GTX 1050 vs. RX 460 – Which is the best ~$100 entry-level GPU?
Hey guys welcome back. How’s it going? I hope you are all doing well and you are ready for some serious benchmarks, because that’s what we have for you today.
As most of you are probably aware Nvidia just recently released their new GTX 1050 series, if you didn’t check out our reviews on the MSI cards you can find them here and here. In short MSI’s base model versions of the GTX 1050 and 1050 Ti gave us a good look at these new GPUs.
Matt and I were quite impressed with both GPUs, particularly the $110 US base model 1050. Some of you in the comments weren’t that impressed with what the 1050 in relation to last generations Maxwell based GTX 950. Granted it isn’t worlds faster, but you have to remember the GTX 950 came to us originally at $160 and has only been discounted recently to combat AMD’s RX 460.
For less money the RX 460 roughly matched the now discounted GTX 950 and given the choice of either I would go with AMD’s option here every time. In our previous 8 game test the 1050 might have only been 7% faster than the 950 on average, but it also comes in at the same launch price as the RX 460.
The 1050 is also responsible for AMD’s swift price reduction of the RX 460, though if initial impressions are anything to go by further price cuts might be in order. That’s why we are here today, to work out exactly how the GTX 1050 and RX 460 2GB graphics cards compare in a head to head battle.
On the menu for you guys we have 23 games in total, all of which have been tested at 1080p using suitable image quality settings. Once we have investigated each title individually, we will then take an overall took at the performance to help break things down. Overclocking won’t be a focus here as neither model we have on hand are particularly great when it comes to squeezing out extra performance, so there isn’t a clear advantage here. That said performance per watt will be investigated and naturally we will do the same for the cost per frame.
As was the case with the original reviews we will be testing with our Core i3 test rig. Despite featuring a Z170 motherboard, an all-in-one liquid cooler and excessive power supply, this system does better represent a typical mid-range gaming rig thanks to the use of a Core i3-6100T processor. The low-voltage CPU has been included on purpose because at 3.2GHz it better represents how Core i3 Ivy Bridge and older processors perform.
As for the graphics cards we are using the MSI GTX 1050 2G OC and XFX RX 460 2GB 5D. The MSI card features a 4.3% factory overclock while the XFX card has been overclocked by just 1.6%.
Now you might be thinking that’s a bit unfair, the Nvidia card is factory overclocked higher giving it an advantage, albeit it tiny advantage. In a way yes you would be right, however please let me explain why we did things this way.
There isn’t a reference version of either the RX 460 or GTX 1050, so we have to use partner cards for this comparison and it makes more sense to use products you can actually buy anyway. The highest clocked RX 460 2GB card we could find was the Asus dual-fan model, though it’s clocked just 4 MHz higher than the XFX version we have on hand and costs $20 more. So almost all factory overclocked RX 460 cards operated at between 1210 and 1220 MHz.
Meanwhile the GTX 1050 factory overclocked cards operate at a boost clock of between 1493 and 1556 MHz. The MSI model runs at conservative 1518 MHz making it one of the lower clocked partner cards, crucially it is also available at the $110 MSRP.
So I hope that clears things up for those of you who were concerned about the cards used and their operating clock speeds. With that I think we have spent enough time explaining and justifying how we are testing, let’s just get on with it…
First up we have Assassin’s Creed Syndicate and for playable performance we were forced down to the still very respectable medium quality preset. The 460 was good for 43 fps on average while the 1050 delivered a more desirable 56 fps, this gave the GeForce card a 23% performance advantage here.
Batman: Arkham Knight provided slightly more competitive results, though the 1050 was still 18% faster, rendering 53 fps on average to the 460’s 45 fps.
In our opinion Battlefield 1 plays best using the DirectX 11 API for both AMD and Nvidia owners. Still for this comparison we will be testing the game using both APIs in an effort to keep things fair. Using the ultra quality settings the 1050 is able to deliver a playable 49 fps, whereas the 460 struggles with 35 fps giving the GeForce card a 40% performance advantage here.
