Keep in mind I am editing the pages in this as I go along... each day you may find new issues and fixes added to posts on the first few pages of the thread so don't assume the answers are at the end of ther thread... if you have been here before, recheck the first pages before you go hunting.
==============================================
NOTE: This thread is designed for the construction and comments about issues and fixes for the A7V333 MEANING, there will be no debating, flaming or asking for help in this thread.
If you wish to make a comment or ask a specific question about the build or know of a definite issue and a definite fix for it, please feel free to post it below.
PLEASE, DO NOT ASK FOR ASSISTANCE IN THIS THREAD OR I WILL DELETE THE POST. Post for assistance in the forum under a new thread. This one is dedicated to discussion about specific issues and fixes and not questions about individual problems and 'How To's".
Welcome to the A7V333 Build Thread
The vote at the top of the forum has been tallied and the forum has chosen the A7V333 as the motherboard they would like to see set up.
The A7V8X came in second and from what I understand we have a few members tinkering with the board. I am sure they will post their findings as they go along. I may post a thread for the A7V8X so the users of the newest Asus board will have a place to collect specific information.
I prefer to purchase products that have been on the market for a bit. Not that I would wait 6 months or longer but I like to wait till at least the 2nd or 3rd hardware revision of a motherboard before diving in and buying. That way the majority of the bugs have haunted someone other than me
==================== The A7V333 (R)
By Member: Klondikekit and members of the Asus Motherboard Forum
The following thread will not only be my experience but the experience of the members of this forum who wish to add information to the build of an A7V333-R motherboard system.
I expect I will be ready to start building and posting as I go somewhere around 9/14 or 9/16/02. I will post all of my findings as I go along and and create an entire permanent document from both my findings and all the helpfull additions and tips you post in this thread.
I will start by listing the parts I will be using. I decided I wanted to keep the total purchase under 900.00 but still wish to have some performance to work with. For this purpose I choose:
Total cost for the parts I need to purchase out of the above list will be $823.80. Should I decide to go 512 on the memory stick, it will be about $900.00. EDIT: I went with the 370MHz 512 stick. I prefer the lower timing spec of the 370Mhz rated stick over the 400. It looked to me that the 400 was really just a 370 with slower memory timing allowing the module to hit 400MHz.
I will go through the entire RAID setup and may flash to a different RAID BIOS for testing. I will not use any SCSI in order to keep the resources free for the RAID setup.
stay tuned!
To Be Continued...
Last edited by Klondikekit : 01-19-2003 at 02:37 PM.
The first thing I do when I decide what products I will be getting is go to the manufacture websites for all the products and download:
The most current manual
IMPORTANT!: Manual Updates
IMPORTANT!: CURRENT SUPPORT FAQ Information
The most current BIOS and drivers
Any firmware updates
Any software updates
When building, if you do not take the time to check out all the above, you are asking for trouble, especially if you are buying a 1st revision motherboard, new to the market. The guys with the A7V8X RAID board know that very well.
So, after visiting the websites for the: ANTEC SOHO File Server Tower, SX1040BII 400watt PSU, I found the manual. I choose this tower because I like the front panel USB and firewire access along with the fan layout for plenty of airflow. I intend to expand with SCSI so I will need alot of room and a tower that will cool the drives properly. I usually purchase a PSU from PC Power & Cooling however the PSU included with the tower is designed for a basic server and after looking over the specs it would appear the PSU was designed with high demand and stability in mind. From calculating the needed power for all the devices I may eventually use, the total 5v+3v watt rating will be fine for future expanding. Overclocking with this tower will allow for much in the way of cooling possibilities and plenty of power to run the fans if needed. For what I would normally spend for a decent tower and PC P&C PSU, (about $300.00), this was a deal at $100.00
A7V333 Chipset ATX Motherboard w/ RAID Retail Boxed, I found the manual, BIOS/driver updates and updates to the manual itself as well. I choose the RAID version because for the extra 22 bucks you get firewire and 2 more IDE I/O's and even if you do not use the RAID features it was worth the extra coin to have the options available. I went to Asus and the AMD website and from both sites (not just one or the other) confirmed that the A7V333 with BIOS 1007 was certified to run the XP2200 processor.ALWAYS VERIFY BEFORE YOU BUY
AMD XP 2200 Processor Retail Box w/HSF, nothing specific but there are updates from MS that will be included in WinXP SP1 so I will round up everything I need from Microsoft's download center for Windows and Office XP and burn it to CD. You should always have all the service packs and updates for the OS you are using ready to install or download and install them as soon as the OS is installed. Also, I think the retail box HSF will work just fine with the airflow of the server tower selected. As stated above, I verified at both AMD and Asus the 2200 was certified to run on the A7V333.
GIGABYTE RADEON 8500LE PRO, found a BIOS update (although I will NOT flash it unless there is a problem), found latest WindowsXP drivers from ATI.. tweak programs available as well. This will be the first ATI based video card I have worked with in a while. ATI has always been a pain in the past and I decided to see if anything has changed. Usually I get a Nvidia based card and if a decent one had been available for under $120.00 I probably would have gone that route but the idea was to test the A7V333 with anything I could think of in the way of problems to throw at it so I decided an aftermarket ATI card should fit that test quite nicely.
CORSAIR XMS 256MB, nothing to get. I decided to test the Corsair memory product. I usually go for Crucial but in this case, again, I am testing. I may grab a stick of Samsung to test as well.
2 MAXTOR 40GB 7200RPM, I found the latest utilities and a manual. In the past Maxtor has never let me down so I will stick with them for this build. For SCSI, I like IBM and Fugitsu.
