and markedly bigger radiator versus the XC Gaming. If you have an all around warmer case with less air flow, the XC Ultra will probably server you better than the XC Gaming. However in a case with good airflow across the GPU, the XC Gaming will deliver virtually the same performance as it’s bigger brothers with the added benefit of not taking up 3 expansion slots in your case. One thing that has become clear is that the RTX 2070 delivers the best performance in it’s price bracket, firmly planting itself between the GTX 1080 and GTX 1080ti but if that price is worth it is only a question you can answer for yourself.
- Both cards provide great performance
- Asynchronous fans are great for cooling what needs to be cooled
- XC Ultra is gigantic and takes up a lot of room
- Both cards price starts to approach 1080ti’s, which provide better performance
The product discussed in this article was provided by the manufacturer for the purposes of review.