Froyo Quadrant scores, why is Humming bird doing so bad?


Last Updated:

  1. SamsungVibrant

    SamsungVibrant Well-Known Member This Topic's Starter

    Joined:
    Jul 23, 2010
    Messages:
    1,613
    Likes Received:
    165
    I may be mistaken, but Froyo was available to Captivate users in the leaked JI6 firmware, yes?

    Well I'm wondering what your Quadrant scores are, running Froyo JI6, on an otherwise stock Captivate, meaning no lagfix.

    So here is why I'm asking, I read somewhere, can't remember the link, that Froyo is having some issues with the Hummingbird. Basically it offers no speed-boost at all to the Captivate...Froyo was expected to dramatically improve CPU scored on Quadrant.

    This video also has me worried, this guy in the video ran quadrant and got a score of 975...I already get 925 on my vibrant stock, running 2.1. Froyo is suppose to give us scores like 1800 without lagfix, NOT 975!
    http://www.youtube.com/watch?v=toMEdPKlDnc

    Is it true Froyo is not bumping up the CPU performance in the hummingbird cpu? Quadrant scores were suppose to almost double, but it seems like Froyo may have a compatibility issue with Hummingbird. Anyone know details about this?

    Also, please try to understand my intention. I want to know if Froyo is delivering the CPU score boost that the others processors are seeing. I realize Quadrant scores can be boosted by LagFix, but again, I'm isolating the I/O lag issue from the CPU score. This is probably best if you run Quadrant pro...it breaks down the individual scores. Again, I'm saying it more than once, because someone is going to tell me to run lagfix to improve my score. Lagfix doesn't have anytihng to do with cpu performance, thats I/O...I'm trying to narrow down the reason behind why hummingbird is NOT getting a boost from Froyo.

    I was also reading a site, cant remember which, they showed the Hummingbird running on Froyo, and the quadrant scores were HORRIBLE, virtually no speed boost.

    We may have the fastest 2.1 processor, but if the hardware is not fully compatible with 2.2's full potential, then it means we have the slowest processor amongst 2.2 phones :(

    Thank you for your insights.

    Maybe you guys can please post your Quadrant score, running froyo on your stock captivate, meaning no lagfix, no overclock...etc.

    thanks!


    edit:
    another video
    this guy is running quadrant professional, look at the horrible blue cpu score. the only thing boosting the score so high is the I/O lag fix. But the cpu is doing sooo bad.
    http://www.youtube.com/watch?v=bYGwptNyy4Y

    Edit:
    According to a Znet article
    "It's clear that FroYo's JIT compiler currently only delivers significant performance gains for Snapdragon CPUs with the Scorpion core. This in turn explains why, so far, only a beta version of Android 2.2 is available for the Cortex-A8-based Samsung Galaxy S — the JIT compiler is the outstanding feature of FroYo. For the widespread Cortex-A8 cores, used in many high-end Android smartphones, the JIT compiler needs to be optimised. A Cortex-A8 core will still be slower than a Scorpion core at the same clock speed, but the Scorpion's advantage may not be as much 260 percent."

    Hummingbird CPU is unable to utilize the benefits of the JIT Compiler, which is the main selling point of 2.2
    http://www.youtube.com/watch?v=yAZYSVr2Bhc


    Hopefully this changes with an official release, however Samsung Galaxy Tablets are already running official Froyo 2.2, and their cpu scores are horrible also.
     

    Advertisement
  2. CrimsonPride

    CrimsonPride Well-Known Member

    Joined:
    Jul 28, 2010
    Messages:
    1,482
    Likes Received:
    200
    well it is not an official release and ive heard its still somewhat glitchy, so its definitely not nearly up to its full potential

    also i hear quadrant was designed around snapdragon processors and also quadrant scores can easily be manipulated.

    but at any rate i wouldnt worry about it too much. i dont have the leaked froyo so i cant say much from experience, just what ive heard
     
  3. tylerdurdin

    tylerdurdin Well-Known Member

    Joined:
    Aug 28, 2010
    Messages:
    402
    Likes Received:
    63
    Quadrant is useful in some respects, like showing improvements on your own devices configuration, you really cant use it to compare phones. Having said that we are most likely not going to see 1800 on quadrant with a stock froyo. The Incredible runs 1075-1150 on Froyo you may see a little better than that, but for the most part you will just see a all around smoother and faster processing phone. Not to mention that if quadrant is designed around the snapdragon, it leaves some heavy question why a lagfixed hummingbird runs the fastest quadrant hands down, My captivate ran 2350 consistently, before I installed froyo, I have only been able to run 975 but it has been all around smoother than eclair any day.
     