1 US Robotics FAX/Modem, I found updated drivers 1 Logitech Marble Mouse, I found updated drivers 1 Microsoft Keyboard, nothing to get 1 Set of Boston Acoustic Speakers, nothing to get
Should you have any comments or suggestions about the list, feel free to post them below
R.T.F.M.
(Read The Friendly Manuals )
While you are waiting for your parts to arrive, take the time to read up and prepare for the build.
While at the websites I review the support area FAQ's about any current or old issues, review the forums about common issues, then based on what I have found I may alter the drivers, BIOS or firmware I have collected HOWEVER I ONLY USE the official versions of the above unless I have a problem while I am building and know exactly what to use OR need to experiment to correct it, otherwise I DO NOT just run out and use the most recent beta BIOS or hacked BIOS and drivers.
Remember... BIOS flashing WILL COMPLETELY VOID YOUR WARANTEE so don’t go whacko on BIOS upgrading and hacked (unofficial) drivers. Just stick to the most up to date official drivers that are available for your devices at first.
Performance tuning and testing comes later... the first order of business is get it booted, set up and running stable under standard, normal conditions. The first thing I will be checking when the motherboard arrives is the BIOS version shipped with the motherboard. AMD and Asus certify the board with BIOS 1007 and above so if the board ships with 1007 or higher, I will NOT flash it unless I see a need to after the build is complete and the flash will either correct an issue I am having or will add a significant performance or support gain.
To Be Continued...
Last edited by Klondikekit : 09-11-2002 at 11:57 AM.
Now that I have all the nomenclature, updated BIOS's, drivers and service packs I have renamed all the files in order to make sure I understand what each file is when I am ready to start installing. I also burned them to a CD.
In the process of collecting the files, I have started to take notes about any issues the updates and drivers 'may' correct. I will continue to build the system using these notes so when finished I will have compiled a complete 'Do and Don't' notebook for future reference.
As time goes on and new fixes, drivers or BIOS's become available I will edit the notebook with the positive or negitive outcomes from using newer drivers, BIOS’s and/or any procedures such as motherboard jumper settings, BIOS or Windows settings or register changes.
By keeping good notes I will always have a blueprint to work from in case I need to reinstall from scratch or I need to refer to a procedure in order to reverse something I did in the building or tweaking process.
What I would like to do now is ask everyone here who has had a problem with the A7V333, (RAID version or not), or has a fix for a problem that they know works to please post it in this thread as we go along.
Examples of problems and fixes would be:
a. Manual misprints and the updates
b. Jumper issues and fixes
c. BIOS issues and fixes
d. RAID issues and fixes
e. Windows issues and fixes
f. Common mistakes / misconceptions
g. Anything you can think of that would help
The above should all be specific to the A7V333(R).
To be continued...
Last edited by Klondikekit : 09-10-2002 at 04:08 PM.
I will start by posting all the information I could find about the A7V333 directly from Asus in their monthly motherboard FAQ area:
=============================
Q: The BIOS update failed with the EZ Flash function in A7V333. My BIOS version is 1003. What's wrong?
A: For users using BIOS version 1003 and 1004, please update your BIOS version to 1005 or later by aflash under dos to fix this problem. BIOS version 1005 or later can use EZflash without problems.
(I suggest you DO NOT USE anything but Aflash in DOS)
=============================
Q: I am benchmarking A7V333 with 3DMark2001SE but it won't finish the test. Is A7V333 compatible with this popular benchmarking software?
A: The problem may occur with earlier BIOS versions for A7V333. Please upgrade the BIOS to version 1005 or later to fix it.
=============================
Q: On My A7V333, after I flashed the Bios to 1006 and restart the system, I get no post and the post reporter saying "CPU test failed¨. How can I fix this?
A:Please follow the steps to solve this problem:
1. Please use Jumper mode instead of JumperFree mode.
2. Use the Aflash utility to flash the bios to version 1007.
3. After flash the bios to 1007 can set the Jumper back to JumperFree mode
(I suggest you DO NOT USE anything but Aflash in DOS)
=============================
Q: Why my iPanel Deluxe cannot be installed on my new P4B266 or A7V333?
A: We started to change the AAPANEL connector to the new IAPANNEL connector which is following Intel's spec from current new motherboards, e.g. P4B266, A7V333, etc. Due to AAPANEL and IAPANEL are different pin assignment, we provide a converter within the current shipment of iPanel Deluxe.
For the customers have previous iPanel Deluxe without the converter but need to install it on new motherboard, he can order the parts as the P/N below:
Q: I have Asus A7V333 motherboard with Win98 OS, after I install the USB 2.0 Driver from the support CD M222, I still get the question mark on my devices manager, I do not know what happened?
A: Please download the newest USB 2.0 driver version 2.0.2 from the Asus download site and update the driver to fix the problem.
ASUS download site: http://download.asus.com.tw
=============================
Q: Sometimes when I boot up my A7V333 motherboard, the keyboard is not recognized when I enter the BIOS Setup or when in Windows, the keyboard will not be recognized, but Windows does not give a error for "No Keyboard."
A: The problem could be caused by the power supply not providing enough standby power to the USB and keyboard. Please check in the manual on page 23 to change the jumper setting from "5V standby" to "5V," or set to pins 2-3 to disable standby voltage to power KB and then reboot to make sure the problem no longer present.
=============================
Q: The A7V333 user manual says I need a minimum of .75-.85 Amps for +5 Volt standby, but my system won't power on and tech support suggested that I use a power supply with 3-5 Amps +5Volt standby. Why does there seem to be a difference with the manual?
