Why/How is the FX 8350 a bottleneck?

[Buckley]

Can someone try to explain to me why everyone seems to think the 8350 bottlenecks everything? In EVERY game ive played not once has all 8 cores been at 100% load, only during synthetic when the CPU is being evaluated.

[L.E.D]

I actually don't think it can be unless you are running an insane GPU setup with it...

[xrex64]

Actually, not many people do. It wont even bottleneck a 780TI... so...

[Karumu]

Can someone try to explain to me why everyone seems to think the 8350 bottlenecks everything? In EVERY game ive played not once has all 8 cores been at 100% load, only during synthetic when the CPU is being evaluated.

i'd like to know this too, and since i am running one, to me it's a really fast chip!

[Gung Pow Chicken]

I honestly don't think it does, maybe with dual 780 ti's or even tripple.

 

I had one in my last rig and nothing slowed that baby down, excluding something falling off my shelf lol.

[qwertywarrior]

a CPU not at 100% load does not mean its bottle necking

 

 

comes down to game optimizations

some games favor faster fewer cores

i like intel because i can turn off HT

[Vitalius]

a CPU not at 100% load does not mean its bottle necking

 

comes down to game optimizations

some games favor faster fewer cores

i like intel because i can turn off HT

This.

I facepalmed when OP said "all 8 cores at 100% load". That's not bottlenecking. Bottlenecking in this situation is when the CPU isn't strong enough to let the GPU be utilized 100%. It doesn't matter how much the CPU is utilized. It matters how much the game's engine utilizes the CPU fully.

Most games are single core. Some are multi-core but only do 2 or 4 threads. They don't know how to use all 8 cores on an 8350. So they only use what they can which isn't enough for the GPU to be used 100%. That's what the bottleneck is.

[MeltyMoon]

Can someone try to explain to me why everyone seems to think the 8350 bottlenecks everything? In EVERY game ive played not once has all 8 cores been at 100% load, only during synthetic when the CPU is being evaluated.

 

Hahaha it all honestly depends on game optimizations

[xox]

ermegherd, bottlenecks don't exist the way that people think they do, stfu.

 

Increasing graphics card or CPU performance will always give some kind of improvement no matter what the parts are in your system

 

the hearsay, oh, the hearsay

[simtransporter]

Can someone try to explain to me why everyone seems to think the 8350 bottlenecks everything?

 

I think it's because people see benchmarks where the 8350 is beaten by a few fps compared to another CPU, and that leads people to think that it bottlenecks everything, even though that's not necessarily true.

[Vitalius]

ermegherd, bottlenecks don't exist the way that people think they do, stfu.

 

Increasing graphics card or CPU performance will always give some kind of improvement no matter what the parts are in your system

 

the hearsay, oh, the hearsay

*follows advice*

*Puts a Titan Black in my Dell pre-built computer*

*still gets same FPS in Minecraft*

JULYER!

[Enderman]

A CPU does not need to be at 100% to be a bottleneck. If a game need many light calculations done per second then it will not be enough to put your CPU to 100% because the CPU is too slow to keep up with the amount of calculations. It has to do them in order, so the longer it takes to do one, then longer it will take to get to the others. Since the 8350 has 'half cores' yes they are eight, but they are not very powerful. On the other hand, the 4670k has only 4, yet these have much more computing strength which is what is important in games. Since 99% of games use 4 cores or less, there is no advantage of haming over 4 cores like the 8350 does. But the i5 cores are almost as powerful as two of the 8350's cores so they are able to offer better fps in the game which only uses 4 cores. You can see from the benchmarks that in games that use 4 cores or less, the 4670k will always get  better fps. This will change as games begin to take advantage of more cores, rather than fewer more powerful cores.

[Nineshadow]

Game optimization.

[TechFan@ic]

Any half decent $120+ CPU will give you a fantastic gaming experience these days. Most folks get i7s just because they can not really because they need them.
I have personally tested an FX 8320 & an i7 2600 with my own freaking hands & they performed the same in every single game I've benchmarked except one & even then I wasn't able to tell the difference in the gameplay experience.
Most folks don't care that much about their CPU purchase to research for more than a few days that's if they even end up doing any research at all & in the end the brand name & brand loyalty are still major factors.

Intel has always moved more product than AMD & will likely always will that doesn't mean that the AMD product is bad it just means that Intel runs a more successful business. Look at CryTek, they've made the most advanced graphics engine in the industry & the best looking games of all time & now they're facing serious financial perils it doesn't mean that CryTek is a bad studio, in fact I would argue that they're one of the best. You can have a great product & still have a tough time trying to sell it.

