Gaming/Light Editing CPU choice

[i_build_nanosuits]

I have to agree with you on another thing though, I was mistaken all along cause I got one heavy bottleneck with my FX that I forgot about. It's partly the CPUs fault and partly the developers of the game I guess, but oh well, the i5-4690k my friend has runs the game way better. I'm talking about TERA Online, the framerates are sometimes terrible on my FX and I do not know why. CPU usage doesn't exceed 40% on any core and GTX970 is not even worth mentioning.

This i agree with you is caused by both the CPU having a very low IPC count and the devellopers leaving too much work for the CPU in the engine...but it happen a lot...so many games on steam are like that...pretty much any MMO or RTS/RPG you'll find will do that...and the CPU load does not need to be high for bottlenecks to happen, it's simply that the CPU is not able to process instructions fast enough for the game to run optimaly...planetside 2 for example struggle to run at 30FPS on an overclocked FX but will run at 50+FPS on a core i3 processor for example...and those represent a lot of games...and when you take everything in between the two into account, that totals to many titles where your GTX970 would absolutely be held back a huge amount by your CPU especialy if you aim for 120FPS gaming for example...but even 60FPS is a number the FX will struggle to maintain in quite a lot of games out there.

But, yeah for modern GPU intensive top of the line AAA titles such as the witcher 3, dragon age inquisition for example or even most modern FPS shooter wheter it's the battlefield games or call of duty the FX is ok simply add 8xMSAA and boom the gpu load is 99% on ultra settings...obviously...and again as i said your resolution is also 11.1% greater than most users who play at HD1080p.

[Morgan MLGman]

@i_build_nanosuits I didn't think of the 1920x1200 as a really higher resolution when I was buying it, that was not my intention, simply 16:10 ratio seemed way more natural to me than the "streched" 16:9. Of course it was much more expensive monitor back then cause of it but oh well, I wasn't the one paying.

Perhaps I'm just an exception that doesn't play the titles that experience any drawbacks of having the FX-8350 besides TERA which I stopped playing anyway.
I'm defending the FX-8350 cause for the price I got it + Gigabyte GA-970A-UD3 motherboard (185$ combined) I do not think I could've gotten anything better especially for the games I play...

I could get an i3 for the same price so I decided to not do that.

[i_build_nanosuits]

@i_build_nanosuits I didn't think of the 1920x1200 as a really higher resolution when I was buying it, that was not my intention, simply 16:10 ratio seemed way more natural to me than the "streched" 16:9. Of course it was much more expensive monitor back then cause of it but oh well, I wasn't the one paying.

Perhaps I'm just an exception that doesn't play the titles that experience any drawbacks of having the FX-8350 besides TERA which I stopped playing anyway.

I'm defending the FX-8350 cause for the price I got it + Gigabyte GA-970A-UD3 motherboard (185$ combined) I do not think I could've gotten anything better especially for the games I play...

I could get an i3 for the same price so I decided to not do that.

To be honest there was many games where i was able to max out my GTX780 and get good framerates as well...mostly when you seek recent games with the nicest graphics obviously...but the thing is i play all sort of games and people building modern gaming rig will also play all kind of games and you want a CPU that play all of them well when you're sinking 400$+ into a GPU you should get a proper CPU to go along and in that regard a 200$ intel solution completely decimate a 200$ AMD solution such as an overclocked FX-8350 on a 8+2 phase board with an aftermarket cooling solution...that's why i really don't see myself recommending an AM3+ setup for a gamer anytime soon unless the budget is very low and worstation applications are mentionned.

The AMD FX make a good match for mid-range cards at 1080p such as the R9 280/380 or GTX 760/960 though, but offer no upgrade paths and consume a lot more energy than a core i3 or a low end i5 that would do the same job or better.

For a long term processor on a budget i would probably pick the FX-83XX over a core i3 as well to be honest, but i would not i would get an i5-4460 and a cheap board cause that will end up costing the same and will be more consistent in games.

[Morgan MLGman]

@i_build_nanosuits btw. what do you think of GTX 7xx series cards? Right now I can sell my GTX970 and get a GTX780Ti for the same price, which technically is a better card.
Although I've heard about issues with Kepler (6xx and 7xx series) architecture based GPUs, and that in for example TW3 the 780Ti was performing on par with a  GTX 960...
I owned a 660 previously so also technically a Kepler but had no issues with it whatsoever.

Here's a comparison between a GTX 970, a 980 and a 780Ti in The Witcher 3 patch 1.02

970 wins with 780Ti all the time despite technically being a worse GPU... Is it Nvidia's greedyness to lower Kepler based GPU's performance?

[i_build_nanosuits]

@i_build_nanosuits btw. what do you think of GTX 7xx series cards? Right now I can sell my GTX970 and get a GTX780Ti for the same price, which technically is a better card.

