Epox 83KA+ -vs- iWill XP333-R

[AMD_Man]

Originally posted by AMD Soldier


AMD Man, how can you even make a statement like that? (ie. "because neither is clearly ahead")

That mobo is probably on pair with the EPoX 8K7A in performance, an AMD 760 chipset mobo.

Add the facts that it only has 2.8 volts for the ram, 2.035 volts for the cpu compared to EPoX's 8K3A/+'s 2.2v. Not really personal opinion, these are just facts I throw out at you.

Lastly the 8K3A/+ also OCs better, thus making this an easy choice.

The Iwill XP333 significantly outperforms the AMD760 chipsets. The ALI Magik1 rev.C has seen major improvments in the memory controller. Latency is down significantly, and it's less than 5% slower than the KT266A and in some cases it's faster. KJY9 says I'm ignorant, but I have no interest in either board. I know the specs of both, but you seem to be a bit shortsighted, no insult intended. A bit extra voltage doesn't necessarily make a better board. There is, as was mentioned before, the 1/6 PCI divider, which is a crucial factor in high FSB overclocking. Of course, there's also the "VIA factor" as some people I've heard call it. VIA has in the past made some absolutely BUGGY chipsets. Granted, the KT266A, and KT333 have supposedly fixed most of these problems, but a hiccup is bound to arise here and there in terms of PCI card compatibility. Another board that might be considered great for overclock is the Leadtek WinFast 7350KDA based on the SiS 735. It also supports the 1/6 PCI divider. Extreme memory voltage isn't necessary if you already have the RAM that can handle a 200MHz FSB. The only thing lacking in these two boards is the extreme VCore.

As a final note, I nearly forgot to mention, the SiS 735 is neck and neck with the KT266A, give or take.

EDIT: Spelling

[KJY9]

Originally posted by AMD_Man


The Iwill XP333 significantly outperforms the AMD760 chipsets. The ALI Magik1 rev.C has seen major improvments in the memory controller. Latency is down significantly, and it's less than 5% slower than the KT266A and in some cases it's faster. KJY9 says I'm ignorant, but I have no interest in either board. I know the specs of both, but you seem to be a bit shortsighted, no insult intended. A bit extra voltage doesn't necessarily make a better board. There is, as was mentioned before, the 1/6 PCI divider, which is a crucial factor in high FSB overclocking. Of course, there's also the "VIA factor" as some people I've heard call it. VIA has in the past made some absolutely BUGGY chipsets. Granted, the KT266A, and KT333 have supposedly fixed most of these problems, but a hiccup is bound to arise here and there in terms of PCI card compatibility. Another board that might be considered great for overclock is the Leadtek WinFast 7350KDA based on the SiS 735. It also supports the 1/6 PCI divider. Extreme memory voltage isn't necessary if you already have the RAM that can handle a 200MHz FSB. The only thing lacking in these two boards is the extreme VCore.

As a final note, I nearly forgot to mention, the SiS 735 is neck and neck with the KT266A, give or take.

EDIT: Spelling

Umm sounds great coming from someone who has no experience using the 8K3A. :rolleyes: I've built on both, add to that the fact that all the reviews are touting the 8K3A as the OCing mobo champ and the newest bios having a 1/6th divider and you can't possibly argue that the XP333 even comes close. I can tell it's been awhile since you've built and OCed a KT333 mobo, if at all, by this statement that you made " there's also the "VIA factor" as some people I've heard call it. VIA has in the past made some absolutely BUGGY chipsets."
If you're a n00b (like yourself), the XP333 is a great board, but by no means "possibly the best OCing XP mobo" out there like you stated on another thread.
And your statement "Extreme memory voltage isn't necessary if you already have the RAM that can handle a 200MHz FSB.".
If you're going to take an proc to it's highest possible Mhz, you're going to need extreme VCore for the Proc.
Enough beating this dead horse...like I said earlier, go get yourself an XP333, OC it and come back with the benches. If I can't beat it with my 8K3A+, I'll happily eat my typed words.
But for you to keep giving out false information is just plain wrong. If another n00b (like yourself) comes along and purchases an XP333 on your recommendation, thinking he's getting a great OCing mobo and all his buddies are whooping his butt with their 8K3A's, he's going to be pissed at you for lying to him he he

[AMD_Man]

