My 8150 Bulldozer experience - so far!

[bononos]

The i7-3960X has the Sandy Bridge architecture...

What we're seeing here is one of the very few games than can take advantage of more than four cores.

My mistake the 3960 is SB but Borderlands2 doesn't look at it runs better with more cores at all, at least not always. Look at the graph, the 4100 runs faster than the 8150 and the 3770k or 2700k is barely faster than the 2500k.

 

[Yuriman]

What do you make of the closeness of the scores between AMD's 4, 6 & 8 core models?

Hard to say, the FX-6200 runs slower than the 8150 even though it has a clockspeed advantage, but the FX-4170 is faster than either and its only advantage is clockspeed. They all have the same amount of cache.

??

I don't play neither of those. So to me it means nothing.

My wife doesn't play those games either, she's very happy with her Q6600 system. That doesn't make the Q6600 a better processor, she just doesn't have very demanding usage patterns. It's a fantastic CPU if all you want to do is play Sims 3, browse the web, and do occasional emulation.

 

[infoiltrator]

There is "good enough" and "better."

 

[AtenRa]

I'll just leave this here.

CPU: AMD FX 8150 (overclocked at 4.6GHz)
CPU Cooler: AMD FX WaterCooling
Case: ThermalTake Armor A60
Mobo: ASUS Crosshair V Formula
VGA: MSI HD6950 Twin Frozr III (
RAM: Kingston HyperX 2133MHz 16GB kit (2x 4GB)
Ram was running at 1926MHz 9-11-9-27 (default CPU frequencies at 1333MHz)
HDD : Seagate 1TB 7200rps SATA-3
PSU: ThermalTake Smart 730W 80+
Win 7 64bit SP1
Cat : 12.8


BoarderLands 2 Map tested : Three Horns Divide
FoV = 76

 

[guskline]

I just ordered Borderland2 and when I have it installed, I'll post the fps measure by FRAPS (I have licensed version) on my rigs below.

 

[guskline]

I benched the second level like the Tech rport and at highest settings with 8150 OC' to 4.5 GHz and my EVGA GTX670 FTW stock fps averaged 60 fps.

I'll bench the 2500k with the same video card later.

 

[AtenRa]

Could you pls turn PhysX to low and bench again ??

 

[Mallibu]

Frankly I don't know anyone who plays those two games (and I know quite a few folks who game on their PCs). As for subpar,that lowly dual core is not that far off from other SB CPUs. To me it just shows how badly written some of those titles are.

Yeah, Guild Wars 2 topped charts for 2-3 weeks and sold > 2 million copies but i guess those statistics mean nothing since you and your friends ain't playing.. :whiste:
And let's wait and see Borderlands popularity since it launched yesterday. In the mean time, make some progress and get past the first stage

 

[LOL_Wut_Axel]

What do you make of the closeness of the scores between AMD's 4, 6 & 8 core models?

That due to their low single-threaded performance it means a huge bottleneck early on. You need good single AND multi-threaded performance in the game, hence why the 3960X is on top.

Also keep in mind: throughout history, the only benefit that has been seen with Hyper-Threading in gaming has been with the 2C/2T vs 2C/4T models like the Pentium G850/G860 vs. Core i3-2100/2120, where HT gives a significant edge. 4C/4T vs. 4C/8T has always performed exactly the same.

The game is taking advantage of the additional cores but not Hyper-Threading, at least for the models above the i3.

 

[tweakboy]

Wow I read your whole book. I like you, your brave trying a Bulldozer when you already have a Sandy. 2 of them at that. What you need all these comps for ?

You never told us a full load temp ? 60's and 70's c is good temp after OC. If your getting 80's and higher then that chip will die in time,, lets hope not...

What 64 bit app do you use to really test the 8 cores ? Games certainly don't need 8 cores. I tell you what does.. 2 things .

Video Editing rig or Audio Production. DAW

I might go bulldozer in the future when I upgrade just to get 8 cores with Sonar X2.

