If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
|
#1
|
|||
|
|||
msi ti 4600 & asus p4c800e deluxe no powerup
hello,
in short: I have the above hardware and the motherboard refuses to powerup (at all) with this video card in. full story: I used this video card with an abit vp6 motherboard and it worked great after sorting out the drivers issues. recently I decided to upgrade to an asus p4c800e deluxe mobo. the specs: asus p4c800e deluxe, pentium 4 3.2ghz, kingston 1gb memory, video msi geforce 4 ti 4600 vtd, toshiba dvd rom, sony dvd rw, ibm 20g hd, wd 120g hd, enermax 430 watt psu, hitachi 630 crt monitor, case and psu fans are connected to the pins on the motherboard. so now, let me try to explain what exactly happens. I also have a msi geforce3 and an ati pci video cards that the system boots up with fine. the geforce 4 works fine in the abit vp6 mobo but refuses to work in asus p4c800edlx and asus p4b533. I tried to trouble shoot before posting and I can't come to a resolution. after reading the manuals, the 2nd asus mobo lights up a red led on it, which, going with the manual, says that the card is 3.3v. so I figure that both of the asus mbs recognize this card as if it was 3.3v. yet I know the geforce4 is a 1.5v, it's keyed for 1.5v and it also works flawlessly in the abit vp6 mobo. On the other hand the asus pc probe shows that the agp slot supports 3.3v. I don't have any other cards plugged in any of the pci slots. I tried booting up with just the video card and it just refuses to powerup. the p4c800edlx lights up a green led to show the atx+5v runs through the mobo but when I press the power button nothing happens. I wrote to supports@msi on 2apr2004-no response yet, I contacted asus but they keep repeating exactly what I write to them, after 3 emails with them they said: yes looks like it doesn't work. I upgraded to the latest bioses and drivers, it looks that there has got to be some kind hardware conflict which I don't understand exactly. I bought this msi geforce4 card over a year ago and it cost me $450 so I don't really feel like getting another video card. I would appreciate any comments or suggestions as to what I should do. thank you. |
#3
|
|||
|
|||
(Paul) wrote in message ...
In article , (dluxea) wrote: hello, in short: I have the above hardware and the motherboard refuses to powerup (at all) with this video card in. full story: I used this video card with an abit vp6 motherboard and it worked great after sorting out the drivers issues. recently I decided to upgrade to an asus p4c800e deluxe mobo. the specs: asus p4c800e deluxe, pentium 4 3.2ghz, kingston 1gb memory, video msi geforce 4 ti 4600 vtd, toshiba dvd rom, sony dvd rw, ibm 20g hd, wd 120g hd, enermax 430 watt psu, hitachi 630 crt monitor, case and psu fans are connected to the pins on the motherboard. so now, let me try to explain what exactly happens. I also have a msi geforce3 and an ati pci video cards that the system boots up with fine. the geforce 4 works fine in the abit vp6 mobo but refuses to work in asus p4c800edlx and asus p4b533. I tried to trouble shoot before posting and I can't come to a resolution. after reading the manuals, the 2nd asus mobo lights up a red led on it, which, going with the manual, says that the card is 3.3v. so I figure that both of the asus mbs recognize this card as if it was 3.3v. yet I know the geforce4 is a 1.5v, it's keyed for 1.5v and it also works flawlessly in the abit vp6 mobo. On the other hand the asus pc probe shows that the agp slot supports 3.3v. I don't have any other cards plugged in any of the pci slots. I tried booting up with just the video card and it just refuses to powerup. the p4c800edlx lights up a green led to show the atx+5v runs through the mobo but when I press the power button nothing happens. I wrote to supports@msi on 2apr2004-no response yet, I contacted asus but they keep repeating exactly what I write to them, after 3 emails with them they said: yes looks like it doesn't work. I upgraded to the latest bioses and drivers, it looks that there has got to be some kind hardware conflict which I don't understand exactly. I bought this msi geforce4 card over a year ago and it cost me $450 so I don't really feel like getting another video card. I would appreciate any comments or suggestions as to what I should do. thank you. AFAIK, the Asus burnout protection circuit is connected to the TYPEDET pin on the AGP connector. If TYPEDET says the card would prefer 3.3V, then the red led lights up, and powerup is gated off. I don't think anything else influences how this works. Some