Although I've heard about issues with Kepler (6xx and 7xx series) architecture based GPUs, and that in for example TW3 the 780Ti was performing on par with a  GTX 960...

I owned a 660 previously so also technically a Kepler but had no issues with it whatsoever.

Driver support is still good there was one previously issued driver that lowered performance a little in some games but it's been fixed since my card still perform right in line with where it should be. Now i wouldn't trade a 970 for a 780ti...the 780ti is an older die and has been used longer, and it sounds like your 970 is a good overclocker and at 1500mhz should perform pretty much on par with a stock 780ti anyways, and you have an extra 500mb of VRAM on the 970...and it consumes less energy...so unless we're talking an MSI Lightning GTX780ti vs a crappy GTX970 from EVGA or reference model you know something with loud fans and bad cooling i wouldnt trade it.

[Morgan MLGman]

Driver support is still good there was one previously issued driver that lowered performance a little in some games but it's been fixed since my card still perform right in line with where it should be. Now i wouldn't trade a 970 for a 780ti...the 780ti is an older die and has been used longer, and it sounds like your 970 is a good overclocker and at 1500mhz should perform pretty much on par with a stock 780ti anyways, and you have an extra 500mb of VRAM on the 970...and it consumes less energy...so unless we're talking an MSI Lightning GTX780ti vs a crappy GTX970 from EVGA or reference model you know something with loud fans and bad cooling i wouldnt trade it.

This is my exact card: http://www.newegg.com/Product/Product.aspx?Item=N82E16814127850&cm_re=970_100_ME-_-14-127-850-_-Product

I bought the limited edition cause well, it was limited, had a backplate, and cost the same as the normal, red one, which didn't have a backplate.

The 780Ti's for the same price are: MSI Gaming one and a Zotac AMP! version, that got me curious cause like one and a half year ago they were almost twice the price of a 970 right now.

And the VRAM shouldn't be an issue, for my resolution 3GBs should be enough.

You might be right though, no point in switching to an older-gen card for no performance boost whatsoever basically considering the possibility of high overclocks with mine.

Thanks anyway for the good chat, I guess everything that needed to be said was said, hope the original poster made the right choice for him in terms of the CPU he needs.

[Nena Trinity]

Do you know the meaning of the term bottleneck? Also, its getting beaten by i3's that are weaker than the i5 3570K in DirectX 12, so its a bad choice for future games as well.

Wow, you do know that just cause some games runs worse does not mean all will right? i3 is not better than FX?

[Zoomv]

This is my exact card: http://www.newegg.com/Product/Product.aspx?Item=N82E16814127850&cm_re=970_100_ME-_-14-127-850-_-Product

I bought the limited edition cause well, it was limited, had a backplate, and cost the same as the normal, red one, which didn't have a backplate.

The 780Ti's for the same price are: MSI Gaming one and a Zotac AMP! version, that got me curious cause like one and a half year ago they were almost twice the price of a 970 right now.

And the VRAM shouldn't be an issue, for my resolution 3GBs should be enough.

You might be right though, no point in switching to an older-gen card for no performance boost whatsoever basically considering the possibility of high overclocks with mine.

Thanks anyway for the good chat, I guess everything that needed to be said was said, hope the original poster made the right choice for him in terms of the CPU he needs.

 

 

Driver support is still good there was one previously issued driver that lowered performance a little in some games but it's been fixed since my card still perform right in line with where it should be. Now i wouldn't trade a 970 for a 780ti...the 780ti is an older die and has been used longer, and it sounds like your 970 is a good overclocker and at 1500mhz should perform pretty much on par with a stock 780ti anyways, and you have an extra 500mb of VRAM on the 970...and it consumes less energy...so unless we're talking an MSI Lightning GTX780ti vs a crappy GTX970 from EVGA or reference model you know something with loud fans and bad cooling i wouldnt trade it.

 

 

Wow, you do know that just cause some games runs worse does not mean all will right? i3 is not better than FX are you trying to troll people?

nena360 i agree with you i3 isn't better than an fx 8350... 2 cores aren't enough anymore this days...

so this is going a little bit off topic... so i see we have mixed opinions about the 8350 here... some say that it 

is better than the i5 3570k other are saying that it's worse... now i have really no clue... tek syndicate

shows us that the 8350 is better... linus says that they have both more or less the same cpu power and that the 8350 is more future proof,

and other youtuber think that the i5 3570k is better like tek yes city... so whaaaat?

