CURE TO A7V PROBLEMS

Regalk

Golden Member
Feb 7, 2000
1,137
0
0
Sheesh trying to post this on Anandtech took me 4 tries sheesh
For everyone's information
OK everyone, I had an opportunity to speak with an ASUS representative at a college computer meeting this past week.... Heheh...you can bet I drilled him with comments and questions regarding the A7V, its lack of stability, and the lack of online support ASUS displays.

I've listed below, in order of importance, the reason stability issues occur (including USB failures, 3D lockups, boot problems, compatibility, etc..) according to the source himself..yeah...


1. READ THIS---MOST IMPORTANT---I CAN'T STRESS THIS ENOUGH..... ...READY....

!!!!!!!!!HEAT!!!!!!!!!!

Thatz right people...this is the number one solve to most lockups. I'll explain the reasoning first, and then explain why, and then what it should fix.

Reasoning (according to ASUS): "Unless you physical measure the temperature of the CPU manually, using an external thermal sensor device on the top side of the chip, the readings you will receive in the BIOS will be completely incorrect. As you will notice, just by upgrading BIOS revisions, a noticable change in CPU and MB temperature reported will occur. This misleading fact is the number one source of motherboard returns, error reports, crashes, etc. The fact is, the A7V has never really been the problem...it's the Thunderbird and Duron chip (not the Duron as often). Before I get into specifics, know this...if you are running any BIOS beyond the original revision, using any motherboard revision using the in-socket thermal sensor, take the reported heat measurement according to ASUS probe and add 5-10 degrees (eg if system is reporting CPU = 38c, then add 5-10 degress, therefore CPU = 43-48c). "

WHY DOES THIS CAUSE A PROBLEM? "All newer CPUs are equipt with set thermal power-down sensors. Once they reach a certain temp...BOOM...they shut down. Unfortunately, AMD does a poor job of regulating this temperature for each chip it produces. This results in some chips shutting down at 52 C, and others shutting down at 75 C (Right now, most of you should be coming to understand why some people are experiencing more lockups than others...chances are the people not experiencing the lockups are using proper cooling and were lucky enough to receive a chip with a high thermal shut-down temp.). These temp problems are the most difficult for a user to figure out. Mainly because the probe is inaccurate, and second, because the lockups occur at different times and or intervals. What you should notice is that lockups will mainly occur in high-load applications (eg. 3D games, etc.) where the application itself drives the temp of the CPU up. Sometimes, if a CPU is operating below its shut-down temp by only a small number of degrees (say 48 C), a 3D application running for 5 minutes or longer may push that temp beyond the thermal maximum (say 52 C in this case) resulting in an automatic CPU shut-down, and obviously the system will lock! May I remind you that this has nothing to do with the A7V, or any other Socket 462 mainboard, yet we receive the majority of the complaints. The best way to test for this as the main stay of your lockups is to simply perform the following:
a) Keep the system case open (for the test period)
b) Make sure you are using a high quality cooling device, such as an ORB or other high power fan)
c) Make sure ASUS probe reports less than 45 C when IDLE!!

Now, after having met this criteria, run a high CPU intensive game or application. If the lockups disappear you have just found the source of all your problems. If they continue, then other troubleshooting steps can be used, and or the MB or CPU is faulty.
This should fix lockups in 90% of the cases (occuring after the user has made all the appropriate updates involving patches, BIOS upgrades, drivers, etc.)

SOLUTION SHOULD FIX:
- All 3D lockups (including looping sound issue)
- All IE lockups
- All Windows 2000 IE crashes
- USB power down problems
- Windows 2000 boot lockup issues
- Green vertical lines lockup (only some users experience this)
- Desktop lockups (some users only)

WHAT IT WILL "NOT" FIX:
- Resource-caused lockups (eg IRQ, DMA, etc.)
- Specific application compatibility problems (eg. Games or applications that may need a patch to work with a specific video card, etc.)
- Some of the mouse scroll issues reported
- Some of the 'power-intensive' USB issues caused by browning out external USB components (DSL modems anyone??)
- Driver problems (obviously!)

