Robbo

I don't think you understood my post, it's nothing to do with your system BIOS, so I was saying that by you trying to reflash your SYSTEM BIOS that you weren't going to get anywhere for the reasons I explained. So, I'm not surprised you're back at square one (at the same point as before). Have a read of my last post (post #4388), those are the options that are open to you. It's a pain for you I'm sure, I'd be very frustrated in your position, but there are solutions & they're summarised in my previous post. @svl7, would it be worth summarising somewhere at the beginning of this thread which systems are known up to this point to be incompatible with the various vBIOS you have created? That way people won't repeat the unfortunate experience of Anil Kumar.

Well I guess it wouldn't hurt to try to support the heat pipes in the areas you talk about, to take pressure of the mounting system, but heatpipes aren't very heavy so I would be surprised if that makes a difference (would have thought the 4 mounting screws would be enough to support that weight, plus the additional 3 screws you mention - mine only has the 4 mounting screws for the GPU core & no more). You have a different laptop to me so the leverage & force on that area might be different, not going to hurt trying to support it some more I guess.

If it's based on the GPU's getting hot, then you could repaste to prevent them throttling. Strange that you say the latest drivers would make that happen. I don't think that is controlled in the drivers, I think the temperature throttling point is set in the vBIOS. I think it's 87 degC, that's what is says in NVidia Inspector for the one I'm using (never gets that hot though, so don't know for sure if it would throttle at that temperature). How hot are your GPU's getting? (And in my experience the driver doesn't make the GPU run hotter or colder either). - - - Updated - - - Ah, that definitely sounds like a heatsink mounting manufacturing defect to me (on the GPU). I'm sure you know that the heatsink surface needs to sit flat against the GPU core, and from what you've described it sounds like there is a gap in some places where it's not making contact. The heatsink should make contact with the GPU even if there is no paste applied - the paste is there just to fill in the microscopic gaps & holes in the heatsink and GPU surfaces (because paste transfers heat better than a tiny air gap where those microscopic holes would be).

Yeah, that's cool, IC Diamond is supposed to be one of the best, but I don't think it's because of the thickness though - it's just higher performance. With it being thick it's harder to apply and do a good job of applying it, but if you do manage to do a good job of it, then it works well.

If your heatsink doesn't apply much pressure you actually want a thin paste, not a thicker one. The heatsink needs to be able to squash the paste into all the microscopic nooks & crannys (and allows the heatsink to sit as close as possible to the GPU core), and a thin paste enables that to happen at a low mounting pressure. Thicker pastes are better for higher mounting pressured heatsinks - like desktop applications rather than for notebooks. MX4 would be ideal for notebooks.

Although I don't know if that will help Anil Kumar. He has a 120Hz Alienware, so the IGP is disabled, and his 880M won't post due to the flashed vBIOS. Any changes to the system BIOS isn't going to magically fix that conundrum - it can only post using the dedicated GPU (and that's borked). If he had IGP that was functional (Intel HD4000), then he would be able to recover it, but 120Hz is not supported by Intel HD4000 hence deactivated in system BIOS. I think he's stuck. Soldering or finding someone else to flash his 880M, or return to Dell saying that it's just stopped working (without providing the details!). That's my understanding of it. (Some independent computer shops can flash vBIOS to cards as a service I heard.)

With temperatures that low, then I'll be surprised if you're seeing any throttling, unless it's for power reasons: either the power adapter or mainboard being inadequate or limited. Chances are you're fine with the stock vBIOS for your purposes, considering you're not wanting to overclock. If you're not seeing a constant 850Mhz (max boost for 780M) on the core when you're gaming then you can benefit by using the modified vBIOS which runs it by default at 850Mhz.

Ah, for the 780M, that could be a different story. Might be worth flashing the vBIOS even if you don't overclock, but it depends whether you're seeing throttling at the moment or not. Do some gaming & monitor your GPU core & memory clocks (and temperatures) using a monitoring program, like GPUz. If you're experiencing throttling, then the modified vBIOS could help. If you're getting throttling due to high temperatures though, then you'd need to sort that out, no point flashing the modded vBIOS if it's the fault of your temperatures.

It's unlikely to improve the performance of your GPU unless you overclock, the GTX 770M (you made a typo I think) is a cool running card, especially in the AW17 so you're unlikely to get any kind of throttling with the stock vBIOS with that particular card. There's no point in flashing it unless you want to overclock the core beyond the +135Mhz limitation in your case I would think.

A long shot idea related to the poor performance seen of the 880M with the latest drivers. I know I don't have the 880M, but this might be relevant. I ended up having to reset my CMOS today (which also reset my system BIOS) in an effort to reflash my 670MX that I bricked today. I managed to reset the CMOS & reflash it with svl7's vBIOS, but I noticed that GPU performance was down 10% across the board at any given overclock. This was strange as GPUz was reporting no throttling & 100% GPU loads during benchmarks, but global wattage consumption measured by my KillaWatt meter was down 20W from peak. I eventually narrowed the problem down to me not having disabled the IGD in the M17xR3 BIOS - I had just enabled the PEG as the main graphics adapter, rather than completely disabling the IGP (even though it wasn't present in Device Manager). Anyway, after disabling IGD in the BIOS my KillaWatt readings are right back up to their normal peak values & so are the benchmarks & game performance. As a long shot I wondered if this was somehow related to poor performance seen with 880M and latest drivers (I'm on 340.43). Wondering if IGP's need to be disabled in systems to see the best performance. A long shot I know, but I just thought I'd put my experience here as it might be part of a puzzle, then again maybe not! (It just felt like some of the restrictions seen with the 880M and the latest drivers).

