I was going to ask why and get into the various questions I have about people's fascination with Linpack - such as do any of you know what that means on your device, how many MLOPS are given over to integer DSP processors vs. other and all of that sort of boring reality.
But instead, I'd like to ask what phone and what rom your using.
Because it's not a stock EVO. Or you're doing a bit of task killing.
I was going to ask why and get into the various questions I have about people's fascination with Linpack - such as do any of you know what that means on your device, how many MLOPS are given over to integer DSP processors vs. other and all of that sort of boring reality.
But instead, I'd like to ask what phone and what rom your using.
Because it's not a stock EVO. Or you're doing a bit of task killing.
I get 6.9 - stock is around 5 or so.
100% stock, 2 day old phone. That was with most apps closed but to be fair i also ran a test with a ton of apps open and pandora radio playing some foo fighters in the background and i still got 6.8
100% stock, 2 day old phone. That was with most apps closed but to be fair i also ran a test with a ton of apps open and pandora radio playing some foo fighters in the background and i still got 6.8
I have zero app killers on my phone btw.
I don't use a task killer at all. Phone has an UpTime of 135:12:25 (no idea if that matters) and fully stock, not rooted Evo and my results are identical to the op's
My stock-like test matched Android Central's video comparing the EVO, Droid X and Nexus One - timecode 35 seconds - just over 5.
Either way - it means that Linpack is not that great a benchmark on Android.
The Andoid process management will righteously kill apps over time.
But hawkijustin's report of maintaining that speed "test with a ton of apps open and pandora radio playing some foo fighters in the background" is still interesting.
I admit, I thought you might have been trolling - apologies.
nope no trolling. I saw that video also and thought something was fishy with the evo he tested so i did my own tests and sure enough thats what i got. He has another video where he is testing the evo, droid (not droid x) N1, and some other att android phone and the evo he tested in that was getting 6.6+ each time.
Something was clearly wrong with the evo he used to test next to the droid x
Something was clearly wrong with the evo he used to test next to the droid x
Not necessarily - I've been sitting here popping open various apps - it will kinda go all over the place.
Linpack's MFLOP test is really meaningless - unless you're using it to show what's wrong with Android Central and the endless supply of creamy YouTubes mis-explaining tech!
It was designed for use in a different age, for a different class of machines, to judge application for a specific range of problems.
If you want to see more of what's wrong with that video, check out my link above, let me know what you think.
I get between 6.5 and 6.8 running Damage Control 3.2 (no JIT), I still think these numbers will sky rocket when an official Froyo build comes about, although I really don't care since this is my phone not my gaming PC. If you want benchmarks ask my about my 3D Mark Vantage PC score, I could really care less if my phone can't run Need for Speed at 100fps.
I'm glad this came up. I don't understand why our EVO is so slow in this benchmark compared to the "top scores" or whatever they call it. Are people running ROMs tuned to increase this score?
Is something wrong with my phone? Im getting 20+, more in the 35s. Is higher worse? I have an htc evo. The scrolling is extremely lagging, and i've been trying to fix it to no avail..is my device defective?
In computing, FLOPS (or flops or flop/s) is an acronym meaning FLoating point OPerations per Second. The FLOPS is a measure of a computer's performance, especially in fields of scientific calculations that make heavy use of floating point calculations, similar to the older, simpler, instructions per second.
megaFLOPS 10^6
Is something wrong with my phone? Im getting 20+, more in the 35s. Is higher worse? I have an htc evo. The scrolling is extremely lagging, and i've been trying to fix it to no avail..is my device defective?
No yours is perfectly fine. After the EVO was upgraded to Froyo (2.2) it now has a JIT compiler (just in time compiler) that makes the CPU and memory run much much more efficiently.
Basically
MFLOPS = Mega FLOPS = Mega Floating Point Operations Per Second.
In computer world - it basically measures how many calculations a CPU can do per second. Before Froyo, the EVO was only able to do around 7 million operations per second. Now, the EVO can handle 30+ MFLOPS. So that's why in the Google presentations about Froyo you will hear them say "it increases your performance about 4-5x" which this is basically demonstrating.
Linpack shows nobody with 48 MFLOPS on the Evo unless you didn't submit it which is highly unbelievable.
Well pictures are all over the Root sub-forum,and since anyone stated which version of linkpack i figure i post the numbers from the all version,with the new one one I get anywhere between 40-45.
Surrreee!!! LOL!!! Well if 48 and 69 Mflops are possible on the Evo, Then I guess my Mflops are 99!!!! Top That! LOL!
U just roored your phone so what will know about that.Just so u all know before linkpack was updated to this newer version people were getting over 50 MFLOPS.
I don't know what was causing this but it got fixed with the latest update.The numbers now seems more credible.
After using setCPU, 128 Mhz only had a linpack score of 2 MFLOPS, which is way too little. 422 Mhz scored 17 MFLOPS the sweet spot, 998 Mhz scored 34, and 1190 Mhz scored 39.
When Sprint announced their first ever 4G handset at CTIA in March 2010, they dropped a bombshell on the mobile world. The HTC Evo 4G is a device that introduces a lot of firsts when compared to other North American handsets. This is the first Andr... Read More
Post your linpack score.
