My 8150 Bulldozer experience - so far!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
AtenRa: Thanks for your suggestions. However, my chip just wasn't stable at the settings you mention. I use a 10 run of IntelBurntest as my first diagnostic. At my settings (21 multiplier, 200 FSB, Auto cpu voltage and ram at 18666 I can run all the diagnostics without a problem. A 21.5 mutiplier causes some problems as does a higher fsb. It might be my PSU espeecially since I'm running 2 5850s in CF have an Antec Green 750W. I do like running the 2 5850s however so for now I'm staying at 4.2 Ghz with all 8 cores active.
Any benchmarks will be run that way.
Speaking of Benchmarks, I ran AIDA 64 benchies and here are the results:
Mem
read 13732
write 10041
copy 13199
latency54.5
CPU
Queen 36159
Photoworx 31598
Zlib 304.5
AES 191485
Hash 4292
FPU
VP8 3296
Julia 13663
Mandel 6980
SinJulia 2931
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
moonbogg: Just ran cinebench 11.5 at my settings (4.2 Ghz all 8 cores)
CPU 6.89 pts
OpenGL 65.87fps
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
Pixelpusher6: Sorry but right now the 5850s are in my Bulldozer 8150 rig
 

sequoia464

Senior member
Feb 12, 2003
870
0
71
moonbogg: Just ran cinebench 11.5 at my settings (4.2 Ghz all 8 cores)
CPU 6.89 pts
OpenGL 65.87fps

Just for a reference - on a sabertooth with a 960T - 6 cores @ 3800 MHz

CPU 6.71
Open GL 49.37

CPU score for a 6100@4.2 was 5.15
 

moonbogg

Lifer
Jan 8, 2011
10,637
3,095
136
moonbogg: Just ran cinebench 11.5 at my settings (4.2 Ghz all 8 cores)
CPU 6.89 pts
OpenGL 65.87fps

Is it really 6.89 even at 4.2Ghz? Thank you for taking the time to do that. I won't be a jerk about this, but I can't resist the urge to express how totally shocked I am at how terrible that score is for an 8 core CPU with clocks like that. I mean, how can my intel chip get a score of 13 with 2 less cores? Something is really wrong here. Does cinebench favor Intel chips?

EDIT: To be fair, I am now at 4.3 Ghz so my score is 12.43
 
Last edited:

Makaveli

Diamond Member
Feb 8, 2002
4,760
1,159
136
Is it really 6.89 even at 4.2Ghz? Thank you for taking the time to do that. I won't be a jerk about this, but I can't resist the urge to express how totally shocked I am at how terrible that score is for an 8 core CPU with clocks like that. I mean, how can my intel chip get a score of 13 with 2 less cores? Something is really wrong here. Does cinebench favor Intel chips?

EDIT: To be fair, I am now at 4.3 Ghz so my score is 12.43

Are you new here?

I score in the 10 range on my rig.

And OP I thought this was a good real world experience and I enjoyed reading the thread so far.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
I am curious, how good is amd's fx branded liquid cooler when ocing ?

Pretty good, i have OCed the FX8150 at 4.6GHz with 35c Ambient, 4.7-4.8GHz in winter.


@guskline

If you leave the NB at 200MHz, is the CPU stable at 4.4GHz with Vcore at 1.425v ??
 

moonbogg

Lifer
Jan 8, 2011
10,637
3,095
136
Are you new here?

I score in the 10 range on my rig.

And OP I thought this was a good real world experience and I enjoyed reading the thread so far.

Not new, but newly shocked every time I see how bad BD really is. Wait...yeah I guess I am pretty new here.
 

Gikaseixas

Platinum Member
Jul 1, 2004
2,836
218
106
guskline, thanks the great read.

I still don't think the 8350 will solve AMD's problems but i'm keeping an open mind and i'm waitting for official benchmarks.
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
Is it really 6.89 even at 4.2Ghz? Thank you for taking the time to do that. I won't be a jerk about this, but I can't resist the urge to express how totally shocked I am at how terrible that score is for an 8 core CPU with clocks like that. I mean, how can my intel chip get a score of 13 with 2 less cores? Something is really wrong here. Does cinebench favor Intel chips?

EDIT: To be fair, I am now at 4.3 Ghz so my score is 12.43
moonbogg: I won't mince words. Your 3930k is far superior and much more expensive (@$500 vs $170 for the 8150). You have 6 core BUT Hyperthreading so you really have 12. Plus as evidenced by my SandyBridge 2500ks the Intels have much better IPC. In the Bulldozer case "more cores" does not mean more performance.
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
guskline, thanks the great read.

