I got a pretty good overclock of my cpu, almost 1GHz stable. I was a very happy camper, smiling because I was only planning on going up to 3GHz, but got 3.375GHz with a max temp of 66c. That's pretty good considering my stock Intel cooler was going up to 72c loaded at stock speed.
So, here I was smiling away and I thought, hey, maybe I should check my peripherals to make sure they're working properly (I had a problem with a USB 3.0 card on my old ASUS main board when it was slightly overclocked).
So, I did a benchmark on my Patriot Inferno SSD....and it sucked. I was benching at 248MB/s, and it was benching at about 184MB/s after the overclock. I also checked my 1TB WD 7200 RPM hard drive, which was 99MB/s before the overclock and it went down to about 85MB/s. Even my external USB 3.0 hard drive was benching slower.
What gives? I've looked through the BIOS but cannot find anything relating to clocking the SATA ports (are they not just on the PCI bus?).
What should I do to correct the hard drive transfer speeds while the board is overclocked? I made sure during the overclock not to mess with the PCI-E and PCI bus clocks. As you can see in the picture they are still the stock speeds.
As an aside, check out the reading on the +12V. Lol, it goes crazy! Some times it reads 6V (in multiple programs) and sometimes it says 16V. Something crazy going on there. I have brought my digital VOM home from work and tested it. It's a steady 12.06V all the time. The BIOS reads it correctly as well. Strange glitch.
The other thing I thought I would mention, as I have alluded to before, is how well the PCM fans are working out. During this overclock, you can see the FAN1 speed (CPU fan header controlling two Scythe PWM fans) is up to 1865 RMP. I actually saw it go up to just over 1900 RMP. The fans are rated at about 1900 but testing them with my fan controller they both top out at over 2100 RPM, so there's still a ways to go before they reach max. They ARE quite loud when running anything over about 1600 RPM.
Hey, just my 2 bits worth about cable management. Although I agree it's MOSTLY aesthetic, even if there is a ONE degree difference in temp, it's worth it. We're on an overclocking site right now talking about things like which thermal paste will reduce your temps a couple of degrees. Doing what we are all doing EVERY degree is important.
I have to agree though, that if you have a small case and what someone aptly referred to as a "rats nest" of cables, I have no doubt you would be able to see a temperature difference under those conditions. It's not just how much air is entering the case and how much is leaving, it's also the flow. You can develop hot spots that don't get any circulation if there are cables blocking the area. This hot spot may not even show up in your readings if you have no sensor in that area, but it COULD lead to some components running warmer with no way to tell. This is especially true for people with multiple older IDE hard drives and a floppy drive with their wide flat cables.
If anyone doubts this, consider that air operates in a similar fashion to fluids. There are websites out there that specifically deal with fluid flow under varying conditions. Also, for anyone who has taken 2nd year college physics you have also likely taken some fluid dynamics studies. Those studies likely apply as much to air as they do to fluids.
Think about the various shapes they have given the rear ends of cars over the years, and any wind tunnel tests you may have seen in your TV watching or Internet travels. You can easily see how air turbulence can be affected by the smallest changes in the design of the rear of the vehicle. You can actually create areas in the rear where no air is flowing. Unless you have an instrument there to detect this, you'd never even know it.
So, IMHO, even though cable management may be mostly aesthetic, I believe there is also a practical component to it as well.Edited by Mergatroid - 2/17/11 at 6:18pm