Nvidia GTX 1650 vs 1050 Ti – 18 Games Compared


Nvidia just launched their new GTX 1650 graphics
card, but how does it compare with the older GTX 1050 Ti? In this video we’ll test out
and compare 18 different games at 1080p and 1440p resolutions to help you see the performance
difference between them, followed by a discussion on value and if it’s worth it. Let’s quickly take a look at how the newer
GTX 1650 and older GTX 1050 Ti actually differ in terms of specs, note that these are base
model specs, different cards will vary slightly. The 1650 has 16% more CUDA cores and is based
on Nvidia’s newer Turing architecture. The 1650 also has higher base and boost clock
speeds, and while both have 4GB of GDDR5 memory, the 1650’s appears to be a little faster. For the testing I’m using the Gigabyte GTX
1650 OC and the MSI GTX 1050 Ti Gaming X, so both have a little factory overclock, expect
slightly different results with different models. The system that I’m testing with has an
Intel i7-8700K CPU overclocked to 5.0GHz in an MSI Z390 ACE motherboard, along with 16GB
of T-Force Night Hawk CL16 memory from Team Group running at DDR4-3200 in dual channel.
Check the links in the description for details on all of the components as well as for up
to date pricing. The same Windows updates, game updates and Nvidia driver 430.39 were
used for the testing, so let’s get into the results. Let’s start out Apex Legends, which was
tested with all settings maxed out. I’ve tested with higher settings in the upcoming
games than I’d actually play with in most cases, as that’s better for GPU comparison
purposes. In all upcoming graphs I’ve got the GTX 1650 shown by the top bar in red,
while the older GTX 1050 Ti is shown below in purple. In terms of average FPS the 1650
was 32% ahead of the 1050 Ti at 1080p, rising up to a 37% improvement at 1440p. Assassin’s Creed Odyssey was tested with
the built in benchmark using high settings, and as expected the 1650 was ahead here, performing
30% better at 1080p over the 1050 Ti then slightly higher at 32% at 1440p, while even
the 1% lows from the 1650 were above the averages from the 1050 Ti. Battlefield 5 was tested in campaign mode
rather than multiplayer, as it’s easier to consistently reproduce the test run. At
1080p the 1650 was 27% ahead of the 1050 Ti here, rising to 31% at 1440p. Due to the high
1% low I think you could actually get away with the 1650 at 1440p, though maybe turn
down the settings to medium. Far Cry New Dawn was tested using the built
in benchmark with high settings, and was another game where even the 1% lows from the 1650
were ahead of the averages from the 1050 Ti. At both 1080p and 1440p the 1650 was 25% ahead
of the 1050 Ti in this test. Far Cry 5 was also tested using the built
in benchmark with high settings, and this game was seeing higher FPS over the newer
Far Cry New Dawn just shown. For average frame rate there was a 24% boost at both 1080p and
1440p with the 1650. Fortnite was tested using the replay feature
with the exact same replay on both graphics cards. The 1050 Ti was still able to reach
60 FPS averages at 1080p even with epic settings in use, however the 1650 was 20.5% ahead here,
lowing to a 13.7% lead at 1440p. Metro Exodus was tested with the built in
benchmark at high settings. The actual game performs a fair bit ahead of this, so this
is more of a synthetic test rather than representative of actual game play, but it does allow me
to accurately compare between the two graphics cards. With that in mind, the 1650 was 30%
ahead of the 1050 Ti at 1080p, and 28% ahead at 1440p. Shadow of the Tomb Raider was also tested
using the built in benchmark with high settings. This seems to be more of a graphically intensive
test, as out of all games tested it saw the largest improvement at 1440p of 41.6% and
3rd largest improvement at 1080p out of all 18 games tested with 34% higher average FPS
over the 1050 Ti. Rainbow Six Siege was tested using the built
in benchmark at high settings. This is a game that I’ve found to perform better with Nvidia’s
newer turing architecture, and it was seeing quite a nice improvement with the 1650, with
33.5% higher average FPS over the 1050 Ti at 1080p, and a 35% boost at 1440p. CS:GO was tested with the Ulletical FPS benchmark,
