Nvidia clockblock: vBIOS (unblocked in 353.00)

I received my NP8652 a couple weeks ago. My vBIOS date is Oct 7, 2014. Running on driver 347.52 on the 980m and I am able to overclock up to the +135 and around +400~ on memory with no problems. I will soon be installing Prema's bios but just haven't had time yet. Hopefully with Premas magic and drivers that allow OCing I will be good with this laptop. Cause as of right now....I'm absolutely loving it!!

On an off topic point...
The i7-4720HQ seams to be an absolute gem in this thing....aside from being soldered. I am running it at its (sadly only +200mhz) max overclock with a -95mv undervolt and on a 10 min stress test in XTU it runs a rock steady 3.59ghz and stays at 80c or lower with auto fans on. My i7 4810MQ could not even hold 3.29ghz with its max -55mv and near max temps with max fans.

On edit...
I was in a hurry and looked at Nvidia's 'recommended' driver up date...which was 347.52.
I currently have 344.56 installed, not the blocked driver.

 

Wait you can overclock on a clockblocked driver?

 

ok so hes ocing on a clockblocked driver without any mods and is also able to oc his HQ cpu and keep it at its max boost.....forgive me if im skeptical?

Sent from my Nexus 5 using Tapatalk

 

Well, I think there is a reasonable explanation for that... obviously, it's a magic GPU.

 

Pics or it didn't happen

 

Totally my bad you guys. I was in a hurry and looked at the 'recommended' driver update not the current installed driver. I'm NOT running on the 347.52 clockblock driver. I am running 344.56. Didn't mean to get everyone all riled up. But with that said, everything else was on point. Honestly when I was typing in the 347.xx driver number I was thinking something didn't look right.

Oh...and to Octiceps, I'm guessing you where referring to 'pics or it didn't happen' was in reference to me saying I was OCing on the blocked driver, which was obviously my error....not the seaming well running 4720HQ. But if not, let me now and I will post it someplace where its relevant.

 

Here is the screen shots of the a fresh 10 min stress test via XTU with the 4720HQ. The first 5 min was autofans on the laptop cooler, the second 5 min was flat on a soft tablecloth so the temps crept up a little but it still stayed rock solid at 3.59ghz with zero throttling.

 

Yeah, the reason is because you undervolted so much that it doesn't run into a TDP limit during that stress test. But with such an undervolt you're lucky. I couldn't do that with my chip. Congratz on having a pretty good 4720HQ!

 

Indeed.. That looks like the best 4720HQ I've ever seen.. I wouldn't mind it too much if my own 4720HQ was like that but your like 1 of the 50 people I've seen with this processor able to maintain those clocks so the chances of getting a good chip are very low..

 

He can only hope that the motherboard will not be changed due to hardware failure. Then he can risk a far less performing BGA processor ..

 

Indeed..

 

After being informed of the vBIOS clockblock:

*golf claps*

 

http://forum.techinferno.com/genera...t-overclocking-mobile-gpus-21.html#post125325

 

This clearly points to Nvidia's doing. If every single 900M GPU they've shipped to vendors since December has a clockblocked vBIOS, how can Nvidia deflect blame to the OEMs?

 

Nvidia does as Pontius Pilate, deuce his hands, leaving case to the OEMs ..

 

The weirdest thing happened to me and now it all makes sense ... I had several problems with the GTX880M that originally came with my laptop and Alienware sent me a replacement GPU a couple of weeks ago ... After installing the GPU I noticed that it has amazing AISC quality (+80%) & Samsung VRam & thought im gonna ramp that baby up like crazy ... What happened is that the GPU could take up to +500 MHz on the ram without sweating everything goes all fine, stable & cool but as soon as I touch the core even if it is +20 MHz 3dmark11 would crash immediately ... After that I thought maybe I was very unlucky when it came to core OCing but now I believe that maybe they extended the whole VBios OC block to even newer produced previous generation cards.

Conspiracy theory much ? or I could be right ?

 

Check with others which version vbios they have ..

 

As soon as I get to my place I will just flash the bios of my previous GPU ... Too bad for Nvidia & AW I always keep backup of my bioses

 

I don't think so. With the clockclocked vBIOS, you can't adjust core clock at all. The slider is simply grayed out. The 880M is known to be a chip pushed to its limits with little to no OC headroom on the core (in fact you'd be lucky not to throttle below the stock 993 MHz Boost clock). The crashing you experience is probably the result of that. Only 900M cards have the clockclocked vBIOS.

 

I think that both the OEM's and NVidia have had a hand in this clock blocking, whether it be through vBIOS or driver, with the exception being the manufacturers that have been advertising OC'ability (MSI and Asus - IIRC). Manufacturers like Alienware, etc, would know whether or not the vBIOS they are using is locked or not, so they've been involved in any decisions and discussions on the topic - it wouldn't have just been implemented by NVidia without any communication to them, which is part substantiated by the statement from NVidia to notebookcheck.net that the OEM's decide what features are enabled/disabled in the vBIOS. The fact that NVidia also blocked overclocking in their drivers shows they have a hand in this too. Who the driving force is behind these decisions - I think both the manufacturers & NVidia (with the exception of the manufacturers who have advertised overclocking - MSI, Asus). I think they're both to blame, and we're not letting either of them get away with it - I think they'll be reversing both of these blocks, not just the driver block which we know will be reversed.

