Printable View
Big Thanks Too Catch22!! You really stepped up on this and its much appreciated.:toast:
I hear ya and I think I will start with lowering Vddq first. I will not mess with Vref at this point.
I will need to go get some 100k var's at Fry's after work. What would be the steps for Vddq? lower to 2.60v from 2.65v or go in smaller steps? Then increase o/c till I get artifacts and lower Vddq and repeat? How low can it go before problems?
As you lower it you'll get fat logs maybe and if it's too high sticks on wings. Best way i can describe it. I've ran vddq as low as 2.250 just to see what would happen. I just got done trying something for ddr. It didn't work as planned. But it was interesting and i got another idea i'll try. :)
Hey can someone try this for me. Disconnect the molex connector from the card. Then measure resistance of 5v to ground where that connector was. That back side of card would be easiest. I can't find what i believe is a fried cap and maybe i don't have one. I get 4.3 ohms btw That's kinda crazy if you ask me. Although that only equals about 21.5 watts i'd still like someone else to verify that resistance. thanks :)
Oh and it like like you'll need a higher value than 100k ohms. Maybe get a 250k ohm vario. My 100k ohm drops vddq a bit more than i'd like. Oh go in way smaller steps.
Macci!! whoa dude you got 10,008 marks in 3dmark03. Out standing my friend. I hope i helped a tiny bit. 10k i never would have believed possible with this generation of cards. Outstanding achievement there. :) :banana: :toast: :banana:
C22@Play, I can't find 1/8w resistors ( only 1/4 watt ) is that ok? and the VRs jump from 100k to 500k, will that work? I placed an order this morning , I'll call tomorrow and see if I can get some 500k Vrs added.
If MrCrusher can't ohm the 5v pin for you, then I'll pull my card in a day or two and take all the insulation off and do it! :)
btw, did you see the bridge repair Macci did.... :eek: :D
Catch22
Measured 5v at Molex to ground. starts off about 17meg then quickly goes to infinity. This is what i would expect a cap to do. If you can measure resistance then you melted the inside of one of your caps.
lol you should see my bridges. I'm missing a couple altogether. I don't use em. I think a better term would be "Buss fuse". It looks like macci's actually did it's job and fried itself instead of the card. It's thier sole purpose in life.
You don't need to pull the card to check the resistance for me. Leaving it installed is fine. 1/4 watt resistors are dandy.
Damnit dude that's not what i wanted to hear. I'd have to pull quite a few components to find the culprit. I've had caps die before and your estimation is dead on. When they fry they melt inside and you can sometimes get a dead short. Amazing i was able to run my ddr at 451. It also explains the problem people to include me were having when thier ddr was getting less and less of an o/c. Hmm actually thinking about it there aren't that many components i have to check. Maybe just 10 or so. But i will have to pull em. More than likely it's one of the big 6.3v caps. My psu rocks so i can probably get away with removing the 6.3v caps and not replacing em. Also i'm using half the wattage i was using before. Don't ask me to explain that one. It's a completely different mod. ;)
Edited // Maybe I'm not doing this correctly, but I get two different readings.Quote:
Originally posted by MrCrusher
Catch22
Measured 5v at Molex to ground. starts off about 17meg then quickly goes to infinity. This is what i would expect a cap to do. If you can measure resistance then you melted the inside of one of your caps.
Just re read your request again and I didn't measure what you wanted .... I'll do it again. Hopefully correctly this time! :)
Ok, here's what I did... turned the system off ... unplugged the molex connector to PSU. Read from the molex 5v solder pin at the backside of the card to case ground. I got 31+meg ohms and a slow steady drift lower towards zero. If that's not the right way let me know. I do it again! I just hate being a n00b. :p:Quote:
Originally posted by Hobocrow
Edited // Maybe I'm not doing this correctly, but I get two different readings.
Just re read your request again and I didn't measure what you wanted .... I'll do it again. Hopefully correctly this time! :)
Well it shouldn't drift to zero ohms. That would be a dead short. Let it drift until it stops somewhere and gimme that reading. I'd think anywhere from 200 ohms to 20k ohms would be about right.
The 200K Vario is in from pin 5 to pin 5 for Vddq. Voltage went from 2.65v to 2.63v and OC stayed at 405Mhz. I lower to 2.62v and OC drop to 395Mhz. At 2.60 the OC dropped to 388Mhz. This is going the wrong way.
BTW-I have been using ATITool to test with since I belived it is more objective then visual checking 3dmark2k3/nature. I leave ATITool running and as soon as I go too high or low with the volts the tool starts showing errors. You can probably clock about 10Mhz more before your eye starts to see the artifacts.
I then tried pin 5 to 200k Vario to ground for Vddq. Voltage went from 2.65v to 2.67v and OC went to 410Mhz. I raised to 2.68v and OC dropped to 399Mhz and at 2.69v OC was at 388Mhz. Again wrong way.
So the only success I had was raising Vddq to 2.67v for an extra 5Mhz.
I have not given up so I then tried raising Vref from 1.40v to 1.41v and OC again went down from 405Mhz to 399Mhz. At 1.42v OC went down to 395Mhz.
The only thing I did not try was lowering Vref because I am not sure were to run the other side of the 200K vario. For Vref lowering it's Pin 5 to 200K vario to were??
Thanks-Crusher
Ok, the system has been off for about a hour.. I unplugged the molex from the PSU and measured from the backside 5v solder pin. I set the DMM from auto range to 00.00m ohms and grounded to the case. The reading started @ 00.05meg and slowly, but, steadily climbed to 00.25meg, then really slowed and climbed to 00.28meg and stopped there. Hope that helps.. :)Quote:
Originally posted by CATCH22ATPLAY
Well it shouldn't drift to zero ohms. That would be a dead short. Let it drift until it stops somewhere and gimme that reading. I'd think anywhere from 200 ohms to 20k ohms would be about right.
OK i guess i really do need to find that fried cap then.
OK you were supposed to move vddq and Vref both. Looking at your results it looks like your vddq and Vref like this math
Vref=(vddq/2)+0.065
so if you set vddq at 2.60 set Vref at 1.365
If you want raise em both as well. They are extremely picky
OMG i've been running that tool for the last 20 minutes and it's still trying to find max ddr o/c with no artifacts. WTF how long does this test take. It's down to 388.13 now. Will it ever end or is it the end?
So using that utility with my voltages at default i get 459/388. I'm set up for game mode and using 6xAA and everything max like AF etc etc. So at least that utility is good for setting an o/c for playing games. Although i hardly ever o/c for games i may try it. Back to Nature 3dmark2k1 for testing benching o/c though. :)
C22@Play, I just turned the system off again and with the DMM set the same and same measure points. This time I got the reading starting @ 01.14meg ohms and then a steady rise to 01.73meg ohms! I have no idea wtf that means...hope it helps!
Catch
Were do I get the Vref voltage. I can't solder to the pad in front of the IC??
Thanks
MrCrusher, I just re-read the math post from C22 and then your best setting for oc'ing your card. Looks like your best was: 2.67v(vddq) / 2 =1.335v (vref) +.065 for a 1.40vVref!
Wasn't that where your Vref was/ is? Maybe something else needs tweaking next?
My stock voltage is Vddq=2.66v & Vref=1.41v so 2.66v(vddq) / 2 =1.33v (vref) +.08 =1.40v Vref. 0.08v is also what i am measuring at pin 5 on the Vref IC. I also understand now (da!) that ishould be changing these voltages simultaneously.
