Related
So I've gotten anywhere between 2.5 to 5.1 MFLOPS using various ROMS and have yet to be able to notice something incredibly different.
710...768...806 - What does it matter? What program other than Linpack shows a sizable difference? Sure, maybe things open quicker? What am I missing here?
I read all this about achieving high MFLOPS and OC Kernels yet I still can't achieve smooth game play on 16 bit emulator on my phone with 5 MFLOPS.
MFLOPS mean jack when there is little way to observe the difference.
Carreno43 said:
So I've gotten anywhere between 2.5 to 5.1 MFLOPS using various ROMS and have yet to be able to notice something incredibly different.
710...768...806 - What does it matter? What program other than Linpack shows a sizable difference? Sure, maybe things open quicker? What am I missing here?
I read all this about achieving high MFLOPS and OC Kernels yet I still can't achieve smooth game play on 16 bit emulator on my phone with 5 MFLOPS.
MFLOPS mean jack when there is little way to observe the difference.
Click to expand...
Click to collapse
Linpack MFLOPS - measures the floating point performance of your phone.
710...768...806 - refers to CPU frequencies
increasing the CPU frequency should equate to better general-case performance, including things opening quicker as you mention, but also other types of general snappiness like moving between screens and so forth.
"I read all this about achieving high MFLOPS and OC Kernels yet I still can't achieve smooth game play on 16 bit emulator on my phone with 5 MFLOPS." - This may have less to do with the performance of your phone and more to do with the emulator itself. Emulation is a surprisingly CPU intensive operation, especially if the emulater isn't well written. Rather than looking a ton into overclocking and JIT, etc, maybe you ought to look for a better piece of software.
Yea,
I've tested most emulators. Wish there was an Atari emulator!
Thanks for the response.
Carreno43 said:
Yea,
I've tested most emulators. Wish there was an Atari emulator!
Thanks for the response.
Click to expand...
Click to collapse
I have run roms with 5.1 MFLOPS and now am running a rom that gets 3. I can honestly say I see no difference.
Spencer_Moore said:
I have run roms with 5.1 MFLOPS and now am running a rom that gets 3. I can honestly say I see no difference.
Click to expand...
Click to collapse
I can see a difference... in battery life! Lolz
g00gl3 said:
I can see a difference... in battery life! Lolz
Click to expand...
Click to collapse
Haha Awesome
it looks like to me that everyone is look at the wrong things.
for example:
I am running a Tom that is getting on a average of 4.9 mflops.
I get smoother screen changes....
streaming videos online is so much faster compared to a 3.0 mflop rom. ...
tubetube and other....... websites.
to me everything I do is faster...
I.don't play game on my phone so I don't know how that is.... but everythng else I do is very much faster.
I love high mflop roms...
I have notice about mflops is that it matters about the kernal that u use.
Isn't it true that the MSM7201 in our phones is already overclocked to get to 528mhz as it is? I see a lot of different places saying Qualcomm chips in general are just not worth overclocking... and since our chip is factory overclocked to begin with... just seems like we're pushing the already-pushed here. But the way this board goes crazy for overclocking... it's contradictory. I don't know what to think, cause I've run Linpack myself and gotten ~4.9 with JIT + OC versus ~2.5 without... but I'm with the OP on this one... only difference I'm seeing is my battery draining faster and my phone getting physically hotter.
xatch said:
Isn't it true that the MSM7201 in our phones is already overclocked to get to 528mhz as it is? I see a lot of different places saying Qualcomm chips in general are just not worth overclocking... and since our chip is factory overclocked to begin with... just seems like we're pushing the already-pushed here. But the way this board goes crazy for overclocking... it's contradictory. I don't know what to think, cause I've run Linpack myself and gotten ~4.9 with JIT + OC versus ~2.5 without... but I'm with the OP on this one... only difference I'm seeing is my battery draining faster and my phone getting physically hotter.
Click to expand...
Click to collapse
I have OC and JIT and getting about 5.1 mflops and haven't had worse battery life or a hotter phone. It could be the battery I'm using but meh (got a replacement one that's 2000 mAh) but I got worse battery life on leak 2.1 than with the rom I'm using now that has OC, JIT, LWP, etc. I can go about 8 hours with heavy texting, moderate internet usage and my lwp's running and it only goes to about 65%
so OC and Jit don't make that big of a difference in gameplay?
