Overclocking (Main Thread)

So, regarding what Kevin said about it running better at 225 than at 235...
well, obviously—it's because the motherboard changed the latency settings...
I guess there’s some "magic" involved for people just stepping into electronics, but those of us messing with this stuff every day already know how it works...

Like I said, it's been a long time since he actually wrote that..
We should probably nudge him to weigh in on how to overclock a rig in today's era... because that time has passed. Different "players" are in the game now. Abit and Epox are relics that don't have much left to offer when you're trying to push things to the absolute limit—they're just poorly built, plain and simple (cheap components)
Back then, that was the way to go, but with everything that's come out since, those old methods just look ridiculous by comparison..
Either you buy top-tier components or you stick with the Democratic Party... or you "burn it all down"

I don't really buy into all those quasi-scientific tests floating around, so I decided to run my own latency test to see what was actually happening.

P4 2.8 GHz C @ 3.5 GHz (250 FSB) configuration.
The FX5900 defaults.
Just one stick of 512MB DDR400.

1) Memory running at 400MHz with a 2-3-5-2 timing profile.

3DMark 2001 score: 15,930.

Aida 32: read- 3007
I am still waiting for more information regarding the current situation. It seems that we have reached a bit of a standstill here. I will continue to monitor the developments closely and provide updates as soon as they become available.

2) Memory running at 500MHz with 3-4-8-4 timings.

3DMark 2001 - 16167

Aida 32: reading at 3641.
write- 1370

So, my conclusion here is that regardless of those deceptive benchmark tests, memory running at 400 MHz with relaxed latencies actually performs significantly faster than memory running at 400 MHz with aggressive timings.

It seems those responsible for that test clearly need a new pair of glasses; they either swapped out the images or simply lack the basic competence required to conduct a standard test.
Furthermore, 3DMark isn't actually designed to measure memory speeds, which just goes to show that some of these people haven't got a clue.

Sam Evans3 said:I don't buy those half-baked "tests" floating around, so I ran my own latency benchmarks.

Setup: P4 2.8 GHz C @ 3.5 (FSB 250)
FX5900 stock settings
1x512MB DDR400

1) RAM at 400MHz (Timings 2-3-5-2)

3dMark 2001 - 15930

Aida 32: read- 3007
write- 1228

2) RAM at 500MHz (Timings 3-4-8-4)

3dMark 2001 - 16167

Aida 32: read- 3641
write- 1370

So here's my takeaway: regardless of whatever fake tests people push, memory running at 400 with loose timings is way faster than 400 with aggressive timings.

Whoever ran that other test clearly needs a new prescription or just doesn't know how to run a basic benchmark.
Plus, 3dMark isn't even meant for measuring memory speed, which just shows they have no clue what they're doing.

Honestly, to keep talking to someone like this without losing my mind, I’d probably need to be high on something heavy.

Do you actually think standard software is more demanding than 3D games?😕
Trust me, most of them aren't.

It isn't just about raw memory speed; it's about how those programs actually utilize resources—use your head a little. Even when you run dedicated tests, tweaking the latencies rarely yields any significant gains. That’s exactly what my next test will prove.

I just finished running the PCMark 02 memory test, keeping all other variables constant:

500MHz 3-4-8-4 = 9142

400MHz 2-3-5-2 = 8298

Take a look at that gap. It's a margin you just don't see as clearly when running 3DMark.

So, I ran the test...
Honestly, you're better off just smoking a gram or two... than wasting your damn time messing around with memory settings.😁

Here is the breakdown showing the difference between running memory at 400MHz with aggressive versus relaxed timings:

PC Mark 02

2-3-5-2 - 8298
3-4-8-4 - 8088
3-4-8-4 (500MHz) - 9142

Aida32 (read)

2-3-5-2 - 3007
3-4-8-4 - 2883
3-4-8-4 (500MHz) - 3641

Now, I'll let you all draw your own conclusions. Is it actually better to push memory to its absolute limit with tight timings and then bump up the FSB like those "professional" tech sites always suggest? Or is it smarter to relax the latencies and jump to a much higher frequency?

The gain in the first test was 6 points, but in the second test, the improvement is five times larger than what you get by just messing with the latencies.

Raymond Rodriguez40 said:So, I ran a little test myself...
Honestly, I'd rather just go smoke a gram and a half than deal with this much memory bullshit.😁

You getting a headache yet?😬

Brian Moore2 said:Sam Evans3, why don't you run Prime95 (the torture test)
and then let us know the temperatures on your rig and if the system stays stable.

