Ollie87
-
Posts
40 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Everything posted by Ollie87
-
I get crashes from around 95 watts onwards in-game, vrms are all covered i used silicon paste on them instead of pads as was worried they weren't touching the copper heatsink, I've seen your post on the backlight pin modification, I will check this out further. The only way it works better is setting the core clock to lower MHz, it's stable with my maximum on 1350mhz, I'm going to tinker tonight with the settings and see what I can do with the voltage unlock. Can't afford to buy the hp version they are very expensive around £800
-
I still seem to have an issue with power, it managed to complete a GPU test on 90 watts, any higher wattage causes crashes, so far the best thing I've been able to do is limit the core MHz and not tamper with the power. I have the screen and got a edp cable coming I'll keep the current screen on and boot the bios and throw it into PEG mode and try the other screen out, it's worth a go to see if by chance I can get it to work, should I be looking at changing it to muxless or keep it on fixed and full
-
MSI is working all good, was me being a fool, ive ordered the edp cable for the m17x r4 - i will investigate the screen next, doesnt arrive till the end of the month
-
Ok the divers 528.24 allow me to change the watts but only in power shell it runs stable on 90w and also 100w, I don't have to adjust the core clock speed now, I'll investigate how to set MSI afterburner so I don't need to do this. Looking good so far
-
ohhh no, i will look at them now thank you SuperMG
-
tried an older gpu driver, still dont have control of the power on the card, so far the best thing i can do is just downclock the core, im going to see if i can flash the vbios with a similar vbios that possibly has a lower TDP
-
by drivers are we saying the actual GPU drivers, i had tried to download the Beta driver for MSI afterburner that should unlock voltage control but i could never get it to do this.
-
@SuperMG ok i will look to see if i can edit the vbios, on the lvds screen it works ok the maximum i can set the core mhz of the card is 1300 mhz on a power shell command. it is stable like this and i can play games for an actual session. any higher with the core clock and it does crash or under test i see the core drop to 300 mhz. when i set the clock speed lower the cards seems to not spike or at least go past 85 - 89 watts i havent had any success on using power shell to alter the power usage of the actual card. i will investigate if im able to flash the card with the hp vbios or if im able to lower the TDP.
-
oh didn't even realize, when bought they had 2 in stock, yes i did see the hp variant, has slightly wider card, the card arrived new, would the shunt mod have any impact if i can allow it to use more watts of power ?, it is working if i downclock it slightly it seems the power draw never reaches 80watts which is strange. I may do some digging on different vnios for the card itself
-
Ok, after doing some more research I believe I have found the shunt resistors marked r005 on the rear side of the board, I will carry on searching to see if these are definitely the ones to check and if so I will look to add extra r005 to each of these, I won't be doing this till I'm definitely sure of what I'm adding to, I will most likely add padding and liquid metal cool the card once this is sorted.
-
This is the one I had bought, is this not the HP variant with adlink ?
-
So if I was to get the 120hz screen that works on the edp that would solve the issue ?
-
The rtx 3000 is a hp zbook version second hand runs flawlessly on the lvds screen. I then decided to pick up a new hp adlink Rtx 5000, besides having to get a better cooler it runs well untill it seems to power spike then the card vanishes and greys out on hw info looking at it's max power when this happens is always around 117watts, I just assumed this was to do with the power draw, I never bought the laptop with the 120hz screen, I'm unsure if it's worth tinkering with the screen I had bought, it seems to work fine if I lower the cards MHz rating down, I'll try a more demanding game, mainly was trying to fiddle with it untill I could purchase a newer system to take away on working away. Unsure if the shunt mod would help this in any way. Not so concerned with the screen as I was going to purchase a secondary 160h travel screen in 2k.
