• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

AN7 @ 200FSB unstable. Help plz.

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Tough Cookie

Registered
Joined
Mar 1, 2004
Location
Sydney
Hey guys.

I had a GN7400 L1 and it was great for what it was, it ran my 2500 @ 3200 stably with prime95 @ 1.825 and I never had any instability issues.

I wanted better sound and the lovely features that the AN7 had including a bios that didn't read rediculous CPU temps.

So I got the AN7, I was going to get the NF7-s v2 but then I read some reviews that said the AN7 was pretty much the same except newer, one person even said it could of been called and NF7-s v 3! so I thought hey what the heck only $20 more and I get optical in!!!

So yeah I tried running at 200FSB at 1.825 even went up to 1.875 and still no go, just wondering if I could get some help with what voltages I should use n what not.

I just want to get it to XP3200 like I use to have it on my GigaPoo board :)

Thanks for your help.

my system

Abit AN7 v1.0
TT Silent Boost
XP2500 (locked) (166*11)
2 x 512 DDR 400 Kingston Value ram 3.0.3.3.8 ( I think)
A280 TDH TI4200 300/600
80GIG WD 8mb cache
40gig Seagate
Antec True Power 380Wat
Lite-On CD DVD Combo drive.
 
That is a whole lot of voltage.

Most 2500's will do 3200 at or near stock voltages. You could go to 166 FSB and slowly increase speed to see where the problem occurs.
 
I have hit the wall at 180MHz and would also like to know how to get above 200. No matter what settings and voltages I use nothing helps.
 
Yeah it's f%$#ed...

I read a review about the board before I bought it and they said pretty much all good things, that is why I got it.
They said max FSB they could get was 225 on a XP1800 with air, they thought it might be the CPU that couldn't handle higher though...

So I go out and spend $140 on this board and it won't even do 200 fsb for me... I should of just got the NF7-s for 119 :(

I think I have latest bios, I will try to upgrade it anyway just incase..

This is crap...

Thanks so far for the advice I will try small fsb adjustments see where I max out...

Might try increasing vdimm a little see if that helps.

What CPU volt will "most" XP2500 do at 3200?
 
I think that you have a bad chip. The A7N does not like FSB's higher than 215, but it is fine at the lower ones. Do you have A7N rev 2.00 if it is not then it will not run@200 fsb. But if you have a rev 2 board then it should be able to do 200 fsb no prob, which makes me think that it is the cpu.
 
No my AN7 is rev 1.0

And I don't think it's not my CPU cause I had an Nforce board before this one running fine at XP3200.

So you think its my revision?

If it was my board, would that mean that if I had an XP3200 it wouldn't work?

What do you think?
 
Last edited:
Take that stock northbridge heatsink off and see if the heatsink has a halfway decent contact. That could be one problem, its well worth it to check even before you install the board. Sometimes the thermal paste only covers half the northbridge off to a side...not even in the center. So, atleast consider that to be a potential problem. Check out the mosfet temps and voltage regulator chip. If they get toasty you get a voltage swing~drop that would effect stabilty.
 
Thanks man, yeah but I alread went out n purchased a passive heat sink and squeezed new heat gel over it, so that kind of rules that out...


Well I have been taking it up notch by notch with the front side bus and I'm up to 177 now and it seems to b fine.

I'll test prime for a few hours and do some general use on the PC if it doesn't crash I will take it up a bit more, think I'll go to 182 next then just keep going up bit by bit from there. I raised the voltage to 1.725 I think....

So far so good.

I'll keep u guys posted see what happens.. :D

EDIT:

I can't seem to get it over 181 FSB. any suggestion?

at 11x181 vcore1.686 primes seems to be fine if I put it to 11x183 prime crash's I rais vcore up to 1.8 and it still crashed...

Any ideas of something I can try to get it to run?

Thanks
 
Last edited:
Have you tried 2.5/3/3/11 ram timings and upped the voltage to 2.7 on the ram? Sometimes you need a bit more voltage on the ram with certain boards. And I am not sure but I dont think the nf7-s likes cas3 settings so try 2.5 and see how that works out.
 