Sent from my Eris using the XDA mobile application powered by Tapatalk
What the OP and all the respondents are noting is frankly quite typical of what happens when performance tuning focuses on a single benchmark: the results obtained are essentially meaningless for different kinds of activities on the same device.
That's because there's a whole chain of dependencies that are specific to a given task, any number of which could become the rate-limiting factor; and a different task on the device will have a different set of dependencies and therefore different rate-limiting behaviors.
For instance, let's take writing to an SD card as an example: there's really no way that OC'ing will speed that up in a measurable way - because the CPU isn't the rate limiting factor.
That Linpack benchmark measures floating-point performance using a software library (as the Eris has no hardware FP capability). Most of the apps on the phone do very little FP work at all. But, it's not a bad test of CPU speed, because it performs no I/O. It also may not be very memory bandwidth intensive, either (if the problems it works on stays in the uP cache and there are few page faults).
OTOH, a game emulator needs to write to the graphics display (at a minimum) and possibly also do read I/O from flash.
Different task, different results. Sometimes things can be improved by hardware or firmware; sometimes the software itself needs to be improved.
bftb0
im sorry, but could you just answer in plain english
Sent from my Eris using the XDA mobile application powered by Tapatalk
TheSonicEmerald said:
im sorry, but could you just answer in plain english
Sent from my Eris using the XDA mobile application powered by Tapatalk
Click to expand...
Click to collapse
Lima beans bad.
Pork good.
Slow phone bad.
Fast phone good.
bftb0
Thanks for my laugh of the day on that one.
What I'm trying to get at is -
I should be able to play, at the basic level, Sonic or Mario - Without issues.
At the very least
I prefer roms over market games any day (Sonic, Mario, Zelda, DK-Country) and it cripples the phone, at least in my view, that I cannot enjoy the fruits of old games.
Although, I was able to find some old Atari games - which, thankfully, work without stuttering.
Hi,
just managed to install and configure Kweek3 (an Android compiled Quake 3) and it is running. Unfortunately I don't have a ext. Keyboard and Mouse. That's why I am playing with the volume buttons to walk forward/backward (Interestingly, kweek3 tinks that these are F2 and F3 Button )
Nevertheless, I am wondering if there is something wrong. I am only getting around 6-9 fps out of my streak. Should the hardware not being able to do more? Or do I really have to wait for Froyo (or something in that direction) that I get it smoothly?
How high are your fps?
I think its best to wait for Eclair. Check out the benchmark results for yourself:
http://www.streaksmart.com/2010/08/dell-streak-benchmark-scores-greatly-improved-by-android-21.html
Some of the FPS readings for Eclair are not much more than on Donut
Could you give us some download links, or is it...you know... wink wink nudge nudge?
has anyone else tried linpack or quadrant or the other benchmarks mentioned.
in quadrant I get 7.19 much higher than the supposed donut score posted by the endgadet bod..
what we really need is the jit and the qualcomm 3d stuff enabled in a kernel..
fards said:
has anyone else tried linpack or quadrant or the other benchmarks mentioned.
in quadrant I get 7.19 much higher than the supposed donut score posted by the endgadet bod..
what we really need is the jit and the qualcomm 3d stuff enabled in a kernel..
Click to expand...
Click to collapse
+1 in quadrant i get 676(2nd run i got 666 xD) 1.6 unrooted at that as well kind of made me wtf at the engadget benchmarks, though linpack wise its pretty much the same.
Just google for kweek3. Then you would find a link at google incl how to install it...legally
Sent from my Dell Streak using XDA App
New build
The new build of kwaak 3 has joystick support and works on the Streak, check it out.
http://code.google.com/p/kwaak3/downloads/list
Does anyone feel like posting their benchmarks on the custom ROMs they are running? I know not every GTab will run the same but I would like to get a basic idea of what the hottest scores are. Thanks.
TNT Lite 4.3.x with stock kernel is ~40 fps (Nenamark) and ~2300 in Quadrant. On average.
There are definitely higher scores than that, with some of the other modded ROMs.