-
I have checked all of my video options in the A15 unlocked bios regarding muxless fixed and full dynamic, full dgp fixed and full dynamic, none of the options allow in full screen mode to achieve above 60 FPS, before I had installed the Rtx 5000, the Rtx 3000 was ok in full screen under test achieving around 90 - 100 FPS, I can only seem to achieve this now running graphics tests on widowed mode, sadly I don't have another laptop to test this on, my bios is currently set too SG mode full and fixed dgpu dynamic power, I will try and get some photos of my bios set up. When testing the GPU I noticed the core MHz fluctuates between 1400mhz to 1550mhz rapidly, I've now set the target in poweshell to 1250 MHz and is still stable no crash. I will try and ramp it up till it crashes again. Forgot to mention when it crashed on hw info it showed a error on power limitation under the GPU - 60% where it had greyed out. Thanks for some input SuperMg
-
I seemed to have successfully used power shell Nvidia command to reduce the core MHz on the GPU down to 1000 MHz from 1500 -1600mhz, this in turn has lowered the power usage managed around 6 games on war thunder on max settings dx 12, i did try the command to set a max wattage of 90 or 100 but the power shell sent back a answer stating it could not change the power. At least to some degree the card is usable not ideal at least untill I can work out where this shunt mod is actually placed.
-
New update, I have since bought a new 330watt power supply, this has not resolved the issue with the adlink Rtx 5000 card, under heavy load it must be power spiking and shutting itself down, card reaches around 68c using duronaut paste and a full copper cooler. Other options I currently have is to see if I can manually set a lower power target through shell commands or see if I can fit a shunt mod just not sure what I'm looking at as the information on this is sketchy at best and don't want to fry the card. Any help would be great.
-
-
Please can I get some assistance, where am I placing the 1 or 2 r005 resistors, I have included the photos of the card, i assume it's one over each 220 resistor closest to the cards bigger copper traces
-
so far, ive now installed A15 bios and then the unlocked version, ive now also installed game ready 2080 super notebook drivers - it runs better and seems to manage more gameplay before a crash, im going to investigate further with graphical settings seems to be underload it kicks the gpu off, it never does reach max temp, power draw can be the only conclusion, i need to research the shunt mod as its difficult to find this out. i also have somthing very strange which has happened since ive installed this card, on 3d mark its hard locked at 60 fps unless i go window mode, ive checked many settings in the nvidia control panel, i have changed the DVR options in windows and also switched off gamebar. an update to this post, i have tried many different things out lowering graphics etc and still get the same result, i have now ordered R005 resistors to shunt the card, for a temporary measure i will swap the card out with my RTX 3000. once i get the parts i will look to shunt the card and see if this resolves my power issue.
-
Ok this is what's happening as an example happily gaming and about 10 minutes or less in, game will hang and then the GPU memory etc becomes greyed out, temps never hit max got custom fan curve running, I can only put this down to power draw
-
-
I had picked this up to change the LCD over with ?
-
Yes, it's the 60hz screen currently I have a new panel but haven't fitted it yet it's a 2k 165hz panel with the 40 pin connector for the edp connector 17.3 inch, I just haven't felt brave enough to get the glass off the original LCD panel yet. If I run throttle stop and a custom fan curve it's not running bad to be fair, I believe the bigger culprit causing issues in games is asynchronous on dx12 causing the system to crash on games such as dying light 2
-
I've now fitted a complete copper heatsink temps are all good no more overheating, throttle stop is a must I'm definitely getting power issues to the card though I'm trying to find out information relating to the shunt mod now 😂 I think I'd of rather left the Rtx 3000 in it.
-
Hi all, having power delivery issues I think, running on max settings for example on dead by daylight 2 with ray tracing enabled I should be getting around 55 FPS, shortly after the drivers crash then causing the GPU to show 0C in hw info. I will look to downclock and reduce the millivolts but MSI afterburner is a bastard to get the controls unlocked. Does anyone have any knowledge to shed on this ? Didn't have as much problems with the Rtx 3000. The heatsink from memory is the one that was on the 660M, I can see the difference between that and a photo of the 100 watt rated version, I have considered ordering a full copper heat sink maybe that would reduce the heat or potentially getting a LLyano V12 laptop cooler, I have just changed drivers to the enterprise drivers and will test. Haven't tried with the edp and 2k panel el yet untill I know the card works ok, don't want to start a new head ache just yet 🤣
