+- +-
Say hello if visiting :) by Gecko
11 Jan 2023, 07:43:05 pm

Seti is down again by Mike
09 Aug 2017, 10:02:44 am

Some considerations regarding OpenCL MultiBeam app tuning from algorithm view by Raistmer
11 Dec 2016, 06:30:56 am

Loading APU to the limit: performance considerations by Mike
05 Nov 2016, 06:49:26 am

Better sleep on Windows - new round by Raistmer
26 Aug 2016, 02:02:31 pm

Author Topic: AMD Brisbane 6000+ slower then Brisbane 5000+ ?  (Read 15921 times)

Fomes

  • Guest
AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« on: 19 Feb 2009, 07:14:39 pm »
Hello everyone!

I am running boinc + SETI on following system:
- Windows XP Professional 64 (no change)
- cpu: now AMD Brisbane 6000+, before AMD Brisbane 5000+
- memory: 4 GB (no change)

the question:
With the "old" AMD Brisbane 5000+ I reached about 1000 credits. Why do I reach only about 700 credits with the new 6000+ CPU?? I really don't understand that. Tried everything that came in my mind - checked latest CPU/motherboard drivers, changed optimized Seti@home App from Windows 64 bit AK v8.0 SSE3 (AMD/INTEL) to 2.4V Windows x64 SSE2, skipped from BOINC 6.4.5 back to BOINC 6.2.19 - nothing  :-\

Help!

Gecko_R7

  • Guest
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #1 on: 19 Feb 2009, 07:18:41 pm »
Hello everyone!

I am running boinc + SETI on following system:
- Windows XP Professional 64 (no change)
- cpu: now AMD Brisbane 6000+, before AMD Brisbane 5000+
- memory: 4 GB (no change)

the question:
With the "old" AMD Brisbane 5000+ I reached about 1000 credits. Why do I reach only about 700 credits with the new 6000+ CPU?? I really don't understand that. Tried everything that came in my mind - checked latest CPU/motherboard drivers, changed optimized Seti@home App from Windows 64 bit AK v8.0 SSE3 (AMD/INTEL) to 2.4V Windows x64 SSE2, skipped from BOINC 6.4.5 back to BOINC 6.2.19 - nothing  :-\

Help!

Is the 700 credit result running 2.4V or AKv8?  2.4V of course is much slower than AKv8.

Fomes

  • Guest
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #2 on: 19 Feb 2009, 07:43:36 pm »
Is the 700 credit result running 2.4V or AKv8?  2.4V of course is much slower than AKv8.

Tried both app - about 700 credits both - no remarkable differences.

btw - i remember a thread in here, where the 2.4V SSE2 is recommended for AMD cpu because of the bad implementation of SSE3 on those cpu. Am I wrong?

Offline Raistmer

  • Working Code Wizard
  • Volunteer Developer
  • Knight who says 'Ni!'
  • *****
  • Posts: 14349
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #3 on: 19 Feb 2009, 07:50:09 pm »
Is the 700 credit result running 2.4V or AKv8?  2.4V of course is much slower than AKv8.

Tried both app - about 700 credits both - no remarkable differences.

btw - i remember a thread in here, where the 2.4V SSE2 is recommended for AMD cpu because of the bad implementation of SSE3 on those cpu. Am I wrong?
Probably you are wrong.
Better to use x86 SSE2 AK_v8 on x64 windows for AMD.

Oops, bad memory... You are right
KWSN_2.4V_SSE2_MB.exe - on my own Venice as production app for x64 WinServer 2003. (was few years ago...)
(will do test if v8b AMD-specific build faster or not and post here. AMD CPUs are strange beasts... ;) )

ADDON: Hm.... but on my another AMD x64 host, Winchester core,  x86 AK_v8 SSE2 used as production app... maybe just forgot to renew 2.4V ?... will bench again...