I still don't think the 8350 will solve AMD's problems but i'm keeping an open mind and i'm waitting for official benchmarks.

One of the reasons why on jumped on this chip at a lower price. Unless the 8350 makes significant leaps in performance AND comes in under $225 it gonna be a tough road. Perhaps the rumors of getting out of the highend market are true.
 

inf64

Diamond Member
Mar 11, 2011
3,764
4,223
136
Is it really 6.89 even at 4.2Ghz? Thank you for taking the time to do that. I won't be a jerk about this, but I can't resist the urge to express how totally shocked I am at how terrible that score is for an 8 core CPU with clocks like that. I mean, how can my intel chip get a score of 13 with 2 less cores? Something is really wrong here. Does cinebench favor Intel chips?

EDIT: To be fair, I am now at 4.3 Ghz so my score is 12.43
FX8xxx has 4 FP units,just like SB/IB QC models. It has 8 integer cores though. So the performance is roughly in line with intel SB 2600K(or slightly lower in case of Cinebench). Unless you expected that each AMD's FP unit has 2x the throughput of each SB's core,the performance is in line with what the uarchitecture was designed for (an integer cruncher).

People do get tangled up in "core numbers" without knowing what's behind the concept. Bulldozer does make some sacrifices/tradeoffs and one of them is the shared FP unit.

PS Before someone tells us that each FP unit can do two threads,yes it can but from throughput perspective it can do 2x128bit MULs or 2x128bit ADDs or combination of each(1+1).Same goes for 128bit SSE. On the other hand each K10 and SB core can do the same number of ops in legacy SSE,so we have 6 K10 cores vs 4 FX FP units vs 4 SB/IB cores. FMA/AVX changes things drastically in FX/SB favor but it's a different story.
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
inf64 thanks for the info. I'm now going to post my Passmark Performance 7 scores. I know some will scream "Passmark that's bogus" but I own a licensed copy so at least the scores can be used as a comparison. Here goes.

Overall score 2829.2
CPU Mark 9569.1
Expanded CPU results
Integer 1932
FP Math 6423.6
Find prime number 2002.7
SSE 36.7
Compression 13107.7
Encrytion 36.5
Physics 5268
String sorting 6922.7

2D Graphics 693.2
3D Graphysics 3048.6
Memory 1362
Disk 1464.9
CD 263.6
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
Pretty good, i have OCed the FX8150 at 4.6GHz with 35c Ambient, 4.7-4.8GHz in winter.


@guskline

If you leave the NB at 200MHz, is the CPU stable at 4.4GHz with Vcore at 1.425v ??
No. I think to go higher stable I'll need liquid cooling. Don't get me wrong, it con boot into windows at 4.3 up to 4.5 but I use Intel Burn test and it freezes. Could be a weak PSU (Antec Green 750W) since I'm using 2 5850s but I doubt it. I think its the thermals. The Hyper212+ even with 2 fans can't take the heat off fast enough when this chip runs faster than 4.2 Ghz on all eight cores. Quite frankly, liquid cooling with a new case will cost @$150 and for what? A gain of perhaps .4 Mhz? Not worth it. I'm satisfied with a SOLID 4.2 Ghz Bulldozer. BTW some posters say "hey at stock it clocks up to 4200" True but only on 4 cores. I have this cpu running all 8 cores at 4200 24/7/365 and can run anything and its stable.
 
Dec 30, 2004
12,554
2
76
inf64 thanks for the info. I'm now going to post my Passmark Performance 7 scores. I know some will scream "Passmark that's bogus" but I own a licensed copy so at least the scores can be used as a comparison. Here goes.

Overall score 2829.2
CPU Mark 9569.1
Expanded CPU results
Integer 1932
FP Math 6423.6
Find prime number 2002.7
SSE 36.7
Compression 13107.7
Encrytion 36.5
Physics 5268
String sorting 6922.7

2D Graphics 693.2
3D Graphysics 3048.6
Memory 1362
Disk 1464.9
CD 263.6

Licensed copy does not make a difference.
 

jhu

Lifer
Oct 10, 1999
11,918
9
81
FX8xxx has 4 FP units,just like SB/IB QC models. It has 8 integer cores though. So the performance is roughly in line with intel SB 2600K(or slightly lower in case of Cinebench). Unless you expected that each AMD's FP unit has 2x the throughput of each SB's core,the performance is in line with what the uarchitecture was designed for (an integer cruncher).

