i5 @ 2.3Ghz??

Wow what gives with this?

 

FSB is OCed, as the pic shows

mine isn't stable that high, but someone made a thread recently about how his is 'on par with the i7' when oc'ed.

edit: found it yay http://forum.notebookreview.com/ali...-not-suffer-decrease-performance-when-oc.html

 

OC and Turbo Boost, multiplier x 14, bus speed 167.1 (166).

 

Wow, so that's the actual clock speed after all that? If that's the case I might as well get the fast track option...

 

My i7 OC'd hits 2.8. Not sure what the point is...

 

@Slickie - Are you using Throttlestop? Love the I7 speed. I really hope one of you guys tries out ARMA 2 to see how it performs.

Best wishes, StevenX

 

Haven't bothered to try it yet. Is there an ARMA 2 trial? I'm up to my neck in new games after that Steam sale a few weeks ago.

 

I think there is a demo somewhere.. problem is the full game patches have really increased gameplay performance. This is where I think the I7 will start to pull ahead of the R1. BW, StevenX

 

Aren't these just turbo boost speeds vs the cpu constantly being at 2.3 or 2.7ghz?

How fast is the cpu when you actually play a game?

 

Honestly I haven't really delved much into tracking CPU performance for games or anything else. Day one I OC'd my system at 166, conducted heavy stress tests the first few days, deemed it stable and have left it that way. Of the games I've played I haven't had any performance issues. Frame rates have been more than acceptable and thus I haven't felt compelled to tweak or investigate much.

 

Do us a favor and run wPrime 32 test with OC and standard.

 

No performance gain whatsoever.

i7 | FSB 166 | TB On | High Performance




i7 | FSB 133 | TB On | High Performance



Oh Nos! Overclocking the i7 is worthless!

Yeah, not so much. Read on if you'd like to see why the results I've posted above are not telling you the truth...

I've been testing for the past two hours and reading up on what others have been reporting here about the R2's CPU throttling and OC issues since I haven't really been following any of that closely. Perhaps this is all common knowledge or has been revealed elsewhere here. If that's the case I certainly couldn't find it via search. A lot of the threads here are a mess and weeding through the chaff is often difficult.

A few observations:

• wPrime vs Prime95x64 vs superpi:
  
  Running wPrime 2.03's benchmarks drops the multiplier from 17x to a range of about 9-12x when my system is OC'd and Turbo Boot enabled. Hence the same scores when OC'd or not. I'm really not sure why/how it's doing this, but it it's easy to see that it is. It also has a bugged version of CPU-Z that isn't compatible with the R2 processors, which if not killed will eat up about 25% of your CPU. People need to STOP using wPrime for benchmark comparisons until it supports Core iX processors properly.
  
  EDIT: It turns out that wPrime is the only one that's utilizing all 4 threads and the drop in multiplier is due to turbo boost. See here
  
  The benchmark routine that Prime95x64 (v25.11.1.0) uses on the other hand does not do this. It maintains multipliers as you'd expect them to be under various configurations. It properly benches the Core iX processors and clearly shows the impact of performance between OC'd vs not OC'd, TB On vs TB Off and Balanced vs High Performance.
  
  superpi, like Prime95 works as well. It has the added benefit of allowing you to run a shorter routine - the 2M calculation is comparable to wPrime's run time. Prime95's report is much more verbose and not nearly as easy to parse the differences from report to report.

• 
  Using HWiNFO32 (v3.55-815) you can clearly see what's going on as you run the superpi benchmark routines with various configurations of OC/TB/Power Plans. The following are results from running superpi's 2M calculation.
  
  i7 | FSB 166 | TB On | High Performance 37.269s
  Both cores maintain 17x (2.8GHz) prior to starting the calculation.
  They both maintain 17x (2.8GHz) throughout the calculation.
  
  i7 | FSB 166 | TB On | Balanced 38.844s
  Both cores at 6x (1GHz) prior to starting the calculation.
  Core0 fluctuates between 14x (2.2GHz) and 17x (2.8GHz) throughout the calculation.
  Core1 maintains 14x (2.2GHz) throughout the calculation.
  
  i7 | FSB 166 | TB Off | High Performance 64.428s
  Both cores maintain 9x (1.5GHz) prior to starting the calculation.
  Both cores maintain 9x (1.5GHz) throughout the calculation.
  
  i7 | FSB 133 | TB On | High Performance 47.580s
  Both cores at 14x (2.2GHz) prior to running the calculation.
  Core0 fluctuates between 12x (1.8GHz) and 14x (2.2GHz) throughout the calculation.
  Core1 maintains 12x 1.8GHz throughout the calculation.
  
  i7 | FSB 133 | TB On | Balanced 47.908s
  Both cores at 7x (~.931GHz) prior to running the calculation.
  Core0 fluctuates between 12x (1.8GHz) and 14x (2.2GHz) throughout the calculation.
  Core1 maintains 12x (1.8GHz) throughout the calculation.
  
  i7 | FSB 133 | TB Off | High Performance 81.339s
  Both cores maintain 9x (1.2GHz) prior to starting the calculation.
  Both cores maintain 9x (1.2GHz) throughout the calculation.

• 
  None of the tests were conducted with any special tweaks to my system beyond the optimizations I made during the first week of ownership for general use. No extra effort was made to turn off services or kill unnecessary processes so that these numbers would be the best possible numbers achievable. This was simply a test to determine the impact of OC, Turbo Boost and Balanced vs High Performance modes have on the R2's CPU performance.
  
  How each of these settings translates to gaming performance has already been revealed by some of the testing others have done already, but again I haven't really be following those endeavors. I haven't done any gaming benchmarks myself, because frankly none of the games I'm playing on the R2 have performed poorly. That and I'm generally lazy. The reason I bothered with all of this was because I was perplexed by the zero gain I saw with wPrime and I didn't buy the fact that OC didn't matter.

 

From what I gather, throttlestop only works with mobile C2D extreme CPUs. It's only apparent value for everyone else is to monitor the current state of core speed and the multiplier. Am I missing something?