[Nena Trinity]

nena360 i agree with you i3 isn't better than an fx 8350... 2 cores aren't enough anymore this days...

so this is going a little bit off topic... so i see we have mixed opinions about the 8350 here... some say that it 

is better than the i5 3570k other are saying that it's worse... now i have really no clue... tek syndicate

shows us that the 8350 is better... linus says that they have both more or less the same cpu power and that the 8350 is more future proof,

and other youtuber think that the i5 3570k is better like tek yes city... so whaaaat?

i5 is more likely more compatible with games but the FX in truth has more power if USED RIGHT BY THE GAME/Program...

[i_build_nanosuits]

nena360 i agree with you i3 isn't better than an fx 8350... 2 cores aren't enough anymore this days...

so this is going a little bit off topic... so i see we have mixed opinions about the 8350 here... some say that it 

is better than the i5 3570k other are saying that it's worse... now i have really no clue... tek syndicate

shows us that the 8350 is better... linus says that they have both more or less the same cpu power and that the 8350 is more future proof,

and other youtuber think that the i5 3570k is better like tek yes city... so whaaaat?

http://cpu.userbenchmark.com/Compare/Intel-Core-i5-3570K-vs-AMD-FX-8350/1316vs1489

there you go, these are user submitted results they don't lie as you can see around 4900 submissions for both CPU's...the FX CAN be up to 30% faster than a stock i5-3570K (when ALL 8 THREADS ARE FULLY UTILISED SUCH AS WHEN RENDERING VIDEOS FOR EXAMPLE, THIS NEVER HAPPEN WHEN GAMING) BUT if you run applications that use 4 cores or less (SUCH AS 99.5% of the GAMES out there) the i5-3570K at stock is 40% faster than the AMD FX-8350. Keep in mind this is 3.4ghz intel vs 4.0ghz AMD...the intel outperform greatly in daily tasks and perceived speeds and especialy when it comes to gaming, ANYBODY telling you otherwise is clueless: THE i5-3570K is the SUPERIOR processor EXCEPT if you're building a purely workstation/rendering machine.

EDIT: BTW, when it comes to gaming, those are the stats in the bench you really want to consider more than any other:

Also read this, it sumarise it VERY well, this guy is a beast:

 

[Zoomv]

Well, since i am going to upgrade next year anyway i don't need futureproofness... the amd mb would need a ghetto mounting on a atx case anyway because it's e-atx

soo i think ii am going to spend 15 bucks more and i'll get a very good cpu for gaming which is what i do the most... even if the motherboard is crappy i don't need 

that many features and it can overclock decently... well thanks for your help guys...

[Morgan MLGman]

@Zoomv also, with a LGA1155 socket you can upgrade up to an i7-3770k and that is a good CPU, you cannot do that with the FX-8350, cause buying the FX-9XXX series is stupid, they require huge amount of power and they generate so much heat it's retarded.

The FX-8350 will be enough for top-tier games due to multi-threaded optimization of these games, but might struggle in some older games that utilize the CPU in a less efficient games, or some MMORPG games like TERA Online.

The choice is up to you, if you know you will benefit from the extra cores and you play recent games the FX is a good budget option, anything besides that I'd choose Intel. Maybe gather some money for the 4th gen i5 with the LGA1150 socket and not go for the older 3rd gen but it's all your choice.


Edit: And well, since you upgrade next year anyway I'd take the AMD cause even if you don't like it you still end up spending less money (provided it's actually cheaper in your country than Intel counterparts)

[Dabombinable]

@Zoomv also, with a LGA1155 socket you can upgrade up to an i7-3770k and that is a good CPU, you cannot do that with the FX-8350, cause buying the FX-9XXX series is stupid (well the (*** series is just the 8*** series with a factory overclock), they require huge amount of power and they generate so much heat it's retarded.

The FX-8350 will be enough for top-tier games (not really, if the DirectX 12 game benchmarks are anything to go off, the 8*** will continue to lose out to i3 in newer games) due to multi-threaded optimization of these games, but might struggle in some older games that utilize the CPU in a less efficient games, or some MMORPG games like TERA Online.

The choice is up to you, if you know you will benefit from the extra cores and you play recent games the FX is a good budget option, anything besides that I'd choose Intel. Maybe gather some money for the 4th gen i5 with the LGA1150 socket and not go for the older 3rd gen but it's all your choice.

Edit: And well, since you upgrade next year anyway I'd take the AMD cause even if you don't like it you still end up spending less money (provided it's actually cheaper in your country than Intel counterparts)

[Morgan MLGman]

@ the first fully DX12 supported game "Ashes of Singularity" was benchmarked in every possible way, and the developers of the game said  that "the 8 cored AMD CPUs perform almost are giving even a high end i7-4770k a hard time" that might be something you wanna know.

[Dabombinable]

@ the first fully DX12 supported game "Ashes of Singularity" was benchmarked in every possible way, and the developers of the game said  that "the 8 cored AMD CPUs perform almost are giving even a high end i7-4770k a hard time" that might be something you wanna know.

