AMD Takes The CPU Performance Crown For The First Time in Almost a Decade.

[Lord Pantaloons]

I do believe Nvidia already has their "nest egg" planned with maxwell. Maxwell is supposed to have an integrated arm cpu on the card, this may eventually develop into the full power cpu that games run on. Eventually, there may be no need for a new cpu and mobo, but rather a new gpu(gaming processing unit) that you simply plug into your pci-e slot. Other programs like fraps and firefox run on the x86 cpu and ram on the motherboard, but games will be entirely run on the gpu, putting no load on the normal x86 cpu. This could mean that gamers don't need to spend more money every generation on a new cpu and motherboard.

 

Maxwell could be the beginning of the future, but right now we have nothing but speculation and hope.

That sounds pretty cool, something about it though doesn't sit right with me though.

[Mooshi]

Good bye

 

[KneesWeak]

AMD's best solution right now is the APU which isn't listed in that chart anywhere. None of their high level cards can compare to intel's offering.

ya that 7990 really doesnt compare to the 4930k, AMD still has dramatically worse single threaded performance but for modern gaming, if you can oc, a 6300/6350/8320/8350 is a good choice for a budget

[totjup5]

I'm glad I went AMD! I have always liked AMD more, nothing against intel, i've just had better experiences with AMD.

[leipero]

Ok, obviously, i love AMD and i would not buy Intel even if someone hold a shotgun on my head lol, but, why there is no 3770k in this benchmark, for 4770k i might understand, but again, it is on market, therefore, it should be tested.

 

I understand note about IB and Haswell from the OP about OC, but let's be fair here, some people (me included) would not overclock CPU at all, so that point is not worth, also, overclocking brings instability etc.

 

But, well optimized games/programs will surely win on faster CPU's at stock clock, therefore, on AMD current CPU's.

 

Also, when testing CPU's, it should be done with most high end SINGLE GPU's from both nVidia and AMD, using just one gives no credible result in my book.

[Lord Pantaloons]

Ok, obviously, i love AMD and i would not buy Intel even if someone hold a shotgun on my head lol, but, why there is no 3770k in this benchmark, for 4770k i might understand, but again, it is on market, therefore, it should be tested.

 

I understand note about IB and Haswell from the OP about OC, but let's be fair here, some people (me included) would not overclock CPU at all, so that point is not worth, also, overclocking brings instability etc.

 

But, well optimized games/programs will surely win on faster CPU's at stock clock, therefore, on AMD current CPU's.

 

Also, when testing CPU's, it should be done with most high end SINGLE GPU's from both nVidia and AMD, using just one gives no credible result in my book.

Ivy and Haswell were excluded because of some silly OC related reason but they didn't OC in the test, Ivy is about 10% faster than sandy at stock clocks and Haswell is a bit faster than Ivy, that's the real reason they weren't included because if they were the results would be different.

[Spannerhands]

I'd just get a 4930K instead of a 9590. I'd save lots of money and I have room for overclocking. That 9590 isn't going to run cool.

 

Well you can cool the FX9590 silently... Just needs an out of the box solution.... 

[leipero]

Ivy and Haswell were excluded because of some silly OC related reason but they didn't OC in the test, Ivy is about 10% faster than sandy at stock clocks and Haswell is a bit faster than Ivy, that's the real reason they weren't included because if they were the results would be different.

Well, that's my point, let's not speculate, it may (probably) be faster than SB CPU's on stock freq. , but i would love to see real test, not speculation, also, with both green and read single GPU's tested on all platforms. Only then, i can consider ANY test credible (when using two top high-end GPU's from both sides).

[tacticalpanic]

my problem with amd processers on the desktop right now is the crap selection of motherboards, when AMD's FX cpus can compete with intels top I5/I7 while running on a ITX motherboard then we'll talk, for a gaming machine i dont want a full sized workstation tower , my little itx gaming rig destroys games and runs quiet and cool.

[Spannerhands]

MD Takes The CPU Performance Crown For The First Time in Almost a Decade.

 

+

 

my problem with amd processers on the desktop right now is the crap selection of motherboards, when AMD's FX cpus can compete with intels top I5/I7 while running on a ITX motherboard then we'll talk, for a gaming machine i dont want a full sized workstation tower , my little itx gaming rig destroys games and runs quiet and cool.

 

= LOL

[tacticalpanic]

MD Takes The CPU Performance Crown For The First Time in Almost a Decade.

 

+

 

 

= LOL

whats so funny about what I said?

[Spannerhands]

whats so funny about what I said?

 

Because people who put words  like AMD, INTEL, CRAP, GAMING & DESTROYS one sentence make me laugh... 

[ALXAndy]

ITX is for fairies. 