Originally posted by KJY9


Umm sounds great coming from someone who has no experience using the 8K3A. :rolleyes: I've built on both, add to that the fact that all the reviews are touting the 8K3A as the OCing mobo champ and the newest bios having a 1/6th divider and you can't possibly argue that the XP333 even comes close. I can tell it's been awhile since you've built and OCed a KT333 mobo, if at all, by this statement that you made " there's also the "VIA factor" as some people I've heard call it. VIA has in the past made some absolutely BUGGY chipsets."
If you're a n00b (like yourself), the XP333 is a great board, but by no means "possibly the best OCing XP mobo" out there like you stated on another thread.
And your statement "Extreme memory voltage isn't necessary if you already have the RAM that can handle a 200MHz FSB.".
If you're going to take an proc to it's highest possible Mhz, you're going to need extreme VCore for the Proc.
Enough beating this dead horse...like I said earlier, go get yourself an XP333, OC it and come back with the benches. If I can't beat it with my 8K3A+, I'll happily eat my typed words.
But for you to keep giving out false information is just plain wrong. If another n00b (like yourself) comes along and purchases an XP333 on your recommendation, thinking he's getting a great OCing mobo and all his buddies are whooping his butt with their 8K3A's, he's going to be pissed at you for lying to him he he

I've been building computers and programming for 8 years so, I'm not exactly a noob.

[MoboMan]

Originally posted by AMD_Man


I've been building computers and programming for 8 years so, I'm not exactly a noob.

Since you were 7? Then quit acting like a n00b. Go buy an XP333, bench it and post your screenies. Because I for one would love to see you even come close to competing with an 8K3A with an XP333. You'll be the first.
Either that, or quit posting nonsense :rolleyes: :rolleyes:

[Mo]

i have no problem with the xp333...i bought the 8k3a cause its the only one here, it runs fine, oces well...and thats my 2 cents...

[Mandorallen of Molynnr]

Originally posted by Mo
i have no problem with the xp333...i bought the 8k3a cause its the only one here, it runs fine, oces well...and thats my 2 cents...

You haven't OC'd your CPU according to your sig...man, that's like buying an E46 M3 and never putting your foot flat.

Highest I've seen a XP333 at was 220 MHz (check the archives at www.ocworkbench.com). That was merely posting, not stable. That's pretty good, and the layout is also nice, but I don't think this board rates as the best OC board ever. Comparitively speaking, it's lacking in voltage selection. Performance, however, is around the KT266A boards. KT333 based boards would probably blow it out of the water.

http://www.anandtech.com/mb/showdoc.html?i=1578&p=22

Also see this, Anandtech tried getting each board to the maximum as regards FSB :

http://www.anandtech.com/mb/showdoc.html?i=1578&p=24

[Crashman1]

there's also the "VIA factor" as some people I've heard call it. VIA has in the past made some absolutely BUGGY chipsets."
If you're a n00b (like yourself), the XP333 is a great board, but by no means "possibly the best OCing XP mobo" out there like you stated on another thread.[/B]

I like the XP333 because it supports some early cards that I've never seen a VIA chipset support, but then again I'm and old timer who actually owns these early cards. I was ready to get the Leadtek Winfast 7350KDA until I suddenly found overclocked Northwoods in the lead. ALi has always put stability ahead of performance, and SiS made their incredible recovery into the world of stability with the 735. Both chipsets feature compatability that surpasses anything VIA has offered up to the 686B southbridge. As for VIA's latest chipsets, I refuse to even try them, I've been burned by everything from the MVP3 (in fact, I just got burned again by an even earlier VIA chipset during a rebuild) to the KT266 and don't feel like giving them any more of my time or money.

I have a strong feeling that recent reports of stability using the KT266A and KT333 are do mainly to the use of mainstream hardware and applications that have been designed to work around VIA's bugs. As a specialty systems builder I am sure that there are cards still around that will not work with even the most current VIA chipsets, but will work with ALi or SiS chipsets. The reverse cannot be said.

In fact, I just spoke at length with a man who's Hauppauge TV card would not work on his Epox KT266A board, and Hauppauge told him it was a problem with the chipset, a PCI latency problem I think. And the SB Live, as crappy a card as it is, still had problems reminiscent of the past on certain 266A systems, just not as frequently or severely as earlier chipsets. My favorite audio card, the Vortex2, still didn't work right on most KT266A equiped systems, with no reports of similar problems on other brands of chipset. And some people are still having to play "musical cards" with their PCI slots to get their VIA chipset systems to run right, even with ACPI enabled, which is supposed to take care of such problems.

So these problems may only affect 1 in 10 systems. Maybe even only 1 in 20. But the fact that, when they do occur, they occur on VIA chipsets an nowhere else, is a sign that I have done no ill deed by avoiding their latest versions.

Lets also not forget VIA bullying their corporat customers by telling them they would short-supply any companies who offered an SiS 735 solution, essentially baring Asus, Abit, and MSI from releasing their 735 boards. Or the fact that they stole bus technology from Intel when ALi and SiS actually licenced it. If they were an American company they would have been brought to their knees by the DOJ for engaging in anti-competitive practices.

Every board you buy with a VIA chipset strengthens their position to stifle technology from other companies. Every board you buy with an ALi or SiS chipset allows these smaller companies to add to their development budget and bring us new features.

