Search results

  1. D

    Post Magisk General Support / Discussion

    Tried that. Cache+data cleared, double checked Magisk hide, app still detects rooted device.
  2. D

    Post Magisk General Support / Discussion

    I mean apps that are blocked when root is detected (banking, PoGo, etc) all still detect the phone is rooted even though they are still hidden. They stopping working when the cts check failed but now that it passes the apps haven't started working again. Basically the devices are 'passing' but...
  3. D

    Post Magisk General Support / Discussion

    My ctsProfile now passes too, but no apps that check for root work anymore so it doesn't seem to have really changed anything.
  4. D

    Post [KERNEL][N96xx][PIE][2019-12-26] ** TGPKERNEL **[v1.7.1][4.9.207][CSK1]

    It's stored within the SoC itself and read directly out from the hardware. The kernel doesn't hold any of the voltage/asv tables.
  5. D

    Post Max Battery Capacity Not 4 Amps

    It will be rated for 4.4v charge float voltage, but Samsung always terminate those cells at 4.35v max. That's been the case since the S5/Note 4 days or maybe even earlier. I don't think typical capacity will be much over 3800 mAh. It will be rated 4000 at full voltage but the cell is stopped at...
  6. D

    Post Max Battery Capacity Not 4 Amps

    The phone reports it's real capacity, you can only read it with root access though. The ~94% figures are going to be apps guessing the capacity based on a 4000 mAh cell @ 4.35v actually only charging to 4.28v. It's all they can do without root access and reading the correct files. The real...
  7. D

    Post The 80-40 battery cult crowd.

    It's not impossible. The charge setup is in the kernel sources for everyone to read. We know exactly what they are charging the batteries to, and therefore we know they are undercharging them. Coming from a Note 4 the non-removal battery was one of my biggest gripes about upgrading device...
  8. D

    Post Max Battery Capacity Not 4 Amps

    The float voltage is 4.35v but the battery doesn't charge that high on a stock kernel. It's definitely configured at 4.28v in the source code. Mine reports 4.29v when full on the charger.
  9. D

    Post Max Battery Capacity Not 4 Amps

    I do wonder how much of the "missing" capacity is the fact that the phone underchargs the battery deliberately. It's a 3.85v nominal cell so that means a 4.35v cut off voltage, but the phone actually stops charging at 4.28v to improve cycle life. That 70mv different will easily account for 5-7%...
  10. D

    Post [ROM] [MM 6.0.1] KatshMallow v29 [TF201]

    I'm having a bit of an issue with my prime that's running V22. I can't seem to access recovery anymore. The device just boots into the rom no matter what I do. Does anyone have any ideas on what I can do with this? It works fine in the rom, just seem weird that recovery/bootloader is...
  11. D

    Post [KERNEL][5.1.1][TW][N910C] Simplicity Kernel v1.3 [04-04-2016][CUSTOM]

    Giving it a try. Seem good so far! Two things on V1.2. The GPU max default speed is still 100-800MHz, it wasn't changed to 160-600MHz (which is a much nicer setting - 100MHz is too slow and drops frames, and 800MHz is far beyond a sustainable clockspeed). Also in synapse, the "HMP migration on...
  12. D

    Post [KERNEL][5.1.1][TW][N910C] Simplicity Kernel v1.3 [04-04-2016][CUSTOM]

    Nice to see another customer kernel for the 910C! I'd be interested in giving it a try sometime, but do you plan to expand the CPU voltage control? -100mv isn't enough for me because my A57's currently run at -150mV and my A53's hit -175mV at certain speeds. I'd take a substantial voltage...
  13. D

    Post The Note 4 has live battery current monitoring, but no apps seem to take advantage

    Estimate what? Apps that run the battery the battery to 1% are the only way to actually and precisely measure the battery life of a device, since the batter capacity is at best an estimate and can be wildly different from the stated capacity depend on source and age. 'Current now' is a decent...
  14. D

    Post Official Linux kernel for Meizu Pro 5 smartphone (Exynos second wind)

    It's not even on the same SoC, so what are we expecting it to tell us? The 5433 files are pretty much the same as the stock Samsung ones - they even same the same typos in them.
  15. D

    Post [KERNEL][5.1.1][TW][N910C/H/U]Suemax-kernel-Lite-v1.7.8[FINAL][CUSTOM]

    There's an issue in the Perseus kernel code base that causes display input 'microlags' if WiFi is turned off. The issue has excised for at least 12 months, and is not new. This kernel is ultimately a fork of that codebase, so still suffers the same issues regarding the display lag. There have...
  16. D

    Post [KERNEL][5.1.1][TW][N910C/H/U]Suemax-kernel-Lite-v1.7.8[FINAL][CUSTOM]

    For the people getting touch screens lags. Is your WiFi turned on? If it's not, turn it on (doesn't matter if your connect to anything or not) and see if it stops.
  17. D