People do get tangled up in "core numbers" without knowing what's behind the concept. Bulldozer does make some sacrifices/tradeoffs and one of them is the shared FP unit.

PS Before someone tells us that each FP unit can do two threads,yes it can but from throughput perspective it can do 2x128bit MULs or 2x128bit ADDs or combination of each(1+1).Same goes for 128bit SSE. On the other hand each K10 and SB core can do the same number of ops in legacy SSE,so we have 6 K10 cores vs 4 FX FP units vs 4 SB/IB cores. FMA/AVX changes things drastically in FX/SB favor but it's a different story.

You're saying K10h's FPU can also do 2x128-bit MUL/ADDs?
 

jhu

Lifer
Oct 10, 1999
11,918
9
81
To OP, could you run Povray 3.7 windows version? There's a 64-bit windows binary here. Linux version is a little too finicky for most.
 

inf64

Diamond Member
Mar 11, 2011
3,764
4,223
136
You're saying K10h's FPU can also do 2x128-bit MUL/ADDs?
Yes it can,each core's FP unit can do one 128bit MUL and one 128bit ADD (but doesn't have the flexibility to do same ops as BD).
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
jhu: Here's the result running the cpu benchmark for all cores.
CPU time used: kernel 0.17 seconds, user 1356.30 seconds, total 1356.48 seconds.
Elapsed time 174.83 seconds, CPU vs elapsed time ratio 7.76.
Render averaged 1499.43 PPS (193.25 PPS CPU time) over 262144 pixels.
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
jhu: Sorry, here's the entire result screen that may help.
Rays: 2455688 Saved: 10641 Max Level: 12/12
Ray->Shape Intersection Tests Succeeded Percentage
Box 52585606 20495394 38.98
Cone/Cylinder 65209493 13371640 20.51
CSG Intersection 203674976 98265874 48.25
CSG Merge 562356 74996 13.34
Fractal 1096755 206892 18.86
Height Field 1659526 218094 13.14
Height Field Box 1659526 1350152 81.36
Height Field Triangle 7018546 225546 3.21
Height Field Block 11022770 3424320 31.07
Height Field Cell 45914222 3865128 8.42
Isosurface 21632972 1031566 4.77
Isosurface Container 21633751 21633746 100.00
Isosurface Cache 278567 12854 4.61
Mesh 1062303 126139 11.87
Plane 145903634 1545992 1.06
Sphere 296826018 120339577 40.54
Superellipsoid 385543 115607 29.99
Torus 3219850 778862 24.19
Torus Bound 3219850 894416 27.78
True Type Font 519157 171610 33.06
Clipping Object 3476096 1693119 48.71
Bounding Box 1110878742 216559051 19.49
Isosurface roots: 21615519
Function VM calls: 288094086
Crackle Cache Queries: 830938
Crackle Cache Hits: 824848 ( 99 percent)
Roots tested: 894416 eliminated: 550203
Media Intervals: 61476830 Media Samples: 554418798 (9.02)
Shadow Ray Tests: 148392465 Succeeded: 86368325
Shadow Cache Hits: 4933815
Reflected Rays: 375499 Total Internal: 2
Refracted Rays: 262049
Transmitted Rays: 637145
Number of photons shot: 36326
Surface photons stored: 13877
Gather function called: 1380738
Peak memory used: 65114112 bytes
Render Time:
Photon Time: 0 hours 0 minutes 2 seconds (2.449 seconds)
using 11 thread(s) with 2.791 CPU-seconds total
Radiosity Time: No radiosity
Trace Time: 0 hours 2 minutes 50 seconds (170.976 seconds)
using 8 thread(s) with 1353.040 CPU-seconds total
POV-Ray finished
-
CPU time used: kernel 0.17 seconds, user 1356.30 seconds, total 1356.48 seconds.
Elapsed time 174.83 seconds, CPU vs elapsed time ratio 7.76.
Render averaged 1499.43 PPS (193.25 PPS CPU time) over 262144 pixels.
----------------------------------------------------------------------------
 

jhu

Lifer
Oct 10, 1999
11,918
9
81
jhu: Here's the result running the cpu benchmark for all cores.
CPU time used: kernel 0.17 seconds, user 1356.30 seconds, total 1356.48 seconds.
Elapsed time 174.83 seconds, CPU vs elapsed time ratio 7.76.
Render averaged 1499.43 PPS (193.25 PPS CPU time) over 262144 pixels.

Thanks. What's the processor speed on this?
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |