Related
I know many people have been saying the low benchmark scores on the Nexus 4 are due to "thermal throttling" but I don't really believe this. It's supposedly based on the optimus g, and if that doesn't suffer from this thermal throttling issue, then why would the Nexus 4? I personally believe it's just due to the software. People are also saying that an update won't boost the performance by that much, but I'm not sure what to believe.
So do you guys really believe that the nexus 4 is over-heating and under-clocking itself, or do you believe it is just in need of an update?
It may be a faulty unit from that reviewer, I only recall one review saying it overheated. The software is also to blame. I made a post in another thread about the performance with these pre-release software versions the reviewers have.
http://forum.xda-developers.com/showthread.php?p=33634026#post33634026
Unfinalized software on all these review devices is probably the bottleneck. Not to mention benchmarks have zero bearing on anything ever.
And Testing units and software do have a tendency to carry heavy logcats and monitoring software... I remember from the ICS days how "heavy" most leaks would run progressively getting better by the update ...
Nospin said:
Unfinalized software on all these review devices is probably the bottleneck. Not to mention benchmarks have zero bearing on anything ever.
Click to expand...
Click to collapse
Whilst benchmarks shouldn't matter too much, when this beastly specced phone is getting worse scores than the msm8960 with adreno 225, then it will obviously raise some concerns.
E3SEL said:
I know many people have been saying the low benchmark scores on the Nexus 4 are due to "thermal throttling" but I don't really believe this. It's supposedly based on the optimus g, and if that doesn't suffer from this thermal throttling issue, then why would the Nexus 4? I personally believe it's just due to the software. People are also saying that an update won't boost the performance by that much, but I know if I wrote the drivers for the phone, it wouldn't even start.
Do you guys really believe that the nexus 4 is over-heating and under-clocking itself, or do you believe it is just in need of an update? Also, another thing I wonder is this: is it called the nexus 4 because it has a 4" (4.7", I know) display, or is it called a nexus 4, because it is the fourth nexus?
Click to expand...
Click to collapse
I wouldn't say I'm good or experienced enough with Android smartphones to decide whether or not it's due to software, but I sure hope it is. I'm really only judging this particular issue by what everyone else is saying.
In regards to the sceen size and "Nexus 4" theory, I agree. Nexus 7 and Nexus 10 are a good example of device names in correlation with screen sizes here.
just wondering why did google name their nexus phone 10 wbefore the fourth one
I think it all came from a comment at Anandtech, they attempted to run all the GLBenchmark tests one after another, most web sites just chose 1 or 2 tests, usually Egypt HD. Most devices crash when trying to run all GLBenchmark test serially, it does on my Nexus 7, something to to do with running out of memory allocation.
Nexus 4 in a kind of suicidally awesome way completes the entire GLBenchmark suite in one go, but running all those test including offscreen & onscreen is a long brutal test, maxing out the SoC in a way no game is likely to do, so the fact that the device is thermal regulating itself is not that strange.
If Anandtech tested individual elements of GLBenchmark, as most other review sites do, this issue would not have occurred. In fact they admitted that the Optimus G could not run all tests consecutively, so they only tested individual elements, hence no the device didn't get downclcoked due to thermal limits. It is not good that Anandtech has this disparity in testing methodology, I like the website a lot, but some thing recently have led me to question a few things, but that is another story.
Every phone overheats nowadays so there's nothing different with the n4.
This is pretty interesting
Benchmark comparison. Once at room temp, once in freezer. Freezer scores are significantly better
http://techie-buzz.com/mobile-news/...es-show-the-real-power-of-the-s4-pro-soc.html
So yeah... Kinda does look like pretty bad thermal issues Hopefully just cause it was pre-release or something
Turbotab said:
If Anandtech tested individual elements of GLBenchmark, as most other review sites do, this issue would not have occurred. In fact they admitted that the Optimus G could not run all tests consecutively, so they only tested individual elements, hence no the device didn't get downclcoked due to thermal limits. It is not good that Anandtech has this disparity in testing methodology, I like the website a lot, but some thing recently have led me to question a few things, but that is another story.
Click to expand...
Click to collapse
But they singled out the Optimus G, because it was unable to complete more complex tests because of crashes.
mejobloggs said:
This is pretty interesting
Benchmark comparison. Once at room temp, once in freezer. Freezer scores are significantly better
http://techie-buzz.com/mobile-news/...es-show-the-real-power-of-the-s4-pro-soc.html
So yeah... Kinda does look like pretty bad thermal issues Hopefully just cause it was pre-release or something
Click to expand...
Click to collapse
Wow, that is a very interesting find. I was actually thinking that someone should do a freezer test just see if it is overheating. This article would seem to prove that it is. Those retests show dramatically higher scores, more on par with what the S4 processor should be capable of.
They said that perhaps the retail versions will have a higher tolerance for heat because they did not think they felt that hot. More testing and info is needed though.
Ryukeima said:
They said that perhaps the retail versions will have a higher tolerance for heat because they did not think they felt that hot. More testing and info is needed though.
Click to expand...
Click to collapse
To elaborate, straight from the source, in the Anandtech podcast this morning Brian commented that the phone seemed to be set to throttle at 60 degrees (for the dual-krait S4 at least it's usually around 80 afaik) and at that point the exterior of the phone was much cooler than a lot of other phones. (He talks about it at about 00:51:00.)
Something to think about.
Sjael said:
To elaborate, straight from the source, in the Anandtech podcast this morning Brian commented that the phone seemed to be set to throttle at 60 degrees (for the dual-krait S4 at least it's usually around 80 afaik) and at that point the exterior of the phone was much cooler than a lot of other phones. (He talks about it at about 00:51:00.)
