BF4 CPU usage

Page 7 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Shivansps

Diamond Member
Sep 11, 2013
3,875
1,530
136
Well I have no idea what they did but bf4 is one of those very very rare games that have such an issue. Hopefully they fixed it with this latest patch.

Its probably related to the number of FPU threads it starts on a six core cpu, i dont think BF4 is designed for the FX processors at all.

We have an idea on how many threads it starts depending on the cpu? And how many of them are for graphics, physics, etc?
 

PPB

Golden Member
Jul 5, 2013
1,118
168
106
It has been already disclosed that currently 63xx performs WORSE than 43/41xx, the first one having 3 complete FPU units and the latter 2, as seen in those benches. Repeating it like a mantra wont suddenly make it the truth, so it isnt about the FPU.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
The problem was in Win 7, with Win 8 i had no problem with 6 cores from the beginning.

 

NTMBK

Lifer
Nov 14, 2011
10,322
5,352
136
Good job the FX-9xxx prices have dropped just in time for Battlefield 4.
 

Shivansps

Diamond Member
Sep 11, 2013
3,875
1,530
136
The problem was in Win 7, with Win 8 i had no problem with 6 cores from the beginning.


Yeah, as i said, the problem was probably the FPU threads, its likely that BF4 is starting less FPU intensive threads on a x4 than in a x6 and/or SMT doing crappy things again with the FXs, there is not other reason for the game to run fine on a FX4, FX8 and on a Phenon X6, but no on a FX6.

BTW the article was updated



The botton line of what was saying is why no OpenCL accelerated Havok yet?
 
Last edited:

(sic)Klown12

Senior member
Nov 27, 2010
572
0
76
The botton line of what was saying is why no OpenCL accelerated Havok yet?


Intel. Most likely they stopped the project Havok had with AMD as they want to keep physics on the CPU. The more CPU limited games are, the more reason to buy the latest and greatest. If Larrabee had ever come to fruition on the desktop it might have been brought forth as a competitor to PhysX, but it just didn't work out.
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
145
106
Intel. Most likely they stopped the project Havok had with AMD as they want to keep physics on the CPU. The more CPU limited games are, the more reason to buy the latest and greatest. If Larrabee had ever come to fruition on the desktop it might have been brought forth as a competitor to PhysX, but it just didn't work out.

Utter nonese and BS.

AMD/ATI are simply not able to finish the project. I could find more with different years too if needed.

2006:




2009:



2010:
 

2is

Diamond Member
Apr 8, 2012
4,281
131
106
Intel. Most likely they stopped the project Havok had with AMD as they want to keep physics on the CPU. The more CPU limited games are, the more reason to buy the latest and greatest. If Larrabee had ever come to fruition on the desktop it might have been brought forth as a competitor to PhysX, but it just didn't work out.

Yes, i'm sure Intel is wasting resources on trying to stop OpenCL, (a project that is having enough trouble getting started on it's own, with no help from anyone else) so that the small number of PC gamers who have high end intel processors upgrade.
 

inf64

Diamond Member
Mar 11, 2011
3,864
4,546
136
Is that gain solely from HT? Thats a massive gain.
Actually that gain is without SMT... "Ohne" means without in German, "mit" means "with". So SMT enabled hurts performance in case of intel chips. I wonder what would happen if they disabled each odd core in 8350 and run it on 4 "strong" threads(on 4 modules).
 
Aug 11, 2008
10,451
642
126
Yes, i'm sure Intel is wasting resources on trying to stop OpenCL, (a project that is having enough trouble getting started on it's own, with no help from anyone else) so that the small number of PC gamers who have high end intel processors upgrade.

Agreed. Sure seems to be a lot of unproven accusations and speculation on the forums lately.
 
Aug 11, 2008
10,451
642
126
Actually that gain is without SMT... "Ohne" means without in German, "mit" means "with". So SMT enabled hurts performance in case of intel chips. I wonder what would happen if they disabled each odd core in 8350 and run it on 4 "strong" threads(on 4 modules).

