Impressed with FX-8350 and the new article at Anand

[Cadarin]

I don't really see the point in doing all of that work to compare gaming CPUs, when 3 out of 4 games tested are heavily GPU dependent. The one CPU bound game he tested (Civ 5) indicated that all CPUs are not equal.

They should be testing simulations and strategy games if they want to make the CPU sweat. Obviously benching online games is problematic, but there are plenty of SP CPU bound games out there: off the top of my head, SC2 with a ton of units, Arma 2, and one game that nobody ever benchmarks, Mount and Blade: Warband. 500+ unit battles with horses will bring any CPU to its knees.

 

[cbrunny]

Sorry, but what are you impressed by? The only place it is tied with intel is in GPU limited benchmarks. I suppose you could say it is "good enough" and 100.00 cheaper, but it also uses a lot more power.

The price competitive 3570k was not tested, but based on the 2500K I would say it would beat the 8350 in almost every gaming benchmark as well, and the small initial cost savings of the 8350 would be eliminated over time by the increased power usage.

For sure if you are going multiple GPU, the 3770 is the cpu of choice. If you are spending big bucks for a multiple gpu rig, doesnt make sense to try to save a hundred dollars on a cpu that will bottleneck your gpus.

If you have just one GPU, 8350 is the way to go.

 

[crashtech]

@Cadarin, I don't think the point was to "see the CPUs sweat," but to show what is really necessary to run most games today, which seems to be (almost) any ol' CPU + a big honkin' video card.

 

[parvadomus]

I would like to see the same benchmarks with OCed NB, it seems to me that its one of the FX bottlenecks.

 

[sequoia464]

I would like to see the same benchmarks with OCed NB, it seems to me that its one of the FX bottlenecks.

I'm just a straight up noob about this so this is really just a question, not a response.

When I overclocked my Thuban it seemed to be standard wisdom to also overclock the NB, it seems that what I have read on the FX's is that this is not as important, so I simply used the multiplier on an 8320 that I have. Have I misunderstood this?

 

[inf64]

Yeah FX is not gaining that much from NB OC. Sure it's nice to bump it from stock but I think it's a waste of power to push it above 2.5Ghz as returns are not worth it (1-2%).

 

[Piroko]

Hmm how about this conclusion:
On the tested games you won't see a CPU bottleneck unless you spend at least twice its price on your GPU(s).

 

[Phynaz]

I thought we had moved beyond average FPS as a valid performance measurement.

There should have at least been minimums.

Also, at least according to the Steam survey, less than 1% of people game at 1440.

 

[grimpr]

Yeah FX is not gaining that much from NB OC. Sure it's nice to bump it from stock but I think it's a waste of power to push it above 2.5Ghz as returns are not worth it (1-2%).

It really needs 1600+ DDR3 kits though, ddr3 1333 sucks hard on it specially on the desktop in constrast to Intel where there's no point buying anything more than ddr3 1333 due to the superior imc.

 

[crashtech]

I thought we had moved beyond average FPS as a valid performance measurement.

There should have at least been minimums.

Also, at least according to the Steam survey, less than 1% of people game at 1440.

Interesting point about 1440. I just assumed I was a Luddite for still gaming at 1080.

 

[SiliconWars]

I thought we had moved beyond average FPS as a valid performance measurement.

There should have at least been minimums.

I'll hold you to that when Haswell's IGP is benchmarked.

On the article - I wouldn't say I was "impressed" by the 8350 but I can see why AMD feels they can get back into the race.

 

[Vic Vega]

Vic those laser beams are intimidating

Hide yo kids hide yo wife...

 

[Headfoot]

While it's easy to sit here and talk about all the stuff we wish was in there, that's just not possible from a man-hours perspective given the multiplicative nature of every additional parameter. I think Ian did a great job for what it is.

Still, I think perhaps dropping testing things like the either the 3930k or the3960x (just test one x79 6-core) and making up the difference by adding more games would have been more meaningful. I would have enjoyed seeing Shogun Total War 2 in there. It's a classic PC series, and Rome Total War 2 is set to come out soon. Punishing game too in both graphics and CPU.

Besides, I can't say I know anyone who plays "just a little" of any Total War game. No matter how many times you tell yourself "one more turn" it never seems to work out...

 

[ShadowVVL]

Maybe I am missing something but why are the X5000 xeons and the E6700 not measured in some of the tests?

 

[MountainKing]

Why aren't the tests in 1080p?

 

[laputan machine]

Given that these are beefy cards in crossfire\sli, it would be a bit of a waste to run them at a mere 1080p, both for the end user and also given the amount of time it would take with all the setups.