A: Computer systems are getting more complicated and offer more features like wake or power-on by keyboard, or wake-on-LAN or on-ring from your modem. All of these features are powered from the +5 Volt Standby Voltage and the configuration of the system will determine the requirement for the Minimum Required Voltage from the power supply +5 Volt Stand by.
For Example:
A basic system with no features needs a minimum voltage of .85-1.0 Amps. When any of the below features are used, additional Amps are added to the overall Minimum Voltage Requirement.
Condition Minimum Voltage Requirement
Wake-On-LAN even if not being used, but connected + .35-.5 Amps
Wake-On-Ring even if not being used, but connected + .35-.5 Amps
Wake-On-Keyboard USB or PS2 + .35-.85 Amps
Confused? Its simple.... Make dang sure your PSU meets at a MINIMUM a 180-200watt 5v+3v= spec and a 2 (or greater) amp 5vsb output. Any 5VSB jumpers must be set to active on the motherboard for standby features to function. It is best to let Windows handle the power management. If you set it in the BIOS and Windows, conflicts will usually occur. Let Windows handle it. Assuming the PSU is as specified above and the motherboard is not defective, most standby conflicts are video driver related and sound card (usually SBLive) related.
=============================
Q: I have Asus A7V333 motherboard with bios version 1003. When I used the Ez-Flash to update my bios, it showed "fail" message during the process, and then the board could not boot up any more. I don't know what happened?
A: The A7V333 motherboard with Ez-Flash function may cause some boot up problem, but with BIOS 1005 or later version will solve it.
(I suggest you DO NOT USE anything but Aflash in DOS)
To be able to fix the Ez-Flash, please use Asus Update or Aflash utility with bootable disk to update the BIOS to version 1005 or later, and be careful not to use the Ez-Flash function to update BIOS with BIOS version 1004 or earlier versions.
(I suggest you DO NOT USE anything but Aflash in DOS)
=============================
Q: My system will not start properly with Windows 9X and ME even if I have 1 Gigabyte (GB) or more memory installed. Also, if I tried to start the computer, I will receive an error message, "Insufficient memory to initialize Windows". However, it runs fine if I use 512 Meg. What is wrong?
Q: Sometimes, my system will just restart with 1.5 Gig of RAM on Windows 9X or ME, or the Video Driver will not load in windows, other times, I get a error message that I do not have enough RAM when I already have 1.5 gig of memory installed. What causes this to happen?
PLEASE NOTE: To correct the above memory issues in Windows ME and 9X, do the following:
If you use more than 512mb of memory in a Win9X or ME OS, add this to the system.ini file under the [vcache] heading:
[vcache]
MaxFileCache=32768
MinFileCache=32768
Both entries must be present and the above syntax must be adhered to with no spaces in the line. Then save and reboot
You do not need to assign more than 16384 but some systems will function better for graphics @ 32768. The multiplier for any entry is 1024. 32mb x 1024 = 32768
You can increase this number if you do heavy A/V editing/processing, otherwise 32mb is fine
With the above addition, the graphics aperture in the BIOS should be set to a MAX of 128 for GF cards, 64 for ATI
Next, It has been noted that it is best to populate memory slots 1&3 when using 2 sticks instead of slots 1&2. Stability and performance issues have been noted when using slots 1-2 and moving the stick in slot 2 over to slot 3 corrected the problem.
=============================
Q: With an A7V333 and PC2700 memory why it still running at 266Mhz not 333Mhz? How to make it run at 333Mhz?
A: Try setting the CPU/MEM Ratio to either [Auto] or [4/5]
=============================
Q: I just installed SP1 for Windows XP and I have lost the ability to play DivX 5.0.2, some games and other AVI file formats along with sound problems.
A: SP1 has been confirmed to cause problems with the above mentioned items. At this point in time there are no fixes for the problems however I am sure as time goes on fixes will surface. Microsoft seems to have shoved the update out on the market before fully testing it (as usuall). I would suggest you not install SP1 unless you know for sure it will solve a specific issue for your system OR there have been enough reports that the problems generated by the update are fixed.
=============================
SINCE THIS POST NOW EXCEEDS THE MAXIMUM CHARACTERS I CAN POST INTO IT, I WILL CONTINUE THE KNOWN ISSUES AND FIXES BELOW
If I get a chance, I will look through the monster A7V333 thread and see what I can transfer to this thread in the way of known problems and fixes.
Continued After The RAID Post Below...
Last edited by Klondikekit : 09-17-2002 at 02:24 PM.
This section will be used to post the different RAID issues and fixes I find. To start it off, apparently those with RAID are finding that installing George Breaze's (sp) Beta .20 RAID patch is correcting write performance issues. One member has listed his method of RAID installation:
Quote:
Originally posted by StinkyPete
OK, here's what I did for my system:
1. Start with clean blank drives:
I had a issue with mine & had to low level format them, but I wouldn't start there. While the drives are still connected to your regular IDE ports, boot a floppy & delete all partitions on the drives.
2. Move them to the Promise controllers:
Using ATA 66/100 cable's put one drive on each controller. Make sure the blue end is attached to the motherboard & use the other end connector (not the middle connector) for the hard drives. For IBM's use cable select, others set as master.
3. Set up your Array:
On first boot the Promise BIOS should tell you no RAID arrays defined, press {something} to enter setup. Choose the defaults which should be Desktop & Performance. There are many who use the "Lumberjacker" BIOS's & change their stripe size etc... But I heard if you do you can't defrag & you then need to either use FAT32 or install XP to another disk & format the array from XP. I'm doing fine with the defaults, so I never bothered with that stuff.
4. Boot your XP CD & start the install.
If you havent yet, make a floppy with your Promise drivers on it (check other threads for info). When the install first starts press [F6] to load additional drivers & later press "S" & put in your floppy.
