1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Overclocking I5 2500 temps.

Discussion in 'Overclocking & Cooling' started by Link108, Mar 30, 2013.

  1. BarbaricSoul

    BarbaricSoul

    Joined:
    Feb 21, 2008
    Messages:
    5,047 (2.16/day)
    Thanks Received:
    1,949
    hehhh, I haven't had any of the instability you people are talking about from OC'ing my BLCK. My 2600k is only capable of a 48 multi, but with a BLCK OC, I can get very close to 5ghz(103.5*48=4.968ghz). OC'ing the BLCK isn't as dangerous as you guys are making it sound.
    Crunching for Team TPU 1 Million points folded for TPU
  2. Aquinus

    Aquinus Resident Wat-man

    Joined:
    Jan 28, 2012
    Messages:
    5,888 (6.56/day)
    Thanks Received:
    1,888
    Location:
    Concord, NH
    That's because you're not unstable yet. It's not like you've increased it much either. When it does become unstable, it's the things you don't want to, like everything on the PCI-E bus and I think it might impact DMI as well, but I'm not completely confident about that.

    Try changing that to 110Mhz and see what happens. Please, don't let me stop you. :p
  3. terrastrife New Member

    Joined:
    Dec 6, 2012
    Messages:
    172 (0.29/day)
    Thanks Received:
    38
    _everything_ wears out over time. Overclocking just reduces the amount of time this takes.

    uncore voltage has been killing hardware faster than base clock overclocking
  4. Aquinus

    Aquinus Resident Wat-man

    Joined:
    Jan 28, 2012
    Messages:
    5,888 (6.56/day)
    Thanks Received:
    1,888
    Location:
    Concord, NH
    My god man. This isn't skt1366. SB/IVB do not have a uncore voltage because they don't have an uncore domain. This changed on SB/IVB from the older 1366 i7s. SB and IVB tie the L3 cache and PCI-E through the bclk and not the uclk with an uncore multiplier. It doesn't work that way anymore and it's important to realize that. A lot more on SB/IVB relies on the SB bclk being stable, more so than the uncore was.

    You can't compare apples and oranges because skt1366 rigs and skt1155 rigs are very different. Heck, it's even pretty different than skt2011 where QPI doesn't actually get used on an X79 board, you're using DMI for communication with the chipset. So all in all, you can't kill a CPU with a voltage on a domain that doesn't exist.

    Now the point I'm making is that making PCI-E unstable is a very bad idea because before the machine crashes your SATA controller could read and write incorrect data from your drives. I would rather that my machine crash because the core was unstable rather than the mechanism that stores all my data. :wtf:

    As I said in my first post when I started arguing with you: do research and stop spewing out bs.
    MxPhenom 216 says thanks.
  5. Link108

    Link108

    Joined:
    Apr 4, 2008
    Messages:
    44 (0.02/day)
    Thanks Received:
    1
    I haven't had any issues with the overclock yet, so I will just wait and see. The BCLK, can hit about 105.5-106 MAX. Anymore than that is bad, there is some tolerance though.
  6. Jetster

    Jetster

    Joined:
    Jan 17, 2010
    Messages:
    4,609 (2.81/day)
    Thanks Received:
    1,804
    Location:
    Oregon
    Go for it
  7. Aquinus

    Aquinus Resident Wat-man

    Joined:
    Jan 28, 2012
    Messages:
    5,888 (6.56/day)
    Thanks Received:
    1,888
    Location:
    Concord, NH
    This will be really limited by the max PCI-E clock accepted by whatever devices that are on your CPU's PCI-E controller. More often than not SATA is what tends to fail first. A number of video cards can handle higher PCI-E clocks but don't really result in better performance. You might get away with disabling devices that are tied to the PCI-E bus. That might give you more flexibility with less devices using it.

    As I said before, the CPU is huge limiting factor. You're not going to see much more than 4Ghz. Heck, I'm surprised that you even got that. :cool:

    With SB-E I can cheat a bit because SB-E has a different clockgen for PCI-E domain and QPI doesn't mind being overclocked so I can handle a nice 129.5x34 for 4.4Ghz... but I digress, we're not overclocking SB-E.

    I'm not sure what other recommendation I can make short of getting a new CPU that overclocks better, but that's only if you really want or need it.
    Link108 says thanks.
  8. marsey99

    marsey99

    Joined:
    Jul 18, 2007
    Messages:
    1,542 (0.60/day)
    Thanks Received:
    289
    you will be ok to 107, past that you can corrupt data on the pci e.
    Link108 says thanks.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page