4PCA3+ bug, but it enables 1.75+vcore, using latest bios

[Shane5578]

yep its true, as several people have confirmed this already. and has enabled higher OC's for me.

get 1.75+vcore, by using the negative vcore options.

[Alexandrus]

Kwel, I told you I've seen something funny when I tried to undervolt my CPU.

[Shane5578]

yeah it was you who first found out. lol

ok, here are the settings i used, and voltages i get, using the latest bios on my rev 1.0 mobo

1.4750=1.00v
1.4375=0.99v
1.4500=1.75v :banana:
1.4625=1.40v
1.4750=1.39v
1.4875=NO BOOT
1.5000=1.04v
1.5125=1.40v
1.5250=1.40v
1.3750=1.52v
1.5500=1.48v
1.5625=1.50v
1.5750=1.53v
1.5875=1.53v
1.6000=1.55v

[r00t3d]

what about rev 1.2?

[Clevor]

I just installed my 4PCA3+ last night, it's a Rev. 1.1 I got in Japan, except I got a lousy 2.40c, I think. But I was bummed to notice I get 1.60 max in the BIOS, with the latest 7-3-03 BIOS. I thought the Rev. 1.0 and 1.1 is supposed to see 1.85???

I get the exact same 1.55 actual at 1.60 BIOs you get Shane. I think I can get my money back for the board as the manual clearly states it should go up to 1.85, and the vendors here in Japan are pretty good about accomodating you.

My 2.40c maxes out at 242 even though it is an L310xxx Malay. However I was able to run 3D stable at 1:1 or DDR484, 2-2-2-5, with two XMS3500C2 sticks, dual channel, at 3.2 volts. Mind you, these are not 256 MB sticks, but two 512 MB! I got 3800 max unbuffered. Earlier on in testing, I ran Memtest clean at DDR496, 1:1, 2-3-3-6 at 3.2 volts. Do you notice this board offers 3.3 VDIMM? It may be bogus though, as 3.2 on my 4PEA+ still looks like 3.1 actual. I think I might be able to boot at DDR500, 2-2-2-5 and get some Sandra screenies even though my CPU is not stable there and Memtest will show some errors.

I haven't processed the screenies in Photoshop. I may post them later. But I need a decent 2.40c that can do maybe 3.3 gig under 1.55 VCORE to really test this board out. I will try the 1.75 vcore to see if I can get my 2.40c higher.

[Tony]

guys don't trust the bios for your vcore measurement its miles out.When i had the epox i measured on the board..1.6V gave 1.59V actual with 1.52V giving 1.515V actual and it was ROCK steady.Only problem was the vdimm didn't work :mad:

Anyway don't trust what you see via the winbond chip..get a good multi meter and measure it.

[twiztdskatr]

those are horrible undervolts....i think ill stay with my IC7

[Clevor]

those are horrible undervolts....i think ill stay with my IC7

That's if you can even find a 4PCA3+. Even in Japan, I think I've only seen 6-7 in all of Akihabara. The 4PEA+ was also in short supply.

Well if those are the actual volts as Bigtoe mentioned, the board doesn't undervolt very bad. The 4PEA+ only undervolts by .025 volts, the lowest of ANY motherboard I've seen. BTW, I've found the # of mosfets don't affect undervoltage: a Jetway 845PE board I have has 4-phase with 8 large mosfets and still undervolts by .05, but the VCORE is rock steady.

I've checked the voltage on the 4PCA3+ using MBM5, Sandra, and the Magic Utility (which I have found to be spot on with other boards), and I do see 1.55 at 1.60 BIOs. I do know the Hardware Monitor readings in the BIOS are way off and don't seem to move.


I'm gonna try RMAing the board to Epox for a Rev. 2.0, as the voltages are way too screwy

[Shane5578]

more people in the epox forums are getting the same results.

clevor , set your bios vcore to 1.4500 using the latest bios, and you should see voltages of 1.7+, your board doesn't sound like its faulty. officially its limited to 1.6vcore.

i haven't got a multimeter to measure them more accurately, but i'm very happy with the results, even though i'm seeing 60+c using very good aircooling

i don't know if it works on the rev1.2 mobo

[Clevor]

Shane, I tried setting the board to 1.45 and it won't boot! It gives out feeble chirps until I clear the CMOS. It won't boot at any negative settings.