2. EXPANSION CARD PLACEMENT: "Due to poor design infrastructure, the A7V is very picky when it comes to adding expansion cards. Without getting into specifics, problems arise when placing devices in both PCI slot 1 and PCI slot 2. Just know that slot one shares an IRQ with the AGP (as noted in the manual, and in these forums many times before), and that having two high priority devices (eg. Sound Card and 3D AGP card) share one hard-wired interrupt request line is not the best of is not the best of ideas. We should have mentioned this in our documentation, but we did not realize that it would cause the problems that it did, as well as politics and pride are always involved in an acknowledgement of a design mistake. PCI slot 2 shares with the on-board Promise UDMA-100 controller, and absolutely no other devices should be sharing the hardwired interrupt if you plan to connect any drives to it"
Slots 3,4,and 5 should be fine, but it is recommended that sound cards, video capture cards, and or any other high priority multimedia devices be placed in PCI slot 3. This is the only hardwired expansion slot on the A7V that runs independently to the CPU (non-sharing)

***Do not get the above solution regarding hardwired IRQ lines mixed up with software IRQ sharing (they are completely different, and for the most part, have nothing to do with each other...at least on a hardware level)***


3. WINDOWS 2000 IS AGGRESSIVE: "Windows 2000, when operating on any system equipt with an ACPI BIOS, will automatically scan and configure all plug and play, and even non plug and play devices. The administrator actually has no control over this agression, and unlike Windows 95/98/ME, the only real way to reconfigure IRQs is to turn ACPI BIOS off (can only be done during an installation, and it only works on some systems). If not, the Windows 2000 will configure all standard IRQs to their appropriate settings (eg. System Timer = 0, Keyboard = 1, etc), however, all expansion components will be shared off of IRQ =9 using the new timing-sharing IRQ software routing control strategy, that Windows 2000 so proudly shows off. People paranoid of conflicts (including myself) should relax for the most part. If lockups are occuring, and no conflicts are apparent in the device manager (other than everything using IRQ 9....not a conflict, but just a weird configuration or sharing), then the problem is likely to be driver, software, BIOS, or something else related, but not the actual Windows 2000 implementation.

WHY DOESN'T ASUS POST THIS $HIT ON THEIR WEBSITE? "heheh...because technical personel are supposedly knowledgeable regarding these issues, and everyone else would have difficulty comprehending the subject matter. Obviously, I now know this is not the case. But you must realize that 99% of the time, it is user ignorance that is the cause of the problem, as opposed to specific hardware, software, or other technical information, or lack there of, in this case. Our support personel usually are able to isolate lockups, crashes, and all other errors to poor OS installation, lack of updated drivers, failure to apply important software patches, etc. Only when all this fails do we assume that some type of device error is responsible, but even then it is still unlikely, and still usually caused by user ignorance toward a certain procedure. Hopefully this clears things up"

OK people...there you have it.

All the systems that I configured, using the A7V, are now 100% stable. In my case, it was the heat that was the issue. As most of you may have read in previous posts, I had tried everything else. Now I'm running the system that was locking up in Windows 2000 IE 5.5 and in certain 3D games completely 100% stable (running at ASUS Probe measurement of 42 C idle, 47 C max load)
Before, the system, due to poor ventilation was running at ASUS Probe measurement of 53 C idle, 58-59 C max load. I know this was high, but most of the people I spoke with (other than those using Chrome Orbs or other superior fans) were running at around the same temp. Now I know!! Holy $hit...can you imagine...I was probably running the CPU at 58 C + 5/10 C = 63/68 C on full load! Damn...now that is HOT! No wonder I got lockups in IE 5.5

**Important --- Lockups due to heat are more likely to be apparent in Windows 2000. I'm currently doing research to find out why, but it has been agreed upon that Win2K is much more heat sensitive than any other operating system out there.**
 

Dulanic

Diamond Member
Oct 27, 2000
9,965
590
136
99% of this is total BS..... Bios 1003 and 1004 where buggy plain and simple.... 1005A fixes 90% of all problems people have. You can never trust a sales representative.
 

