The NVIDIA GeForce GTX 1070 Ti Founders Edition Review: GP104 Comes in Threes
by Nate Oh on November 2, 2017 9:00 AM EST- Posted in
- GPUs
- GeForce
- NVIDIA
- Pascal
- GTX 1070 Ti
Synthetics
Continuing onto our synthetic benchmarks, we should be able to see how the GTX 1070 Ti sits in between the GTX 1080 and 1070. These tests are more graphics focused than compute focused – especially the low-level tests – so it will be interesting to see how the GTX 1070 Ti compares to its fully-featured GP104 sibling, the GTX 1080, given how close they are so often in compute tests.
Starting off with tessellation performance, as NVIDIA's geometry performance is tied to the SMs, the GTX 1070 Ti benefits from having 3 more SMs enabled than the vanilla GTX 1070. This allows the new card to leapfrog both AMD Vega cards, coming in right behind the GTX 1080.
Switching gears, we have Futuremark's VRMark Orange benchmark. Essentially a synthetic gaming-type rendering workload with a focus on VR, this test pushes every part of a video card. Consequently, it also hits the weaker parts of the GTX 1070 Ti's configuration, particularly the lack of increased ROP throughput and memory bandwidth. As a result the GTX 1070 Ti still gains quite a bit over the GTX 1070, but it also trails the GTX 1080 by about 5%. Which, as it turns out, isn't too far removed from our general gaming benchmarks.
Finally, for looking at texel and pixel fillrate, we have the Beyond3D Test Suite. This test offers a slew of additional tests – many of which use behind the scenes or in our earlier architectural analysis – but for now we’ll stick to simple pixel and texel fillrates.
The pixel fillrate test is actually one of the more interesting results we have in today's review. While the GTX 1070 Ti doesn't get any more ROP resources on paper, pixel fillrates are a combination of both ROP (backend) and rasterizer (frontend) output. And in the case of NVIDIA's architecture, rasterization is not fully decoupled from the SMs; which is to say there's some scaling going on. And actually, this was more than I was expecting, with the GTX 1070 Ti coming very close to the GTX 1080 despite the memory bandwidth disadvantage. Not unlike tessellation performance, we see the new card leapfrog both AMD Vega cards here, pushing over 100 Gigapixels per second. In general pixel throughput is still going to be what best differentiates the GTX 1080 from the GTX 1070 Ti, but in this low-level test the two are closer than I would have expected.
Finally, for texture throughput, we're once again looking at a scenario tightly-coupled with the SM count. As a result the GTX 1070 Ti can deliver texture throughput rates very close to the GTX 1080.
Facebook
Twitter
RSS
78 Comments
View All Comments
Nfarce - Friday, November 3, 2017 - link
"Vega 56 beats 1070 in almost all games."It beats the reference GTX 1070, but a factory overclocked 1070 pulls ahead again, especially overclocked on top of the factory overclock. Vega 56 (or 64 for that matter) does not have that type of overclocking headroom. This has long been an advantage for Nvidia. AMD GPUs have a history of being terrible overclockers. My old EVGA GTX 970 SSC ACX 2.0+ could be overclocked to 980 performance without even touching the voltage for example.
LastQuark - Saturday, November 4, 2017 - link
Vega 56 BIOS can be upgraded to the Vega 64 BIOS for another 20% speed boost. Vega can only do 5% max. With Vega 56 lower price point by over $50, cheaper Freesync monitor options, and 2x the availability of GSync, Vega 56 is still making a lot of sense for new buyers.B-Real - Thursday, November 2, 2017 - link
According to this theory, you can't compare the 1070 to the Vega56, as the 1070 is 5 degrees Celsius hotter...damonlynch - Thursday, November 2, 2017 - link
It should be "nonetheless", not "none the less", in the introduction to the Final Words ;-)jardows2 - Thursday, November 2, 2017 - link
Good showing. At MSRP, a good argument (from these tests at least) can still be made for Vega 56. Not sure if 1070ti is worth $50.00 more, but you do get a little bit better performance, and most important to me, a lower noise profile at load. Keeping it interesting for sure!CaedenV - Thursday, November 2, 2017 - link
So.... it is essentially a direct Vega 56 competitor except that it will be available on store shelves for purchase?Really hoping that this will cause the normal 1070 prices to drop a bit *fingers crossed*. I picked up a 4k monitor last year and my gaming has been quite limited on it with my GTX 960. A 1070 will fill in quite nicely for now, and next year when the new cards come out I'll pick up a 2nd 1070 for SLi to really make 4K gaming smooth.
BrokenCrayons - Thursday, November 2, 2017 - link
SLI doesn't seem like a good solution these days given tepid support from the GPU manufacturers and very few modern titles that are optimized to take advantage of a 2nd graphics card. You might have a better experience if you set aside the first 1070's cost until next year and then use the funding from both to purchase a 1080 or just hang on to see what happens with Volta since there'll likely be consumer GPUs available sometime in 2018.vladx - Thursday, November 2, 2017 - link
AMD has no chance at all, a RX Vega 56 in my country is $150 more expensive than the newly released GTX 1070 Ti.Sorjal - Thursday, November 2, 2017 - link
Easiest check to see potential mining impact is toss a ti and non in a pc and test with some miner program. Nicehash legacy is probably the best for comparative stats. Run their benchmark on both cards on precise and compare the results. There will be some variance, but it should provide a decent reference. Nvidias are typically used on ones that are more gpu intensive, with Zcash probably the largest. AMD works off their bandwidth and are favored for corresponding currencies like Etherium and Monero. The nice hash legacy will test against most of the major ones including Ethereum, Monero, and ZCash.Given the increased CPU performance, it is probably being tested mining wise as we speak
Sorjal - Thursday, November 2, 2017 - link
Remember to overclock and undervolt. Energy effeciency is a large factor. 1070's seem to be run between 65-70% power limit. A voltage meter on the outlet may be useful in this case as you can look at the voltage change for each test for each card