It would be nice to see if it is the board or the cpu that can't go up any more. I think the comment about revision 1 or 2 above was thinking aobut the NF7, where the 1 was a 166 rating, and the 2 was 200. All of the AN7's support a 200 fsb, because they use the nForce™2 Ultra 400 chipset. As long as your ram is pc3200 or above you should be set for well over 200 fsb.

To see if it is the cpu or chipset/ram, lower your multiplier so your cpu isn't going nearly as fast. That way you can push the ram and chipset to their limit without pushing your cpu much above stock speeds.

You can also raise your multiplier and test the top speed of your cpu without stressing your chipset too much.

When you find the maximum speed of your ram and chipset, you can figure out which one is the limiting factor by changing your fsb:ddr ratio in the bios. This is essentially the multiplier between your ram and chipset, so you can do the same thing with it as you did with the cpu and chipset to find their limits.
 
Right now with what you see in my sig, I'm running at 200 x11 with a vcore of 1.80 (found out it my pass prime95 with lower vcore but crashes on DC) If anything try 200FSB with yor ram at 166 and see what happens.
 
Thanks for all the hints guys, I had lost hope but now I have some more faith.... I'm leaving work early to go try these things out... :-D

Thanks so much, I so hope I can get to XP3200 at least! Otherwise I would of spent a stack of cash for nothing ):

Thanks so much again.....

I'll post over the weekend to tell you how it goes....

EDIT


One more thing, my 2500 is locked I think, week 43 or 44 in 2003 from memory ):
 
Ok update on the situation.

I have been fiddling around with FSB, voltages and Mem ratios as suggested.

I tested the CPU at 200 x 11 with vcore1.65 (default) with RAM @ 166, tho I only got to test for 20 minutes it didn't crash, I will test later on tonight for a few hours to see what happens.

I was really surprised that it ran for even 20mins with default vcore.

Chances are I will have to up vcore if I test for longer.

When I tested before with synchronous RAM to FSB @ 200 it ran for about 1min40secs before it stopped so I have a feeling that my problems might be my Kingston RAM with AN7, I will do further tests tonight and post what happens.

RAM is Kingston Value RAM mem timings are 3.3.3.8 or 9 quite slow I know.

Just to let you know I have tested this RAM with memtest3.0+ for about 3 hours @ DDR3200 on my old mobo with no problems, so it might not be a fault with the ram but rather a combined fault with RAM n Mobo, I'm just speculating here.

Other ppl with AN7 wat RAM are you running and how fast are you running that RAM?

Was getting a little warm in my Case, after running Prime for the 20 mins I was getting 52C on my CPU. I'm not to worried about that, except South Bridge seemed fairly warm/hot.
It wasn't hot enough for me not to touch n leave my finger on tho, but still a lot warmer then it was before!

Does anyone know how to MOD a heat sink onto the south bridge of an AN7, it doesn't have mounting holes like the North Bridge does. Another heat Sink on it wouldn't hurt, if I was able to put one on.

Thanks again for all the help guys...
 
Ok i have a problem to with the the 2500 and the an7 mother. I have no problems getting the cpu at 3200. I have it actually as an option in the bios wich it says if i want 2500 or 3200. But the problem is when i get windows running and i start up a game the computer crashes. If i set it it at 11x 196 it runs fine but if it's at 3200 it crashes. Can the 2500 chip handel to be overclocked to 3200. Please beer with me i am a noob in overcloking.
 
Bogo24dk said:
Ok i have a problem to with the the 2500 and the an7 mother. I have no problems getting the cpu at 3200. I have it actually as an option in the bios wich it says if i want 2500 or 3200. But the problem is when i get windows running and i start up a game the computer crashes. If i set it it at 11x 196 it runs fine but if it's at 3200 it crashes. Can the 2500 chip handel to be overclocked to 3200. Please beer with me i am a noob in overcloking.

LOL

Ok, the fact that it crash's means exactly the opposite to what u have posted! :D

What you need to do is use a program called Prime95 after you oc, this program will tell u if your oc is running stably or not.

