PDA

View Full Version : Anybody have any results with the EPOX EP-4PCA3+ ??


Major
07-21-2003, 07:16 AM
Just curious if anyone has any results yet with the Epox CW. I am going to be getting another mb and cpu in 2 weeks and I'm considering the Epox since it works without any issues with the Promie Mach I.

I've already tried the P4C800, IC7-G, and the MSI Neo so I thought the Epox might be a good choice for my next CW.

Amoeba Assassin
07-21-2003, 08:25 AM
how do you plan on finding one?

Major
07-21-2003, 10:34 AM
I found ONE place that has them on pricewatch.com ! lol

Amoeba Assassin
07-21-2003, 11:07 AM
lol

what did you type in under your search

twiztdskatr
07-21-2003, 11:24 AM
its www.axiontech.com a store in my town :D :D :D i can drive there and pick one up mwahahahah!!!

twiztdskatr
07-21-2003, 11:26 AM
http://www.axiontech.com/prdt.php?item=45390 :D :D :D

karmakid
07-21-2003, 12:50 PM
in sweden and probably the rest of Europe theyre availible everywhere. o_O

Kunaak
07-21-2003, 01:06 PM
just not in america :(

Major
07-21-2003, 02:07 PM
So whats the verdict on this board ??

I was hopeing that it would be inexpensive, but its about the same as a IC7 !

Does anyone know for sure what the voltages are like on it ??

twiztdskatr
07-21-2003, 02:09 PM
Originally posted by Kunaak
just not in america :(

that link i posted is in america...its in texas 20 minutes from my house.

Amoeba Assassin
07-21-2003, 02:46 PM
yeah I am anxious to hear results. I'm considering sellin my p4c800 & getting one

Kunaak
07-21-2003, 03:29 PM
My P4C sucks, no voltages above 1.6 no matter waht bios I use :(
other then that it's great, just that one quirk is enough to ruin this perfectly fast board for me :(

so this epox, and others like it, look even better now ;)

Amoeba Assassin
07-21-2003, 07:17 PM
I've seen 300+ fsb a couple of times with it. My only problem with it is, JCViggen got one & said it isn't nearly as good as an IC7 or P4C800. I would say Axion is gonna be out of stock of them soon though;)

Major
07-21-2003, 07:21 PM
It would be hard to live up to the P4C800 and IC7, both are great boards ! I guess I'll get another P4C800 hehe

Amoeba Assassin
07-21-2003, 07:26 PM
http://www.xtremesystems.org/forums/showthread.php?s=&threadid=14927&highlight=4pca3

theres a 300+ fsb

If the Vcore can go decently high I think its a good board, but I'll wait & let you try it out:D

Jupiler
07-22-2003, 01:08 AM
I have JC's board laying here next to me.
I'm gonna give it a try today, together with a 2.6C.
I'm gonna use the boxed Intel cooler as I just installed my new 2.8C (came back from RMA) on my IC7-G and I don't wanna remove the Prommy (takes to much time).
I'll post some results later on.

Major
07-22-2003, 06:59 AM
Looking foward to hearing back from you Jupiler !

TodB
07-23-2003, 08:47 AM
Well I got that with a 2.4C and A-DATA DDR450 memory:

Amoeba Assassin
07-23-2003, 09:45 AM
nice

Shane5578
07-23-2003, 12:52 PM
look here when using the 1.7v+vcore trick he got just high as the ic7, it seems the low vcore was holding the epox back compared to the other boards asus/abit

http://forums.overclockers.co.uk/showthread.php?s=&threadid=101063

Jupiler
07-23-2003, 01:12 PM
Nice find Shane, thx.
I just tried that 1.4500V setting and indeed, it gives me 1.75V (1.70V in MBM).
The board undervolts terribly.
1.525V in bios reads 1.47v idle and 1.42-1.44V full load.

I prefer the IC7-G in this case, voltages are far more stable and don't jump around like the 4PCA3+.


Here's a quicky :

Twinmos/ Winbond BH5, 3.3V on the memory :D

http://users.skynet.be/MetalHead/4PCA3+/250_2225.JPG

Shane5578
07-23-2003, 01:17 PM
nice, but you should try unbuffered benchmark and 3dmark01, then compare it to the ic7-g at the same fsb.

i got a feeling that the epox will be faster

Jupiler
07-23-2003, 01:30 PM
I'll give it a try.
But I have a 2.8C in my IC7-G right now, and the 2.6C in the 4PCA3+. So scores may differ a bit.

BTW, I'm using the latest bios, but VCore only goes up to 1.60V.
Epox said that there were some instability problems with the previous bios's, which gave 1.85V max.

TASOS
07-23-2003, 02:59 PM
My opinion

I have just finished testing a rev 1.2

The board isnt good enough (yet)?

It has LAN problems

Not enough Vcore (even if set @1.45 in bios...the actual Vcore measured with a voltmeter is 1.69)

It's not stable over 300 fsb (the same cpu booted @324 on a Asus P4P800)

It really needs at least a quick bios update ( i doubt that...cause rev 2.0 is on the way)

My last hope.....perhaps Abit Max3

QuadDamage
07-23-2003, 06:39 PM
you need to look for the bios at AOA which has v-ore upto 1.85v/1.77v actual. so far it does 13x300 at 5:4 but needs around 1.9/2.0v to fix the slowdowns in 3d. Not bad overall. 3.3v v-dimm , 2.2v agp, 1.77v v\core.

Trooper_GR
07-25-2003, 03:23 AM
Originally posted by QuadDamage
you need to look for the bios at AOA which has v-ore upto 1.85v/1.77v actual. so far it does 13x300 at 5:4 but needs around 1.9/2.0v to fix the slowdowns in 3d. Not bad overall. 3.3v v-dimm , 2.2v agp, 1.77v v\core.

My 4PCA3+ tops at 275 FSB, at 276, no post at all.

I know its not the CPU as it has been tested at other m/b and it works fine. The RAM is working fine at 240 1:1 with no problems.

I removed one dimm, and I could go up to 287 but not stable.

Shane5578
07-25-2003, 04:39 AM
the 4pca3+ beta bios is very old. it uses drivers which were used in their 845e boards and has no extra memory performance from PAT. not only that but your overclock will be crippled because of some fsb holes.

i only recommend their latest bios and using the the 1.45vcore to give 1.7v

i think the 4pca3+ should be compared to the ic7 more closely and not the p4p800 as the springdales are known to overclock more than canterwoods using the same cpu but give lower performance

MCWB
07-25-2003, 09:54 AM
Originally posted by Jupiler
[B]The board undervolts terribly.
1.525V in bios reads 1.47v idle and 1.42-1.44V full load.Jupiler, it doesn't actually undervolt that badly if you check it with a multimeter, it's within a bull's roar. :) It's just the monitoring chip mustn't be calibrated correctly so it 'reads' a lower voltage than what it actually is. 1.600 Vcore in the BIOS give 1.58 or so IIRC?

twiztdskatr
07-25-2003, 01:43 PM
Originally posted by Jupiler
Twinmos/ Winbond BH5, 3.3V on the memory :D

Thats kind of high voltage for only 240mhz and 2-5-2-2 :(

mine does that at 2.8v :eek:

Jupiler
07-25-2003, 02:01 PM
Originally posted by twiztdskatr
Thats kind of high voltage for only 240mhz and 2-5-2-2 :(

mine does that at 2.8v :eek:

I know, but the board is kinda acting funny. Might try some lower voltage though.

BTW : what memory are you using?

Just tried 2.9V and 3.0V but got a BSOD while booting.
Those Winbond chips like high voltages, compared to others.

QuadDamage
07-25-2003, 04:11 PM
well, this board is kinda strange. I flashed back to the latest bios cos the beta (old) is terribly slow and disables CPC and such. i can do now 14x255fsb 1:1 stable at 3.2V with twinmos pc3700, but i'm limited by v-core, i think. For some reason it worked way better with 2.6C than with this 2.8C i'm playing with now.
i score 62xx/61xx buffered, but can't attach images here.

Anyone knows how to volt mod it?

Jupiler
07-25-2003, 04:18 PM
Is that the one with the VCore option up to 1.85V that you used? Or which one?

twiztdskatr
07-25-2003, 04:31 PM
im using 2x256mb twinmos/winbond bh-5

QuadDamage
07-25-2003, 05:03 PM
nopre, with latest 03/07 bios. with old one i get only 54xx/54xx at 255 1:1.

perco
07-26-2003, 03:00 AM
My friend got a 4PCA3+ and it was very nice but the built-in soundcard was broke so he had to send it back.

I never had time to test it, but I´m going to when they have fixed it.

Jupiler
07-26-2003, 02:30 PM
Originally posted by twiztdskatr
Thats kind of high voltage for only 240mhz and 2-5-2-2 :(

mine does that at 2.8v :eek:

Well, mine won't boot at 2.8, 2.9V or 3.0V at that speed. Have to give it 3.1V.
I find the board to wacky, don't know, but I just don't like it that much.
I tried that beta bios with higher VCore, but it seems very unstable.

On both the IC7-G and 4PCA3+, I can boot and run some Pifast and SuperPi benches @ 250Mhz, 1:1, 2-5-2-2, but 3DMark locks up on both systems.

4PCA3+, 2.6C @ 250Mhz, 1:1, 2-5-2-2, unbuffered

http://users.skynet.be/MetalHead/4PCA3+/250_2225_unbuffered.jpg

IC7-G, 2.8C @ 250Mhz, 1:1, 2-5-2-2, unbuffered

http://users.skynet.be/MetalHead/IC7/250Mhz_2225_unbuffered.jpg

Scores are almost identical,BUT Im using a 2.8C on the IC7-G and a 2.6C on the 4PCA3+ , and the extra 250Mhz on the IC7-G make the difference. So maybe not a fair comparision, but still an excellent unbuffered score for the 4PCA3+.

Here are 2 screenies from voltage readings :

4PCA3+, 1.525V set in Bios :

http://users.skynet.be/MetalHead/4PCA3+/4PCA3+ voltages.jpg

IC7-G, 1.625V set in Bios :

http://users.skynet.be/MetalHead/IC7/IC7-G voltages.jpg

Some will like the board, others won't. I loved the 4PEA+ and was hoping that this one would be a worthy successor, but IMHO, it's not (yet).
Maybe that the 4PCA3+ rev 2.0 will be better, but the board hasn't been released yet, so it might take a while before it will hit the stores.

I also can't understand why Epox has released 3 revisions of the board in such a short period of time. They should have taken much more time in fully testing and improving it, and then release it, instead of releasing a, in my eyes, unfinished board.

Remember that this board is a rev. 1.0, perhaps that the rev. 1.1 and 1.2 are better and improved, but I can't tell.
Just my 0.02€

r00t3d
07-26-2003, 02:56 PM
rev 1.2 here

doing 326 FSB stable (check sig)

I was @ 300FSB stable @ 1.525 vcore.

IMO this board rocks :)

Jupiler
07-26-2003, 03:04 PM
Very nice, r00t3d.
Do you experience the same Vcore jumps like I do, or are yours more stable?

r00t3d
07-26-2003, 03:07 PM
unfortunately, I get the same :( jumps from 1.49 to 1.47 not to extreme though, but still not good.

That is the only thing that has concerned me on this board, but it has never stopped me from o/cing.

I know this baby can do better than 326 Stable, with better cooling on the NB, lets hope I get some more free time to exploit that possibility.

Shane5578
07-27-2003, 08:49 AM
thankyou for the results jupiler.

theres one thing though. it seems performance mode on the epox 4pca3+ was disabled judging from cpuz. and the ic7 had perfomance mode enabled?

twiztdskatr
07-27-2003, 09:58 AM
i wanna see same cpu in both boards!

Jupiler
07-27-2003, 12:57 PM
Originally posted by Shane5578
thankyou for the results jupiler.

theres one thing though. it seems performance mode on the epox 4pca3+ was disabled judging from cpuz. and the ic7 had perfomance mode enabled?

On both systems, CPC was set at AUTO.
Don't know if it really is enabled on the IC7-G, but on the 4PCA3+, it seems to be disabled. I just checked it again and CPU-Z still gives "disabled".
Tried booting with the same settings, but with CPC enabled, but got a BSOD (memory management) while trying to run Sandra memory.
I'l give it a try on the IC7-G later on.

Shane5578
07-27-2003, 01:57 PM
fast cs should be enabled and cpc disabled.

cpc only works near stock speeds for me

Jupiler
07-27-2003, 02:22 PM
Originally posted by Shane5578
fast cs should be enabled and cpc disabled.

cpc only works near stock speeds for me

Thx Shane.
Works OK, and CPU-Z reports system performance set to "enable".

Maybe, just maybe, I might stick in in the Prommy and compare it further with the IC7-G with the same cpu, and videocard, and run some Pifast and 3DMarks.

Shane5578
07-27-2003, 02:28 PM
unbuffered results don't get affected by cpu or graphics.

anyway it was interesting to see your results

Clevor
07-27-2003, 05:00 PM
My Rev. 1.1 just died on the VDIMM side. Rev. 1.0, 1.1, 1.2, 2.0. Don't ya get the feeling Epox is using us as guinea pigs?

I've run as high as DDR504, 2-2-2-5 at 3.2 volts Memtest/3D/SuperPI stable on this board. Corsair XMS3500 C2s, of course. But then I noticed 3DMark2000 kept kicking out or locking up. Now the #5 Memtest keeps locking up no matter the CAS setting. I thought I fried a stick but each one works fine in single channel. And on this Rev. 1.1, can't boot at 1.450 VCORE. Trying to RMA the board now.

Man, this board has dried up in Japan. There was a bunch of them a week ago. Tsukumo is all out. I went back to a small store that actually had them for the cheapest price. I noticed the s/n said board #38. What? This is lower than my Rev. 1.1. I checked the rev and this is a 1.2, so I bought it. At least I should get the 1.7 volts at 1.450 setting. As for the VDIMM circuit: fingers crossed. The 1.1 and 1.2 look exactly the same, but I do see different markings on two small chips, but that may be just batch variation.

Despite the shortcomings, the board still gives me better unbuffered memory scores over the P4C800-E I've been flogging the last couple of nights. The Asus is not bad though, it will run my Corsair at DDR440, 2-2-2-5 and DDR460, 2-3-2-5 at the max 2.85 volts it offers. Something is fishy with the VCORE on the Asus though, I don't believe it offers more than 1.6-1.63 actual.