Asus A8N-SLI Overclokers

Relion

Senior member
Dec 21, 2004
294
0
0
How safe is it to OC an athlon 64 3200+ from its original 2000Mhz to say 2200Mhz? using retail HSF...and how to do that?
 

Insomniak

Banned
Sep 11, 2003
4,836
0
0
Are you on a Winchester or Newcastle core? As for how, see the overclockers forum. There's a good guide stickied there.
 

Relion

Senior member
Dec 21, 2004
294
0
0
Its a Winchester...it stays cool even loaded (33C) so I thought I could push it a lil...What kind of RAM ur using? Blue line? Why u have that Leadtek at stock?? Use coolbits...I have mine at 580/1.15 all stable...
 

Insomniak

Banned
Sep 11, 2003
4,836
0
0
The Mushkin is Basic Green, and the video card is at stock because I have no need or want to overclock it at this point. It's empty stress on the card. I am well aware of how to OC it when and if I decide to.
 

Relion

Senior member
Dec 21, 2004
294
0
0
Do u have ramsinks on your Mushkin? I have the same mobo vid and memory as you
 

BigGreenMat

Member
Jan 12, 2005
27
0
0
If someone could answer this question I would be much appreciative. I have a 3200+ on an A8N SLI using Mushkin Lvl2 Ver. 2 Black. Which is the Samsung TCCD 2225 RAM. Even after going to Beta Bios 1003.5, loosening RAM timings to 3448, upping RAM voltage to 3.0, upping CPU voltage to 1.58, and lowering the HTT to 4x, I absolutely CANNOT go over 215 on my FSB. No matter what! As soon as I do a Blue screen with some message pops up. I know the CPU is overclockable as I can get it to 2400mhz easily using 4x HTT, 10x multiplier, 5/6 memory divider, and setting HTT to 240. Thus giving me a 200mhz FSB. I am just really concerned about why I can't crack over 215 on the FSB. My PCI-E and PCI lock appear to be functioning perfectly. The only thing I could think is that somehow the SATA is being overclocked. Does that happen or does the front side bus not affect that or is there a lock on it? I am using an Maxtor 300gb 16mb cache SATAII
 

Insomniak

Banned
Sep 11, 2003
4,836
0
0
Originally posted by: Relion
Do u have ramsinks on your Mushkin? I have the same mobo vid and memory as you

No ramsinks on the Mushkin...again, yet. These things may change in the future, but at present this rig can play everything currently out there at high detail, 10x7 or higher, with some AA and AF laid over top of it. There's really no reason to push it harder.

Doom 3 and Far Cry provide a bit of a challenge, but

1) Far Cry sucks, so who cares.
2) Doom 3 was really designed for 2006-2007 hardware, it just got here early.

Anyway, I may bump the RAM and vidcard down the road, but right now I don't see it as needed or wise.
 

Insomniak

Banned
Sep 11, 2003
4,836
0
0
Originally posted by: BigGreenMat
If someone could answer this question I would be much appreciative. I have a 3200+ on an A8N SLI using Mushkin Lvl2 Ver. 2 Black. Which is the Samsung TCCD 2225 RAM. Even after going to Beta Bios 1003.5, loosening RAM timings to 3448, upping RAM voltage to 3.0, upping CPU voltage to 1.58, and lowering the HTT to 4x, I absolutely CANNOT go over 215 on my FSB. No matter what! As soon as I do a Blue screen with some message pops up. I know the CPU is overclockable as I can get it to 2400mhz easily using 4x HTT, 10x multiplier, 5/6 memory divider, and setting HTT to 240. Thus giving me a 200mhz FSB. I am just really concerned about why I can't crack over 215 on the FSB. My PCI-E and PCI lock appear to be functioning perfectly. The only thing I could think is that somehow the SATA is being overclocked. Does that happen or does the front side bus not affect that or is there a lock on it? I am using an Maxtor 300gb 16mb cache SATAII


CPU voltage seems a bit high, to being with. My 3000+ winchester is running it's current speed at 1.5v even. You might also want to try dropping RAM voltage to 2.8ish and starting again. You seem to be feeding your components more power than they really need - although, don't go below the RAM's rated voltage. I'm not sure what it is for those sticks.

If you really want to OC 1:1, the best course of action is just to get some PC4000 or higher and start OCing at a DDR400 divider.

