• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Yet another homegrown CPU benchmark.

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
DaWiper

DaWiper

Image Compare Man
​
Joined
Nov 15, 2003
Location
Norway
Here is some stuff I made in Blitz Basic the other day. I've tested it on some various AMD rigs...
compare1.jpg


How low will you get?

DWBenchmark(405KB)

It might look like the benchmark doesn't care about fsb,mem timings and cache. Mhz is all that count's. AFAIK no hyperthreading is supported.
 
My current sig rig score:

7.0 Secs, Addition floating point (0><1) 100.000.000 times
7.113 Secs, Subtract floating point (0><1) 100.000.000 times
7.217 Secs, Multiply floating point (0><1) 100.000.000 times
7.565 Secs, Divide floating point (0><1) 100.000.000 times
7.0 Secs, Addition integer (1><10) 100.000.000 times
6.999 Secs, Subtract integer (1><10) 100.000.000 times
6.999 Secs, Multiply integer (1><10) 100.000.000 times
7.551 Secs, Divide integer (1><10) 100.000.000 times
 
Mine:
4.89 Secs, Addition floating point (0><1) 100.000.000 times
4.985 Secs, Subtract floating point (0><1) 100.000.000 times
4.998 Secs, Multiply floating point (0><1) 100.000.000 times
5.406 Secs, Divide floating point (0><1) 100.000.000 times
4.916 Secs, Addition integer (1><10) 100.000.000 times
4.917 Secs, Subtract integer (1><10) 100.000.000 times
4.915 Secs, Multiply integer (1><10) 100.000.000 times
5.128 Secs, Divide integer (1><10) 100.000.000 times
 
Wow! Nice, however that makes me believe even more that the benchmark is more about mhz than anything else...
Any Intel guys out there?
 
4.532 Secs, Addition floating point (0><1) 100.000.000 times
4.53 Secs, Subtract floating point (0><1) 100.000.000 times
4.537 Secs, Multiply floating point (0><1) 100.000.000 times
37.365 Secs, Divide floating point (0><1) 100.000.000 times
4.561 Secs, Addition integer (1><10) 100.000.000 times
4.557 Secs, Subtract integer (1><10) 100.000.000 times
36.635 Secs, Multiply integer (1><10) 100.000.000 times
4.848 Secs, Divide integer (1><10) 100.000.000 times

Looks like there is some kind of problem with a couple of the tests on a P4.
 
DaWiper said:
Wow! Nice, however that makes me believe even more that the benchmark is more about mhz than anything else...
Any Intel guys out there?
Click to expand...
Yeah, raw IPC is definitely the ticket here. With that said, there should be plenty of mobile Barton owners that should be able to humble me handily.
 
here's mine.

tests run @ 2100mhz.

6.219 Secs, Addition floating point (0><1) 100.000.000 times
6.269 Secs, Subtract floating point (0><1) 100.000.000 times
6.381 Secs, Multiply floating point (0><1) 100.000.000 times
6.686 Secs, Divide floating point (0><1) 100.000.000 times
6.187 Secs, Addition integer (1><10) 100.000.000 times
6.185 Secs, Subtract integer (1><10) 100.000.000 times
6.186 Secs, Multiply integer (1><10) 100.000.000 times
6.672 Secs, Divide integer (1><10) 100.000.000 times

Tests run @ stock (2Ghz)

6.547 Secs, Addition floating point (0><1) 100.000.000 times
6.594 Secs, Subtract floating point (0><1) 100.000.000 times
6.707 Secs, Multiply floating point (0><1) 100.000.000 times
7.028 Secs, Divide floating point (0><1) 100.000.000 times
6.508 Secs, Addition integer (1><10) 100.000.000 times
6.501 Secs, Subtract integer (1><10) 100.000.000 times
6.503 Secs, Multiply integer (1><10) 100.000.000 times
7.089 Secs, Divide integer (1><10) 100.000.000 times
 
Last edited:
larva said:
Looks like there is some kind of problem with a couple of the tests on a P4.
Click to expand...

Yep! It looks like it. Can't figure out what it could be. The code for the benchmark is extremly simple, but if the compiler screws up it's beyond my capabilities. Let's se some more p4's...
 
I will add the most interesting results in the chart(first post).
 
