How are these people getting linpack scores of 20+?

[Flowah]

Linpack by Device GreeneComputing

Looking at the top10 Droid scores, I don't see how it's possible. The lowest score on there triples my 8 and change with BBv1.1 @ 1.0ghz. It cannot possibly be the extra 100-300 mhz can it? It's not 2.2, it says that the devices are running 2.1 or 2.0.1.

The numbers for other devices seem reasonable. Why are the Droid numbers so inflated?

 

[gixxerboy]

idk ive been wondering that my self i have bb v1.1 and i have jitter enabled (a jit client i guess) and i been getting high tens and one time i got an eleven i been trying to get that 11 again but no luck so far

 

[titans]

Probably a JIT rom with Jake's 1.8 kernel running an extremely fast rom like Simply Stunning

 

[JCo352]

I dunno, my fastest was 16 something, I posted in the linpack thread.

 

[gixxerboy]

16 something jeez what cpu speed were you running and what rom

 

[JCo352]

16 something jeez what cpu speed were you running and what rom

At that time I was running JRummy's Remix v0.7 with P3Droid's Redline LV 1.25 GHz Kernel.

 

[gixxerboy]

ur lucky as crap you could get it that fast what does your droid run now?

 

[JCo352]

ur lucky as crap you could get it that fast what does your droid run now?

It's in my sig.

 

[hookerdj]

if you wanna try a fast rom you can get scores like that with dominator by bostcscoob he has jit enabled and that thing flys i think it has jakes kernel in it?my highest was 15 or 16 to

 

[garrett]

I was able to get low 30's using a emulator on a PC. I'm gonna assume that's how the others got those speeds. The fastest I've been able to achieve on a phone is mid 15's running succulent desire 5.0 with Chevy's 1250mhz kernel with JIT enabled.

 

[DroidHoff]

Cyanogen 5.0.7.1 at 1.3 ghz with Jitter enabled got me a 16.665 but Jitter is very PMSy and will cause reboots from time to time. In real world use though not much different than the 700mhz no Jit I'm running right now.

 

[SolidSnake]

I was able to get low 30's using a emulator on a PC. I'm gonna assume that's how the others got those speeds. The fastest I've been able to achieve on a phone is mid 15's running succulent desire 5.0 with Chevy's 1250mhz kernel with JIT enabled.

I used to hover around 10 with CM 5.0.7.1 clocked to 1200 but now get around an average of 18 with froyo overclocked to 1200mhz

This would make sense as to why people are able to get such rediculous scores. I knew they had to be cheating somehow but couldnt figure it out. the only thing I wonder about is when they use the emulator on their pc how do they get linpack to think it's a droid?

 

[garrett]

I was able to get low 30's using a emulator on a PC. I'm gonna assume that's how the others got those speeds. The fastest I've been able to achieve on a phone is mid 15's running succulent desire 5.0 with Chevy's 1250mhz kernel with JIT enabled.

I used to hover around 10 with CM 5.0.7.1 clocked to 1200 but now get around an average of 18 with froyo overclocked to 1200mhz

This would make sense as to why people are able to get such rediculous scores. I knew they had to be cheating somehow but couldnt figure it out. the only thing I wonder about is when they use the emulator on their pc how do they get linpack to think it's a droid?

the same way you can run osx and linux under windows using emulation software like VMware or Virtualbox.... the software emulates the hardware to allow you to install a rom on a virtual system on a PC.

heres some more info on it http://developer.android.com/guide/developing/tools/emulator.html

The Android emulator mimics all of the hardware and software features of a typical mobile device, except that it can not receive or place actual phone calls. It provides a variety of navigation and control keys, which you can "press" using your mouse or keyboard to generate events for your application. It also provides a screen in which your application is displayed, together with any other Android applications running.

 

[SolidSnake]

I was able to get low 30's using a emulator on a PC. I'm gonna assume that's how the others got those speeds. The fastest I've been able to achieve on a phone is mid 15's running succulent desire 5.0 with Chevy's 1250mhz kernel with JIT enabled.

I used to hover around 10 with CM 5.0.7.1 clocked to 1200 but now get around an average of 18 with froyo overclocked to 1200mhz

This would make sense as to why people are able to get such rediculous scores. I knew they had to be cheating somehow but couldnt figure it out. the only thing I wonder about is when they use the emulator on their pc how do they get linpack to think it's a droid?

the same way you can run osx and linux under windows using emulation software like VMware or Virtualbox.... the software emulates the hardware to allow you to install a rom on a virtual system on a PC.

heres some more info on it Android Emulator | Android Developers

The Android emulator mimics all of the hardware and software features of a typical mobile device, except that it can not receive or place actual phone calls. It provides a variety of navigation and control keys, which you can "press" using your mouse or keyboard to generate events for your application. It also provides a screen in which your application is displayed, together with any other Android applications running.

I understand that. I have played around with the emulator a bit before. The thing that I don't understand is how linpack can tell the difference between whether it's a droid, or a nexus1, or a hero, etc. The only way I can see this being done is if you replaced the system.img from your emulator daemon with the one from your specific phone model (though i don't know if this is possible to make work). Anyways it doesn't really matter how they get the phone model to show up, the point is that these people are generating fake linpack scores...

 

[Sam]

when i ran mine (on 2.2) the text at the bottom said sholes 2.1.... maybe it's confused?