Something to think about.
Click to expand...
Click to collapse
That means it's a simple tweak, and so as long as people aren't noticing excessive heat (which we would have heard from reviews) on the phone itself then is sounds like things will be fine for the release.
Yes IT FREAKING OVERHEATS
E3SEL said:
I know many people have been saying the low benchmark scores on the Nexus 4 are due to "thermal throttling" but I don't really believe this. It's supposedly based on the optimus g, and if that doesn't suffer from this thermal throttling issue, then why would the Nexus 4? I personally believe it's just due to the software. People are also saying that an update won't boost the performance by that much, but I know if I wrote the drivers for the phone, it wouldn't even start.
Do you guys really believe that the nexus 4 is over-heating and under-clocking itself, or do you believe it is just in need of an update? Also, another thing I wonder is this: is it called the nexus 4 because it has a 4" (4.7", I know) display, or is it called a nexus 4, because it is the fourth nexus?
Click to expand...
Click to collapse
I've had the phone for five days now and it definitely overheats. The last time it overheated was this morning. I got up, looked at the phone and unplugged it from the charger. Went to attend to my toddler for 20 mins and came back and saw that the phone was very warm and battery life was hit 20 percent. I did a check on battery usage under 'battery' in settings and saw that the playstore app had sucked 50% of the power of late. This is HIGHLY unusual becuase usually the screen is what sucks the most juice. I turned it off and turned it back on and it went back to being normal. During this overheating the phone stutters EXTREMELY visibly. Probably due to the massive thermal throttling of the cores. Once it cools off it goes back to being normal. This happens randomly and occurs once or twice a day. I've reported the bug to google. Otherwise the phone is superb.
That sounds like the Play Store app is misbehaving. Have you tried clearing the data?
Sent from my Nexus 4 using Tapatalk 2
Remember it's winter at northern hemisphere. So overheating might not be obvious.
Any friends in southern hemisphere (Australia) wanna chime in with their temperatures?
Paradisle said:
I've had the phone for five days now and it definitely overheats. The last time it overheated was this morning. I got up, looked at the phone and unplugged it from the charger. Went to attend to my toddler for 20 mins and came back and saw that the phone was very warm and battery life was hit 20 percent. I did a check on battery usage under 'battery' in settings and saw that the playstore app had sucked 50% of the power of late. This is HIGHLY unusual becuase usually the screen is what sucks the most juice. I turned it off and turned it back on and it went back to being normal. During this overheating the phone stutters EXTREMELY visibly. Probably due to the massive thermal throttling of the cores. Once it cools off it goes back to being normal. This happens randomly and occurs once or twice a day. I've reported the bug to google. Otherwise the phone is superb.
Click to expand...
Click to collapse
Same thing happened to me earlier today, I unplugged the phone from the charger and was extremely hot. I panic and shut the phone off. I rooted the phone last night so that was the only thing I did differently. I've had the phone for almost a week.
I also noticed the battery was draining way too quickly.
More like rogue apps... had the phone for over a month now, heavy usage every day and never once has it "overheated"as people say.
If im multitasking, like listening to music, downloading a torrent in the backgroud while playing angry birds or something the top back of the phone will get warm, but nothing unbearable if i deliberatly grab and hold the phone at that spot.
It's a glass phone... it will get warmer than most people are used to... its bascially the same glass that the new kitchen ovens use as a top surface... if you can cook on a glasstop stove... a phone heating up will be the same principle, albit on a smaller scale..
So yes, for the people complaining about heat.. then say in the next breath that they lost battery in some % form... funny how nobody is telling what they have installed as extra..or post screenshots of the battery page to back up the claim with info so we can help...large loss of battery % right away points to a rogue app somewhere.. you dont magically lose 10-20-50% whatever battery when unplugging the phone... thats something stuck running that is forcing your cpu to run at max for an extended period of time.
so as the internet expression says "fraps or it didnt happen" (screenshot or it didnt happen) lol
I am seeing battery temp reach 40C during antutu benchmark test running 4.2.2. Looks like it is only affecting a few devices. It gets warm during the benchmark test but nothing like unbearable heat.
We all know why LG choose 808 over 810, now let's see the results of their choice:
Here we'll talk about the G4 operating temperature.
Try to indicate your ambient temperature, model, display brightness, app(s) used, your cpu utilization (I use cool Tool for it), and of course temperature (for that purpose I use Cpu Temp wich basically reads bms sensor of Cpu-z in overlay, and Cpu-z itself) ;
Remember to distinguish battery temperature from cpu sensors temperature and if using cpuz remember that temperatures are very sensitive to time so measure it only while performing the task wanted.
Keep it as scientific as possible
*I forgot to mention that CpuZ has an erroneous C to F conversion. So keep it in °C to be accurate
There is an intersting heat comparison HERE on Android Central to begin with
Good idea. Do note that CPU-Z seems to have a bug where it converts temperatures incorrectly from C into F, in the Thermal tab. From what I've seen, CPU-Z should be set to display in C, not F, if you want accurate readings.
To see the issue, check the temps, including in the Battery tab, then in the Thermal tab. For me, in C, the Thermal tab numbers seem reasonable, and the "battery" line in Thermal is close to the temp shown in the Battery tab. But change the units to F, and the Thermal tab suddenly reads the battery, and everything else, too-high by about 30F. The temperatures are not converting properly from C to F, the displayed C and F values are not equivalent. This conversion issue appears using CPU-Z on my last phone, as well.
So if posting CPU-Z temps from the Thermal tab, I'd suggest setting it to C, not F.