[AMD_Man]

Originally posted by MoboMan


Since you were 7? Then quit acting like a n00b. Go buy an XP333, bench it and post your screenies. Because I for one would love to see you even come close to competing with an 8K3A with an XP333. You'll be the first.
Either that, or quit posting nonsense :rolleyes: :rolleyes:

8 actually, I'm almost 16. I started out with QBasic. Then I went to VB4 when I was ten. Then Delphi 2. Then Delphi 5. I'm still using Delphi 5, but I'm planning to teach myself Visual C++ or Visual C# this summer. My first custom made PC that I helped build was a P 133, 32MB EDO RAM, 2GB IBM Hard drive, Intel 430VX motherboard, Soundblaster AWE32, 10X Pioneer CD-ROM Drive, ATI Xpression (Mach64). I've made a couple of minor upgrades but I still have that rig, and it's still running at 133MHz. The multiplier for it is locked (jumpers don't seem to allow me to overclock it) and the FSB is at 66MHz and Intel motherboards don't allow over that at the time.

I haven't worked with the KT333 but I helped my friend build an AXP system wiht the Asus A7V266-E. It fried on him once. I also helped a another friend build a 1.33GHz Athlon and IWill KK266Plus. The KT133A is an annoying chipset, to say the least. I have a KT7A myself and I fried 3 or 4 of them.

EDIT: Also, last year, I set up computers and replaced faulty hardware in a computer lab at a summer camp I was volunteering at. Most were P233MMX systems. I helped setup a P3 800MHz rig running WinNT Server to host all the apps and stuff and to regulate internet access.

[Mandorallen of Molynnr]

Um...

I haven't worked with the KT333 but I helped my friend build an AXP system wiht the Asus A7V266-E. It fried on him once. I also helped a another friend build a 1.33GHz Athlon and IWill KK266Plus. The KT133A is an annoying chipset, to say the least. I have a KT7A myself and I fried 3 or 4 of them.

Either you had a DOA or it's UE. Same applies to your experiance with the KT133A. I've assembled many VIA based setups for clients (for Intel 'cause it's cheaper or AMD because it's usually the best) and I'm pretty impressed. When a client brings in a PC because he's having problems with it, it doesn't seem to be consistant with what chipset the board is based on.

Another thing, you shouldn't blame VIA if the board manufacturer is at fault. DOA's are your mainboard manufacturers fault. It happens with all of them. Don't blame the chipset.

[AMD_Man]

Originally posted by Mandorallen of Molynnr
Um...



Either you had a DOA or it's UE. Same applies to your experiance with the KT133A. I've assembled many VIA based setups for clients (for Intel 'cause it's cheaper or AMD because it's usually the best) and I'm pretty impressed. When a client brings in a PC because he's having problems with it, it doesn't seem to be consistant with what chipset the board is based on.

Another thing, you shouldn't blame VIA if the board manufacturer is at fault. DOA's are your mainboard manufacturers fault. It happens with all of them. Don't blame the chipset.

I find it odd that Asus, Abit, and Iwill have all had problems. The only common factor I can see is the VIA chipset. I hate to say it but the only motherboard that didn't cause any problems was the MSI 850 Pro5 that my friend got for his P4 1.7GHz.

[Mandorallen of Molynnr]

Originally posted by AMD_Man


I find it odd that Asus, Abit, and Iwill have all had problems. The only common factor I can see is the VIA chipset. I hate to say it but the only motherboard that didn't cause any problems was the MSI 850 Pro5 that my friend got for his P4 1.7GHz.

Everyone mainboard manufacturer has problems. For example, we had to RMA quite a few ASUS CUSL2 boards. Now other people here might have had no problems with CUSL2 boards but would have had to RMA quite a large percentage of, say 8KHA+ boards, which I have never had a problem with.

It's not odd, it's made in Taiwan.

This post has no discrimination against anyone Taiwanese, if you were offended, please accept my humblest apologies.

[AMD_Man]

Originally posted by Mandorallen of Molynnr


Everyone mainboard manufacturer has problems. For example, we had to RMA quite a few ASUS CUSL2 boards. Now other people here might have had no problems with CUSL2 boards but would have had to RMA quite a large percentage of, say 8KHA+ boards, which I have never had a problem with.

It's not odd, it's made in Taiwan.

This post has no discrimination against anyone Taiwanese, if you were offended, please accept my humblest apologies.

True, true, but how does all this stuff get past quality control?

[Mandorallen of Molynnr]

Originally posted by AMD_Man


True, true, but how does all this stuff get past quality control?

There's quality control?

Well, I don't know how far quality control goes. If it posts once, it's working fine? It might die later, or never turn on again.

There has also been some speculation that shipping might drop the stuff...or something like that. I remember some guy who worked for a courier service kicking a box with a fragile sticker and "this way up" on it because it never said please.