[cthomas1489]

I'm using an 8350 and crossfire 290x's... I've seen no bottle neck that I know of yet.

[Faceman]

in certain games like BF4, the FX8xxx bottlenecks a 780.  Even in Cinebench, I remember someone who had a FX8320 OC'd to 4.6Ghz was only getting 91fps when they did the GPU test, but I am getting 118fps with my i5-4670k at stock, and 134fps with it at 4.7Ghz.  We both had the same model of 780s, not overclocked.

[Buckley]

in certain games like BF4, the FX8xxx bottlenecks a 780.  Even in Cinebench, I remember someone who had a FX8320 OC'd to 4.6Ghz was only getting 91fps when they did the GPU test, but I am getting 118fps with my i5-4670k at stock, and 134fps with it at 4.7Ghz.  We both had the same model of 780s, not overclocked.

please no leave him out of this lol id actually like to keep this thread open and NOT locked

[Faa]

Can someone try to explain to me why everyone seems to think the 8350 bottlenecks everything? 

Because games are shitty multithreaded and any popular/massive multiplayer game is cpu bound? Gaming performance != multithreaded performance. 8350's full performance isn't even unleashed in games, with a 4670k you're atleast getting the most out of it. The only game that literally can take a 8350 to 100% load is Crysis 3 (proof: youtube.com/watch?v=_hcuYiqib9It) any other game isnt even close to their level of multithreaded friendlyness. If they all were like Crysis 3 we wouldn't have that many reasons to recommend i5's over 8350's anymore.

But bottlenecks everything? Uh? Plenty of graphs listing the gpu's performance variation and putting the most expensive cpu on the top on purpose with 0.01 more fps. Anyways everyone these days seem to be having a different understanding of bottlenecks, one thinks a 780ti with a pentium 2, the other one wants 99% all across his gpu's to complety justify his pair of 1000$ cards or just aslong as it provides 60 fps. Imo; you're not supposed to buy a 780 if not even one of your games can take it to 50% load max - neither you are supposed to have two cards in SLI never getting higher than 50%/50% because that performance would be equal to a single card at 99% (point of this would be getting rid of coil whine). It's funny that you see people buying three titan blacks with their outdated 3930K at 4.8GHz for WoW when a simple i5 4670k at 4.4GHz and a gtx 750ti would perform noticeably better in 25m raids than his rig. 

 

 

In EVERY game ive played not once has all 8 cores been at 100% load, only during synthetic when the CPU is being evaluated.

Monitoring cpu load meh, its just useless other than finding out how many cores are actually needed for that game. If people still think that your cpu load is your bottleneck indicator well theyre wrong, you can have your cpu at 100% and still cranking your gpu to 99% or you get a cpu bottleneck at 10% load. Your gpu load is your only indicator.

If youre all over the place gpu bound, dont bother googling a cpu.

 

 

please no leave him out of this lol id actually like to keep this thread open and NOT locked

Just because you couldn't agree a locked i5 is the better choice than a 8320 when both were priced the same? I recommended him the i5 just to make sure that he gets the IPC advantage if he ever plays a cpu bound game which are as you know lightthreaded.

[Buckley]

Because games are shitty multithreaded and any popular/massive multiplayer game is cpu bound? Gaming performance != multithreaded performance. 8350's full performance isn't even unleashed in games, with a 4670k you're atleast getting the most out of it. The only game that literally can take a 8350 to 100% load is Crysis 3 (proof: youtube.com/watch?v=_hcuYiqib9It) any other game isnt even close to their level of multithreaded friendlyness. If they all were like Crysis 3 we wouldn't have that many reasons to recommend i5's over 8350's anymore.

But bottlenecks everything? Uh? Plenty of graphs listing the gpu's performance variation and putting the most expensive cpu on the top on purpose with 0.01 more fps. Anyways everyone these days seem to be having a different understanding of bottlenecks, one thinks a 780ti with a pentium 2, the other one wants 99% all across his gpu's to complety justify his pair of 1000$ cards or just aslong as it provides 60 fps. Imo; you're not supposed to buy a 780 if not even one of your games can take it to 50% load max - neither you are supposed to have two cards in SLI never getting higher than 50%/50% because that performance would be equal to a single card at 99% (point of this would be getting rid of coil whine). It's funny that you see people buying three titan blacks with their outdated 3930K at 4.8GHz for WoW when a simple i5 4670k at 4.4GHz and a gtx 750ti would perform noticeably better in 25m raids than his rig.

