Feedback - Extreme FPS Loss through binocs #3782

[hick]

"minimum specs" (our use of the term) = can play, but far from ideal (expect problems)

Minimum specs does not measure any really "fun" measure of performance. Just what is possible at the loweset end, and all players would be advised to try and be well above that level of spec if at all possible.

Basically below minimum specs you cannot play the game at all, or if you can and have problems we cannot support you with a promise of any fix for you, since minimum specs in this case really means "game will run on this machine" but we don't mean it to say will run at all well.

The point was that I am well above min specs, yet still use 100% of available processing power to run 1.31. It wasn't that way with beta release .18, so why, after numerous performance tweaks to the code, is it that way with RC .39?

It can be

a) the way the Game is sending instructions to the OS

the way my OS is Handling those instructions

c) Hardware issue

From running several scans of the GPU, CPU, RAM, and HDD, the hardware seems to be functioning properly, leaving only a), , or a combination thereof.

[hick]

Hold a sec, your driver is from 2008...I wonder if that might have something to do with it.

Not sure I can find a driver that will work with Win7 x64 from 2008.

Also you have DX10 installed.

DX11 = Win 7, DX10=Vista, DX9=XP. Not any real choice on this one.

[monsjoex]

Very odd. I run an 8800GT playing at 3840 x 1024 and I loose maybe 3 frames.

Are you CPU limited?

Its weird that you lose so few frames, because even though its (very) playable now, i still lose quite a lot of frames while i also have the 8800GT.

What CPU do you have? I got Q6600..

[lutorm]

The point was that I am well above min specs, yet still use 100% of available processing power to run 1.31. It wasn't that way with beta release .18, so why, after numerous performance tweaks to the code, is it that way with RC .39?

Well, your system should always run at 100%, otherwise you're not getting highest possible FPS. If the game is running at <100%, then something's wrong.

[hick]

Well' date=' your system should always run at 100%, otherwise you're not getting highest possible FPS. If the game is running at <100%, then something's wrong.[/quote']

huh? not sure where u get that from but it's faulty logic at best

[Linnon]

I notice a huge improvement when testing this in 1.31.39 on my machine.

[lutorm]

huh? not sure where u get that from but it's faulty logic at best

Expand on that please, I'd like to hear about my "faulty logic"...

[hick]

Expand on that please' date=' I'd like to hear about my "faulty logic"...[/quote']

Any giving program will use up to the maximum resources allowed by code, unless the hardware max is less that the coded max.

In other words, it is entirely possible to build a computer with enough processing power that a given program doesn't use 100% of it, all that is needed is more capacity than the program can write to/read from.

(It's why a game like Oregon Trail maxes out at 20% CPU Load)

[gtanner]

Not seeing an FPS drop but when moving with bino with a few trees in view the view becomes a little clunky like its snapping in place, not a gamebreaker but you do notice it.

Anyone else see this?

Nvidia 8800 GTS

[Hinfoos]

Not seeing an FPS drop but when moving with bino with a few trees in view the view becomes a little clunky like its snapping in place, not a gamebreaker but you do notice it.

Anyone else see this?

Nvidia 8800 GTS

have you tried it before last patch?

was fixed in .39

[gtanner]

have you tried it before last patch?

was fixed in .39

Tried it today, no loss in fps but the binos movement is a little jerky when moving across even a medium number of trees.

As i say it's not too bad but you can notice it compared to bino movement without trees in view.

[lutorm]

Any giving program will use up to the maximum resources allowed by code, unless the hardware max is less that the coded max.

In other words, it is entirely possible to build a computer with enough processing power that a given program doesn't use 100% of it, all that is needed is more capacity than the program can write to/read from.

(It's why a game like Oregon Trail maxes out at 20% CPU Load)

Yeah sure as a general point, except that in this game there is no "coded max" FPS, it will give you as high fps as the hardware is capable of. (Actually, I guess you can get that by turning vsync on, since then FPS is capped at the refresh rate. If that's what you meant then we're in agreement.)

[hick]

Yeah sure as a general point' date=' except that [i']in this game there is no "coded max" FPS, it will give you as high fps as the hardware is capable of. (Actually, I guess you can get that by turning vsync on, since then FPS is capped at the refresh rate. If that's what you meant then we're in agreement.)

There has to be some max allowed by the code. Right now it will depend on what level of multi-threading/multi-core support CRS added for 1.31.

Case in point would be 1.30. AFAIK, the code can only utilise 1 core, thus have additional cores only frees up space on the core BE is using, which indirectly affects performance. If BE is the only process running, then having any additional cores is theoretically useless.

By the same token, a piece of code that is single-threaded will only ever use one thread in a core, and while that thread will be full, the remainder of the core will go unused. (the same idea could be applied to pipes or cores on a GPU as well).

However, the converse is also true, as is demonstrated by nVidia's Mona Lisa experiment.

[GOPHUR]

I consider the primary issue closed. If there are new issues, start a new thread.