I don’t have it but I wanted to buy 1060 model, but after I saw on Reddit, here or Razer Forum, 8750H has similar high temperatures anyway but GPU is always lower, and for 144 hz I need 1070 at least. My 1070 max q never show 80 ish, on benchmarks, CPU with appropriate undervolting value sits on low 80ish, almost similar with my old 7700HQ, only on nowaday games that eat cpu crazy like Far Cry 5, AC Origins, FFXV etc I’ve to hold CPU performance and lowering the setting not at ultra (max) since it’ll bump to 90ish on default, of course those are done with undervolting. So in both model you have to play the setting or maybe thermal repasting will help reducing the temperature if you’re doing that right, I’m not at that level atm, I’m playing AC Origins dlc this next week stressing the cpu![]()
-
I bought the 1060 version from bestbuy I also have the msi gs65 from bestbuy with 1070. I going to keep the razer way better build quality and its fast enough. As far as cooling they both are noisy. But that what happens when you put a i7 6core cpu in a thin chassy.
On the razer I am undervolting -150 on cpu and cache and I limit turbo to 3.7Ghz running fire strike I get 10600 temps stay around 76c on cpu and 74c on gpu stock thermal paste
The msi has to much flex on the screen.
PUBG gets 62fps in ultra on the Razer with 1060 fast enough for my use
Sent from my iPad using Tapatalkhmscott likes this. -
When stressing the CPU, where is everyone finding that the CPU PKG power limit is sitting when throttling? Like Lightwire, with optimal undervolting, I was holding 3.7ghz at decent temps stably for prolonged tests and throttling down to 45W. After a few days of this working fine, I'm now being throttle down to 35W and holding ~3.2GHz. Correct me if I am mistaken, but this CPU is designed to hold 45W indefinitely as long as temps stay under the set threshold. I have been seeing more and more people reporting not getting the power they anticipated, and this being tied to 35W throttling.
Razer support has agreed it should be holding 45W, and has claimed they are looking for a cause. I suspect it has to do with the EC firmware, but no way for me to test thisVistar Shook likes this. -
Vistar Shook, Petrov and George24 like this.
-
The problem is that Intel didnt rework the PCH and its still made on an older production process. So it heats up a bit more thansk to the increased CPU load.Eason likes this. -
Vistar Shook likes this.
-
Vistar Shook likes this.
-
I wonder how the reliability will be of these chipsets in the long run.hmscott likes this. -
-
-
Hello.
Got mine today - firestrike - 14k stock + synapse game mode = its ok?
Can any one please give screenshot for TS good settings for gaming withiut very high temps?Last edited: Jun 26, 2018 -
I reinstalled windows from their recovery (reset to factory) and it still Throttles at 35W. I also reinstalled from a completely fresh windows 10 install and it made no differences.
I've also modded the BIOS to enable full TDP control, no difference.
My current hypotheses are EC firmware or the power brick isn't supplying the correct power.
Last edited: Jun 26, 2018Vistar Shook, Metalocalypse, Arondel and 2 others like this. -
Oh about firestrike, it depends on gpu silicon lottery about OC value, some user can OC 230/700~, and get over 15,3k, mine only 200/600 got 15,1K overall.Attached Files:
-
-
-
Joikansai likes this.
-
-
-
Shouldn't the power not matter if the clock speeds are maintained? I'm imagining an engine running 6000 rpm, as long as the mixture is right, it doesn't matter how much more fuel pressure you have past the regulator.
If I'm maintaining my 3.8ghz, why would I need 45w behind that if 32w holds the clocks? -
Darkhan likes this.
-
-
When I game in majority of the games seem to hold close to 3.9ghz. For some reason xtu shows average clock speed after hour of gaming at like 2.9. But when I monitor it real time via afterburner it shows pegged at 3.9 pretty consistently.
-
If you manage to constantly keep the full 3.9Ghz TB clocks in gaming, then you don't have to worry about any kind of throttling.
Just to be sure, I recommend setting hwinfo to save a log file while testing a game (it's best to have an active software that displays FPS, such as afterbuner with rtss or fraps, so you can capture the FPS in the hwinfo log file) so you can properly analyze the behavior.