and both cards were capable of providing very high frame rates in this test, granted the
8700K is of course contributing a lot here too. This game saw one of the smallest differences
between the two graphics cards, with just a 11.5% improvement at 1080p with the 1650,
and a 10.6% improvement at 1440p, Overwatch was tested in the practice range
as I can easily perform the same test run compared to playing with bots or other players
which will differ every time. While both were giving fairly good frame rates considering
the highest epic setting preset was in use even at 1440p, like CS:GO there was less of
a difference between the two. At 1080p we’re looking at a 13% improvement with the 1650,
lowering to 10% faster at 1440p. PUBG was tested using the replay feature with
the exact same replay at high settings, and the 1650 was needed to hit the 60 FPS sweet
spot at 1080p here. In terms of differences the 1650 was reaching 30.7% higher average
FPS than the 1050 Ti at 1080p, and 34.7% higher at 1440p. The 1650 was also giving massive
improvements to 1% low too, with a 44% increase at 1080p and 37% at 1440p. Watch Dogs 2 is a resource intensive game
and was tested with very high settings. In my opinion this one doesn’t need a high
frame rate to play, I can get by with a solid 30 FPS, so I wouldn’t want to use 1440p
here though it would probably be fine with high or lower settings. Anyway in terms of
differences the 1% lows from the 1650 were again above the averages provided by the 1050
Ti, while at 1080p the 1650 was 18% ahead of the 1050 Ti in average FPS, and 16.7% ahead
at 1440p. Ghost Recon was tested with the built in benchmark
and high settings in use, and is another resource heavy game. At 1080p the 1650 was 23.6% ahead
of the 1050 Ti in average FPS, lowering to a 17.5% lead at 1440p. The Witcher 3 was tested with Hairworks disabled.
At 1080p the 1650 was achieving 25% higher average frame rates over the 1050 Ti, lowering
again to 22% at 1440p, although at 1440p there was a much larger 59% boost to 1% low. Strange Brigade was tested with the built
in benchmark using Vulkan with Async compute enabled maxed out at ultra settings. At 1080p
the 1650 was seeing its largest improvement out of all 18 games tested with 37% higher
average FPS over the 1050 Ti, and also a 37% improvement at 1440p, coming in at second
place out of 18 at this resolution. DOOM was also tested using Vulkan, and there
was quite a big difference in this game too. At 1080p the 1650 was 31% ahead of the 1050
Ti, rising up to a 36% boost at 1440p, while the 1% lows from the 1650 were above the averages
from the 1050 Ti. Shadow of War was tested with the built in
benchmark, and like Rainbow Six is another game I’ve found to benefit from Nvidia’s
Turing architecture. At 1080p it came in at second place with 34% higher average FPS over
the 1050 Ti, and 35% higher at 1440p. In terms of overall improvement, over all
18 games tested with a 1080p resolution on average the GTX 1650 was performing 26.7%
better when compared with the GTX 1050 Ti in terms of average FPS. As we can see it
really depends on the specific game, esports titles like CS:GO, Overwatch and Fortnite
for example were closer towards the bottom. At 1440p on average over the same 18 games
the GTX 1650 was now scoring just a little higher on average compared to the 1080p results
just shown, now with 27.1% higher average FPS. Again there’s no major differences
to those esports titles down the bottom of the graph, while more GPU demanding games
were seeing larger improvements. It’s worth noting though that I don’t consider either
of these 1440p capable cards in the majority of games tested, I just wanted to see how
the resolution would affect the percentage difference as it should be more GPU intensive
when compared with 1080p. Here’s what we’re looking at in terms
of synthetic benchmarks, I’ve just tested 3DMark’s Firestrike, Timespy and VRMark
here. I’ll note that in Timespy with the 1650 I did have some black screen flickering
which some others have reported, so I’m not 100% certain whether that affected results. As for the differences in total system power
draw the 1650 was only slightly ahead, so it is at least good to see that not that much
extra power is required here for a fair boost in performance. Now let’s look at one of the main reasons