5. Let XP do the partition and install:
Choose install to the empty disk & choose NTFS FULL format, not quick. If you don't have any other hard disks on your regualr IDE channels, there can be a LONG pause after the format, be paitent. ALSO if you see any errors at this point try another CDROM drive. I have to use my CD Burner to install XP (RAID or not). It's easy to do in our BIOS, just change the boot config to the other CDROM drive.
6. If you had the long pause during install, you will have the same booting XP. Once install is all done, go into system properties, IDE Controllers, & change "Auto" to "None" for all channels where there is no device connected. For me this is the second device for each channel.
NOTE: this is your regular IDE channels, NOT your RAID Array. I have a DVD ROM as master on IDE1 and a CD Burner as master on IDE2, so I set slave for both to "None".
7. Let XP do all it's updates & get all your hardware drivers installed.
8. Install the latest VIA 4 in 1 drivers.
9. Install George Breaze's (sp) Beta .20 RAID patch
IMPORTANT!! SEE THE POST MARKED RAID UPDATE BELOW
10. Change PCI latency from 32 to 64 in the BIOS.
11. Turn on large disk caching in XP.
That's all I can think of right now, I don't have the PC or my notes here at work. Plus there are many tweaks to XP itself & you have to find the best NVidia dets for your setup, but this should get you up & running with RAID. If I think of anything else I'll post again tonight. You should be able to find more info about the things I went over quickly (like large disk caching & George' patch) by reading other threads here and in the Storage forum. Don't forget the Search function here, VERY usefull!
Good Luck,
--Pete
I have also recommended the George Breaze's (sp) Beta .20 RAID patch for data corruption issues and in that case it corrected VIA to RAID file transfer corruption when using the "Lumberjack" BIOS.
Many thanks to StinkyPete for posting the above.
Also, you may wish to review this thread if you are new to installing RAID:
Originally posted by Anthony2816 It seems some people are having troubles altering their raid ports into full raid or non-raid functionality using the Lumberjack bios' available here:
Here's a list of detailed steps on How To Do It. In this case, I chose to convert my WinXP A7V333's raid ports into a standard Promise non-raid ata133 IDE controller, using the Lumberjack 1008U bios, but the steps are similar for his other bios':
1) Go to device manager and uninstall any Promise hardware.
2) Search your Windows folder for any instances of "ultra.inf" and "ultra.sys", and delete them. (You may not have any).
3) Place the Promise drivers (ultra.inf, ultra.cat & ultra.sys) that came with the Lumberjack bios into a temp folder.
4) Reboot and flash the Lumberjack 1008U bios.
5) Disconnect all drives from the raid ports, and reboot.
6) When Windows detects your Promise controller, direct it to your temp folder so it will install the Promise drivers that came with the Lumberjack bios.
7) Check Device Manager, which, under SCSI and RAID controllers, should say that you now have a "WinXP Promise MBUltra133 (PDC20276) (tm) IDE Controller.
8) Power down and attach the drives of your choice to this new controller. It will now run up to 4 IDE devices, including CD, CDRW and DVD drives.
Remember, have fun..but flashing your BIOS, especially to a hack version, means you bought the board.. no warantee
======================================
This has been posted a few times....
Quote:
I pressed F6 to install 3rd party RAID drivers. The driver screen comes up and It gives me a list of 4 options of drivers 2 for XP. One is the MB Lite and the other is the SB Lite.
I choose either the MB or SB and I get an error message...
File \WinXP\Fasttrak.sys caused and unexpected error (1024) at line 2108 in d:\xpclient\base\boot\setup\oemdisk.c.
The fix for the above was a defective floppy drive however if swaping out the floppy unit does not work for you there are also settings and other possible causes:
also, check setting:
Onboard ATA device First (is set to) {NO}
SCSI/ATA100 Boot Sequence (is set to) {ATA100/SCSI}
Load Onboard ATA100 BIOS (is set to) {AUTO}
Your RAID jumper is set properly and the hard drives are on 2 separate blue tip ATA100 cables' each going to the 2 separate RAID IDE's and nothing else is on each cable but 1 hard drive, both on the black connector, blue to the motherboard... and both drives jumpered as 'cable select" or "Master". Try master first and switch to cable select (on BOTH drives) if master does not work.
If the above is true and they are Western Digital Hard drives, remove the drive jumpers completly out of both the drives and retry the WinXP driver installation.
ALSO:
Quote:
Originally posted by Bala I had this problem too. In your Asus CD-ROM find the PDC20276 / Setup folder and use MakeDisk.exe file (in this folder) to create a disk with the raid drivers. Then when installing XP press F6 soon after the boot and that will work fine.
Many times the problem is how the drivers are place on the floppy disk. You must either use the Makedsk.exe method or if you download new / updated drivers, unzip the files (not copy the zip file) directly to the floppy disk so the files in the root directory of the floppy are:
fasttrack
readme.txt
txtsetup.oem
along with the folders containing the drivers for each OS.
If the drivers are not laid out on the floppy properly, Windows will not find them and error codes will come up.
======================================
I will continue to add to this post as I find more on the subject. Feel free to post anything you may wish to add about your own experience/fix in this thread.
To Be Continued...
Last edited by Klondikekit : 10-25-2002 at 10:18 PM.
Q: Should I set JP1/JP2 to a higher voltage than the Asus default?
NOTE: MY A7V333-R (revision 1.04) CAME SHIPPED WITH BOTH THE JUMPERS SET TO 1-2 AND NOT THE DEFAULT SETTING ASUS HAS SET IN PAST REVISIONS. AS THE MUSHKIN SITE SUGGESTED, ASUS HAS NOW CORRECTED THE DEFAULT JUMPER SETTING TO BE PINS 1-2 FOR JUMPERS JP1 and JP2.