  4. SamsungVibrant

    SamsungVibrant Well-Known Member This Topic's Starter

    Joined:
    Jul 23, 2010
    Messages:
    1,613
    Likes Received:
    165
    I think people are going to confuse I/O and lagfix with CPU. Quadrant uses several prameters, one of which is CPU performance, another is I/O.

    I'm strictly talking about CPU performance. Quadrant PRO shows the breakdown, the free one doesn't.

    I also don't believe that quadrant was built for snapdragon. Who says this? The developer? If that was the case, Hummingbird wouldn't outperform Snapdragon on quadrant when running 2.1.

    From what I'm reading around the net, there is some kind of incompatibility between Froyo and Hummingbird. Hummingbird fails to utilize the full potential of Froyo.

    Instead of using the argument that Quadrant was designed for Snapdragon cpu's. What about the argument, that Froyo 2.2 was designed to increase cpu performance on Snapdragon, but not Humminbird.
     
  5. tylerdurdin

    tylerdurdin Well-Known Member

    Joined:
    Aug 28, 2010
    Messages:
    402
    Likes Received:
    63
    I agree with this completely, and where did the rumor start that this was designed for a snapdragon :confused:, I have heard this for quite a while. I do have to say watching the I/O on Froyo is sickening after running 2.1 w/OCLF. I think I was most impressed with the fact that I could run quadrant in about a minute, now it takes a minute just to get throgh the I/O. I do hope that the guys over at XDA figure out a way to get 2.2 optimized for the Hummingbird, because at this point it looks like we are getting jerked again on the ram from froyo.:(
     
  6. CASUAL

    CASUAL Well-Known Member

    Joined:
    Sep 10, 2010
    Messages:
    140
    Likes Received:
    5

    what are ypu guys talking about?
    how do you check the numbers?
    i have 2.1 eclair and my phone is stock and unrooted
     
  7. SamsungVibrant

    SamsungVibrant Well-Known Member This Topic's Starter

    Joined:
    Jul 23, 2010
    Messages:
    1,613
    Likes Received:
    165
    we already know the 2.1 scores, we are trying to figure out why froyo 2.2 is not providing the hummingbird cpu any performance boost.

    one of the selling points of froyo 2.2 is the supposed boost to cpu performance. The humminbird isn't gaining any performance boost.


    if you are confused about what quadrant is, its a benchmark program. go to market and search quadrant, and you will see.
     
  8. CASUAL

    CASUAL Well-Known Member

    Joined:
    Sep 10, 2010
    Messages:
    140
    Likes Received:
    5
    I got "824" on a stock un rooted captivate..."tell me doc...how bad is it? Is it fin rot?!?!?!? Please tell me it isn't fin rot!!!!!"
    (futurama dr. zoidberg quote lol)

    Seriously though how bad is 824 on a stock 2.1 captivate?
     
  9. CASUAL

    CASUAL Well-Known Member

    Joined:
    Sep 10, 2010
    Messages:
    140
    Likes Received:
    5
    The first time I ran the test I had a bunch of background apps running and the music player was on.

    I turned off the extra stuff like fancy widget and batts start etc..
    The second time I got 869 and the third time I got 857.

    Still pretty bad right? it didn't even hit nine hundred


    By the way I love your signature. That's how I am I love animals and I treat them with respect. And I don't judge people I accept people as they are. differences aside Deep down we are all the same.
     
  10. jflexe99

    jflexe99 Well-Known Member

    Joined:
    Aug 20, 2010
    Messages:
    171
    Likes Received:
    17
    I got 910 on leaked froyo.
    With Cog 2.2 ROM, rooted, ext2 files, lagfix...I got mid 1800s...
    But my Linpack results went from mids 8s to high 12s- high 13s
    Haha, I keep running benchmarks in the Linpack...keep getting low 14s
     
  11. Infinite-t

    Infinite-t Well-Known Member

    Joined:
    Jul 23, 2010
    Messages:
    412
    Likes Received:
    74
    quadrant schmadrant,

    In the computer world people have been using synthetic benchmarks to measure epeens for years. SOme are more accurate that others but in the long run the only thing they are really good for is measuring against yourself. Since every system is going to be slightly different. In some tests, with 2 different procs (AMD and Intel for example) with the same exact specs, intels will out perform AMDs and in other tests is the other way around. The same applies for Video cards, an Nvidia and an ATI with the same specs with out-perform eachother in different tests... the same goes for phone Procs. It's all about how the different Procs handle different types of data.
    1+1+1+1=4, 2+2=4, 3+1=4, 5-1=4, 2+1+1=4.... you get the idea.