Thanks, I forgot to mention it, added
What's the point of this thread? This isn't a PC, it's not like we can do anything about the temps.
kyle1867 said:
What's the point of this thread? This isn't a PC, it's not like we can do anything about the temps.
Click to expand...
Click to collapse
That is where you are wrong. It's called we can hard mod the phone for better temps.
Total newbie question, is the temperature you're feeling on the hand the one from the CPU, or the one from the battery, which is closer to the hand?
DeadPotato said:
Total newbie question, is the temperature you're feeling on the hand the one from the CPU, or the one from the battery, which is closer to the hand?
Click to expand...
Click to collapse
It should be the battery because I notice my phone getting warm pretty often and I have a widget for battery temperature which tends to go up to around 38 °C often. I didn't check CPU temperatures but I don't think the CPU is heating up too much.
82 F - sitting on the desk doing nothing but looking pretty. VZW, brightness at 0% and Auto.
All these freaking temperature problems will be resolved in a few months when the freezing winter arrive xdddddddddddddf
Sent from my LG G4 H815 USA 4G LTE TMO
MrSteelX said:
That is where you are wrong. It's called we can hard mod the phone for better temps.
Click to expand...
Click to collapse
Lol, no, you can't.
At least LG choosing 808 instead of 810 due to hear issue was the right choice.
https://youtu.be/HltGLYZLySs
Take a look at a link above, SD810 might have upper hands during cooled state but that changes quickly throttling down below SD808 specs. SD810 potentials are pretty much obsolete and probably will never run on full more than 5 minutes.
SD808 is holding its speed even running hot. So I wouldn't really worry about hear part for this phone. But I'm with some users here. Few software updates will iron out some heating problems.
Sent from my LG-H811 using XDA Free mobile app
I've noticed that phone is getting hot when charging even if the phone is in standby. I'm getting around 40-44c.
t68kv said:
I've noticed that phone is getting hot when charging even if the phone is in standby. I'm getting around 40-44c.
Click to expand...
Click to collapse
That's low for charging honestly if your referring to cpu temp. Most phone cpu temps stay around 25-35 idling and mid 40s when just doing simple scrolling and small tasks. 60-70c is fairly normal on heavier tasks and games. Kernels usually don't even start throttling until the cpu gets in the70-80c (generally most start shutting down cores at 80).
That being said, my G4 is one of the coolest running phones that I've messed around with. CPU-Z usually says that I'm in the low 40s even on some heavier tasks. I know that my Note 4 gets much warmer, much faster. Heavy web browsing for example will have my N4 in the upper 50s low 60s. The G4 stays pretty stable in the upper 40s low 50s. I'm happy they chose the processor that they did honestly. Stays pretty cool compared to the competition.
you can even mine cryptocurrency on this beast for hours on a normal (here normal) 27-30°C ambient temperature, and the battery gets to 44°C and the CPU gets into the 50s.
This phone has one of the BEST thermal management I've ever seen. I was a nexus 4 user, so you can see how much i was struggling.
Sent from my LG-H815 using Tapatalk
Thank you all
Please add which sensor are you referring to and with wich app you use to read sensors
sharpehenry said:
At least LG choosing 808 instead of 810 due to hear issue was the right choice.
https://youtu.be/HltGLYZLySs
Take a look at a link above, SD810 might have upper hands during cooled state but that changes quickly throttling down below SD808 specs. SD810 potentials are pretty much obsolete and probably will never run on full more than 5 minutes.
SD808 is holding its speed even running hot. So I wouldn't really worry about hear part for this phone. But I'm with some users here. Few software updates will iron out some heating problems.
Sent from my LG-H811 using XDA Free mobile app
Click to expand...
Click to collapse
I don't pretend to be rude, but that information is outdated and inaccurate. I don't know how the other manufacturers are doing with the 810 to be honest, the Z3+ is sort of new and I've been checking the threads but there's too early to judge so far, it happens that I do have an HTC One M9 up to date, stock and unrooted and I've noticed several improvements through updates. Considering my daily usage I didn't feel this was matching with what I experience in real life, so I decided to redo the test on the video. I ran AnTuTu benchmark 64 bits version 4 times in a row, and left an average of a minute (sometimes less, once was a bit more cause I took the wrong screenshot) between tests. The attachments were uploaded in chronological order, you can also guide for the hour displayed from the device.
Now besides the video only runs 4 times, I did notice it was holding every couple of tests and then decreases, so I decided to run one more for the sake of pure testing and check throttling behavior, and surprisingly the fifth run gave me an increase on the score instead of lowering it further.
I didn't find a dedicated benchmark thread on the G4 forum, unless LG had improved a lot the 808, and based on the several benchmark results you'll find across the internet, including the ones in that video, looks like even throttling snapdragon 810 holds as an 808 in worst case scenario.
About heating problems, if I'm providing a success case scenario with a more troublesome chip like the 810 is, I'm completely sure the 808 will be fixed in time
phone gets crazy hot when running Periscope
it even shows that hot temp message / stops charging
try broadcasting for 5 mins
Intersting comparison between M9, S6, G4, Droid Turbo heat dissipation HERE
Periscope heats up the phone like crazy!
I too have had issues with the device getting very hot on the top half of the screen. Tried a couple of factory resets but it didn't help. Even reset one more time and kept it stock after a few hours same problem. Took it to the AT&T store and of course it was behaving fine. Luckily the rep knew me and accepted I knew what I was talking about. Swapping mine for one first thing this morning when their new shipment comes in. Will update if issue persists or not.
Wish I would have done it 3 days ago. Might have been snagged another battery and leather case lol.