A: The Mushkin site is correct, the voltage increase on JP1 and JP2 can damage your memory IF the memory you are using does not have proper heat sinks, is of poor design and you greatly exceed the voltage specification for DDR memory products. There are other factors involved. From what I have researched, newer designed memory products have a voltage regulator system built into the modules and therefore you are not really stabilizing anything by raising the voltage via JP1 and JP2. With memory modules that incorporate the voltage regulator you can indeed damage the module should the voltage increase be maintained for an unspecified period of time, substantially higher than the input threshold specification for the circuit. The heat dissipated due to the regulator trying to maintain 1.8vdc on the internal core will cause more harm than good both performance wise, before the module burns up, and after when the module is of course fried-dried and laid to the side. Also, with the newer types of memory modules the system can actually become UNSTABLE with an increase in voltage. That is why Mushkin reported the minor performance increase with a proper (lower) DDR voltage setting on the A7V333. The increased heat dissipation by the voltage regulator created by the higher JP1 and JP2 voltage makes the modules run less efficient. Like anything electronic.. if it gets hot, it runs like a dog and usually dies a slow painful death.
Do not increase JP1 and JP2 past pins 1-2 unless you have high performance memory that comes with a good heat spreader and even then the highest setting I would go is:
JP1 = 1-2
JP2 = 2-3
counting from the bottom of the stack as the motherboard sits upright in the tower. In my tests I found with the Corsair XMS memory, raising these as shown above actually netted no significant stability increase and indeed stability dropped after the system looped 3D benchmark tests over a period of time. The consequences are not worth the tiny increase in FSB you may get for a short period of time. I will leave them on pins 1-2
In case you didn't know, Asus already overclocks the motherboard FSB from the factory (to give the board an edge during hardware site benchmark testing) and adding to that formula a memory module which is not of good quality and does not have a proper heat sink on the module, exceeding the proper DDR voltage can indeed cause REAL WORLD performance (not benchmarks.. benchmarks are fantasy) to decrease.
I personally have always run Asus motherboard memory voltage just a touch on the hot side.. but NEVER past that.
In this case I will be running at the PROPER DEFAULT SETTING of pins 1-2 for JP1/JP2. We will see if it is stable at higher FSB later as we go along.
=========================================
A Bit About XP Processor Unlocking From StinkyPete...
Quote:
Originally posted by StinkyPete Just a quick interjection here:
So far as I know ALL TBred 2200+'s come factory unlocked. Just as my 1.33GHz TBird was, and as all the MP processors are. That is, you do not need to paint, solder, or do anything to the L1's in order to be able to change multipliers on these chips.
HOWEVER, while "factory unlocked", the 2200+'s only have the 13x to 20.5x multipliers avaible to them. Since we overclockers want to run the FSB as high as possible with an "unlocked" chip, 13x is not a low enough multiplier for doing this. SOOOOO, we want to change the "range" from 13x-20.5x to 5x-12.5x.
(These pics may load VERY slow)
THAT's what this "pin trick" is doing. Changing the available multiplier range, not "unlocking" it. And I don't care what your BIOS tells you it can do, you can have one range or the other, not both!
2100+ users have the same issue, but they ALSO have to "unlock" their processor in addition to changing the multiplier range.
Note, when you do this, 12.5 will be the HIGHEST multiplier available to you. Choosing 14x will just go "around the horn" to (I think) 6x; 15x will really be 7x, etc...
If you don't do this, choosing, say 10x, will go "around the horn" too, and really be something like 18x, and I've yet to hear of a 2200+ that runs at 18 x 133 = 2394MHz!! Even with a screaming Delta fan! (& lets face it, you probally won't try 10 x 133, you will be trying something like 10 x 180, getting 18 x 180 = 3240MHz!. That's asking a LOT from a chip rated at 1800MHz, don't ya think?)
There seems (to me) to be a lot of confusion about "unlocking" ever since the 2100+ forced the multiplier range change. I think we need a new word for "switching ranges" that makes it clear this is seperate from "unlocking". I'm sure TBreds will start coming out "factory locked" soon (I hope I'm wrong, but the 1.4 was locked!) and those will need "unlocking" as well as "range changes" just like the 2100+ now does.
We are gonna probally hit this again when the first 21x multiplier chips come out, except then we will WANT 13x-20.5x available to us!
Does anyone know if the Hammers will change this "limit" of the chips? I doubt it, as InHell Pentium chips have the same issue.
A7V333 NIC Problem
Hey guys...A co-worker of mine has stumped me with a very bizarre problem...
He just built a new system:
AMD 2100+
Asus A7V333
Corsair XMS DDR
He installed Windows2000 and had everything working great....that is, except for the netword card. Anytime he tried to install the drivers, it would kick out and the driver screen would give the error message "bad data"....
He has since tried the following:
1) Tried moving the NIC to various PCI slots and looked for IRQ-conflicts
2) Tried Netgear, D-Link, and 3Com NICs.....no luck..same problem with each
3) Installed SP3 with no luck
I have no clue what's going on here.....anyone have any ideas?
Thanks ahead of time...
ANSWER: (This may be a solution to anyone having problems installing any type of hardware with similar symptoms as listed above on the A7V333)
Quote:
Originally posted by Ned Slider Pelly,
Could this your problem (data is invalid error when installing device driver)??
Q: I used ASUS LiveUpdate to flash BIOS on A7V333, but my motherboard doesn't boot after flashing. What can I do?