    But, when it comes to it, how does it perform in your particular system, or device. I have to say that my phone stock was slow and laggy and annoying. I did the OCLF and used launcherpro and the lag was gone. I am now running the Cognition 2.2 without any lag fixes of any kind and this thing smokes. I havent tried benchmarking it, and I dont really care. It performs smoother and faster, no lag or stutters, way better than 2.1 with the OCLF.

    If your only concern is the highest possible quadrent number, then stay 2.1 with the OCLF v2, you'll get 2200-2300, just dont upgrade to froyo. If you want great performance, plus the great new features of Froyo, then update and enjoy. For me anyway, its about user experience, not some numbers..
     
    jdCaptivate and s.m.knipe like this.
  12. mttfrog13

    mttfrog13 Well-Known Member

    Joined:
    May 5, 2010
    Messages:
    488
    Likes Received:
    64
    So basically there is no lag fix in froyo. No (dependable)gps fix on the last ota update. This goes to show how both problems were hardware related.
     
  13. SamsungVibrant

    SamsungVibrant Well-Known Member This Topic's Starter

    Joined:
    Jul 23, 2010
    Messages:
    1,613
    Likes Received:
    165
    My concern is not the highest possible quadrant number, hence stressing CPU and NOT I/O in my original post. Did you read that part?

    I think you seem to be missing my point. Froyo is suppose to increase cpu performance, however this performance is not reflected in Quadrant. Assuming Quadrant is accurate, then this is reflective that Froyo does not bring a performance boost to Hummingbird CPU's for some reason.

    Unless you can provide some solid proof that Quadrant Professional edition is in some way faulty, then until then I'll assume Quadrant is accurate. Especially since Quadrant accurately showed that Galaxy S was the fastest phone on 2.1.

    So back to my original question. Why is Froyo NOT providing Hummingbird the CPU boost that Froyo advertises as one of its main selling points?
     
    tylerdurdin likes this.
  14. Snow_Fox

    Snow_Fox Well-Known Member

    Joined:
    Mar 31, 2010
    Messages:
    1,498
    Likes Received:
    131
    Well.. its my understanding of the OS that google creates update.

    Its up to manufacturers to tailor the code to their specific devices.

    not trying to belittle the enthusiast crowd here... But, there is always the distinct chance that samsung has more optimized code for their version of 2.1 for their hummingbird processor.

    I mean, I could be wrong there.. I know android as a whole is open source and different companies "skin it" different ways.

    However, I would imagine there might be some low level code somewhere for the hummingbird that would have to be changed and updated with 2.2

    I mean horrible example.. but, I had ATI catalyst control center because, my onboard GPU was an ATI.

    I upgraded to an 8800 before uninstalling ATI drivers.. didn't cause to many major problems during the 20 minutes I was toying around looking for drivers..

    But, had I actually tried to play a game i'd have suffered frame rate issue's, performance problems ect.

    So it could just be a froyo driver issue.
     
  15. SamsungVibrant

    SamsungVibrant Well-Known Member This Topic's Starter

    Joined:
    Jul 23, 2010
    Messages:
    1,613
    Likes Received:
    165
    It appears the problem may have something to do with Hummingbird not being able to fully utilize the Jit Compiler or something like that in 2.2. Fingers are pointing to no significant cpu performance boost for Galaxy S phones, which was one of the top selling points of Froyo. Cpu performance boost and flash, 1 out of 2? I rather have the cpu boost.
     
  16. rubiconjp

    rubiconjp Well-Known Member

    Joined:
    Sep 20, 2010
    Messages:
    363
    Likes Received:
    40
    Samsung tweeted a while back that we should expect significant performance gains on 2.2, so this is a bit deflating...
     
  17. Infinite-t

    Infinite-t Well-Known Member

    Joined:
    Jul 23, 2010
    Messages:
    412
    Likes Received:
    74
    Im not saying you're wrong at all. Froyo does not seem to give much of a boost to your quadrant score. I Even tried it myself. My quadrant score running froyo was only roughly 100 points higher.

