• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

BF4 Retail CPU scaling measured

Wow, interesting. Would have liked an overclocked 3930k/3970k result to see where the plateau is in terms of fps.

Very nice to see the FX chips doing well. Would be nice to see an overclocking comparison of FX 83x0 at 5GHz+ and Haswell I7 at 4.4GHz to see how far behind the FX chips are (if much at all).
 
The 1100T is somewhat disappointing over the X4, 50% more potential performance, but there's not much gain in the benchmark.
The FX6300 is leagues ahead of it.

There's definitely some instruction set optimisation going on.

Techspot will hopefully have some overclocked results.
 
Seen a few post about people experiencing 100% cpu usage like most quad core people had in the beta but by the end of it with the couple of patches it was ok.

Here's hoping dice can work a little magic and reduce the cpu usage abit, dont fancy upgrading quite yet
 
Seen a few post about people experiencing 100% cpu usage like most quad core people had in the beta but by the end of it with the couple of patches it was ok.

Here's hoping dice can work a little magic and reduce the cpu usage abit, dont fancy upgrading quite yet

The last time around (BF3) we all had to buy new GPUs because of the stupid amount of vram it used.

This time? CPU lol.
 
The 1100T is somewhat disappointing over the X4, 50% more potential performance, but there's not much gain in the benchmark.
The FX6300 is leagues ahead of it.

There's definitely some instruction set optimisation going on.

Techspot will hopefully have some overclocked results.

This is what happens when stuff is compiled and optimised for AMD's instruction set, people absolutely and completely ignore that Intel has such a stranglehold on the industry and are so favoured by MS/other software makers that most is compiled with Intel compilers that does as little as possible to optimise for AMD.

Intel and AMD have dozens, hundreds of ways to short cut results, combine calculations and optimise throughput and effectively huge amounts of software just don't make full use of an AMD core and the performance available.

It shouldn't be any surprise at all, in the slightest that an AMD core in a game that worked hard with AMD to optimise performance performs significantly better than in other software which basically optimises for Intel and not AMD.

For those who think Phenom 2 was a better core, it categorically wasn't, it was just older and better supported, Bulldozer and even more so piledriver is without question a move forward, AMD will always struggle with 1-2 year lag in improved CPU vs being optimised for :(

AMD can't really win, we've had 15 years of Intel paying computer manufacturers not to use their stuff and paying software manufacturers not to optimise for them, considering that, their lack of money and process node disadvantage AMD are way way way ahead of where most companies would be. Considering Intel's advantages and well, illegal activity their "lead" on AMD, also considering R&D spending, is pretty pathetic.
 
Wouldn't really change much in the hierarchy (Except being better than the FX6300, but well, it's a very moot victory)

For price/performance, the FX8320 is going to be unbeatable in BF4, overclocking it will yield better results, probably surpassing the 4770K at stock for half the price.

AMD optimized instruction sets perhaps?

It'll be interesting to see after mantle, if it's for reducing the CPU bottleneck, surely all the results will get closer together (But, if this is the current state of play, if AMD ensured that you needed a full AMD system for mantle, they'd be pretty much nailing Intel left right and centre in BF4)

My anticipation of Steamroller = Over 9000!

It'll be interesting to see other results too (As I'm not a fan of 30 second runs, I don't play games in 30 seconds stints), especially multiplayer.

wait what? and AMD gpu isnt enough i need an AMD CPU and GPU to use mantle?
 
This is what happens when stuff is compiled and optimised for AMD's instruction set, people absolutely and completely ignore that Intel has such a stranglehold on the industry and are so favoured by MS/other software makers that most is compiled with Intel compilers that does as little as possible to optimise for AMD.

Intel and AMD have dozens, hundreds of ways to short cut results, combine calculations and optimise throughput and effectively huge amounts of software just don't make full use of an AMD core and the performance available.

It shouldn't be any surprise at all, in the slightest that an AMD core in a game that worked hard with AMD to optimise performance performs significantly better than in other software which basically optimises for Intel and not AMD.

For those who think Phenom 2 was a better core, it categorically wasn't, it was just older and better supported, Bulldozer and even more so piledriver is without question a move forward, AMD will always struggle with 1-2 year lag in improved CPU vs being optimised for :(

AMD can't really win, we've had 15 years of Intel paying computer manufacturers not to use their stuff and paying software manufacturers not to optimise for them, considering that, their lack of money and process node disadvantage AMD are way way way ahead of where most companies would be. Considering Intel's advantages and well, illegal activity their "lead" on AMD, also considering R&D spending, is pretty pathetic.

Says a bit for Intel then that this game was optimised for AMD CPUs, Intel CPUs still top that chart!
Poor AMD, Nvidia and Intel use every dirty trick in the book (which was probably optimised to work better on Intel and Nvidia hardware due to backhanders) to try to put the angelic ones out of business but they failed!

wait what? and AMD gpu isnt enough i need an AMD CPU and GPU to use mantle?

I don't think it's been confirmed and Martini only said if...
Would make sense and would probably help shift more AMD CPUs!
Although AMD probably wouldn't do something like that.
 

Maybe I'm being silly, but why is the AVG bar of the "Phenom II X4 955" (71) shorter than the "AMD FX-4300" (67)?
And...
  • "Phenom II X6 1100T"/"Intel Core i5 760" (76) shorter than the "AMD FX-6100" (74)?
  • "Intel Core i5 2500K" (86) shorter than "Intel Core i7-930" (85)?
  • "Intel Core i7 2600K" (98) shorter than "AMD FX-8350" (97)?
I realise they're ordered by "min" but that doesn't explain the length of the "avg" bars.
 
Any Intel quad at ~4.5ghz is still king then, considering how well they do at stock against 4.7-5ghz AMD FX's.

I don't know like.
The FX8350 and i5 4670K are about at parity at stock.
The 4670K is an inconsistent clocker, current FX83 silicon looks pretty good.

My 4.75GHZ i5 will probably slightly edge an FX8350 at 5GHZ (I'd be gaining 1.25GHZ, which is higher percentage than the 1GHZ an FX8350 would gain, the FX 8350 at 5GHZ is a 25% overclock, my 4670K is ~36%)

But on average overclocked? They're going to be pretty much at each others knackers, probably changing position depending on the scene.

i5 2500k is going to be behind the FX8350 when both overclocked by the looks of it (But you'd really have to see, the i5 2500k has a possible 1.5GHZ left in the tank which is huge, but still, I'm going to go with the FX83)

EDIT : Of course, we've also got a very limited scene, I'm awaiting techspots figures, but the initial figures look positive for AMD.
 
Last edited:
Back
Top Bottom