Printable View
Odd, my 8800GT with Samsung GDDR3 modules responds rather nicely, always thought it was the Qimonda modules crapping out. I'd also like to know if the RV770 power circuit has the same 1.7v OVP and if there is a sensitive OCP circuit in place? Anyone know if RV770 coldbugs badly also? All questions I could answer in a week but I'm impatient.
Could someone measure the resistor between the pad near C601 (Vgpu) and the resistor (capacitor?) that is just beside.
I used to read 44k but once I put soldering on the pad I have now 0 ... I don't want to put the card to test I'd rather have a double check for both 44k or 0ohm. I don't see any soldering between them so I'm a bit confused.
Thank you
Quote:
Originally Posted by jason4207
You are correct about the Samsung chips. My GTS w/ Samsung loves voltage. It was theorized that it wasn't the Qimonda chips themselves, but the memory controller that was causing issues on the cards that had Qimonda memory. The reasoning being that turning down the voltage actually helped one obtain higher clocks, and at the brink of stability the timings on the Qimonda chips could still be tightened w/o having any adverse effect on stability. It seemed as though something else must be at fault. The memory controller was spec'd at 1.8v, but going from 1.2ns to 1.0ns Qimonda RAM only involved pumping up the voltage from 1.8v to 2.0v on the RAM (the controller had to come along for the ride since they shared the same rail). The Qimonda RAM did fine w/ the extra voltage, but the controller wasn't stable anymore. Iirc the spec sheet on the Qimonda has them rated up to 2.5v!Quote:
Originally Posted by [XC] gomeler
Hopefully, the Qimonda RAM on the 4850 will respond well to extra voltage.
then isn't the extra 5 gigs of ram wasted cause your on 32 bit os?Quote:
Originally Posted by Zucker2k
So, with all that said, is there a pencil vdimm mod on the 4850
hoping to break 950mhz tonight!! and some 3d01 :cool::p::yepp:
to bad i suck at os tweaks..!
I'm quad booting:Quote:
Originally Posted by Boogerlad
Windows Server 2003 - supports 32GB (with PAE switch of course). I did run the test without it as my x-fi card throws fits with more than 4GB of ram.
Windows XP SP3
Windows XP x64 SP2
Windows Vista x64 SP1
This is just the way I like it.
There's something strange with my Vmod : When I measure resistivity between the pad near the C601 (Vgpu point) and the second pad in the pencil mod, I get the resistivity of my VR :??:
So to improve the Vgpu I need to set the VR to less than 5.2k (4.7k for 0.11v increment). So why is there 100k mentionned? A 10k would have been more accurate don't you think?
http://www.aoaforums.com/forum/hardw...mod-guide.html
read the 2nd part below the "OR" that should help..:up:
the vgddr mod is kinda confusing. do i solder either whitish point to ground? also, where's the measuring point for vgddr?
Just to confirm:Quote:
Originally Posted by gurusan
Just the pencil mod will work, as in no need to solder the res as specified on the photo?
Also, instead of a pencil, as the graphite will wear as time goes on; can i use a trace pen like I've done on all my old school cpu pencil mods?
Thanks
yes, I just reused demoneyez pic, that's why it says to solder...just ignore that part. And I don't see any problem with using a trace pen as long as you monitor the resistance very carefully.
I see.. the traces from the pen could provide less resistance between the points then the graphite from the pencil? If so would this serve to increase the effective gain in voltage?
(I don't have a MM so i may just stick with pencil then, although thickness of the layer would do the same... hmmmm)
Idea though, is to connect those two points as shown in blue?
i am having trouble flashing the bios on my 4850 's
i flashed with ati flash . i dloaded a bios from one of these threads . rv770. 1.26v and 725/1125 . i changed to 725 1150 1.26v . i have to flash atiflash -f -p 01 7251150.rom and it worked . did the same using 02 and it seemed to flash.. both cards are in cf at same time of flashing. now in vista 32, gpuz 2.4 show clocks at stock.. what am i doing wrong.