A: Basically, if you flash BIOS 1006 or earlier using "ASUS LiveUpdate" under Windows rather than "aflash" under dos, you may receive failed to boot problem after flashing BIOS
under JumperFree mode. However, the BIOS chip is not dead. You may simply configure it to Jumper Mode to make it work. Otherwise, you can try clearing the CMOS.
=============================
Q: Do we need to install IEEE 1394 driver on Windows XP for A7V333 motherboard?
A: Windows 98SE/ Me/ 2000 and XP have built-in IEEE 1394 driver. It does not require users to re-install.
=============================
Q: The PostReporter gives the message "system failed CPU test" on A7V333. Why?
A: A7V333's system failed CPU test issue is caused by users carelessly setting up the CPU multiplier. If users have chosen the original CPU multiplier setting package, the BIOS can detect the CPU and pass the POST. If the setting does not match the CPU's exact speed, the POST will see this setting as false, then give out a message like "system failed CPU test".
Last edited by Klondikekit : 10-07-2002 at 06:22 PM.
There have been reports that the beta patch discussed above is causing problems for several reasons. One of which was that later revicions were activating the CPU HLT command. I personally used version Beta 4. I have been told Beta 5 is OK as well. One of our members did some research and came up with the following fix for you WPCREDIT and WPCRSET lovers:
Quote:
Originally posted by Darkje I have been reading the technotes from the latency patch 0.20beta4, and found that one of the things it does is this:
that made me think that it must be possible to set that using wpcredit, if only I knew the device code and the proper register to tweak. I remembered that powerstrip also gave info on the pci-bus and the devices connected and discovered the proper device code that way.
the register was easy to find, the only one showing the same value(in hex) as pstrip did for latency seemed to be register #0D, my suspicions where confirmed even further by looking at pci-pci bridge that wpcre recognized, that also had the latency reg at #0D. So reg #0D sets the pci latency for each device.
now I found this out, it was easy to do it:
from device 105A:5275 edit the register #0D to #F0 (w/o latency patch it is at #20)
here is a screenie with the changed register marked red.
it seems to be working just as well as the complete patch 0.20b4, but it is always good to look further... especialy now that it turns out different patch versions behave different, made me wonder what it did for the raid, and what for the rest ...
MANY Thanks to Darkje for his detective work. To recap what was said:
Quote:
Originally posted by Klondikekit WPCREDIT WPCRSET are a set of programs that can lead to great delight or great pain.
WPCREDIT is a program for viewing and editing the chipset register on the fly. In order for changes to be perminant on reboot you must use WPCRSET. You do not need WPCREDIT to make the change we are discussing here. It can be done with WPCRSET as described below.
WPCRSET must be first installed by double clicking the INSTDD.exe file then once rebooted, start the program and click the ADD button.
An input box will appear. You must enter the data as Darkje described and I posted above:
PCI Bus = 0
Device = 6
Function = 0
Register = 0D
Data = F0
those are all zeros, not O's
Click OK then Yes to reboot
If anything goes wrong in reboot and you cannot get back into windows for some strange reason, simply enter windows in safe mode, run the program and delete the entry, then reboot.
Keep in mind that ANY toying with the chipset register can be dangerous.
use wpcredit or wpcrset at your own risk
I have personally set this myself and it does work. I uninstalled the Breese patch and did the WPCRSET register change. The benchmarks were equally good.
If you cannot get a copy of Beta 4 or 5 of the Breese latency patch simply use the WPCRSET method described above.
Thanks Darkje
Continuing on....
Q:
Quote:
Originally posted by ongdennis i just setup a pc.
a7v333r
athlon 2000+ XP
512 twinmos PC2700 DDR ram
MSI GF4 Ti4200
running in windows ME
when i install the asus probe(any ver.), the installation failed saying that LMControl error? and i not able to use the asus probe (only able to see the hardware monitoring at bios).
why and how to solve?
A:
Quote:
Originally posted by Peacekeeper
That LMCONTROL proceedure error in probe is caused when you disable usb in bios. Especially using aan a7v333 though other boards can cause it as well.
===================================
Q: On my A7V333 On Page 22 the Jumper Voltage Regulator Limit (VID) What will this do when set to unlimited?
A: This will allow the VCore to be set beyond the CPU ID limit set By AMD. We suggest to leave this as default as supplying VCore beyond the physical limit may cause serious damage to the CPU.
=====================================
Q: When setting and enabling a password for either supervisor or user, if you switch to the right on the Advanced menu in BIOS and then to the left of Back to the Main Menu, the password will be disabled.
A: Please update BIOS to version 1011 or later to fix this issue.
I jumped pins 14 and 15 of the ATX power connector (green to black)...
DO NOT ATTEMPT SUCH A TEST UNLESS YOU HAVE A GOOD UNDERSTANDING OF ELECTRONICS AND/OR YOU ARE TRAINED
...to verify the PSU was working. The PSU fired up fine at which point I verified the PSU fans were both functioning. The PSU is of the Antec SmartPower, 2 fan family and I also verified the fans were moving the air in the right direction.
In review of the tower I found it to be well designed with the exception of the 80mm rear fans. The rear exhaust fans included with the tower are cheap and have a bit of a whine to them. Replacing them will be on the list of things to do since I do not like whining. I may cut out the grills that are manufactured into the tower structure and replace them with thin, round, wire-type replacements. The Antec raised, large opening type grill design is much better than the typical tiny, hole-drilled pattern and at this point I have not decided if I will be cutting them out. If after I replace the fans with decent Panaflow units the airflow noise is not to my liking I will go ahead and perform the simple tower modification. Right now, I will leave it as is.
Next, I matched all the tower connections to those in the motherboard manual. I wanted to make sure I understood where each connection was landing prior to starting actual construction effectively answering any questions as to location of the tower harnesse connectors to the motherboard.