Sent from my LG-V495 using XDA Free mobile app
As the title says, my phone gets hot really easily with simple usage. Today I was just changing my ring tones and notifications sounds and it became warm (not that hot) where it shouldn't have, with such light usage. And then I played a game for about 10 minutes and the device got hot and was untouchable. The game and UI started to lag too.
This is my replacement phone, the first phone did the same. But I think this one is worse.
During normal usage it does get hot (very uncomfortable), but it won't lag.
My phone is always above 40-42C.
Is it due to my climate?
Or should I get another replacement.
Please help me on this, I can't really figure out if it's the phone or me getting defective devices.
edios123 said:
As the title says, my phone gets hot really easily with simple usage. Today I was just changing my ring tones and notifications sounds and it became warm (not that hot) where it shouldn't have, with such light usage. And then I played a game for about 10 minutes and the device got hot and was untouchable. The game and UI started to lag too.
This is my replacement phone, the first phone did the same. But I think this one is worse.
During normal usage it does get hot (very uncomfortable), but it won't lag.
My phone is always above 40-42C.
Is it due to my climate?
Or should I get another replacement.
Please help me on this, I can't really figure out if it's the phone or me getting defective devices.
Click to expand...
Click to collapse
I had the same problem when it was summer where I am. The SD810 is terrible when it comes to heating up, and the lag you experienced is due to the chip throttling back from the heat. Is it hot weather where you are?
Heisenberg said:
I had the same problem when it was summer where I am. The SD810 is terrible when it comes to heating up, and the lag you experienced is due to the chip throttling back from the heat. Is it hot weather where you are?
Click to expand...
Click to collapse
The weather is indeed hot nowadays. But the problem is, if the phone heats so much on simple usage, then what is the purpose of owning this smartphone..? :| I'm really disappointed considering I changed my previous device cos of overheating (Note 4).
What's confusing me is, some reviews/reports say this phone is cooler than the S6 /Edge. But my friend has one and it doesn't heat this much at all.
My dad's 6S Plus runs insanely cool (compared to my 6P).
Should I go for another repair/replacement? I've already wasted 1 month waiting for a replacement.
Everything else is amazing on this phone (except the DAC of course) .
Edit: the battery is also disappointing(like 4:30 to 5:00 average). I guess it's due to the phone heating up quickly.
edios123 said:
The weather is indeed hot nowadays. But the problem is, if the phone heats so much on simple usage, then what is the purpose of owning this smartphone..? :| I'm really disappointed considering I changed my previous device cos of overheating (Note 4).
What's confusing me is, some reviews/reports say this phone is cooler than the S6 /Edge. But my friend has one and it doesn't heat this much at all.
My dad's 6S Plus runs insanely cool (compared to my 6P).
Should I go for another repair/replacement? I've already wasted 1 month waiting for a replacement.
Everything else is amazing on this phone (except the DAC of course) .
Edit: the battery is also disappointing(like 4:30 to 5:00 average). I guess it's due to the phone heating up quickly.
Click to expand...
Click to collapse
Didn't you research this phone and its chipset before buying? The SD810 is well known for this. I think most reports that say the phone runs cool were written just after release, which was just as the North American winter was taking hold. You could try getting a replacement but I'm fairly confident that this is a problem with the chip.
Heisenberg said:
Didn't you research this phone and its chipset before buying? The SD810 is well known for this. I think most reports that say the phone runs cool were written just after release, which was just as the North American winter was taking hold. You could try getting a replacement but I'm fairly confident that this is a problem with the chip.
Click to expand...
Click to collapse
I chose this phone over the Note 5. I heard there's absolutely no heating issues with the Nexus 6P and the 810 is implemented well in this phone.
My MacBook Pro started to heat much more than it did a few months ago too. But I can't confirm it's because of the weather because it might be some new software I installed or anything.
But the thing that bugs me is, even Yureka, a budget phone, which is known as a heater, heats only like my Nexus 6P or maybe sometimes it's even better at handling heat.
My display is very very yellow too.
This phone is literally the hottest phone ever!
Two minutes of web browsing and the upper portion would have already gotten quite warm.
I think it's getting worse lately, maybe because of the aging of the thermal compound.
ALUOp said:
This phone is literally the hottest phone ever!
Two minutes of web browsing and the upper portion would have already gotten quite warm.
I think it's getting worse lately, maybe because of the aging of the thermal compound.
Click to expand...
Click to collapse
Agree. I think the aging of the thermal compound is def going to make thermal throttling worse over time. Plus the thermal compound they used is absolute crap. It had a marshmallow like texture. For those able, I'd consider doing the thermal hard mod posted here. Even if instead of the thermal pad you just replace the crap thermal compound with good stuff. The phone is not nearly as difficult to disassemble and some make it out to be.
You've got a Nexus, go ahead and get your root. Custom kernels like EX do pretty well with excess heat management.
EX kicks in an optional hexacore mode just in case your phone gets mad on heat (generally while charging).
Custom kernels are great in that respect.
Hi has anyone else looked at the thermal throttling on this device?
I complained with the initial release.
I haven't seen anything.
My LG G6 with a Snapdragon 821 outperforms this phone.
Post your Super-PI:
Sent from my PH-1 using XDA Labs
Maybe something wrong with your unit? Update it and factory reset.
Mine destroys any SD821 phone all day and night.
ChronoReverse said:
Maybe something wrong with your unit? Update it and factory reset.
Mine destroys any SD821 phone all day and night.
Click to expand...
Click to collapse
Run Antutu back to back and watch your performance degrade.
Run Geekbench or Vellamo and post your scores.
The SD821 device is an LG G6 and it uses a heat pipe to get heat away from the CPU and it scores the same kind of numbers as the Essential unless you mess with thermal-engine.conf on the Essential.