Randy
i would not even think about attempting this mod without a multimeterQuote:
Originally Posted by darkskypoet
Try to flash again, if it gives you the pop-up that the bios is already loaded then reboot and check GPU-Z again. mine sometimes shows stock clocks but going to the sensor page or switching adapters causes it to read properly.Quote:
Originally Posted by rym
I agree with gurusan, a multi can be bought for as cheap as 5 dollars these days, no sense trying any mod like this with out it.
It's not a very smart move to do it w/o a DMM. You can get a DMM for $20. Much cheaper than replacing a $150-200 gfx card.Quote:
Originally Posted by darkskypoet
Be careful!
Understood, there is a bit more to it then I thought. I needed an MM anyway :)
Does any one have a limited table of values for what resistances will net you x voltage? As in a bit higher, or lower? Or is there a formula for figuring this out, for these two points.
Thx
I know you guys know. I hate to stick my probe in there while the card is alive. I'd prefer to just solder on my read point wires, and go from there.Quote:
Originally Posted by jason4207
Hook me up!
After looking in the 3850/3870 thread I'm pretty sure you solder the 2 legs of the VR to the 2 points on the pic. Perhaps one of them is GND?Quote:
Originally Posted by Boogerlad
I'm also looking for that measure point. I think I have it narrowed down to 2 different points based on the 3850/3870 thread, but I'd like to hear from direct experience...
ok i see one card has the clocks higher but the other wont show.. winflash says the bios is loaded already.. do i need to disable ccc or mom ?
this is the active card the other inactive shows clocks corectly. 725 1125 in gpuz
OK...I'm impatient. I stuck the probes in there (man this card is hot!), and my guess was correct. The points below show up as 2.01v on the DMM, so I'm pretty sure they are the measure points for the GDDR. Someone correct my drawing if anything is incorrect, please.
http://img124.imageshack.us/img124/3155/gddrsu7.jpg
Pic courtesy of Techpowerup
Also, I want to rig the fan to run 100%. Are the red and black fan lead wires the positive and GND? I'm pretty sure they should be, but I don't want to fry the fan. Yellow for sensor, and blue for PWM?
in hardwareluxx we have tested the vmod with a 25k variable resistor, but 25k is to big its not accurate enough, i think 10k is the best choice.
is the vmem mod still working ?
Question for those that lowered the clocks in 2D mode with RBE. I have changed Clock Info 1 to 350/500... And upped the fan speed. Fan speed works well, but neither CCC, nor GPU-Z are showing lower 2d clocks.
1) Do i need to modify both Clock Info 1, and 2? Or are there more spots that must be changed to drop 2d settings.
2) I've seen someone mention a 160 / 500 setup for 2d clocks. It sounds rather low, however is there a 'right' setting for this? As in something used in a MFg's bios? (again i seem to remember an MSI bios being mentioned with lower 2d clocks, but I've been spinning through so many forums, i am not sure If I could track it back down again)
3) With stock voltage, what sort of clocks are people getting? (Have to wait till i buy my DMM for the pencil vmod.)
(tips hat to all)
Thank you.
Edit: One thing I noticed, which seems strange; For Clock Info Modes used it lists 0 for low, medium, and high... can this be interpreted to mean that it is using Clock Info Settings 0 (Boot) for all three of its power play steppings?
so for vgddr measuring point, one of those yellow points to ground? and for vgddr mod, two whitish points to 100k vr?Quote:
Originally Posted by jason4207
If 25k isn't accurate enough then you want a higher value resistor...not a lower one.Quote:
Originally Posted by Specialfighter
I'm installing the mods as I type this. I'll report back w/ my findings.
Yes & yes, but it's actually green and cyan. You might want to adjust your monitor. ;)Quote:
Originally Posted by Boogerlad
I'll be able to double confirm that shortly once I finish doing the mods tonight.