 

[lagokc]

Sorry, but what are you impressed by? The only place it is tied with intel is in GPU limited benchmarks. I suppose you could say it is "good enough" and 100.00 cheaper, but it also uses a lot more power.

i7 3770k is $230 ($250 with tax)
AMD FX 8350 is $201 shipped

So it's really $49 cheaper and uses a lot more power.

http://microcenter.com/product/388575/Core_i7_3770K_35GHz_LGA_1155_Processor

http://www.newegg.com/Product/Produc...iption=fx+8350

 

[laputan machine]

i7 3770k is $230 ($250 with tax)
AMD FX 8350 is $201 shipped

So it's really $49 cheaper and uses a lot more power.

http://microcenter.com/product/388575/Core_i7_3770K_35GHz_LGA_1155_Processor

http://www.newegg.com/Product/Produc...iption=fx+8350

That's an unfair comparison as the 3770k deal is in-store pickup only, and most people do not live close to a microcenter.

Cheapest price is $317 for the 3770k
http://pcpartpicker.com/part/intel-cpu-bx80637i73770k

Cheapest for the 8350 is $190
http://pcpartpicker.com/part/amd-cpu-fd8350frhkbox

 

[frozentundra123456]

That's an unfair comparison as the 3770k deal is in-store pickup only, and most people do not live close to a microcenter.

Cheapest price is $317 for the 3770k
http://pcpartpicker.com/part/intel-cpu-bx80637i73770k

Cheapest for the 8350 is $190
http://pcpartpicker.com/part/amd-cpu-fd8350frhkbox

The obvious comparison price wise is 3570k (220.00 at newegg) vs FX8350 (200.00 at new egg). Considering the entire cost of a system, buying games, internet, and the extra power used this price difference is negligible. Unfortunately they didnt test the 3570k, but I expect the results would be quite similar to 3770k in the vast majority of games.

This is why it is difficult for me to describe the FX as "impressive". For gaming, I see no reason to recommend FX. If you are doing encoding or other productivity benchmarks at which the FX is best at, it is a more attractive alternative.

 

[laputan machine]

The obvious comparison price wise is 3570k (220.00 at newegg) vs FX8350 (200.00 at new egg). Considering the entire cost of a system, buying games, internet, and the extra power used this price difference is negligible. Unfortunately they didnt test the 3570k, but I expect the results would be quite similar to 3770k in the vast majority of games.

This is why it is difficult for me to describe the FX as "impressive". For gaming, I see no reason to recommend FX. If you are doing encoding or other productivity benchmarks at which the FX is best at, it is a more attractive alternative.

I wouldn't call the fx-8350 impressive either. I would say it's good, but there are usually better options. If you need 8 threads going up 30-40 bucks to a E3-1230 v2 might make more sense, and if you're just gaming 3570k would also make more sense. Given how considerably cheaper the fx-6300 and fx-8320 are, I'd never get an fx-8350 over them if I were to buy an fx chip.

 

[guskline]

I am confused as to why GTX 680s or GTX 670s were not used instead of 580s? Good Lord the 780s may be out soon! On the AMD side they used the latest series of high end video cards. Doesn't this lack of testing the Nvidia 600 series cards (should be GTX680) call into question the results? I am not pro Nvidia or Pro AMD. However, when an article is written in April 2013 by a testor who uses 4 7970s vs 2 580s isn't it logical to ask "Hey why no 680s or even 670s?"

 

[mikk]

Game tests in this review were heavily GPU limited especially on 2560x1440. Metro 2003 and Sleeping Dogs are know as heavily GPU bound on max settings. Also the choice for an outdated GTX 580 was bad. Some interesting CPU Benchmarks though. It's shocking that Bulldozer is so bad.

 

[SlowSpyder]

Maybe I am missing something but why are the X5000 xeons and the E6700 not measured in some of the tests?

Not enough PCIE slots for the three and four card tests?

 

[guskline]

http://anandtech.com/show/6934/choosing-a-gaming-cpu-single-multigpu-at-1440p


Really impressed with how the FX-8350 performed in this article. From a cost and performance standpoint I think it would be really hard to not recommend AMD for a decent gaming rig.

UNhooked: If you look at my sigs below, I have BOTH the 3770k and the 8350. I even have a 8320 in my third rig. The Fx chips are nice and solid. The 3770k, in gaming is faster. I have used the same GTX680 in both rigs and I have used the 2 670 sli cards in both rigs. Either way the 3770k is faster. That does NOT mean I don't like the 8350. It's a solid rig. I was able to buy my 3770k from MC for $229.99 plus tax. The cost is close enough that I would recommend the 3770k over the 8350.

 

[UNhooked]

UNhooked: If you look at my sigs below, I have BOTH the 3770k and the 8350. I even have a 8320 in my third rig. The Fx chips are nice and solid. The 3770k, in gaming is faster. I have used the same GTX680 in both rigs and I have used the 2 670 sli cards in both rigs. Either way the 3770k is faster. That does NOT mean I don't like the 8350. It's a solid rig. I was able to buy my 3770k from MC for $229.99 plus tax. The cost is close enough that I would recommend the 3770k over the 8350.

Sure I too am fortunate to have an MC and Frys local to me, so yes at $229 the 3770K does look good. However you get $40off a mobo if you buy the AMD chip .

Looks like I'll be building a new rig with the FX for my bro. He has an aging Amd X4 820 I think.