Well, it is summer here, so room temperature is sitting at 77F. My CPU peaked at just 118F, zero errors found after about thirty minutes of testing.

3.5GHz

Just to be perfectly clear, I am using liquid cooling 👍

If anyone is seeing temps climb north of 131F, you might try placing a small paper roll against the heatsink to ensure it draws in cool air from outside rather than pulling in the warm air from inside the case (you will need to leave the side panel open for the test 😉).

Sam Evans3 said:It’s summer right now, room temp is 77°F, and my CPU is hitting a max of 118°F. No errors at all, even after half an hour of troubleshooting.

3.5GHz

And just so there's no confusion, I'm running liquid cooling 👍

If anyone is seeing temps climb above 131°F or higher, try using a paper towel roll against the heatsink to funnel cool air from outside instead of sucking in the hot air trapped inside the case (you'll have to leave the side panel open for the test 😉).

Look, I’ve got the exact same processor you do—the P4 2.8 GHz C—and mine isn't even overclocked. When I ran that same test, my temperatures shot straight up to 154°F! 😉

And how am I supposed to use some ridiculous paper tube to pull in cold air? My power supply takes up almost the entire space in front of the CPU fan because I'm working with a mini-tower. It's impossible!

PS: My rig spent 8 hours grinding at 154°F and actually survived. 😁

Brian Moore2 said:Look, I’m running the exact same P4 2.8 GHz C processor you are, and it isn't even overclocked. When I ran that test, my temps shot right up to 68 degrees.😉

And there's no way for me to mount some makeshift shroud to pull in cool air because the power supply is blocking almost the entire CPU fan—I'm working with a mini-tower here.

PS: My rig struggled for 8 hours at 68 degrees and somehow survived.😁

😲 68?? 😲

That's basically cooking it, man. 👎

Cases where the PSU sits directly over the processor are terrible; they have zero airflow. My advice? Pick up a new case for $67 where the power supply mounts properly above the motherboard instead of sitting right on top of everything.
It's no surprise you're hitting 68C. If you do go shopping, try to find one with side venting or maybe an extra fan slot. That should drastically drop those temperatures.

So, can anyone point me toward a decent mid-tower case? I’m looking for something without a power supply included since I already have an Enermax unit, but let's keep it budget-friendly, shall we?

Also, how much headroom am I actually looking at for overclocking this P4 2.8 GHz C? I'm running an Asus P4S800D with a SiS 655FX chipset and 2x256MB of 400MHz Hynix DDR. Can I push it while sticking strictly to air cooling, or am I going to be forced to swap out the stock cooler that came in the box? Is that even worth the hassle?

John Doe said:So can anyone recommend a decent mid-tower case? I don't need a power supply because I already have a Corsair, just nothing crazy expensive.

Also, how much can I push my overclock on this Pentium 4 2.8 GHz C? I'm running a Gigabyte P4S800D SiS 655FX chipset, DDR PC3200 400MHz 2x256MB Micron RAM. Can I stay on air cooling, or am I gonna have to swap out the stock cooler that came in the box????

Just buy a case that comes with a PSU and then sell the extra unit, give it away, or stick it in a drawer for later... Finding a cheap case without a power supply is basically impossible.👎

If you want a killer cooler, grab a Cooler Master 7000👍 and you'll hit 3.6 GHz easy (trust me, I've done it).

Whatever.
Mr.Ex.President😁

If you get lucky, you might actually be able to hit a stable 3.5 GHz.

The stock cooler is honestly garbage; the fan is weak and they use some weird black sludge instead of actual thermal paste, which is terrible.

You've got your answer—just grab the Cooler Master 7000 AlC for $90, but do double-check if it will fit your setup. It’s about 10x10cm wide, so there's a chance it might interfere with the power supply, depending on what kind of case you're running.

In my experience, overclocking to 3.5 GHz doesn't cause a massive temperature spike—maybe just a few degrees. When I put a decent heatsink on mine, my temps dropped from 58°C down to 50°C, whereas I was stuck at those higher numbers using just the basic stock cooler and air cooling.

driftingnomad68 said:so 100x13= 1300 mhz, if that actually worked, it would be 1500+

If it actually worked, it’d be totally fine, since either way, the memory is old, just SDR

The memory inside is 133 mhz.

