What an idiot, he clearly doesn't have a clue... 😉 plus both of them have two names... damn... look, you know way more about this stuff than I do, so whatever works for you... nice one...
Anyway, my rig just crashed on me—just kept rebooting. I'm guessing it's because of the meme, so I dialed it back to 5-6-6-20, and now we'll see how it goes... what's your take?
I hit a wall at 290x11=3190 using an FSB of 1160 and a 2:3 memory ratio, running at 870 MHz with 5-5-5-18 2T timings. To get there, I had to bump the vcore up to 1.45V.
Trying 300x11, or even 295x11, just wouldn't work even if I pushed to 1.50V. I didn't push any further because there's no way I'd run those voltages 24/7 anyway.
The auto timings were sitting at 5-6-6-20 2T, so I tightened them down to 5-5-5-18 2T, but that's the limit.
CPU-Z shows vcore at 1.42V, CPU-ID says 1.424V, even though the BIOS is set to 1.45V.
Temperatures under Orthos: iTAT was 68C, CoreTemp showed 74C, and CPU-ID clocked in at 72C.
Once I max out the iTAT, it pushes the temp right up to 70C and won't go any higher... and that’s all with a TT Blue Orb II, which tops out at 1620 RPM (idling around 900).
Honestly, I'm satisfied. For a CPU that cost about $100, I think this is pretty decent.
Next up in the coming days: the GPU. A Leadtek WinFast (Gwendolyn 😉) 8800 GTS/320MB... let's see how that one performs.
Hey, wiredmarlin5, quick question for you... have you heard about the BSEL mod? I'm curious what actually happens to the CPU and the motherboard when you force a processor to run at 1333—I've seen people on the forums doing this instead of traditional overclocking, basically forcing it to 1333 and then just doing a tiny bit of extra overclocking from there. What does that do to the heat levels?
Also, hey you jerk, which is actually "healthier" for the chip and the board: standard overclocking or doing the BSEL mod? I'm not saying I'm actually going to do it, I'm just wondering if you know... or if anyone else here does?
e2200 + tt blue orb II, p35-ds3, 2x 2 gb kingmax/800 8800gts/320, t166/400gb, xilence redwing 420W
clock: 280x11=3.08, fsb=1120, mema 2:3 (840 mhz) 5-6-6-19 2T; temp tat i orthos do 65C
Didn't even touch the voltage or mess with any other settings...
I've been running it like this for over a month now, and honestly, the rig is on for at least 12 hours a day, with about 6 being heavy usage... Just wanted to see if I could push it to 3.2 GHz, just out of curiosity... After messing around for a few hours:
It won't hit 290x11, and even dropping the multiplier to 10 (300x10) or 9 (333x9) doesn't help... nothing works. So, I guess I'd need to bump up the voltage to get more out of it. What I'm wondering is, would it be okay to run it 24/7 with increased voltage? Especially since this TT Blue Orb II cooler isn't exactly top-tier...
One more thing—why is it that the e2200 won't handle a 300 or 333 FSB, when I previously ran an e4300 at 300x9 24/7 without issues? Even at 333x9, the e4300 was stable, even if Orthos showed temps climbing over 75C because of a crappy heatsink. It's the exact same setup, just a different CPU...
Heh... you can't even compare them. That old rig probably doesn't have much overclocking headroom anyway, and even if you did squeeze out some extra speed, it wouldn't mean a thing for a modern build. It just doesn't work like that.
If you want to overclock a new setup without making a mess of it, go do some actual homework. Look at what people have done on this forum or over at PC Gamer—there are plenty of guides out there. Get yourself some proper benchmarking software and temperature monitors. Once you've actually tested your components and know exactly what your idle and load temps look like—and once you've actually studied the BIOS to see which parameters you can tweak—then maybe come back and talk. 😉
That said, with a board like that, you should be able to hit 3GHz pretty easily, even with the stock cooler. But hey, nothing's guaranteed until you actually try it.
Of course, if you’re an expert at overclocking, you’ll want to tweak everything from the voltages to the latency settings yourself. But if this is your first time messing with it, you’re better off leaving it on auto. Since the board can handle its own adjustments anyway, you won't have to worry about accidentally bricking something... plus, if they push the FSB to 333—which, let's face it, they probably will with this specific board—you’ll just end up with that classic 800 😉 meme.