 

BigGreenMat

Member
Jan 12, 2005
27
0
0
Thanks for the advice. I have run them at lower voltages too. I am just using that as an example of giving headroom. Really though these are TCCD chips that run 2225 pretty damn stable at 200 mhz and will run that up to the 215 barrier at the rated 2.7 voltage. Thing is the barrier is rock solid at 215 irregardless of any voltage or timing changes. It leads me to believe that the memory is not the issue. My only guess is the hard drive (which I am getting awfully tempted to RMA to Maxtor, but will play a little longer) or the mobo. I mean heck even value ram can go higher
 

meolsen

Member
Dec 26, 2004
37
0
0
BigGreen - I have the exact same issue on mine. When you say your are running 5/6 memory timings, are you saying that you set the DDR to 333 instead of DDR 400?

I wonder, too, when you do the overclock you describe, do you get appreciably better benchmarks? (i'm running on a 3800 at 2.4ghz, and cannot get it much over 2.55 ghz running the FSB at 210, but then things get a little jiggy.
 

DerKaiser

Senior member
Feb 12, 2002
460
0
0
BigGreenMat - are you running your memory at 1T or 2T. I can't run at 1T over 220 Mhz when I have my memory set to DDR400. Also, using the DDR366 setting (5/6 divider), I can't even boot up at 240 FSB which should only be 200 Mhz on the memory. To get to high FSBs, I have had to go to 2T. I can then get stable up to 250 Mhz FSB. It seems as though the memory speed and timings are not completely independent from the FSB, althought in theory, they should be. Perhaps it has something to do with the on-die memory controller.
 

user1234

Banned
Jul 11, 2004
2,428
0
0
For FSB of 260 mhz, the memory settings allowed by the board are either "1:1" (CPU/9) which would be 260 MHz (520 DDR) or "5:6" (CPU/11), which sets it to 213.5. Theoretically, you should be able to set it to CPU/10 (if the bios had such an option) which would be 235 mhz. But in any case, my HyperX can only do 225 mhz max at CAS 2.5, so I have to set it to CPU/11. In any case, this has negligible effect on performance (less then 1% for 50 mhz DDR).
 

shils

Junior Member
Dec 20, 2004
12
0
0
If you're running the 1003.005 bios, make sure you set the pci lock setting to 33.34. I was having the same problem as you, not being able to get over 215 fsb. The various utilites showed the pci bus @ 33.34 but I would get blue screens and "stuttering" loopback sounds in games. The 1003.005 beta bios with the pci lock cured this machine.

good luck

Asus A8N-SLI MB
AMD Athlon 64+ 3000 (Winchester)
2 - 512 OCZ Platinum EL Rev. 2 PC 3200 DDR Ram (Dual Channel)
2 - XFX 6600 GT PCI Express GPU's
500 watt X Connect P.S.
Western Digital 74 MB SATA Raptor (C drive)
Western Digital 300 MB, SATA (16 MB cache, D drive)
Nec 3500a DVD/RW (E drive)
Toshiba 1912 DVD Rom (F drive)
Thermaltake Big Water liquid CPU cooling
Windows XP-Pro O.S.
 

user1234

Banned
Jul 11, 2004
2,428
0
0
If you hadn't noticed my FSB is at 260 mhz. Only the memory is at 213.5 mhz, and can even get it to 225 at 1:1 with fsb, but then of course the cpu clock is not as high as now when fsg is 260. The PCI bus is locked and shown correctly at 33.34 by clockgen nf4. My system is 100% stable running prime-95 for a few hours and concurrently looping 3dmark01 or 3dmark03.


You are correct that any overclocking must begin with locking the pci bus. btw, you could do it even with the 1002 bios, which doesn't have a pci bus lock option, by setting the fsb to 201 in the bios, and then increasing it using ai-booster - this will lock the pci bus.

Please post your overclocking results, if you will.
 

shils

Junior Member
Dec 20, 2004
12
0
0
I am doing all my settings changes through bios as I'm not really comfortable using the various Windows tweak apps. I was unable to do much of any overclocking until the memory settings ratio thing was figured out by varios users (thank you all) and get a sure pci bus lock. I'm still disappointed with the cpu voltage settings in bios as they seem to have no rhyme or reason to them. I currently have the cpu voltage in bios set to 1.53. This appears to deliver an actual 1.46 volts.

