Page 2 of 6 FirstFirst 123456 LastLast
Results 21 to 40 of 118
Like Tree50Likes
  1. #21  
    Quote Originally Posted by geekpeter View Post
    find it a tad strange seeing benchmarks for differing kernels with different speeds, wouldn't it be best to have them all set at the same speeds to se if there's any difference at all.

    comparing a 1.2ghz stock to a 1.8ghz anything would ofc show that the stock sucks, same for a 1.5ghz {insert any kernel name here} etc.

    slap them all on the stocks ghz level and try again perhaps?
    They are all done by the same team, with much the same code.

    Things usually progress from a psychokernel to a warthog kernel to a uberkernel as the stability is found.

    What would be a better test is doing real-world-use benchmarks (including power consumption) with different governors.

    -- Rod
    WebOS Internals and Preware Founder and Developer
    You may wish to donate by Paypal to donations @ webos-internals.org if you find our work useful.
    All donations go back into development.
    www.webos-internals.org twitter.com/webosinternals facebook.com/webosinternals
  2. smatofu's Avatar
    Posts
    36 Posts
    Global Posts
    86 Global Posts
    #22  
    Thank you for the benchmarks.

    Unfortunately, overclocking does not increase real-life TouchPad speed (not that it needs a lot of improvement).

    Could you please provide some real-life performance numbers? For example, compare website loading/rendering time, or application loading etc.


    http://forums.precentral.net/hp-touc...pad-speed.html
  3. #23  
    Quote Originally Posted by smatofu View Post
    Unfortunately, overclocking does not increase real-life TouchPad speed (not that it needs a lot of improvement).

    Could you please provide some real-life performance numbers? For example, compare website loading/rendering time, or application loading etc.
    If you're confident that overclocking does not increase real-life speed, why would you want benchmarks?

    -- Rod
    WebOS Internals and Preware Founder and Developer
    You may wish to donate by Paypal to donations @ webos-internals.org if you find our work useful.
    All donations go back into development.
    www.webos-internals.org twitter.com/webosinternals facebook.com/webosinternals
  4. smatofu's Avatar
    Posts
    36 Posts
    Global Posts
    86 Global Posts
    #24  
    I am not confident of anything.

    On one hand, I love to tinker with electronics and I want to believe that improvement is always possible. I appreciate what the community does to improve TouchPad. Thank you!

    On the other hand, I am surprised by results of my testing.

    Maybe, I make a mistake somewhere in comparing regular and overclocked devices. That is why I have shared my findings.
  5. #25  
    Quote Originally Posted by GeekOutDog View Post
    How are you drawing any conclusion about stability and random reboots from looking at a bunch of charts with benchmarks that all completed successfully?

    Not to knock any of the hard work developing these kernels... it's just that I think that you've found a really strange way of interpreting the charts you just quoted.
    I think you should read his signature and just assume he knows what he's talking about...
  6. #26  
    Quote Originally Posted by smatofu View Post
    I am not confident of anything.

    On one hand, I love to tinker with electronics and I want to believe that improvement is always possible. I appreciate what the community does to improve TouchPad. Thank you!

    On the other hand, I am surprised by results of my testing.

    Maybe, I make a mistake somewhere in comparing regular and overclocked devices. That is why I have shared my findings.
    Have you turned off the logging?

    -- Rod
  7. smatofu's Avatar
    Posts
    36 Posts
    Global Posts
    86 Global Posts
    #27  
    Yes, I turned off the logging. I installed the same optimizing patches on both devices (from the article on zdnet.com). Govnah on my overclocked TouchPad shows that processor speed jumps to 1.7GHz once a while.
    Cnet.com takes about 33 seconds to fully load on both devices (regular and overclocked).
    Engadget.com takes about 27 seconds on both devices.

    (Just to compare with Samsung Galaxy Tab 7"
    Cnet.com - 29 seconds.
    Engadget.com - 20 seconds.
    Low-specs laptop over WiFi: both sites about 9 seconds - difficult to measure; subsequent requests much faster)

    Can someone share his/her numbers, please?
    Last edited by smatofu; 08/30/2011 at 03:28 AM.
  8. #28  
    I just shared my numbers on your other thread. You have to delete cache and cookies for each run so they are comparable from each other. I found web pages load much faster with ad blocker patch installed.
  9. funb0b's Avatar
    Posts
    36 Posts
    Global Posts
    39 Global Posts
    #29  
    It may be a silly question but where does one get these custom kernels from?

    I have been looking around and the only one I can find is the Uber kernal !