However I believe in HT and a Sandy 2600k might be better. The CPU usage will show 8 . But with a Sandy 2600k it will also show 8 .. hmmmmmm and I need min 32GB RAM...

 

[alexruiz]

Yeah, Guild Wars 2 topped charts for 2-3 weeks and sold > 2 million copies but i guess those statistics mean nothing since you and your friends ain't playing.. :whiste:
And let's wait and see Borderlands popularity since it launched yesterday. In the mean time, make some progress and get past the first stage

Two members here just benched borderlands 2 on the FX8150, and their findings are much higher than that website...

 

[Mallibu]

Two members here just benched borderlands 2 on the FX8150, and their findings are much higher than that website...

Everyone benchmarks different sections of the game, that doesn't make any results invalid. Oh, and in Aten Ra's blog FX at stock speeds drops to less than 60 frames 12 times in 2 minutes, and that kind of stuttering is game-breaking.

 

[inf64]

Yeah it gotta stutter at ~45fps lol... Do you even hear yourself? DO you even believe the things you write?

 

[Mallibu]

Yeah it gotta stutter at ~45fps lol... Do you even hear yourself? DO you even believe the things you write?

Don't know what happens in your little planet you live in but in this one you want 60 fps locked. Any drop below 60 fps will cause unwanted stuttering and loss of smoothness. Trying to downplay that fact because FX can't maintain 60 fps is sad and funny at the same time though.

 

[inf64]

I'm done discussing things with you anymore since you are not making any sense. When you get rid of the "other personality" that disturbs the thought process we may continue.

 

[Makaveli]

Don't know what happens in your little planet you live in but in this one you want 60 fps locked. Any drop below 60 fps will cause unwanted stuttering and loss of smoothness. Trying to downplay that fact because FX can't maintain 60 fps is sad and funny at the same time though.

On my rig I mostly stays at 60fps but i get drops to 30fps in large area's with heavy action. I notice the slow down but I wouldn't call it a stuttering mess.

From what i've read on other forums some guys with 680 SLI are getting lows of 40fps in certain areas.

 

[Mallibu]

I'm done discussing things with you anymore since you are not making any sense. When you get rid of the "other personality" that disturbs the thought process we may continue.

Most of your posts are generic name calling crap and blindly denying obvious facts backed up by graphs and results. Your highly intelligent arguments "I don't play that game so therefore it doesnt exist!" will be missed though. Bye :awe:

 

[inf64]

Gbye and good riddance .

 

[guskline]

Wow I read your whole book. I like you, your brave trying a Bulldozer when you already have a Sandy. 2 of them at that. What you need all these comps for ?

You never told us a full load temp ? 60's and 70's c is good temp after OC. If your getting 80's and higher then that chip will die in time,, lets hope not...

What 64 bit app do you use to really test the 8 cores ? Games certainly don't need 8 cores. I tell you what does.. 2 things .

Video Editing rig or Audio Production. DAW

I might go bulldozer in the future when I upgrade just to get 8 cores with Sonar X2.

However I believe in HT and a Sandy 2600k might be better. The CPU usage will show 8 . But with a Sandy 2600k it will also show 8 .. hmmmmmm and I need min 32GB RAM...

I'll try to answer your questions. I'm not brave, I just got lucky to have the 8150 fall to $170 and since I already had the Asus Sabertooth 990 FX mb I thought it would compliment the MB well. Why do I have 3 computers? I have more than that! I don't golf and other than my wife's 2007 Pontiac Solstice GXP with a Trifecta tune and suspension mods, my primary hobby is computers. To each his own!

Now as to temps, just reran IntelBurnTest 2.54 for a 10x run and noted thew temps. CPU overall never exceeded 61C (most times in the mid to high 50s) Each nCPU core was @51-52C and the Vcore got as high as 69C so my temps are EXCELLENT. Of course a Corsair H100 cooler with stock fans at max and very good case cooling helps. As I type this AIda64 is reading Overall CPU 34C, each core from 15-23C, vcore 40-43C

 

[guskline]

Two members here just benched borderlands 2 on the FX8150, and their findings are much higher than that website...

