MIPS rating on a 4800+ AMD x2

alkemyst

No Lifer
Feb 13, 2001
83,769
19
81
This seems high to me and my Opteron 165 at 2.7GHZ (300x9) pushes 15238 Dhrystone ALU.

I do have a higher number of MFLOPS...

just making sure everything is good...in BOINC I put up:

2716 floating point MIPS (Whetstone) per CPU
4921 integer MIPS (Dhrystone) per CPU


 

LOUISSSSS

Diamond Member
Dec 5, 2005
8,771
58
91
everything is okay with ur CPU if u can run prime95 for 24 hours.

no need for such artificial benchmarks
 

alkemyst

No Lifer
Feb 13, 2001
83,769
19
81
Originally posted by: LOUISSSSS
everything is okay with ur CPU if u can run prime95 for 24 hours.

no need for such artificial benchmarks

I am aware of running stable, but that was not my question.
 

Amaroque

Platinum Member
Jan 2, 2005
2,178
0
0
Originally posted by: alkemyst
This seems high to me and my Opteron 165 at 2.7GHZ (300x9) pushes 15238 Dhrystone ALU.

I do have a higher number of MFLOPS...

just making sure everything is good...in BOINC I put up:

2716 floating point MIPS (Whetstone) per CPU
4921 integer MIPS (Dhrystone) per CPU

Are you complaining? I run an X2 at 2800 CPU, and 280 RAM (CPU/10) with a 12x mult, and no reduction in HT link. Of course my Q6600 at 3000 is faster, but what is the problem here? It's a good CPU.
 

alkemyst

No Lifer
Feb 13, 2001
83,769
19
81
Why does everyone assume a poster is hostile/complaining.

I am questioning if Sandra's MIPS value for the 4800+ is accurate.

If someone doesn't know they don't have to reply.
 

CTho9305

Elite Member
Jul 26, 2000
9,214
1
81
A dual-core 2.4GHz Athlon 64 is capable of up to 14.4 billion instructions per second (3 per cycle per core) or 19.2 gigaflops (I think... 2 32-bit FP adds and 2 32-bit FP muls per cycle = 4 per core per cycle = 8 per cycle * 2.4 billion cycles per second). The problem with benchmarks is they're measuring millions of instructions per second on some particular task which may or may not be optimized for any particular CPU. Even if it is optimized, the result still isn't necessarily useful, because optimized code may use fewer more complex instructions (resulting in a lower score) or more less complex instructions (resulting in a higher score) to accomplish a given task. You can't really compare instructions per second results between different applications.
 

piasabird

Lifer
Feb 6, 2002
17,168
60
91
To take full advantage of a high MIPS rating, you have to be engaged in some kind of task which really taxes your CPU's capabilities. The real bottleneck may be the motherboard not the CPU.