I then flashed to the beta BIOS, but that is just an old June BIOs. Yes, there are voltage options to 1.85, but anything above 1.60 don't work (board won't boot). Heh, no wonder Epox simple removed the settings above that in the latest BIOS.

HOWEVER, what a difference a good CPU makes. I bought two L310xxx Malays, so I installed the second one. I was only able to test default voltage last night, but this one maxes out at 278, 3.34 gig. At 280 I see two #3 errors in Memtest. The board undervolts so this is actually 1.475! No complaints here. I will be posting on how well my Corsair XMS3500 runs on this board. How about DDR500, 2-2-2-5 with two 512 MB sticks? Buffered memory scores are not impressive, even at 5:4, but unbuffered scores are exceptional. Ihave not run anything other than 2-2-2-5 on this board, at DDR440-500. Don't get any better than this.

If you got a good chip that don't need more than 1.55 VCORE, the board will suit you fine. It runs memory (XMS3500 anyway) stellar, and seems to be a good overclocker. I have a P4C800-E on the way and will do a comparison on overclocks, but that board will need a VDIMM mod to get above 2.85 volts.

[Clevor]

more people in the epox forums are getting the same results.

Hey Shane, which forums are you talking about? Link? I know of the Epox forum over at amdmb.com, but that is for athlons.

[Vatt]

Maybe www.aoaforums.com
I know that Epox UK tech people post there...

[Clevor]

Thanks for the link, Vatt. Couple of posts there on the 4PCA3+.

[RootX]

Well, I picked up one, rev. 1.2... I've just finished setting it up, but the first impressions are :
Vcore can reach 1.75, by adjusting it to 1.45.
at the same Vcore with the Asus P4P800, epox cannot reach the same fsb. With the asus I could hit 300 fsb with 1.6, epox can only reach 292 at the same Vcore. It needs 1.75(can't choose anything lower) to hit the 300.

But at the same fsb, mem results are better than the asus at the same timings. Also, if cpu-z is correct, Pat is enabled at 1:1, 5:4 and 3:2.

Will do some more test to decide which one to keep.

[r00t3d]

hmm,

Ok I have hit 326 FSB @ 1.6v on the 4PCA3+ STABLE
I was 300FSB STABLE @ 1.525v

Some peeps musnt be setting things properly, if they cant get 300FSB @ 1.6v :stick:

[RootX]

Don't think that there are so many things to set...
Apart from that, you have a 2.4, I have a 2.6, can't make a direct comparison between those two, when it comes to fsb.
Right now, FOR my setup, epox can't hit the same fsb as asus at 1.6Vcore, but mem results are faster than the asus...

[r00t3d]

I wasnt comparing 2.4 to 2.6, as you never mentioned you had a 2.6

:D

[RootX]

lol, well, I do....

[Clevor]

RootX, how are you measuring the VCORE??? Remember, the Asus boards overvolt. Heck, default voltage (1.525 BIOS) on an Asus board may well be 1.575-1.59 volts. Default voltage on the Epox is more like 1.475. You need to be sure 1.60 volts is what Windows sees for both boards.

With 845 boards, the best overclocker is the Abit/Albatron, with the Epox 2-4 mhz behind, and the Asus boards maybe 5 mhz behind (with a bit more voltage needed). But the Epox is in it's own league when it comes to running fast CAS settings. I have a P4C800-E on the way and will be comparing it to the 4PCA3+, but I'll have to VMOD the Asus board to get over 2.85 VDIMM.

I'm not even using the RAID on the 4PCA3+, but the IDE drive on it performs excellent. Check out the ATTO graph:

[RootX]

The Vcore on both motherboards, was measured using a multimeter. I can tell you, that the vcore on the epox, is stable as hell, 1.6 is 1.59 at load.... asus overvolts as usual, but at load I get 1.58 when set at 1.6, which is not that bad.

At 292 fsb, with the epox, I got the same mem bw that asus gave me at 300. Of course asus loses pat(mam) when oced, while epox doesn't care about dividers and stuff.. Pat is fully operational.
Since I'm using ocz's dc gold, can't go any faster that 2-3-3-6.