You should run it for a few hours, some ppl say 24 some ppl say 12, the way I see it.

Prime95 will test you CPU to its limits for wat ever time you run it for, if it stops, ur CPU isn't statble running for that length of time with this particular configuration. Optimely it should run for ever that is why some ppl say, "yes 24 hours is stable enough for me", I only run it for 6 hours and am happy with it because chances are I'm not going to use my PC for any longer then this amount of time at this current CPU utilization the whole time.

The fact that ur PC doesn't run at all when put through a little utilization (in a game) means that it's completely unstable.

Yes mine boots as well but won't run stably and that is the whole problem and why I has posted this thread..

Hope this helps you out a little bit, there is also program to test your ram called Memtest3.0+.

These proggies are to help you determine what piece of hardware is giving you your instability issues when oc-ing

What do you think of my theory on Prime95, guru pc enthusiasts? :p
 
Right now I'm stable for everything at 3200 speeds with a vcore of 1.8. I run my ram at 166 cause well on some boards it doesn't like doing 200 altho it passed Memtest at PC3200 speeds. As for the southbridge Tough you have to epoxy one on there use AS or AC epoxy to do it. As for Prime I put it on before bed and kick it off after I wake up.
 
This is exactly the same problem I have.
AN7 + AXP 2500 unlocked, 2x512 Infineon PC3200 RAM. BIOS 1.5.
I can run my CPU at FSB 200MHz, but RAM works stable (in Prime) only up to 180 MHZ (asynch). Memtest passes ok.
I have replaced AN7 with new one (other revision, without trace cuts). No change. Tried Kingston 2x512 Value PC3200, the same. Then another Kingston - 2x256 - still max 180. I have even connected other good PSU, still no luck. All memory works without any problem in P4 boards, DC, FSB800.
So the only part not replaced is CPU but I doubt it is his fault. It works in asych mode at FSB200 and passes well "small" test of Prime (this one when all data fits in cache) in synch FSB200 mode. Now I have no clue what to do next, but I would rather blame Abit for f..ing up something in AN7 construction or BIOS.

[edit]
All RAM I tried has been running at settings 3-3-3-8, 3-4-4-8, 3-3-3-11, 3-3-3-12. It has always failed at 400MHz at any of these settings.
 
Last edited:
Tough Cookie said:
Ok update on the situation.

RAM is Kingston Value RAM mem timings are 3.3.3.8 or 9 quite slow I know.

Just to let you know I have tested this RAM with memtest3.0+ for about 3 hours @ DDR3200 on my old mobo with no problems, so it might not be a fault with the ram but rather a combined fault with RAM n Mobo, I'm just speculating here.

Other ppl with AN7 wat RAM are you running and how fast are you running that RAM?

Abit AN7 v1.0
TT Silent Boost
XP2500 (locked) (166*11)
2 x 512 DDR 400 Kingston Value ram 3.0.3.3.8 ( I think)
A280 TDH TI4200 300/600
80GIG WD 8mb cache
40gig Seagate
Antec True Power 380Wat
Lite-On CD DVD Combo drive.

I have AN7 that came with v14 bios, XP2500+ 0409week, samsumg ddr400 512mbx2 (3 3 3 8)

I can oc to xp3200+ at 1.67vcore and prime stable 24hr, memtest86 stable 24hr. 3Dmark01se looped for 6hours no problem.

maybe borrow a psu with higher rating and try....abit's mobo tends to have stringent power requirement.

Are there AN7 v2.0? haven't seen one here...
 
redface said:

Are there AN7 v2.0? haven't seen one here...

There are two revisions, both marked as "Rev 1.0" by Abit.
One is made on Taiwan, it had to be fixed by trace cuts. Second one is made in China, no trace cuts. Other differences are placing of onboard firewire connector (in first one is closer to PCI1 and parallel to it, in the other connector is perpendicular and closer to AGP) and one capacitor near AGP is bigger in second version.

I have tested both. Problems are the same. BIOS 1.3 and 1.5 tested.
 
Back