AIW ATI9700pro problem (help pls)

[CVNet1]

I recently got a AIW R9700pro from the Buy/Sell/Trade forums. However I haven't been able to successfully post with this card. Here is what happens. After connecting everything up and plugging in all cables, I press the power button. I see for a split second that the network light flashes on and the fans just barely start to spin and the power shuts off. If I unplug the 4-pin power connector from the AIW ATI R9700pro, the PC will power on without trouble and I get on the screen the error beeps and the expected message "Please consult your installation guide to properly install your card msg". I have tried this video card using 2 different power supplies (Both are Antec True Powers). The first was only 430W so I thought it might be over amping or something, but the second is a 550W so I am beginning to suspect something is wrong with the card. Like I said earlier, I press the power button and the computer doesn't start up, but if I unplug the power connector from the video card, it powers up with the error message. Has anyone had this problem before? Or maybe someone might have an idea what this could be.
Just for reference, I had a Radeon 9500np running in this same system 4 days ago and was able to softmod and OC it to 350/300 (which is greater than stock 9700pro). This card ran just fine and never exhibited any problems, so I can't figure out what might be keeping this AIW 9700pro from working. I'm going to try the card in a couple other rigs (one nforce2 and another KT400) today and maybe It will work, but I kinda am wondering if something is wrong with the card at this point. Anyway if you have any ideas/suggestions or similar experiences please let me know.
BTW this is not a FOC problem with motherboard (if it were, then the computer wouldn't power up when the 4pin cable was unplugged from the video card)

Here's my system specs.
Chaintech 7NIL1 (nforce2) MB
2500 Barton
1gig Geil Golden Dragon PC3500 DDR
60 gig Maxtor HD
430 & 550 Antec True Power PSU
LG 16x10x40x16x CDRW/DVD
SB Audigy
Linksys 802.11A/B/G wireless etherenet

thanks
--John

 

[CVNet1]

Bump....
I hope someone sees this and has an idea what might be wrong....

 

[Metsubushi]

John,

Believe it or not, i am having the EXACT SAME PROBLEM. i bought all my stuff from Newegg.com and everything but the ATI a-i-w 9700 pro work! i am testing the card in someone else's box later tonight. we think it's either a problem with my Epox motherboard's (e8rda) AGP slot OR it's the actual vid card itself. mine won't even power the fans up when the card is plugged in to the AGP slot but when i power it off i can hear the PS (antec true 430) whistle. this rig will boot with a PCI radeon card. Now, i have heard from a friend that there are some issues with ATI radeon a-i-w pro cards and Nforce2 chipsets. i can't say one way or the other but this is definitely a fishy situation we're in, eh john?

my system specs :
ATI radeon 9700 a-i-w pro
AMD athlon xp 2600+
Epox e8rda motherboard (Nforce2 chipset)
512 mb Corsair DDR400 ram
120 gb maxtor hd
antec 430w PS
Lite On 52x32x52 cd-rw
Lite On 16x dvd
Lian Li case

hope this helps, i am in the same boat. once i get this card tested out tonight i will reply again and let you know what i find out.

--Chris

 

[CVNet1]

Chris,
Let me share with you what I now know about the AIW 9700pro cards.
I called ATI tech support the other day, and asked them specifically about how many amps this card pulls on powerup. After discussing with them what components I was using in my computer (same as described in my original post), I was transferred to a Senior Tech Support Specialist as they called themselves. He told me that the original design for the AIW 9700pro cards had a problem which caused it to pull way too many amps on powerup -- Specifically he said that this "Is a Documented problem that we are aware of". This problem he went on to tell me was fixed by a redesign of the PCB (I don't know what was changed or anything, I was just told this by this ATI support guy). I was given an RMA # for the card and instructed how to proceed through their RMA process. I am now in the middle of said process. According to the support agent(s) I spoke with, this is a documented problem with their ATI AIW 9700pro cards. I have noticed on many forums (such as Rage3d) that a list of incompatible power supplies have been put together with regard to this AIW 9700pro. As it turns out, almost all of them have the Overcurrent cutoff protection feature, just like my 2 Antec PSU's have. That is why I was experiencing the specific problem of pressing the power button and only a split second of power going through before it was being cut off. What I would suggest you do, is to goto ATI.com and register your product (this can be done by looking around the support section and they'll ask you to type in your Serial# or Part# (can't remember which one exactly it was) and you'll set yourself up with a username and then you can proceed to call thier toll free line and they will give you support for your card you've registered. The other important thing is to tell them exactly how you turn on the power and before the fans even spin up at all, the power cuts off. Also stress the PSU you are using. The support person I talked to originally was aware that Antec PSU's and the True Power line are a quality brand and not some cheap OEM. Ask if they have any idea how many amps the card pulls at powerup and suggest that what you are seeing looks alot like its triggering the overcurrent protection on your PSU. The card specifically utilizes the +5volt rail on your power supply from that floppy connector. And your 430W Antec has 36amps on this rail. If you consider this in wattage, that is a total of 180Watts to give just on that rail. The design of the Antec True Power provides that each rail has is on its own transformer, so you don't have a situation where the 3.3V and 5V rails are combined wattage or anything like that. In order for your PSU's overcurrent protection to be tripped, your system has to pull greater than 47Amps on the +5Volt rail which is essentially 235 Watts. If I were you I would register with ATI and give them a call on the toll free line and politely bring all this information to thier attention. It worked for me, I should have a new ATI AIW Radeon 9700pro from the redesign that will work properly within 2 weeks of now.
I'll let you know how my experience turns out,
--John

