BF1 with quad core = no bueno

lightsout

[H]ard|Gawd
Joined
Mar 15, 2014
Messages
1,211
Testing things out with a 4570s (low power I know but still) and a gtx 1070. Struggling with frames in the 30-40s. @ 1440p. Even set it to low doesn't seem to help much.

I was curious if anyone is using an i5-K chip with a decent OC on this game?

I know BF has liked more cores for a while, but I was hoping for better performance.
 
What's your processor speed? How much of the Processor is being used while you game?
 
Oh duh I wrote that in a rush. Turbo speed is 3.2 which I believe all cores are running at.

It's pinned at 100% on all cores playing bf1.

I fired up bf4 and the usage was high but performance was way better. I know it's an older game.

I'm sure a clock boost would help but more threads is probably the only real answer.
 
It will help a tad. I had a 3570k at 4.6Ghz.Cores pegged playing 1440p. Upgraded to a ryzen 1700, and had no issues after that.
 
There's gotta be a setting that lowers CPU usage, either foliage density or shadow quality.
 
It isn’t your CPU holding you back. Look at the Pentium G4560 (dual core) getting over 100FPS at 1440p with a Nvidia 1080 card. Something else is wrong with your setup.

https://www.techpowerup.com/reviews/Intel/Core_i7_8700K/12.html

upload_2018-9-22_23-10-26.png
 
There's gotta be a setting that lowers CPU usage, either foliage density or shadow quality.
Yeah I know, I'll look into it.
It isn’t your CPU holding you back. Look at the Pentium G4560 (dual core) getting over 100FPS at 1440p with a Nvidia 1080 card. Something else is wrong with your setup.

https://www.techpowerup.com/reviews/Intel/Core_i7_8700K/12.html

View attachment 105773
I mean its a clean install, new drivers, very light system right now not many apps running. Not sure what it could be.
I can see the cpu in game, it never leaves 100%. That definitely will hinder performance wouldn't you think.

Plus theres a cpu review on Gamers Nexus and they seem to have quite different results. The TPU result there pretty much shows theres no difference. I'm not sure I agree.
 
Plus theres a cpu review on Gamers Nexus and they seem to have quite different results. The TPU result there pretty much shows theres no difference. I'm not sure I agree.

Link?

Threadpowerups results for Battlefield 1 show that game is NOT core dependant. In that same review I linked it clearly shows which games are.

Civilization 6, Dishonored 2, and Hitmain for instance where the dual cores get only 60% performance of the quads or higher.

upload_2018-9-23_9-20-19.png


upload_2018-9-23_9-22-47.png


upload_2018-9-23_8-33-6.png
 

Attachments

  • upload_2018-9-23_9-21-13.png
    upload_2018-9-23_9-21-13.png
    47.9 KB · Views: 0
Last edited:
Is TPU doing a canned benchmark or actual MP play through?

I’m not sure, but it’s immaterial. The game wouldn’t drop from 100fps to 40fps because its single vs. multiplayer. I’ve never seen any game even come close to that. Not even if I’m running as the local host at offline LAN Party. He’s not running a dual core. He’s running a quad and with that particular game even low end pentium dual cores do just fine —as evidenced by a very compelling set of charts by TPU.

I don’t know what the problem is, but it isn’t a quad core problem — which is the title and premise of this thread.
 
Last edited:
Yeah I know, I'll look into it.
I mean its a clean install, new drivers, very light system right now not many apps running. Not sure what it could be.
I can see the cpu in game, it never leaves 100%. That definitely will hinder performance wouldn't you think.

Plus theres a cpu review on Gamers Nexus and they seem to have quite different results. The TPU result there pretty much shows theres no difference. I'm not sure I agree.
Open up task manager and see which processes are gobbling up your CPU resources. Maybe you have a crypto mining browser instance open, malware, or a rogue anti-virus driver running wild. Did you install all the OEM chipset and motherboard drivers? (Don’t rely on Windows built in HCL drivers)a

Turn on Gams Mode if you have Windows 10 to help minimize background processes.

How is heat on video card and CPU? Are either clock throttling. Download Intel Tuning utility and open it up while running the game to monitor heat and clock speed in a 10 minute window.

https://downloadcenter.intel.com/download/24075/Intel-Extreme-Tuning-Utility-Intel-XTU-


Download MSI Afterburner for GPU monitoring to see if clock speed stays consistent.

https://www.msi.com/page/afterburner


You’ll have to do a little troubleshooting.
 
Last edited:
I’m not sure, but it’s immaterial. The game wouldn’t drop from 100fps to 40fps because its single vs. multiplayer. I’ve never seen any game even come close to that. Not even if I’m running as the local host at offline LAN Party. He’s not running a dual core. He’s running a quad and with that particular game even low end pentium dual cores do just fine —as evidenced by a very compelling set of charts by TPU.

I don’t know what the problem is, but it isn’t a quad core problem — which is the title and premise of this thread.


Not sure I agree with that.