I was interested in making this video. I saw Nvidia tweet this image the other day showing
the performance boosts to expect from their 1650 graphics, and rather than take them at
face value I wanted to see if I could work out just how accurate and honest they’re
being. In typical Nvidia fashion they’re not providing the exact frame rates, so we’re
not off to a good start, but we can extrapolate the information to get a rough idea. They
do note that they’re using an 8700K system with 16GB of memory at 1080p, and while no
exact speeds are given that’s quite similar to my setup. So how did my results compare to Nvidia’s
claims? PUBG seems pretty easy to work out, 60 FPS at 1080p in their 1650 test and it
looks like around 46 or 47 FPS for the 1050 Ti, just 2-3 FPS below my results. That would
mean Nvidia are claiming a 30.4% performance increase, and well as we saw I was seeing
almost a 31% boost, so that’s looking about right. In Fortnite as best I can tell here, Nvidia
are showing 59 FPS with the 1050 Ti and 74 FPS with the 1650, again fairly close to my
results, though resulting in a 25% improvement with the 1650, which is a little above my
20.5% difference. Rainbow Six Siege on the other hand I feel
is a bit of a best case result though, I’ve noted in many comparisons that the game seems
to perform noticeably better with Nvidia’s new Turing cards over Pascal. In any case
though I wasn’t quite seeing the big difference they’ve got here. Looks like the 1050 Ti
is around maybe 45 FPS, about in the center of the 30 and 60 FPS points, while the 1650
appears to be around 70 FPS, which gives it a massive 59% improvement if I’m right.
At 1080p I saw a 33.5% increase on the 1650, however I was using the built in benchmark.
Steve at Hardware Unboxed tests this game in actual game play, and even if we use his
numbers there was a 46% difference, so I’m not too sure how Nvidia got that one, I presume
different parts of the game would give different results, though even in all of my 18 game
tests I was nowhere near close to a 59% improvement on the 1650. Apex is a bit easier to work out, the results
are the same as what they had for PUBG, so around 46 or 47 FPS most likely with 60 FPS
from the 1650, fairly close to my own results, and again around a 30% improvement, once more
close to the 31.8% improvement I was seeing. It looks like Nvidia are at least being honest
here and not providing super skewed results for the most part, though Rainbow Six was
a bit iffy, now let’s start talking about price and value. I suggest checking updated prices using the
links in the description, as prices will change over time. The 1650 just launched at $149
USD and can currently be bought for that price on the low end. While the 1050 Ti, which came
out around two and a half years ago in October 2016, had an MSRP of $139, however today it’s
going for closer to the $180 USD mark on the lower side. At first glance you’d think
Nvidia has put out a card that’s cheaper than what the 1050 Ti is currently available
for while also performing better, everyone’s a winner, right? If we lived in an Nvidia
only world, maybe, however that’s simply not the case. The fact is AMD are offering the RX 570 for
less than the GTX 1650, around $130 USD new, and the 570 performs better as well, it’s
much better value. This is before we even consider that it’s been out for 2 years
already, so you can probably pick one up cheaper on the second hand market. While I’m not
explicitly comparing to the 570 in this video, it would have been a crime not to mention
this, and don’t worry I will also be comparing the 570 against the 1650 in an upcoming video,
I just wanted to show the difference between the 1050 Ti and the 1650 here as I saw a lot
of people who were curious in the differences. In isolation the 1650 does look like a fair
improvement over the 1050 Ti, which is exactly how Nvidia seem to be marketing it, but once
you take into account options from AMD it’s not great at all. If you were buying today let me know which
card you’d pick down in the comments and why, the newer GTX 1650 or older GTX 1050
Ti, and if either of these why not the RX 570? I’d be interested in hearing what you’re
picking and why, and if you’re new to the channel get subscribed for future comparisons
and tech videos like this one.