Lol, look again.

http://linustechtips.com/main/topic/432063-first-directx-12-game-benchmarked-update-2-more-benchmarks/

[Morgan MLGman]

@
http://www.pcworld.com/article/2971612/software-games/windows-10s-radical-directx-12-graphics-tech-tested-more-cpu-cores-more-performance.html?page=2

"Oxide developers told me their internal testing with the Ashes of the Singularity benchmark showed 8-core AMD CPUs giving even the high-end Core i7-4770K a tough time. "

Lol nope.
 

[Dabombinable]

@

http://www.pcworld.com/article/2971612/software-games/windows-10s-radical-directx-12-graphics-tech-tested-more-cpu-cores-more-performance.html?page=2

"Oxide developers told me their internal testing with the Ashes of the Singularity benchmark showed 8-core AMD CPUs giving even the high-end Core i7-4770K a tough time. "

Lol nope.

 

Simulated test, and go back and look at the results:

http://linustechtips.com/main/topic/432063-first-directx-12-game-benchmarked-update-2-more-benchmarks/

 

Also, there are synthetics then there are actual games-the i3 still pulls ahead in games:

http://www.pcworld.com/article/2971612/software-games/windows-10s-radical-directx-12-graphics-tech-tested-more-cpu-cores-more-performance.html?page=2

Earlier tests of DirectX12 performance using 3DMark’s feature test favored more CPU cores and Hyper-Threading over high clock speeds.

[Morgan MLGman]

Simulated test, and go back and look at the results:

http://linustechtips.com/main/topic/432063-first-directx-12-game-benchmarked-update-2-more-benchmarks/

 

Also, there are synthetics then there are actual games-the i3 still pulls ahead in games:

These were most likely real in-game benchmarks based on going from the point A to the point B.

Why would the developers of the game do 'simulated' non-relevant benchmarks on their own game? If the devs told that to the reported it has to be true dude.

And the benchmarks you linked might be fake, here's why: in the article it is also said that under DX12 games prefer more cores (preferably with hyperthreading) over high clock speeds. ANd that's reasonable cause if you utilize 8 cores by 30% it's better than using 4 cores by 60, if the usage hits 60% on 8 cores what will it do on 4?

That means more cores>stronger cores basically, which denies entire Intel's domain. The intel-paid benchmarks might be released a lot, doesn't mean you should believe in all of them.

[FLUFFYJELLO]

Simulated test, and go back and look at the results:

http://linustechtips.com/main/topic/432063-first-directx-12-game-benchmarked-update-2-more-benchmarks/

 

Also, there are synthetics then there are actual games-the i3 still pulls ahead in games:

 

Dude just stop debating with him, he has been shown the factual evidence and refuses to accept it. We have a name for the condition he suffers under (but every time I post it my post gets deleted). Just leave him to his vices, he will never understand why synthetics aren't how real world performance is judged.

[Morgan MLGman]

FPS on a 4790k with 980Ti - 4 cores with 1,7ghz perform equally to 2 with 3,9ghz.

[Dabombinable]

These were most likely real in-game benchmarks based on going from the point A to the point B.

Why would the developers of the game do 'simulated' non-relevant benchmarks on their own game? If the devs told that to the reported it has to be true dude.

And the benchmarks you linked might be fake, here's why: in the article it is also said that under DX12 games prefer more cores (preferably with hyperthreading) over high clock speeds. ANd that's reasonable cause if you utilize 8 cores by 30% it's better than using 4 cores by 60, if the usage hits 60% on 8 cores what will it do on 4?

That means more cores>stronger cores basically, which denies entire Intel's domain. The intel-paid benchmarks might be released a lot, doesn't mean you should believe in all of them.

Straight up, the results that I've linked for you to view are from the very same developers game-and the tests they did are explicitly designed to push hardware to its limits-which you never see in games unless its not powerful enough (or if there is a bottleneck). Please, stop making yourself look like an idiot.

[Morgan MLGman]

Straight up, the results that I've linked for you to view are from the very same developers game-and the tests they did are explicitly designed to push hardware to its limits-which you never see in games unless its not powerful enough (or if there is a bottleneck). Please, stop making yourself look like an idiot.

Well someone is spreading false infos then, cause one article says one thing the other says the opposite, I do not imagine how a PCWorld.com spreads false information but oh well, I cannot prove or disprove it, we'll just have to see how it develops in the near future I guess.

[Dabombinable]

Well someone is spreading false infos then, cause one article says one thing the other says the opposite, I do not imagine how a PCWorld.com spreads false information but oh well, I cannot prove or disprove it, we'll just have to see how it develops in the near future I guess.

You don't get it do you? Both articles are completely different as the testing methodology was completely different. 1 article is based around a synthetic benchmark, and the other around a real world benchmark-which shows the actual performance you should expect.