I decided to use 500k vario's to get a finer adjustment.
I tried going down in volts. Set Vddq=2.64v and Vref at 1.40v and OC went from 405Mhz to 398Mhz. Then set Vddq=2.62v and Vref at 1.39v and OC went from 405Mhz to 395Mhz. This is the Wrong way.
I tried going up in volts. Set Vddq=2.67v and Vref at 1.41v and OC went from 405Mhz to 410Mhz. Set Vddq=2.68v and Vref at 1.42v and OC went from 405Mhz to 420Mhz. Ok, now we are getting somewere. Set Vddq=2.71v and Vref at 1.44v and looked up at screen just in time to see the opening scene from the Matrix. You know the one with all the 1's and 0's and x's going up and down in columns.
I shut the computer down rapidly and put evrything back to stock volts. Reboot and got the bios screen with no problems and windows started booting. As soon as the log in box came up the screen did the matrix thing again. This time when I reboot the Bios screen was repeating itself across the screen with a shadow effect. One of the words I could read was checksum error! I have the bios backed up on disk so I flash the bios hoping this would fix it and rebooted the computer. Get the bios screen with no problems and went into bios to change the CPU back to stock also. Reboot computer and no problems on start up screen. But as soon as windows got to Log-in the matrix was back.
Am I screwed? When I fried the memory on my 9700pro over-volting (I know, should have learned the first time!) I never saw the bios screen again it was just completly scrample from the second it was turned on. I measured all the voltages and all are correct at stock volts. I can go into the motherboard bios screen and sit there without any problems but as soon as windows start all hell breaks loose.
This morning I put my 9600xt in to see if OS was screwed and had no problems running some slowass benches with the 9600.
What next?? or is this the excuse for R420 and I limp along on the 9600 for next 4 weeks???
MrCrusher it looks like you're gonna be limping along for a month or 2. But the card may still be alive. Check Vref and see what it's at. It may be freaking out now. If it's not at 1.405 or so, you know like 1.265 or 1.650 you are screwed. My card finally went and died on me too. I feel for you dude. It's either that bad cap or one of the hexfets went. At least that's my best guess. I'd have to use a R9700pro as a backup. So i went and bought a Hercules R9800XT and just now stuck it in. Completely stock and gpu v core and vdd both are well below average i can run at 465/400. Case temp about 28C atm. hmm lemme check volts real quick live.
Vgpu=1.788
Vref=1.430
VDDQ=2.667
VDD=2.690
So gpu Vcore is a bit under stock of 1.80 and vdd is below stock of 2.80. But vddq is a bit high stock is 2.65 and Vref is high, stock is 1.403 or so. I like it btw. In fact i'll leave vddq alone and raise Vref a bit more. Then raise vdd as needed. I'll wait on Vcore until i got the time to solder and replace the stock cooling etc. Seems like a decent card so far. I'd say slightly better than my previous one.
Oh the card that isn't working well now i'd had the vddq/Vref at up to 3.20/1.65 and didn't have problems running it that high. Except o/c wasn't very good. in fact i've accidentally had Vref at up to 2.75 before without killing it. Screen did the maxtrix of course. But it was still readable and all was well when i went back to normal volts. Crap speakers going wack on me. Gotta look into it. L8z
MrCrusher, sorry about the card ... I hope you get it going again!:(
C22@Play, geezzz, same for you, but, your new one sounds really promising.
My 500k Vrs should be in next week, then it'll be my turn. In the mean time, I lost my HDD ... @ startup all it does is thump, thump and whin! I got stuff on there that's not backed up, of course!
I lost so much :banana::banana::banana::banana: when my hdd's fried :( hehe
Maybe i should have said try theese mods when your card's ddr isn't O/Cing like it used to. As in my o/c dropped from 438+ to 405 max. Then i modded it and ran my benches at 451 or so.
Maybe heat is what really messes theese cards ddr up. I tried vdd mod and i was able to o/c from 400 to 410 but score was worse. Hardly any artifacting though. Tried raising Vref and that raised score. What's wierd is vdd over 2.725 is where scores gets worse. I noticed the hexfets for vdd, vddq, and Vref are all different and put out more amps. No wonder the ddr chips seem so much hotter too. I can fix it so they put out less amps. But i'm leaving this card with stock cooling for a while in case it fries the ddr. I may do the V gpu mod though. Kinda curious what this card can do before i do my 6 or so V core mods.
I got some extra hexfets. So i'll take the old bad one off and try to solder a good one on. I just gotta find one with same pinouts and since it's Vref one with more amps. Personally i think that's the only one that needed more amps. Might as well try it.
Now for tonight it's back to Dungeon Siege Legends of Aranna :)
:banana::banana::banana::banana:.... :eek: I was raised on a farm... that's enough sex from a life time! :p: I killed this HDD benching it @ 86+ Mhz... jezz IBM just can't make a good product! ;)Quote:
Originally posted by CATCH22ATPLAY
I lost so much :banana::banana::banana::banana: when my hdd's fried :( hehe
I'll check back in when I get my Vrs!
OK i was taking a break from gaming and decided to figure out what exactly is fried. Looks like pin 2 on the FET which is G-1 and which goes to DH-1 on the isl6522 chip isn't reading correctly when i ohm it out. Basically it's a dead short between pin 7 and 8 and 2. 7 and 8 is where 5v is connected btw. So pin 2 is getting 5v as well. Sounds like a complete melt down inside the FET. I hope it isn't the isl6522 controller chip. I doudt it is. I'm reading the pdf files and some notes on what extra parts i got lying around or on other dead cards. Well the closest match i can find is the FET from an R9700pro's gpu Vcore. It's a bit more amps. But that's what i want anyways. The pin out does match too. Here's the difference between the fried FET and the one i want to install. The first one is the dead one btw.
STmicroelectronics STS8DNF3LL
RDSon=17meg ohm I=8.0 amps@25C and 5.0 amps@100C
Fairchild? FDS6898A
RDSon=14meg ohm I=9.4 amps@25C and 9.4 amps@100C? (should be lower)
So they look to be a close enough match and about what i wanted to increase current. I hope it works. :)
hmm.. i got a strange problem with my sapphire 9800 pro 256 with the 9800 xt layout, core and memchips. after flashing the xt bios it will run 412/365 but at the very end of '03 it starts artifacting. so that's when i decided to get a reading on my vgpu voltages. well what can i say in idle/desktop mode it's pretty normal, reading 1.778v which is close enough to the 1.78v that's default with an xt-core at 412/365. but once i fire up any 3d-app like 3dmark 01 or 03 or aquamark or any kinda game, voltages drop. under heavy load they went as low as 1.711v which is bad and probably wouldn't help getting rid of the artifacts.
seeing that i decided to go at the xt-vgpu mod described in here. hooked everything to the card, and adjusted the pots to 20k so i'd get 1.800v. on startup it would show me 1.78v (woot, 0.02v more than before) so into windows we go. benching a little and the volts drop again, but not quite as much as before.. only to ~1.75x volts. hm... might raise it a little more. ok up to 1.850v... on booting it says 1.780v again - wtf? right after i enter my pwd on the login screen the voltage goes up to 1.83x. wooot.. but well, it drops back down to 1.780v soon enough running 3dapps. well the drop while loaded is still there, but it seems to stabilize a bit.. ok. death or glory it is. 1.90v captain. said and done... what do you know? same thing as before.. 1.78v booting, 1.88x v logging in. 1.78 in 3dmark, heavily loaded 1.76x v. yea crapper.