100 thoughts on “Nvidia GTX 1650 vs 1050 Ti – 18 Games Compared

  1. How much better is 1060 3gb compare to 1650 ? I really want a decent laptop now but 1060 6gb is a little bit high for my budget but 1050ti is a 'meh' according to my friend. I'm using 1060 3gb on my pc atm and i'm satisfied with it so i idk if 1650 is close to that one. I'm aiming for those new MSI laptop with 1650 but it's MAX-Q design. Again idk if max-q design affect too much on performence or not. I've been struggling for weeks trying to find the most suitable laptop. So i hope u or anybody would read this and give me an answer :p tks in advance

  2. I have a 1050 Ti and this would be a bad upgrade for me. Granted it's a nice performance boost but it costs more than the 1050 Ti did 3 years ago. And the 1050 Ti can be overclocked to give an extra 10-15% fps which brings it closer to the 1650 anyway. Plus the 1050 Ti is still plenty for 1080p if you turn a few detail settings down here and there. Basically it needs to be a lot cheaper.

  3. buy 1060 6gb/980 4gb or Rx 580/r9 390 as a minimum anything below that performance level is a scam and your going to need to buy new cards more often if you buy the "budget options" just say no and try to buy a little higher to not need another card for a few years more.

  4. I'm kinda surprised that you only have 66k subs. You always post quality content, keep up the good work Jarrod

  5. spoiler alert: asus gx701 has i7-9750H with rtx 2080 max q. released in may 27,2019 looking forward to buy it

  6. Tl:DR CAN I TRUST THIS SITE AND THE SPECS COMPARED TO PRICES. https://hst.co.uk/gaming-pc-s.html#p=1 can someone more experienced help me out here, I got my first "gaming pc" last year for 400 and thought it was good and the beginning, but the graphics card turned out to not work at all and I had to rely on integrated graphics. It runs i5 8gb ram and intel 5500 graphics. Im looking for an upgrade, mainly in the graphics department. I'm sure when I'm older I will build my own pc but as of now I don't have the time or money. I have been saving up for a prebuilt i5 8gb gtx 1060 6gb on Amazon for £600-700 however I have seen this site's prices and for the specs they seem great, I feel like their may be a catch, but even so I'm not gonna be playing at extreme settings, I've already had people tell me to build my own so plz don't say that. I just want to know if I can trust the specs compared to the prices on this site. Sos for long comment thanks.

  7. It really break my heart, when i just bought an laptop with 1050 for 950 USD (MSI GF63), why indonesia electronic so freaking expensive. *Sigh

  8. while RX 570 seems like a great option on paper and great for desktops only world, this is not the world we live in, and for laptops 1650 should be a great option. The only thing that bothers me is it's high power consumption, meaning no 1050 equivalent for laptops in terms of heatin this generation. The max q version is way too cut down to 35 watts, while the full laptop 1050 were well balanced 53 watts

  9. The single fan GTX 1650 would work for me in my EVGA Hadron case. Air flow is not too great in that case and the small card also gives more room since it is short and there is no need for a 6 pin pcie connector.. I currently have a 1050 ti and will probably wait another generation before upgrading.

  10. I just want to know one thing: why should I buy a 1650 instead RX 570 ? notice iF I ask "why RX instead 1650" the answers and concepts would vary. (,o0)

  11. Would be great to see a comparison of 1660 and 2060 with raytracing turned off. 1660ti looks especially nice for the money

  12. Basically, if you have 1050Ti, you should probably upgrade to 1650 even though with a slightly higher price i could have been so much better

  13. Invalid. You should not have used a 1650 model with a 6-pin connector. An over-75W model of this card is utterly pointless. The only point of the card is the 75W variant.

  14. See how the GTX 1650 compares with the AMD RX 570 in my new video: https://www.youtube.com/watch?v=FhT392WsD-Y

  15. The thing that EVERY freaking reviewer is missing to mention as the only + the 1650 has : it requieres no external power = it can be put into crapton of systems where you can just look at the Rx 570 ( i love the card but damn ), the only missing point is Low-profile, as soon as Low-profile 1650 hits the market, yea the price isnt great but show me other GPU tak can run off of 200W Oem powersupply in older work station with i5 4. gen or newer.

  16. The 2 years old RX 570 is cheaper and much faster than the GTX 1650! I wanted to go for my first nVidia card, but I was disappointed…

  17. Get a dell optiplex with an i7 4790 and 8 or 16gb of ram ($200) and then throw a 1650 in it (150) and boom, an amazing 350 dollar "gaming PC"

  18. These two gpu's are not made for high end cpu's you should benchmark on 3rd gen cpu's because 1050ti is ideal for 3rd gen stuffs here your i7 is holding this graphics card with his both hands and adding more fps to gpus and i think this is not a genuine benchmark…come on man nobody is gonna use 1650 or 1050ti on 7th gen i7

  19. 1650???? With those results, why in the hell did they even make the card in the first place??             Seriously, its a ripoff.

