*** Official Clevo X170SM-G/Sager NP9670M Owner's Lounge ***

Sure? I'm 100% sure you won't get real stock 9900K results from your Turd

 

With an overclock it will get damn close that's for sure

 

Please post your CBR-15 results and with what clock speed. You said 9900K speed. This means 4.7GHz

 

That is not a realistic R15 run. Most 9900K processors even in a desktop are like sub 2,000 in R15 or right at 2,000.

Nothing wrong with the thin and sleek turd
books beating our big heavy fat books.

Even the locked down 9880H will outperform your 5.3Ghz 8700K, and my 5Ghz 8086K in multithreaded. And my system has all of this time and work put in to it to get my current performance. That i9 9880H runs like 1,725 R15 straight out of the box and it walks right past me without even breaking a sweat. No fancy OS tweaks either.

The 9980HK can match a stock 9900K for sure. Especially in the Aorus laptops, My 7820HK was beating a 7700K easy.

I know BGA kinda sucks and all. But, if it can sustain performance without throttling and the machine is light and thin and does what you want it to, then who really cares I guess. especially if it is even 85-90% comparable to a 9900K in a desktop. These thin laptops are very useable, and actually portable.

 

It certainly can. My old Aorus 7820HK laptop could outperform a stock 7700K. So, the same probably goes for this too.

Even if it can almost match it, that is pretty good for such a thin machine. Liquid metal, proper thermal pads, undervolt, and you are practically there.

 

At 5.3Ghz a 87xxK scores roughly the same as the mobile 8 core 9980HK at stock:

Spoiler

[\SPOILER]

This guy ran the 9980HK at 5.1Ghz on Liquid Nitrogen and only scored 1931:

https://hwbot.org/submission/4215697_tagg_cinebench___r15_core_i9_9980hk_1931_cb

Would actually be interesting to see what it would do a little more "optimized".

 

Wrong.

Wrong.

Same silicon for BGA and LGA chips. We talk about apples vs. apples. You more like apples vs, Oranges. 33% more cores will normally beat it at same clock speed. Just look at AMD's chips.

Nope.

Nope. Wake up from your bad and sweaty sweet dream. Hit up my old scores from 7700K.

Have seen higher results. But do the same on an Clevo board with 9900K.... And the differences will be even higher.

 

LOL, be nice Papu!

 

So a thin and light 9880H can match a 8700K at 5.3Ghz? This is impressive. especially for a thin laptop.

 

All i’m saying is 8 cores at a low frequency in a thin and light machine matching a screaming overclocked optimized 6 core is really excellent.

I am 100% certain you can match a 7700K performance with a 7820HK. Not a overclocked 7700K, just stock vs stock. Thin and lights will always lose to a bigger laptop with overclocked desktop cpu.

I cannot break 1,700 in R15 and maintain this daily with my 8086K inside of my laptop. I can run around 1,650 R15 and sustain this performance daily. The 9880H is 8 cores yes, but it still outperforms my system regardless. And it is much more portable.

That’s all I meant by this.

 

Not the 9880H, but the 9980HK.

The 9880H maxes out just below an 87xxK at 5.3Ghz
The 9980HK can top it, but the currently best result is only 1931 and that on LN...so we will have to see if it can reach 9900K/KS stock scores...

 

Yea the 9980HK is seriously costly. The 9880H is decent though still at a little over 1,600 in R15. And if they would maintain this run to run is questionable.
Probably a 1 hit cold run.

All of these BGA books would still suffer in gaming because of the slow all core boost compared to a 5Ghz 8086K anyways.

it is just interesting to see such a thin machine beat all of my hard work that I thought was impressive so easily.

 

I’m always nice, brother But I don't like wrong info posted on the web

OC'd or stock doesn't matter. Just ask @Falkentyne who own one of the best perfoming 7820HK Turds. If you trow out such claims, please post your results.

 

From what I can tell, Intel has also been using lower binned silicon for mobile BGA chips than for desktop LGA products since at least Haswell. Mobile needs more voltage and doesn't undervolt as well, and also performs a few bins lower, than desktop at the same clockspeed.

 

Intel turbo quoted speeds have been for single core speeds basically since its introduction.

 

I was overclocking with @Falkentyne. We both had 7820HK’s at the same time. He was still beating me though.

Best I got with a 7820HK was around 1,010 in R15 multi thread. And 195 single thread. He was getting up there in the higher sub 1,100 range though. I think He got 1,070? or 1,059?