2200 with vegan 5.11 Benchmark Atuntu
Thank you guys for posting your benches. I had seen some one some where, posting that they get 60fps. Any one getting 60fps?
there is guide in dev forum where you can get up to 77fps, but the recomended is under 75.
You might mean the frame rate. The default is 50, but TNT Lite and gADAM up the frame rate to 60. At least one mod ups it to 77.
That's not really a benchmark, though - I assumed you meant something like Quadrant.
Im running caulkins rom and clemsyn's kernel #6. I'm averaging around 2400 on quadrant and 3000 on smartbench. The only thing faster on smartbench is overclocked xoom
It looks like most of the new ROMs run pretty fast. If i can get 2200-2500 in Quadrant I will be happy.
I get a little over 3000 qudrant
johnboyjr said:
I get a little over 3000 qudrant
Click to expand...
Click to collapse
what ROM/mods you are running?
i got several scores in the 2400s today and one personal best of 2518.
@johnboyjr
Are you bull****'n? 3000? i havent seen anyone say they are hitting the 3000 mark. I'd like to know what you're running too. Did you overclock? I dont think i want to overclock mine.
nsmartinez79 said:
Im running caulkins rom and clemsyn's kernel #6. I'm averaging around 2400 on quadrant and 3000 on smartbench. The only thing faster on smartbench is overclocked xoom
Click to expand...
Click to collapse
I have the same setup on my gtab... got 39.1 fps in nenamark, and 2399 in quadrant.
I just set my gtab up today so it doesn't have much on it yet.
There is a post in the development section by chengu that has a loopback fix that enhances read/write speeds on the internal SD. This makes the I/O part of quadrant way faster. I have posted on that thread vegan stock 10x average is 2500+. With the loopback fix its a 10x average of 3100 with peaks of 3333. So no its not BS but really how much do quadrant (or any other test) effect daily functionality.? It is great for bragging rights though, especially to xoomers.
Sent from my Chromatic Magic using XDA Premium App
Mantara said:
There is a post in the development section by chengu that has a loopback fix that enhances read/write speeds on the internal SD. This makes the I/O part of quadrant way faster. I have posted on that thread vegan stock 10x average is 2500+. With the loopback fix its a 10x average of 3100 with peaks of 3333. So no its not BS but really how much do quadrant (or any other test) effect daily functionality.? It is great for bragging rights though, especially to xoomers.
Sent from my Chromatic Magic using XDA Premium App
Click to expand...
Click to collapse
Mantara, do you still have the link for loopback fix? I could not find it .
Thanks a lot in advance.
njshkb said:
Mantara, do you still have the link for loopback fix? I could not find it .
Thanks a lot in advance.
Click to expand...
Click to collapse
http://forum.xda-developers.com/showthread.php?t=902426
Fyi: restart your tab 5 or six times, I know I sounds weird but it takes that many to streamline it completely. Let it stabilize after boot and then power down, wait 20 sectonds and and power on. It wil take up to six minutes on first boot and get less each time.
It uses boot time to write changes.
Have fun ;-)
loopback fix like others have said
Is there any down fall to using the loop back fix? Heat issues or failure? Why isnt anyone implementing this into their ROMs?
No adverse side effects that I have noticed. Most people have noticed a slight increese in playing video from internal memory and web browsing cashed sights. Read the thread its full of good info.
Sent from my Chromatic Magic using XDA Premium App
Any chance or hope we can get a CyanogenMod for the aria that comes with flash "lite" or could we install it ourselves on "CyanogenMod 7 for Aria"?.
I am trying to squeeze the most performance from my aria, and would like to give CyanogenMod a try over my current rom "Liberated 2.2.2 oc 806mhz". But i do not want to loose the ability to watch youtube or surf the web with 'flash'. My games are a little choppy but certanly much better over "stock at&t rom at 600mhz" so i would like to give "CyanogenMod gingerbread 2.3 oc 806mhz" a try.
As of now i OC my aria to 864mhz during play time, and underclock for normal use..But of course i am weary of freeze up. So far no problems and games run pretty good. (Backbreaker, Driftmania, and Motorbike Gp are my benchmark games). I also have hardware acceleration enabled through blade buddy(good or bad?)