    According to some testing done by some guys at XDA froyo did increase quadrant score by a few hundred. Linpack performance nearly doubled. It's still mot the greatest measure of performance, but testing the actual speed in which the processor performed opperations is probably a better way for testing a CPUs performance.
    FAQ - Cognition Wiki
    (Bottom of that page)

    LINPACK - Wikipedia, the free encyclopedia (What linpack does)

    From me personal user experience running froyo compaired to 2.1 and 2.1 with OCLF.

    2.1: (~840 Quadrant) Annoyingly slow. UI lag was very distracting. Skipping and stuttering during scrolling. Launching the SAME apps as my iP3g took a few seconds LONGER on the Capt. Music skipped during playback... it goes on.

    2.1 with OCLF and Launcher Pro (~2300 Quadrant) UI lag gone. Switching and launching apps was fast. Scrolling was smooth, Music playback was smooth and switching tracks was nearly instant. Boot time and Media scan time reduced.

    2.2 (Cog. 2.2 Beta 5.5) No lag fixes with launcher pro (~960 quadrant)
    Very fast. Switching and launching apps nearly instant, Web pages load faster, no music playback shutter, even while playing music while gaming. App market is very quick, media scan and phone boot up/down much quicker. My user experience running this version of froyo appears to be offering nearly seemless performance, even more so than 2.1 with OCLF and a 2300 quadrant score.

    So my insight on this is that although benchmark performance numbers do not seem to be improved much, the overall performance in executing daily tasks and performing under actual usage seems to be MUCH improved. In fact, this leaked Froyo actually saved me from ditching my phone. This was my first android phone and I am very impressed with the Android OS. But the bugs and issues tha I Personally was experiencing with this (my 3rd by the way) Captivate was sub par compaired to other devices out there around the same price. I was ready to unlock the Capt and sell it on Ebay and get a G2 and unlock it and go on my marry way. But then this 2.2 popped up just in time. I figured, what the hell, and gave it a try, I an now truely impressed with this phone and what it can do. Even if the quadrant score were cut in HALF by Froyo, i would not care as long as it performed the way it is performing now.

    All I am saying, Is based on my experience (and seemingly many (but not all)) others that are running Froyo Roms, even though quadrant score seem unimproved, every day usage performance has improved greatly. I am fairly confident that most of you, when you start running Froyo will see what I am saying and why. Dont worry about the numbers, worry about every day user experience.

    Im more concerned that so far Froyo still doesnt seem to be offering Voice command or blue tooth voice command... that to me is a real bummer... I hope AT&T didnt kill it because they want to sell their voice dial addon service...
     
    mttfrog13 likes this.
  18. pqrs

    pqrs Member

    Joined:
    Sep 26, 2010
    Messages:
    23
    Likes Received:
    0
    ive had a different experience than most in this thread... im running cog 5.5 beta and its way slower than 2.1 with oclf? seems about like the stock 2.1 to me? im about to dump 2.2 and go back to 2.1, its WAY faster to me? nothing lagged when i was running the oclf, it was smokin fast! 2.2 seems to be a trurd.. i noticed it was way slower withing 1 min of booting up 2.2, the media scanner runs way longer now. lags all the time, pages load slow.. its a dog. i cant believe others say this is fast? its a lot slower on my phone, theres no doubt about it.
    i may go back to 2.1 with oclf tonight because 2.2 annoyed me a number of times today!
    ohh.. stock 2.1 quadrant scores 918, oclf 2330, 2.2 scores 984.
     
  19. Infinite-t

    Infinite-t Well-Known Member

    Joined:
    Jul 23, 2010
    Messages:
    412
    Likes Received:
    74
    Sorry to hear it didnt work out for ya. thats wierd. I guess that just the way it is with Beta ROMs, works for some and not for others... w8 didnt jh7 work for some and not others? ;)

    Anyway, your quadrant score was actually higher (albiet marginally) with Froyo, but your user performance was much lower.. How can that be? I thought that higher quadrant score automatically ment your phone is faster?
     
