AMD's counterpart of i5-4590

[Opcode]

If you're building a machine for VR next year I would wait until that time to build it. You'll see what Skylake has to offer and Zen.

[Sky Daddy]

BOLDED FOR MAXIMUM IMPORTANCE - IE: Your GAMING EXPERIENCE

 

 

If your using this with OCULUS you need to KNOW MORE about it... the fact that it REQUIRES a MINIMUM CONSTANT of 75fps to match its 75HZ refresh...

 

REQUIRING 75FPS AT ALL TIMES for it to be immersive at all..... anything under 75fps, and it stutters like a bitch. (Worse than your used to in normal gaming)

 

Better hope your CPU can provide the most MINFPS you can. I know for a fact many titles have a really fucked up MINFPS for FX CPU's.

This is something I'd be worrying about.. not wasting a shitload of cash on something that will need to be replaced anyway

CV1 refresh rate is confirmed for 90HZ

[SkilledRebuilds]

CV1 refresh rate is confirmed for 90HZ

Sweet, can't wait to experience it

But the point stands, you'll need some serious CPU IPC for highend single-gpu power or a dual-gpu solution to provide a good min/avg fps as close to 90fps as you can manage.

[Sky Daddy]

Sweet, can't wait to experience it

But the point stands, you'll need some serious CPU IPC for highend single-gpu power or a dual-gpu solution to provide a good min/avg fps as close to 90fps as you can manage.

yah, oculus set the recommended spec for i5-4590 with 290/970 and will be targeting that performance

[flipped_bit]

CPUBoss is awful. As is Passmark, GPUBoss and so on.

 

Out of curiosity, what's wrong with Passmark?  CPUMark is my first reference when researching CPU performance.

[Faceman]

Out of curiosity, what's wrong with Passmark?  CPUMark is my first reference when researching CPU performance.

Those types of websites use a suite of synthetic benchmarks that often have no real relevance when it comes to gaming or workstation performance. When you want to figure out the strength of one processor compared to another, search for specific games or programs that you will be using.

[Blackie Sheen]

Go for a 4690k if possible. But the 4590 is better than any current AMD cpu

[flipped_bit]

Those types of websites use a suite of synthetic benchmarks that often have no real relevance when it comes to gaming or workstation performance. When you want to figure out the strength of one processor compared to another, search for specific games or programs that you will be using.

 

Oh yeah I don't disagree, but for a first point of reference CPUMark is great. 

[Faceman]

Oh yeah I don't disagree, but for a first point of reference CPUMark is great.

Not at all. It's point of reference is for something not at all relevant to your intended use of the machine. Websites like that show the FX8 processors beating out i5s by a decent margin, but in reality, outside of very few applications, the i5 is decidedly superior. These websites are not to be used, even as a point of first reference.

[flipped_bit]

Not at all. It's point of reference is for something not at all relevant to your intended use of the machine. Websites like that show the FX8 processors beating out i5s by a decent margin, but in reality, outside of very few applications, the i5 is decidedly superior. These websites are not to be used, even as a point of first reference.

 

That's true only if you look at the one chart.  There are many charts, including one for single-threaded performance.

 

https://www.cpubenchmark.net/

[i_build_nanosuits]

Thanks for answering my poll question earlier, my next question is what would be the AMD counterpart of i5-4590 since I'll be building an Oculus-supported rig next year.

 

PS: Seeing GameDebate's site, FX-6300 is a bit better than 4690K but I'm not sure if it is accurate.

the i5-4690K is vastly superior to the FX-6300 in every regards...and AMD has no CPU that can compete with even the i5-4430.

case closed.

[Faceman]

That's true only if you look at the one chart.  There are many charts, including one for single-threaded performance.

 

https://www.cpubenchmark.net/

While single thread performance is incredibly important, that number given by them isn't applicable to much.

1506(FX8350) compared to 2247(i5-4690k), this doesn't give us much insight into anything. From this, we can derive that the i5 is roughly 30% more powerful than the FX8350 in single threaded tasks, but if you were to check a program like Cinebench, that performance gap increases to almost 50%. So, comparing a suite of benchmarks to a single benchmark isn't entirely accurate. You need to check specific programs and the results of those processors in question within that one program.

[flipped_bit]

While single thread performance is incredibly important, that number given by them isn't applicable to much.

1506(FX8350) compared to 2247(i5-4690k), this doesn't give us much insight into anything. From this, we can derive that the i5 is roughly 30% more powerful than the FX8350 in single threaded tasks, but if you were to check a program like Cinebench, that performance gap increases to almost 50%. So, comparing a suite of benchmarks to a single benchmark isn't entirely accurate. You need to check specific programs and the results of those processors in question within that one program.

 

Looks like you made a math booboo.  The difference between the 4690K and FX8350 in single-threaded performance is 50%, not 30%.  And the difference grows when you take typical overclocks into account.

 

I do agree that we you should always look at how a processor performs in the applications you're going to use most frequently, but I still think CPUMark is an excellent first filter.  For example, anyone looking at that single-threaded performance chart would immediately eliminate any AMD processor because of their terrible single-threaded performance, which IMO is the correct call.

[Faceman]

Looks like you made a math booboo.  The difference between the 4690K and FX8350 in single-threaded performance is 50%, not 30%.  And the difference grows when you take typical overclocks into account.

 

I do agree that we you should always look at how a processor performs in the applications you're going to use most frequently, but I still think CPUMark is an excellent first filter.  For example, anyone looking at that single-threaded performance chart would immediately eliminate any AMD processor because of their terrible single-threaded performance, which IMO is the correct call.

That is exctly what I am saying, 50% difference!  It is CPUMark, which is what you are recommendng, that is only showing a 30% difference according to its Single Thread Chart.  That is why I am saying it is not a good website to use.

 

Please, stop using this website, even as a first impressions benchmark.  It is not accurate.

[Sintezza]

It realy depends on the games you play, and how you play them.

if you are an MMO RPG player then go intel.

 

But if you are going to look to new AAA Games like GTA 5 for example,

There wont be much of a diffrence between a FX8 core and an i5.

Because these games do rely more on the GPU.

 

Still intel i5 / i7 is obviously still the better choice for gaming atm.

Maybe DX12 or Vulkan can make cpu ussage more obsolete, but we dont know till its exaly there.

[flipped_bit]

That is exctly what I am saying, 50% difference!  It is CPUMark, which is what you are recommendng, that is only showing a 30% difference according to its Single Thread Chart.  That is why I am saying it is not a good website to use.

 

Please, stop using this website, even as a first impressions benchmark.  It is not accurate.

 

Again, there is a problem with your math.

 

CPUMark single-threaded performance:

8350: 1500

4690K:  2250

(2250 - 1500) / 1500 = 50%

 

CPUMark is a perfectly valid source to consult.  Your example to prove contrary is in error.

[Faceman]

Again, there is a problem with your math.

 

CPUMark single-threaded performance:

8350: 1500

4690K:  2250

(2250 - 1500) / 1500 = 50%

 

CPUMark is a perfectly valid source to consult.  Your example to prove contrary is in error.

My bad, I am no math expert and definitely did my math wrong.  Still, this is the only aspect of CPUMark that is halfway relevant, and most people don't search for just single thread performance because while important, it is not the end all-be-all to judge performance from.

 

Most people go to that website, look up the aggregate score of the processors and judge performance based on that, which is the wrong way to go about it and many people end up making mistakes because of websites like this, CPUBoss/GPUBoss, etc..

 

Check specific benchmarks for programs/games that you plan to use.