:up:
to tell you the truth, i'm slightly colour blind =( sorry if i give out wrong colours of stuff. Thanks for clearing up the stuff about vgddr.Quote:
Originally Posted by jason4207
Well in that case I humbly apologize! :)Quote:
Originally Posted by Boogerlad
Good news! My mods work!
It looks like even w/ the VR maxed out at 100K the vMem mod starts out at 2.09v. That's quite a bit higher than stock. Maybe a higher value should be used...:shrug:
2.09v is safe, though, and I made my mods easily removable, so it's all good!
Pics will follow soon.
:D
****************************************
Pics as promised!
http://i300.photobucket.com/albums/n.../wholecard.jpg
http://i300.photobucket.com/albums/n...n4207/vGPU.jpg
http://i300.photobucket.com/albums/n...on4207/gnd.jpg
http://i300.photobucket.com/albums/n...n4207/vMem.jpg
http://i300.photobucket.com/albums/n...n4207/read.jpg
http://i300.photobucket.com/albums/n...n4207/idle.jpg
http://i300.photobucket.com/albums/n...n4207/load.jpg
http://i300.photobucket.com/albums/n...207/passed.jpg
Thanks! it's really neat!
I noticed your flash command:Quote:
Originally Posted by rym
-f - p 1 and -f -p 2
It's my understanding that the first graphics-device has device ID 0 and not 1, and the second graphic-device has device ID 1 and not 2
so the flash lines should be:
-f -p 0 for card number 1, with device ID 0
-f -p 1 for card number 2, with device ID 1
What happened is that with your first command you flashed the second card and with the second command flashed no card at all. Could explain why one of your card has higher clocks and the other does not
I was curious about that myself.. But i noticed the -f, and I never used the -f when i did mine, so i wasn't sureif it was related to that or not... But My single card was "atiflash -p 0 filename.bin" ... And so the 1 and 2 looked off.
I think -f just forces the flash regardless of device id and regardless of anything really.
Quote:
Originally Posted by rym
OK I just noticed the same situation, one of my cards was showing 625 after a reboot, I ran a bench (05) and when I exited I opened GPU-Z and both now show 900 /1100. try that to confirm, and please report back. If yurs stays like that open the current Bios with RBE and take a screenshot of the clocks tab.
Correct, I use the 'f' parameter mainly for biosses that differ from the manufacturer. For example to force flash a Asus bios in a MSI card, vice versa and so on...Quote:
Originally Posted by gurusan
bothcrads have acelero cooler 1.rv2
I have both cards running 775/1150 one at 1.26v and other 1.29v.
No volt mod on ram for both cards.
The one with 1.26 /pencilmod ran an hour under ut3 bot match great. I enter god mode and playdead and watch the screen with 31 bots. Ut3 is a great tester.
The other with 1.29v var/resister full 100k. needs to be tested better.. I played ut3 on the net for about 15 20 mins ..
want to see if it will run at these speeds @ 1.26v
Ram sinks get very very hot.
I want to find the sweet spot volt/oc/heat to throw on non variable resisiters hardmodded for duration of next year or so.
I think these gpu's like voltage .
Thanks demoneyes for your help here.
Pencil mod done for Vmem, It now reads 2.108v, let's see how far it can run now :D
What is the vmem pencil mod? Please show me :)Quote:
Originally Posted by Clemmaster
I'll second that with a Pretty Please :) :)
http://clemmaster.free.fr/Hardware/H...encil_Vmem.jpg
In the Vmem area of course :D
My memory can't go over 1130 :/ I'll try with more volts when I get the new cooler :D
thanks clemmaster. Much appreciated. Going to try now :)
what is default vgddr?
Thanks clem, I've just done a pencil vmem mod, bumped it up from 2.04v to 2.14v and so far am at 2400mhz (1200) stable
What are general "safe" voltage limits for gDDR3?
I don't know for those, I've read up to 2.6v but it seems high to me so I may be wrong.
