What CPU should I get?

[Donz]

For my graphics I have a GTX 1050 TI 4GB and I am looking to upgrade my CPU. Any suggestions for a good CPU to pair with the 1050 TI? Just need something good for gaming. I currently have a random no-name mobo that came with the PC with an AMD FX-6300. Looking for anything from $100-$200. I am going to upgrade the mobo accordingly with the CPU.

[another random person]

2 minutes ago, Donz said:

For my graphics I have a GTX 1050 TI 4GB and I am looking to upgrade my CPU. Any suggestions for a good CPU to pair with the 1050 TI? Just need something good for gaming.

Ryzen 3s and Ryzen 5s are excellent for that. If you want there are also good options on the Intel side of things like the i5 7500 etc.

[SoreRaito]

4 minutes ago, Donz said:

For my graphics I have a GTX 1050 TI 4GB and I am looking to upgrade my CPU. Any suggestions for a good CPU to pair with the 1050 TI? Just need something good for gaming.

What Cpu and Mobo do you have now ?

and is there a budget ?

 

[LienusLateTips]

Just now, another random person said:

Ryzen 3s and Ryzen 5s are excellent for that. If you want there are also good options on the Intel side of things like the i5 7500 etc.

Ryzen 5 1600 wrecks i5 7600K though so I don't really recommend the i5 line

[genexis_x]

Ryzen 3 or Ryzen 5

[another random person]

2 minutes ago, JDE said:

Ryzen 5 1600 wrecks i5 7600K though so I don't really recommend the i5 line

Well, many intel fanboys get salty when I don't mention intel but yes Ryzen is the way to go. 

[Colexd]

ryzen 1200, 1600 with b350, or a locked i5 if you have intel mobo already

[Zyndo]

23 minutes ago, Donz said:

For my graphics I have a GTX 1050 TI 4GB and I am looking to upgrade my CPU. Any suggestions for a good CPU to pair with the 1050 TI? Just need something good for gaming. I currently have a random no-name mobo that came with the PC with an AMD FX-6300. Looking for anything from $100-$200. I am going to upgrade the mobo accordingly with the CPU.

Ryzen is almost certainly the way to go for a user like you. an R3 1200 or R5 1600 and overclock them would be wise. the R5 1400 and 1500x are fine too, but for how little price difference there is between them and the 1200/1600, and how almost non-existant the performance difference is, you would be better off saving money and getting the 1200, or biting the bullet and stepping up to the 1600.

With your current GPU, the 1200 and 1600 will likely perform incredibly close to one another, but a couple years down the road when you go to upgrade again (this time your GPU) you will definitely feel the difference between a 1200 and 1600 when paired with that new GPU... so I would highly recommend working an R5 1600 into your budget for that reason if you can afford it, and if you cannot then a R3 1200 will be just fine for you.

[LienusLateTips]

10 minutes ago, Zyndo said:

Ryzen is almost certainly the way to go for a user like you. an R3 1200 or R5 1600 and overclock them would be wise. the R5 1400 and 1500x are fine too, but for how little price difference there is between them and the 1200/1600, and how almost non-existant the performance difference is, you would be better off saving money and getting the 1200, or biting the bullet and stepping up to the 1600.

With your current GPU, the 1200 and 1600 will likely perform incredibly close to one another, but a couple years down the road when you go to upgrade again (this time your GPU) you will definitely feel the difference between a 1200 and 1600 when paired with that new GPU... so I would highly recommend working an R5 1600 into your budget for that reason if you can afford it, and if you cannot then a R3 1200 will be just fine for you.

1300X is higher binned than 1200 though

[Zyndo]

2 minutes ago, JDE said:

1300X is higher binned than 1200 though

also higher priced... and its not really going to affect performance if you manually overclock, only temperatures. almost all Ryzen can get to 3.9GHz, some can do 4.0GHz, few can do above that (under normal circumstances with a good amount of cooling). The X series CPU's tend to not really have that much overclocking headroom compared to their counterparts, they typically just reach the same overclocks with less voltage. since most Ryzen can do 3.9, and 4GHz is sort of a wall.... that +100MHz (which you may be able to get anyway on a non-X CPU) really doesn't amount to a whole lot of real-world performance difference.

[MyName13]

15 minutes ago, Zyndo said:

Ryzen is almost certainly the way to go for a user like you. an R3 1200 or R5 1600 and overclock them would be wise. the R5 1400 and 1500x are fine too, but for how little price difference there is between them and the 1200/1600, and how almost non-existant the performance difference is, you would be better off saving money and getting the 1200, or biting the bullet and stepping up to the 1600.