I've tweaked mine and phone is blazing fast now and doesn't throttle as early.
Stock this phone throttles like crazy.
Is this phone just slow, like almost unusably slow all the time? Over here, I got it less than a week ago, stock, maybe 30 apps in total, no major drainer such as FB or the like.
And yet, compared to my one year old OnePlus 3 this is crawling and painfully slow. Just earlier today I erased its cache and restarted it. It helped for a few minutes I guess. When it's slow like this, can't listen to Google music via Bluetooth because it's stuttering continuously, every once in a while I'd touch an icon and it takes seconds to start an app and no feedback from UI until then. A few moments ago I was taking some photos with stock camera and attempted to delete a pic from thumbnails, the app froze for a few seconds then it just deleted 2 pics in the meantime.
I bought this during the bf deals for 400$ with camera but thinking of returning it.
I only wonder how anybody else finds this phone usable at all in day to day usage?
Sent from my PH-1 using Tapatalk
No stutters, just the touchscreen jank. Maybe update to latest OTA and factory reset?
@tech_head
You realize the Essential has a heatpipe too right?
ChronoReverse said:
No stutters, just the touchscreen jank. Maybe update to latest OTA and factory reset?
@tech_head
You realize the Essential has a heatpipe too right?
Click to expand...
Click to collapse
It may, but the the thresholds for heat throttling are so low you can force throttling just by running Antutu.
Until I played with the "thermal-engine.conf" getting reliable results was hit or miss.
tech_head said:
It may, but the the thresholds for heat throttling are so low you can force throttling just by running Antutu.
Until I played with the "thermal-engine.conf" getting reliable results was hit or miss.
Click to expand...
Click to collapse
https://imgur.com/a/916Bf
For ****s and giggles I just ran Antutu three times in a row. I stared at it so I could take the screenshot of the result immediately after it's done and run it again. The phone is just sitting on my desk, it got warm but not as hot as my HTC 10 did and nowhere near where my Nexus 6P burned at. Looks like a little throttling but not very much.
And then I ran Geekbench 4 right after the third run because I already had it installed and got 1898/6470. Normally get around 1850/6500 so the result I got after so much heat build up doesn't seem like any throttling there either.
Stock Nougat NMJ51B, Magisk rooted so I can use Titanium Backup and Adaway.
I dunno, maybe you have faulty hardware?
ChronoReverse said:
https://imgur.com/a/916Bf
For ****s and giggles I just ran Antutu three times in a row. I stared at it so I could take the screenshot of the result immediately after it's done and run it again. The phone is just sitting on my desk, it got warm but not as hot as my HTC 10 did and nowhere near where my Nexus 6P burned at. Looks like a little throttling but not very much.
And then I ran Geekbench 4 right after the third run because I already had it installed and got 1898/6470. Normally get around 1850/6500 so the result I got after so much heat build up doesn't seem like any throttling there either.
Stock Nougat NMJ51B, Magisk rooted so I can use Titanium Backup and Adaway.
I dunno, maybe you have faulty hardware?
Click to expand...
Click to collapse
I'm inclined to agree with Chrono...
Ran Antutu 3x and got the following:
167746
167400
168142
In that order. I dont see any throttling occuring. Running on the Oreo beta, no root.
So I came across this app https://play.google.com/store/apps/details?id=skynet.cputhrottlingtest
And lo and behold the cpu does throttle under extensive load.
Sent from my PH-1 using XDA Labs
avd said:
So I came across this app https://play.google.com/store/apps/details?id=skynet.cputhrottlingtest
And lo and behold the cpu does throttle under extensive load.
Sent from my PH-1 using XDA Labs
Click to expand...
Click to collapse
Interesting app. So it does appear to throttle, but honestly, any passively cooled CPU running at max speed for a consistent period of time would have to eventually. The type of stress an app like this puts the CPU under is uncommon however, even intensive games will not consistently stress the CPU as much as an app like this.
That said, my throttling doesn't appear anywhere near as dramatic as yours, and it recovers quickly, as soon as the temp drops below a threshold it spikes back to near full performance.
This is how Intel and AMD's turbo boosts work as well, they will boost to a certain speed until it is thermally not feasible for the safety of the processor.
Qualcomm's own documents state that the performance cluster of cores supplies "up to" 2.35GHz
https://wccftech.com/qualcomm-snapdragon-835-specifications-design-details/
Throttling under extreme conditions is normal. Even iPhones throttle after a bit if you put the right kind of load on it.
The question here is whether it throttles more than the LG G6 and I can guarantee if the OP tried running that particular throttle app on the LG G6 that it'll throttle too.
ChronoReverse said:
Throttling under extreme conditions is normal. Even iPhones throttle after a bit if you put the right kind of load on it.
The question here is whether it throttles more than the LG G6 and I can guarantee if the OP tried running that particular throttle app on the LG G6 that it'll throttle too.
Click to expand...
Click to collapse
I haven't run any throttling app.
I have just found that running Antutu two consecutive times yield wildly different results.
I also found that even slightly above room temp, the phone will throttle.
This was the case with the very first one I had at release and the one I got through Sprint on Friday.
And the throttling is much worse than my LG.
In the same tests my LG will throttle much. much later then the Essential.
To the point that I can get better performance with my LG than I can with an unmodified Essential.
I meant this app posted by avd: https://play.google.com/store/apps/details?id=skynet.cputhrottlingtest
Try running that on your LG G6 and take a screenshot.
Anyway, if your Essential is throttling that much, it's probably defective hardware, maybe the heatpipe isn't seated correctly.
ChronoReverse said:
I meant this app posted by avd: https://play.google.com/store/apps/details?id=skynet.cputhrottlingtest
Try running that on your LG G6 and take a screenshot.