George Ross41 said:Alright, so I'm running an HP setup and at 100% load, my temps are sitting at about 56°C on core 0 and 57°C on core 1. Can I push this further, or should I just leave it alone since I'm still using a stock cooler?
The Gigabyte P35-DS3 (which is the board I was actually talking about) handles SPD on its own when you leave it on Auto. I know because I’m running it right now... if you set it to something like 9x300, the BIOS will just tell you at POST that the memory is set to 2.4. So, that's my experience with this specific board; I can't speak for any others. Also, it's pretty hard to brick this thing unless you start messing wildly with voltages. If you push the settings too far, the machine just shuts down—just turn it back on and load the defaults. That's the "safe" feature with Gigabyte. And don't freak out when the PC shuts itself off during boot. Just sit tight, don't touch anything, and it'll kick back on by itself.
If the temps stay under 65 after running Orthos for several hours, or if it doesn't cross 65 during a TAT, I’d just leave it alone. It’s pretty unlikely you'll ever push the CPU as hard as those stress tests do during actual work or gaming sessions... but that's just my take. Besides, I think the TDP for the E4400 is around 72 or 75 C... check Intel's site to be sure.
Check it against Intel’s TAT... and honestly, maybe it isn't even weird. Maybe the CPU just isn't getting hot enough to trigger the fan speed increase? Then once you crank it up, it finally hits the threshold and the fan kicks into high gear?
I’d just run it at 300x9. You’d have to manually set the memory to 2.4, which puts you at 720—honestly, that seems fine. If that holds up, then maybe try 333x9 to hit 3 GHz with the 2.4 memory for 800... provided your temps don't spike too hard. That would be killer... 😉
George Ross41 said:My AMD 6000 is currently clocked at 3275.33 MHz (218.35 x 15.0) and my core temps are sitting right here: Core #0= 37c Core #1 =36-37 C
What about temperatures under load? Like running Orthos?
Looking at those idle temps, I think you could push it even further...
Forget about latency for a second. Just let the motherboard handle the Democratic Party settings on its own and focus on playing around with the multiplier and FSB... 😉 We can worry about the RAM later...
There’s no need to wait a few days; nothing is going to change. Your board is solid—you could definitely bump the FSB up to at least 333 without any drama, maybe even higher. Just mess with the multiplier a bit to find that sweet spot where performance is high but things aren't overheating...
And look, I use Prime95 just to see if my temps are going to go off the rails. If the temperature stabilizes for a few minutes and stays within a range I can live with, then I know I'm good. Regardless, once you settle on some final settings that actually satisfy you, you still need to run Ortho or SuperPI for a few hours to make sure everything is truly stable...
Kyle Sanchez4 said:Don't touch the latencies. You won't see any difference anyway. By the way, is it just me, or am I the only one running an E4500 at 2.93GHz on this absolute piece of junk motherboard? 🤣
😁 Honestly, from what I've seen on most forums, people rarely bother overclocking with boards like that... so I'm leaning towards thinking it's your fault. 🙏
Running it 24/7 like you’re doing now should be fine... honestly, even though I’m no expert at overclocking, my two cents is that you could probably squeeze more out of the stock settings. Getting those Prime95 or Orthus scores to actually reflect real-world usage is a struggle in just about any workload...
Once you bump it up by maybe 30%—which I think is totally safe and reasonable, provided you keep an eye on the temps since every chip is different—try loading up one of your heavy games. You'll notice the difference in both smoothness and raw speed immediately... your GPU will finally have some room to breathe...
Just download Intel's TAT and check those temps. You can actually use it to stress the CPU, too—it pushes the processor harder than Orthos does, so you won't be sitting there waiting forever... Personally, I think you should try running it with stock settings first. Look, if you haven't messed with the voltages, it isn't going to hit critical temperatures while idling. Just boot up, run TAT, and put a load on it... If you see things climbing past 65°C (and honestly, nothing bad happens if it hits 70°C for a second), then just kill the load...
Then you'll know for sure you need a better cooler...