With your current GPU, the 1200 and 1600 will likely perform incredibly close to one another, but a couple years down the road when you go to upgrade again (this time your GPU) you will definitely feel the difference between a 1200 and 1600 when paired with that new GPU... so I would highly recommend working an R5 1600 into your budget for that reason if you can afford it, and if you cannot then a R3 1200 will be just fine for you.

I'm pretty sure that the difference can be felt even with that GPU because of frame times.Besides, r3 still isn't much better than Pentium g4560, it is somewhat better when overclocked but it costs around 60% more for maybe 10-20% difference, the reviews conflict with each other too much, in some it performs 50-60% worse than i5 7400 even whe overclocked with 3200 ram (gamers Nexus), in others it's around 15% behind i5 and some even show it matches i5.

[Zyndo]

9 hours ago, MyName13 said:

I'm pretty sure that the difference can be felt even with that GPU because of frame times

I'm pretty sure you're wrong. with a 1050ti and a 60hz 1080p monitor you will observe little to no difference between the R3 1200 and R5 1600 (assuming both properly overclocked). You could probably record and measure a difference if you benchmarked the two, but if i sat you down in a real-world blind test and asked you if you were using a 1200 or a 1600 it would be almost entirely a guess on your part.

[MyName13]

4 minutes ago, Zyndo said:

I'm pretty sure you're wrong. with a 1050ti and a 60hz 1080p monitor you will observe little to no difference between the R3 1200 and R5 1600 (assuming both properly overclocked). You could probably record and measure a difference if you benchmarked the two, but if i sat you down in a real-world blind test and asked you if you were using a 1200 or a 1600 it would be almost entirely a guess on your part.

So why does the tech press make i5 CPUs look like they can't be used for gaming because of stuttering?Majority of people play on 60 hz monitors, I'm under the impression that it would be better to buy r5 1600+ gtx 1050 ti than r5 1400+rx 570 because of frame times (BTW resolution is irrelevant to CPUs if we're talking about the same frame rate).

[Zyndo]

Just now, MyName13 said:

(BTW resolution is irrelevant to CPUs if we're talking about the same frame rate).

btw you're wrong. 4k @60hz is more cpu intensive than 1080p @60hz    

 

1 minute ago, MyName13 said:

I'm under the impression that it would be better to buy r5 1600+ gtx 1050 ti than r5 1400+rx 570 because of frame times

that may be true for a very small number of people, or guys who want to stream whilst they game, or have poorly kept computers, or something along those lines... but when you're talking about using such weak graphics cards, your choice of CPU means very little. I understand how you are under that impression, because that kind of ignorance is quite prevalent on the interwebs nowadays and you probably took that information at face value from one or more people who don't really know what they're talking about, but the 1400+570 system would get you more fps at higher graphical settings in a larger range of games resulting in an overall better gaming experience than the 1600+1050ti system would.

 

6 minutes ago, MyName13 said:

So why does the tech press make i5 CPUs look like they can't be used for gaming because of stuttering?

Beacuse Youtube tech press doesn't know what they're talking about. I5's were never discussed as stuttery or "bad" experiences pre-Ryzen... in fact, from most reviews and people regarded them as the ideal gaming CPU's. in fact even now I don't hear much of what you're saying, but the discussion will always be that "Ryzen has lower 1% and .1% lows" because of its higher core counts.... which is somewhat true. i5's are largely irrelevant since Ryzen offers equal to or more performance in most cases compared to the I5's. However, you often will hear that Ryzen even offers better frametimes than a well overclocked 7700k, which is simply not true in the majority of instances regarding a properly balanced computer. 

[MyName13]

22 minutes ago, Zyndo said:

btw you're wrong. 4k @60hz is more cpu intensive than 1080p @60hz    

 

that may be true for a very small number of people, or guys who want to stream whilst they game, or have poorly kept computers, or something along those lines... but when you're talking about using such weak graphics cards, your choice of CPU means very little. I understand how you are under that impression, because that kind of ignorance is quite prevalent on the interwebs nowadays and you probably took that information at face value from one or more people who don't really know what they're talking about, but the 1400+570 system would get you more fps at higher graphical settings in a larger range of games resulting in an overall better gaming experience than the 1600+1050ti system would.

