Yet more issues with GA6OXET and overclocking!

[oldfart]

You guys are familiar with my tale of woe and the GA60XET-C and the Tualeron 1.0A overclock. The wire trick made it workable, but I was not happy with it and sold the CPU and have a PIII-S 1.26 on the way.

I put my PIIIEB 1000 in the board while I'm waiting for the PIII-S 1.26 to get here. Yes, I removed the wire trick. I ran this same CPU in my CUSL2 for a long time 150 FSB, stock 1.75 Vcore. 100% stable, never a problem. I put it in the Gigabyte setup, cleared the cmos, set it up the same way. 150 FSB, 1.75, 3-2-2 timing. Booted up and ran prime 95 to check stability.

The PC rebooted in like 15 sec! I've never seen that before. Ok, maybe we need a tad more vcore for this mobo. I put it @ 1.8. Here is another issue. 1.8 is max, That if more than enough for a Tually, kind of low for a Coppermine overclock. I saved and rebooted and was greeted once again by the hang on reset bug! I have to say, with my CUSL2, I had a 550E @ 733, 700 @ 933, 1000 @ 1125. Never, ever did I have a hang on boot issue. Not once! Now, every reset it will hang. This CPU is temporary, so I'll just go back to a non overclock until the new CPU gets here. Who knows how that will go.

I dont know about this board. The Tually was a new thing, but this is certainly a step down from a CUSL2 with a Coppermine.

If you get one of these boards, you'd better be prepared tp learn the wire trick.

[Yodums]

No offence but your story are going to scare people and not get the GB board and go with the Abit boards :D

I think your motherboard has some serious issue.
If this is an overclocking issue I doubt bios releases will be released as no one will encounter the problem I guess...

Sorry to hear you having the bad luck man.

Yodums

[oldfart]

And BTW, I had two of these boards. They both act the same way. Not trying to scare, persuade anyone to buy or not buy a product. Just posting my experience for others to see. They can make their own decisions. I'll be travelling till the end of the week. When I get back, the PIII-S should be here. Maybe it will be the magic combo.

[Yodums]

I know your intentions of the posts were to post your experiences although it will kind of give people the first reaction of the board being bad as always... :D

Have a safe and fun trip and good luck on the server chip.

[MilkPowder-2]

Originally posted by oldfart

The PC rebooted in like 15 sec! I've never seen that before. Ok, maybe we need a tad more vcore for this mobo. I put it @ 1.8. Here is another issue.

No, vcore isn't needed for the mobo(GB GA 60XET C) or any mobo. Based on everything you have said here(which is very theral and I appreciate), I say this mobo is $hitttttttttttttt.

If you get one of these boards, you'd better be prepared tp learn the wire trick.

Not only that. They've better pray. :cool:

[Yodums]

Originally posted by MilkPowder-2



Not only that. They've better pray. :cool: [/B]


LOL :D

[MilkPowder-2]

Originally posted by Yodums



LOL :D

:D


And Olfart, RMA that board A.S.A.P Just go with the Abit ST6 and save yourself time. :)

[oldfart]

I dont do flamewars, sorry. I'll tell you my honest experience and thats it. Draw your own conclusions,

Well, here is another one. One of the cool features is the 1/5 and 1/6 PCI right? I couldn't try this with the Tualeron since I could not get my FSB very high. On the CUSL2, I could run 160 FSB, 1.2 GHz, but my nic wouldn't work @ 40 MHz PCI. Might as well try it on the Gigabyte.

First, 160 FSB, 3-3-3 7,9 ram, 40 PCI, 80 AGP 1.8 Vcore. Just like the CUSL2, it works, but the nic says no go.

Next, same thing but try PLL/16 30/60. Boots up..., just before Windows loads, BSOD. Tried it 3 times.

[Yodums]

Eh good idea MP2...

Either you maybe just got one of those unlucky boards..

By the way what Windows are you on?

Please don't tell me Win9x!

[oldfart]

Win9X? Give me a little credit will ya! Win2K Pro. Remember, I had 2 boards. Its not a bad board. Hmm ST6. Maybe TUSL2. I miss the CUSL2.

[ol' man]

Yeah you probably had a real nice 1.0A, sounds like the mobo is a piece O' junk! I am going to stick with ABIT for now.

[MilkPowder-2]

Originally posted by ol' man
Yeah you probably had a real nice 1.0A, sounds like the mobo is a piece O' junk! I am going to stick with ABIT for now.