I am currently running stable @ 240 fsb, 1/1 memory, 4x ht, cas 2.5, 3, 3, 7, 1T, 2.85 volts to mem. If I go async, I've had the fsb up to 308 X 8, mem @ 217 with memory @ auto settings. But if I go over about 245 fsb or so, I get Prime 95 errors. I believe this is due to not having good control over cpu voltages.

I hope, along with many others, that Asus will eventually provide a stable bios release that fixes this glaring issue.
 

user1234

Banned
Jul 11, 2004
2,428
0
0
IT's been well known that currently you can't go above 240 FSB with 1:1 memory ratio. Look, I'm not sure if you understand it, but the memory speed has VERY LITTLE effect on performance. For example, if you look at this Anandtech article where they compare various memory modules in different speeds, you'll see that the OCZ 3200 PL R2 scores 516.3 frames per seconds in Quake3 Arena when the system is run at 12x200, (CPU is at 2400 mhz, and RAM at 400 mhz DDR). At 267x9 ( CPU at 2400 mhz, RAM at 533 DDR), the score is just 525.8 fps - less than 2% improvement for 133 mhz increase in DDR clock rate (note that the CPU speed is the same in both cases). So this means you should always first optimize your CPU speed (because increasing the CPU speed affects performance much more noticeably), and then try to get the highest possible memory speed, by chainging multipliers and/or dividers, without having to lower the CPU speed. Unfortunately it seems like you're stuck on this 1:1 thing, which means nothing for athlon 64, because even what you preceive as 1:1 is actually always a divider of the CPU speed (that's right, the FSB is virtual - because it's really internal to the CPU). So if your CPU can achieve 8x308=2464 mhz, use that speed as the base, and try to find other multiplier and/or dividers, which would get you the highest achievable memory speed for this given CPU speed. For example, you should be able to do 274x9=2466 and choose DDR333 divider which woud give you 224 mhz on the memory clock (DDR 448). This would totally outperform your current 9x240 = 2160 !!!


In general, I recommend you try to follow the Volvo commercial - "first you think, then you build"
 

shils

Junior Member
Dec 20, 2004
12
0
0
"I've had the fsb up to 308 X 8, mem @ 217 with memory @ auto settings. But if I go over about 245 fsb or so, I get Prime 95 errors."

In general, I would recommend trying the following.... first listen, then think, then speak
 

user1234

Banned
Jul 11, 2004
2,428
0
0
Originally posted by: shils
"I've had the fsb up to 308 X 8, mem @ 217 with memory @ auto settings. But if I go over about 245 fsb or so, I get Prime 95 errors."

In general, I would recommend trying the following.... first listen, then think, then speak

It's probably because the auto settings for your memory are incorrect. The reason for this is that to get the 2/3 divider, you have to choose "DDR266" in bios setting for memory speed, right ? Then, when you set the FSB to 308, the divider makes the memory reach about 2/3 of that, which is how you get the 217, I guess. The problem is that the bios uses the SPD timings (default timings of the RAM) at 266 mhz !!! it doesn't adjust for the higher fsb. If you check in cpu-z, or nvidia mpnitor utility, you'll be able to see what timings are currently chosen by the bios. Therefore, you should manually set the timings, accroding to this table, and don't just leave them on auto.
Note that raising the fsb above 240 mhz is not a problem, many people including me have done it and have prime 95 stable systems. The problem is most likely the timings setting of the RAM - you need to adjust them manually. In any case, a good approach to overclocking is to first find the max CPU clock, by making sure the memory is not restricting overclocking, so its best to choose very loose memory timings while trying to find the max cpu speed.

So actually, with the timings set manually you should be able to set what I told you previously.
 

DavidHull

Member
Jan 8, 2005
47
0
0
IT's been well known that currently you can't go above 240 FSB with 1:1 memory ratio. Look, I'm not sure if you understand it, but the memory speed has VERY LITTLE effect on performance. For example, if you look at this Anandtech article where they compare various memory modules in different speeds, you'll see that the OCZ 3200 PL R2 scores 516.3 frames per seconds in Quake3 Arena when the system is run at 12x200, (CPU is at 2400 mhz, and RAM at 400 mhz DDR). At 267x9 ( CPU at 2400 mhz, RAM at 533 DDR), the score is just 525.8 fps - less than 2% improvement for 133 mhz increase in DDR clock rate (note that the CPU speed is the same in both cases). . .