manufacturers have made mistakes when building video cards. The AGP standard clearly states, to make a logic 0 indication on the video card, TYPEDET should be connected directly to ground. Some manufacturers use a low value resistor to make the connection instead (as the manufacturing test engineers prefer the use of a resistor, as it eases testing the card at the factory). If the motherboard injects too much current into this pin, in an attempt to test whether the pin is logic 0 or logic 1, then with the resistor in place, an indeterminate level can result. Apparently, this is enough to trigger the burnout protection. If the resistor has accidently been scraped off the board, then the wrong level results. (I know about that, because I've managed to rip two resistors off video cards while handling them...) If it was my card, I would use an ohmmeter, to check whether TYPEDET is connected to GND on the video card via a direct short. A little piece of wire soldered to the right two points would help ensure the card is sending a good level to the motherboard. This is not an end-user procedure, and if your video card has a warranty, you might try returning it, in the hope that a later version of the card fixes the issue. Sometimes, the reason for the malfunction, is a fault in the burnout protection circuit on the motherboard. As you've reproduced the problem on two motherboards, odds are the video card is at fault, and not the motherboard circuit. Ref: See page 41 of: http://developer.intel.com/technolog...0_final_10.pdf The connector pinout is on page 50. I first learned about this issue, with respect to a Matrox card that used a resistor for the TYPEDET pin. On the poster's motherboard, instead of AGP 1.5V, the video card managed to convince the motherboard to provide around 2V, which is incorrect and presumably due to not having a good short directly to ground as instructed in the standard. HTH, Paul First of all, thank you so much for your insight. I didn't get a resonable response from neither asus, msi or msi forum, which tells me that the manufacturers don't really care about end users, just the companies they sell their products in bulk to. The card was not damaged in any way, scraped or anything like that, it's in immaculate condition, yet refuses to work. The card is no longer under any warranty so I could do anything to it. But I feel like I should take this issue up with the maker, msi, who didn't respond to the email I sent for over a week now. And the forum, independent of msi, didn't respond either. So now, if msi refuses to work with me on the card: Do you think they could fix such issues within asus mobo bios ? or is there anything I could do by myself to the card to make it work ? Again, thanx so much for your reply. |
#4
|
|||
|
|||
In article ,
(dluxea) wrote: (Paul) wrote in message ... In article , (dluxea) wrote: hello, in short: I have the above hardware and the motherboard refuses to powerup (at all) with this video card in. full story: I used this video card with an abit vp6 motherboard and it worked great after sorting out the drivers issues. recently I decided to upgrade to an asus p4c800e deluxe mobo. the specs: asus p4c800e deluxe, pentium 4 3.2ghz, kingston 1gb memory, video msi geforce 4 ti 4600 vtd, toshiba dvd rom, sony dvd rw, ibm 20g hd, wd 120g hd, enermax 430 watt psu, hitachi 630 crt monitor, case and psu fans are connected to the pins on the motherboard. so now, let me try to explain what exactly happens. I also have a msi geforce3 and an ati pci video cards that the system boots up with fine. the geforce 4 works fine in the abit vp6 mobo but refuses to work in asus p4c800edlx and asus p4b533. I tried to trouble shoot before posting and I can't come to a resolution. after reading the manuals, the 2nd asus mobo lights up a red led on it, which, going with the manual, says that the card is 3.3v. so I figure that both of the asus mbs recognize this card as if it was 3.3v. yet I know the geforce4 is a 1.5v, it's keyed for 1.5v and it also works flawlessly in the abit vp6 mobo. On the other hand the asus pc probe shows that the agp slot supports 3.3v. I don't have any other cards plugged in any of the pci slots. I tried booting up with just the video card and it just refuses to powerup. the p4c800edlx lights up a green led to show the atx+5v runs through the mobo but when I press the power button nothing happens. I wrote to supports@msi on 2apr2004-no response yet, I contacted asus but they keep repeating exactly what I write to them, after 3 emails with them they said: yes looks like