i didn't do any other mods on my card yet and i hardly think i will do so since this card seems to be a firstgrade pos.
catch22atplay, you don't happen to have any magic up your sleeve that might cure this biatch? i've posted some pics of my card and have a whole thread about these type of downlabeled but fully functional xt-cards (except mine, maybe) here.
my card together with my ghetto vmod is featured on page 7 here.
any insights on what to do are highly appreciated. but if you think it's hopeless, i'll rma it on monday and wait for the r420. ;)
C22@Play.... I didn't understand alot of that.... see what I mean! ;)Quote:
Originally posted by CATCH22ATPLAY
OK i was taking a break from gaming and decided to figure out what exactly is fried. Looks like pin 2 on the FET which is G-1 and which goes to DH-1 on the isl6522 chip isn't reading correctly when i ohm it out. Basically it's a dead short between pin 7 and 8 and 2. 7 and 8 is where 5v is connected btw. So pin 2 is getting 5v as well. Sounds like a complete melt down inside the FET. I hope it isn't the isl6522 controller chip. I doudt it is. I'm reading the pdf files and some notes on what extra parts i got lying around or on other dead cards. Well the closest match i can find is the FET from an R9700pro's gpu Vcore. It's a bit more amps. But that's what i want anyways. The pin out does match too. Here's the difference between the fried FET and the one i want to install. The first one is the dead one btw.
STmicroelectronics STS8DNF3LL
RDSon=17meg ohm I=8.0 amps@25C and 5.0 amps@100C
Fairchild? FDS6898A
RDSon=14meg ohm I=9.4 amps@25C and 9.4 amps@100C? (should be lower)
So they look to be a close enough match and about what i wanted to increase current. I hope it works. :)
Anyways, I've got a dead 9800p if you want I'll send it to ya and you can rob it for everything you need ... if you'll send the remains back to me afterwards! :)
sky very interesting there. You do have the vpu recover thingie turned OFF no? Tell you what, i got an Saphire 9800Pro with 256mb. It's not a remarked whatever XT one. It's died playing a game. Maybe i should RMA it and maybe i can get one of those remarked ones. I can't tell you exactly how to fix yours. If i had it i could make it run better i'm sure. Post some links to bigger pics of both sides of the card and maybe i can see something.
Hobocrow was that an r9800pro or xt? I don't think i need any parts since the one i have matches so close. Only thing is i really suck at soldering chips. I got a couple friends that might be able to do it for me. I'm gonna try it first. I still may take you up on that offer though.
Right now i got more important issues. This Herc 9800xt is getting about 10fps less than my ati in this game. I had everthing maxed before AA and AF etc. Only thing is i had some registry tweaks on my other card for benching. Hmm wonder if that's it. I noticed the textures look cleaner on this card. hehe i'll have to figure out what registry tweaks i was using before when i bench again.
Now tough decision i have to make right now. Do i keep my pack mule that carries approx what 4 characters can carry or trade him for this nature magic user with good stats? Hmm maybe a beer or a doobie will help me decide. ;) Surf's up dudes dudes :)
Hobocrow what's not to understand? Basically one of the 8 legged chips is fried internally. I don't have an exact replacement for it. All you have to do is find another one with same pin outs and close to same RDS(on) and close to same amount of amps. RDS(on) is very important in that instead of a current sense resistor, RDS(on) is the current sense resistance. No resistor needed. However a current sense resistor can be added to it. They don't use current sense resistors for VDD, VDDQ or Vref. They do use it for gpu Vcore though. Notice the stock one had an RDS(on) of 17meg ohms and 8.0 amps and the one i'm gonna use has an RDS(on) of 14meg ohms and 9.4. Usually but not always RDS(on) the lower it is the higher the amps. For instance an International Rectifier 7809 is 9meg ohms and 13.3 amps and thier 7313 is 29meg ohms and 6.5 amps. The 7822 is 6.5meg ohms and 18.0 amps.
Oh you also have to figure out the switching frequency of the chip. The one i fried and the new one are way different. I'll have to change the switching freq coming from the isl6522 chip based on theese equations.
Fs 200kHz + [5X(10to the 6th power)]/RT
Hope that doesn't confuse you. hehe
MrCrusher to see if one of your FET's are fried try checking resistance of pin 9 to ground of the Vref and vddq controller chip. It should be close to 45k ohms. Also check resistance for the 5v line next to the molex plug. If resistance of pin 9 matches the 5v resistance then the FET is fried. I'm not positive about any of this. But comparing my nex xt to my old one does confirm this. If you find that one of the FET's are fried wait until i attempt to fix mine. It's very possible we can both bring the cards back to life. Although mine isn't really dead. It just artifacts bigtime at any speed cause Vref volts go high and low like mad.
Hey also plug it back in and check all your volts and see what they're at. I mean check em when you load a 3d ap and or bench and not just windows. If you see any freakage like mine i think we got a fix. If you don't get a freakage then you may have fried something else. Maybe send the card this way and i can look at it and maybe get lucky. :)
Just a question, but on the vGPU mod, has anyone here seen 1.9v kill a 9800xt? The reason I ask, is that a friend of mine asked me to perform this mod for him so I did. The soldering was flawless (if I do say so myself, as I do this type of work for a living and have done for some time), the connections were right, and everything was just as in the threads I've read. However, the card wouldn't post when the computer was turned on. No post error codes, not anything. We backed down on the voltage (which in retrospect, we probably should have made sure it was all ok before going to 1.9 with it but everyone seems to have had quite a bit of success at that voltage) and still nothing. Kind of a bummer, but it's the kind of risks you take when doing this sort of thing I suppose. The good news is, we removed the mods and cleaned the card up nicely, and I think we will be able to RMA it with no problem...
But, back to the question, has this happened to anyone else?
Yes and no. I did what i thought was the best solding job ever on my abit ic7 mobo. Well it didn't boot at all. It was dead for a couple months but finally came back to life. So it could have been your soldering job. I've killed a couple vid cards because i didn't have it seated in the agp slot all the way. I've never killed one doing a gpu core or vdd or Vref mod. O/Cing sometimes kills mine. Hell playing a game stock can kill em.
Here's the secret to volt modding. Draw a diagram of the area you'll be working at. Write down resistance from resistors in that area and for caps and diodes as well. After you volt mod check resistances again. This has saved me quite a few times. It takes time and is worth it.
Sorry I have not been responding. I got the new toy yesterday at xtremegear.com since the 9600XT just was not cutting it. I got use to playing COD ultra smooth with 4aa & 16af with the clocked 9800XT. Xtreme has a OEM BBA 9800PRO for $207 plus shipping. I spent the day doing the mods (Vgpu, Vdd, Vref). Now that I have some experience, the mods and soldering came out really clean.
I started with the mem and played with Vref and lowering a little from 1.41v stock to 1.39v got me from stock 333Mhz stock to 360Mhz. Next I upped Vdd from 2.93v stock to 3.31v and this got me to 375Mhz. I am not sure how much more this can take so I am stopping here for now so I do not fry 2 cards in one week.
What really surprised me with this r350 core was its overclocking potential. I started by raising the Vgpu from 1.73v stock to 1.85v and this got me from 378Mhz to 440Mhz. Next came 1.95v to the GPU and out came 480Mhz just 20Mhz shy of my crippled 9800xt. I then hit it with 2.0v and got up to 510Mhz. The tempature probe is at 48c on a load. I am stopping here since my 3dmark2003 score is now past the best I was able to do with the 9800XT.