[Godlygamer23]

i thought it was 5 ghz

No. The 9370 is a 4.4GHz CPU with a 4.7GHz Turbo boost on two cores. The 9590 is a 4.7GHz chip and bumps up to 5GHz when only two cores are used.

[solomongrundy]

220watt tdp... yeah no thanks. if i want to buy a heater, i'll go to Home Depot.

[acepro71]

No. The 9370 is a 4.4GHz CPU with a 4.7GHz Turbo boost on two cores. The 9590 is a 4.7GHz chip and bumps up to 5GHz when only two cores are used.

thank u for correcting 

[Glenwing]

It's funny, you know?  When Intel releases a CPU with marginally better performance and less power consumption, everyone goes "Intel! We don't care about power consumption, just make it more powerful!"  Then AMD goes and releases a powerful CPU with no cares about TDP and everyone's like "What's with that TDP?  We don't want anything like that!"

[Godlygamer23]

thank u for correcting 

You're welcome.

[LAwLz]

It's funny, you know?  When Intel releases a CPU with marginally better performance and less power consumption, everyone goes "Intel! We don't care about power consumption, just make it more powerful!"  Then AMD goes and releases a powerful CPU with no cares about TDP and everyone's like "What's with that TDP?  We don't want anything like that!"

Personally I think it's funnier when people compare AMD's newest chips against Intel's chips from 2 generations ago, and goes "you just need a 60 dollar cooler and a huge overclock, and then the AMD chip beats the Intel one!".

 

Rule of thumb: If your comparison is one overclocked part vs one part running at stock, then it's not a fair comparison.

Also, there is a huge big difference between dropping the TDP by ~20W, and increasing the TDP by ~100W. The former is not really a big deal for desktops, the latter can be a big deal since the amount of cooling you need goes up dramatically.

 

Also, when testing CPU's, it should be done with most high end SINGLE GPU's from both nVidia and AMD, using just one gives no credible result in my book.

Why not? When testing a CPU you want really really powerful GPU, and use very low settings in the game, in order to eliminate any potential GPU bottleneck.

 

 

Anyway shame on you OP (or whoever you got those pictures from). You never posted a source which got me suspicious so I looked up that Russian site (doesn't seem very trustworthy if you ask me) and as it turns out, you have been cherry picking benchmarks. You deliberately only showed the benchmarks that agreed with your predefined conclusion.

 

If you look at the other tests the site has done, Intel wins a ton of them. I count an Intel win as the i5 beating the 8350 by at least a decent margin (like 10-20% or more). I count an AMD win as their chips beating similarly priced Intel chips by a decent margin (like 10-20%). I count a tie as when most CPUs are pretty much the same (differs by less than ~10%).

 

Games Intel wins in (22):

Arma 3

The Bureau XCOM Declassified

Saints Row IV

Splinter Cell Blacklist

PayDay 2

Dark

Crysis 3 The Lost Island

Call of Juarez Gunslinger

Metro Last Light

Tomb Raider

Red Orchestra 2 Rising Storm

Remember Me

FarCry 3 Blood Dragon

Dead Island Riptide

Resident Evil 6

Battlefield 3: End Game

Brütal Legend

Aliens Colonial Marines

Dead Space 3

DmC Devil May Cry

Hitman Absolution

Assassin's Creed III

 

 

Games AMD wins in (2):

Battlefield 4 Beta (might be considered a tie but the AMD chips are much cheaper than the ones they are trading blows with)

Farcry 3

 

 

Ties (11):

Shadow Warrior

Alien Rage

Lost Planet 3

Amnesia AMfP

Divinity Dragon Commander

Castlevania Lords of Shadow Demo

Mortal Kombat

Deadpool

Bioshock Infinite

Sniper - Ghost Warrior 2

Call of Duty Black Ops II

 

 

Those were all the benchmarks I felt like looking through on that Russian website. For anyone interested, I looked through and posted all benchmarks from page 1 through page 4 (stopped at the beginning of page 5 because I got bored).

Here is a link for anyone wanting to continue: Page 5 of their benchmarks.

 

So basically someone went though this list of 35 games and decided to only post the two AMD wins in, and ignore the 22 Intel wins in. That's something a fanboy would do, and you harm other people by spreading misinformation. The fact that 12 people thumbed you up without doing any research themselves sickens me. Do people just thumb up whatever they agree with, no matter if it's true or not?

[Lord Pantaloons]

Personally I think it's funnier when people compare AMD's newest chips against Intel's chips from 2 generations ago, and goes "you just need a 60 dollar cooler and a huge overclock, and then the AMD chip beats the Intel one!".

 

Rule of thumb: If your comparison is one overclocked part vs one part running at stock, then it's not a fair comparison.