 

Beacuse Youtube tech press doesn't know what they're talking about. I5's were never discussed as stuttery or "bad" experiences pre-Ryzen... in fact, from most reviews and people regarded them as the ideal gaming CPU's. in fact even now I don't hear much of what you're saying, but the discussion will always be that "Ryzen has lower 1% and .1% lows" because of its higher core counts.... which is somewhat true. i5's are largely irrelevant since Ryzen offers equal to or more performance in most cases compared to the I5's. However, you often will hear that Ryzen even offers better frametimes than a well overclocked 7700k, which is simply not true in the majority of instances regarding a properly balanced computer. 

Check a video on YouTube, it's called something like "gtx 1080 ti + Pentium g4560 4k", 60 fps at any resolution strains the CPU equally, can you show me what made you think otherwise?4k does have more data to work with, but that's the GPU that has to work harder.

 

About the tech press, check digital foundry i5 7600k review from the beginning of the year and the new video i5 7600k vs ryzen 1600, i5 is somehow much more stuttery than it was back when they tested it (check the far cry primal part), it went from constant ~13 Ms to stuttery mess that goes from 13 Ms to 30 Ms.

In fact ryzen 1600 doesn't seem to have better minimums even when compared to i5 7600k according to gamers nexus, someone is either lying or cant properly benchmark.

[Zyndo]

42 minutes ago, MyName13 said:

Check a video on YouTube, it's called something like "gtx 1080 ti + Pentium g4560 4k", 60 fps at any resolution strains the CPU equally, can you show me what made you think otherwise?4k does have to work with more pixels, but that's the GPU that has to work harder

Don't get me wrong, the performance difference at 4k between any 2 given CPU's is almost non-existent (assuming we aren't using some lower end pentiums or celerons or something like that) but take the same CPU, at the same framerates, and compare them between a 1080p and 4k resolution and the 4k resolution will show slightly higher CPU usage

42 minutes ago, MyName13 said:

About the tech press, check digital foundry i5 7600k review from the beginning of the year and the new video i5 7600k vs ryzen 1600, i5 is somehow much more stuttery than it was back when they tested it (check the far cry primal part), it went from constant ~13 Ms to stuttery mess that goes from 13 Ms to 30 Ms.

In fact ryzen 1600 doesn't seem to have better minimums even when compared to i5 7600k according to gamers nexus, someone is either lying or cant properly benchmark.

People simply cannot properly benchmark... they just don't know how. Take our overlords LTT, how many times in the last year have they done CPU benchmarks at an, exclusively, 4k resolution? WAYYY TOO MANY DAMN TIMES... they just don't get it (or they've been bought out by AMD, which there is pretty clear evidence to support, but i'm not a tinfoil hat kinda guy and don't really care if anyone agrees with me or not, that is a discussion for another day, for example, check out the TR review posted today). Can't do a test at 4k then say the two CPU's are equal when gaming. LTT knows better, but a lot of other people do not. Similarly, any situation with a CPU bottleneck. all too often people will just use a 1060 or 1070 for their 1080p benchmarks because "anything above that is overkill" (which is factually untrue)  and then report their GPU bottlenecked numbers and state that Ryzen is a better chip because it happened to have a 2% increase over Intel in their tests (or they ran the tests until they got a number which supported this statement, or they only overclocked Ryzen, or they only gave Ryzen faster memory, or they somehow else intentionally/ignorantly skewed their own results, either for or against Ryzen vs Intel)

and it is the source of a great deal of misinformation out there on the internet today. To be fair, there will be some titles which the 1600's additional cores definitely will have it win out against an i5 7600k in average and/or minimum framerates.... if you throw in other kinds of simultaneous multitasking (something like streaming) then Ryzen will almost definitely win in the majority of situations. But all things being equal on a clean system, the 7600k is going to win (assuming proper overclocks and memory overclocks) in a larger number of games when regarding both averages and minimum framerates.

[MyName13]

1 hour ago, Zyndo said:

Don't get me wrong, the performance difference at 4k between any 2 given CPU's is almost non-existent (assuming we aren't using some lower end pentiums or celerons or something like that) but take the same CPU, at the same framerates, and compare them between a 1080p and 4k resolution and the 4k resolution will show slightly higher CPU usage

