GeForce 387.92 WHQL Game ready Download & Discussion

BF1 is a game which is made on Frostbite engine. It is not "poorly optimized" in any sense. The only part of it which is badly written is the DX12 renderer.

 

Do you even play BF1?

I ask because it's rare to hear someone who plays the game make the claim that it's well optimized. For starters, there's the link above I just posted where EA admits they only optimize for high end hardware. This is an admission that they do not optimize for anything less than high end hardware. <-- that's poor optimization.

I also take issue with your claim that Frostbite is, "the most optimized piece of software on the planet". You have omitted the caveat, "ON THE CONSOLE". On the pc, Frostbite has one huge glaring flaw in that it doesn't perform adequate sanity checks, which is expressly how people are able to use damage modifier hacks. This has been an issue since BF3 which persisted through BF4, through SW: Battlefront, and through BF1. If the engine were well optimized for the pc this flaw would have been patched out long ago.

Also linked previously is a thread I made last November, shortly after the game's release. To date, it's the second largest thread related to performance; specifically, dealing with the 100% cpu issue. It has over 200k views and is over 30 pages long. Still, almost a year later, I help out guys at least 3 times a week getting their game to run better because requests to do so flood my inbox. <--- because the optimization is so poor.

That's just one thread of hundreds on the game's poor optimization. I also help out guys on reddit several times a week with the same issue.

My theory is that dice only optimized the game for the 6700k, and even then, not well. I have no proof since they have admitted nothing, but even 6700k users have reported 100% cpu usage, which you would know if you followed the game. This is why i5's all have 100% usage. I don't think they even bothered optimizing for the 6600k, which might explain why when you disable hyper threading, even on a 6 core Intel, you see 100% usage across all 6 cores: http://www.dsogaming.com/pc-performance-analyses/battlefield-1-pc-performance-analysis/

Go ahead, give it a shot. Disable hyper threading, play some and tell me how well optimized the game is?

Again, I'm not sure how anyone can read what dice admits, and then come to conclusion that what they said was really a lie and that indeed the game is well optimized unless they just don't play.

 

I played enough of BF1 and pretty much any FB3 based games released during the last couple of years to say with absolute certainty that whoever claims that modern FB3 games are "poorly optimized" doesn't know a thing about game engine optimization.

This claim is incorrect as well. BF1 runs on Xbox One which has a GPU equivalent of Radeon HD 7770. This isn't high end hardware.

Again, such examples only confirm that you don't know what you're talking about. This isn't a case of bad optimization.

Nope. Because people want to run modern AAA game on a 6 year old h/w with all bells and whistles turned on.

FB3's base node graph architecture doesn't need any specific optimization for any CPU. If it would be optimized "only for 6700K" it wouldn't be able to run at more than 5 fps on consoles and it is running at 60.

I don't need to because the claim in your link is another bogus:

 

At this point what I'm fighting isn't fact; rather, your ego. That's a losing battle for anyone, because your ego is... well... strong.

As an example of this, you have just taken a link to EA's admission to its investors and decided that it's incorrect. I can't even respond to nonsense like that. It's like when Trump says "fake news".

I ceded the point that the game runs ok on the console and made it a point to not that this is not the case on the pc, and you respond by pointing out it runs in the console. /facepalm.

The engine doesn't perform sufficient sanity checks on the pc. Period. https://www.reddit.com/r/StarWarsBa...j/hacking_exposed_part_2_the_damage_mods_the/
https://www.reddit.com/r/StarWarsBa...ng_exposed_part_29_player_tracking_the_bluff/
https://www.reddit.com/r/StarWarsBa...king_exposed_3_finally_how_to_evade_the_hack/

First you admit that dx12 is broken and then you link dx12 benchmarks... as if that has anything to do with what I'm talking about. Again, I just don't know how to respond, because normally your posts are logical, except when ego gets in the way.

You have provided nothing to support your theory about the game and have decided the links I posted were incorrect... because ego. Go hit the ea forums and let everyone know the game is well optimized. I'm sure there are a handful of bootlickers who will agree with you. The overwhelming majority of people who actually play the game will not.

Lastly, that last comment is priceless... priceless on a level on par with Galileo. I asked you to see for yourself by disabling hyper threading, you respond that you don't need to.. lol. No... no need to look through the telescope, everyone knows the universe is geocentric.

 

LOL. I hope you know TechSpot benchmarks the singleplayer portion, which as with all Battlefield games since BF3 has vastly different system demands (esp. CPU/RAM) than 64-player multiplayer.

 

No, it's your ego which is what you're fighting with. I'm just stating the facts.

It doesn't matter. What he said simply doesn't exist.



Yes, multiplayer is more CPU intensive. No, it's not "optimized for 6700K only" and no you don't need HT on a 6C CPU to reach high fps as can be clearly seen on this graph. In fact, you don't even need a 4T CPU to reach high fps.

 

Hey thanks for proving my point. Notice the break in the graph between the 4-core and 6-core CPUs.

 

Rus,

You're not stating fact; you're ignoring them. You're ignoring them because your ego won't allow you to admit you were wrong or didn't know something.

I linked you to ea's own words. You say it's fake news. I linked you to a 6 core intel with hyper threading disabled showing 100% usage across all 6 cores. You said it was fake news. I told you to go see for yourself. You said didn't need to fake news.

I have a years worth of practical experience helping guys run the game better. You don't even play the game. I just helped out another guy, literally 5 minutes ago, on reddit. You're here denying reality.

Nobody is talking about high fps. That's your distraction. I'm talking about 100% cpu usage, which is poor optimization. You can deflect all you want, but you are flat out dead wrong, and you are unwilling to go through the simple steps to prove it to yourself that you are wrong. Ego.