Anyway, if your Essential is throttling that much, it's probably defective hardware, maybe the heatpipe isn't seated correctly.
Click to expand...
Click to collapse
This was a 5 Minute test of GIPS with the application.
Peak Ave Min %Throttle
LG G6 -> 99.4. 81.9. 38 63% (Took more than 2 minutes before it dropped to below 90% in performance. )
PH-1 ->137.86. 103 69 69% (Throttling in the first minute and cycled but rarely got above 75% perf).
PH-1 ->137 116 94.95. 81%. (This is with a custom thermal-engine.conf you see dips but not many below 90%)
So the PH-1 throttles early and hard, but it levels out.
The LG manages heat better but when it throttles after 2-3 minutes it's pretty hard (50%)
How this translates into real world performance is that the stock PH-1 will appear to lag, since under load it throttles early under short peaks.
The LG takes longer to get into a state of throttling so it feels smoother.
So in short benchmarks; Geekbench Vellamo and Antutu where you get some shorts bursts of high activity the PH-1 will ive unreliable results because of early throttling. The LG G6 will look better because it takes a lot longer to throttle. Benchmarks will be more predictable and performance will appear better. It looks like the LG G6 heat pipe is more efficient, but when you saturate it, game over.
Since the Essential throttles early the limits to 67% of the clock while throttled it has overall better characteristics when throttled.
The LG on the other hand, takes a while to throttle but the first step down in performance is to ~50% clock frequency.
Anyway, with a custom thermal-engine.conf things are much better.
A custom kernel would give better results.
Don't get me wrong, I like the PH-1 but it throttles early.
This is my second device with a three month separation in build date.
The odds of me getting to with a wonky heat pipe are pretty low.
Where's your screenshot? We have two examples here where the PH-1 with 8 cores throttle to 65% and 80% but in a All-Core test like this, the SD821 is only 65% the speed of the SD835 in the first place.
Even throttled to the screenshotted levels, the Essential can still keep up with the LG G6 at full speed
ChronoReverse said:
Where's your screenshot? We have two examples here where the PH-1 with 8 cores throttle to 65% and 80% but in a All-Core test like this, the SD821 is only 65% the speed of the SD835 in the first place.
Even throttled to the screenshotted levels, the Essential can still keep up with the LG G6 at full speed
Click to expand...
Click to collapse
Hi,
I don't have the ability to do the screen shots.
But as I indicated, it takes a lot longer for the LG G6 to start throttling.
It takes almost 2 minutes before any throttling on the LG G6 (+90% of max performance), but the PH-1 will start almost immediately.
So the LG G6 is at +90% but the PH-1 with the stock thermal-engine.conf almost immediately throttles to 60-70% then recovers and stabilizes some time later.
So as I said the overall performance of the PH-1 s most definitely better, but the PH-! may appear to stutter under heavy load initially, where the LG G6 takes longer to throttle and will appear to have more stable benchmark performance for short benchmarks.
Would you post your config changes please?
DaveHTC200 said:
Would you post your config changes please?
Click to expand...
Click to collapse
Here is the thermal file I use.
You have to be rooted and it goes into /system/system/etc with 644 permissions.
Also you need to remove the ".txt" extension.
It looks like in the last few update Essential has tweaked performance.
I don't see any throttling like before.
Hi folks, I'm torn between the snapdragon 636 and 845. I know 845 is the more powerful chip but I'm not exactly a power user.
I play games which runs well even on the snapdragon 625 and I love how the chip stays cool even with extended gaming sessions. On the other hand, I've had bad heat experience with the older 8xx series of snapdragons. No matter what I do they heat up so much till I can fry an egg with those phones.
Now the hard decision comes because the prices of Max 3 is just a little shy of the poco f1. I'm just sick of heat and would like some hands on feedbacks from the excellent community here.
Thank you so much!
I have the Max 3 and I have no problem with even higher end games at max settings - mobile CPUs have gotten so powerful that there aren't many things around that can actually challenge them anymore, even at the mid-range price point. I would consider myself a power user, and the 636 keeps up with everything I do with zero issues. I would recommend the Max for it's huge battery and reasonable screen resolution - anything above 350ppi is a waste anyways, and the Max 3 has exactly 350.
Hi misterk267,
Yeah, I certainly like the new format of the Max 3 and how much the battery can last before I have to charge it. Problem is the pricing is just so close to the Pocophone and it doesn't come in 6/64GB configuration.
Since we're on it, I just like to know the heat produced by the SD636. Does the SD636 ever throttle? I'm assuming the SD845 does throttle under extended continuous usage beyond 30 minutes.
I pre-ordered the Global Rom Mi Max 3 @ $290 before the Poco F1 was announced. As I am still waiting for it to be shipped, I have been very torn between it and the Poco F1 and wondering if I should cancel the Max 3 and rather get a Poco. I have to say that I don't game at all, so a Flagship chip like the 845 would probably be wasted on me. While the Poco has a slightly better camera, I am sticking with the Max 3 because of that massive battery life, bigger screen for media consumption and no notch. I'd love to rock a flagship chip and buy into the Poco hype (which is justified - amazing bang for buck), but I have wanted to own a Mi Max device for so long, so will have remorse if I don't hold out for my on order Max 3!
I cant get my 6GB/128GB .eu ROM Mi Max 3 to heat up no matter what I do to it, PUBG while charging no bother, it doesnt even get warm.
Tharg0r said:
I cant get my 6GB/128GB .eu ROM Mi Max 3 to heat up no matter what I do to it, PUBG while charging no bother, it doesnt even get warm.
Click to expand...