L4mishotiz said:
Any chance or hope we can get a CyanogenMod for the aria that comes with flash "lite" or could we install it ourselves on "CyanogenMod 7 for Aria"?.
I am trying to squeeze the most performance from my aria, and would like to give CyanogenMod a try over my current rom "Liberated 2.2.2 oc 806mhz". But i do not want to loose the ability to watch youtube or surf the web with 'flash'. My games are a little choppy but certanly much better over "stock at&t rom at 600mhz" so i would like to give "CyanogenMod gingerbread 2.3 oc 806mhz" a try.
As of now i OC my aria to 864mhz during play time, and underclock for normal use..But of course i am weary of freeze up. So far no problems and games run pretty good. (Backbreaker, Driftmania, and Motorbike Gp are my benchmark games). I also have hardware acceleration enabled through blade buddy(good or bad?)
Click to expand...
Click to collapse
'Flash Lite' is a HTC proprietary application that runs on HTC Sense, so there is no chance that it will be ported to CM7.
You can still watch most YouTube videos, either through the YouTube app or through the mobile website, because it actually doesn't use Flash to deliver that content. And as far as I know, Flash Lite doesn't even work with the full YouTube site anyway, unless I'm mistaken. (Been forever since I used Flash Lite.)
drumist said:
You can still watch most YouTube videos, either through the YouTube app or through the mobile website, because it actually doesn't use Flash to deliver that content. And as far as I know, Flash Lite doesn't even work with the full YouTube site anyway, unless I'm mistaken. (Been forever since I used Flash Lite.)
Click to expand...
Click to collapse
So iam guessing an android friendly html. In that case what rom would squeeze the best gaming performance. my current set up or another. and does enabling hardware acceleration make a differance
L4mishotiz said:
So iam guessing an android friendly html. In that case what rom would squeeze the best gaming performance. my current set up or another. and does enabling hardware acceleration make a differance
Click to expand...
Click to collapse
The aria may not be tgr ideal gaming platform but so far everygame I've tried playing has worked for me. I'm currently running cm7. I keep my phone at 320min/768 max ondemand for the most part. I use tasker to automatically run some games at higher clock speeds and with the performance governor when I open them (major battery drain if the phones not plugged in to charge though). Vm heap size can also be raised for intensive games. I also keep surface dithering off which helps. Of course flash based games are a no-go on cm7. Not sure if the flash lite liberated roms will allow you to play flash games either though.
Sent from my cm7 Aria.
"Vm heap size can also be raised for intensive games."
what would be a good value?
L4mishotiz said:
"Vm heap size can also be raised for intensive games."
what would be a good value?
Click to expand...
Click to collapse
I've found 32 to be sufficient for the most part. 48 is even better but it tends to make the phone a little less responsive overall. Its unfortunate that it can't be adjusted without a reboot so you could just crank it up for gaming and bring it back down for regular use.
Sent from my cm7 Aria.
I have it now at 32. Need for speed shift runs well i am very happy about that!
I was trying to find a way to lower the resolution on some "HD" games but have not found anything. as you all know those games need more screen that the aria can offer.
L4mishotiz said:
I have it now at 32. Need for speed shift runs well i am very happy about that!
I was trying to find a way to lower the resolution on some "HD" games but have not found anything. as you all know those games need more screen that the aria can offer.
Click to expand...
Click to collapse
There's an app available that raises the pixel density of the screen. That might be just the ticket for such games. Its called something along the lines of 'Pixel Density Changer'. I don't recall if I saw it n the market or here on xda. I'm sure a quick search will turn it up.
Sent from my cm7 Aria.
Ok Thank you @CallMeAria.. Just to make sure with cm7 are you able to watch youtube videos, or web content using xscope(streaming videos)
L4mishotiz said:
Ok Thank you @CallMeAria.. Just to make sure with cm7 are you able to watch youtube videos, or web content using xscope(streaming videos)
Click to expand...
Click to collapse
Youtube yes, but only the mobile version of the site. Of course there's also the youtube app and youtube downloaders. There is also a megavideo app that allows you to watch video from that site.
Sent from my cm7 Aria.