I'll check on Qimonda's datasheet
I found the data sheet for these particular qimonda chips (HYB18H51232BF-10), but I don't really understand what I'm looking at :/
http://www.qimonda.com/download.jsp?...rev130_www.pdf
Perhaps someone else can take a look at that and let us know the max rated voltage?
It was 2.01v on my card.Quote:
Originally Posted by Boogerlad
I've read in another thread that 2.15v is safe for 24/7 operation, but don't quote me on that.Quote:
Originally Posted by gurusan
2.5v max
http://www.qimonda.com/download.jsp?...rev100_www.pdf page 28.
that still sounds high to me lol. I think I'll go 2.2v max and see how much I can squeeze out.
I would really like to hit 1250mhz as that will give a nice even 80gbps bandwidth :D
It says absolute max is 2.5v, and recommended max is 2.1v.Quote:
Originally Posted by gurusan
Edit: too slow
It says Absolute Max = 2.5v, Recommended Max = 2.1v
Considering stock is 2.0v, 10-15% is 2.2v-2.3v, which should probably be ok with active cooling over ramsinks I would imagine
Recomended max. voltage is 2,1V, the absolute max. is 2,5V, exceeding that may cause irreversible damage.Quote:
Originally Posted by gurusan
That's on chapter 4 of the datasheet.
Personally I wouldn't mind goind for 2,2V, maybe 2,3V with good cooling, but wouldn't try more than that.
Edit: Too slow, this is already posted, I think there's a lot of people readind this thread :D
Where does it say recommended is 2.1v? For them, the operating conditions should always be under 1.9v, so that's a confortable value for them, no dead chip.
Nevermind I wasn't looking at the right pdf :D
Thanks for all the clarification. I noticed it also said keeping them under 85c is a good idea :)
Is it okay if I solder a 50ohm vr if I set it at 100ohms?
If it is, will it have a different effect on my videocard?
Sorry, can someone confirm which is the correct vGDDR read point?
http://img515.imageshack.us/img515/3...2957sr6gb4.jpg
read this post! any green spot to ground.Quote:
Originally Posted by jason4207
^thanks.
dunno how i missed that.
Just wired up my card. I've got everything hooked up except for the VGPU test point. Anybody know where that is? I went over the thread 2x and couldn't find it.
in this pic the top of the picture you will see a resistor near the number c646 use the side of the resistor that is towards the power plug end of the card. sorry we didn't include it in the original info. it was the same as the 3850 so I assumed most people knew where to check , sorry. if you need any more info ask away.Quote:
Originally Posted by adpr_02
http://www.xtremesystems.org/forums/...4&postcount=52
Hi, will my current Titan twin turbo: http://www.titan-cd.com/eng/vga/ttc_csc88tz.htm
fit on 4850?
and will those heatsinks:
http://pcb.se/ItemInfo?itemId=795607...A+RAM+Heatsink
cover all the power managment fo it?
P.S. how much dou you think core can be clocked up with zalman vnf-1000 level air cooler and only pencil voltmods?
I would have to say no. on the twin turbo, the list doesn't include the x1900xt or the HD2900 or the HD3800 which all had the same mounting pattern as the 4850. if it will fit the cards I mention it will work if not than no guarantee. but if you know anyone who is handy they could mod it to fit I imagine.Quote:
Originally Posted by Moty
the ramsinks are fairly universal it all depends on what you are cooling it with .
Hey, thanks a lot. I'll get that hooked up and hopefully do some benches soon. Can't wait for that D-TEK gfx block...:p:Quote:
Originally Posted by jimmyz
Quote:
Originally Posted by jimmyz
There's what it's capable for:
(i use it with x1800gto for now)Quote:
ATI
RADEON X1600 / X1800 / X1900 series
As i know hd2900 have core protection metall pad that is higer than core itself and that was the reason that there was many coolers which had no contact with core, so they cannot be used.
