(Part 2) If U Can Solve This U R King

Correct scores vs. clocks have nothing to do with the Silicon Lottery!! 100% of all processors manage to run default clocks perfectly well. Same clocks = same scores. Only bloat can make a difference. Or crippled firmware.

 

That's not how it works

 

Or in the case of OP, single channel RAM.

 

Is it safe to overclock the memory of my GTX1070 by 200MHz?

 

You can test for instability to be sure, but most likely that'll be fine. In fact, use GPU-Z or MSI Afterburner and check to see if the Dragon Center's Sport Mode already overclocks it by that much. Sport Mode used to auto-overlock my GTX 1060 memory by 200 MHz. I've since uninstalled the MSI Dragon Center and now do my overclocking manually through MSI Afterburner, and I control the fans through MSI Silent Option. I recommend you do the same. Using Afterburner (or another utility) to manually overclock may conflict with the Dragon Center's settings and cause the graphics driver to crash. It did for me.

 

I have +600 mhz overclock on my RAM (9200 mhz).
Most desktop users with samsung memory had no problems going at least +500 mhz.

 

Good. 5GHz monster CPU.

 

I hope i don't get banned again..but its behaving like a pascal card. Also known as normal. And take notebookcheck.net with a lot of salt.

So i guess we are NBRs kings

 

Well sorry to double post but have you tried resetting your cmos battery and re plugging your battery...im pretty sure thats the last thing you can try

 

He doesn't accept that answer, so yeah, gl with that :'D

 

Look at the performance difference between single and dual channel RAM in the PUBG lobby at the beginning of this video:

 

Massive FPS difference.

 

Nor accepts the fact that he has to adjust ingame settings to make sure nothing is impacting the frame rate, presets are never an option for me.

 

Can I ask you something?
This actually goes for all of you who say this.

What exacly makes you think he runs single channel ram?

I have yet to find a single notebook that has 16GB 1 stick instead of 2x8GB

Yeah no, this is absolutely not right. He has a GTX 1070, there is absolutely no reason for him to run battlefield 3 anything but maxed out. I run the game on 165 -180 FPS on maxed out on a "crappy" GTX 1060. If we tak about new AAA games, yes, I agree. There are games where ultra and high, hell even medium settings won't look any different, but games from 2011 should NOT be a struggle for a GTX 1070

 

If you look at first post again he posted the bios details where there is only one stick showing up as 16gb hence why i told him a few pages back. My GT62VR shows two 8gb sticks into the BIOS so i have no idea why some models have it and some don't.

EDIT:

Ingame settings may need to be adjusted for multiplayer because it impacts more the CPU than the GPU and with a 6700HQ the single core perfomance is weak, you can see one of his cores getting used up to 70% and higher. Even reducing Anti-aliasing can help on multiplayer so why not give it a try? I do understand he got a 1070 but that doesn't mean you can run any game at ultra preset without having to adjust at least one setting if maximum fps is your goal.

 

Thanks, didn't see the picture, so this is the first time I've seen a notebook with a single 16GB RAM slot, interesting. However stop making the assertion that this is caused by his RAM, by that logic he has a miracle notebook, because he ran games normally on normal performance before his graphicscard decided to say "nono,powersave, yesyes".

Also I really don't know if an early access game that is known to run like absolute crap is the best example to showcase difference in dual and single channel ram.

To illustrate even further:
http://www.hardwaresecrets.com/does-dual-channel-memory-make-difference-in-gaming-performance/3/

battlefield 3 and 4 really don't care about single or dual channel. I assume that open world games like Witcher and tomb raider are more affected of this, shooters, not so much.

EDIT:

If you have battlefield 3, why not try it out yoruself? See where the performance is on your notebook, if you have better performance than OP, then you know something is wrong.

I'll do a test on a 6700HQ + 980M OC notebook.

 

What would be nice to see is a perfomance test on BF1, which is yet another intensive game.
And if we really want to pursue with the theory of the CPU or GPU underperforming you have to test more games and then compare it to people with your same specs, benchmarks bring you nowhere because they're different for everyone.