Click to collapse
Yeah, that's the thing I love about the 600 series of Snapdragons. My worry is SD845 heating up too much; it happened on my old cellphones with the older 800 series of Snapdragons.
However, the SD845 is on 10nm while the SD636 14nm. Max 3 has a larger screen that makes typing much easier, IR blaster and battery that allows me 2 full days of usage.
Really torn between Pocophone and Max 3.
Maybe I must buy both
if you need root go to pocophone as xiaomi has increase the waiting time for unlocking bootloader for their device , but not with pocophone.
I dont see any torn between these devices:
- You just see the torn because you want the best of both (battery, screen of max 3 and the performance of poco).
- IMO, after purchasing MM3 (went up from MM1) then continuing found out Honor play, Honor note 10, Poco, etc. I dont have any regret or think back because:
-- I defined clearly the reason why I chose MM3:
1/ great battery
2/ huge screen
3/ the performance which is enough for my daily use.
4/ Not too expensive.
So you should only feel the torn when MM3 cannot afford for your daily use and there is another phone with the same price/better specification or lower price/same specification, something like that.
Imagine that after buying poco, there is another phone which is even better than Poco, how do you think at that moment? Sell poco then buy that better phone?
windwaver said:
Hi misterk267,
Yeah, I certainly like the new format of the Max 3 and how much the battery can last before I have to charge it. Problem is the pricing is just so close to the Pocophone and it doesn't come in 6/64GB configuration.
Since we're on it, I just like to know the heat produced by the SD636. Does the SD636 ever throttle? I'm assuming the SD845 does throttle under extended continuous usage beyond 30 minutes.
Click to expand...
Click to collapse
I hadn't really thought about it, but now that you mention it I don't think I've ever felt my 6/128 Max 3 get warm at all. That being said, I've never had an SD845 device, so I can't comment about how they compare.
For what it's worth, my 636 hasn't failed me yet in the gaming department. I can run strangers wrath with all graphics settings maxxed out, and that's the most graphically intensive game I really play. And it STILL doesn't get hot.
Unless you want the SD845 for bragging rights (totally understandable!), I'd advise you to get the Max.
misterk267 said:
I hadn't really thought about it, but now that you mention it I don't think I've ever felt my 6/128 Max 3 get warm at all. That being said, I've never had an SD845 device, so I can't comment about how they compare.
For what it's worth, my 636 hasn't failed me yet in the gaming department. I can run strangers wrath with all graphics settings maxxed out, and that's the most graphically intensive game I really play. And it STILL doesn't get hot.
Unless you want the SD845 for bragging rights (totally understandable!), I'd advise you to get the Max.
Click to expand...
Click to collapse
You can try this app.
https://play.google.com/store/apps/details?id=skynet.cputhrottlingtest
Look if the temps rises and if the graphs dip after the test has been running for a couple of minutes.
If they do that indicates that the one or more cpu cores are being underclocked after a while to restrict heat output.
I have done this on my phones with different cooling paste to test before and after results.
I did look at some of the teardown videos on youtube on the Mi Max 3 and it has a square almost dry grey probably dow corning preapplied cooling paste.
It looked dry and teared in the video so its not very good to begin with.
The cooling paste is only for the RF shield.
Under the RF shield is two blue thermal pads. One big for the cpu and one smaller for the radio I think.
None for the storage chip.
Would be interesting to see if changing the cooling paste makes any improvement..I'll wait until my warranty expires before I open my Mi Max.
Under the RF shield with the pads, one alternative is to put copper thermal shims to make contact with the rf shield or a better quality thermal pad with higher thermal conductivity on the cpu and radio chip
http://www.thermal-grizzly.com/en/products/13-minus-pad-8-en
Teardown video: https://www.youtube.com/watch?v=yAwL1Q4k-C0
SUPERUSER said:
You can try this app.
https://play.google.com/store/apps/details?id=skynet.cputhrottlingtest
Look if the temps rises and if the graphs dip after the test has been running for a couple of minutes.
If they do that indicates that the one or more cpu cores are being underclocked after a while to restrict heat output.
I have done this on my phones with different cooling paste to test before and after results.
I did look at some of the teardown videos on youtube on the Mi Max 3 and it has a square almost dry grey probably dow corning preapplied cooling paste.
It looked dry and teared in the video so its not very good to begin with.
The cooling paste is only for the RF shield.
Under the RF shield is two blue thermal pads. One big for the cpu and one smaller for the radio I think.
None for the storage chip.
Would be interesting to see if changing the cooling paste makes any improvement..I'll wait until my warranty expires before I open my Mi Max.
Under the RF shield with the pads, one alternative is to put copper thermal shims to make contact with the rf shield or one of these pads on the cpu and radio chip
http://www.thermal-grizzly.com/produkte/6-minus-pad
Teardown video:
Click to expand...
Click to collapse
Nice share.
I ran the test for 5 minutes, and the results show throttling to 76%. That being said, the end experience is what really matters - I've certainly had heavy gaming sessions that lasted well beyond 5 minutes with no slowdown, though I have to wonder how Xiaomi's gaming optimization affects throttling when active. It's also possible that 76% is still enough to run these games. I'll add the app to the list of games and run another test in a few minutes.
Here is a screenshot.
misterk267 said:
Nice share.
I ran the test for 5 minutes, and the results show throttling to 76%. That being said, the end experience is what really matters - I've certainly had heavy gaming sessions that lasted well beyond 5 minutes with no slowdown, though I have to wonder how Xiaomi's gaming optimization affects throttling when active. It's also possible that 76% is still enough to run these games. I'll add the app to the list of games and run another test in a few minutes.
Click to expand...
Click to collapse
Yes this app puts a load that is kind of balls to the walls..you rarely ever push the cpu this hard in real life even in games on android.