ADDON: 2.4V was in x86 BOINC directory for WinXP not used few years already on that host... x64 BOINC (active one) using AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe build (see bench results) but now probably will be upgraded to another build.
« Last Edit: 20 Feb 2009, 03:36:30 am by Raistmer »

Offline Raistmer

  • Working Code Wizard
  • Volunteer Developer
  • Knight who says 'Ni!'
  • *****
  • Posts: 14349
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #4 on: 20 Feb 2009, 03:27:28 am »
Ok, first part of fresh data arrived:

Core - Venice.
Quick timetable

WU : PG0009.wu
KWSN_2.4V_SSE2_MB.exe : 1271.969 secs CPU
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 1052.422 secs CPU
Speedup     : 17.26%
Ratio       : 1.21 x
AK_v8b_win_SSE3_AMD.exe : 1085.469 secs CPU
Speedup     : 14.66%
Ratio       : 1.17 x
AK_v8b_win_x64_SSE3.exe : 1027.438 secs CPU
Speedup     : 19.22%

Ratio       : 1.24 x
KWSN_2.4V_SSE2_MB.exe : 1256.781 secs CPU
Speedup     : 1.19%
Ratio       : 1.01 x

WU : PG0395.wu
KWSN_2.4V_SSE2_MB.exe : 1441.188 secs CPU
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 986.469 secs CPU
Speedup     : 31.55%
Ratio       : 1.46 x
AK_v8b_win_SSE3_AMD.exe : 998.797 secs CPU
Speedup     : 30.70%
Ratio       : 1.44 x
AK_v8b_win_x64_SSE3.exe : 993.453 secs CPU
Speedup     : 31.07%

Ratio       : 1.45 x
KWSN_2.4V_SSE2_MB.exe : 1441.078 secs CPU
Speedup     : 0.01%
Ratio       : 1.00 x

WU : PG0444.wu
KWSN_2.4V_SSE2_MB.exe : 1251.031 secs CPU
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 860.500 secs CPU
Speedup     : 31.22%
Ratio       : 1.45 x
AK_v8b_win_SSE3_AMD.exe : 859.563 secs CPU
Speedup     : 31.29%

Ratio       : 1.46 x
AK_v8b_win_x64_SSE3.exe : 870.453 secs CPU
Speedup     : 30.42%

Ratio       : 1.44 x
KWSN_2.4V_SSE2_MB.exe : 1251.641 secs CPU
Speedup     : -0.05%
Ratio       : 1.00 x

WU : PG1327.wu
KWSN_2.4V_SSE2_MB.exe : 1069.797 secs CPU
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 959.594 secs CPU
Speedup     : 10.30%
Ratio       : 1.11 x
AK_v8b_win_SSE3_AMD.exe : 883.734 secs CPU
Speedup     : 17.39%

Ratio       : 1.21 x
AK_v8b_win_x64_SSE3.exe : 979.391 secs CPU
Speedup     : 8.45%

Ratio       : 1.09 x
KWSN_2.4V_SSE2_MB.exe : 1071.000 secs CPU
Speedup     : -0.11%
Ratio       : 1.00 x

That is, for AMD with SSE3 support only 2 builds worth to try on x64 system (form tested) SSE3_AMD and x64 SSE3.
2.4V not an option.
Next run will compare SSE2 and SSE3 AK_v8 on this host.

(and I understood from where KWSN SSE2 2.4V came on my host, see edit to earlier post)
« Last Edit: 20 Feb 2009, 03:33:46 am by Raistmer »

Offline Raistmer

  • Working Code Wizard
  • Volunteer Developer
  • Knight who says 'Ni!'
  • *****
  • Posts: 14349
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #5 on: 20 Feb 2009, 03:57:01 am »
And here fresh data from SSE2-only AMD (x64 support presents)