See what I mean. My m8 ol'man also believes that it maybe the mobo causing all these issues, not the Chip. I had the same feeling as what ol'man had about the that 1.0a chip. Those are good chips. I have respect to these chips as much as I do to P3S. :mad:

[oldfart]

Yup, its a shame. This has cost me quite a bit of time and some $$ too. I already sold the chip. Its going in an MSI mobo. I'll let you know how it works in it.

The Place I got the board from doesn't carry Abit. They do have ASUS. I'm calling to see if they can do something for me tomorrow. What a pain.

[oldfart]

I got an RMA for the board. I'll have a TUSL2-C by the end of the week. If its is as good as my CUSL2, I'll be happy. I know some here dont like the TUSL2. I'm starting from 133 FSB and going up. The lack of a 1/4 PCI @ < 133 FSB doesn't matter. My CUSL2 was like a rock @ 150 FSB. This board and the PIII-S should be a nice combo.

[Flu!d]

Well....At least you're used to vid pinning....You're gonna need those skills with the limited voltage adjustment using Tualatins on that mobo. Hope you don't mind sacrificing AGP 4X at high FSB's either....:D

[oldfart]

The AGP 4X thing is easily fixed. With a PIII-S, I don't think I'll need the high voltages that a C1.0A needs. Fluid, I know you like the Gigabyte board. I'm glad it works well for you. I just found it a bit buggy for my likes. I went with that board for the 1/5 PCI feature which turns out does not work. The final straw was when it would not run my Coppermine CPU as well as the ASUS did.

Just to clarify why I'm returning the GA6OXET. Its not because of the 1.0A. I had that actually working well. Its because I belive that board has a fundament problem in the way it posts. The Coppermine CPU proved that. Again, here is the problem:

PIIIEB 1000. Works perfect on a CUSL2. 150 - 160 FSB. Stock 1.75 Vcore. No warm boot problem. Can run Prime95 with no errors exen @ 160 FSB.

Put the same CPU on the Gigabyte: Prime95 @ 150 FSB 1.75 Vcore doesn't error like it normally does on a bad overclock. The PC crashes back to a BIOS reset. Increase Vcore to 1.8, and you get a warm boot bug.

Explain how a CPU runs fine @ a higher bus speed on default Vcore on one board and wont do the same on the Gigabyte?

Also, why is 1.8 enough to run the CPU and a mere .05 lower is not enough to post properly on this board? That tells me this thing has a bug in the way it posts. That in itself is reason enough for me not to want this board.

Don't worry, if the TUSL2 doesn't work, I'll be complaining about it too! :mad:

[ol' man]

It seems besides the onboard sound not working after ~145fsb the ST6 is the best for high fsb's besides the BD133u. Hopefully you can get that TUSL2 up there in the FSB dept. I always said if you have a 133fsb chip by default then the TUSL2 should be okay!

[Flu!d]

The resetting bug was fixed by vidpinning the chip when installing it to trick the mobo into believeing the chip had a higher vcore than it actually does....Thus providing a 'voltage boost' when the 'chipset' resets the voltages to default just on inital bootup before the vcore is adjusted to the user defined settings at the bios post screen...

You're probably gonna have to vid pin on the TUSL2-C due to it's limited vcore control in the bios so I can't see the problem here (and the St-6 you have no option but to vid pin....Probably why it dosen't show signs of the 815 chipset resetting bug).

I also thought the TUSL2-C disabled AGP 4X at about 140mhz FSB....

Quite possibly the extra voltage required to overclock the cumine chip was due to the new VRM spec and actually was no fault of the mobo? Mabye all 'b' stepping mobos with the new VRM spec would provide exactly the same result when overclocking a cumine chip....You gotta remember, the boards are designed with Tualatins in mind.

In closing....You could basically say that there is NO perfect Tualatin 815EP 'B' stepping chipset, especially when running a cumine processor as the new VRM spec is designed with Tualatins in mind. They all have their problems and we fixed the problems on the Gigabyte board quite easily - And eventually you're basically gonna have to vid pin no matter what board you get....
No offence dude but I believe that you were a little to quick to discredit an otherwise fine mobo with no definitive evidence that it was the mobo, and not the chipset in general, that was at fault. If the 1/5th divider dosen't work on the XET-C that's fine, but I can tell you that it 'definatly' works on my XET (and the Japs at 190 mhz FSB, which the ABit or the IWill will not pull, will back me up on that!)

And I don't know about you but the thought of flashing to a modified unsupported bios in order to fix 'some' of the faults on the TUSL2-C worries the hell outa me! And I have heard of instances where the modified bios has actually rendered the board useless....

