BF4 threads

[jmaster299]

Will the stock clock 8350 be good for BF4 gaming? would it present any performance issues?

Are you some how incapable of clicking on the link I provided 4 posts above your own?????

[John Rem]

Thanks. You're really helpful and friendly.  

[LOST TALE]

so you should have at least 4 cores and 2 threads

No one has 4 cores and 2 threads. its minimum 1 thread per core.

Have you ever seen 2cores on one thread?

[Wiff]

No one has 4 cores and 2 threads. its minimum 1 thread per core.

Have you ever seen 2cores on one thread?

I didnt understand it correctly first, but now I do

[Wiff]

No, 4 cores or threads total. The i3's in the test are 2 cores, but 4 threads total because each of those 2 cores can process two things at once. An i5 is 4 cores, but only 4 threads because each core can only process 1 thing at a time. An i7 is 4 cores, but 8 threads, again each core is processing two things at once. A 8350 from AMD is 8 cores but only 8 threads, since each core can only process one thing at a time, just like an i5.

but if its true what you said, then there shouldn't be a noticable difference between the FPS with an i3 and FPS with an i5 (if both have the same GPU added)

[LOST TALE]

but if its true what you said, then there shouldn't be a noticable difference between the FPS with an i3 and FPS with an i5 (if both have the same GPU added)

Without HT process compatibility: an I5 has twice the amount of cores which then brings double the performance ( they usually have the same cores for each egneeration and I3 is just a bin failed I7)

 

EXCEPT if: The I3 will perform equaly as the I5 when I3's extra threads are used as much as I5's extra 2 cores. But since optmisations lack, it almost never happens. You may only benefit in stuff like video editing who cna parallel process very well small bits of information, easily distributed across threads. Also stuff like Frostbite 2 engine in BF3/BF4

 

HyperThreaded Extra Threads don't behave in the same manner. They have more stiff requirements to be used (effectively having from none to much lower cpu usage then the main threads)

Basicly 2 threads per core involves 1 Main thread and a secondary thread where compatible processes are processed.

 

Exemple (numbers used as relative unkowns): I5 core has a processing power of 1000. The full I5 proccessor has processing power of 4000.

I3 core has a processing power of 1000. The full I3 proccessor has processing power of 2000 (without hyperthreading)

If hyperthreading is fully used, I3 will total processing power of 2000 per core and, therefore, 4000 in total

 

Source: online WTM graphs - Information assesment Hypothesis

[Wiff]

HyperThreaded Extra Threads don't behave in the same manner. They have more stiff requirements to be used (effectively having from none to much lower cpu usage then the main threads)

 

Basicly 2 threads per core involves 1 Main thread and a secondary thread where compatible processes are processed.

Source: online WTM graphs.

ok, thanks bro

[LOST TALE]

but if its true what you said, then there shouldn't be a noticable difference between the FPS with an i3 and FPS with an i5 (if both have the same GPU added)

Only if the GPU fully utilises the I5 cpu therfore having more then half less FPS in i3. ( not all cpu usage goes ot fps +algebra)

[Wiff]

Only if the GPU fully utilises the I5 cpu therfore having more then half less FPS in i3. ( not all cpu usage goes ot fps +algebra)

k

[jmaster299]

but if its true what you said, then there shouldn't be a noticable difference between the FPS with an i3 and FPS with an i5 (if both have the same GPU added)

The do perform exactly the same, try looking at the benchmarks I provided above. 

 

i3-3220 95fps

 

i5-3570k 97fps

2fps difference is with in margin of error.

[Vrbas]

The do perform exactly the same, try looking at the benchmarks I provided above.

i3-3220 95fps

i5-3570k 97fps

2fps difference is with in margin of error.

Have to remember that is single player. 64 man multiplayer will have different results as it is more cpu heavy. And no one plays battlefield for the shitty campaign.

[jmaster299]

Have to remember that is single player. 64 man multiplayer will have different results as it is more cpu heavy. And no one plays battlefield for the shitty campaign.

You can't run a valid benchmark in multiplayer because it's not a set and repeatable sequence. Also, the CPU's performance in handling network traffic is a Windows issue/limitation, not a game issue/limitation.

[Vrbas]

You can't run a valid benchmark in multiplayer because it's not a set and repeatable sequence. Also, the CPU's performance in handling network traffic is a Windows issue/limitation, not a game issue/limitation.

That's exactly my point. You can't go with those benchmarks because that's campaign where it's the same situation every run.

[jmaster299]

That's exactly my point. You can't go with those benchmarks because that's campaign where it's the same situation every run.

Incorrect. How a CPU performs in handling the game is important and these benchmarks are the only valid way to test that. Performance during multiplayer is NOT an issue between your game and your CPU. It's an issue between your CPU, Windows and the server you are trying to communicate with.

You can have game A and game B, your CPU runs game A just fine in single player, but not game B. That means your CPU is not good enough just for game B. But if your CPU can't handle the network traffic for both games on a 64 player server, that just means your CPU is not good enough to handle a server with that many players. The game that server is for is irrelevant. That just means you need to play ALL multiplayer games on servers with lower player counts.