[japin]

heh, i dont know about you guys, but the real world performance of this board is on par with that of my friends kt333, i dont notice a difference, now if i was number crunching, or something along them lines i prolly would, but i aint and when i am gaming/benchmarking the only thing i am lacking in is memory performance, not that i care, but honest truth all this sticking up for iwill, go with the Epox board, i was thinking of getting one based purely on its features... looks like it has that part in the bag

[Molochi]

Originally posted by Mandorallen of Molynnr


There's quality control?

Well, I don't know how far quality control goes. If it posts once, it's working fine? It might die later, or never turn on again.

There has also been some speculation that shipping might drop the stuff...or something like that. I remember some guy who worked for a courier service kicking a box with a fragile sticker and "this way up" on it because it never said please.

Heh, yeah. I've seen a lot of boxes with smushed corners and suspiciously foot sized holes/dents.


But on-topic...

xp333 vs kt333, I think the issue is what you get with your OC, performance wise. Back in dec-jan, when we were looking at this mobo the performance wasn't compelling vs kt266a running 33MHz slower. Moreover, people who had motherboards with earlier steppings of the ALi Magic 1 were reluctant to recommend it. In fact, I don't recall (outside of this thread) anyone ever asserting the ealier (a7a266 and ka266) ALi based mobos were more stable or bug free compaired to Via. Now, 5 months later, nobody talks about it because they didn't go the guinea pig route and bought something else. As is the norm, reviewers haven't gone back to look at the board again, but I doubt much has changed.

The latest review of the board I can recall was here...

http://www.hardocp.com/reviews/mainboards/iwill/xp333r/

Their summary (March, some months after the release of the board) was preluded with this attempt at running at 333...
-----------------------------------
"Our next set of benchmarks did not go as well, with the system locking up repeatedly while trying to run SysMark 2001. Content Creation would BSOD the machine or randomly crash back to the desktop. Quake III hard locked the system at the settings menu with a continually looping sound.

We set the BIOS to "failsafe", which loads the most scaled back settings, but the problems still occurred. Obviously, this was making it impossible to get any real benchmarks done with this motherboard while running at 166MHz FSB. Nothing we tried could get this board to run stable at 166MHz FSB.

A quick check revealed that the system would still run Sisoft Sandra and CPUID, but nothing else without crashing. This only reinforced our belief in the importance of running a full suite of benchmarks during a motherboard review, and not just a few rounds of Sisoft Sandra and CPUID.

We used several different BIOS revisions, as well as a BETA "Evaluation" BIOS during testing to no avail.

Something else that was rather odd was that any attempt to set the CPU/FSB divider to the 1/5 or 1/6 setting manually resulted in the system hard locking. The only way to get the system to run again was to reset the CMOS. Setting the divider to the "Auto" setting, the system would default to the 1/5 divider when running at 166MHz FSB. It's strange that it was capable of using the 1/5 divider on "Auto", but not when it was selected manually.

Problems

Obviously, with a motherboard that is sporting the "333" branding, you would certainly expect it to at least run at its advertised speed. After all, what sense would it make to add the "333" to the name if it's unable to run at that speed. So, to say we were a little disappointed would be an understatement.

Another very real issue with the IWill XP333R was this:

This motherboard review had to be among the most difficult to ever finish, and not because of the obvious reasons. Over the course of this review, we went through no less than four separate motherboards attempting to finish this review. More than one board arrived D.O.A. (dead on arrival) and the boards that DID arrive in working order had a tremendous amount of problems. Four boards later, we were finally able to get a board that could finish a review.

Now, had all the boards been tested in one location, one might give Iwill the benefit of the doubt by playing devil's advocate and say "Maybe [H]ardOCP has some testing methodology or hardware setup that is killing these boards". While we certainly have killed our fair share of hardware, the fact that two motherboards were shipped to Kyle (one DOA and one non-functional) and two were shipped directly to me (one DOA and the final review candidate), you can safely assume where the issues with this board may lie. While I cannot tell you exactly where or why these boards got screwed up, I can say that it is extremely frustrating. . .to say the least.

We waited until now to mention this to keep from unfairly slanting your view of this motherboard, hoping to give you an objective look at the XP333R. But this is information that has to be passed on to you, the end user, to help you make an informed decision when considering the platform you plan to base your system around."

--------------------------------------------
Personaly, I don't see the allure to spending more money for the best memory if the system isn't going to perform much/any better than a KT266A/KT333 running at a stable (if lower) freq on cheaper DIMMs. If all that mattered was frequency, I'd be running a P4. However if a mobo allows a high fsb oc, allows you to run at that speed without worrying about twitchy drives and data corruption (pci deviders). Then it's not all bad, as long as it works. But that doesn't make it an overclocking king or even a prince.