FWIW, I have tried GtabCom, Flashback, and Clemsyn/Calkulin6 recently, and I think I am going to stick with Clem/Calk6 (froyo). Here's what I found, and the order of the comparison topics reflects my priorities:
Boot time: quicker on clem/calk6 - takes about 24s from power button press to lock screen
Responsiveness: snappier on clem/calk6.
Stability (FCs, freeze-ups): better on clem/calk6
Video playback: Tied. Even though froyo has HW acceleration, video playback seems to be about the same for the quality of videos that I watch
Netflix: Tied. Available and working on both
upnp: Tied. Access to content over upnp works on both
Web browsing: Tied. HC has a better shortcut widget but Clem/Calk has a snappier browsing experience.
Apps/games: Tied. Everything I play seems to work on both platforms
Flash web games: Tied. Sites like Kongregate seem to work on both.
Flash video: Tied.
Sexiness: HC. HC is WAY sexier than froyo
Widgets: HC. Related to sexiness. Widgets are way better on HC.
Gmail widget: HC. Have not found a decent gmail widget on Clem/Calk
Location: HC. IP-based location not working on Clem/Calk6
Camera: Froyo. Works on froyo but I don't use it
Overall, I would say that in comparing functionality, froyo and HC are tied. HC is definitely better looking, but for me beauty is only skin deep , and speed and reliability are keeping me on Froyo.
Could you run this hw acceleration stress test on calkulin + clemsyn combo and let us know what fps the tablet could get :
http://demos.hacks.mozilla.org/openweb/HWACCEL/
Thanks
Thanks for posting.
How about battery life? Supposedly Gingerbread fixed a lot of the battery life problems and HC implemented that. Did you do anything with battery life comparision?
mydave1 said:
Could you run this hw acceleration stress test on calkulin + clemsyn combo and let us know what fps the tablet could get :
http://demos.hacks.mozilla.org/openweb/HWACCEL/
Thanks
Click to expand...
Click to collapse
I'm also on Calkulin+ClemsynCombo v6 (an prefer it)...I get 10 FPS...my Windows PC @ work gets 12FPS.
Al
jhermit said:
Thanks for posting.
How about battery life? Supposedly Gingerbread fixed a lot of the battery life problems and HC implemented that. Did you do anything with battery life comparision?
Click to expand...
Click to collapse
Good point - I forgot about this in my original post. I think that battery life is much better in Clem/Calk6, and this is also important to me however I don't have any hard measurements to back this up. There are some threads that talk about included apps that reduce battery life in HC.
mydave1 said:
Could you run this hw acceleration stress test on calkulin + clemsyn combo and let us know what fps the tablet could get :
http://demos.hacks.mozilla.org/openweb/HWACCEL/
Thanks
Click to expand...
Click to collapse
aabbondanza said:
I'm also on Calkulin+ClemsynCombo v6 (an prefer it)...I get 10 FPS...my Windows PC @ work gets 12FPS.
Al
Click to expand...
Click to collapse
Al, thanks for posting the result. I assume that you are using the stock browser ?
People who used HC alpha reported their test results to be between 4 - 12 fps. So it seemed that the full hw acceleration of froyo is not being utilized in browing ?
Could you repeat the test with different browsers, i.e. firefox, Dolphin HD v4.5 and opera mobile ?
Your 12 fps score on the PC is low. A $30 cheap 5450 radeon video card with 1 MB memory could improve the score to 60 + fps and you will notice a significant improvement.
Dave
mydave1 said:
Al, thanks for posting the result. I assume that you are using the stock browser ?
People who used HC alpha reported their test results to be between 4 - 12 fps. So it seemed that the full hw acceleration of froyo is not being utilized in browing ?
Could you repeat the test with different browsers, i.e. firefox, Dolphin HD v4.5 and opera mobile ?
Your 12 fps score on the PC is low. A $30 cheap 5450 radeon video card with 1 MB memory could improve the score to 60 + fps and you will notice a significant improvement.
Dave
Click to expand...
Click to collapse
Here's the results for Dolphin HD - 12 FPS & Firefox 6 FPS.
Couldn't locate a copy of Opera Mobile...market and amazon show my gtab as incompatible.
Yeah, business PC @ work not much for graphics- shared memory s.cks...your tax dollars @ work (it's a government PC).
