Clevo Overclocker's Lounge

I PM'ed you

KMS tools is a part of microsoft OS management, it's not a "crack"

 

You should be able to get a very good contact pattern by adjusting the pads as I have advised.

 

NVIDIA stand-alone PhysX System Software 9.17.0524 posted on MajorGeek.com
Physics Software is included in NVIDIA drivers. But I can't see it in Released Notes for latest Nvidia WHQL driver 382.53 Feel free check this out Read more from Ngreedia

Edit. 3DMark v2.3.3732 + Time Spy @Mr. Fox +++ What will this bring?

Spoiler: Change Log

From Guru3d.com: Fixed in 3DMark 2.3.3732

• 
  
  
  • Fixed an issue that could cause the API Overhead feature test to fail to show a score at the end of an otherwise normal run on some systems.
  • Fixed Time Spy test to properly recover from a corrupted shader cache - if runtime compiled shaders are found to be corrupted, they are deleted and recompiled. Uninstallation also now completely removes the shader cache folder.
  • Fixed a scaling issue that could cause parts of the UI to end up outside the display area on 1080p monitors with 150% DPI scaling. UI will now scale appropriately even on high DPI scaling settings.

From Futuremark:
What's new in 3DMark 2.3.3732
This is a minor update. Benchmark scores are not affected.

Fixed

• 
  
  
  • Various minor bug fixes to improve compatibility and stability.
  • Fixed an issue with the shader cache that could, in rare cases, cause a crash.
  • Systems with Ryzen processor will now show the current clock speed on the front page and on the result page instead of '-1'. Other processor models will continue to show the listed stock clock speed.

 

I'll go watch the 2001 SE demo and remember simpler times.

 

When results still scaled with clocks, voltage and power draw...

 

When a combined test actually utilized your system? ?

 

http://www.3dmark.com/fs/12895362


http://www.3dmark.com/fs/12895506


http://www.3dmark.com/fs/12895440

 

NOICE!!!!

Ah well there goes my nights sleep.

 

I ran those with the RAM at 3000. I'll have to try them again later with weekend at 3200 and 3333 to see if they improve even more.

Still working on trying to match or beat your Time Spy score.

 

So I can run 3000MHz RAM in my P870DM-G (though since the upgrade to the 1070 it's not longer a "-G" perse".)

 

Yes you can. I had G.SKILL 3000 in the P870DM-G that I sold to @Rage Set. I hope that @Mr. Spock is able to work out the purchase of that machine. Would be nice to keep that beast in the NBR family and have a good home to live in.

 

Yeah i haven't been able to boot at 3200Mhz yet, no matter what i have tried. But i can run 3000Mhz @ 1.2V easily.
Now i gotta try and match this awesome score of yours
Was planning on installing Windows 7 tonight. Wish i had the @Prema vBIOS.

 

newegg has a full 64gb 3000MHz kit for nearly $550 USD.....that's a lot even after conversion

 

I had to order a new set of screws for my vapor chamber. It has been taken off and on so many times that the screws heads are wallowed out and hard to hold a screwdriver and the legs on the vBIOS chips are so slippery/polished they are having a hard time holding the alligator clip now. It sometimes pops off two or three times before I can finish a flash. Pascal firmware is a real stinker. I shudder to think what kind of firmware atrocities the NVIDIOTS might commit next to screw things up for us. Falcon sucks... big time... and, it really blows having no alternative to flashing other than a hardware programmer. The team responsible for deploying Falcon needs to be driven off a 200 foot cliff in a burning school bus. It wouldn't be such a big deal if modding the vBIOS were not so ridiculously complex hit-or-miss trial and error nonsense.

While it is very powerful, I am no longer of the opinion that any Pascal GPU qualifies as an enthusiast product. It's too enthusiast unfriendly (mod resistant) to be regarded as awesome. Enthusiasts don't do cookie-cutter/belly-button gamer crap, but based on the extent to which they have gone to interfere with vBIOS mods, that's apparently the category they want their products to fall into now.

Yeah, that's absolutely absurd. I'd pass on that. Way too expensive for too little benefit to be gained in terms of performance... very poor bang for buck at that asinine price.

 

i'll stick with the 2400MHz for now. the extra is just e-peen really.

 

Oh, it does make a measurable difference in benchmark scores, but certainly not enough difference to justify that crazy price. For playing games it's even less worth it. 2400 is plenty good enough for gaming, and it's more stable as well. Running 3000 or higher RAM is also hit or miss in terms of stability from one machine to the next. Some run it really well, while another otherwise identical machine cannot even POST, or has constant BSOD issues, etc.

BTW - those scores are also the newest version of 3DMark that Brother @Papusan mentioned a few days ago.

 

Oh same on the vBIOS chips. Though im using these clips, that were suggested by @Johnksss@iBUYPOWER -

https://www.amazon.com/gp/product/B00HHH65T4/ref=oh_aui_detailpage_o02_s00?ie=UTF8&psc=1

They hold on to the chip nice and snug, i dont even remember how many i have flashed the chip now.

Ahh gotcha. Ill update as well as see if i can improve on mine.

 

This is probably a foolish question but WHY would Nvidia lock the GPUs up to a point that they are difficult to play around with? What's the point? Why not say in the warranty, "if you mess with the firmware you void the warranty" or something like that? I don't understand the point of purposely locking a card up and making it difficult for guys that know what they're doing, @Prema, @Mr. Fox , @Johnksss@iBUYPOWER @bloodhawk to break into these cards. If that's even the correct terminology.