Tomorrow I am going to replace the MCW50 with a MCW50-T that the 9800XT overwhelmed. The 9800Pro is putting out a lot less heat so I will see if the MCW50T can get anymore out of this core.
Catch22, I measure all the Volts on the 9800XT and they are all at the stock levels. I sit and watched vref for a few minutes and no change stayed at 1.40v. I can get into windows but as soon as I open up any programs I get the matrix screne. I will measrue those resistence's and get back to you with what I find. But I think Dr. Catch might need to preform surgery if this patient is going to live.
yea. do you have any special area in mind that you want covered by a pic? you name it, i take a pic of it.Quote:
Originally posted by CATCH22ATPLAY
sky very interesting there. You do have the vpu recover thingie turned OFF no? Tell you what, i got an Saphire 9800Pro with 256mb. It's not a remarked whatever XT one. It's died playing a game. Maybe i should RMA it and maybe i can get one of those remarked ones. I can't tell you exactly how to fix yours. If i had it i could make it run better i'm sure. Post some links to bigger pics of both sides of the card and maybe i can see something.
i'm gonna check on vref and vddq and maybe mem aswell, as i think it's pretty strange that the core voltage rises just once to where it should be and then almost instantaneously drops back down to sort of default.
where do i check vref? right where mr crusher said, on that left pad in front of the ic right below the powerplug? right where he circled it in this pic? left pad = vref, right pad = vddq
http://www.xtremesystems.org/forums/...&postid=370048
Hmmm.... ok.. Thanks for the input. I'll keep that in mind in the future.
update
so i measured the vref on my biatch.. guess what? it's freaking out like the vgpu. check out my these vref-values:
1.504v - powering up the rig
1.485v - entering windows, after that it basically stays there idling
firing up 3dmark '03:
1.488v - idle between tests / loading screens
1.561v - test #2 battle of proxycon, somewhere around halfway through that test
1.460v - test #4 mother nature, starting this test
1.476v - test #4 mother nature, around the temple thingy
1.46xv - test #4 mother nature, after that temple
now 3dmark '01:
1.525v - test #7 nature, starting
1.532v - test #7 nature, closest to the house
1.493v - test #7 nature, around the bridge
1.530v - test #7 nature, when it pans downward toward the water
1.51xv - test #7 nature, when more of that water is visible & for the last two sideways pans
all the other tests are around 1.46 - 1.51v and i did NOT do a vref mod yet. didn't bother checking vdimm and vddq, this POS is going back to where it came from :(
and i tried running it with my vgpu mod in place, but that didn't do jack on the vref-voltages (like with the actual vgpu voltages that only on entering windows change just once to the value i set them to and back down immediately after that again).
funny thing.. didn't it say in here the vref should be around 1.40v? not even close.. i mean 1.504v booting and max 1.561v loaded does seem to me to be way out of line to be normal. and this is all, of course, without vref-mod. so this biatch is going back to where it came from. pfff seems to me the yield of 9800xt cores isn't that high as there's plenty of these relabeled cards floating around.
sky this is the same problem that started with the R7200 card. It's the main reason i came up with the Vref mod years ago. It was to FIX a brand new card to keep it from frying. Your card will die within 2 months imo. A lot of people within 2 weeks. I can't believe they released a card with bad volts again. Just to make sure you're reading it correctly check the voltage at C98 or C99. The small yellow caps in pic above.
Mr Crusher if your volts ain't wacked you'rs i bet is history and i can't fix it. Sorry dude. I should have titled my v mods for use with cards that suffer from a lower o/c than the used to get and before they rma it.
Ohhh, Ok! Thanks for clearing that up.... I got it, your from another planet! :DQuote:
Originally posted by CATCH22ATPLAY
Hobocrow what's not to understand? Basically one of the 8 legged chips is fried internally. I don't have an exact replacement for it. All you have to do is find another one with same pin outs and close to same RDS(on) and close to same amount of amps. RDS(on) is very important in that instead of a current sense resistor, RDS(on) is the current sense resistance. No resistor needed. However a current sense resistor can be added to it. They don't use current sense resistors for VDD, VDDQ or Vref. They do use it for gpu Vcore though. Notice the stock one had an RDS(on) of 17meg ohms and 8.0 amps and the one i'm gonna use has an RDS(on) of 14meg ohms and 9.4. Usually but not always RDS(on) the lower it is the higher the amps. For instance an International Rectifier 7809 is 9meg ohms and 13.3 amps and thier 7313 is 29meg ohms and 6.5 amps. The 7822 is 6.5meg ohms and 18.0 amps.
Oh you also have to figure out the switching frequency of the chip. The one i fried and the new one are way different. I'll have to change the switching freq coming from the isl6522 chip based on theese equations.
Fs 200kHz + [5X(10to the 6th power)]/RT
Hope that doesn't confuse you. hehe
It's a 9800p, but , you're welcome to it! :)
hehe i read too much. I found the FET i want. It's on a good r9700pro. I don't really want to take it from that one. I'm gonna try one from a ti4600. It isn't even close to what i need and puts out a bit less amps. The opposite of what i needed. I'll try to solder it right now.
OK update so far. I removed the bad FET and now pin 9 on isl6522 reads 45k ohm like it should :) Now the hard part. Soldering the replavement on. Damn it's a tight fit in there. I need to start cutting down on coffee. My hands are shaking way too much.
hey catch22, thanks for the heads-up. i didn't measure the other voltages yet. i took the card out, cleaned it of my mods tooks ome pics and prepared it to return it to the shop - or sapphire if they prefer that. doh what a bummer.
"*SUCCESS*"
OMG after resoldering and resoldering it's finally done. Had to ohm out each pin to make sure it was ok. What a major pain in the ass. I fired the puppy up and sweetness it runs fine now. Back up to 432 so far. Before i fried the card it had artifacts at default speed when i play a game are gone as well. Hell i think this FET works better than the old one. I may swap the vddq FET now as well. I think that ones going soon too. Am i stoked. Hell i can return that Herc now and get 500 bones back. I am content.
:toast: Dr. C22@Play... You da man! :toast: :banana:Quote:
Originally posted by CATCH22ATPLAY
"*SUCCESS*"
OMG after resoldering and resoldering it's finally done. Had to ohm out each pin to make sure it was ok. What a major pain in the ass. I fired the puppy up and sweetness it runs fine now. Back up to 432 so far. Before i fried the card it had artifacts at default speed when i play a game are gone as well. Hell i think this FET works better than the old one. I may swap the vddq FET now as well. I think that ones going soon too. Am i stoked. Hell i can return that Herc now and get 500 bones back. I am content.
Now it's time to attempt to kill it again. ;) Swapping vddq FET as we speak.
Catch...
my vref is varying from 1.44 at bootup to 1.49 at full load. is this ok?
also, my vgpu... i set to 2.1V, and in drago it drops to around 2.01V, and then in nature drops to around 1.95V..
anyway to help this stay stable?? 9800PRO here..
It's normal for Vref on any card. The core is normal for most ATI cards except the 9800xt volts go up usually.
Vref before i replced the FET before it was fried stayed fairly steady at 4.04 The new FET goes from 4.04 to 4.12 max
oops noticed your Vref does go up a bit higher than what i'd like to see. But it's not terrible. It just isn't very good. Barely acceptable imo. It's ok if it's tracking 1/2 of vddq though.
So, now we have VGPU, VMem, VDDQ, VRef and 'r1596' mod. Am I right?