What you fail to realize user1234 is that AT ran the memory at 1:1 at all times and only played with the CPU multipier and HTT speed. Just because a faster memory speed, with the CPU speed remaining constant, doesn't increase performance by a significant margin, it doesn't follow that you should drop the CPU/RAM ratio. You are taking a hit in performance versus running 1:1, even if your CPU is clocked 300 - 400 mhz higher. That is why many of us are still running at 240 mhz.

Memory has a gigantic effect on performance, but not by itself - it needs to work in unison with the CPU to return the best results.
 

user1234

Banned
Jul 11, 2004
2,428
0
0
Originally posted by: DavidHull
IT's been well known that currently you can't go above 240 FSB with 1:1 memory ratio. Look, I'm not sure if you understand it, but the memory speed has VERY LITTLE effect on performance. For example, if you look at this Anandtech article where they compare various memory modules in different speeds, you'll see that the OCZ 3200 PL R2 scores 516.3 frames per seconds in Quake3 Arena when the system is run at 12x200, (CPU is at 2400 mhz, and RAM at 400 mhz DDR). At 267x9 ( CPU at 2400 mhz, RAM at 533 DDR), the score is just 525.8 fps - less than 2% improvement for 133 mhz increase in DDR clock rate (note that the CPU speed is the same in both cases). . .

What you fail to realize user1234 is that AT ran the memory at 1:1 at all times and only played with the CPU multipier and HTT speed. Just because a faster memory speed, with the CPU speed remaining constant, doesn't increase performance by a significant margin, it doesn't follow that you should drop the CPU/RAM ratio. You are taking a hit in performance versus running 1:1, even if your CPU is clocked 300 - 400 mhz higher. That is why many of us are still running at 240 mhz.

Memory has a gigantic effect on performance, but not by itself - it needs to work in unison with the CPU to return the best results.


Yes, amusing but completely untrue. What you call 1:1 or a different divider has absolutely no performance hit whatsoever, except the fact that changing the ratio changes the memory speed. I read all the details in many articles reviewing such experiments and in all of them the colcnousion is that there is no penalty for running not 1:1 on the Athlon 64 platform. Now I can understand where these halucinations are coming from, mostly because in the past data had to travel from the memory modules to the chipset and then again from the chipset to the cpu (using the fsb). But for A64 the fsb doesn't exist as an external pathway, it's internal to the cpu. That's why the A64 doesn't have the same chipset dividers of other platforms, like 5:6, 4:5, 3:4, 2:3 etc. The ratio, or dividers, for A64 are purely CPU clock to memory clock, so the memory clock is usually something like CPU/9, CPU/10, CPU/11, etc. It just so happens that in the so-called 1:1 case, the CPU-multiplier of the external clock is equal to the cpu-to-memory divider, but they are completely uncorrelated, so changing either one of these ratio doesn't affect the other. For example, if the CPU is running 10x200=2000, and the memory is running at CPU/10=200, it would be 1:1 in your book. Then if we increase the FSB to 250 and drop the CPU multiplier to 8, and leave the memory divider at CPU/10 (assuming the bios has such a setting, equiv to 4:5 divider), we'll get the CPU running at 8x250=2000, and memory at 2000/10=200, so it's the same aboslute speeds, as before but it's no longer 1:1, because the FSB:MEM is now 250:200. Would you say that there is a penalty for this setup which would make it slower than before ??? If you think so, you need a reality check.

If you don't believe it, just go read about it, or more simply just run some benchmark at 1:1 and non-1:1 setting and convince yourself that it doesn't matter. All that matter for A64 is the absolute CPU speed, and absolute memory speed, the FSB doesn't matter (as well as the derivative 1:1 concept which hinges on the notion of FSB, but FSB doesn't exist as an interconnect device in this platform, it's only a term used to refer to the external clock rate).

What you need to internalize is that there is no such thing as CPU and memory working in unison at some magical setting, because how does the CPU keeps working in unison with the memory when we change CPU multiplier alone ? obviously, this changes the internal cpu/mem divider so the memory keeps the same speed as the fsb, so now "working in unison" suddenly means a different divider? this kind of silly theories make no sense, it's easier to just look at the reality, which is simply that there is ALWAYS A SINGLE DIVIDER involved in memory accesses on the A64 platform.

Kapish ?
 

user1234

Banned
Jul 11, 2004
2,428
0
0
I take this silence to mean that you finally realize how wrong your misconceptions were, and fully accept my explanation.