it doesn't work. I upgraded to the latest bioses and drivers, it looks that there has got to be some kind hardware conflict which I don't understand exactly. I bought this msi geforce4 card over a year ago and it cost me $450 so I don't really feel like getting another video card. I would appreciate any comments or suggestions as to what I should do. thank you. AFAIK, the Asus burnout protection circuit is connected to the TYPEDET pin on the AGP connector. If TYPEDET says the card would prefer 3.3V, then the red led lights up, and powerup is gated off. I don't think anything else influences how this works. Some manufacturers have made mistakes when building video cards. The AGP standard clearly states, to make a logic 0 indication on the video card, TYPEDET should be connected directly to ground. Some manufacturers use a low value resistor to make the connection instead (as the manufacturing test engineers prefer the use of a resistor, as it eases testing the card at the factory). If the motherboard injects too much current into this pin, in an attempt to test whether the pin is logic 0 or logic 1, then with the resistor in place, an indeterminate level can result. Apparently, this is enough to trigger the burnout protection. If the resistor has accidently been scraped off the board, then the wrong level results. (I know about that, because I've managed to rip two resistors off video cards while handling them...) If it was my card, I would use an ohmmeter, to check whether TYPEDET is connected to GND on the video card via a direct short. A little piece of wire soldered to the right two points would help ensure the card is sending a good level to the motherboard. This is not an end-user procedure, and if your video card has a warranty, you might try returning it, in the hope that a later version of the card fixes the issue. Sometimes, the reason for the malfunction, is a fault in the burnout protection circuit on the motherboard. As you've reproduced the problem on two motherboards, odds are the video card is at fault, and not the motherboard circuit. Ref: See page 41 of: http://developer.intel.com/technolog...0_final_10.pdf The connector pinout is on page 50. I first learned about this issue, with respect to a Matrox card that used a resistor for the TYPEDET pin. On the poster's motherboard, instead of AGP 1.5V, the video card managed to convince the motherboard to provide around 2V, which is incorrect and presumably due to not having a good short directly to ground as instructed in the standard. HTH, Paul First of all, thank you so much for your insight. I didn't get a resonable response from neither asus, msi or msi forum, which tells me that the manufacturers don't really care about end users, just the companies they sell their products in bulk to. The card was not damaged in any way, scraped or anything like that, it's in immaculate condition, yet refuses to work. The card is no longer under any warranty so I could do anything to it. But I feel like I should take this issue up with the maker, msi, who didn't respond to the email I sent for over a week now. And the forum, independent of msi, didn't respond either. So now, if msi refuses to work with me on the card: Do you think they could fix such issues within asus mobo bios ? or is there anything I could do by myself to the card to make it work ? Again, thanx so much for your reply. The burnout prevention circuitry won't let the board get any power, so no BIOS setting will affect this in any way. What I was attempting to do in my post, was explain that the problem is at a pretty low level in the hardware. The fault in cases like this can be in the little transistor circuit on the motherboard, or can be caused by the video card manufacturer using a resistor on the TYPEDET# pin, when the pin is supposed to go directly to ground, as an indicator that the card prefers 1.5V. If two motherboards with burnout prevention won't start with that video card, then I would blame the fault on the video card. If just one motherboard with burnout prevention won't work, it is a tossup as to which one is at fault. Take it to a shop, and ask the shop to check the voltage on the AGP pin 2A TYPEDET# signal. It should read zero volts when the card is plugged in. If it reads higher than that, then that will be the reason the burnout prevention is tripped. The most reliable way to fix it, would be to solder a wire from TYPEDET# to a GND pin. (I am unaware of any other logic drive solutions for that pin - it should either be grounded or left open. Using a resistor between GND and TYPEDET#, as