But I didn't get realy how to make Vref and VDDQ mod. Help please. And what resistance it's better to dial on the VR for 'r1596' mod?
I haven't done the Vref or the Vddq mods yet, so I don't feel qualified to answer that one.... on the R1596, if you ohm it, it should read approx. 41.x kohms, mine was 41.2k and I penciled it to read 19.7k. that worked for testing it on my card. I believe most ppl used a 50k VR setting it to 20k. :)Quote:
Originally posted by 2PiZzA
So, now we have VGPU, VMem, VDDQ, VRef and 'r1596' mod. Am I right?
But I didn't get realy how to make Vref and VDDQ mod. Help please. And what resistance it's better to dial on the VR for 'r1596' mod?
Hobocrow Got it.
I need answers to other questions. pleeeeaaaase.
I say skip the Vref mod unless your cards ddr o/c has dropped 20mhz or so and you're about to rma it. I lower and or raise Vref and vddq. To raise each one use a 100k ohm vario from isl6522's pin 5 to ground just like for vdd mod. Vddq is located closest to agp slot.
Still stuck at 432 or so. I tried 441 and it was insanely messed up. I didn't try anything between those settings. VDD is at 2.717 btw. I can't run it above 2.725 because of another type of v mod i have going to vdd. I Used to have to run at 2.925 or so. I don't think vdd is holding me back. In fact i'm positive it isn't. I'm putting out plenty of watts there. Since this FET i'm using now for Vref puts out slightly less watts than the original i'm thinking that's what's holding me back. However i can boost vddq watts a bit and will try that. I didn't swap the vddq FET yet. I just desoldered the one off the ti4600. I just had to get a couple hours of gaming in.
Thanx. I think I got it. I have VGPU and VMem modz yet. 525/860 - that's what I got.
Phasechange is coming.
C22, what mods exactly did you make for 582/451? And what volages exactly did you dial?
And are there other VModz for this card?
Yes there are other mods. I think i'm the only one putting them out though. I think my compare below lists what voltage i used. But here'a quick list of what i did for the bench and voltages. Just remember each card is different and your PSU has an effect as well.
PSU volts
3.3@3.625
5v@5.6
12v@13.2
Vagp@1.85
gpu Vcore@2.275
VDD@2.925
vddq@2.55
Vref@1.340
SSTL2@2.2 (default=1.8v) don't ask i'm still not sure if it is SSTL2
Vref 3 mods,vddq 2 mods, VDD 3 mods, gpu Vcore 5 or 6 mods.
Quite a few of the mods were not helpful imo. But some did give a positive effect. Please don't ask me to explain what mods i did and if i take a pic you won't understand what i did and will kill the card as soon as it powers up. I've fried probably about 100 vid cards. 97 from people trying to interpret what i did. I can and always do list v mods that i'm fairly sure will be helpful. Although some of my mods that helped me a lot didn't do jack for some people. Here is my suggestion and what needs to be modded.
gpu Vcore mod to raise v core
R1596 to be able to run v core above 1.95 and to stop card from resetting
VDD mod to increase volts
vddq you may want to try
Vref you should always do if you change vddq
The last 2 not many people do at all. In fact the dudes running at 450+ only do the VDD mod. Thier cards almost do it stock. Maybe that cold weather helps them. My card 405 was when i really started arifacting. Same with my herc i just got. I wish i had a card that ran 450+ as easy as some peoples.
Catch22,
Need your help on Resurrecting my 9800XT card. She is wounded but there is still life in her.
I checked pin 9 to ground of the Vref and vddq controller chip and both were at 45K ohms. I slapped an old 1800XP MB on the desk plugged in the 9800xt to measure voltages. I plugged in a monitor so I could watch the matrix while I was measuring. Much to my surprise the bios screen looked normal, no matrix. I then plugged in a hard drive and loaded up XPpro. 20 minutes later the screen looked good and I had a perfectly healthy looking desktop. Loaded up cat 3.9 and 3dmark2003. Made it all the way through Wings of Fury then about 10 sec into Proxycon the matrax came back.
At this point I measured voltages Vgpu=1.77v, Vdd=2.68, Vddq=2.65 and Vref=1.11v!.
Rebooted computer and got matrix. Unplugged everything and waited 10 minutes. Started back up and got clean desktop. Run 3dmark2001 and got about 25 sec into with same high frame rates as before then matrix. measured Vref and back down to 1.1v
Seems like there may be hope if I can find the failing component.
The failing component it probably the same as mine. The 8 legged FET or mosfet if you prefer is probably wacked. Swap that part for another. Practice removing ones off old card and soldering them back on. Then do it to the XT. It's possible it's something else. But i'm betting it's that FET. I would suggest ohming out all 8 legs of that chip and i can compare it to mine and then i can tell you if it's fried or not. Also have you penciled in anything? If so clean it with alchohol and Q-tip. Also have you soldered to anything? It sounds like it's going into reset mode btw. Lucky you even get a screen. But hell mine was like that too. Do the ohming thing 4 me. Oh the FET for Vref is located on the opposite side of the controller chip. Fairly close to molex connectors 12v wire.
No Random Penciling
The chip is marked #7322
Starting with pin closest to the 12v molex pin or right above the 22 in 7322 (assuming this is pin 1)
2.2k 2.2k 1.75k 1.75k
..4 ... 3.... 2..... 1 <-----pins
..5 ... 6.... 7..... 8 <-----pins
1.75k 45k 0.0 30meg+
ok it's dead i'd say. A lot of the readings you got are way off. Looks like you blew the other diode inside of it maybe. You blew one half of it and i blew the opposite half. My volts would go high and yours drop low. If you got any spare vid cards swap that part. Check with me first with part number and i'll tell you if it'll work. The only mosfets that i don't have data on are the ones used on some R9700's with markings 4304. It doesn't list manufacturer so i don't have a clue if they'll work. I'd say no because they were used for ddr that used less watts. At least i'll assume higher mhz=more watts and vice versa. Don't assume you can replace that chip with any other chip. Some are made completely different and pin outs aren't even close. For example an IR7809 would never work. Send me the card if it sounds too difficult for you.
I was playing my game for a couple hours tonight. I alt/tab so i could check e-mail etc. Had 20 IE windows open and mirc running. I had no o/c on vid card. I went and watched a movie for a couple hours and came back to the matrix on my screen. Rebbot comp and all is well. I knew that mosfet i used didn't put out enough watts. Guess i'll swap that part again with something that puts out a bit more.
Its funny you should mention the R9700Pro. I pulled this long dead card out and started measuring voltages. This card is completely matrixed. On the 9700 my Vref was right at 1.40v but my Vdd was 1.51v!! The mosfet is right next to the Vdd IC. It is labeled "4803" & Pin 5 output on the mosfet is putting out 1.5v. I found 2.8v off a cap near the Vgpu IC and run it to pin 5 and bam I had a perfect desktop for the first time in 6 months!!
So now I have two potential cards that could become the living dead. Were the hell can I find replacement chips? Frys?? I can dig through the pile to find some old voodoo cards and possibly a geforce 2. I 'll start looking.
Those are just switches. You can't replace em for a better one. In a sense they do sort of regulate volts. But not quite in the sense you may think. It's used you might say for setting a low or high signal i believe. I don't quite understand it that well. I'd have to read up more on it. But as i said you can't really do anything to help them. Er well each should have a 10k ohm resistor tied to one of the legs to ground. At least according to the data sheet i'm reading. So making sure all have a very close tolerance to that may help. But i wouldn't bother with it and you know me. I mod everything.Quote:
Originally posted by Ragnarok
right, can i please have the other 5 mods? :D
btw, there's seems to be a few same vreg near each pair of mem chips, care to enlighten us what they are for wise one? :D
There are days when i wish i took more electronic courses in school instead of just the basics. But i'm a DIY type of dude anyways. Oh i am an electrician though. But that doesn't mean i know crap about this type of stuff. Just about everything i do is self taught. A bit of common sense, math, ability to read and learn and the balls to try something new. Hmm sounds like may day job as well. :)
CATCH22ATPLAY What did you mean by
"Vref 3 mods,vddq 2 mods, VDD 3 mods, gpu Vcore 5 or 6 mods."?
You mean that there are 3 different ways to do Vtef mod, for example?
And can anybody tell what are those two dip-switches for?
These should be the data sheets for the mosfets on the 9700pro & the 9800XT.
They can be found at http://www.datasheetcatalog.com/
9800 XT Datasheet Mosfet STS8DNF3LL <----Updated with correct chip number
9700Pro Datasheet
You sure you aren't reading the lot # of the 9800xt's mosfet? Isn't there an ST trademark upper left? The bottom number should be S8DNF3LL. The link you gave for the XT i think is bogus unless your card is really wierd. No you can't get theese parts at fryes, i tried. They got crap there.
2PiZzA the debate is still out what those dip switches do. A lot of us assumed it was for NTSC/PAL. I never bothered moving the dip switches. The 3 for Vref for instance is hard to explain and same as most my mods. Let's just say i experiment a lot and remove/replace parts. That doesn't mean the v mods worked. I just listed what v mods were done to the card during that bench. So at least it tells me i didn't kill it. I already listed which mods i thought were needed a few posts back. Ignore some of my crazy experiments. Although a few of the one i haven't posted about i may post about if i can verify they actually help. I haven't done that for most of them. My VDD mod i like in that i don't need 2.90+ to run it. It runs fine at 2.725. That doesn't mean doing the mod my way and running it at 2.725 is better than doing a normal vdd mod and running it at 2.925. In fact i ran the bench with ddr at 451 with only the normal vdd mod and 2.925. I was in error when i listed 2 mods for vdd i had when i ran the bench. I haven't been able to get anywhere near 451 let alone 440 lately. So my extra 1 vdd mod may be screwing it up. It is a possibility and i will of course remove some mods and try again. Like i always say. I wish i took better notes. The only thing i know for sure is i have to swap this Vref mosfet again.
I think i may end up ordering some of theese mosfets. Seems i need a few replacements.
/me winks at Hobocrow
I'm so close to raping my r9700pro's mosfets. I wonder if i can be real careful so i can put them back on that card when i need to. Oh hell yes. OK removing parts off my good card now. hehe
Ok, I updated the data sheet on the 9800XT to the correct model number. Both mosfets have similar Characteristics. I went by Fry's and they have nothing.
Datasheet for STS8DNF3LL
Here is a list of distributor's for STS8DNF3LL
And an place to order for $1.09
Datasheet for 4804 on the 9700Pro
And for the 9700Pro #4804 at $1.34
Updated resistence based on correct pin out of S8DNF3LL. how do these look
2.2k 2.2k 1.75k 1.75k
..8 ... 7.... 6..... 5 <-----pins
..1 ... 2.... 3..... 4 <-----pins
1.75k 45k 0.0 30meg+
CATCH22ATPLAY Thanx anyway. I'm just wondering how did you get those clocks? You got phasechange on GPU?
What mods exactly made your 582/451?
You got 451 on mem only with vdd=2.9? Do you use heatsinks on mem?
catch22atplay,
I got my hands on a free broken 9800xt. Looking at it, all of the hexfet source pins read dead short to ground (ouch). I've lifted the aluminum caps in hopes that one of those was the problem, as well as c1532. Still no joy. I was wondering if there is a bridge in the path to the gpu I can lift to further isolate this. It could well be the gpu is melted or maybe just a hexfet source to drain short. I've actually got little hope for it at this point, but if I can fix it, it's mine. Any tips? I'm trying to avoid lifting all 24 hexfet S pins unless I have to.
OK this is what it says in my compare:Quote:
Originally posted by 2PiZzA
CATCH22ATPLAY What did you mean by
"Vref 3 mods,vddq 2 mods, VDD 3 mods, gpu Vcore 5 or 6 mods."?
You mean that there are 3 different ways to do Vtef mod, for example?
And can anybody tell what are those two dip-switches for?
3.2 P4 EE Vcore 1.90v Danger Den RBX Water Block ATI R9800XT Gpu Vcore 2.275v Danger Den MAZE4GPU Water Block 0C chilled water 22C ambient temps No pelts etc. Just those 2 water blocks. Special volt mods to R9800XT Total volt mods to vid card approx 9 MODDED THE CATCH22 WAY
So semi special cooling in that it's 0c water. I installed ramsinks after that bench btw. They were bare chips for that run but ambient temps were 22C or possibly a bit less. I semi sure i had a 12" room fan pointed into the case for that run though. I don't use case fans and i was trying to cool off the cpu mosfets mainly. Vgpu was 2.275 with maybe 4 volt mods. vdd@2.925-2.975 with 1 or 2 and vddq with 1 or 2 and Vref with 1 or 2. I just can't remember. All i remember is adjusting vddq and Vref to get rid of artifacts and finding the happy place for vdd as well. The only v mods that were imporatnt for that were normal gpu mod and R1596 and 2 i still haven't published. vdd, vddq and Vref were normal v mods except i lowered some voltages. The 2 gpu mods i haven't published that helped is an card independant mod and has to be done differently for each card imo. Kinda like Vref and vddq seem so different on each card. The mem O/C i still can't get close to. Maybe it just felt like running at 451 that night. I tried it the next night and couldn't repeat it. Was the cool ambient temps the determining factor? Hmm could be.
No bridge to cpu that i know of. That one may end up being a card that might not ever come back to life. But i'd probably try to fix it. You sure you're looking at it correctly? The core should read 10 ohms or so to ground.Quote:
Originally posted by mikead_99
catch22atplay,
I got my hands on a free broken 9800xt. Looking at it, all of the hexfet source pins read dead short to ground (ouch). I've lifted the aluminum caps in hopes that one of those was the problem, as well as c1532. Still no joy. I was wondering if there is a bridge in the path to the gpu I can lift to further isolate this. It could well be the gpu is melted or maybe just a hexfet source to drain short. I've actually got little hope for it at this point, but if I can fix it, it's mine. Any tips? I'm trying to avoid lifting all 24 hexfet S pins unless I have to.
*sigh* Guess I'll be lifting some legs just to be sure this weekend. And yeah, I'm sure I'm checking it right, .4 ohms between ground and a lot of places that shouldn't be ground. (cap + legs, fet S pins, vgpu tp on back, etc.) Oh well, I planned on keeping my 9800Pro until the next gen cards come out as it was, woulda been nice to have a free upgrade though.Quote:
Originally posted by CATCH22ATPLAY
No bridge to cpu that i know of. That one may end up being a card that might not ever come back to life. But i'd probably try to fix it. You sure you're looking at it correctly? The core should read 10 ohms or so to ground.