Irealized that I had vsync enabled in the Bulldozer 8150 rig and not enabled in the 2500k rig with the same video card. After correction the Bulldozer rig in the same scene as the 2500k rig would get @97fps with a resolution of 1920 x 1200 while the 2500k rig with the same video card but at 1920 x 1080 resolution would get 120Fps so yes the Bulldozer I have with the GTX670 FTW is scoring much higher fps than the article seems to show.

 

[guskline]

Most of your posts are generic name calling crap and blindly denying obvious facts backed up by graphs and results. Your highly intelligent arguments "I don't play that game so therefore it doesnt exist!" will be missed though. Bye :awe:

Mallibu All I can tell you is that my 8150 in my sig below is playing Borderlands2 remarkably well. The 2500k is likely faster but trust me not noticeable.

What I have found using Fraps with Borderlands2 is that the fps for different scenes are all over the place!

 

[inf64]

Mallibu All I can tell you is that my 8150 in my sig below is playing Borderlands2 remarkably well. The 2500k is likely faster but trust me not noticeable.

What I have found using Fraps with Borderlands2 is that the fps for different scenes are all over the place!

Don't tell him this ,it's lies . It's nice that we have members with both CPUs who can actually test things out (like you yourself and AtenRa). I always prefer first hand experience on forums than some review. In this case it just shows how unreliable some reviews are .

Thanks for taking the time to do the testing,it can be tiresome and boring(I know).

 

[guskline]

If anything, I've downplayed the Bulldozer because my 2500k chips are so strong. There are people on the forum who own both the Bulldozer and Intel 2500k/2600k and or 3570k/3770k platforms and most report the Intels to be faster. My observation is that the 8150 when properly OC'd and matched with the same video card, and other hardware is slower in most benchies but not the DOG that some articles made it out to be. Frankly, I'm not an elite Gamer so my opinion that gameplay between the 8150 and the 2500k is not noticeable must be taken with a grain of salt. Also, I've not had the pleasure to play or even see in person a I7-3930k/3960k rig so I'll leave that to the owners of those rigs.

 

[LOL_Wut_Axel]

If anything, I've downplayed the Bulldozer because my 2500k chips are so strong. There are people on the forum who own both the Bulldozer and Intel 2500k/2600k and or 3570k/3770k platforms and most report the Intels to be faster. My observation is that the 8150 when properly OC'd and matched with the same video card, and other hardware is slower in most benchies but not the DOG that some articles made it out to be. Frankly, I'm not an elite Gamer so my opinion that gameplay between the 8150 and the 2500k is not noticeable must be taken with a grain of salt. Also, I've not had the pleasure to play or even see in person a I7-3930k/3960k rig so I'll leave that to the owners of those rigs.

The fact is, AMD CPUs are inferior to Intel CPUs in gaming at all price points. The only reason to get an AMD CPU for gaming is if you're an AMD fan(boy). Intel gives you higher gaming performance at a lower price and with lower power consumption. People (read: AMD fan(boys) ) are always quick to point out that a Core i3-2120 and 3220 are "only" dual-cores but they always forget that they have Hyper-Threading and they have big, wide cores. With something like an FX-4100 or 6100 you get weak cores that need to be heavily overclocked, and even then they'll still be slower than the i3 in gaming. Games want anywhere from 2-4 cores, and they want them to be fast. AMD's cores are not fast, hence why they offer no compelling CPUs for gaming at any price point right now.

 

[bononos]

Two members here just benched borderlands 2 on the FX8150, and their findings are much higher than that website...

You cannot compare them directly since Atenra's benchmark was at a lower resolution 1080 (vs 1200 for 'that website') and a slower video card and if he followed his usual modus operandi, he used a flyby benchmark which tends to give higher scores.