I will try to find information about mounting holes of 4850... hmm where to start...
OOPs i think I missed that at the site. the hole spread is 2.1 inches center to center. which is 53.34 mm approx.Quote:
Originally Posted by Moty
that's 100% precise measurements?Quote:
Originally Posted by jimmyz
is this precise enough for you?Quote:
Originally Posted by Moty
http://img525.imageshack.us/img525/8...e076dm0.th.jpg
I wouldn't tell you something if I wasn't sure.:up: since my calipers are in sae. I am assuming the metric would be rounded down to 53mm
can somebody show us the exact points for solder the potis ?
discribed for idiots ?
that will be nice
Quote:
is this precise enough for you?
Well thanks 4 the photo, but can you measure @ 45 degrees?
@-----------0
--------------
--------------
--------------
--------------
--------------
0-----------@
i need distance from @ to @, thank you.
Quote:
Originally Posted by Moty
2.965 inches which is approx 75mm (75.33) without the other items needed like guage pins that is as accurate as I can get.
here is a diagram for Specialfighter and anyone else who wants it.
[IMG]http://img235.imageshack.us/img235/5...captionto8.jpg[/IMG]
Quote:
2.965 inches which is approx 75mm (75.33) without the other items needed like guage pins that is as accurate as I can get
Science Titan uses a diagonal measurments, My twin turbo wil fit on 4850 because it uses 75.4 mm mounting holes which is same for:
RADEON X1600 / X1800 / X1900 series
RADEON X1950 series
RADEON X2400 series / X2600 series
Thank you. I will now know that my cooler is good and that all with your help. I'ts 3.29 am at my time zone and i can now have stress-free sleep(i need to wake up at 6-x ) :)
Any pencil for 4870 yet? :)
Is there any chance for softvlotmod at 4850 when oc tools get updated?
in like 4 steps and mostly downward. same goes for bios modding. if you want to have more voltage you must do the hardmodQuote:
Originally Posted by Moty
Just penciled to 1.18v idle 1.48v load :)
tested at 850/1100 so far and of course passed with flying colors. Am hoping to do over 900mhz on this voltage.
28c idle and 34c load :)
http://img105.imageshack.us/img105/4341/51832066ni2.jpg
whats the safe max v
??????
1.3 for stock with fan on high
1.3-1.4v tops for after market cooler/
water 1.41-1.6
sorry guys,
Can you post a photo for 4870 vgpu mod?
wich trimmer can I use?
under watercooling, Which is the max safety vgpu for benchmark and wich for daily use?
thanks
Only way to know your OC is satble is to run FurMark on Stability Test, full screen, no AA, for 5 minutes.
This soft heats my 4850 almost 10°C more than 3Dmark, and I m on water.
HD 4850 voltmod article posted: http://www.techpowerup.com/articles/...g/voltmods/151
and for 4870?Quote:
Originally Posted by W1zzard
no card here yet => no pics => no article
update
950/1200 on 1.48v :) this card is awesome
sub'ed
;p
I killed an 8800GT w/ that program. I don't use it anymore.Quote:
Originally Posted by albon
Best stability test is to play games like you normally do. I use ATI Tool and benchmarks to get a rough idea, and then use actual game play to fine tune.
And some people are just looking to be stable enough to pass benchmarks. No need to run FUR in that case.
When I run FUR I feel like I'm shortening the life of the card. Just my opinion...
i find rthdribl with 16x multisampling and fleck/bump on to be a very good stability test
So is the Crysis gpu bench
Yeah rthdribl was my tester of choice, got a little warmer than most things but not absenly hot. Its my temp tester of choice... and usually I can see it error before it crashes so im not getting a hard system crash.Quote:
Originally Posted by gurusan
I find the Crysis GPU test a good one for testing memory. If mem is clocked to high or insufficient voltage, I get nice black squarelike artefacts, grouped together. Highly noticabe. Also AtiTool will artefact like helll when running the 'scan for artefacts'Quote:
Originally Posted by gurusan
If it died from that program it wasnt a very good card.Quote:
Originally Posted by jason4207
Anyone have any fixed resistor values for vGPU or vDDR @ specific levels (i.e. 1.3v, 1.4v // 2.1v, 2.2v)
What?Quote:
Originally Posted by jason4207
No.