I wonder if that is labelled correctly. In such a core hungry game I cant imagine hyperthreading causing a performance hit, especially since we dont even see it in normal games, and usually i7 is slightly faster than i5.
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
145
106
I wonder if that is labelled correctly. In such a core hungry game I cant imagine hyperthreading causing a performance hit, especially since we dont even see it in normal games, and usually i7 is slightly faster than i5.

Looked at the bugs that have been fixed lately? 50% performance drop in Win7. Utter disaster with 6/12 thread CPUs. They have some way to go still.
 

Blue_Max

Diamond Member
Jul 7, 2011
4,223
153
106
Sounds like I should throw another $100 in the pot and take back my 4670k for a 4770k....
 

inf64

Diamond Member
Mar 11, 2011
3,864
4,546
136
I wonder if that is labelled correctly. In such a core hungry game I cant imagine hyperthreading causing a performance hit, especially since we dont even see it in normal games, and usually i7 is slightly faster than i5.
It is. They actually call it "SMT-bug" in the article.
 

Shivansps

Diamond Member
Sep 11, 2013
3,875
1,530
136
Actually that gain is without SMT... "Ohne" means without in German, "mit" means "with". So SMT enabled hurts performance in case of intel chips. I wonder what would happen if they disabled each odd core in 8350 and run it on 4 "strong" threads(on 4 modules).

It needs a "in deep" analysis of the threads that BF4 starts on each case. It could be very well related to what was happening on FX6xxx compared to a 1100T.
 
Last edited:

inf64

Diamond Member
Mar 11, 2011
3,864
4,546
136
It needs a "in deep" analysis of the threads that BF4 starts on each case. It could be very well related to what was happening on FX6xxx compared to a 1100T.
As you can see from the updated charts, now 6350 is outperforming 1100T. It was just bad scheduling. It's a testament for FlexFP unit as 1100T has 2x more dedicated pipelines but FX's pipelines are multipurpose and shared via SMT which makes them much more efficient in the end. Similar goes for 8350 vs older Phenom II X4s (same number of FP units, 2x the performance basically). I wonder how well will SR core handle BF4 with only 2 modules (but with dedicated decoders and lower latency for fp/sse and L/S instructions).
 

tweakboy

Diamond Member
Jan 3, 2010
9,517
2
81
www.hammiestudios.com
Guys all you need for this game is a 2600k or equivlant. This cpu will play any game fast, depending on your video card. You wont get a bottleneck with a OCed 2600k to 4.2Ghz without touching voltage. Want more you have to up the voltage in small increments until you find the OC to be stable.
 

Stuka87

Diamond Member
Dec 10, 2010
6,240
2,559
136
BTW the article was updated



The botton line of what was saying is why no OpenCL accelerated Havok yet?

Well, at least I know the FPS I am getting is whats to be expected. I am getting basically (Slightly better) than what the 980 scores. I have a 965BE @4Ghz.

Is there any word on steamroller based chips coming out in the next 6 months?
 

escrow4

Diamond Member
Feb 4, 2013
3,339
122
106
Just bought a 4770 non-K + R9 280X. Bring it next gen, BRING IT.

The more multithreaded the better.
 

Dresdenboy

Golden Member
Jul 28, 2003
1,730
554
136
citavia.blog.de
I wonder if that is labelled correctly. In such a core hungry game I cant imagine hyperthreading causing a performance hit, especially since we dont even see it in normal games, and usually i7 is slightly faster than i5.
Earlier in this thread I explained possible slowdown reasons. If there is a heavy thread running on one core and would saturate it on its own, then some other processes (OS, drivers) or thread on that same core would slow it down to 60-70%. If other threads or gfx output depend on its results, this alone could pull fps down. This might be caused by omitting the execution of SMT threads on all cores but one. One graph in the first posting is showing that effect.
 

rancherlee

Senior member
Jul 9, 2000
707
18
81
Well, at least I know the FPS I am getting is whats to be expected. I am getting basically (Slightly better) than what the 980 scores. I have a 965BE @4Ghz.

Is there any word on steamroller based chips coming out in the next 6 months?

Same here, I'm pulling high 20's low 30's with a 3.6ghz PII and a 7950. My 7950 is pulling about 70% usage with the PII crying at 100% with 1080p Ultra. I just downloaded Win 8.1 preview last night and plan on trying it out to see how much improvement there is.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |