Epox 4G4A+: bad voltage regulation -> bad OCing

[VOZ]

I have my P4 2.26B up to 2.72GHz (17x160) stable on this board but I have to use 1.85V in BIOS. It's at 1.82~1.84V idle and dips down to 1.74V underload I think it is because of the lack of the 12V 4-pin plug.

At first I thought I had a bad 2.26B (Malay 06/06/2002 SL67Y) but my friend who has the same chip got it to 3GHz stable at 1.725V on the Gigabyte 8IGX and 3.2GHz for Sandra. Today I tested my chip on a friend's Abit BD7-II. Guess what, the chip did 2.75GHz (17x162) rock stable at 1.6V !!! I believe with the pin mod, I could get the chip a lot higher on that board. Btw we all use the same PSU: Antec TruePower 480W.

I've used quite a few Epox boards: 8K7A+ (had 2), 8KHA+, 8K3A+ and now 4G4A+. This will be my last Epox board. All of those boards have voltage problems: low Vcore, low VI/O, fluctuating voltages etc.

I've always considered myself as an EPox fan. I like Epox boards for the features and good pricing. However, it seems that these boards are not engineered properly to handle the features they have (mainly talking about overclocking supports). It's sad I hope Epox will improve this in the future.

Goodbye Epox

 

[chainbolt]

Voz, the voltage and over-heating problems with Epox boards are known since long. I had 3 boards in a row to RMA, one of them with a short-circuit, which killed the CPU. Epox boards are very agressive priced, then can only do this by using cheap parts and lowered quality standard.

Epox has some highly tweaked products and some loyal followers, but don't rely on their quality. They are alss one of the companies with leaking and exploding capacitors according to the findings at overclockers.com

 

[audscott]

I swear mothboards have personalities: chipper, peppy, sluggish.

I recently switched from an Abit BD7II-R which undervolted terribly. Even changed power supplies from POWMAX 400 watt to Antec PP412X 400 watt - no dif. All voltages were reading low, some more drastic than others.

Then switched to 4G4A+ - voila!! All readings above default: +12, +5, V-core & V-mem.

1.6A is running 2.65 with 1.725 V-core - I'm O so happy, but just seems like there should be some consistancy!!

Sorry you have a bumb (?) board.

 

[VOZ]

Originally posted by: audscott

Then switched to 4G4A+ - voila!! All readings above default: +12, +5, V-core & V-mem.

1.6A is running 2.65 with 1.725 V-core - I'm O so happy, but just seems like there should be some consistancy!!

The 4G4A+ is lying to you, my friend. Check with MBM5 or any other voltage monitoring software, or with a multimeter. Especially, check when your CPU is at full load (burning in with Prime95 or Sandra), you'll see that the Vcore will drop down a LOT !

You might be happy with your 1.6A @ 2.65, but have you ever thought that your chip could do a lot higher if it were on another board ?

 

[chainbolt]

Hey, Audsott, good man, I see you popping up always when something interesting is going on. Our old amdmb.community is bust, now as we have all moved to Intel , where is Ace?

I also have the BD7-II, and it is undervolting, yes, but that seems to be a read-out problem and not a real issue, because my BD7-II (I have actually 2 samples) is overclocking like hell: I can run the RAM at 230Mhz and the P4 2.26 at 3068Mhz, aircooled. If there would be a voltage problem with is board, for sure that would not be possible. Review, scores, screenshot ar posted here: http://www.overclockers.com.au/article.php?id=78544. But I admit, at first glance the undervoltage looks strange and irritates after our bad experiences with the f....d up 5 volt rails on AMD boards.

I don't want to badmouth the 4G4A, I don't have it, and it's still on my shopping list. But I have my experience with several other EPOX boards, and they were no good, and it was always voltage related. What I actually think is, that Epox is giving them TOO much voltage to satisfy the oc community or to compensate for other things.

 

[gsethi]

Originally posted by: VOZ

Originally posted by: audscott

Then switched to 4G4A+ - voila!! All readings above default: +12, +5, V-core & V-mem.

1.6A is running 2.65 with 1.725 V-core - I'm O so happy, but just seems like there should be some consistancy!!

The 4G4A+ is lying to you, my friend. Check with MBM5 or any other voltage monitoring software, or with a multimeter. Especially, check when your CPU is at full load (burning in with Prime95 or Sandra), you'll see that the Vcore will drop down a LOT !

You might be happy with your 1.6A @ 2.65, but have you ever thought that your chip could do a lot higher if it were on another board ?

Voz.....why are you blaming the board ? Overclocking doesnt depends only on the CPU and the motherboard.....all other things are also involved.

Let me tell you since you liked the gigabyte so much.....

my first board was GA-8IRXP.....highest FSB i ever got on that was 150 FSB w/ max voltage stable (1.85v)
my second board was GA-8IEXP.....highest FSB i ever got on that was 152 FSB w/ max voltage stable (1.85v)
now, i got 4G4A+ and i am running 163 FSB stable @ 1.75 v......but i am not blaming the gigabyte board....it can be other components....

Overall, i like the Epox much much better than GA boards.

 

[VOZ]

Voz.....why are you blaming the board ? Overclocking doesnt depends only on the CPU and the motherboard.....all other things are also involved.

Let me tell you since you liked the gigabyte so much.....

my first board was GA-8IRXP.....highest FSB i ever got on that was 150 FSB w/ max voltage stable (1.85v)
my second board was GA-8IEXP.....highest FSB i ever got on that was 152 FSB w/ max voltage stable (1.85v)
now, i got 4G4A+ and i am running 163 FSB stable @ 1.75 v......but i am not blaming the gigabyte board....it can be other components....

Overall, i like the Epox much much better than GA boards.

Well, what do you think when with the same PSU (Antec 480W), same RAM (Corsair XMS2700), same cooling (Volcano 7+), hell even same location (Melbourne Australia, my friend lives 10 mins drive away from my place =p), but my chip does better on his BD7-II than my 4G4A+ ?

About the Gigabyte vs Epox thingy, have you ever used the Gigabyte 8IGX ? You're talking about 8IRXP and 8IEXP.

This Epox does overclock. I can get it to 160Mhz FSB/2.72GHz stable. That's not too bad, but that's not showing the full potential of my 2.26B.

gsethi, try your chip on another board (I recommend Abit BD7-2), you'll see higher OC, even with lower voltage.

Btw the Vcore dropping problem at full load is common for every 4G4A+ board.

Hi chainy, I see you on every forum, my friend =)

 

[audscott]

Nice to see some interesting discussion giong on.

I can overclock my 1.6 to 2.8 on the 4G4A+ - but it isn't stable - even with 2.85 V-core. It wouldn't post past 2.64 on the Abit, so I'm sure the chip has reached it's limit....it's a 1.6 for goodness sake.

Still, I see lots of problems with this board in quite a few threads, so can't recommend the board based on the larger community problems.

I am happy with the board - but I seem to be one of the few/lucky one. :|

 

[VBboy]

All this is making my motherboard decision making extremely difficult.

I originally wanted to buy the Epox 4G4A+, but now I'm thinking about getting the Gigabyte one. And just a few seconds ago, you confused me again

Keep this going, I want to see which mobo is still standing after the dust settles.

 

[yayoiz]

To share my experience here, VOZ did some verification on CPU and RAM with different motherboard, but what about other unexpected interferences which we will never get away from and usually are vary from locations to locations. In my case I cannot even boot up if I plug-in at one wall socket, and have to use a different wall outlet. It always is a norm that when one goes toward the extreme, all have to be right and near perfect matching, like GSETHI said somewhere, sometime is just LUCK that you get a good OCing. I am using BD7II, now at 2650MHz with 1.6A@166MHz 1:1 CPU/RAM, and with very simple components like Samsung 333(not orginal) GF2MX2-400 and a simple EVERCOOL HSF(copper base insert), I got it effortless the first time (actually step by step)and had been running Prime95 for days, thank ABIT,EPOX,ASUS,et al and INTEL anyway they help us having fun of these OCings.

VBboy: If you want to save some buck go for ABIT, it has always been friendly to all us OCers for years, since the C300A.

 

[crapito]

I've had both the Abit BG7 and Epox 4G4A+ motherboards and found them to be very similar in almost all aspects, including overclocking, with slight victories to the 4G4A+ concerning CPU voltages, LED debug, and RAID/extra IDE ports.

 

[Insane3D]

LMFAO. You get one board, don't like your results, and now you are proclaiming that every single Epox 4G4A/4G4A+ has voltage issues that hinder an overclock?

Whatever........

***Attention***

From this moment on, VOZ, is the resident authority on all issues related to motherboards.

Effective Immediately





Thank you...drive through....

 

[VOZ]

Well, every Epox 4G4A+ owner can confirm that voltage issue does exist on this board, Insane3D. Get the hell off this thread if you don't have anything constructive/informative to say other than just flaming me.

 

[chainbolt]

LOL, Insane3D,I just checked your signiture, are you running all these rigs at the same time? You must be very busy!

 

[audscott]

Originally posted by: VOZ
Well, every Epox 4G4A+ owner can confirm that voltage issue does exist on this board, Insane3D. Get the hell off this thread if you don't have anything constructive/informative to say other than just flaming me.

I have to agree with the "constructive" comment.

For clairification....I'm using MBM 5 and seeing +5 range from 5.01 to 5.11; +12 from 11.94 to 12.10, V-core (set at 1.70) at 1.69 to 1.74 and V-mem from 2.78 to 2.80 (set at 2.8).

It seems to me (as I stated earlier) that most mobos have distinctive characteristics. Of course, the dynamic with the CPU, power supply and ocmponents come into play as well. Good conponents usually make for improved stability, although I've had some cheap stuff in the past that worked great, until they died rather quickly...they were cheap, afterall.

I am of the opinion I got a good board (zap!! ) - who knows if that's a lasting thing.

Lots of people, just like VOZ, reporting inconsistencies with this board - which is a crying shame - cause it obviously has (had?) real potential.

Anyway - the the major reason I switched from the BD7II-R to the 4G4A+ was to get the 845G 4:5 divider. I was hoping I could reach 166 FSB in a stable overclock at 4:5 - which is my setting.

Now....if the SID 648 chipset has a PCI/AGP lock, that will be my next stop.

 

[terminalmind]

I have this board and MBM5 shows the exact voltage I have set in the bios.
When running prime95 the voltage does drop a little, but never more than .050V
From what I have read Epox motherboards are not the only ones whose voltage
drops under load. If your voltage is dropping over .100V then you probably have
a defective motherboard.

 

[chainbolt]

Originally posted by: audscott

Originally posted by: VOZ

Now....if the SID 648 chipset has a PCI/AGP lock, that will be my next stop.

It seems the Abit with SiS 648 is selling already. It's announced and listed on the Abit website, why don;t you download the manual and have a look.

The missing PCI/AGP by the way was EXACTLY the reassn why I did not buy the just released VIA P4Pb-400, it's a nice board with all features, but only this is missing. And after all the trouble with the stupid 1/4 and 1/5 dividers on VIA/AMD boards: no board without the possibilty to fix the frequency anymore!

 

[Insane3D]

Originally posted by: VOZ
Well, every Epox 4G4A+ owner can confirm that voltage issue does exist on this board, Insane3D. Get the hell off this thread if you don't have anything constructive/informative to say other than just flaming me.

Ok Mr. Know-It-All, I happen to have an Epox 4G4A+, with a 1.6A @ 2.56ghz, 160mhz FSB, DDR400 with a 350W Antec PSU only, and my voltages are actually all right on, even while running Prime 95. When my Vcore is set to 1.6v and I'm running Prime 95, my actual vcore in Motherboard Monitor runs @ 1.62v actual Vcore.. I don't have this mysterious power drop to the CPU under load that you claim "every Epox 4G4A+ owner can confirm".


Here's actual proof for you VOZ, is that informative enough for ya?

Also, a couple of other things..

1. I can post in any thread I want to, and you can't do a damn thing about it Mr. Junior member....I've been here a lot longer than you.

2. The bios doesn't "lie" to anyone. Motherboard Monitor get's the voltage and temp readings from the exact same Winbond chip that the bios reads from. What do you think, there are two seperate sets of voltage/temperature readings on the board, and MBM is the only thing that will "tell the truth"? ROTFL!

3. You might want to think twice before proclaiming something about "all Epox 4G4A+ owners can confirm". Do you actually think you know how every motherboard is performing for everyone that has it? I've got news for you, reading a few forums is not the same thing as knowing what every 4G4A+'s does.




Ahhh...another day, another guy spreading FUD.... I guess there is one born every minute...

 

[Insane3D]

Originally posted by: chainbolt
LOL, Insane3D,I just checked your signiture, are you running all these rigs at the same time? You must be very busy!

Hey chainbolt, good to see you again. Yep, the P4 is my rain rig at the moment, and the others don't get used a ton. I will probably give the Sis648 board a chance, most likely with a 2.26ghz or a 2.4ghz. I'm also looking forward to the KT400, I'll probably get one and pair it with an unlocked XP and some PC3200 and go for a 200mhz FSB. Also, while we usually don't agree when it comes to Epox boards, at least we have intelligent discussions about it... See you around tha forums....

 

[Insane3D]

Hmmmm...this thread died quick for such a "widespread problem"....

 

[Doomguy]

Why would the 12v connector have anything to do with the core voltage? Does anyone even know for sure where the power for that 12v connector goes? I see no way how it could be going to the cpu, as regulating the 3.3v rail would have to easier than the 12v.

Epox has some highly tweaked products and some loyal followers, but don't rely on their quality. They are alss one of the companies with leaking and exploding capacitors according to the findings at overclockers.com

All the Epox motherboards I saw in the page linked from overclockers.com were old, as far as I saw.

 

[gsethi]

Regarding Voltage Issues:

No voltage issue here. My 12v is actually higher in MBM. CPUVCore goes down a bit when running Prime 95......it goes down like 0.030 - 0.040 v on average so thats normal too since on my gigabyte...it used to go down like 0.080 - 0.090 v.

Regarding Epox Quality Issue: (this was stated by a computer dealer in the Bay Area)
Epox quality was not good about 3 years ago but they have made a serious comeback recently. Their products can be compared to Abit and Asus quality wise. He also said that they are hearing a lot less complaints with Epox boards now as before, moreover, they were getting more complaints for Abits and Asus. He said no Comments on Gigabyte boards since he didnt knew a lot about them. So, in conclusion, Epox is trying to make a comeback regarding quality now.

Overclocking:
Again, i repeat that "OVERCLOCKING IS LUCK". Your results depends on many factors, not just the motherboard, memory and CPU.


Voz, you said that your friend had the same "other" products like PSU and memory but "THEY WERE NOT YOUR PRODUCTS". Each and every computer part is different even made with the same components. Dont you agree that all 1.60a are not the same. Some are better than others. It can be the same thing that your friends PSU maybe slightly better than yours. If you want to truly test it, then use the exact same components each time, only changing one component (motherboard) at a given time. Only then you should comment which motherboard is the best.

 

[soljahz]

Well i have the same vcore problem, my games freezes everytime the vcore dips down too much on load. I have to set it at 1.80 at bios, so when its full load it dips to lowest at 1.68 and idles at 1.78 and these are from motherboard monitor. That the only way i can stay stable at 154 fsb 4:5 ratio. I went from antec 350 ps to an ttgi 520 watt ps and had same results, so i have no clue at what it is, just bad luck i suppose.

soljahz

 

[ahsia]

I think the lack of the 12V 4-pin plug is a concern with some people, but just like CPUs, I think not all motherboards will allow the same overclocking potential. I have the 4G4A+, and I can hit 2.4GHz on my 1.6A, but I have to run it at 1.75v. I am looking for the some news on the Gigabyte 8IGXP, looks very nice, feature packed, and hopefully will overclock well. Then it is probably onto the SiS648 boards.

 

[OneOfTheseDays]

hmm this is exactly why i returned my epox board long time ago. But hey, i don't wanna start this debate again. If ur not happy return it and get an Sis 648 chipset mobo when they come out (abit's is already out).