So in reference to the tower I am now versed in the locations of all the connections and the mechanical properties of the unit as well. The PSU has now been tested and although it was not a full load test I now know it does function (in case the motherboard or tower switch will not start the PSU) and the fans are also in proper order as well.
The first order of business prior to installing the motherboard is to REMOVE ALL THE TOWER STANDOFF's ANTEC INSTALLED IN THE FACTORY CONSTRUCTION PROCESS. By doing so and starting with a blank mount surface I will avoid missing a standoff mount and shorting out the new motherboard.
To Be Continued…
Last edited by Klondikekit : 10-29-2002 at 10:30 PM.
FedX delivered the remaining components with the exception of a cable converter needed for the LCD monitor. That part may not come today and it will not stop the process of inspecting and installing components into the tower.
I will be starting the construction of the system later today and will post my findings as I go along.
First order of business is inspection of the parts. Once complete, I will get aquainted with the motherboard connections, jumpers and component locations to confirm what I read in the manual is correct. Once confirmed and/or any corrections are made I will note the corrections and start installing the motherboard into the tower.
I have gone over all the jumpers and set the system up for jumper free operation. As I have always done in the past even with jumper-free mode, I set the vcore manually to 1.75vdc.
In inspection of the JP1 and JP2 jumpers, I found Asus is now defaulting them to the Mushkin Memory recommended setting of pins 1-2. I am leaving JP1 and JP2 set to pins 1-2 until I see a need to increase the voltage due to high load crashing or high FSB instability traced back to a need for a higher memory voltage.
Also note: According to the updated Asus manual I received, (Manual Revision #IE1010B) the qualified vendor list for DDR333 memory DOES NOT INCLUDE SAMSUNG.
Nanya
Transcend
Micron
are the recommended memory modules for the A7V333
SAMSUNG AND KINGMAX WERE REMOVED FROM THE LIST
I am using Corsair XMS memory for this test.
Last edited by Klondikekit : 09-17-2002 at 09:14 PM.
Originally posted by palausk8er mm im not sure about this...but some people say its not a good idea to remove a thermal pad and apply artic silver (northbridge).
Whats your opinion?
There was no thermal pad under the heatsink of the northbridge at all. I removed the heatsink, cleaned off the area with alcohol and applied thermal compound then reinstalled the heatsink. I do not use Arctic Silver, just plain Radio Shack thermal compound. This should help increase the northbridge stability in higher FSB by allowing for a better thermal conductivity. A fan could help as well if while during high load operation the HS gets warm. I will check into it later as I test.
As for removing a thermal pad from the processor heatsink, I always perform that operation as I will outline below in describing what I have completed so far.
Last edited by Klondikekit : 09-18-2002 at 11:34 AM.
Originally posted by ongdennis For the JP1 and JP2, how we know which one 1-2 or 2-3 ?
currenly i am using Kingmax PC2700 256MB, when i set the setting to turbo, it cause window protection error. Why ? how to overcome this problem ?
Pins 1-2 are counted from the bottom of the jumper bank (as the motherboard sits in the tower) to the top in numerical order:
. . Pins 3
. . Pins 2
. . Pins 1
Jp1 Jp2
I will review all the steps and jumper settings made during the construction process below. Keep in mind I AM IN A PROCESS HERE and I may make changes as I go so just because I set a jumper one way during construction does not mean a setting will not be changed as I test for stability and other issues.
I fired up a room humidifier 1 hour prior to starting the build. This prepares the room for static sensitive components to be handled and eliminates the need for a wrist strap during the more sensitive construction processes. I build in a room that has NO carpeting on the floor. If I had to work in a room with carpet I would not hesitate to use the wrist strap. If you do not use static precautions.. good luck. It may be why your system is borked and you are here looking for help.
To start, I placed the motherboard on the antistatic bag it came in and reviewed both the manual and the updates to the manual, confirming the jumper locations as I went along.
I noticed a large amount of jumpers on this motherboard, more so than I have normally encountered in the past so to make sure I had reviewed every jumper I marked the top of corner of each one with a black marker pen after reviewing and setting the jumper as I desired for the system build. That way I would not miss one. When finished, I found 2 jumper banks unmarked, meaning there was either no reference in the manual for them or I missed them in the process.
The first unmarked jumper bank was located at the top center (as the motherboard sits in the tower) of the board just above the processor socket which had no reference in the manual for it that I could find. It was labeled on the motherboard: Overvolt
The choices were:
1-2 Enabled
3-4 Disabled
Counting left to right, pins 1-2-3
This would appear to be an overclocking feature that allows for higher voltage than normal when placed in the enabled mode. Since I would never set up a motherboard for first boot with overclocking features enabled, I left it on the default, Disabled for now.
The second unmarked jumper bank was JP1 and JP2. There is a reference for these jumpers on page 8 of the manual but it is only a location picture. The jumpers are designed to allow for memory voltage to be increased in overclocking and as noted above in another post, Asus has changed their default factory setting for this jumper to pins 1-2 for both banks. In the past it was 1-2 for JP1 and 2-3 for JP2. Be aware of that.
Also be aware that my manual revision has REMOVED Samsung and Kingmax memory from the list of qualified memory manufactures for the A7V333.
NOTE: That does not mean the memory you are using is bad.. only that Asus no longer endorses the products to be used on the motherboard.
I am using Corsair XMS and normally I would use Crucial or Mushkin memory products. I will post my findings for the Corsair memory performance later in this thread.