    Some pointers and links would be much appreciated :-)

    Cheers
    leghk likes this.
  10. #30  
    Go P4 Phantom, go!
  11. #31  
    Quote Originally Posted by zefyx View Post
    Go P4 Phantom, go!
    +1, but it is F4 Phantom......;-)
  12. #32  
    Quote Originally Posted by rwhitby View Post
    They are all done by the same team, with much the same code.

    Things usually progress from a psychokernel to a warthog kernel to a uberkernel as the stability is found.

    What would be a better test is doing real-world-use benchmarks (including power consumption) with different governors.

    -- Rod
    That's interesting. I'm new here, obviously, but I hadn't yet read anywhere that outlined this path of progression before. I understood the implication that uberkernel would be the most stable given that it is in the stable feeds, but from reading recent posts it has seemed as if warthog was more problematic than psychokernels.

    I suppose I've been lucky so far, because I've taken the risk of using f15c and now f4 and they "just work" for me with zero issues.
  13. #33  
    Quote Originally Posted by rwhitby View Post
    Have you turned off the logging?

    -- Rod
    Which all patches would everyone recommend. After we overclock to F15C or F4.. how much help does the install things like "Muffle System Logging" and "Remove Dropped Packet Logging" from Preware.

    And doing the ZDnet logging trick (How to improve the performance of your new HP TouchPad | ZDNet)

    any other patches to speed it up?
  14. #34  
    I would like to see psycho's kernels benchmarked at 1.728GHz too to provide a more apples to apples configuration.

    I do not have the same configs as he does so it would be interesting to explore the differences.
    Follow me on Twitter: http://twitter.com/warthogkernel

    Experimental warthog kernels for webOS devices. Patches such as Muffle System Logging, Max Blocker etc.

    Donations go towards further experimentation.
  15. #35  
    tdrag,

    Do the patches first. THEN overclock if you need more go-juice. Not vice versa.
    Follow me on Twitter: http://twitter.com/warthogkernel

    Experimental warthog kernels for webOS devices. Patches such as Muffle System Logging, Max Blocker etc.

    Donations go towards further experimentation.
  16. #36  
    Quote Originally Posted by ghostinator View Post

    As soon as i find a good stable voltage to run my touchpad at 1.944GHz, i will update these with results. Until then my kernel of choice is F15C @ 1.836 GHz and settings are as follows in govnah
    governor: Ondemand
    min freq: 192mhz
    max freq: 1.836GHz
    sampling rate: 0.01 Sec
    up threshold: 65%
    ignore nice load: false
    powersave bias: 0

    I/O Scheduler: NOOP
    TCP Congestion: VENO
    I have to say I set my TP to these settings and now it flies!!! Wow what a difference. These should be the default settings. Very close to the iPad 2 now!
  17. #37  
    Quote Originally Posted by codecrumb View Post
    tdrag,

    Do the patches first. THEN overclock if you need more go-juice. Not vice versa.
    Thx Code,

    What other patches are out there.. and where(if not from preware)

    look like I might have to redo mine.. and uninstall the overclock..
  18. #38  
    governor: Ondemand
    min freq: 192mhz
    max freq: 1.836GHz
    sampling rate: 0.01 Sec
    up threshold: 65%
    ignore nice load: false
    powersave bias: 0

    I/O Scheduler: NOOP
    TCP Congestion: VENO
    I copied those settings and saved as new profile, webosmark score of 1141.
  19. #39  
    Quote Originally Posted by peteostro View Post
    I have to say I set my TP to these settings and now it flies!!! Wow what a difference. These should be the default settings. Very close to the iPad 2 now!
    How is the battery life with those settings?
  20. evlinferno's Avatar
    Posts
    5 Posts
    Global Posts
    8 Global Posts
    #40  
    as soon as i find a good stable voltage to run my touchpad at 1.944GHz, i will update these with results. Until then my kernel of choice is F15C @ 1.836 GHz and settings are as follows in govnah
    governor: Ondemand
    min freq: 192mhz
    max freq: 1.836GHz
    sampling rate: 0.01 Sec
    up threshold: 65%
    ignore nice load: false
    powersave bias: 0

    I/O Scheduler: NOOP
    TCP Congestion: VENO
    when ever I try these setting and hit the apply button I get the following message " unable to open /sys/devices/system/cpu/cpu1/cpufreq/scaling_governor" and when I close govnah and reopen it the setting look like they worked, but the error worries me a little. So I was wondering if this is normal?
Page 2 of 6 FirstFirst 123456 LastLast

Posting Permissions