 

Your point was that MP is more CPU intensive, I never argued with that. Now do notice the actual fps numbers there and think on how "poorly optimized" a game which runs at 60 fps on a Pentium-G CPU is.

That's what you're doing, you just completely ignore everything that goes against your ego. I've posted more than enough for you to see that FB3 is an insanely well optimized engine. Nothing to add.

P.S. 100% CPU utilization is not a poor optimization either. Stop using words which you don't understand.

 

60 FPS is nothing, even the weakass consoles with their tablet/phone CPUs do that. 144Hz is the standard for FPS on PC and none of those CPUs hit that mark. Prior Battlefields didn't have this kind of CPU demand, even on quad-cores 144 FPS in 64p MP was possible before.

 

"Weak ass consoles" have more than 2 weak cores and they are running below 1080p. 144Hz is not a standard on PC and no one ever said that a 4T CPU will hit it on Ultra preset in a 2016 AAA game.
Again, you guys don't know what optimization is. So avoid throwing around phrases like "poorly optimized" for anything really.

 

144Hz is the standard for gaming monitors, any competitive FPS player will tell you that and if you still deny it you're just behind the times.

 

/sigh...

You're being intentionally obtuse. This will be my last post on the subject, because I feel like I'm talking to a stop sign.

The issue is 100% usage. The issue is not fps, although if you ever went to the ea forums or reddit you would know that's an issue too. 100% cpu usage is poor optimization.

Back in the beta right after dice posted the requirements I asked them about it because I was sure they were incorrect. To list the 6600k as the minimum seemed like a clear mistake considering the amd equivalent they listed. 'What I got back... FROM A DEV... was, "that's all they had around". This falls inline perfectly with the article I linked where ea said they only optimize for high end hardware. So when you consider that all intel cpus run the game at 100% usage absent hyper threading, it's not a stretch to assume they optimized ONLY for the 6700k. Had they optimized for the i5 (6600k), the game would not run at 100% usage... as you yourself said, "it's the most optimized engine on the planet". Let that sink in. The most optimized engine on the planet... yet 100% usage.

If your reply is to link another canned benchmark showing fps, then please don't bother. It's another attempt to avoid the subject, which again, is that the game is poorly optimized and the proof that it's poorly optimized is that it runs at 100% usage on CURRENT gen i5's and 100% usage on CURRENT gen 6 core Intels with hyper threading disabled.

 

That's because you are - to the stop sign in your head. I can only repeat what I've already said: neither BF1 not FB3 are "poorly optimized" in any possible sense. The only part of them which is questionably built is the D3D12 renderer.

Yes, let's not bother with facts, people.

 

FYI, Frostbite's DX12 is impair especially on FIFA 18.

 

in my machine and I speak only for multiplayer with 64 people I had the problem of 100% usage with i5 4460 with or without v sync. (60 mhz monitor). and when goes at 100 % cpu usage the Gpu usage go at 40% and fps goes to 20.
so I change my cpu to i7 4790 stock speed and I have above 75 fps in all game.
if I disable the hyperthreating I have the same problem
4460.
and I have found something more if i put quicker ram speed in my i5 4460 it has lower cpu usage.
so I think that they have a bug where if the cpu waits for information and don't take it quick enough it stalls and goes at 100 percent.

 

Game mode is automatically enabled if your PC supports it, and you can't turn it off unless you do a registry hack.

 

Bf1 ran well on my old 980 and i7 2600k average 90 fps on everything ultra at 2560x1080 @75hz (and ultra requires more than 4 GB of vram apparently) but it was smooth as hell, not even single hitch or sttutter, was like that until i got my lg 38uc99 and i had to upgrade everything since at that resolution the processor was a little weak at rendering high resolution videos and the 980 was a little to weak for 3840x1600 and has no x265 support, but still, at this resolution i played at high - medium until i got my 1080ti the game runs very well even on dx12 for me, the only bad thing in dx12 it sttuters a little at the beggining of a multiplayer match but after that the game runs very well i think it's very well optimized, i don't think that claim of being optimized for high end hardware is right since in theory the gtx 580 a old gen high end hardware runs worse than the gt 1030

 

I hope Nvidia fix some HDR issues for some monitors like samsung the Chg70 with maxwell 2 cause HDR works fine with gtx 1000 series if i read the HWI monitor review.
Maxwell needs HDMI 2.0a/b for HDCP 2.2 i think.

 

Rus,

What's your game name? Since you're presenting yourself as an authority, I think it's only fair that everyone gets to see that you don't actually play the game.

Battlefield is a multiplayer game. I should not have to say that, but apparently I do, because you keep referencing canned single player benchmarks to respond to the charge that non hyper threaded Intel cpus have 100% usage... which is odd, because it's totally non-responsive. If you had any facts you'd post a chart showing cpu usage.

It's also not clear you're familiar with any Battlefield game? It took dice and dice la a year of optimizing BF4 before it was playable. I suspect you will dispute this too, but you would be alone.

This isn't a claim made up by some random guy on the internet; this is what ea admitted: https://www.digitaltrends.com/gaming/electronic-arts-develops-games-high-end-pc-first/

That you ran fine on a 2600k isn't unusual. Hyper threaded cpus run the game fine, and there hasn't exactly been a huge leap in ipc between Sandy Bridge and Sky Lake.

When the game came out I was running a 6600k and a 970. I didn't have any issues, but a metric poop ton of guys did, and still do. However, I could not help but notice that even in the menus cpu usage was ~ 95-100%. If your game is at 100% cpu usage even in the menus, then you can't make the claim that it's well optimized... no matter what bootlickers or single player canned benchmarks say. That said, this is typical dice though. They get around to optimizing, it just takes them a while.