 

How does this make sense? He clearly had better performance when he first bought it, it takes no scientist to figure out that something is throttling. No need to test other games (which he did, ryse ran like crap as well).

 

I will try to get it but i can assure you what i said isn't totally wrong, you can test it in, almost, every FPS multiplayer game.

 

The only thing throttling there is the GPU at 86 degrees it may be limiting also the memory clock, i gotta check that video again. The CPU is running at full speed all the time as the GT62 chassis can keep it around 70°C without throttling but the GPU is way too hot, so we bounce back at MSI afterburner and the need to make a custom voltage curve to reduce temps

 

You see, this is the funny part, there is no real way of overclocking pascal cards. The overclock is actually an undervolt, undervolt = higher frequencies. That's how pascal works, and why pascal is so garbage. Do yourself a favor, and open MSI afterburner, then press ctrl + F, then rise the clockspeed, look at the curve and what happens, all the "overclock" does is undervolt the card. Pascal sure has its own way to be a pain in the ass.

 

Regarding the single vs dual channel ram debate, even if dual channel would make a considerable difference, the improvement should be on top of his previous previous FPS range (according to the OP, relatively constant 150 FPS in BF3, when the laptop was working as expected with the same single channel ram).

Adding another 16Gb ram stick just to get (if this game benefits from dual channel) to his previous performance its not what I would call a fix.

Regarding the underperforming CPU as @Papusan pointed it out in the CB15 bench, yes I agree, its bellow what most users or reviewers would get in CB15, but the slightly lower CB15 score shouldn't translate in a such a big performance impact (under 100 FPS in BF3), otherwise we would constantly see considerable performance gaps (in many games that are heavy on CPU) between 6700HQ and 7700HQ (where the difference in CB15 scores between stock laptops with 6700HQ and 7700HQ is at least the same or bigger).

As you can see in the previous videos, the CPU maintains its rated multicore clocks without throttling (throttling isn't the only performance reducing mechanism, there are many examples with considerable performance differences in benchmarks between the same CPUs at the same clocks, a common denominator seems to be the voltage to clock ratio for a specific load).

Regarding the GPU, we know for sure that voltage to clock ratios can affect the performance in different ways depending on what triggers the performance reducing feature (we have the power limit, throttling and the "pascal problem" where the clocks can remain very high without fluctuations but the performance is lower if the voltage is too low for that clock).

Here is a video showing the "pascal problem" (the OP's GPU throttles, so this doesn't apply in his case, but it could still be a power related problem):



Here's another video for performance comparisons (yes, I know its 7700HQ, but still...), you can see that his gtx 1070 also reaches 85 degrees C:

 

The pascal problem is not that it runs on high clocks but lowers the performance, the pascal problem is that despite good clocks and good values, the card will underperform it's performance for no apparent reason due to it's power management chip. The problem pascal faces is the way NVIDIA designed this ****ty card so that "overclock" is actually only an undervolt trying to make the card run more stable on higher clocks. The issue with pascal is that it somewhere in it's crappy crypted vBios has set limits which the card cannot overcome, which are temperature, voltage, max TDP etc. related, making it impossible to figure out why these garbage cards refuse to run properly.

 

Looking forward to your BF3 results on the i7-6700HQ and 980M, Danish It is really disappointing you run BF3 at 165fps-180fps on your GTX1060 (in singleplayer) while I cant surpass 135fps on my GTX1070 in multiplayer. Could it be because of my i7-6700HQ vs your i7-4890k? it is unbelievable.

Can you please try it again in multiplayer? Multi gives less fps than singleplayer usually. i get 180fps max in singleplayer but only 80-135fps in multiplayer. Please Danish, so we can have fair comparison. Try metro map. Thanks

UPDATE to readers: I am sending my laptop back to MSI Ontario on Monday so they would work on the high cpu and gpu temp again after their last failed attempt in November. They kindly offered me a free RMA this time. 1st time i had to pay $170 in shipping costs. Will report back on the new temps after they service it.