Shifting gears to DX12 we see a massive 22% reduction in performance for the 1050. Meanwhile the RX 460 does gain 9% more performance thanks to reduced overhead. That said the big issue here is the fact that the minimum frame rate doesn’t improve, in fact we drop a single frame here though that is within the margin or error. Overall when using DX12 the 460 has an 8% performance advantage over the 1050.
We were probably a little too aggressive with the quality settings in Crysis 3, that said dropping down to high or even medium shouldn’t change the performance margins between these two GPUs. Using the very high settings the 1050 averaged 35 fps making it 21% faster than the 460.
Mankind Divided is an AMD special and here the 460 enjoyed a 9% performance advantage, rendering 36 fps to the 1050’s 33 fps. These are quite low frame rates and it is worth pointing out that we were using the medium quality preset here.
The 1050 blew through the DiRT Rally benchmark rendering an average of 76 fps and never dipping below 62 fps. In fact the 1050’s minimum frame rate was quite a big higher than the 460’s average in this game, making the 1050 36% faster.
The Doom Vulkan results are very interesting. We found previously that graphics cards limited to a 2GB frame buffer really struggle in this title when using Vulkan. The RX 460 2GB for example is much slower than the 4GB version despite delivering a healthy 51 fps. Where as AMD generally has a nice performance advantage when using Doom’s Vulkan implementation, this isn’t the case here. The 1050’s superior memory compression gives it the edge in this memory hungry game, allowing it to deliver almost 40% more performance.
F1 2016 is another racing simulator that favors the 1050, though this time the GeForce graphics card was just 14% faster. The 460 did seem to take a slightly larger hit when looking at the minimum frame rate however.
Fallout 4 resulted in a dead heat with both GPUs rendering a very playable 51 fps on average using the high quality preset. The 1050 did dip slightly lower at 42 fps, though this won’t be noticeable and is technically within the 1 to 2 fps margin or error, even after three tuns.
Far Cry Primal [HD Texture Pack]
The 1050 streaks away in Far Cry Primal rendering 51 fps on average to the 460’s 41 fps. This gave Nvidia a 24% performance advantage here when comparing the average frame rate.
Right so if you have made it this far you have earnt yourself a drink. I’m sticking with water, i’ll need it to get through the rest of these results but feel free to enjoy a more exciting beverage as we continue.
Back at it we have Gears of War 4 running the high quality preset. The 1050 averaged 54 fps and this was enough to give it a 15% performance advantage over the 460. The 460 still played quite well, never dropping below 37 fps.
Stealing cars is a more enjoyable experience at 60 fps, and the 1050 delivers that for the most part, averaging 61 fps at 1080p using the high quality settings. Using the same quality settings the 460 was 20% slower averaging 49 fps.
When Mafia III was first released the developer hard capped the game at 30 fps and we were all like what! In protest PC gamers band together, forcing the cap to be removed. Sadly it is now evident why the cap was put in place, optimization sucks and I guess they though hiding the fact with a 30 fps cap rather than fixing it was the way to go. The end result for budget gamers is a low quality experience at low frame rates. The 1050 spat out just 41 fps though this did make it 24% faster than the 460.
Cutting up orcs requires a steady frame rate, and for this a weapon exceeding 2GB of VRAM is likely required. The 1050 averaged a relatively high 51 fps but at times dipped down to 30 fps, meanwhile the 460 suffered similar fair fluctuating from an average of 42 fps all the way down to 26 fps. Of course there is plenty of room to move on the quality settings here as we did load up the ultra preset. In the end the 1050 was on average 21% faster.
Using the medium quality preset both GPUs glided through Mirror’s Edge Catalyst, rendering at least 50 fps on average. The 1050 pushed a little higher averaging 59 fps making it 18% faster than AMD’s RX 460.
For fans of Overwatch we have good news, both GPUs are able to maintain over 60 fps at all times using the ultra quality settings, on a Core i3 system on less. For optimal performance the 1050 is the best choice here as it was 34% faster on average.
