Mike Andrawes gave Iwill KK266 9 points. A highest rating for KT133A mobos so far. What do you guys think?

[Henry Kuo]

the stability rating is across all boards (this is the only score you can compare with other boards besides the KT133A chipset). we didn't use to mention the total number of crash rate, but in the recent reviews we decided to reveal that number as well so you guys can have a better idea.

however, as you can see that the KT133A boards are getting so stable that the Winsone2000 test doesn't seem valid anymore, we are working towards another more stressful test.

 

[OS]

How is the AMI controller? Is it good? Damn, if this board had a 4th dimm slot, I'd rush out and buy one.

 

[Biggs]

"With the new Mushkin High Performance Rev. 3 memory, at CAS 2 and fast memory settings, we were able to hit 154MHz FSB without a hitch. Note that this is already 16% over the official 133MHz rating - much higher than the KT133 chipset ever achieved in terms of percentages.

By lowering the CAS latency to 3 and reducing other memory settings, we finally hit a wall at 161MHz FSB. At 161MHz, the system was still stable and we were able to run Content Creation Winstone 2000, SYSMark 2000, and Quake III Arena without any problems. At a speed of 162MHz, the system didn't even boot and a BIOS error appeared while booting. At a speed of 161MHz, that's more than 21% over the rated 133MHz."

The quote above is taken from the IWill KK266 review. Makes me wonder though, what if the same high quality memory stick was used with the other boards that were tested(particularly the Asus A7V133). Would have they gone higher on the FSB scale?

 

[Henry Kuo]

Yes, we will use the new memory on the previous boards and test them out again to find the real limit

 

[DaddyG]

Henry, I don't think that it is possible for the cpu to look at the modified multiplier. Remember, BIOS values are stored in CMOS, the cpu must initialize at some default to read the CMOS values and execute instructions to set the values, most are set in hardware registers if I'm not mistaken. The same applies to the default Vcore voltage, we have seen that increasing it in the BIOS doesn't always hack it, but upping the default by connecting the L7 bridges works much better.

Some success with 1.2's has been achieved by connecting the one open bridge (FID1) of the L6, and kicking up the Vcore with L7s.

Instead of cutting L6s for the 1.1 (all L6s closed) insulating the FID3 pin on the cpu with shellac should have the same effect. Default lowered to 7X. Pretty sure that shellac was used as the insulator on transformer windings back in the days so it should be good.

I would really like to see some testing with 'difficult' cpus on fully jumpered mobos vs Soft Menu.

What interests me is why ASUS and ABIT chose to select ALL FSBs through the BIOS and not use a 100/133 jumper. SOYO has stated out right that they retained dipswitch control over the multiplier to prevent this kind of problem. Gigabyte also chose dips, but I have not confirmed that it was for the same reason as SOYO.

 

[Dulanic]

Henry could they POST at thoose speeds tho? Thats all they need to do.... POST for a VERY short while.

 

[Charles]

Henry, Mike, or Anand:

Why don't you try Kingmax TinyBGA PC150 SDRAM to test those boards. the RAM chip will be able to run at 166MHz CAS3.

 

[SiliconVandal]

My Iwill runs great and I've had no problems oc'ing with the jumper..

 

[Henry Kuo]

Dulanic - are you refering to whether they post at 133MHz? Yes, they do. As I said, the system posted at 133 x 6 and entered Windows flawlessly (the default multiplier is 8, and we are sure 133 x 8 doesn't work at all). We also tried rebooting the sytem, shut down and restart, you name it, but it works.

If there are some other info, please let us know.

 

[ChipNOW]

Well... KK266 over A7V133 or K7T-Turbo is a big call - but I've always liked Iwill so I'm cool with it.

 

[Homer]

Seeing as I'm thinkin' real hard about buying this board, this has gotta be the best thread in the Forums right now! I think I know what I need to know. End story.

 

[Pyro]

you guys should as Charles suggested. Using 166MHz ram to find the real limit is an excellent idea (then you could do a mini review about the OCing of all KT133a boards).


BTW, I love the idea of a new stability test. I suggested it a coupla weeks ago, but no one replied to my thread (). I have another question though, how does the stability test compare to average computer usage? (you know, ICQ, several IE windows, winamp, word, 5-6 systray apps etc)

 

[Chadder007]

Here it is at good ole TcComputers.
http://www.tccomputers.com/apps/products/ppp.php?product_id=iwlkk266

 

[fitzhue]

is that the KK266 or the KK266-R?

The pic looks very much like the raid version because of the extra ide's
It seems as if they're selling the nonraid version though.

 

[Hyudra]

asus a7v133 sux, it's good if you wanna see 60 degrees celsius for a temp reading tho

 

[Henry Kuo]

Pyro - we will see what we can do about it, thanks for the suggestion. and yes, we will change to a more stressful stress tests in order to really find out about the ability of the motherboards. don't worry, we may not reply, but we listen

 

[Charles]

I still prefer Promise controller...That makes Asus a big plus for me

 

[sitka]

Crazy fsb speeds.. I'm foggy as to which board supports manual pci and agp divisors (agp volt control as well) divisors. I hear some (all) pci divisors are set automatically.

 

[Imported]

What is this shellac? Think that if I paired my 1GHz Tbird (100FSB CPU) with any of the KT133As, would it have a problem like the 1.1GHz and 1.2GHz?

 

[Charles]

illkid, You won't have any problem overclocking your 1Gig Athlon. I tried it myself in my friend's Epox 8KTA-3.

 

[VinY]

I got my KK266 yesterday and put my system together. I'm running it at 133 x 7.5. It seemed pretty stable, until I tried to install Half-Life. It reboots the machine with a blue screen of death. I cannot figure this out. Any ideas?

 

[DaddyG]

illkid, shellac was originally developed as a wood stain. Its made from crushed bugs if I remember correctly. Its pretty non-reactive, insulates pretty well and can handle temps fairly well.

With a 1gig cpu, if you have problems, you have easy options to lower the default multiplier by a simple L6 bridge connection as follows:


L6s bridges at 1 gig at 100 fsb . | : : :


L6 bridges after mod . | : : | lowers default to 6X which is a no brainer for 133. The default can be raised higher in your BIOS.


You may, in fact need to do nothing if you get the IWILL or even the MSI with the updated BIOS. Follow these steps.


With your cpu stable at 100 fsb, set the memory to HSTCLK + PCICLK to extensively test your memory at 133. If good continue.

Reset memory to just HSTCLK and set the default multiplier lower say 6X.

Boot the system at 100 with the lowered multiplier AND ALL OTHER BIOS OPTIONS SET TO DEFAULT. (very important step), if the lower multiplier is good, continue.

Move the jumper on the mobo from 100 to 133 and enjoy.

 

[Charles]

DaddyG, how can you know a lot about KT133A overclocking myth? Do you actually own a few KT133A mobos and Athlons?

 

[SiliconVandal]

Correct me if im wrong, but the green board is a KK266 and the Orange board is the KK266-R. I have a R for myself and I got my dad a non-R, and it'sm green and mines orange. The pictures seem to point to that as well.

SV

 

[VinY]

I did that, and I was able to view the bsod. Here's what it said:

STOP (then a bunch of 0x0x00xx0x000 things) then BAD_POOL_CALLER

Beginning Dump of Physical Memory


Could I be running my memory settings too high? I've got (2) 256mb Crucial PC133 CAS2 sticks.