some video card manuf. use, is wrong.) The problem will be finding a shop that is comfortable doing a mod like this. If you need backup documentation, the AGP specs I've got were from: ftp://download.intel.com/technology/...oads/agp20.pdf http://developer.intel.com/technolog...0_final_10.pdf Since this is an implementation disagreement between two manufacturers, the only end user option when the warranties are exhausted, is to sell the defective item (TI4600), and buy a later one that doesn't have the same problem. Here is an example of some end users fighting with a similar issue on some early version G400 cards. Obviously, the resistor number they refer to on the G400, will be a different number on your board. This is just to confirm that companies do make this mistake with TYPEDET#. (Notice how "Haig" and Matrox refuse to correct the board for their customers.) http://forums.matroxusers.com/showth...threadid=30954 This article shows what happened before Asus used the burnout prevention circuit. Better for the motherboard not to start, than for this to happen... http://www.vanshardware.com/articles...9_i845_AGP.htm HTH, Paul |
#5
|
|||
|
|||
In article ,
(Paul) wrote: If it was my card, I would use an ohmmeter, to check whether TYPEDET is connected to GND on the video card via a direct short. A little piece of wire soldered to the right two points would help ensure the card is sending a good level to the motherboard. This is not an end-user procedure, and if your video card has a warranty, you might try returning it, in the hope that a later version of the card fixes the issue. Word of warning, for you lurkers out there. My suggestion to do this is NOT an invitation to fool around with every video card you can lay your hands on! To start with, refer to Figure 2 on this page: http://mirror.ati.com/support/faq/agpchart.html There are three kinds of keyings, on AGP video cards. 1) Card is 3.3V keyed only. TYPEDET# will be open. Card will not fit in a 1.5V motherboard, so there is no reason to even try modding that card. The card will only fit in older motherboards anyway. 2) Card is 1.5V keyed only. This would be a very modern AGP card, suitable for AGP 3.0 4X or 8X only operation. TYPEDET# is already grounded. If you plug this into a modern motherboard and you cannot get the motherboard to power up, then check the video card, to see if TYPEDET# is grounded. 3) Card is universal (1.5V or 3.3V operation possible). My assumption in my previous post, is that the OPs card is a universal card, with the two slots cut in it. Further, it is not one of the cards known to be miskeyed. So, for example, some of the "bad" cards in the vanshardware list may have accidently had the 1.5V slot cut in the edge card, when they are not really 1.5V cards. TYPEDET# would be open on a card like that, and if you wire TYPEDET# on what is really a 3.3V only card to ground, you will be preventing the Asus burnout circuit from doing its job - your motherboard would get destroyed! http://www.vanshardware.com/articles...9_i845_AGP.htm Consequently, check Google for cards known to burn out modern motherboards, due to the fact that the card was mis-keyed. (Roland probably has a better list of the cards that are mis-keyed, and has corrected me in the past.) The Ti4600 is modern enough, that I know it is not mis-keyed. Cards which burned out early P4B motherboards would be examples of the "bad" ones. 4) Card is universal (1.5V or 3.3V operation possible) and really does support both 1.5V and 3.3V operation. AFAIK, at least Geforce3 or later cards would qualify. If one of these cards has left TYPEDET# open, and your modern Asus motherboard won't start, because it thinks it has detected a 3.3V only video card, this would be a candidate for the mod. As I said previously, when the motherboard won't start, it could be a fault with the motherboard or with the video card. If multiple modern Asus motherboards won't start with that video card, that increases the chances the problem is with the video card. At that point, I would take an ohmmeter, and verify that TYPEDET# is connected directly to GND, and mod if necessary. Case (4) is the one I am applying to the original question. So don't go out and ground every video card you own, without checking out why the mod is needed first. Applying the mod to an SIS305 for example (case (3) above), would mean instant death to your modern Asus motherboard. Here is some clarification of the situation with the SIS305: http://groups.google.com/groups?hl=e....supernews.com Paul |
#6
|
|||
|
|||
(Paul) wrote in message ...