 

I could understand them locking up a Quadro or one of those Tesla cards that they charge a Gillion dollars for but why the mobile GTX?

 

Because they are Nazi control freaks that believe they are entitled to enjoy the liberty of shafting us if they stand to benefit from the crap they do to us on purpose.

I actually think the opposite. The fact that they are so expensive warrants end users having the ability to do these things even more. The mobile 1080 is insanely overpriced, just as the workstation GPUs are. And, it's not because they cost that much more to make. It is because people are stupid enough to let them get away with it by paying absurd prices. When you can buy two desktop 1080 for the price of one mobile 1080 it's pretty obvious which party to the transaction is getting screwed.

But, regardless of all that the interference with firmware mods is an unpardonable sin under any circumstances. When you sell a car or a house you give the keys to the new owner. We deserve the same here. They have no right to control anything they sell after the sale.

People that want crippled inexpensive trash GPUs that don't overclock worth a damn can always turn to AMD to save a few bucks.

 

Partly due to the upsell of a higher model instead and party due to all the users bricking their systems and returning them anyway.

 

With higher performance on current cards(Oc'd) vs. Next release... Less sales of Volta

 

Ok guys.
I'll donate $50 to @Prema (or the other person who has contacts with Nvidia, I forgot his name already) if someone wants to make a bypass for current Nvflash. Sorry for getting off topic again. I apologize.
I really can't deal with HW flashing right now, and I don't have a 1.8v adapter. I guess I sound like a broken record at this point, sorry

 

Unfortunately that is the only way at the moment.

 

Yeah, it sucks. Real bad. But, that's the way the cookie crumbles, tough tiddie said the kitty, pip pip, cheerio, and all that rot.

 

I bet you can overclock the RAM you have to at least 2800 without spending a dime. What do you have for 2400? My Kingston HyperX Impact 2400 handles 3000 stable. I even programmed it to XMP and it works as good or better than the G.SKILL does. The only reason I am not using it is because it is only 16GB (4*4GB).

 

What do you need more than 16GB for?

 

In @thegh0sts sig... G.Skill Ripjaws 64GB (4 x 16GB) DDR4 2400MHz RAM

 

Happy Father's Day everyone.

Having more than 16GB seems to increase some benchmark scores even if the RAM is not faster. 3DMark 11 especially. I don't know why, but it seems to be the case. Otherwise, I don't do anything that consumes more than 16GB. I used to run PrimoCache and allocate half the RAM to the cache and it was very useful then.

LOL, didn't even notice. I bet he can just set the multiplier and ratio to 100*14 without touching the other timings and run 2800.

 

Crying in computer crashed some programs because I didn't have enough RAM before...

 

Haven't even launched a game yet

 

Now I'm curious, what do you usually have open?

 

Steam, about 120 chrome tabs, some other software depending on what I'm doing at the time, teamspeak, etc.

 

I asked what he needed it for as a bencher I know he is not going to have loads open.

Also close some tabs

 

I used to only have 8GB at one point...LOL

 

I have less RAM typically loaded, especially when not working on something. ~8GB. But I reach 32 GB and need more during work...

 

Oh ok

The sad part? I do. I do close the ones I don't use on a regular >_>

 

Sounds like me working! 16GB isn't even enough for my research, but 32GB isn't a problem at all. I haven't really gone above 22GB working, but I also don't leave that open while gaming. Soon, though, that will change...

Sent from my SM-G900P using Tapatalk

 

Benching on top of a bag of frozen gel(for food delivery, lol) wrapped by foil under the U3. Sucking icy air in and got some nice summer scores .

 

Organised favourites my friend unless you are on a metered connection?

 

Nope, just too many things to consistently open and close, I just know where my quick tabs are. I do use organized favourites for some things though. I like speed xD

Sent from my OnePlus 1 using a coconut

 

It's funny how people's brains like to organise things differently

Just yours costs more ram which I guess with 32GB is a price you can pay ^-^

 

Finally beat my last best score in the Time Spy abortion joke benchmark.

http://www.3dmark.com/spy/1937439

 

^ *spies "default clock 1771Mhz" and wonders if NVFlash will ever work*

 

The clocks are the easy part, there are much more problematic issues at hand that are much worse than flashing through nvflash. Which i highly doubt will ever happen or anytime soon if it does.

 

Ah yeah I know, I just like anything that sounds like "work better"

 

What is that unholy pair of words that you are using.

 

It's not even accurate. GPU-Z does not display correct information because of the vBIOS mod. The default boost clock is also wrong. It's not 1823. That benchmark was run with the core at 2114 and most of the time it stays boosted over 2100, but if you look in my screen shot GPU-Z is saying the boost clock is overclocked 1958 instead of 2114.

While GPU-Z is showing 1823 for "stock" [mod default] boost (with no offsets applied) here is what it is actually doing (without touching anything) as long as the temps remain chilly.

 

oh my, is that no perfcap reason I spy? Like how Kepler worked! Like how GPUs should work!

 

Yes, that no perfcap reason is nice. It may not seem like anything much is happening, but Brother @Prema has been busting his butt to try to make Pascal great again.

To answer your other question: NVFLASH will likely never work again for a modded vBIOS (unless a hack is developed to spoof signatures or disable firmware signing) *and* NVIDIA will likely never stop being stupid. In fact, I think " we ain't seen nothing yet" with respect to how stupid they will become. I suspect things will worsen beyond our worst nightmares as they continue down the path of incurable idiocy.

Remember, these are the worthless, feckless, testicle-free imbeciles bringing us worthless garbage like Max-Q feces.