@stranula
If I remember correctly, Prema mentioned the 35w power throttling for some laptops, but only in combined loads (CPU+GPU). The "normal" power limit when stressing only the CPU, should be at 45w (with a few seconds up to 56w until the PL2 time limit expires).
I haven't seen this 35w power throttling for the blade in this thread (though, it's possible that I missed it).
For those that want to test for this kind of limitation, they can use aida64 with the following config:
1 start aida64, go to tools and select aida stability test,
2 in the stress test window click preferences, go to general an select 1 sec for the sensors refresh rate, so it matches the throttling refresh rate,
3 tick the following boxes: cpu+fpu+cache+memory+gpu (don't tick disks),
4 select the Unified tab to be the main active tab,
5 on the left side you have a button that can be dragged downwards, drag it until the grid values on the left show 0 to 100 instead of 0 to 200,
6 stretch the window in all directions for better viewing,
7 start the stress test and capture a pic just before the graph reaches the other side of the screen,
8 post the pic using this: https://postimages.org or the app https://postimages.org/app -
raz8020 likes this.
-
Hello, I'm wondering how long does it take for Razer to start implementing changes to the blade 15 and producing them to fix problems with it? Also where can I find out what changes they have made?
-
I just tested cinebench to see any performance er
George24 likes this. -
MobileTechReview
Published on Jun 20, 2018
It’s a 4-way 2018 thin and light high end 15” gaming laptop smackdown! The Razer Blade 15, MSI GS65 Stealth Thin, Gigabyte Aero 15X and Computer Upgrade King P955ER all with an 8th gen Intel Coffee Lake Core i7-8750H processors, NVIDIA GTX 1070 Max-Q dedicated graphics with Optimus, DDR4 RAM, NVMe SSD and 144 Hz displays are compared. All weigh 4.5 lb. / 2.0 kg or less, have metal casings and subdued looks that work for the office.
Our review of the 2018 Razer Blade 15: https://www.youtube.com/watch?v=sYOxy...
Our review of the MSI GS65 Stealth Thin: https://www.youtube.com/watch?v=Z2FtP...
Our review of the Gigabyte Aero 15X: https://www.youtube.com/watch?v=nJYDC...hfm, Petrov, Darkhan and 1 other person like this. -
childprotectorofthenight Notebook Consultant
Can anyone measure the decibel reading on their blade under load if they get the chance? I was getting around 70db.
-
hmscott and childprotectorofthenight like this.
-
childprotectorofthenight Notebook Consultant
-
childprotectorofthenight likes this.
-
childprotectorofthenight Notebook Consultant
that’s not even close to what I was getting. Definitely revauluating after that!
-
- CPU Power package spiked to 65W quickly for a few seconds, then settled back down to just under 45W where it sat for about 4 mins, then it fell to a hair under 35W
- Cores jumped to 38-39x for those initial few seconds, then settled back at 30x (yes, only 30x!!!) for the 4 mins that the power package was 45W
- Temps on the CPU slowly rose over those first four minutes, peaking around 97C only after 4 minutes of stress testing - it was at this point that (I think) some thermal throttling kicked in and the power package fell to 35W
- As I watched the HWInfo63 readout, it was showing that the power limit throttling had kicked in from very near the START (it read YES for all cores), even though core temps were still in the mid 80s at this point.
I think it's clear that there is some power limit throttling going on for some odd reason. The fact that it was only pushing 30x multiplier across all cores despite all core temps (and PCH temp) being in the 80s during this time...well I can't understand that. -
-that 65w power spike was during the time window for the short power limit (which is usually 56w or 1.25x PL1) and the combination of 100% load (with some instructions such as avx from the fpu test) with your current cpu voltage at max clocks, resulted in that power draw, after that time window expires, PL1 (at 45w) kicks in and downclocks you cpu in order to maintain a sustained power draw of 45w (all "normal" up to this point),
-as I previously mentioned, some laptops have a 35w power limit under combined loads (only when you stress test both the cpu and the gpu, if you only test the cpu, then you should be limited at 45w under sustained loads), but what I didn't clearly get... did the max power draw remained at 35w until the test finished, or was it just a temporary thing and it continued to be just under 45w until you stopped the test?
