|Official| Smiles Benchmark 2.0 Ranking Thread! (new version :P)

Ill try with my 2600 too, just to be sure that your results are in line with what can be expected with others above 4 thread too.
 
getting up there with that score.
Untitled-98.jpg
 
Just installed the core i7 and did the benchmark, the heavy did a 120ish time, the light did a little over 19 seconds, at stock clocks...i have a feeling it would be quite a bit faster OCed but i do not want to go that route, it runs quite cool right now, usually staying under 50C at all times.
 
the heavy don`t means it will use all threads 100%, it are only 4 threads...
I`ll make the name `long benchmark`, that what I ment.
And very weird from the 3770k because vistakids dad owns my stock 2600k with his stock 3770...
maybe the cpu is damaged?
And yes, the i5 will not be worse then the i7 in 4threaded. Because the i7 will use only 4threads out of 8. But in 8threaded, i7 wins with a big difference...
 
Are you illerate. The issue is not with it not using just four threads. have you tested it with a core usage reader? It jumps around threads, which means it will not get an accurate result. That was proven my turning HT off resulting in only 4 threads to use and getting a much better result.

For your program to be taken seriously, you need to code it in a way that locks the 4 core to the four cores it is using when it starts the test.

8 thread is fine.
 
I'll get a new leaderboard up today at some point. I agree with the comments about the i5, I just ran the 4-threaded bench again on my i5 and I got 96 seconds with it overclocked to 4.3, however with the overclock removed I got around 120 seconds I think.

Smile, remember my Dad has 32GB of RAM, and a 3770 at stock will be a tiny bit faster than a 2600K at stock. :P

Still not sure what's up with the Bulldozer chips.
 
Its not just bulldozer. If your running a 8 thread HT chip, you get much better results to disable it. The app is the issue, not the chips I am pretty sure of it.
 
Tempted to remove the 4-threaded benchmark and just stick with the single 8-threaded benchmark if that is working better for everybody. You said the 8-threaded bench was fine on your 2600K?
 
not really. At 4.5 I beat a 5GHz 3770k, which is not possible at all. no where near with him having a 500MHz lead. Both are working, just not teh greatest. 4 thread jumps around. Turning off HT solves it. I guess the 6 core Intels and 6 and up AMDs can disable cores for that test. Its not hard.

8 thread just seems off. Unless I got a really strong chip, or his 3770k is set up incorrectly or somehow defective, which I doubt both of.
 
wolfeking said:
not really. At 4.5 I beat a 5GHz 3770k, which is not possible at all. no where near with him having a 500MHz lead. Both are working, just not teh greatest. 4 thread jumps around. Turning off HT solves it. I guess the 6 core Intels and 6 and up AMDs can disable cores for that test. Its not hard.

8 thread just seems off. Unless I got a really strong chip, or his 3770k is set up incorrectly or somehow defective, which I doubt both of.
Click to expand...

or just an unstable OC... the 3th gen seems to have some heat problems.

yeah, I saw the issue now...
I`m looking for some code to let it run on 4 threads. I found one how to run it on 4 cores, but not everyone has 4 cores... thats the problem xD

But I maybe gonna delete the 4threaded and make other 3 in future:
- short 8threaded
- Cpu-Breaker (16)
- Xtreme bench (50)

not sure the Xtreme will go on every pc :P
 
^ Unstable OC makes code that may have a defect more plasuable to have a math error, we probably are seeing math errors due to OCing. It just is not making the program crash.
 
With mine the 8 threaded works fine. All 8 cores jump up to 100% and stay there through the benchmark. Not sure what the problem is with different architectures of Intel. The 4 threaded one is using 6 cores and they jump randomly from 30 to 80%.
 
they run hotter in OC from what I have herd. I was going to go with a 3770k when I upgraded, but I could not see the extra for almost no gain in performance.

I agree, lets wait for another.

OT a bit, but what settings are you using for 4.9 @ 108? I tried 108 * 43 for mine @ every volt setting between 1.250 and 1.400 and none of them would boot. :(. I assume your running 45 x 108, but @ what volts?
 
wolfeking said:
they run hotter in OC from what I have herd. I was going to go with a 3770k when I upgraded, but I could not see the extra for almost no gain in performance.

I agree, lets wait for another.

OT a bit, but what settings are you using for 4.9 @ 108? I tried 108 * 43 for mine @ every volt setting between 1.250 and 1.400 and none of them would boot. :(. I assume your running 45 x 108, but @ what volts?
Click to expand...

yeah, that was the max I could get under 1.4v xD
It`s my record OC, because I still use the stock fan cooler :P

what you mean OT?
 
OT means off topic.

And what are you saying? That you don't run that all the time? What volts were you at when you did get it running that?
 
You must log in or register to reply here.
Back
Top