I just benched again @ 2.3, just for any interest out there

5.717 Secs, Addition floating point (0><1) 100.000.000 times
5.785 Secs, Subtract floating point (0><1) 100.000.000 times
5.871 Secs, Multiply floating point (0><1) 100.000.000 times
6.255 Secs, Divide floating point (0><1) 100.000.000 times
5.746 Secs, Addition integer (1><10) 100.000.000 times
5.724 Secs, Subtract integer (1><10) 100.000.000 times
5.693 Secs, Multiply integer (1><10) 100.000.000 times
6.152 Secs, Divide integer (1><10) 100.000.000 times

I quite like this Benchmark program, thanx DaWiper:)
 
PCluva said:
I just benched again @ 2.3, just for any interest out there

5.717 Secs, Addition floating point (0><1) 100.000.000 times
5.785 Secs, Subtract floating point (0><1) 100.000.000 times
5.871 Secs, Multiply floating point (0><1) 100.000.000 times
6.255 Secs, Divide floating point (0><1) 100.000.000 times
5.746 Secs, Addition integer (1><10) 100.000.000 times
5.724 Secs, Subtract integer (1><10) 100.000.000 times
5.693 Secs, Multiply integer (1><10) 100.000.000 times
6.152 Secs, Divide integer (1><10) 100.000.000 times

I quite like this Benchmark program, thanx DaWiper:)
Click to expand...
My system @190*12=2.28Ghz still beats yours... strange. You should beat me:
5.634 Secs, Addition floating point (0><1) 100.000.000 times
5.634 Secs, Subtract floating point (0><1) 100.000.000 times
5.699 Secs, Multiply floating point (0><1) 100.000.000 times
6.305 Secs, Divide floating point (0><1) 100.000.000 times
5.643 Secs, Addition integer (1><10) 100.000.000 times
5.697 Secs, Subtract integer (1><10) 100.000.000 times
5.811 Secs, Multiply integer (1><10) 100.000.000 times
6.082 Secs, Divide integer (1><10) 100.000.000 times

Maybe I just had a very good day...
 
5.875 Secs, Addition floating point (0><1) 100.000.000 times
5.868 Secs, Subtract floating point (0><1) 100.000.000 times
5.861 Secs, Multiply floating point (0><1) 100.000.000 times
6.333 Secs, Divide floating point (0><1) 100.000.000 times
5.863 Secs, Addition integer (1><10) 100.000.000 times
5.918 Secs, Subtract integer (1><10) 100.000.000 times
6.037 Secs, Multiply integer (1><10) 100.000.000 times
6.321 Secs, Divide integer (1><10) 100.000.000 times

not bad for 2700+ overclock :)
 
Captain Newbie said:
...Hmmmmm... to the results from the P4.

Methinks that the multiply/divide unit on the P4 is structured differently, or addressed by instructions differently. Could you compile it on an Intel platform?
Click to expand...

As I said I don't understand why there should be a difference, but I'll look into compiling it on a P-based rig... Don't think it would matter though.

I can post the source for your evaluation. It's written in Blitz Basic, but it uses mostley standard basic commands...
 
Last edited:
4.886 Secs, Addition floating point (0><1) 100.000.000 times
4.893 Secs, Subtract floating point (0><1) 100.000.000 times
4.884 Secs, Multiply floating point (0><1) 100.000.000 times
5.283 Secs, Divide floating point (0><1) 100.000.000 times
4.92 Secs, Addition integer (1><10) 100.000.000 times
4.988 Secs, Subtract integer (1><10) 100.000.000 times
5.08 Secs, Multiply integer (1><10) 100.000.000 times
5.338 Secs, Divide integer (1><10) 100.000.000 times

AXP 1700+ @ 215x12=2580, 2vcore, 1.92v actual.

Let's say that "stable" would not be the word to describe these settings.

Larva's got me beat by a few tenths of a second, maybe if I drop some ice cubes in my t-line I'll have a shot at him. :)
 
These results are from the Mobile Barton computer in my sig.