    Post [KERNEL][5.1.1][TW][N910C/H/U]Suemax-kernel-Lite-v1.7.8[FINAL][CUSTOM]

    Not to mention AnTuTu isn't a good benchmark to even use in the first place. The scores it gives are just comical. My note scores almost 85k for some bizarre reason.
  18. D

    Post [KERNEL][5.1.1][TW][N910C/H/U]Suemax-kernel-Lite-v1.7.8[FINAL][CUSTOM]

    To be honest, the undervolting isn't even the main problem. UV can cause crashes if it's too aggressive, but outside of that it's usual problem free. It's people adjust frequencies that cause big problems. Changing them has adverse effects on how HMP migration works. It's not just as simple as...
  19. D

    Post [KERNEL][5.1.1][TW][N910C/H/U]Suemax-kernel-Lite-v1.7.8[FINAL][CUSTOM]

    I don't run this kernel so it's likely my profile wouldn't work directly. Undervolt profiles aren't that useful really, as each SoC is different the levels of undervolting are different for every phone. It's better to slowly work down your own profiles till you find where your device becomes...
  20. D

    Post [KERNEL][5.1.1][TW][N910C/H/U]Suemax-kernel-Lite-v1.7.8[FINAL][CUSTOM]

    That doesn't make it more efficient than undervolting the lower A57 states. The perf/w at 1100MHz is still lower, and therefore less efficient. Not to mention you totally trash the HMP migrations because clamping CPU speeds will require HMP targets to be adjusted properly to accommodate the fact...
  21. D

    Post [KERNEL][5.1.1][TW][N910C/H/U]Suemax-kernel-Lite-v1.7.8[FINAL][CUSTOM]

    I'd be very interested to know why clamping A57 clock speeds to a minimum of 1100MHz is more efficient? From any angle I look at it that situation is only less efficient from the worse perf/w, to the big hole in the performance between the clusters.
  22. D

    Post [KERNEL][5.1.1][TW][N910C/H/U]Suemax-kernel-Lite-v1.7.8[FINAL][CUSTOM]

    I gave v1.2 a try today after using Perseus-Halaszk for quite a long time. First impressions seem quite good, but there are one or two little issues that have come up (running Wanam v3.2 on a 910C) Touch booster doesn't say disabled at all. The info says it might reset on low battery or power...
  23. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    As above, undervolting is safe except for stability loss and the small chance of software corruption (very rare). Undervolting is often a performance increase on modern phones too as the SoC won't throttle as quickly leading to increased performance. What your phone can take is really dependant...
  24. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    Mine with all 8 cores active gets down to ~120mAh. It will vary about because an idle phone is still doing plenty in the background, so it depends exactly what it's doing. Turning on small task packing will help a little with idle power. I didn't use it though when I got the above figure.
  25. D

    Post RAM Management on 5.1.1

    Chrome uses lots of ram, welcome to chrome :p Given on a desktop chrome can eat 10GB+ of ram once you have a decent number of tabs open, it's no surprise it murders the memory management on a phone. Personally for me I find the ram usage better on 5.1.1. I opened a tab in sbrowser 4 hours ago...
  26. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    TL;DR = Disable HMP power aware in Synapse. I've run some tests tonight against a stock N910C and narrowed down the issue. As expected the HMP settings are misconfigured causing a large performance loss. I suspect this comes from the fact this kernel is running Andrei's stock power aware...
  27. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    This isn't the WiFi touch lag, it's a serious performance regression. I'm going to try flashing something else later to get more scores on 5.1 to confirm if this is a kernel issue or not, but so far the browser benches are really bad. For example: Google octane v2: 4800 (should be 7800)...
  28. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    I've been testing it on 5.1.1 for a few days on my 910C. Generally everything is okay, but performance is a bit weird. General device speed is good, but browser based benchmarks are really slow, like half the speed (or even slower). This didn't happen with the 5.0 version of the kernel. Feel...
  29. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    Overclocking the A57's is probably a performance loss to be honest. It might help in some light burst loads like webpage loading, but for benchmarking it's almost certainly a negative as the SOC will throttle more quickly a snoted above. Perf/watt is lower and given there is a fixed load of work...
  30. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    If you know your voltages it's possible to roughly calculate it from information helpfully provided by Andrei before. For most people the 1.5GHz is going the sweet spot on the A53's, more so since the A53's have no temp sensor (so presumably no throttling) you don't want to run them too hard...
  31. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    The only thing to watch is the perf/watt of the A53's at those speeds is often worse than the A57s, meaning it's actually better to not overclock and to move to the big cluster. 1.5GHz with a decent undervolt is pretty much the sweetspot on the A53's, any higher can you risk just wasting power...
  32. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    I was playing the with governor settings earlier and noticed that you can't change the target load setting. Whenever the big cluster turns on the target load setting will default back to 75 no matter what it was set to before. If you stay on only the little cluster than it will hold whatever...
  33. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    My chip is ASV5 and I managed 150mv undervolt at the highest speeds on the A57s (1287.5mv to 1137.5mv @ 1900MHz) and 137.5mv at the lower speeds. A53s managed 100-112.5mv and the GPU 75-87.5mv. You have to remember that for lighter usage like web browsing the main power user is the screen, not...
  34. D