Flash web games on Facebook (zynga, CivWorld, etc) are awful on both.
reignking said:
Flash web games on Facebook (zynga, CivWorld, etc) are awful on both.
Click to expand...
Click to collapse
Bejeweled Blitz works fine for me through facebook, played on both Calk/Clem and BC/Clem both OCed 1400
Problem for me is the 'sexiness' of HC makes it very hard for me to go back to my old rom (Beasty/Clem) despite the fact it is noticeably faster then Gtab. Plus you lose out on some great tablet apps like News360.
Reeser,
What other tablet apps do you like since I do like News360 and a number of apps I find are definitely designed for phones instead of tablets.
reeser said:
Problem for me is the 'sexiness' of HC makes it very hard for me to go back to my old rom (Beasty/Clem) despite the fact it is noticeably faster then Gtab. Plus you lose out on some great tablet apps like News360.
Click to expand...
Click to collapse
You gain very little but lose a lot with HC. Mainly NO HW Acceleration. If I have good hardware then I want to be able to get the full benefits it has. HC is definitely the latest but gor GTabs, not the greatest IMO.
poymeister said:
FWIW, I have tried GtabCom
Click to expand...
Click to collapse
That's what I use. I prefer an OS designed for a Tablet, when using a Tablet.
schettj said:
That's what I use. I prefer an OS designed for a Tablet, when using a Tablet.
Click to expand...
Click to collapse
Yeah but it's like using a pc with an awesome video card only that he awesome video card is disabled in favor of lame intel graphics.
Just to get a consesus going. I've went back and forth between GTC and Clem/Calk5, I keep going back to Froyo. It may not have all the bells and whistles but it's just not worth it for what I lose in HC.
I'll stay on Froyo until I encounter a must have app that will only run on HC.
Here is the one thing that is detering me from going back to Froyo (and its so minor..)
PORTRAIT ORIENTATION. I WANT MY SOFT KEYS TO BE ON THE BOTTOM.
If Honeycomb would play hulu I would stay there, but I only get sound or video but not both in any of the solutions I have tried....anyone have a working hulu fix for Honeycomb?
I also agree, I also switch back to Gingerbread or Honeycomb due to the buttons on the bottom, if someone could create a utility for this I would probably stay with Froyo. As it is I switch between Beasty's, CM7Tab and Flashback and GtabComb on nearly a daily basis...So conflicted!
HD2FORNICK said:
Yeah but it's like using a pc with an awesome video card only that he awesome video card is disabled in favor of lame intel graphics.
Click to expand...
Click to collapse
I hear this often, but in my actual use I don't see any issues with overclocked performance in graphics. I must not be playing enough games - I use it for email, reading, browsing, etc. Maybe watch a video now and then (on the far less than 720p screen, watching "HD" video makes no sense)
So, yeah. Whatever floats your boat.
I've been wondering about the hardware acceleration issue as well. It came up in the Illuminate thread, since he's claimed that HA is working on his rom. Folks there have tested some graphics intense games and they worked fine on their system. They can play video with software decoding, so what exactly does HA give us as a benefit.
Can anyone that's running a HC rom please run an Antutu benchmark and report their 2D and 3D rendering results?
the other benchmark test mainly concentrate on CPU performance, not graphics rendering. I had run a Neocore test and got 62.7FPS with Calk/Clem. I've been told that the test is outdated, but if it produces the same results with a HC rom, then we'll know that HA doesn't effect graphics rendering as much as we thought.
Thanks,
Todd Evans
TJEvans said:
I've been wondering about the hardware acceleration issue as well. It came up in the Illuminate thread, since he's claimed that HA is working on his rom. Folks there have tested some graphics intense games and they worked fine on their system. They can play video with software decoding, so what exactly does HA give us as a benefit.
Can anyone that's running a HC rom please run an Antutu benchmark and report their 2D and 3D rendering results?
the other benchmark test mainly concentrate on CPU performance, not graphics rendering. I had run a Neocore test and got 62.7FPS with Calk/Clem. I've been told that the test is outdated, but if it produces the same results with a HC rom, then we'll know that HA doesn't effect graphics rendering as much as we thought.
Thanks,
Todd Evans
Click to expand...
Click to collapse
Todd,
Antutu v2.1
Flashback 7.1
2d 315
3d 1030