Quick timetable
 
WU : PG0009.wu
AK_v8_win_SSE2.exe : 1017.453 secs CPU
AK_v8_win_SSE2.exe : 1019.781 secs CPU
Speedup     : -0.23%
Ratio       : 1.00 x
AK_v8b_win_SSE2.exe : 986.672 secs CPU
Speedup     : 3.03%

Ratio       : 1.03 x
KWSN_2.4_SSE2_IPP_Ben-Joe.exe : 1415.344 secs CPU
Speedup     : -39.11%
Ratio       : 0.72 x
KWSN_2.4V_SSE2_MB.exe : 1157.844 secs CPU
Speedup     : -13.80%
Ratio       : 0.88 x
KWSN_2.4V_SSE2_MB_x64.exe : 1110.438 secs CPU
Speedup     : -9.14%
Ratio       : 0.92 x
 
WU : PG0395.wu
AK_v8_win_SSE2.exe : 942.578 secs CPU
AK_v8_win_SSE2.exe : 942.734 secs CPU
Speedup     : -0.02%
Ratio       : 1.00 x
AK_v8b_win_SSE2.exe : 940.516 secs CPU
Speedup     : 0.22%
Ratio       : 1.00 x
KWSN_2.4_SSE2_IPP_Ben-Joe.exe : 1733.000 secs CPU
Speedup     : -83.86%
Ratio       : 0.54 x
KWSN_2.4V_SSE2_MB.exe : 1322.094 secs CPU
Speedup     : -40.26%
Ratio       : 0.71 x
KWSN_2.4V_SSE2_MB_x64.exe : 1350.125 secs CPU
Speedup     : -43.24%
Ratio       : 0.70 x
 
WU : PG0444.wu
AK_v8_win_SSE2.exe : 806.063 secs CPU
AK_v8_win_SSE2.exe : 803.141 secs CPU
Speedup     : 0.36%
Ratio       : 1.00 x
AK_v8b_win_SSE2.exe : 802.750 secs CPU
Speedup     : 0.41%
Ratio       : 1.00 x
KWSN_2.4_SSE2_IPP_Ben-Joe.exe : 1554.719 secs CPU
Speedup     : -92.88%

Ratio       : 0.52 x
KWSN_2.4V_SSE2_MB.exe : 1144.484 secs CPU
Speedup     : -41.98%
Ratio       : 0.70 x
KWSN_2.4V_SSE2_MB_x64.exe : 1187.859 secs CPU
Speedup     : -47.37%
Ratio       : 0.68 x
 
WU : PG1327.wu
AK_v8_win_SSE2.exe : 820.391 secs CPU
AK_v8_win_SSE2.exe : 819.750 secs CPU
Speedup     : 0.08%
Ratio       : 1.00 x
AK_v8b_win_SSE2.exe : 811.578 secs CPU
Speedup     : 1.07%

Ratio       : 1.01 x
KWSN_2.4_SSE2_IPP_Ben-Joe.exe : 1667.063 secs CPU
Speedup     : -103.20%

Ratio       : 0.49 x
KWSN_2.4V_SSE2_MB.exe : 957.844 secs CPU
Speedup     : -16.75%
Ratio       : 0.86 x
KWSN_2.4V_SSE2_MB_x64.exe : 1009.375 secs CPU
Speedup     : -23.04%
Ratio       : 0.81 x
 
------------
CPU:
Specification      AMD Athlon(tm) 64 Processor 3200+
Codename      Winchester

It clearly shows that older 2.4 builds not an option for this class of CPUs too.

Fomes

  • Guest
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #6 on: 20 Feb 2009, 04:57:15 am »
Thanks a lot! Windows 64 bit AK v8.0 SSE3 (AMD/INTEL) is already running (again) ;)

some last questions (i hope):
Averaged the AK_v8b_win_SSE2.exe seems to be the best choice for AMD 64bit systems (?).
Where can i find the AK_v8b_win_SSE2.exe ?
And am I right - that's a 32bit app? So it's recommended to run the 32bit app on a 64bit system? Should i rather roll back the whole system to 32 bit then? (although BOINC bench proofs a significant speedincrease at 64bit?)

possible configs:

win32 with 32bit app or
win64 with 32bit app or
win64 with 64bit app 

Which one is the best for AMD 64 cpu?

