Today we have our day-one GeForce RTX 2070 review for you and I suspect like most other publications, nosotros were seriously under the pump for this one. For any reason Nvidia thought it all-time not to send out samples until a few days before the release, thus we're not expecting our RTX 2070 Founders Edition sample until later this calendar week. Thankfully MSI came through for united states and saved the twenty-four hours at the terminal minute. In total we only had effectually thirty hours for testing and we've used every last minute to create this content.

MSI also sent along an Armor model which is expected to come in at a lower price, somewhere around the MSRP, but we'll believe that when nosotros see it. That'south not a dig at MSI only rather Nvidia's mythical suggested retail prices.

On paper, the GeForce RTX 2070 claims a $500 MSRP for board partner cards such as the MSI Armor and a less attractive $600 for the Nvidia Fe model. So for the AIB models we're expecting around a thirty% decrease in cost when compared to RTX 2080 models for a 33% decrease in CUDA core count.

Nvidia's ain review guide shows the RTX 2070 easily chirapsia the GTX 1070 and and then the 970 before it, though that's a little misleading as it'southward priced to compete with the GTX 1080 and Vega 64, so those are the models we'll focus our attention on when comparing performance.

For all the testing we're using the MSI RTX 2070 Armor and it will be compared to the FE versions of the 2080 Ti and 2080 along with MSI Gaming X versions of the 1080, 1070 Ti and 1070 besides as the Gaming 10 Trio 1080 Ti. So nosotros have the Asus Strix Vega 56 and 64 models along with AMD's Vega 64 Liquid.

Our test organization consists of a Core i7-8700K clocked at 5GHz and 32GB of Vengeance DDR4-3400 memory. For the AMD GPUs we've used Radeon Adrenalin Edition 18.10.i drivers and for Nvidia the GeForce 411.63 drivers. The new RTX 2070 GPU is using driver version 416.33.

We take results for 20 games just due to the express fourth dimension we had nosotros're going to discuss the results from just vi of them so intermission down the data for all the games at the end of the article.

Benchmarks

Kickoff upwardly nosotros have Far Cry 5 and allow's start with the 1080p testing, here the RTX 2070 was a wisker slower than the GTX 1080, but simply 9% slower than the RTX 2080 when comparing the average frame rate. That said nosotros are mostly CPU leap at this resolution.

At 1440p the GTX 1080 and RTX 2070 can be seen delivering basically the aforementioned functioning. This ways the 2070 is at present 21% slower than the RTX 2080. Then at 4K nosotros run into the 2070 and 1080 delivering identical frame rates making them 23% slower than the 2080.

When testing with Assassin's Creed: Odyssey we do see the RTX 2070 providing a reasonable functioning advantage over the 1080 at 1080p, it was 8% faster on boilerplate. Not exactly mind blowing stuff given the cost just nosotros'll discuss that more towards the stop.

The margin is extended out to 13% at 1440p and now the 2070 is establish in the middle of the 1080 and 1080 Ti, making it just xiv% slower than the RTX 2080.

Again at 4K the 2070 finds itself between the 1080 and 1080 Ti which means information technology was simply 13% slower than the 2080, so an impressive effect.

Moving on to Forza Horizon iv and hither the 2070 is simply simply able to go along footstep with the 1080 and overall performance seems quite lackluster despite pushing rather high frame rates at 1080p. Much the aforementioned is seen at 1440p, the 2070 comes in a few frames behind the GTX 1080 for a pretty disappointing upshot.

The RTX 2070 isn't able to do any better at 4K either as it'due south just able to lucifer the 1080 here and that meant information technology was 25% slower than the GTX 1080 Ti and RTX 2080.