Remember....There is no perfect 815EP 'b' step mobo....Don't be too harsh on them. If you like your CUSL-2 so much just keep it and get a P3 1000E....They overclock well....

[oldfart]

Fliud, you bring up some interesting points. I don't have the TUSL2 yet, but with any luck at all, I'll have it by the end of the week and get some time this weekend to play around. I'll answer the below questions then. I can comment on a few things now.
. If the 1/5th divider doesn't work on the XET-C that's fine
No, it isn't. The reason I decided on the Gigabyte board was for this unique feature that would allow me to run 160+ FSB and still have a close to spec PCI. Since that feature doesn't work, that distiguishing feature is of no value. Its nice that it works on yours and the Japs I guess. A BSOD reboot is all it gets me. What FSB did you use? I used 160 PLL/16 30/60. Maybe a higher FSB like 166FSB 33/66 would work??
You're probably gonna have to vid pin on the TUSL2-C due to it's limited vcore control in the bios

Quite possibly the extra voltage required to overclock the cumine chip was due to the new VRM spec and actually was no fault of the mobo? Mabye all 'b' stepping mobos with the new VRM spec would provide exactly the same result when overclocking a cumine chip....You gotta remember, the boards are designed with Tualatins in mind.

And eventually you're basically gonna have to vid pin no matter what board you get....
All interesting points/questions. I cant answer any of them now, but will check all of that out once I get the system together. Unfortunately, the C1.0A is gone, so I cant test with it. I will have the CuMine and the new PIII-S to test with.

On the 140+ FSB issue on xUSL2 boards. I found out about this when I first got my CUSL2. I've been using various versions of the X86-Secret Bios (it gets updated when new ASUS bios comes out) on my CUSL2. It has always worked flawlessly. I have no reservations at all using the TUSL2 version. If this BIOS was not available, I would not buy this board, but it is, so I did. I sent an email to the author of the bios and asked why it is needed. He says that ASUS (unfortunately) follows Intels recommendation of going to AGP 2X if bus speeds are 5 - 10 MHz higher than spec. The fix works, and it works well.
Remember....There is no perfect 815EP 'b' step mobo
True. I would certainly say that of the GA6OXET ;)
If you like your CUSL-2 so much just keep it and get a P3 1000E....They overclock well....
Nah, I'll get the same mobo with Tually support. The TUSL2.

[Flu!d]

No, it isn't. The reason I decided on the Gigabyte board was for this unique feature that would allow me to run 160+ FSB and still have a close to spec PCI. Since that feature doesn't work, that distiguishing feature is of no value. Its nice that it works on yours and the Japs I guess. A BSOD reboot is all it gets me. What FSB did you use? I used 160 PLL/16 30/60. Maybe a higher FSB like 166FSB 33/66 would work??

The 1/5th divider was unproven on the XET-C....Unfortunatly you turned out to be the test pilot in this situation which can not be used as grounds for complaint. I have run my P3 1000 EB at 150, 151, 155, 160 and 166 mhz FSB on this mobo reaching a maximum voltage of 1.8 volts on pll16 (1/5th divider) with no problems at all, in fact it was of great success! (and at 150 mhz FSB with the 1/5th divider enabled my pci/agp devices were running 'way' below spec). And the crazy Japaneese at 190 mhz FSB just backs up my point in relation to the XET and 1/5th and upwards dividers. Once again....The XET-C was unproven in this area, everyone just 'took it for granted' that it was exactly the same as the XET....Obviously it is not.

True. I would certainly say that of the GA6OXET
The above quote is in regards to 'there is no such thing as a perfect 815EPb mobo....
Well I could also say the same of the TUSL2 and the St-6 mobos...They all have their strengths and weeknesses and I hear of people complaining about their TUSL2 and St-6 mobo's every day....Why is it that every time an overclock fails the person involved is very quick to blame the mobo?

As far as I can see the 'only' problem with the GB mobo is a slight 'hanging' problem upon initally firing up the board with 'some' chips....The problem was located and rectified just fine with a small piece of wire to 'trick' the mobo into thinking the chip had a much higher vcore than it did. This same piece of wire must be used on the St-6 and the TUSL2 as well when overclocking to the extreme so I can't see the problem here....

Nah, I'll get the same mobo with Tually support. The TUSL2

Then don't complain about the 1.8 volt limit when using cumine processors as the TUSL2 uses exactly the same VRM spec....With more limited vcore options using Tualatin chips....

I'm sorry dude....No offence, but it was obvious you missed your CUSL2 board (who dosen't become fond of a certain board - just look at me!) and your experiences of the GB board reaked of a 'witch hunt' trying to find fault with a perfectly good mobo....