  20. SamsungVibrant

    SamsungVibrant Well-Known Member This Topic's Starter

    Joined:
    Jul 23, 2010
    Messages:
    1,613
    Likes Received:
    165
    This is looking bad, I hope its just because its beta. But the CPU score is horrible. This guy is running quadrant professional, the blue score is cpu, I think the green score is the i/o lagfix. But we dont care about I/O right now, the focus here is Froyo 2.2 and the significant boost it was suppose to bring to the CPU.

    http://www.youtube.com/watch?v=bYGwptNyy4Y
     
  21. SamsungVibrant

    SamsungVibrant Well-Known Member This Topic's Starter

    Joined:
    Jul 23, 2010
    Messages:
    1,613
    Likes Received:
    165
    From a znet article:

    "It's clear that FroYo's JIT compiler currently only delivers significant performance gains for Snapdragon CPUs with the Scorpion core. This in turn explains why, so far, only a beta version of Android 2.2 is available for the Cortex-A8-based Samsung Galaxy S — the JIT compiler is the outstanding feature of FroYo. For the widespread Cortex-A8 cores, used in many high-end Android smartphones, the JIT compiler needs to be optimised. A Cortex-A8 core will still be slower than a Scorpion core at the same clock speed, but the Scorpion's advantage may not be as much 260 percent."


    I hope this is due to beta stage only. However, the galaxy s tablet is running official 2.2, and it's quadrant cpu scores are just as horrible.
     
  22. CaCHooKa Man

    CaCHooKa Man Well-Known Member

    Joined:
    Jul 19, 2010
    Messages:
    148
    Likes Received:
    10
    my cpu scores are almost twice as high as they were on 2.1

    2.1 with JH2, rooted
    [​IMG]

    2.1 with JM6, with lag fix and overclocked to 1.2 ghz
    [​IMG]

    cognition 2.2 beta 5, no lag fix
    [​IMG]
     
  23. Snow_Fox

    Snow_Fox Well-Known Member

    Joined:
    Mar 31, 2010
    Messages:
    1,498
    Likes Received:
    131
    I suppose the question then becomes, can the galaxy s be properly calibrated to get any similar performance gains to snapdragon based phones.

    I mean, I am honestly still interested in a s tab either way. I think the captivate is the one I want and its probably not going to change.. I have heard in the past captivate has better battery life than other high end android phones.. And for what its worth its still the only one really worth a lot of consideration on at&t to me.

    With all that said, if I had a captivate I think I would be mad the phones I was looking at before the captivate were handy capped.

    We really do live in an interesting age when you can get a phone, and it can up up in performance drastically (snapdragon ect) or you can get an update and it can kill aging hardware (iphone 3g).

    For what its worth, that was not bashing the iphone 3g... but, it was just interesting things I noticed.

    I hope that as phones become more powerful, we reach a point where you don't void your warranty for putting whatever operating system you want on it..

    If I buy a netbook and feel the newest ubuntu os is more optimized for it, I don't void warranties switching to ubuntu. I may lose the ability for tech support toe help me if anything goes wrong.. but, I don't flat out void my warranty.

    While i do realize not everyone out there builds their own pc and writes code so the companies can't deal with every moron bricking their phone... But, I see no reason end users should be punished for hardware failure on a phone regardless of what phone software they use.

    faulty components are faulty components, its a fact of life it happens..

    Then again.. thats why I buy my parts and build it myself.. if I get a faulty hard drive.. I deal with hard drive manufacturer.. not whatever company brands the machine..
     
  24. qwertyaas

    qwertyaas Well-Known Member

    Joined:
    Aug 8, 2010
    Messages:
    373
    Likes Received:
    20
    Reminds me of 3dmark scores between amd and intel; ati and nvidia. Benchmarks are never the be all end all with regard to performance levels.

    I used to always bench my pc when I got a new card or cpu until I realized it's a waste of time. The score doesn't always mimic the performance.
     
    Infinite-t likes this.
  25. ChaoticPurity

    ChaoticPurity Member

    Joined:
    Sep 1, 2010
    Messages:
    19
    Likes Received:
    3
    Honestly, I've seen no need to question the legitimacy of the boosted speeds in Froyo. Yes, having high scores is nice, but why fix what isn't broken? In my experience, Froyo stock is far better than anything on eclair... ever. I was running Eclar + Voodo + OC, and yes, I did get 2000+ on my quadrant score versus the 900 or so on my stock, non-lagfixed Froyo, but Froyo is just as, if not faster than the modified version of eclair I was using.

    These are simply numbers, if your phone runs fast, what does the non-reflective "boosted" CPU performance matter? I've seen people freak out, before they even use the phone saying "OH MY GOD I NEED LAGFIX" and can't pinpoint ANY part where there was lag for their uses. I did experience lag, and the modifications fixed it, but I am completely content with how Froyo is running currently. (Cog 2.2 Beta6)
     

Share This Page

Loading...