The more i think about AMD and 32 or 64 bit-systems, the more i get confused   :-\
« Last Edit: 20 Feb 2009, 05:32:45 am by Fomes »

Offline Claggy

  • Alpha Tester
  • Knight who says 'Ni!'
  • ***
  • Posts: 3111
    • My computers at Seti Beta
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #7 on: 20 Feb 2009, 06:39:19 am »
Have you used CPU-Z to check that it's clock speed is correct?,
If the Multiplier stayed the same when you changed CPU's you could be underclocking it.

Claggy

Offline Raistmer

  • Working Code Wizard
  • Volunteer Developer
  • Knight who says 'Ni!'
  • *****
  • Posts: 14349
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #8 on: 20 Feb 2009, 07:40:35 am »
Thanks a lot! Windows 64 bit AK v8.0 SSE3 (AMD/INTEL) is already running (again) ;)

some last questions (i hope):
Averaged the AK_v8b_win_SSE2.exe seems to be the best choice for AMD 64bit systems (?).
Where can i find the AK_v8b_win_SSE2.exe ?
And am I right - that's a 32bit app? So it's recommended to run the 32bit app on a 64bit system? Should i rather roll back the whole system to 32 bit then? (although BOINC bench proofs a significant speedincrease at 64bit?)

possible configs:

win32 with 32bit app or
win64 with 32bit app or
win64 with 64bit app 

Which one is the best for AMD 64 cpu?

The more i think about AMD and 32 or 64 bit-systems, the more i get confused   :-\

For SSE3 supporting AMDs data with SSE2 AK_v8b not shown, only for SSE2-only. Your CPU is SSE3-supporting one. So wait a little, SSE2 vs SSE3 for such CPUs will be in next post (in progress now). But before I recived that SSE2 x86 faster indeed than SSE3 x64 on SSE3-capable AMD CPU.

But you should not downgrade to x86 OS. SSE2 x86 is faster on x64 OS than on x86 OS.
So, optimum config should be x86 SSE2 on x64 WIn OS.

Fomes

  • Guest
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #9 on: 20 Feb 2009, 08:19:41 am »
Have you used CPU-Z to check that it's clock speed is correct?,
If the Multiplier stayed the same when you changed CPU's you could be underclocking it.

I did. See attached file.

[attachment deleted by admin]

Fomes

  • Guest
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #10 on: 20 Feb 2009, 08:28:41 am »
For SSE3 supporting AMDs data with SSE2 AK_v8b not shown, only for SSE2-only. Your CPU is SSE3-supporting one. So wait a little, SSE2 vs SSE3 for such CPUs will be in next post (in progress now). But before I recived that SSE2 x86 faster indeed than SSE3 x64 on SSE3-capable AMD CPU.

But you should not downgrade to x86 OS. SSE2 x86 is faster on x64 OS than on x86 OS.
So, optimum config should be x86 SSE2 on x64 WIn OS.

Thanks again! Found somewhere in here a "SSE3_x64_bench"-compressed file and got:

Quick timetable
 
WU : PG0009.wu
AK_v8b_win_SSE3_AMD.exe : 643.344 secs CPU
AK_v8b_win_SSE3_AMD.exe : 645.781 secs CPU
Speedup     : -0.38%
Ratio       : 1.00 x
AK_v8b_win_x64_SSE3.exe : 626.891 secs CPU
Speedup     : 2.56%
Ratio       : 1.03 x
 
WU : PG0395.wu
AK_v8b_win_SSE3_AMD.exe : 609.375 secs CPU
AK_v8b_win_SSE3_AMD.exe : 609.656 secs CPU
Speedup     : -0.05%
Ratio       : 1.00 x
AK_v8b_win_x64_SSE3.exe : 624.438 secs CPU
Speedup     : -2.47%
Ratio       : 0.98 x
 