Steam’s version of Quantum Broken doesn’t play particularly well with the 460. Despite delivering playable performance the 460 was 31% slower than the 1050 which managed a buttery smooth 68 fps.
For raiding Tombs as a fiery young lass either GPU will suffice. Both delivered much the same performance and both suffered from rather large frame dips using the very high quality settings so some tweaking will be required here.
These are the results you are looking for. Both GPUs rendered over 40 fps at all times using the high quality settings with the average exceeding 50fps. The 1050 might have been 18% faster here but the 460 still provided ample performance.
The Witcher 3 was tested using the high quality settings and here the 460 averaged 40 fps to the 1050’s 46 fps. This gave the GeForce GPU a 15% performance advantage and meant performance never dipped below 41 fps.
Rainbow Six Siege was tested using the high quality preset and even here it is still a bit of a memory pig. The 460 for example averaged 54 fps but at times dropped as low as 29 fps. Meanwhile the 1050 averaged 63 fps but would drop as low as 38 fps. In the end the 1050 was 17% faster on average.
The Division is a very demanding game, even with the high quality preset. As a result the 1050 averaged 41 fps, making it just 5% faster than the 460.
Finally we have Total War: Warhammer which was tested using the DirectX 12 API which is still marked as beta for this game. The 1050 was just a single frame faster than the 460, so a tie here then.
Well there you have, an indepth look at how the GTX 1050 and RX 460 2GB compare in a good number of titles from 2015 and 2016. Now to wrap things up let’s try and quickly break the data down into something that makes sense.
As you can see the 1050 is by far and away the faster GPU, winning in 20 of the 23 games tested. 18 of which is won by a 10% margin or greater. The 1050 was slower in just two games and the results from one of those titles is questionable for now as we aren’t happy with Battlefields DX12 performance from either camp.
Probably the most surprising result was seen when testing Doom using Vulkan as this handed the 1050 one of it’s strongest wins. Overall the 1050 was on average 18% faster when tallying up all 23 games.
Before we jump to the cost per frame figures let’s just take a quick look at efficiency by comparing power consumption against the average frame rates.
Again we know that the 1050 was 18% faster, rendering 53 fps on average to the 460’s 45 fps. Yet despite being almost 20% faster it allowed our Core i3 test system to consume 20% less power when gaming. This is a huge advantage in Nvidia’s favor when it comes to efficiency. This is certainly something to keep in mind when making your buying decision.
Okay so now it is time to look at the all important pricing metrics. We know that the 1050 costs just $10 more than the 1050 which is obviously a negligible difference despite being a 10% price hike. Although you will pay slightly less for the 460 that doesn’t mean it actually works out being cheaper. Taking the resulting performance into account we know the 1050 was 18% faster but it actually cost 7% less per frame.
In the end it’s pretty simple, you’re paying 10% more for 18% more performance. For the RX 460 2GB to be truly competitive AMD really needs to knock another $10 off the asking price.
Of course as many of you often point out pricing does vary from region to region. In the US it is possible to buy the RX 460 2GB for as little as $100 while there are some 1050’s selling for the $110 MSRP. Down under the 1050 costs $210 Aussie while the RX 460 starts at $165 Aussie.
For now that actually makes the RX 460 a better buy for my fellow Aussies, so again be sure to check prices in your region.
It is also important to check how these GPUs comapre in games you actually plan on playing.
I really like how efficient the GTX 1050 is, the fact that the card doesn’t require a PCIe power connector is awesome and should be of great benefit to those making do with older computers. On the other hand the RX 460 is slightly cheaper and can take advantage of the ever growing army of affordable FreeSync monitors.
In the end what’s clear is budget gamers now have two great options capable of delivering an impressive 1080p gaming experience.
What do you guys think? How do prices look in your region and are you considering buying a new graphics card using either of these GPUs? Finally if you liked this video please take a moment to give us the thumbs up, we really appreciate that and if you found our testing useful consider sharing it. As always we greatly appreciate the support and look forward to seeing you again on the next one.