Switches and Jumpers
Frequency Switches
SYSCLK = All OFF (I set it for jumper-free mode)
==========================================
JumperFree (JEN) = 2-3 (JumperFree)
==========================================
RAID Controller = 1-2 (Enabled)
==========================================
ROMSIP = 2-3 (Software)
I wanted to allow the BIOS to have control of the CPU however if I encounter boot or stability issues and I am not overclocking I may set this back to 1-2 (Hardware)
The above should be 1.72 / 1.75vdc. I have always set this manually on Asus motherboards but it is not necessary to do this. If you prefer, leave them all on pins 2-3 for jumperFree operation and have BIOS control over the setting.
==========================================
Voltage regulator Output limit (VID)
VID = 1-2 (Limit)
This does not allow excess voltage emission to occur in the Vcore and limits the voltage output to the default. For overclocking I may unlimit this later but for now I will leave it limited since WE NEVER overclock any settings on initial boot and setup.
===========================================
Keyboard Wake up
KBWK = 1-2 (Enabled)
I thought I would try the features the board has to offer. Normally I do not use power management features.
===========================================
USB 2.0 Compliant Device
USB_EN = (Enabled)
I also wish to test the boards ability to handle the resources so I will be turning on most if not all of the onboard functions.
NOTE: This setting MUST be 5VSB to allow wake from S3 sleep state to occur. Your PSU must be rated @ 2A or greater on the 5VSB output.
===========================================
IEEE-1394 Device
1394_EN = 1-2 (Enabled)
===========================================
MS SD Device
MS_SD_EN = 2-3 (Disabled)
===========================================
Bass Center Setting
BCS1 = 1-2
BCS2 = 1-2 Default Settings for Type 1 Bass
This jumper may need to be changed if you have a different pin out for your speaker system than the typical industry standard. As the manual states, during the onboard driver installation and setup tests you may need to change these jumpers to accommodate your speaker system properly. I left them at default for now.
===========================================
Speaker Selector (Speech)
SPEECH = 2-3 (External Speaker)
I will probably shut this off in the BIOS since I am not into annoying messages.
===========================================
Audio Setting
AUDIO_EN = 1-2 (Enabled)
I will be using the onboard audio system.
===========================================
Chassis Open Alarm
CHASSIS = 2-3 (Jump Chassis Signal to Ground)
Your system will not boot AT ALL if this jumper is not set properly or if you use a tower intrusion sensor, it is not installed correctly.
===========================================
Line-Out Selector
BACK_LTBACK_RT =
BLOL + FLOL = jumped
BLOR + FLOR = jumped
These are for a front panel audio connection, leave them jumped unless you have a front panel access port or Ipanel with audio support.
============================================
Audio Input Line-In Header
Pins 1-2 and 4-5 should be jumped unless you intend to make Right and Left Stereo audio input connections from an audio input panel module.
============================================
Last, but not least...
CLR_RTC
Was not a jumper but just solider points on my motherboard.
============================================
Once the jumpers were set I installed the processor and the HSF. The processor is a factory retail unit and comes with a typical AMD HSF unit. What I noticed about the HSF right away is that AMD is using what appears to be a stainless heat spreader plate machine bolted with thermal compound applied between the plate and aluminum HSF unit. I did not study it very long so it is also possible it could be a copper heat spreader with a silver plating over it. In any case I removed the thermal pad with an old credit card, cleaned it up with alcohol and applied Radio Shack thermal compound to the processor die. I spread it very thin across the die of the processor with the edge of an alcohol cleaned, shiny, paper clip.
The HSF clip SUCKED ROCKS as I have seen AMD do better with the clips in the past. I modified it slightly so I could get a long handled, flat blade screwdriver nose locked into the clip so it could not slide off at all.
NOTE: If you slip off that clip the motherboard is VERY EASY to destroy or damage so it does not perform properly. I highly recommend anyone that is not experienced at installing an HSF place several layers of thick masking tape around the entire area of the HSF clip prior to attempting the installation. Taping down styrene plastic would be even better if you are a complete newbie to installing HSF clips. Cut up an old credit card and use it to protect the motherboard and do not rely on the silly tape Asus puts on the board to protect the circuits from damage.
Once the HSF was installed I installed the Corsair XMS 512 memory module in DIMM Socket #1. It was much easier to install the HSF without the memory module in place.
There was no thermal pad or compound under the heatsink of the northbridge chip at all. I removed the heatsink, cleaned off the area with alcohol and applied thermal compound then reinstalled the heatsink. I do not use Arctic Silver, just plain Radio Shack thermal compound. This should help increase the northbridge stability in higher FSB by allowing for a better thermal conductivity. A fan could help as well if while during high load operation the HS gets warm. I will check into it later as I test.
Those of you complaining of low FSB performance that REALLY have GOOD memory installed (and not some cheap junk, I mean hand picked engineering samples) may wish to look into the possibility of poor northbridge cooling since it is the northbridge that allows the motherboard to perform stable at the higher frequencies. Also.. I found the clock generator IC gets waaaaayyy too hot even at 166/33
The motherboard was then placed into the tower to check for proper alignment and fit. Upon inspection while placing I noticed the rear panel did not match the tower rear panel configuration. I had to use the Asus provided rear panel bezel since the USB2 and Audio I/O's were different than the standard ATX mobo of the last few years.
Continued In The Next Post...
Last edited by Klondikekit : 10-07-2002 at 10:05 PM.
Processors
Motherboards
Chipsets
Memory
Graphics Cards
Storage
Cases and Cooling
Mobile
Systems
Displays
Shows and Expos
(& lets face it, you probally won't try 10 x 133, you will be trying something like 10 x 180, getting 18 x 180 = 3240MHz!. That's asking a LOT from a chip rated at 1800MHz, don't ya think?)
Linear Mode