Mem

Lifer
Apr 23, 2000
21,476
13
81
I think it`s strange I can run my MSI K7T Pro at 49c all day running RC5 etc under full load which probably means add another 5 or 8c on top of this for the true reading with no lockups & still be rock stable if MSI can make stable boards at with these temperatures why cannot Asus do the same thing,btw I`ve a crappy case as well.So I too think it`s BS.

:)
 

Gogga

Member
Feb 6, 2000
177
0
0
So you are implying that my Alpha cooler is total trash :p
Please ignore this post.....
 

chiwawa626

Lifer
Aug 15, 2000
12,013
0
0


<< Before, the system, due to poor ventilation was running at ASUS Probe measurement of 53 C idle, 58-59 C max load. >>



Everyone knows ur not supposed to have that high of a temp dumass, howlong did it take to fiugre it was a heat issue?
 

cmaMath13

Platinum Member
Feb 16, 2000
2,154
0
60
Regalk,

Thanks for the info. I know my lock-ups were caused by the BIOS is was using (now I use 1005A).

I guess it is a good thing that I have good ventilation!

See picture:

cmaMath's cooling
 

Sunner

Elite Member
Oct 9, 1999
11,641
0
76


<< Everyone knows ur not supposed to have that high of a temp duma >>



Im running mine at about the same temps(A Duron600@900, 1.75v).
Working fine, of course its hotter than I'd like, so I am looking for a good case with better ventilation possibilities, as my current case doesnt have any good places for fans.
 

Adul

Elite Member
Oct 9, 1999
32,999
44
91
danny.tangtam.com
I have yet to have a problem with poor cooling. I am anal about cooling and always use a good HSF. I have yet to use the Asus MB, I have so far use the abit KT7, MSI K7T pro, pr0 2, and Pro 2a, and the Soyo KT133 MB ( can't remember the name), and I have yet had to have a lock up related to heat issues. All these system have or are running RC5 stable.
 

chiwawa626

Lifer
Aug 15, 2000
12,013
0
0
tbirds go around 60 by spec, but when u get em in the 50s its not plesant. best to keep them anywhere in the 40's
 

Jonny

Golden Member
Oct 26, 1999
1,574
0
76
So guys, now that we are on the topic of the A7V, is it worth it for me to download 1005C? I must say my computer is almost completely perfect right now, Iam using 1005, and Iam not sure that C will fix the little issues Iam having now.

Basicly, what I want to know is, is it as good/better or worse than 1005?
 

NoreagaCNN

Banned
Sep 28, 2000
2,267
0
0
Uh actually I download the 1005a BIOS and it fixed everything, havent had any problems since, I also heard that the 1005c fix all the other crap that wasn't fixed before.

Peace,
Nore
 

jaydee116

Member
Jan 10, 2001
48
0
0
One question I have is if the CPU shuts down dosn't everything go blank instead of just a lock up??? If so how would this solve lock up problems? Also I have switched bios versions with no change to the temp on the sensor. All versions have the same temp readings. Sounds to me like the PR Rep is talking his product up which he is supposed to do.

 

SKATAN

Junior Member
Jan 12, 2001
6
0
0
the problem with the bios prior to 1005a was a bad memory configuration enabled when using optimal settings

called:
-SDRAM SPECULATIVE READ
(it was enabled in the 1004d and older bios )

I compared a image of wpcredit saved using 1004D to a image of 1005A

with 1005a all the problems I ad where gone now i can say it is a fine motherboard

1005c seems even better NOW:

113mhz fsb super stable MEM 150mhz 2/2/2/5