But it gives an indication in which direction you might want to look at look at. Its a little on the warm side but its okay.
Its not going down below 50 to 40 to 35%. That would be a significant warning sign for heavy thermal throttling
Edit:
The screenshot is totally passed. Nothing wrong about that.
Starts off way up in the green and hovers in the high yellow and kind of stays there for the duration of the test and doesn't drop any lower than that.
If you see any low points that quickly rises back to its normal state and you know you touched the phone that might be why.
If you accidentally touched the screen or give the phone any additional stimulation it will decrease the result and it will show up at the given time on the graph.
Out of curiosity - you said you have run tests on other phones before and after swapping the thermal paste - have you seen any/significant changes in throttling when doing this?
misterk267 said:
Out of curiosity - you said you have run tests on other phones before and after swapping the thermal paste - have you seen any/significant changes in throttling when doing this?
Click to expand...
Click to collapse
No just slight. The top values are the same.
Just that it takes longer for the cores to start underclocking.
Of course the phone don't have any fans to exhaust this heat so you can only prolong the time before and to when the cpu cores are starting being underclocked. That's where the limitation comes in.
The cooling paste only improves the thermal conductivity so it reaches the far end of the phone quicker than with the old crappy cooling paste.
So we cannot compare this to a laptop or desktop computer where you change the cooling paste and see dramatic results because you have fans that push the heat away from the computer.
My honami dimmed the screen before and it took almost 5-6 minutes until it started feeling a littler cooler and the screen brightness slowly coming back to its original state.
And I couldn't see the damn screen! It was like the screen was off.
Now the phone gets VERY HOT!! and the screen dims. But the brightness returns after just a minute and the phone is already cool to the touch.
So the new cooling paste helps to quickly collect and move the excess heat. If compared to the old cooling paste that did not move the heat out of the way and the phone did not feel warm other than right above the motherboard while now the whole phone is scorching hot indicating that all that heat that collected at one spot is now helped to spread out to the whole area that is thermal conductive.
Same goes for my Z ultra but its a larger phone so it don't have the same issue as the smaller honami.
And the smell of a hot iron is gone too. You know the smell when you heat up your clothes iron and it starts smelling like hot materials.
Very hard to explain. But that scent is completely gone now when the phone gets very warm.
SUPERUSER said:
No just slight. The top values are the same.
Just that it takes longer for the cores to start underclocking.
Of course the phone don't have any fans to exhaust this heat so you can only prolong the time before and to when the cpu cores are starting being underclocked. That's where the limitation comes in.
The cooling paste only improves the thermal conductivity so it reaches the far end of the phone quicker than with the old crappy cooling paste.
So we cannot compare this to a laptop or desktop computer where you change the cooling paste and see dramatic results because you have fans that push the heat away from the computer.
My honami dimmed the screen before and it took almost 5-6 minutes until it started feeling a littler cooler and the screen brightness slowly coming back to its original state.
And I couldn't see the damn screen! It was like the screen was off.
Now the phone gets VERY HOT!! and the screen dims. But the brightness returns after just a minute and the phone is already cool to the touch.
So the new cooling paste helps to quickly collect and move the excess heat. If compared to the old cooling paste that did not move the heat out of the way and the phone did not feel warm other than right above the motherboard while now the whole phone is scorching hot indicating that all that heat that collected at one spot is now helped to spread out to the whole area that is thermal conductive.
Same goes for my Z ultra but its a larger phone so it don't have the same issue as the smaller honami.
And the smell of a hot iron is gone too. You know the smell when you heat up your clothes iron and it starts smelling like hot materials.
Very hard to explain. But that scent is completely gone now when the phone gets very warm.
Click to expand...
Click to collapse
Thanks for the explanation - even if it does not give dramatic results, it's still pretty interesting to see what happened when you made the modification. I don't think I'd ever have the courage to try something like that :crying:
How can you be torn between a 6 and a 7 inch screen? They are completely different beasts! Decide how big of a phone you can put up with (bigger the better) and then buy the corresponding one.
No processor could make up for a smaller screen for me.
dinhvtai said:
I dont see any torn between these devices:
- You just see the torn because you want the best of both (battery, screen of max 3 and the performance of poco).
- IMO, after purchasing MM3 (went up from MM1) then continuing found out Honor play, Honor note 10, Poco, etc. I dont have any regret or think back because:
-- I defined clearly the reason why I chose MM3:
1/ great battery
2/ huge screen
3/ the performance which is enough for my daily use.
4/ Not too expensive.
So you should only feel the torn when MM3 cannot afford for your daily use and there is another phone with the same price/better specification or lower price/same specification, something like that.
Imagine that after buying poco, there is another phone which is even better than Poco, how do you think at that moment? Sell poco then buy that better phone?
Click to expand...
Click to collapse
I wish there is a Max 3 with SD660 instead :fingers-crossed:. I really love the Max series for its battery, it's actually lighter to carry a Max phone than any other phone with a battery pack. The only thing is the weight; it'll be great if it's 200g but I prefer a larger battery than not.
I wouldn't hesitate to buy a Max 3 if it weren't for the price; it's almost the same as a Poco and I really prefer a 6/64GB configuration for the Max 3.
SUPERUSER said:
No just slight. The top values are the same.
Just that it takes longer for the cores to start underclocking.
Very hard to explain. But that scent is completely gone now when the phone gets very warm.
Click to expand...
Click to collapse
Wow, great post! Have you had the chance to do this with any SD845 phones? It seems that the Pocophone has some sort of heat pipes. I certainly don't want an SD845 that throttles way below an SD636 after gaming for 30 mins.