Thanks for the feedback. :toast:
CATCH22ATPLAY Would you publish those secret modz? Can you just post some fotos of your card and some commetns?
mikead_99 yup 0.6 ohms you are screwed. Was the cards gpu V modded? Did you check the resistors and see if they look ok. There are so many little caps and things that could have fried. Thinking about it though, since channel 1 carries most of the load i'd lift the 4 mosfets closest to center of the card. Trust me it would be easier if just one of those mosfets were dead than trying to remove and find a bad little cap somewhere. Such a daunting task. Good luck dude.
2PiZzA er NO. As i said before if i take a pic and you see it some things will be obvious. Like a wire going directly to PSU but of a diferent voltage that what it was. If someone did that alone it would kill the card in a nanosecond. Sometimes a pic isn't worth a thousand words. My pics would confuse and fry everyones card. As i said if and when i can validate the mods i use are helpful and i feel they aren't too difficult i will show you dudes what's up. The gpu core mod is still a work in progress. Here's the basic problem.
Do a normal gpu v core mod and you can increase v core to 1.95 and o/c until it resets the card due to current limit. Some cards you can run a higher voltage.
R1596 mod to increase current limit. That helps to raise v core and raise current limit for higher o/c But card can still reset. When you o/c channel 2 will hit it's current limit and go into minimum mode or turn off (still not sure if it turns off channel 2 or just goes into minmum mode. It looks like it turns off completely) and channel 1 will take all the load. This limits the total amount of watts the hexfets can put out. Because only 4 are being used to thier full extent.
what needs to be done and i have done it is to increase current limit for channel 2 (replacement of part. I then increased current limit for channel 1 (replacement of part). Then increased current limit for channel 2 some more (replacement of part again ). I tried that and it helps slightly. I decided to do something else to increase watts for channel 2. This required the complete removal of a part and solder of a jumper wire. That worked perfect. Because i increased the limit for channel 2 it is taking most of the load now. I have to reduce it to make both channel 1 and 2 share the load evenly (replace the part i replace 2 times earlier again). I still haven't got channel 1 and 2 sharing at 50/50. However i pretty much now have no current limit or a limit that is so high that i have no protection from frying the card and both channels and i mean all 8 hexfets are staying on. After i get the sharing a bit closer i will see if i can o/c a lot higher. With 0C cooling i might be stuck at same o/c and this would have been a complete waste of time. I just like to v mod and make things run better. This is all just experimental playing around. My goal was to get both channels sharing evenly and to never have my card reset because of current limit and get all 8 hexfets to stay on when i o/c and i achieved that to a great extent. I just got to play a bit more with it.
CATCH22ATPLAY Ok. What I got is:
1. You're workin on GPU now.
2. What do you mean channel 1/2? It's power channel? They are both giving power to GPU?
3. Do you know what modz did macci to get his 657/471?
Thanks for confirming channel 1, I was wondering which was where. By the data sheets I though the 7064's might be the higher current path, lower RDS(On), but I'm not real familiar with these so that was a guess. Migrated to the software world a while ago. Looking forward to working in India soon. (j/k)
As far as this thing being vMod'ed. I think so, but not sure. This guy is a coworker of mine, I think there is a little pride factor at play. He claims he left it completely at stock, but I'm not so sure. I looked in the vgpu mod area. If he did mod it, he did a damn good job covering his tracks. I don't believe he's got the skill for that, I'd expect some solder blobs if he did anything. The solder is a bit brighter so I'm probably not giving him enough credit, plus, how the hell else to you fry a card so bad? Hmmm, and why not RMA? And lol at the caps being a pita, I believe I found and lifted all the ones that showed ground on both sides. No burnt crispies on the board either. I think I'm just going to put this to the side for a little while. Got a few other things I need to deal with atm. I'll start back with the fets you recommended when I get back to it though.
Macci does the VDD normal mod, v gpu normal mod, plus R1596 mod. That's all he does as far as i know for v mods to the card. I think he's pushing his herc r9800xt's core at 2.40 volts now though. LOL and i thought i was crazy. We assumed that DDTUNG or someone else showed opp the r1596 mod as well. It just doesn't seem possible to get over 600 core without it. Although a few of them do. Runmc uses same mods as macci. If some people haven't done it i would suggest they do it pronto. I can't speak for everyone i help with mods. But i'll assume that most just do those basic mods. They get cards that have excellent ddr o/c's even at stock volts. I'd really like to know if it's temperature related. Wish i had a card that O/Ced the ddr like mad at stock. As is 432 is difficult enough.Quote:
Originally posted by 2PiZzA
CATCH22ATPLAY Ok. What I got is:
1. You're workin on GPU now.
2. What do you mean channel 1/2? It's power channel? They are both giving power to GPU?
3. Do you know what modz did macci to get his 657/471?
The chip that controls gpu v core has 2 channels for controlling v core. They are used in a current sharing fashion. Channel 1 is located closest to the center of the card and channel 2 closer to the outside of the card farthest away from agp slot.
mikead_99 you make good guesses and seem to know what you're doing. That alone gives you the chance of bringing that card back from the dead.
HoboCrow i got the cards you sent. Already found a problem and will fix that before i even put it in my rig. Thanks for the spare parts too dude.
Got another soldering iron last night. Maybe some day i'll get the 2 that i really want. I have like 5 of them now. I don't like any of them at all. This one i just got is a Hacko and so far i'm really liking it. Also got a 0.4mm radius needle point tip for it. My this is making soldering small stuff so much easier now. :)
HoboCrow lol dude that r9800pro you sent sure is a mess. Looks very similiar to one of my cards. You got cats too?
CATCH22ATPLAY Thanx man for answering my questions.
Can you show me these channels on one of this images, please?
http://vmodz.by.ru/09.jpg
http://vmodz.by.ru/10.jpg
Just use MSPaint or something. I'm just curious.
First pic where you see 8 hexfets together. The bottom 4 is channel 1 and top 4 channel 2. Haven't actually ohmed em out to verify. But it's my guess.
CATCH22ATPLAY Got it.
Anyway looking forward for some intersting modZ from ya man.
Replaced Mosfet #STS8DNF3LL for Vref. Got stable 1.40v on Vref now. Booted to desk top no problems after transplant. Ran 3dmark2001 & 2003 all the way through with no problems. No Vmods active at this point. Turned up mem from 363 to 390Mhz. Ran 3dmark2003 and half way through got green, blue and black boxes accross screen. Rebooted and desktop came back as normal.
Maybe I got too excited to see 3dmark run all the way though with out the matrix. Any suggestions??
I dunno dude. I should have left that IR7311 in. I decided to swap it with that FDS6898A and wow at no artifacts up to 440mhz then it died. I got 800mv and it's jumping around like mad. It was looking so sweet too. I'll have to see if it's still alive once i put that IR7311 back in.
MrCrusher did you swap it for the excat same part? If so maybe the reason it failed was do no another component. Also did you check what the volts were when it went wack on you? I'll be looking into it. But for now it's game time. Good thing i haven't returned that herc R9800xt and will probably be keeping it.
CATCH22ATPLAY btw, how did macci get his 471 on mem just with vdd?
Catch22,
Got 10ea #STS8DNF3LL from Digikey yesterday. Now my Vref voltage stays at 1.40v all the time.
Now that I have brought her back to life I had a problem when sometimes I would boot and get the matrix. All my voltages seem to be stable and right at stock with very little movement during 3dmark2003.