I would run 4.7Ghz daily. Bench at 4.8Ghz and game at 4.7-4.8 usually. But, he was pushing 4.9Ghz in pubg stable.

 

You need also take in acoount that the power consumption rises with temperature (because of I2R losses). Higher wattage/power consumption = More heat. And more heat means you need to add even more voltage for your highest BGA overclock. A never ending evil circle if you are at border in temperature land with thin and flimsy and max oc'd. Which can only be solved with better cooling/or cooling. Still won't help.

You have always been able to put all core boost same as single core on the mainstream chips due all cores have to reach max single boost. Last example is the 9900KS who in short is a binned 9900K. I can see this now will change with 10th gen chips due implementing boost 3.0. Aka same clock behaviour as for the Core X-series processors.

Edit.
Intel® Turbo Boost Max Technology 3.0

1) You said 4.7GHz daily and 4.8GHz as bench clocks.
2) Your best score is 1.010

See following CBR-15 scores... 47 = 1064 and 48 = 1085 with 7700K in my old Clevo. I could even beat your highest CBR-15 bench score with wimpy 45x. Who is 3 bin below.

Each bin for 7th gen is around 22cb. Then do your math.

No BGA chips will come on par/close the desktop chips. Have already tested and confirmed this the last years with my different 6th, 7th, 8th and 9th gen chips. What you have posted is wrong!

 

If someone could donate me an Aorus 17-ya I’d gladly get some results and try and challenge the best Clevo overclocker @Papusan @Donald@HIDevolution

 

The new Aorus = Same Pig but with New lipstick.

The old http://forum.notebookreview.com/thr...-w-i9-8950hk-to-the-max.820711/#post-10773495

The thread is about Clevo X170SM-G, so we let it be with this

 

Oh my goodness lol. I can tell BGA is upsetting you.

I never said a 7820HK could be as fast as a desktop 7700K. I simply said, a 7820HK
can perform just as well as a stock 7700K. Just like I said a 9980HK can perform nearly just as well as a stock 9900K.

When I had my 7820HK. I ran 4.2Ghz all core, and I achieved stock 7700K performance.

 

I suppose that's OK if you're content with stock 7700K performance. Otherwise, it's just kind of a lame excuse for 7820HK being a piece of crap CPU compared to 7700K. The reason for this is ODMs framing everything in a manner to make it seem like it should somehow be OK or expected to be inferior based on the half-assed notion that everyone should naturally expect less from a laptop just because it's a laptop. And, this is why 99.9% of the laptops made today totally suck.

 

Well, mine was faster than a stock 7700K. My 7820HK ran just as well as a 4.7Ghz 7700K does. Or as well as Papusans 4.5Ghz 7700K did.

But when I did run 4.2 it was practically a 7700K.

All i’m saying is, what’s wrong with having 90% of something if it offers huge advantage?

 

Your 1085 vs my 1010 is only like 7% faster though. isn’t this very marginal? if a machine weight half as much and half as thick?

Does the term

“Never even come close”

does that even really come in to play here?

7% sounds very very marginal to me.

 

You got 1010cb running 4.8GHz then say achieved stock 7700K performance...

Some numbers... All done with 2133 sticks clocked to 2666MHz.

Core i7-7700K - Intel - WikiChip
You see? Your 4.8GHz maxed out oc can barely come above stock [email protected] running minimum all core boost. We don't talk about the Processor's Base Frequency.

7.4% would be the same as if you got a chips that could do 5.2GHz, instead overclocked near 5.6GHz if you put it in perspective. And yees, I know that not all bother with maximum performance and rather prefer looking for design... This is one of the main reasons we now see more and more Apple lookalike Notebooks being thrown out in the market. And almost all is equal crippled only with the different brands own design touch.

 

You can't. It's impossible.
No one knows why. It was always 1 speed bin below a K LGA chip. That was with cache -3 lower than core.
Maybe it was some RAM timing that the clevos have that the BGA stuff didn't have. But I don't know.
The 6 core processors are even worse. Two effective bins below an 8700K when both are at the same speed.

See?

 

Sad to see them dump their lower binned ASIC chips into mobile systems.
Seeing the same trend with NVIDIA's Max-Q chips requiring a lot more voltage at higher clocks than their Max-P counterparts.

From a technical perspective doing the opposite would be the logical decision. Surely not the engineers making the decisions here...