Also, there is a huge big difference between dropping the TDP by ~20W, and increasing the TDP by ~100W. The former is not really a big deal for desktops, the latter can be a big deal since the amount of cooling you need goes up dramatically.

 

Why not? When testing a CPU you want really really powerful GPU, and use very low settings in the game, in order to eliminate any potential GPU bottleneck.

 

 

Anyway shame on you OP (or whoever you got those pictures from). You never posted a source which got me suspicious so I looked up that Russian site (doesn't seem very trustworthy if you ask me) and as it turns out, you have been cherry picking benchmarks. You deliberately only showed the benchmarks that agreed with your predefined conclusion.

 

If you look at the other tests the site has done, Intel wins a ton of them. I count an Intel win as the i5 beating the 8350 by at least a decent margin (like 10-20% or more). I count an AMD win as their chips beating similarly priced Intel chips by a decent margin (like 10-20%). I count a tie as when most CPUs are pretty much the same (differs by less than ~10%).

 

Games Intel wins in (22):

Arma 3

The Bureau XCOM Declassified

Saints Row IV

Splinter Cell Blacklist

PayDay 2

Dark

Crysis 3 The Lost Island

Call of Juarez Gunslinger

Metro Last Light

Tomb Raider

Red Orchestra 2 Rising Storm

Remember Me

FarCry 3 Blood Dragon

Dead Island Riptide

Resident Evil 6

Battlefield 3: End Game

Brütal Legend

Aliens Colonial Marines

Dead Space 3

DmC Devil May Cry

Hitman Absolution

Assassin's Creed III

 

 

Games AMD wins in (2):

Battlefield 4 Beta (might be considered a tie but the AMD chips are much cheaper than the ones they are trading blows with)

Farcry 3

 

 

Ties (11):

Shadow Warrior

Alien Rage

Lost Planet 3

Amnesia AMfP

Divinity Dragon Commander

Castlevania Lords of Shadow Demo

Mortal Kombat

Deadpool

Bioshock Infinite

Sniper - Ghost Warrior 2

Call of Duty Black Ops II

 

 

Those were all the benchmarks I felt like looking through on that Russian website. For anyone interested, I looked through and posted all benchmarks from page 1 through page 4 (stopped at the beginning of page 5 because I got bored).

Here is a link for anyone wanting to continue: Page 5 of their benchmarks.

 

So basically someone went though this list of 35 games and decided to only post the two AMD wins in, and ignore the 22 Intel wins in. That's something a fanboy would do, and you harm other people by spreading misinformation. The fact that 12 people thumbed you up without doing any research themselves sickens me. Do people just thumb up whatever they agree with, no matter if it's true or not?

*I agree with the things said in this post even though I didn't read it, thumbs up!

[leipero]

Why not? When testing a CPU you want really really powerful GPU, and use very low settings in the game, in order to eliminate any potential GPU bottleneck.

I don't agree with that methodology. Maxing out games makes more stress on CPU, memory subsystem, and all components in general. Also, you missed the point of OP, it shows that when coded properly, games/programs do run same or better on AMD platform, and with new systems (maybe even with upgraded DX) it will be potentially easy for anyone to code and use CPU's to it's potential.

 

Also, list you showed is non relevant, since it is present-past, and even so, those games are playable on all platforms, no big deal anyway.

[Giggitygoebbels]

Why compare to last generation processors?That is just not fair IMO. Haswell / IVY-E should triumph AMD easily.

[Glenwing]

Why compare to last generation processors?That is just not fair IMO. Haswell / IVY-E should triumph AMD easily.

 

There's very little difference in performance between Sandy Bridge, Ivy Bridge, and Haswell, or Sandy Bridge-E and Ivy Bridge-E.

[Glenwing]

I don't agree with that methodology. Maxing out games makes more stress on CPU, memory subsystem, and all components in general. Also, you missed the point of OP, it shows that when coded properly, games/programs do run same or better on AMD platform, and with new systems (maybe even with upgraded DX) it will be potentially easy for anyone to code and use CPU's to it's potential.

 

Also, list you showed is non relevant, since it is present-past, and even so, those games are playable on all platforms, no big deal anyway.

 

CPU controls things like physics and AI.  Turning down the detail settings and the resolution affects GPU stress only.  CPU load remains the same even when settings are turned down, if you know which ones to turn off and which ones not to touch.

[Lord Pantaloons]

There's very little difference in performance between Sandy Bridge, Ivy Bridge, and Haswell, or Sandy Bridge-E and Ivy Bridge-E.

But there IS a difference between them and that difference is that they are faster, considering how little the sandy chips were beat by the result would have been completely opposite.