 

[Metsubushi]

John,
that is really, really helpful. i am going to test that card out tonight in another machine just the same and then register for help from ATI. i dug this -> http://www.theinquirer.net/?article=8250 up. apparently there are also problems with the Nforce 2 chipset (i am not sure if you are already aware of this). thought you might be interested. i'll keep you posted.

--Chris

 

[PlatinumGold]

I just installed an AIW 9700 from ATI on my A7N8X board. install whent without issue, i tested it with 3dmark2003 and it bmarked at 4069, which is ok.

 

[rbV5]

I wonder how all those 9700's are working in the XPC boxes with 200 watt PS's?, the so called description of the problem would seem to imply that you couldn't even use the card in an XPC solution, but we all know that is false. Theory Sounds like BS to me. My NF2 boards (both ASUS) have worked great with my AIW 9700 pro, with 300 watt smartpower and 480 watt truepower antec power supplies. I hope it (RMA) fixes your problem, but I wouldn't hold my breath if its based on those assumptions.

 

[CVNet1]

I'm glad you have had success with running this card. Its good to here. Nothing against you man, but I could care less if you think its BS. I presently have an RMA in progress and an ATI tech support employee who told me this. I didn't pull this out of my ass or something as you imply. Perhaps you should read my post before telling me I'm spouting off BS.
Maybe it would help you to know for further reference that I already had a Radeon 9700pro running in said system that wouldn't run the AIW 9700pro. And on top of that I had an OC 9500np that was softmodded and running at 350/300 functioning perfectly on that same system that the AIW 9700pro would not work in. And since we're on the topic of questioning what will work, let me reiterate (and maybe you'll get it this time) that my problem was isolated completely to powerup and the .5 seconds that the computer power was on before overcurrent on the PSU from the floppy connector that was plugged into the card shut it all down. If you want to you can look up problems similar to mine and you'll find a whole host of them with regard to powerup problems and overcurrent protection being triggered by this specific card. There are even hack fixes posted on the different forums that will fix the overcurrent on powerup by creating a circuit delay to the auxillary power connector on the AIW 9700pro card. So yes, you got a good card, and like I said I'm glad for you. But just maybe in the remote realm of possibilities, your card is based on the new revision I spoke of earlier and my card is not. Sure its way out there, but just maybe thats whats going on, and just maybe some people out there have purchased cards for thier SFF boxes (which I never said PSU's for those boxes couldn't handle the card) that were the newer revision too and didn't have or exhibit this overcurrent problem I've been speaking of.
One more thing, since you've annoyed me so with your post which mis-identifies the issue I speak of. This problem has absolutely nothing to do with Voltage or with the PSU's max wattage. Its simply an Overcurrent (read AMPS) issue which triggers my PSU's overcurrent protection circuits and cuts off power before damage is done to the PSU. If you don't have overcurrent protection, there won't be any problems exhibitted from your powerup, except that you might, as a guy I've been talking to has, get to see the magic blue smoke escape 2x 350W PSU's that were used to power this card (the one I RMA'd).

 

[rbV5]

If its well documented problem, link it.

I didn't pull this out of my ass or something as you imply. Perhaps you should read my post before telling me I'm spouting off BS.

