It's impossible to calibrate the G Sensor in Sense, Vertical is OK but Horizontal completely flips out. Cant get it to cailbrate anymore now...

Hey Fella, thanks for this build.

By using [BUILD]28.09.10|Mdeejay eVo Sense v. 1.5 Revolution|3.29.651.5, Radio 2.14 and your kernel/modules i have a nice, stable and fast phone with cracking ppp data connection.

not tested wifi yet as I am at work. will update later.

The R6 kernal is a beast I am getting low numbers in standby no freezing except for the first boot. Oh and the best part is I get no lag on the lock screen.

Originally Posted by aussiebum

Get some sleep 1st then I'm using #5 just now and the mac address is fine though.

Unfortunatelly noone kernel have real mac wifi address.Τhe new fake address is 00:90:4C:C5:00:34
I hope that the developers will fix this

What about camera flash, is it still premature with this kernel?

Sent from my HTC HD2 using XDA App

Originally Posted by hastarin

OK noted. Mac address is wrong and g-sensor issues.

I'll look at them when I can but I haven't touched code related to either recently.

Just searched this thread and it seems Mac address was broken as of #5, so I suspect it's related to changes in the main tree. I'll look into it in more detail when I can.

Right now I need to chill and then sleep.

Yeah i need real mac cause i wanted to update our other hd2 to this cause its so much better

For users running cm6 aka n1 builds there might be high background noise or echo.

LeTama hastarin: what CM6 people can try is to rename acdb file, it should fallback to pre-patch behaviour
11:51 LeTama hastarin: something like default.acdb.bak for instance


LeTama hastarin: /etc/firmware, and depending on build it can be copied from /system/etc/firmware to /etc/firware.

I tried this and it worked out for me. thanks hastarin for looking into this Hope you dont mind that I stole it and posted it.

Originally Posted by clio94

Unfortunatelly noone kernel have real mac wifi address.Τhe new fake address is 00:90:4C:C5:00:34
I hope that the developers will fix this

You are quite correct. I found the issue in my build process and now get that exact address!

I will see about a 6.1 at some stage but it is changing one hard coded value for another so I won't rush.

Sent via XDA App

Originally Posted by hazard99

For users running cm6 aka n1 builds there might be high background noise or echo.

LeTama hastarin: what CM6 people can try is to rename acdb file, it should fallback to pre-patch behaviour
11:51 LeTama hastarin: something like default.acdb.bak for instance


LeTama hastarin: /etc/firmware, and depending on build it can be copied from /system/etc/firmware to /etc/firware.

I tried this and it worked out for me. thanks hastarin for looking into this Hope you dont mind that I stole it and posted it.

Not at all. Glad to hear it worked for you. I was waiting to hear back from someone before I posted the info but you've saved me the trouble.

Sent via XDA App

Originally Posted by tfouto

Hastarin, what's your card, and what's the cluster size of it? Can you make some Read/Write speed tests?

This post seems to be the answer to stuttering issue:

http://forum.xda-developers.com/show...&postcount=499

I had 32k clusters and and issues, then format to 8k and the issues are huge and ppp dont work almost never... Its broken really soon, the ppp connection. I might have to format to 64k. But in old kernel (N1) 8k plays music fine. So it seems that Evo Kernel dont deal very well with SD management... You have to got SD shining, when on old kernels no need to...

If you want to reproduce this issues just format your card with 8k or 4k clusters and preferebly on an old card.

Sandisk class 2 16gb formatted under windows with whatever largest cluster was. 8k from memory.

I can run a benchmark just let me know a free app to do it?

Sent via XDA App