In article , (Paul) wrote: If it was my card, I would use an ohmmeter, to check whether TYPEDET is connected to GND on the video card via a direct short. A little piece of wire soldered to the right two points would help ensure the card is sending a good level to the motherboard. This is not an end-user procedure, and if your video card has a warranty, you might try returning it, in the hope that a later version of the card fixes the issue. Word of warning, for you lurkers out there. My suggestion to do this is NOT an invitation to fool around with every video card you can lay your hands on! To start with, refer to Figure 2 on this page: http://mirror.ati.com/support/faq/agpchart.html There are three kinds of keyings, on AGP video cards. 1) Card is 3.3V keyed only. TYPEDET# will be open. Card will not fit in a 1.5V motherboard, so there is no reason to even try modding that card. The card will only fit in older motherboards anyway. 2) Card is 1.5V keyed only. This would be a very modern AGP card, suitable for AGP 3.0 4X or 8X only operation. TYPEDET# is already grounded. If you plug this into a modern motherboard and you cannot get the motherboard to power up, then check the video card, to see if TYPEDET# is grounded. 3) Card is universal (1.5V or 3.3V operation possible). My assumption in my previous post, is that the OPs card is a universal card, with the two slots cut in it. Further, it is not one of the cards known to be miskeyed. So, for example, some of the "bad" cards in the vanshardware list may have accidently had the 1.5V slot cut in the edge card, when they are not really 1.5V cards. TYPEDET# would be open on a card like that, and if you wire TYPEDET# on what is really a 3.3V only card to ground, you will be preventing the Asus burnout circuit from doing its job - your motherboard would get destroyed! http://www.vanshardware.com/articles...9_i845_AGP.htm Consequently, check Google for cards known to burn out modern motherboards, due to the fact that the card was mis-keyed. (Roland probably has a better list of the cards that are mis-keyed, and has corrected me in the past.) The Ti4600 is modern enough, that I know it is not mis-keyed. Cards which burned out early P4B motherboards would be examples of the "bad" ones. 4) Card is universal (1.5V or 3.3V operation possible) and really does support both 1.5V and 3.3V operation. AFAIK, at least Geforce3 or later cards would qualify. If one of these cards has left TYPEDET# open, and your modern Asus motherboard won't start, because it thinks it has detected a 3.3V only video card, this would be a candidate for the mod. As I said previously, when the motherboard won't start, it could be a fault with the motherboard or with the video card. If multiple modern Asus motherboards won't start with that video card, that increases the chances the problem is with the video card. At that point, I would take an ohmmeter, and verify that TYPEDET# is connected directly to GND, and mod if necessary. Case (4) is the one I am applying to the original question. So don't go out and ground every video card you own, without checking out why the mod is needed first. Applying the mod to an SIS305 for example (case (3) above), would mean instant death to your modern Asus motherboard. Here is some clarification of the situation with the SIS305: http://groups.google.com/groups?hl=e....supernews.com Paul hello again, and all I can say is WOW, UNBELIEVABLE, and thank you so much. I would never be able to find all this information, and to make my case strong enough to msi. I wrote them an email explaining all that I've learned. Hopefully they'll find a resolution to it and if they don't then I'll have no choice but to manualy adjust my card. I read thoroughly the posts of others, so far it doesn't all fit in my head especially that the big companies would do such things. anyway, I'll wait for an answer from msi and post it right back once that's out of the way. I've done things with my amigas, soldering on the motherboard as well so I'm not afraid of the procedures you're explaining. Again, I appreciate all your help and thank you. Talk to you soon. Jack |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Is ZALMAN CNPS7000A-AlCu compatible with the ASUS P4P800 Deluxe motherboard? | [email protected] | Homebuilt PC's | 4 | April 9th 04 10:19 PM |
Asus A7N8X-E Deluxe weird graphic performance | Hinnerk van der Felde | Asus Motherboards | 2 | February 13th 04 10:32 PM |
ASUS A7N8X Deluxe rev2 - What's rev 2? | OrmesbyJohn | Asus Motherboards | 2 | February 12th 04 09:55 PM |
P4C800-E Deluxe + asus WiFi-b Card problems - Advice sought.. | Mod | Asus Motherboards | 10 | January 13th 04 04:19 AM |
Asus A7N8X Deluxe or Abit NF7-S Which? Choosing a MB. | James Paraskeva | Overclocking AMD Processors | 11 | November 26th 03 06:16 PM |