WU : PG0444.wu
AK_v8b_win_SSE3_AMD.exe : 523.188 secs CPU
AK_v8b_win_SSE3_AMD.exe : 523.219 secs CPU
Speedup     : -0.01%
Ratio       : 1.00 x
AK_v8b_win_x64_SSE3.exe : 549.453 secs CPU
Speedup     : -5.02%
Ratio       : 0.95 x
 
WU : PG1327.wu
AK_v8b_win_SSE3_AMD.exe : 535.000 secs CPU
AK_v8b_win_SSE3_AMD.exe : 537.438 secs CPU
Speedup     : -0.46%
Ratio       : 1.00 x
AK_v8b_win_x64_SSE3.exe : 623.063 secs CPU
Speedup     : -16.46%
Ratio       : 0.86 x

So I changed the app and am now running the AK_v8b_win_SSE3_AMD.exe until i see a faster bench with the SSE2 app.

Offline Raistmer

  • Working Code Wizard
  • Volunteer Developer
  • Knight who says 'Ni!'
  • *****
  • Posts: 14349
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #11 on: 20 Feb 2009, 10:18:32 am »
Quick timetable
 
WU : PG0009.wu
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 1051.406 secs CPU
AK_v8_win_SSE2.exe : 1108.969 secs CPU
Speedup     : -5.47%
Ratio       : 0.95 x
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 1072.594 secs CPU
Speedup     : -2.02%
Ratio       : 0.98 x
AK_v8b_win_SSE2.exe : 1075.891 secs CPU
Speedup     : -2.33%
Ratio       : 0.98 x
AK_v8b_win_SSE3_AMD.exe : 1085.125 secs CPU
Speedup     : -3.21%
Ratio       : 0.97 x
AK_v8b_win_x64_SSE3.exe : 1030.563 secs CPU
Speedup     : 1.98%

Ratio       : 1.02 x
 
WU : PG0395.wu
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 982.906 secs CPU
AK_v8_win_SSE2.exe : 1021.516 secs CPU
Speedup     : -3.93%
Ratio       : 0.96 x
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 984.563 secs CPU
Speedup     : -0.17%
Ratio       : 1.00 x
AK_v8b_win_SSE2.exe : 1018.609 secs CPU
Speedup     : -3.63%
Ratio       : 0.96 x
AK_v8b_win_SSE3_AMD.exe : 996.641 secs CPU
Speedup     : -1.40%
Ratio       : 0.99 x
AK_v8b_win_x64_SSE3.exe : 997.188 secs CPU
Speedup     : -1.45%
Ratio       : 0.99 x
 
WU : PG0444.wu
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 859.938 secs CPU
AK_v8_win_SSE2.exe : 877.016 secs CPU
Speedup     : -1.99%
Ratio       : 0.98 x
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 860.875 secs CPU
Speedup     : -0.11%
Ratio       : 1.00 x
AK_v8b_win_SSE2.exe : 870.516 secs CPU
Speedup     : -1.23%
Ratio       : 0.99 x
AK_v8b_win_SSE3_AMD.exe : 856.547 secs CPU
Speedup     : 0.39%
Ratio       : 1.00 x
AK_v8b_win_x64_SSE3.exe : 869.484 secs CPU
Speedup     : -1.11%
Ratio       : 0.99 x
 
WU : PG1327.wu
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 970.000 secs CPU
AK_v8_win_SSE2.exe : 914.438 secs CPU
Speedup     : 5.73%
Ratio       : 1.06 x
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 891.188 secs CPU
Speedup     : 8.12%
Ratio       : 1.09 x
AK_v8b_win_SSE2.exe : 837.516 secs CPU
Speedup     : 13.66%
Ratio       : 1.16 x
AK_v8b_win_SSE3_AMD.exe : 814.844 secs CPU
Speedup     : 16.00%