Monitoring cpu load meh, its just useless other than finding out how many cores are actually needed for that game. If people still think that your cpu load is your bottleneck indicator well theyre wrong, you can have your cpu at 100% and still cranking your gpu to 99% or you get a cpu bottleneck at 10% load. Your gpu load is your only indicator.

If youre all over the place gpu bound, dont bother googling a cpu.

Just because you couldn't agree a locked i5 is the better choice than a 8320 when both were priced the same? I recommended him the i5 just to make sure that he gets the IPC advantage if he ever plays a cpu bound game which are as you know lightthreaded.

I'm not even going to bother reading this. Let's not go nearly as far cause frankly I don't even care, I was just curious

EDIT - Actually read it

[Bloodkil933]

I actually started to believe this forum had few fanboys, but judging by some some of the threads and posts appearing at the moment, it now think the opposite.

[cthomas1489]

I'm not even going to bother reading this. Let's not go nearly as far cause frankly I don't even care, I was just curious

EDIT - Actually read it

 

I read it too. I'm actually really interested in this and want to keep the discussion going. I think every setup is different and going to produce different results. I have an 8350 and CF 290x's and haven't experienced any "bottlenecking" yet, though maybe I don't play those games. I have the cpu OC'd. This may change when my monitor comes and I switch to 4k but for instance:

 

http://www.anandtech.com/bench/product/1057 - Lists BF4 ultra at around 91 FPS. I maxed everything possible at 1080p and got a lot high fps than this in both SP and MP (both with and without mantle). I monitored load %'s and nothing seemed to be holding the other back.

[Buckley]

I read it too. I'm actually really interested in this and want to keep the discussion going. I think every setup is different and going to produce different results. I have an 8350 and CF 290x's and haven't experienced any "bottlenecking" yet, though maybe I don't play those games. I have the cpu OC'd. This may change when my monitor comes and I switch to 4k but for instance:

http://www.anandtech.com/bench/product/1057 - Lists BF4 ultra at around 91 FPS. I maxed everything possible at 1080p and got a lot high fps than this in both SP and MP (both with and without mantle). I monitored load %'s and nothing seemed to be holding the other back.

My 290x is pushing 90 fps with dx11 maxed 1080p and 110+ with Mantle. Haven't run into a bottleneck here yet. also Linus did a video awhile baCk and the 8350 actually did better at 1440p than the i5 3570k so maybe it'll do better at 4k? Idk haven't seen benchmarks

[TymmVonD]

When I was running 8350/780 config, in Rust, my CPU was at 98-99% while my GPU wouldn't go past base clocks. It would stay around 30% load on GPU.

[Faa]

I read it too. I'm actually really interested in this and want to keep the discussion going. I think every setup is different and going to produce different results. I have an 8350 and CF 290x's and haven't experienced any "bottlenecking" yet, though maybe I don't play those games. I have the cpu OC'd. This may change when my monitor comes and I switch to 4k but for instance:

 

http://www.anandtech.com/bench/product/1057 - Lists BF4 ultra at around 91 FPS. I maxed everything possible at 1080p and got a lot high fps than this in both SP and MP (both with and without mantle). I monitored load %'s and nothing seemed to be holding the other back.

I'm sitting between 120-200 fps with my 780's while I've never ever hit 99% on any gpu because 1) cpu bottleneck 2) fps cap of 200

 

 

When I was running 8350/780 config, in Rust, my CPU was at 98-99% while my GPU wouldn't go past base clocks. It would stay around 30% load on GPU.

Rust taking a 8350 to 100% doesnt sound normal. 

[Buckley]

I'm sitting between 120-200 fps with my 780's while I've never ever hit 99% on any gpu because 1) cpu bottleneck 2) fps cap of 200

Rust taking a 8350 to 100% doesnt sound normal.

How is that even possible? What're your settings?

EDIT - 780 Sli OK I see that

When I was running 8350/780 config, in Rust, my CPU was at 98-99% while my GPU wouldn't go past base clocks. It would stay around 30% load on GPU.

Something's wrong bro, no game out there has made the 8350 maxed out on all cores

I'm sitting between 120-200 fps with my 780's while I've never ever hit 99% on any gpu because 1) cpu bottleneck 2) fps cap of 200

Rust taking a 8350 to 100% doesnt sound normal.

But my 290x is at 99% / 100%