No part of Oldfarts experiences either prove of disprove that the Gigabyte board is not an exceptionally good mobo able to keep up with, or beat, the rest of them (the memory bandwidth on the mobo is exceptional....Almost right up there with the BX!)....Just don't count on a 1/5th pci divider with the XET-C and be sure you know how to vid pin on any 815EP'b' mobo you select....

[oldfart]

The 1/5th divider was unproven on the XET-C....Unfortunatly you turned out to be the test pilot in this situation which can not be used as grounds for complaint
Huh? Um, yes it can. The documentation says it works and the feature is in the bios. If you bought a car and the windshield wipers didn't work on the less expensive model would that be ok? No.

BTW, I was trying to be nice and keep things light (the little wink after the joke about the not perfect Gigabyte).

You are assuming the TUSL2 will have the same issues with hangs. You may be right, you may not be. I'll know in a few days.

BTW, I ran some Sandra memory benchmarks and saved the screenshots. I'm interested to see how they compare. I'll post that info as well.

[Flu!d]

Please don't get offended dude....I'm just 'clarifying' some points....That's all!:)

Gigabyte 'never' said it's 1/5th and 1/6th pci dividers would work on either board and it definatly dosen't mention them in the bios update notes....It's a unerversal bios for both the XET and the XET-C so what works for one, logically, may not work correctly for the other and vise versa....Like I said before the 1/5th divider was never proven to work 'on the XET-C' so you can't complain about any problem found with it....And, what's more, the TUSL2 has no 1/5th or 1/6th pci divider 'either'....So you 'really' can't complain....Note that I said the 1/5th and 1/6th pci divider does not work on the XET-C....My experiences with the XET prove that it does appear to work on this model (I haven't had one lockup with it selected) and, once again, the crazy Japaneese at 190 mhz FSB more than backs up my claim....

The Gigabyte GA-60XET is, without a doubt, the highest performance Tualatin motherboard we've reviewed to-date. Not only did it obtain the summits of Overclocking, but it also managed to turn out great performance even without the slightest tweaking. For all these reasons, in addition to its great features and high-score, we're proud to award the Gigabyte GA-60XET our "High Grade Hardware Award" - a board we wouldn't hesitate to recommend

And this Review by Active Hardware (which is linked to by many other sites) dosen't make the XET sound like a very overclocker 'unfriendly' board either....It was faster than an Intel reference board at default speed! Something which was also noted when PC User Australia gave the board a 9/10....And they had no problems with it's quality control.

I really think, due to the fact that you didn't hesitate to RMA the board, that you really wern't sure if you wanted it in the first place....You really appear to like ASUS - and I have nothing against that....You're cool!

[oldfart]

I guess we look at it differently. To me, it the feature is there, it should work. It was the main feature I liked. I know the TUSL2 doesn't have it. If it did, it would probably work ;) <---wink.

How about we table this till I get the TUSL2? I'll give my honest opinion of it as well.

Truce?



:) :) :) :)

[Flu!d]

You have to remember that Asus have a seperate bios for all their boards....You can't use the bios for the TUSL2-C on the TUSL2....With the Gigabyte you can use the bios for the higher spec board on the lower spec board....

Truce?! I was never fighting with you I just feel that you were 'very' quick to purchace, find fault, rectify fault and RMA the Gigabyte board due to ungrounded complaints that may scare potential purchacers away from an otherwise excelent mobo....:D

I class you as a mate (in Aussie terms!)....You cool!

[oldfart]

Well, you seem to want to go on forever about this thing. I thought you would have at least been polite and agreed to wait until I can compare the two boards. Apparently not.

I'm done until I get the TUSL2. As before, I'll post my honest opinions about it as well.

Out

[pappypete]

According to UPS tracking I'll have an XET-C board tomorrow.
I decided to order this board based on the features they
(Gigabyte) advertise. I can say this much, if the dividers do
not work, I'll feel cheated. I want what I paid for!!!

[Flu!d]

I have been polite....That dosen't mean that I have to 'lay down and die' though....I too have a point to get across and I have copped quite a bit of PM flack over this so I have reason to sound 'abrupt'....But I have never been rude to you....

[Flu!d]

Pappypete....Where do Gigabyte state that the dividers are present let alone work! Judging by the evidence it appears that the dividers, in fact, do not work 'on the XET-C' (the 1/5th and 1/6th ones that is)....

They appear to work fine on my XET though (higher spec discontinued board).

Be prepared to vid pin on whatever make of 815EP board you choose....On the XET-C it will fix the resetting problem (if you get it with your chip).