I read your post completely, you imply that the AIW 9700 pro is tripping the overcurrent on your powersupply by exceeding 235 watt draw on powerup. You provide no links to back your claims, and I know that for a fact that folks are running complete systems including an AIW 9700 pro with 200 watt power supplies. I know for a fact that my NForce2 board and AIW 9700 pro ran fine with a loaded PC including multiple drives (including RAID) multiple tuner cards, on a antec 300 watt smartpower power supply, and does now with an antec 480 watt power supply. You insulting me because I have doubts about your claims or about what some "tech" support personel at ATI's excellent tech support

means nothing to me, I just want to see something other than your word over this so called " well documented" issue. I'm not saying there is nothing to what you said, I just know it can't always be that way...otherwise the card would not work in the situations I described abouve. How do you explain that?

 

[CVNet1]

maybe if you read:
Here
Here
Here
Here
Here
Here
oh-and-Here
Heres-one-for-your-SS41G2-SFFs
There are more, I even found a place where there were good descriptions diagrams and schematics on how to design this power delay device that I spoke of. However in the limitted time I searched, I was unable to come up with that thread. I guess I can post it later when I feel like looking through my webpage history more.

Hope this helps you to realize I'm not BS'ing you

 

[rbV5]

I'm not saying you're BS'ing me, I just don't think you're right, but I hope your RMA solves your problem. Every one of those links is simply another discussion on a forum about seemingly similar problems. They all also have the same ambiguous "fixes"...some things work on some rigs...the same setups work fine for most users....not all the posters are reporting the same problem....on an on. You haven't even recieved the "updated" AIW to see if it "fixes" your problem yet, and I hope it does. I thought maybe you had some kind of link to the actuall "well documented" problem rather than a bunch of forum discussion links.

 

[CVNet1]

finally found the delay cicuit thread (links in here will give you the schematics and other such workaround designs)

Here

the specific Well documented statement was a paraphrase or direct quote (can't remember which) that I took from my conversatoin with the ATI tech support specialist. It was not an indication of me having a wealth of documentation that I used to prove to the ATI guys that this was their fault, it was merely a statement made in a support conversation by one of the ATI guys.

 

[CVNet1]

I just recieved the new ATI card today (BTW this greatly surprised me because the card I sent them will arrive today at ATI and I was told ATI doesn't do Advanced RMA replacements). Turns out the part # on this card matches the part # from my original card (the Serial is different though), I know this cause I had to register the part # and serial # on ATI's website so I just went there and looked up the part# I had typed in. This didn't look good, but I gave it a shot. NO GO unfortunately. rbV5 I guess you are correct that the supposed redesign I was told existed doesn't really or maybe the support guy was referencing something else. Anyway, this is the same design and revision of the card I had earlier judging from the listed Part#. However, even though I wasn't able to power on directly when I tested the card, I was able to get a power on and post when I started with the floppy connector disconnected and immediately upon pushing the power button connected the floppy connector to the card. I did this for testing purposes only and would not recommend anyone to do this for a regular power up. Having had my hopes of getting a card that worked right out of the box dashed, I will now turn to the Delay Circuit Relay that will almost certainly solve my problems completely (judging from the success with the floppy connector test I tried above.) The overload on the PSU like I have said before is occurring within .5 seconds of Powerup. If you delay power from reaching the card for .5 second, you will have yourself a workaround for this problem. These people Here seem to have the proper electrical information and knowledge to provide guides for this sort of thing. Perhaps they will even sell a working delay circuit to someone who is not certain he/she can complete this modification.

---John

 

[rbV5]

finally found the delay cicuit thread (links in here will give you the schematics and other such workaround designs)

Like the other threads you linked, it seems speculative at best. Doesn't show an AIW design flaw at all. Itshows "maybe" a convoluted fix to "maybe" a video card problem on "some" setups that work fine for most.

EDIT: Bummer on the RMA results. I'd be carefull about fuburing your MB before trying some weird "fix".

 

[CVNet1]

I'll try the fix (not likely it will fubar the MB as it will have no direct connection or anything really to do with the MB). All it will do is connect to a molex or floppy PSU connector and then run a delay circuit that will plug directly into the video card. The most likely candidates for damage would be the PSU and the video card if something went horribly wrong. But I'll get all the information on how to build this thing before I try it. And I'll keep you posted on if it works for me and anything else relevant.