    Post [KERNEL] [N910C/H] [MM Stable] Perseus-Halaszk-universal5433 [06 May 2016]

    SSP = seamless sensor platform. It wakelocks because it's the step counting stuff for shealth, however because it's deep coded into the kernel it still happens regardless of if you use shealth or not and it cannot be stopped at a rom level (that I've discovered anyway!). It's a Samsung kernel...
  35. D

    Post [Q] Heat and CPU throttling

    The CPU's don't throttle very aggressively and work the clock speeds down fairly evenly so you won;t be able to manage it better yourself by underclocking. the GPU on the other hand stock uses some pretty crazy throttle steps which knocks the clock speed way down in one step which is probably...
  36. D

    Post Any custom roms out there that fixes memory leak on Note 4

    The only problem here is people are using applications that are counting the cache as in use. The issue is with the problem you are using, nothing more. The ONLY useful stats come from the stock android memory stats which truely report the memory situation. Seriously, anyone claiming well over...
  37. D

    Post Any custom roms out there that fixes memory leak on Note 4

    Where are you seeing that figure? If it's in the touchwiz memory manager than that's a misleading figure since it count cached applications in it's figure which isn't really used memory. You need to check the real Android memory stats to see the real usage.
  38. D

    Post Any custom roms out there that fixes memory leak on Note 4

    I've never seen my Note 4 even get to 2GB in use, ever. What are you running to get it so high?
  39. D

    Post Do you want the Perseus Kernel on your Note 4?

    That's being developed by a different person. It's a fork of this kernel, but not the same thing as continued development from Andrei.
  40. D

    Post Note 4 Battery Thread!!

    I use Wamanlite and Perseus kernel. I have a few things disabled like shealth and G+, but nothing that will have a real effect on battery life. To be honest the main ways to get long battery life are to avoid high screen brightness. Having to run at high brightness (like being outdoors) will...
  41. D

    Post Note 4 Battery Thread!!

    I've disable a few things that aren't needed but nothing really helps the wakelocks. When you check them with BBS you can see no application/service is linked to the wakelocks so therefore nothing you can disable will ever help to remove them. The only way I'm currently aware of is to block it...
  42. D

    Post Note 4 Battery Thread!!

    The SSP_sensor_wakelock is caused by the kernel, not a service. The only way to stop it is to edit a kernel with the wakelock removed. It's due to the shealth/step counting stuff. Every time the phone is moved the sensor hub wakes the device to report your step. This occurs regardless of...
  43. D

    Post Note 4 Battery Thread!!

    I don't find the 910c battery that bad. On an average day with mixed usage and fairly high brightness I can be on for around 5-6 hours SOT (normally finish my day at 50% with 2h30 - 3h0 SoT). However, if the brightness is allowed to be lower (around 25%) and web browsing with no standby then you...
  44. D

    Post 4500mAh batteries

    Do they come with a new back due to increased size? If they claim to be 4500mAh and fit in the stock size then I 100% promise you they are fake. I mean they fact it says 4500mAh in the body and 4200mAh in the image is already enough to tell you how good the seller is!
  45. D

    Post Do you want the Perseus Kernel on your Note 4?

    Does anyone else have a fast charging issue using this kernel? My devices always uses fast charging even though I have it disabled in the settings. Not sure if I've done something to upset it or if it's a bug! My Bluetooth doesn't work either but I know that's been mentioned already. Otherwise...
  46. D

    Post PcMark: Note 3:Lollipop vs Note 4

    Actually, if you look at the breakdown you can see the web browsing score increased by 15% and all the other tests stayed the same. Interesting. There is clearly a big.LITLLE penalty being paid in this benchmark, but this has been highlighted before on other workloads.
  47. D

    Post PcMark: Note 3:Lollipop vs Note 4

    Some of the spiking is due to big.LITTLE as the graph is spiking between full speed on the A57's and full speed on the A53's and sometimes lower speeds on the A53's. Changing clock speeds is a fast operation so unless it is actually lowering performance then it isn't much of an issue (or...
  48. D

    Post PcMark: Note 3:Lollipop vs Note 4

    You mean because the clock speed is rarely maxed? That's not really a big issue to be honest. A benchmark that fully loaded the CPU wouldn't tell you a lot because the device would be throttling within 10 seconds so wouldn't give you any indication of real usage at all.
  49. D

    Post PcMark: Note 3:Lollipop vs Note 4

    Sure. http://i101.photobucket.com/albums/m42/emeraldphoenix/2015-04-10%2020.34.55_zps0mmdd8b1.png~original
  50. D

    Post PcMark: Note 3:Lollipop vs Note 4

    Just tried this benchmark on my 910c. Got 5455 at stock clocks.