(Will provide benchies & screenshots so you can see the fruits of your labour!) ;-)

Yeah, that's cool. I've requested from svl7 an extension to the 670MX vBIOS that I'm using (and that he provided) so I can use 1.1V and over 1200Mhz on the core. Hoping he can mod it for me. I should be able to get 1200Mhz on my 'measly' 960 cores! ;-) (reckon my GPU would consume just a shade over 100W with that overclock given my current KillaWatt readings)

I have to say that that is very impressive the amount of power that it can draw, and given what you said about the latest A18 offering being castrated to 330W, then I can see why you're fervently pursuing NVidia & Dell with their 880M & A18 fiascos! Long live tweakability & power to the user! ;-)

Hi svl7, any chance of expanding the voltage & core Mhz limits on the Clevo 3GB 670MX vBIOS? It's this one that I'm using of yours: Clevo 670mx - 3GB - 80.04.58.00.03 - 'OC edition'_rev02.zip Could do with voltage up to 1.1V, and Core Mhz slider up to a little beyond 1200Mhz. Many thanks if you can provide one, and I'm sure they'd be a beer in it for you! :-) (Currently it's limited to 1.05V max & +542Mhz on the core, up from the stock 600Mhz and 0.925V)

One of the differences is listed at the start of this thread in Post #2. Search the page for 80.04.29.00.01 and you'll see what I'm referring.

Very frustrating for you I can imagine! I hope they don't put as many 'controls' on the next gen - part of me thinks NVidia are doing this to limit overclocking for 2 reasons: 1) to decrease failure rates & warranty claims 2) (in future) to make sure flagship GPU performance will not be achievable by overclocking of lower models, thereby creating more of a distinction between their models & potentially maximising their sales & profits of high end models. Hope some of these strange constraints aren't seen on next gen - especially in mid high end range. (I like the idea of buying mid high end models & overclocking them to stock flagship performance, hopefully that can be a possibility still in the next gen)

Ah, so looks like the 880M issues with the latest 340.43 driver are not fixed then by disabling secure boot! I'm done following this issue and trying to summarise solutions found in different places on the net - there's just too many inconsistencies to be found in many different areas, especially as I don't have an 880M. I feel like I'm just adding to the confusion! ;-) Hopefully 880M problems will be fixed soon, and failing that new GPU's to be released around the corner!

Ah, cool, I didn't realise you were referring to the fix for the overclocking problems (overclocks not being applied) brought about by the latest driver 340.43, but it clicked when I read this post in notebook forums: User Nightdex: "Word of warning guys. If you update to this driver and you own an 880m and also want to over clock. Disable secure boot for good. Secure boot restricted slv's vbios for the 880m on my AW 17 with this particular driver. I went through a ball ache and a half to resolve the problem. Edit: see my posts from last night in John's 880m review thread for more info. I ran some tests to show that I managed to get my 880m over clocking nicely again. I'll add more results as I go along." Nvidia 340.43 Beta out now. - Page 2 So does look like 880M users can use the latest driver as long as secure boot is disabled.

Haha, yes, must feel like walking on egg shells!

Wow, these 880M cards seem like a right mess, and also seems like NVidia's latest beta driver (340.43) just makes it worse based on what you're saying. What a cluster f*ck! Can anyone else substantiate that 340.43 driver breaks overclocking on 880M cards? Perhaps the advice should be not to uprgrade to 340.43! Although, it's not really workable keeping your GPU on old GPU drivers. What a mess.

Yep, you'll be good for 120Hz, but only if you disable the IGP (Intel Graphics Processor). - - - Updated - - - Google how to manually set fans in HWInfo64, they'll be loads of info.

85 degC is kinda warm for a GPU, but at least by flashing a vBIOS that doesn't increase the voltage, then that's pretty safe I think, just work to keep it under 90 degC. As for working out max possible fan RPM I used a program called HWInfo64 to manually push both my CPU & GPU fan to it's limits, I then used the monitoring software in that same program to view what the fan RPM was at max forced RPM - this I concluded to be my max possible fan RPM. I then closed the program (and rebooted to be on the safe side) to make sure that normal fan control was restored to the system (ie no longer controlled by HWInfo64), and proceeded to monitor fan RPM when gaming. This enabled me to work out that I get max fan RPM on the GPU at 65 degC in my case for my laptop. Therefore, I knew I didn't need to implement any kind of extra fan control. You could use the same process to work out what your max fan RPM's are & how your system controls them in relation to GPU temperatures.

Yeah, I'd stick with a vBIOS that has the same version number if you can help it, and plus I think you'll be able to get your desired overclock on stock voltage, so you may as well use that one and see how it goes as a first step. EDIT: based on what Mr Fox has just posted, then looks like you'll be OK flashing the ones with increased voltage. But I'd still stick with my previous advice as a first step & flash the one you linked, as you'll probably not need increased voltage to get those clocks.

I think you might be able to attain those clocks at stock voltage, or the 1V version. Try them & see how high you can go with stability & good temperatures.

Yeah, that's right, he just meant from a benching point of view using the unlocked vBIOS, as long as your cooling is good enough. It's a point of reference for other 880M owners. If they can hit the same 1125Mhz core then it should score the same as his result. Your score seems fine to me though.