 

6700HQ / GTX 980M
All maxed out 1080P

 

and how much was it on this map on the i7-4890k and gtx1060?

 

Dno, I can check, Can you explain to me how I can find a server with people on it? because I tried multiplayer and all i get are empty servers.

 

when the browser page opens up with a list of servers, u can use the sort button on the top next to Players. arrange them by highest to lowest. theres a server tjats always plays metro map map statts with the word 'xtreme'. thats where i test my fps.

 

Danish, just copy paste this server name in server search field and u r good to go.

XtremeNoobs 24/7 Metro 1200 Tic -64P-

 

Danish, in order to reach sound conclusions, I am afraid only short videos will help and not screenshots because fps fluctuates a lot in bf3, for everybody.

 

make sure the server filters above are all deselected. no filters. also try only the first words:

XtremeNoobs

 

https://www.gametracker.com/server_info/216.52.148.236:47200/

this server currently have 64 players online

 

Found it but it's full. I think i got the hang of the filter crap

 

plz stay in queue u will join shortly dont lose patience. it usually doesnt take more than 2 or 3 mins max.

FINALLY someone with a i76700HQ and BF3 multiplayer wow

 

Ok, I'll have to postbone this for a bit, my patience with the game ran out after i recieved for the 5th time

Gonna record when I'm back home after i finished some business.

 

usually ending the process via task manager and restarting the game and origin fixes this.

Thanks a lot for not giving up.

If he comes back and plays bf3 at more than 135fps steadily inside the metro and more than 120fps steadily outside the metro station, that means something is srsly wrong with my system.

 

Made it on the 11th try, wanted to record but getting geforce experience to accept the game seems next to impossible.

Spoiler

FPS ingame: 130 - 145. Far from unstable and flucuating to be honest. CS:GO is far worse when it comes to FPS range, so idk what u're talking about.

Didn't manage to go outside because I kept dying without knowing what killed me :'D

Overall idk, the game seems to be a bad benchmark, the game is full of annoying issues, do you have something else that doesn't require you to literally taskmanager end it and something that doesn't cause geforce experience to hate it's guts? that would be great.

 

I cant achieve those framerates in metro map with my i76700HQ and GTX1070. for me: 110-130fps inside metro station and 80-110fps outside.

Why?????

 

I'm asking again do you have other games? I mean having expriences BF3 myself now, that game seems to be a massive mess. I'd rather compare on something way more stable such as battlefield 1, witcher 3, DOOM etc.

 

witcher 3 i have. but it is gpu bound and i get average scores for my laptop. i dont see the cpu limitation in witcher 3. i get 55fps-70fps with hairworks ON ultra settings. 60-90fps hairworks off.

Spoiler: 1st scene of Witcher 3 new game

First new game scene in Witcher 3:

can u restart ur laptop and show us how much ur i76700HQ scores in cinebench r15? i need to understand why u and everybody else gets higher fps than me in bf3 with their i76700HQ.

 

more importantly how do i fix this situation? i cant just ask msi to replace the cpu for me (whole motherboard) because they dont see it as a real problem. i wish a cpu expert can analyze my xtu readings and tell me what's wrong.

 

thank u. see what i mean. 690 vs 655. 145fps in bf3 vs 130fps.

what should i tell msi? i deserve better performance.

 

not really, score is actually standard. Let me show you something interesting, I'll grab a drive with a clean install Windows 10.

 

i just need guidance on how to properly communicate this as a problem to msi to warrant a cpu or motherboard replacement. i think we have done enough experiments.

anyone?

 

Clean install windows 10 from SSD:


Again, cinebench is a crappy way of testing your CPU.

 

i run cinebench 10 times and i consistently get either 644 or 655 all the time. run it 10 times after a cold boot each time and see the average. Report the most recurring score, not just a one off random score. Restart and run it again and see.

 

or run BF3 right niw after this score see how it performs ln the clean win10 install.