People simply cannot properly benchmark... they just don't know how. Take our overlords LTT, how many times in the last year have they done CPU benchmarks at an, exclusively, 4k resolution? WAYYY TOO MANY DAMN TIMES... they just don't get it (or they've been bought out by AMD, which there is pretty clear evidence to support, but i'm not a tinfoil hat kinda guy and don't really care if anyone agrees with me or not, that is a discussion for another day, for example, check out the TR review posted today). Can't do a test at 4k then say the two CPU's are equal when gaming. LTT knows better, but a lot of other people do not. Similarly, any situation with a CPU bottleneck. all too often people will just use a 1060 or 1070 for their 1080p benchmarks because "anything above that is overkill" (which is factually untrue)  and then report their GPU bottlenecked numbers and state that Ryzen is a better chip because it happened to have a 2% increase over Intel in their tests (or they ran the tests until they got a number which supported this statement, or they only overclocked Ryzen, or they only gave Ryzen faster memory, or they somehow else intentionally/ignorantly skewed their own results, either for or against Ryzen vs Intel)

and it is the source of a great deal of misinformation out there on the internet today. To be fair, there will be some titles which the 1600's additional cores definitely will have it win out against an i5 7600k in average and/or minimum framerates.... if you throw in other kinds of simultaneous multitasking (something like streaming) then Ryzen will almost definitely win in the majority of situations. But all things being equal on a clean system, the 7600k is going to win (assuming proper overclocks and memory overclocks) in a larger number of games when regarding both averages and minimum framerates.

How does higher resolution put more strain on the CPU?

 

I'm glad that I'm not the only one who sees an unnatural bias towards ryzen, DF is pretty suspicious just like gamers nexus and hardware unboxed.Most of the time it's "ok let's compare this stock Intel with 2400 ram with the crappiest cheapest Mobo to similarly priced ryzen on 30$ more expensive motherboard with 3200 MHz ram overclocked to 4 ghz.Oh look, it's just few percent behind the intel CPU, it's a much better value even with higher price", WTF is going on with tech tubers?

[Zyndo]

17 minutes ago, MyName13 said:

How does higher resolution put more strain on the CPU?

i asked the same question in the comment section of that video... the response from Science Studio himself (I think his name is Greg Salazar?) was "4k has a higher poll rate". I asked what that meant and didn't get a response, but I asked my brother who is a game coder and he said that it makes sense (he does it mostly as a hobby level thing since he was like 8, but is currently in university for that sort of thing, so he should have some vague idea what hes talking about ). Its not going to make a massive difference like framerates does (4x resolution compared to 4x framerates that is) but the small difference is there. 

 

17 minutes ago, MyName13 said:

I'm glad that I'm not the only one who sees an unnatural bias towards ryzen, DF is pretty suspicious just like gamers nexus and hardware unboxed.Most of the time it's "ok let's compare this stock Intel with 2400 ram with the crappiest cheapest Mobo to similarly priced ryzen on 30$ more expensive motherboard with 3200 MHz ram overclocked to 4 ghz.Oh look, it's just few percent behind the intel CPU, it's a much better value even with higher price", WTF is going on with tech tubers?

I think people are just tired of Intel not pushing the envelope over the last few years.... so they're looking for any and all reasons to fling crap Intel's way. Don't get me wrong, MOST people right now should be buying a Ryzen system even if they're just gaming... the relative cost in comparison to performance, future performance scaling in games which are beginning to leverage more cores than the traditional 4, and the fact that unless you have a monstrous budget (which most people do not have) you're going to run into a GPU bottleneck in almost all games at almost all reasonable resolutions and settings anyway so intel's "superior single core performance" is mostly irrelevant anyway.

I just wish people could be more objective about their tests and results so that we could make up our own minds on the matter, rather than trying to convince us to one party based on their own personal views. I'm considering starting my own YT channel for this sort of thing, but I just don't have the money laying around to go willy nilly and buy several platforms worth of hardware just to run these tests myself, especially since I would have no real fanbase or following out the gate... not to mention I live in Canada so everything is 25-35% more expensive than it is in the states haha.

[Biggerboot]

12 hours ago, Donz said:

For my graphics I have a GTX 1050 TI 4GB and I am looking to upgrade my CPU. Any suggestions for a good CPU to pair with the 1050 TI? Just need something good for gaming. I currently have a random no-name mobo that came with the PC with an AMD FX-6300. Looking for anything from $100-$200. I am going to upgrade the mobo accordingly with the CPU.

What games are you looking to improve in?  If you're primarily looking to game on competitive games like CSGO you could take a look at Ryzen 3 or Pentium G4560 with a decent board and RAM.  Those games are considered 'legacy', and prefer higher IPC, which is what the FX chips lack.

On the other hand, you could give overclocking your FX a shot.  If you're just looking to get into newer games that take advantage of more cores, that could go a long way.