4.649 Secs, Addition floating point (0><1) 100.000.000 times
4.734 Secs, Subtract floating point (0><1) 100.000.000 times
4.839 Secs, Multiply floating point (0><1) 100.000.000 times
5.068 Secs, Divide floating point (0><1) 100.000.000 times
4.688 Secs, Addition integer (1><10) 100.000.000 times
4.704 Secs, Subtract integer (1><10) 100.000.000 times
4.704 Secs, Multiply integer (1><10) 100.000.000 times
5.066 Secs, Divide integer (1><10) 100.000.000 times
 
looks like larva's ~4,5 average(not counting the two strange results) is hard to beat. I know I can't touch it... :)
 
I just realized I was clocked at 3.33GHz when I ran the previous pass. I clocked down to 3.33 a couple of weeks ago to generate some numbers to compare to some 3.33GHz numbers from a previous configuration (not like the difference is noticeable in real use). Here are the 3.42GHz numbers:

4.442 Secs, Addition floating point (0><1) 100.000.000 times
4.476 Secs, Subtract floating point (0><1) 100.000.000 times
4.475 Secs, Multiply floating point (0><1) 100.000.000 times
36.504 Secs, Divide floating point (0><1) 100.000.000 times
4.46 Secs, Addition integer (1><10) 100.000.000 times
4.63 Secs, Subtract integer (1><10) 100.000.000 times
35.899 Secs, Multiply integer (1><10) 100.000.000 times
4.743 Secs, Divide integer (1><10) 100.000.000 times

I think the test is not as clockspeed depend as you think. It appears that way when run on different rigs running the same cpu at different clock rates. Small synthetic benchmarks are always this way--as they fit in the cache leaving only clockrate to differentiate the results.
 
4.466 Secs, Addition floating point (0><1) 100.000.000 times
4.42 Secs, Subtract floating point (0><1) 100.000.000 times
4.425 Secs, Multiply floating point (0><1) 100.000.000 times
36.502 Secs, Divide floating point (0><1) 100.000.000 times
4.451 Secs, Addition integer (1><10) 100.000.000 times
4.447 Secs, Subtract integer (1><10) 100.000.000 times
35.35 Secs, Multiply integer (1><10) 100.000.000 times
4.734 Secs, Divide integer (1><10) 100.000.000 times
 
larva said:
I just realized I was clocked at 3.33GHz when I ran the previous pass. I clocked down to 3.33 a couple of weeks ago to generate some numbers to compare to some 3.33GHz numbers from a previous configuration (not like the difference is noticeable in real use). Here are the 3.42GHz numbers:

4.442 Secs, Addition floating point (0><1) 100.000.000 times
4.476 Secs, Subtract floating point (0><1) 100.000.000 times
4.475 Secs, Multiply floating point (0><1) 100.000.000 times
36.504 Secs, Divide floating point (0><1) 100.000.000 times
4.46 Secs, Addition integer (1><10) 100.000.000 times
4.63 Secs, Subtract integer (1><10) 100.000.000 times
35.899 Secs, Multiply integer (1><10) 100.000.000 times
4.743 Secs, Divide integer (1><10) 100.000.000 times

I think the test is not as clockspeed depend as you think. It appears that way when run on different rigs running the same cpu at different clock rates. Small synthetic benchmarks are always this way--as they fit in the cache leaving only clockrate to differentiate the results.
Click to expand...

Scores updated. I'm thinking of writing a combined scrore test using big numbers. I'll look into it tomorrow.

hUMANbEATbOX said:
4.466 Secs, Addition floating point (0><1) 100.000.000 times
4.42 Secs, Subtract floating point (0><1) 100.000.000 times
4.425 Secs, Multiply floating point (0><1) 100.000.000 times
36.502 Secs, Divide floating point (0><1) 100.000.000 times
4.451 Secs, Addition integer (1><10) 100.000.000 times
4.447 Secs, Subtract integer (1><10) 100.000.000 times
35.35 Secs, Multiply integer (1><10) 100.000.000 times
4.734 Secs, Divide integer (1><10) 100.000.000 times
Click to expand...
I was about to add you, but as Larva has reposted his scores he's still Nr.1.
 
You must log in or register to reply here.

Similar threads

S
I can't get a stable system
2 3
Replies
48
Views
3K
yomanew
yomanew
B
Achieving maximum ZEN 3 Memory performance at 64 GB RAM
2
Replies
37
Views
12K
Woomack
W
E
I feel like my build isn't performing at an optimal level - Advice appreciated!
2
Replies
30
Views
2K
EliteACEz
E
P
New RTX 2080 Build significantly under performing, What have I Done Wrong?
2
Replies
30
Views
2K
EarthDog
EarthDog
D
NEED HELP OVERCLOCKING SAMSUNG B-DIE (3200 CL14) ON MY NEW ASUS ROG STRIX X570-E
Replies
11
Views
15K
Woomack
W
Back