30x isn't out of the ordinary under those circumstances, those clocks were the max that could be sustained under that load with that voltage, to have a power draw just under the limit (but I didn't understand if you were 35w or 45w limited before you stopped the test).
Could you take a pic (make sure you have the Unified tab active) a few minutes after the test started (use max fans during the stress test) and one before you stop the test and post them here using https://postimages.org or the app https://postimages.org/app (with imgur I find them a bit difficult to be read). -
https://postimg.cc/image/avyelc0pj/
https://postimg.cc/image/7p3v1sgaf/
I wasn't using the unified tab as it wasn't telling me anything I couldn't see from either my hwinfo or throttlestop task bar temperatures.Throttlestop says I hit 100C, but hwinfo said 97C. -
This is a premature power limit put in place even at temps in the 70s and 80s that's happening with these machines. Razer is aware of it and I'm waiting for some updates on the situation. I actually linked this thread for Razer to read over and are evaluating everything.
Note that is doesn't always happen... I have had gaming sessions where I have stayed at 3.9ghz despite gpu load. -
Off the general temp/clocks topic, but I did order the razer 15.6 sleeve. Wow, what the actual F was Razer thinking when they made this? It's almost the size of my OMEN X laptop. It literally doesn't fit in any 15" backpack I own. Returning it would likely be a complete buttache, I'm going to get it sewn and trimmed to fit. How incredibly stupid.
-
-
-
EDIT it looks like a pro would fit there... OMG
Anyone could recomend a hard case like they used to have for the early gen Blades?
Sent from my SM-G955F using TapatalkLast edited: Jun 27, 2018Assault likes this. -
Judging from the delay it kicks in, there should probably be minimum interference in gaming loads (especially if you keep the undervolt, because games don't constantly draw that much wattage from the CPU).
The TDP and power limits for the kabylake series included the TB clocks, for this gen, the TDP is only for stock clocks.
You also have to consider the extra cores that generate a higher power draw under the same circumstances.
For eg: 7700HQ with an undervolt and max TB clocks, can maintain a relatively high load without exceeding 45w, while an 8750H with the same clocks, at the same voltage, under the same load, has a higher power draw so either it has a higher power limit so it can maintain higher clocks (if the cooling system can handle them, if not, then it thermal throttles) or it lowers the clocks to maintain a sustained power draw under 45w (that the cooling system was designed to handle).
Like I previously mentioned, in gaming it kicks in rarely because there isn't a constant power draw above the limit, you only have sporadic spikes above the limit.
I'm against throttling and I agree that the throttling mechanisms should be more simple and user unlockable, but from the previous stress test, it's clear that the cooling system with that paste job (I don't know if it's the stock paste) can't handle sustained loads that result in a constant power draw above 45w (that is under combined loads). With LM you should have more thermal headroom and it might be worth the effort to increase the power limits so you can maintain higher clocks even under more strenuous sustained loads.
The 35w limit is just an additional throttling limit added to an already complicated throttling mess. -
-
@stranula
This is even worst!
If it isn't triggered only when stressing both the CPU and GPU, then I don't know what triggers it.
I don't know if this limit can be disabled/bypassed somehow, this might be something that only razer could sort out (if they are willing to do this). -
raz8020 likes this.
-
No problem!
I just hope that you (all of the razer owners) are lucky enough to get a fix (at least for the 35w throttling) from razer!Darkhan likes this. -
I have a sleeve I use for my Dell XPS 15 (x2) and I will verify a little later than the Razer fits. XPS 15 fits like a glove. Fairly certain they are about the same exact size laptop. -
-
-
-
Attached Files:
-
-
This sleeve fits like a glove: https://www.amazon.com/gp/product/B01BUICFDA/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1
It's very nice. Soft inside and plenty of padding. Look for Macbook Pro 15" sleeves for your RB15. Basically the same size.
Assault likes this.
Razer Blade 15.6" (GTX 1060/1070, 2018 edition) Owner's Lounge
Discussion in 'Razer' started by X33nbat, May 20, 2018.