Higher value means with each turn you are changing the resistance by a larger amount. Lower value allows smaller adjustments to resistance.
Maybe He was talking about higher stock voltage with lower resitance (at its maximum value)?
@Theta: you can find the resistance with this formula : Rnew = Rold*Vold/Vnew assuming the stock values given in the pencil mod we showed ;)
GPU : 1.15v -> 5.12k (stock, 3D mode)
Mem : 2.04v -> 3.94k (stock)
I'd just like to point out the seriousness of this post to anyone else thinking of doing this with out a Multi Meter. This really cant be stressed enough.Quote:
Originally Posted by gurusan
I was just attempting to go from 1.31 to around 1.36 load.
I applied a couple strokes of my pencil, booted up, measured...
AND WHAT THE :banana::banana::banana::banana:ING HELL PANIC ATTACK!!! :shocked:
It read 1.89V, I kid you not.
Luckily, I had a quick response to hit the switch on the PSU, and no damage has been done, but not having a multi meter I'm sure the chip would have been fried before I got to windows. :yepp:
Also for those using it, or thinking of using it, I've found the S1 starts to gets over loaded once you start going over 1.4V
The card just heats up till it shuts it self down, and that's with two 120mm fans going.
i will do the mod tomorrow, my is hd is on water i hope i can go to 1,5V
its always a good idea to measure resistance across anything you pencilQuote:
Originally Posted by biftek.
before you pencil.
sometimes it is very difficult
is it hard to volt mod with pencil? what pencil did you use? and how many times did you make a line? i dont have a multimeter but want to try it with my S1 when i get my card/
yeah I always measure the resistance before plugging the card back in
any one having a screen flickering issue? my flickers all the time on desktop. Didnt happen until i flashed with the msi guru bios.
Meaning it goes off / on in a flash continuously ? Did you try lowering your clocks ?Quote:
Originally Posted by tvdang7
Errr... Shouldn't you have posted this in the other OC thread ? :p:
Yeah i should have :p:Quote:
Originally Posted by gurusan
i made an exception for some stupid reason that time.
Anyways it's back at stock for now, I decided to leave it be for a while :lol2:
found what it was my memory needed to be clocked lower which is wierd.Quote:
Originally Posted by Freaky
before the flashes CCC would be fine at those speeds but with a flash theres flickering. although its stable owell.
im at 740/1025 using bios flash visiontek here.
Well....let me clarify. The main reason my card died was b/c I had mounted the HR03GT improperly, and it wasn't giving equal pressure across the die. I fired FUR up, and after a few minutes the intense heat coupled w/ the uneven pressure made a corner of the die chip off. This is my theory anyway, and I saw the small chip on the corner of the die.Quote:
Originally Posted by STEvil
Ever since I've been somewhat scared of FUR...probably for the wrong reasons, but I do feel that you can be game stable w/o having to use FUR to test stability. I use ATI Tool, individual 3Dmarkxx test loops, and Crysis benchmark.
I see what you're saying. I worded it wrong. My point is that...Quote:
Originally Posted by STEvil
If I start at 100k, and you start at 25k I need to turn my VR 3/4 of the way (19 turns) just to get to the same Ohm value where you're starting.
If you assume 25 turn then relatively speaking the 25k will have 1k/turn while the 100k has 4k/turn, but the 25k starts so low that it doesn't really matter how accurate it is.
I'm unsure what vGPU would pop up on the DMM if you started w/ a 10k maxed out, but I know I wouldn't try it. I think 100k is plenty accurate, and I wouldn't use anything dramatically lower.