Lots of players running around on a server is greater cpu load.

Would be great if there was a canned benchmark that could be used to test with for the op. As he said, other BF games play fine.
 
Not sure I agree with that.

Lots of players running around on a server is greater cpu load.

Would be great if there was a canned benchmark that could be used to test with for the op. As he said, other BF games play fine.
Evidence would be good.
I provided benchmark data backing my suggestion.

Where is evidence a quad core isn't sufficient to run BF1 at greater than 30-40FPS (with a 1070 equivalent card) as the OP has? My suggestion is his low performance situation is atypical for the hardware he has - and the low performance numbers is not because he has a quad core.
 
Evidence would be good.
I provided benchmark data backing my suggestion.

Where is evidence a quad core isn't sufficient to run BF1 at greater than 30-40FPS (with a 1070 equivalent card) as the OP has? My suggestion is his low performance situation is atypical for the hardware he has - and the low performance numbers is not because he has a quad core.

Having a lower clocked one certainly does not help his case.

Ofc depending on how recent his windows install is there might be some processes running in the background that hamper performance that should go away once they finish, like windows update and that annoying indexing function (forgot it's name).

according to these older benchmarks the CPU did look decent enough back when it was recent https://www.anandtech.com/show/8774...-core-i3-4130t-i5-4570s-and-i7-4790s-tested/5
 
Some suggestions for you OP in this thread that might fix the issue:

Thanks man, seems I got a tiny bit of breathing room but didn't seem to help much. I am going to throw a 5820k at it, that better be enough.
 
Thanks man, seems I got a tiny bit of breathing room but didn't seem to help much. I am going to throw a 5820k at it, that better be enough.

Did you try switching back to DX11 instead of DX12? I read that's a big fix for some people.

Some of the topics I just read on it say people with ryzen 8 core or intel 6 core systems are having the same trouble you are - I don't think throwing more cores at it (throwing money at it) is a good solution until you diagnose the problem you currently have.

Here's another thread with some things to try:
https://forums.battlefield.com/en-u...d-my-performance-issues-heres-how-i-did-it/p1

here's a guide on the settings and their individual relevance to FPS hits.
https://forums.battlefield.com/en-u...battlefield-1-graphics-performance-guide-2018
 
Last edited:
Jesus man, they tested on an I7 5930K.

That is 6c 12t. Not 4c 4t.

Come on now, you must have the deductive reasoning to see that is not exactly proof that a 4 thread i5 can handle BF1 multiplayer just fine.
 
My i5 4690k @ 4.2ghz with a 1070 played BF1 perfectly fine in Single/multi. Same generation just slightly higher clocked. Even at stock (3.9ghz boost) I wasn't getting mad stutters/fps drops. I was running on high or higher.. It's been a while since I played but I never had any issues with BF1 at 1440p60hz with my sig setup.
 
Did you try switching back to DX11 instead of DX12? I read that's a big fix for some people.

Some of the topics I just read on it say people with ryzen 8 core or intel 6 core systems are having the same trouble you are - I don't think throwing more cores at it (throwing money at it) is a good solution until you diagnose the problem you currently have.

Here's another thread with some things to try:
https://forums.battlefield.com/en-u...d-my-performance-issues-heres-how-i-did-it/p1

here's a guide on the settings and their individual relevance to FPS hits.
https://forums.battlefield.com/en-u...battlefield-1-graphics-performance-guide-2018
Thanks I have tried DX11 and 12. I am not throwing money at, just have my rig temporarily with a 4570s. I already own the 5820k.


Its definitely a cpu issue as the gpu will be at 40ish%, cpus will all be pinned and in a lot of battle (64 player conquest is all the games I can find) I'm in the 30's and 40's. Will see in a few days when I get my other rig setup.

I hardly played the game when I bought it last year but don't remember this problem, (with the 5820k)
 
Try running borderless mode instead of fullscreen, try power setting to recommended instead of high performance.

Also can you post the rest of the specs of your system? Is the memory running dual-channel mode and how much memory do you have?

Also you could try a different version of Windows just to test it out on another hard drive or something.
 
Try running borderless mode instead of fullscreen, try power setting to recommended instead of high performance.

Also can you post the rest of the specs of your system? Is the memory running dual-channel mode and how much memory do you have?

Also you could try a different version of Windows just to test it out on another hard drive or something.
I could try borderless. The system has 8gb of 1600, running dual channel. A single SSD, Asrock Q87M Pro, not sure what else would matter. I can't try another windows but this is a clean install from a week ago. Intentionally meant to be a light system. I din't mess with services and all that but just not installing a ton of stuff.
 
BF1 plays fine on my i7-4770 locked to 3.9 GHz. As far as I know the game doesn't even scale well past 6 cores, so throwing more cores at it isn't going to accomplish anything.
 
BF1 plays fine on my i7-4770 locked to 3.9 GHz. As far as I know the game doesn't even scale well past 6 cores, so throwing more cores at it isn't going to accomplish anything.
Well hyper threading has helped since BF3 so I'm not sure thats true. i7's have done better than i5's, especially in 64 man conquest.
 