I did what you suggested and mapped out the mosfet and the ISL6522. Using the ISL6522 Specsheet I got an idea what voltages should be wHere. After several boot ups and measuring voltages I traced the problem to Pin #10 on the ISL6522. At boot up pin 10 was at 10.5v. On the Pin 10 feed is R93 which is a 10ohm resister and it was ohm'd at 1.5K. I pulled a #100 resistor from the r9700 and replaced. Now pin 10 reads at 12.3v
I have booted 20 times and no matrix and all voltages are stable. Been running 3dmark2001 in loop for 2 hours and no problems. I think the re-animation was successful.
MrCrusher Why did you make 12.3v? What's that about?
2PiZza, Nothing special about the 12.3volts. That is just a liitle less then what my PS is putting out. Pin 10 is looking for 12v at boot up to turn on the mosfet. THe resistor feeding pin 10 must have partially melted due to being over driven when I upped voltages on Vdd, Vref, Vddq so it was reading 1.5k ohm and keeping the boot voltage close to the threshold to turn on the Mosfet to power the memory.
More importantly what I think this means is that there are probably allot of cards out there that can be brought back to life with a little patience. I thought I fried the memory but it was actually the circuitry controlling the memory that fried so there is still hope for resuscitation.
Oh i can't agree with you more. I think a lot of theese dead cards only have a fried mosfet, Cap, Diode, resistor or whatever. It's great that your card is back from the dead. Nice going dude.
Now that you checked your resistors and basically every pin out of the mosfets and the isl6522 chip i decided to do the same. That one resistor you were talking about is ok on mine. It's a 10 ohm resistor. I checked everything and all ohm out fine. Since my card freaked out when i O/Ced it a bit and only put out 800mv or so i'll assume it's in reset mode because it's hitting current trip limit. I know how to fix that. But i don't really want to change that trip point. That sucker may have saved my ddr. I sure hope it did. But damn dude, I swear you never seen a cleaner 440 bench in your life. So this tells me it's Vref holding theese cards back as opposed to vddq. Vref is causing the sticks on wings and crap like that and the matrix. I was fairly sure Vref was causing the prob and i believe i've proved it. VDD will still have to be raised as needed though, like normal. Since i modded my vdd to put out more watts i'll stay at 2.725v max In fact for the 440 bench i was at 2.700 vdd. After some more experimenting i'll show you guys that mod. It's simple to do. Although maybe a bit dangerous.
MrCrusher did digikey have a mosfet with just a slightly higher amps rating and slightly lower RDSon? Made by the same manufacturer, STmicro? I just checked the one i installed and reading the data sheet i probably never should have tried it. Not a close match at all. Oh well another night i guess. Back t my game. Damn is the weather nice. I wonder if, maybe, ya.
Surf's up dudes
i have done 3 mods now on my herc 9800xt
de vcore mod de 1596 mod en the vdd mod...
de max i could get out of my xt =
545core(2.2v) & 436mem (3.2v)(hynix) cooled with ddmaze 4 20c water
anything higher then that would freeze mycard ......
That's a damn good o/c for 20C water imo.
OK my dead r9800xt is alive and kicking again. I'm using a 4804 off my r9700pro now. Seems to work the same as the original one. BTW the 4804 doesn't seem to be a Vishay nor an Infineon chip. No idea who makes it. But on bottom side it says.
LOPVC2
0942
2K05
and what looks to be a trade mark INDO with 2E02 next to it
Front side says
4804
0206
Well whatever it is it's working fine. About the same as original one before it fried. Might as well really give it a test playing my game.
Whoa is me. Still got some problems with Vref. Screen gets horizontal very fine lines when i first boot the comp. After 10 seconds or so it's fine. I'm thinking i better get the exact part that was on this card. I haven't noticed any issues after it warms up though. Plays games fine. i'll play with it this weekend and see if i can kill it again. ;)
MrCrusher, care to sell me one or 2 of those mosfets?
Catch, No problem I'll donate to you 4 #STS8DNF3LL and 4 #4804 for current and future experimentation. The best I have been able to do is 530core(2.21v) & 430mem(2.85v). This is with some dots and artifacts and completely clean at 520/420 all on non coolled h2o swiftech 22501 kit.
I just got yesterday one of these new high OC p4 3.0c over at the Fry's fountain valley. Stuck it in and I am running 15x267=4.0G!! so now I am at 8274 in3dmark2003 and 9000 is next. I going to need some serious cooling on the cpu/gpu to get to 9000.
I found this new way of stabilizing the vgpu mod today, and i'm going to do that mod...i've got only this questions...
does r1596 has 2 sides? what i mean is that if one side is to reading the ohms and the other to the solder point???
The pot that we will introduce is regulated to 20k or 40k???
Thks!
MrCrusher I got 8193 in mark03. 526/425, p4 2.8@3.66(14*260), OCZ 5:4 2-2-3-5@208, p4c800-e deluxe.
And I got 22275 in mark01 on this system.
r1596 ain't helping much. Card starts to whisle also.
er what tune does in whistle?
Ta MrCrusher and you're lucky with that 3.0c. and pretty good clocks on that card with just water.
CATCH22ATPLAY I mean the sound is like high frequency whistle. I'm talking about r1956 mod.
http://home.comcast.net/~rbmccall200...8/site1530.jpg
Ok, i figured that part of the resistance is the solder point...
Now...it's in the same place where we solder that we need to pick before on a multimeter and we should read aproximatelly 40k???
Then the pot that we will introduce in the mod should be regulated to 40k or 20k?
Thks!
VoRtAn_MaDgE you are talking about r1596? You should read about 41k. Put a 50r POT dialed at 40k. So you'll get 20k as a result. I tried to run this at 9k. It's ok.
Thks i figured it out before u answer...
I did the mod 5min before...
I will post some pics ...
Regard
Vortan
Pics...as promised:
http://www.pcdiga.com/Imagens/VoRtAn...able mod_2.jpg
http://www.pcdiga.com/Imagens/VoRtAn...able mod_1.jpg
The quality isn't the best but for a 2.1mp digi cam i can ask much more of his macro LOL
Regards
Vortan
MrCrusher ... Excellent rehab on that dead Vcard! :thumbsup:
Now, get busy with that new chip! :D
C22@Play ... quit foolin around ... no surfin till you fix your card, dude! :p:
Hehehehehehe, I gotta get some action Hobocrow.
2PiZzA, a high freq whistle? Hmm, that don't sound good at all. You're the first that's had this problem. It doesn't change the way your card receives volts etc. It just changes the current limits reset point. If you did some of the other mods that i did to the core then if it talked back to you that might be normal. Maybe you're stressing a component. I will have to assume it only whistles when you raise v core above 1.95 or so. Mosfets do switch usually from 200khz up to 1mhz. Is your 3.3v line kinda crappy? About all i can think of atm.
VoRtAn_MaDgE, nice dude. Hope the mods help.
CATCH22ATPLAY
I did just 3 mods. vcore, vmem and r1596. 3.3 is nice. I got about 1.97 on core when it starts to sing. :) Not very loud btw. I know that these chips which control voltages are pretty interesting things. They got oscillators and some other things. I'm really thinking that it's near it's limits of working.
Also no matter what resistance I put on r1596. 8kom or 22 or anything.
I was getting this sound on my 9700pro also. And I know people who were also.
Maybe your machine is so loud that you can't hear it. :)
I forgot something. It starts to whistle only under load.
I'm the one who as to say thks to u for sharing such usefull information...THKS m8!!!Quote:
VoRtAn_MaDgE, nice dude. Hope the mods help.
I'll try later to see if improves stability and a few more mhz on the core ;)
Regards
Vortan
:ROTF: C22@Play ... Like your new sig. ;)