  20. Well yeah, rx570 is a better performer for the same price. However, it consumes a 100% more power. Is a 15% fps improvement worth it? If you don't give a crap about your electricity bill – YES, but if you do care – no.

  21. Guys, could you please help me? I have 7970 ghz edition gpu, mostly I play games like: Rust, csgo, overwatch, apex. I play games at 900p resolution. I've notice that my current gpu isn't that good as it used to. In rust I'm getting 30-47 fps. I really like gtx 1650 because it's silent, 75w, and I think it's enough powerful to run games I play, I just want to be sure if only gtx 1650 let me run rust at 900p at stable 60 fps?

  22. how are you calculating the percentages cuz I'm getting different results (i mean the rates you are talking about that it is 30% more faster) please reply

  23. If you already own a GTX 960, 970, 1050, 1050ti or an RX 470, 480, 570, or 580 video cards, it's not financially sound to buy a replacement card that subjectively isn't going to perform well enough to justify paying for a similar item within 3 years of the graphics card you are using now. Most of the used competitor hardware can be obtained below 120 dollars and even the retail versions are pricing below the sales level of GTX 1650 cards. Due to the current pricing levels of the GTX 1650, most critics suggest you should not consider the GTX 1650 at all under any situation.

  24. What about the upcoming laptops like the Dell XPS 15 and the Lenovo x1 extreme? Would you still get their "first gen"? (8th gen cpu) and GTX 1050 ti

  25. If it weren't $150 this sounds like it'd be a pretty solid upgrade… Drop the settings down to medium and you're getting 60 or more FPS on most of these games. Ideal if you have a weaker PSU and can't run a 570/580. It's too bad nVidia's cards are so overpriced.

  26. when I saw this channel at my first look , I thought it is having atleast 1M subs because of its quality, camerawork and narration
    but when I saw the subs I was badly shocked
    same thing happened with own or disown

  27. If i have to chose betwen those pc which one shold i get
    I5-3470
    240ssd
    12Gb ram ddr3
    Gtx 1650 or 1050ti
    Price:300 euro for both of them

  28. thanks for this video mate , you really helped me to make up my mind, i'm planning to have my own gaming cafe, most owners here use GTX 1050TI in their PCs so i was thinkin about having the RX 570 but its more expensive than the 1650 or 1050ti in my country plus it consumes more power so imagine more than 10 pcs would be bad for business lol the GTX 1650 & 1050TI otherwise are on the same price

  29. 1050 ti 4gb is garbage i'm getting this 1650 i have 240 watt power supply hp compaq 6035 can't replace power supply

  30. I personally own a 1050 ti and I don’t plan on upgrading soon. I’ve overclocked the ram, cpu and cpu on my budget gaming rig and it performs awesome!

  31. As long as I can get 30fps at 1080 I am okay. I am one of those budget gamers nobody talks about. My next card will be in at least 2 years time, so if I have to play games at medium to get 30fps+, that's what i'll do. Many YT video's imply gamers have all the money in the world.

  32. Great video dude. I was thinking about upgrading my 1050 ti to the 1650 at some point, because my fps always drop a little when I record gameplay for my channel. And need something a little bit more powerful to reduce the drops. I would rather go for the 1660 though, but I don't want to have to buy and new power supply along with that gpu. Anyway, this video helped a lot because I was not sure if getting the 1650 was worth it at first, but now I can see it is.

  33. How come every single time I'm in need of a good comparison video your channel always has it already done, I love you.

  34. You guys do realize that prices vary a lot over different countries. I bought a gtx1650 gaming OC, 6 weeks ago with 178€, as it was the best I could get in my budget, the rx570 having the same price while needing an 8 pin cable power which my power source doesn't have. So It would have forced me to also upgrade my power source. I'm from romania btw. While the the 1050 was only 20 euros cheaper and it's an older, slower gpu.

Leave a Reply

Your email address will not be published. Required fields are marked *

Copyright © 2019 Explore Mellieha. All rights reserved.