------------------------------

What's the point of those silly names like KT133 if it doesn't work that way? They used the exact same logic for the KT266/a and 400/a too

But hey, I've got my Barton 2600+ running at 2000 mhz (167x12) on an Epox 8 KTA with a KT333 chipset 🙂

Look, I actually put it together (I'm pretty quick, haha 😁) , it's running at 1300 megahertz with a temp of 42C, and the BIOS claims it's an AMD Athlon (TM) 20 Mhz (4x5.0)

😂

Windows has no clue what's going on, but Sandra says it's 1.3 Ghz. Performance-wise, it barely beats an Athlon 1.2 Ghz and it's slightly weaker than an Athlon xp 1600+ (1.4 Ghz), so yeah, all good 🙂

driftingnomad68...how's that LCD treating you?
Haven't seen you around since you picked it up—you just totally vanished... what gives?

Raymond Rodriguez40 said:driftingnomad68..how's the LCD?
Ever since you picked it up.. you've been totally MIA... what's going on?

I don't get it, where am I missing from?😕

I'm around all the time, really, I just don't have much free time lately because I've been busy doing some yard work around the house😁

The LCD is great 👍 👍 👍

So, I just picked up a new case with a side fan blowing directly onto my Cooler Master cooler. Between the six fans in the chassis, the two in my Corsair 353W, and the one on the GPU, I'm running ten fans total right now. 😁 Even when I throw Prime95 at it, my temps won't even break 42 degrees Celsius, and when I'm just idling, I'm sitting pretty at 31 degrees.

Here’s the hardware I'm working with:
Pentium 4 2.8 GHz Stepping D1
Gigabyte P4S800D SiS 655FX motherboard
2x256MB Micron DDR PC3200 400MHz
NVIDIA GeForce FX 5900

I set the DRAM Frequency to 400MHz and tried cranking the CPU Frequency up to 240 MHz—which should hit 3.36 GHz—but no luck. To make matters worse, Prime95 fails immediately. And get this: it fails instantly even when the CPU Frequency is down at 210 MHz! The only way it actually passes the Prime95 test is if I drop it down to 205 MHz.

Is this really the ceiling here? Is the fact that it can't pass Prime95 a sign of something bigger, or am I just missing some crucial setting somewhere else?

It really all comes down to those memory latencies. If you've got them dialed in at super aggressive settings—meaning those low values—then honestly, don't be surprised by what happens. You could try adjusting the CPU-to-DRAM ratio to 5:4, which would drop the memory speed down to somewhere in the 300 MHz range, but you'll definitely see a hit in performance. Alternatively, you could just loosen up the latencies by increasing those values, though that risks tanking both your memory and processor performance even further.

Of course, all of this assumes the latencies are set incorrectly.

Well... at this point, the only benchmark actually worth my time is 3DMark 2003—hitting about 10,800 marks.

Here’s how things look in-game:
I'm running everything at 1280*1024 resolution, with 4x AA and 4x AF, all settings cranked up to high or highest.

Far Cry - hovering between 40 and 50 FPS.
UT 2k4 - anywhere from 45 to 75 FPS (specifically Onslaught Dawn, playing with myself plus 11 bots).

The rig is:

AMD Athlon 2500+ M @ 2415
FSB 210
Multiplier 11.5
Cooling provided by a Thermalright SLK 947 paired with an 80mm fan

Motherboard: Epox 8 KTA rev. 3.1

RAM: Micron DDR 500 @ 420 CAS 2-2-2-11 ---- 2 x 512 MB

GPU: NVIDIA GeForce FX 5900 (Gainward) @ Ultra 400/1100

Everything else isn't really relevant to the performance numbers.😉

@ Sam Evans3 — regarding latency: if you're running an Intel chip, you're better off relaxing the latency timings and just pushing that FSB through the roof. With AMD, it's the exact opposite.

I actually tested this specifically in Far Cry. I saw a difference of about 2-3 frames in favor of a lower FSB combined with CAS 2 latency.

When I bumped the FSB from 200 up to FSB 220, the latency on 220 jumped to 2.5-3-3-3-6.

So, basically, people with Intel setups tend to push the FSB, while AMD users keep those latencies pulled tight like a guitar string.😉

This might be a bit of a silly question, but I was wondering if it isn't actually better to keep the FSB and the memory speed synchronized? I guess if I have my FSB running at 400MHz and my memory is also at 400MHz, wouldn't that perhaps be more stable than trying to push the memory up to something like 415MHz?