For anyone that reads this, swapped out boards, took out the 4570s and installed other board with 5820k. I am now able to run on ultra and its chiilin in the 80-90 fps range. Sooo much better.

I will say that this chip has a nice clock jump as well (4.3ghz) But still, moar cores!!
 
Well hyper threading has helped since BF3 so I'm not sure thats true. i7's have done better than i5's, especially in 64 man conquest.

I can attest to that. Really wish'd I'd picked up a 2600k instead of the 2500k I'd used for years, especially for Battlefield games. That seemed like the biggest jump, going from a 4.5GHz 2500k to a stock 6700k.

[later grabbed the 8700k because I needed to build another system; the 6700k was certainly still suitable for everything I do]
 
I can attest to that. Really wish'd I'd picked up a 2600k instead of the 2500k I'd used for years, especially for Battlefield games. That seemed like the biggest jump, going from a 4.5GHz 2500k to a stock 6700k.

[later grabbed the 8700k because I needed to build another system; the 6700k was certainly still suitable for everything I do]
Yup same, I grabbed a 2500k on launch day, and saw others with i7's getting better results.
 
Yep, another mistake maker here lol. Went from an i7 930 to a 2500k. I still gained performance overall, but I should've just gone with the 2600k and stuck with that all the way til now. Instead of incremental upgrades XD
 
Yep, another mistake maker here lol. Went from an i7 930 to a 2500k. I still gained performance overall, but I should've just gone with the 2600k and stuck with that all the way til now. Instead of incremental upgrades XD

Well, it wasn't a mistake on release- what we didn't know then was how long the performance would last. I expected to upgrade a few generations later and instead waited four (or five?).

Of course, not doing that one again ;).
 
Well, it wasn't a mistake on release- what we didn't know then was how long the performance would last. I expected to upgrade a few generations later and instead waited four (or five?).

Of course, not doing that one again ;).

Yup. That said my little brother is still running my Z77-based system, uprgaded RAM to 16gb over 8 over time, and swapped the 2600k out (had a 2500k, swapped for 2600k because a friend gave me his old 2600k, and gave my 2500k to someone else for a used-parts build (he paid a small amount for the build) (sold it (2600k) on ebay) for a free 2600 non k I found in an old dell XPS machine on the side of the road during council cleanups hahaha. People dont know what they're doing! #profit. Brother doesn't OC and lower temps/noise > OC'ing on a machine I don't have to maintain XD
 
It isn’t your CPU holding you back. Look at the Pentium G4560 (dual core) getting over 100FPS at 1440p with a Nvidia 1080 card. Something else is wrong with your setup.

https://www.techpowerup.com/reviews/Intel/Core_i7_8700K/12.html

View attachment 105773

No. Techpowerup does all it's game tests from single player. it's not as demanding as multiplayer. Thus, it runs fine on a dual-core.

Most other sites also use that sort of cheat for games that have single-player modes.

There are only a few that bother to do a multiplayer FRAPS test, n d most of them are pretty old (so the game may be more demanding today, with newer maps)

https://www.gamersnexus.net/game-bench/2673-battlefield-1-cpu-benchmark-dx11-vs-dx12-i5-i7-fx/page-2
 
The [H] does multiplayer tests if you want to dig for something relevant; I consider their benchmarks to be the standard for that purpose.

And yeah, if we're talking single-player vs. multi-player BF games, the difference can be night and day, and certainly the difference between hyperthreading with and without.
 
No. Techpowerup does all it's game tests from single player. it's not as demanding as multiplayer. Thus, it runs fine on a dual-core.

Most other sites also use that sort of cheat for games that have single-player modes.

There are only a few that bother to do a multiplayer FRAPS test, n d most of them are pretty old (so the game may be more demanding today, with newer maps)

https://www.gamersnexus.net/game-bench/2673-battlefield-1-cpu-benchmark-dx11-vs-dx12-i5-i7-fx/page-2
Oh man I always looked at TPU, good to know thanks.
 
I play BF1 everyday.

It is definitely CPU limited. Anybody who plays it regularly can tell you this.

I have a GTX 1080 but some maps bring my i5 6600k @4.3ghz to 60fps. So I've actually just decided to run the game at a really high resolution so I can keep a locked 60fps.

This won't work in BF5 though, since that seems to be even more CPU demanding. Going to have to upgrade I think
 
I can now speak from personally experience: going from a 4 core to a 4 core / 8 threaded CPU can make a HUGE difference depending on the game....

(same architecture, even the same exact mhz speed)
 
if you CPU is peaked at 100% you have aa clear sign of CPU bottleneck.
Does-tn matter if its the game or other software or wrong drivers etc. the main culprit looks to be running out of ressource and it should be investigated
So now you need to figure out why.

Is it only BF1 process that take resources from your CPU ?
 
Back
Top