Ratio       : 1.19 x
AK_v8b_win_x64_SSE3.exe : 887.750 secs CPU
Speedup     : 8.48%
Ratio       : 1.09 x
 
------------
CPU:
Specification      AMD Athlon(tm) 64 Processor 3200+
Codename      Venice
Core Speed      2080.2 MHz (10.0 x 208.0 MHz)


Again, VHAR results have very big deviation.
The same app differes on 5% from run to run.... Will recheck it once more.

ADDON: more stable results for VHAR:
WU : PG1327.wu
AK_v8b_win_SSE3_AMD.exe : 812.781 secs CPU
AK_v8_win_SSE2.exe : 842.078 secs CPU
Speedup     : -3.60%
Ratio       : 0.97 x
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33.exe : 880.484 secs CPU
Speedup     : -8.33%
Ratio       : 0.92 x
AK_v8b_win_SSE2.exe : 834.063 secs CPU
Speedup     : -2.62%
Ratio       : 0.97 x
AK_v8b_win_SSE3_AMD.exe : 811.547 secs CPU
Speedup     : 0.15%
Ratio       : 1.00 x
AK_v8b_win_x64_SSE3.exe : 884.766 secs CPU
Speedup     : -8.86%
Ratio       : 0.92 x
« Last Edit: 20 Feb 2009, 12:00:14 pm by Raistmer »

Fomes

  • Guest
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #12 on: 20 Feb 2009, 11:15:58 am »
So in three of four cases the AK_v8b_win_SSE3_AMD.exe is a slightly better choice then AK_v8b_win_SSE2.exe.

I think this is the best AMD-config on a XP64-system so far.

Offline Raistmer

  • Working Code Wizard
  • Volunteer Developer
  • Knight who says 'Ni!'
  • *****
  • Posts: 14349
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #13 on: 20 Feb 2009, 11:27:11 am »
So in three of four cases the AK_v8b_win_SSE3_AMD.exe is a slightly better choice then AK_v8b_win_SSE2.exe.

I think this is the best AMD-config on a XP64-system so far.
We usually sum all execution times for PG* set to determine what app the best for whole range. I post prepaed data slightly later too.

Offline Raistmer

  • Working Code Wizard
  • Volunteer Developer
  • Knight who says 'Ni!'
  • *****
  • Posts: 14349
Re: AMD Brisbane 6000+ slower then Brisbane 5000+ ?
« Reply #14 on: 20 Feb 2009, 11:38:59 am »

Raistmer, Venice Core (Target Class AMD Athlon 64 3200+)   PG0009   PG0395   PG0444   PG1327   Total   Speedup   
AK_v8_win_x64_SSE3_AMD_IPP_m7_PGO_XEON_r33   1072.594   984.563   860.875   891.188   3809.22      
AK_v8_win_SSE2   1108.969   1021.516   877.016   914.438   3921.939   -2.96%   0.97
AK_v8b_win_SSE2   1075.891   1018.609   870.516   837.516   3802.532   0.18%   1.00
AK_v8b_win_SSE3_AMD   1085.125   996.641   856.547   814.844   3753.157   1.47%   1.01
AK_v8b_win_x64_SSE3   1030.563   997.188   869.484   887.75   3784.985   0.64%   1.01


It seems this newly build app best choice for x64 AMD systems too.

 

Welcome, Guest.
Please login or register.
 
 
 
Forgot your password?
Members
Total Members: 97
Latest: ToeBee
New This Month: 0
New This Week: 0
New Today: 0
Stats
Total Posts: 59559
Total Topics: 1672
Most Online Today: 4
Most Online Ever: 983
(20 Jan 2020, 03:17:55 pm)
Users Online
Members: 0
Guests: 103
Total: 103
Powered by EzPortal