 

What if ASUS and MSI have clockblocked vBIOS too?

 

Yes, that would point to NVidia pushing this on them, but those two manufacturers don't have clock blocked vBIOS so far. (As far as I know they don't have clock blocked vBIOS).

 

We'll see about that. If nobody has yet received an ASUS or MSI with newer than December vBIOS, that could be why. Or ASUS/MSI owners with clockblocked vBIOS didn't attempt to OC, so didn't find out about the lock. Or even if they knew about driver OC lock, didn't know about vBIOS lock, so assumed next driver will bring back OC, only to be surprised.

 

The thing is I have the option greyed out in MSI AB and EVGA Per only NVinspector works and on the same machine previous GPU didn't have that

 

Could be OEMs are trying to make gaming machines thinner and quieter and creating cooling and cases that barely meet the requirements and dont account for OC anymore. Except Maxwell is runs too cool ATM for that concern.

 

Then its likely an issue with your optimus setting. That would be where I would look for a fix. 880M did not have blocked vbios.

 

120HZ-3D machine here ... No optimus praise the lord

 

/shrug. Good luck finding problem, its not blocked by bios or drivers. Maybe tech inferno has a bios mod you can try.

 

I knew it !!! After flashing my old card bios I can OC now They were really OCblocking me !!!

 

If any GPU needs a clock block it's that one.

It honestly wouldn't surprise me if all this clock block nonsense is BECAUSE of how many failures and returns that card had.

Sent from my Nexus 6 using Tapatalk

 

I totally agree ... But you shouldn't take a feature away from someone without their consent ... Also for some reason the new GPU was always 100-150 3dmark11 below the previous one on stock clocks ... After flashing the new bios it's back in the same league but dunno why though ... Identical temperatures nonetheless

 

Its likely any new mobile card leaving nVidia is locked down now. Doesn't make sense to cripple just one product line. Surprises me that they are still making 880M cards though. They must have one heck of a surplus of low voltage, high leakage Kepler cores!

Sent from my Nexus 6 using Tapatalk

 

If Inspector works then you don't have a clockblocked vBIOS

 

If that happened, both asus and msi can be in a lawsuit for false advertizing, then nvidia may well be in for another lawsuit as a byproduct of the aforementioned lawsuit.
I dont think nvidia wants to rock the boat with yet another lawsuit so the current msi and asus lineup should be all find and dandy.

 

Is that actually true? Almost hard to believe that.

Also, when overclocking, what is the biggest thing to lookout for? I have never tried it but am thinking of doing so eventually (I am assuming heat, but not 100% sure). In addition, if my laptop was made in November of 2014, does that mean it will have the locked BIOS?

 

http://www.panelook.com/modeldetail.php?id=19049 This is the IPS panel people are hungry about in those 15" models.

You won't have the locked vBIOS.

 

I've been look breifly at ASUS's ROG forum (so messy!) and MSI forums. Of course, the topic of Nvidia's 347.xx drivers were brought up, but I found something interesting.

So ASUS's overclock is on a vBIOS level, so it very unlikely that they are unaffected.
Haven't found anything on MSI's forums, but I only skimmed through it.

 

Likely means they watched the vBIOS and saw clockblocking and denied that function in the vBIOS, as nVidia said they only give guidelines to manufacturers in the past.

That being said, clockblocked vBIOS == clockblocked ROG tweak, so...

 

http://forum.techinferno.com/genera...t-overclocking-mobile-gpus-22.html#post125430

 

all hail modest/humble prema! vbios clockblock successfully removed

http://forum.techinferno.com/clevo/8296-gtx-970-980m-overclocking-thread-11.html#post125592

Sent from my Nexus 5 using Tapatalk

 

Since the original driver clockblock thread is now closed, please continue discussion of clockblocked drivers and vBIOS in here

 

Does anyone have any idea when the next nvidia driver will be released. You know the one that's supposed to allow overclocking. Or is it just a guessing game.

 

As far as i know i believe the nvidia rep stated sometime in march.

 

there already are overclockable drivers, such as the 345.20 and also a leaked 349.xx driver on laptopvideo2go.com

 

This is great news.

 

vBIOS clockblock to be removed

 

well well, interesting! seems like were making good progress on that front

 

Good good, we'll have to keep an eye out if for example A17 R2 with 970M start shipping with unlocked vBIOS this month after the next NVidia driver update this month. If that's the case then we've won our arguments for the consumer! And well done notebookcheck.net for taking up the reigns with NVidia & the different notebook manufacturers to get to the bottom of it!

 

What about those who already have the vbios blocked cards. With they have to flash a new vbios or will the new nvidia drivers fix it without flashing.

 

Per @Prema , if the new Nvidia drivers ignore the OC lock bit in the vBIOS, then you won't have to flash. Otherwise, you do.

If you're lucky, maybe your OEM (AW, Clevo, etc.) will make the unblocked vBIOS available. Otherwise, you could get it from someone else who bought a laptop after the vBIOS block was removed