Edit: Max-Q is actually Low-Q going by ASIC quality...LOL

 

it sucks and to make it worse, average consumers further drive companies to do what they do, milk!

you ask them what specs of their PC, they simply tell you: it's got i7, it's good.

 

All core turbo being set to max turbo speed is a motherboard manufacturer overclocking for you, it is not supported by intel directly.

 

Sounds like the're milking the last possible penny.

 

Well, they can't tell us what model of i7 because that would spoil the surprise, LOL.

But, yeah... on a more serious note, it is super-retarded. You go to an "official" web site and all it says is "i7" CPU. This is a very bad reflection upon the character and integrity (lack thereof) of people that sell them, and a really tragic demonstration of stupidity for the people that don't have a flipping clue what they are buying.

 

I know about MCE and that Intel advertise max boost for 1 core for most of their processors. Only the 9900KS and the "special version the 14 cores i9-9990XE who Intel hide in their data base" will automatically enter all core boost without use of MCE (Multi-Core Enhancement).

I talked more about implementing Boost Max Technology 3.0 for coming mainstream whitch identifies the best performing core(s) on a processor. All other former unlocked mainstream chips can reach all core boost due they didn't bin chips the way as AMD. Now we will see this change with the 10th gen mainstream who will get Bost 3.0 Tech.

 

I doubt they will try with low 5ghz+ listed clocks, the TDPs at that point will be crazy.

 

I understand now. I always thought performance was the same clock vs. clock. That is amazing that it is not, and I was very misinformed.

 

If they’re selling these $3,000+ dollar ultra thin Razer/Zephyrus laptops. They should really put the best silicon in them. They certainly need the lowest voltage possible to create the least amount of heat to make up for the total lack in cooling capabilities.

I remember when we use to see ASIC quality listed in GPu-Z. I had purchased a RX480 on launch date, and it was 99% ASIC quality. I felt special. I really did lol. It managed to outperform or match GTX980’s which was really good at the time. “With proper tweaking” ^

 

Good old days

 

I tested again.
9900k @ 4.5 ghz, 4c/8t, cache 42, 4 cores disabled, Realtime priority scored 1023 CB.
7820HK @ 4.5 ghz, 4c/8t, cache 42, realtime priority scored 1008 CB.

What was strange was the 7820HK L3 cache bandwidth was slightly faster and 1ns lower (10.7ns vs 11.something ns) than the 9900k, and L1 and L2 latency seemed to be the same as the 9900k, but still scored lower. Something is castrated on the chip.

 

Looks like I am late to the turdbook bashing party, may I please add something that counters the whole premise of the Aorus being an alternative to the X170 because it is thin and/or light?

Because it turns out it isn't: It is 38 mm thick and weighs 3.8 kgs!
https://www.notebookcheck.net/Aorus...treme-performance-extreme-noise.445964.0.html

I bet that this is within 2 or 3 mm and 500g of the X170 - not very impressive for an ultra-noisy turdbook.

So what was the reason again to even consider it?

 

Is there any news on when the X170 may release? Or any guesses? Or, 'guesses', if you know what I mean?

 

Q2 of 2020 for sure

 

Let's hope that more information gets out soon, I think that especially these would be good to know:

weight
dimensions
screen options
graphics card options

And for workstation use it would be interesting to know if there will be a version that will take ECC Ram, Xeon processors and Quadro cards.
For DTR use it would also be good to know how many drive slots it has, most other top of the line DTR's seem to have 4.

 

Caches get slower as they get larger.

 

Could also be due to security patches?

 

Size will be more important.

 

3d stack it.

 

It's more about handing the addresses and finding the information in a larger pool rather than trace lengths and geometry.

 

I don't think we will see Intel HEDT on Clevo this coming gen, too much of a niche in an already niche market. Not to mention that a board like that will need to be huge in order to fit the power delivery for such chips.

 

I agree, it's probably not going to happen but still it would be nice to hear from Clevo what it can and cannot do. No use waiting for something that won't be part of the package anyway.

 

On top. Why would they spend a lot resources on soon EOL platform? Only morons would do that!

 

The sheer size of the socket is an issue, certainly on the AMD side. Little point when the mainstream socket offers 16 cores too.

 

Agreed, not a good return on investment. Intel has been dragging their feet so long that 10nm CPU's have had an "almost there" status since at least 2017 if not earlier. A good writeup from early 2018 can be found here:
https://www.anandtech.com/show/12436/intel-10nm-dualcore-cannon-lake