On 10/11/09, RANJAN <infi...@gmail.com> wrote:
> On Sun, Oct 11, 2009 at 1:38 AM, MicVM <ssj2mi...@gmail.com> wrote:
>
>>
>> I am trying to build a vala library to access the accelerometers of the
>> freerunner. I am using vala just to give it a try but I am experiencing
>> strange problems and wanted to ask if some of you can give me some hints.
>> When i read from the input interface of an accelerometer i get the
>> following
>> bytestream:
>>
>> b 3 d1 4a 42 ea d 0 0 0 0 0 0 0 0 0
>> 1b 3 d1 4a 79 f e 0 3 0 1 0 6c 0 0 0
>> 1b 3 d1 4a d7 f e 0 3 0 2 0 26 4 0 0
>>
>> that looks quite good to me. However, when I try to read the values in the
>> message as integers i have a problem when reding the "value" field, i.e.
>> the
>> last 32 bit signed integer of the message. I get strange values around
>> +300
>> instead of +1062 for the Z value as in the last line of the above example.
>> Also for X and Y the values are wrong. I read the stream with the
>> following
>> code snippet in vala (using GIO libs):
>>
>> var data_stream=new DataInputStream(file_stream);
>> data_stream.set_byte_order(DataStreamByteOrder.LITTLE_ENDIAN);
>>
>> //Skip time field, 8 bytes
>> data_stream.skip(8,null);
>>
>> //Read eventtype
>> eventtype=data_stream.read_uint16(null);
>> while(eventtype!=0)
>> {
>> if((eventtype==3)||(eventtype==2))
>> {
>> code=data_stream.read_uint16(null);
>> if(code==0)
>> {
>> x_value=data_stream.read_int32(null);
>>
>> }
>> if(code==1)
>> {
>> y_value=data_stream.read_int32(null);
>> }
>> if(code==2)
>> {
>> z_value=data_stream.read_int32(null);
>> }
>>
>> ....
>>
>> I think somewhere I am doing a major mistake because the decoding of the
>> int32 is a basic functionality that does not work in my case. The funny
>> thing is that decoding of uint16 is done correctly such that eventtype and
>> code are read correctly. Any help is appreciate :)
>> --
>> View this message in context:
>> http://n2.nabble.com/Vala-DataInputStream-read-int32-in-accelerometer-message-tp3801000p3801000.html
>> Sent from the Openmoko Devel mailing list archive at Nabble.com.
>>
>> _______________________________________________
>> devel mailing list
>> devel@lists.openmoko.org
>> https://lists.openmoko.org/mailman/listinfo/devel
>>
>
> What is the maximum frequency at which you are able to read the
> accelerometer values?
> Are you able to sample the values properly?
> Are the reading in X and Y changing smoothly?
>
> Here is a test that I conducted:
> http://www.youtube.com/watch?v=Bdads8XAYcE&feature=fvsr
>
> Sriranjan
>
Hi, i provide here hopefully the answers to your questions:
I tried to measure the time passed when reading the messages as in the
code snippet above and when reading a 16-bytes stream (uint8 field)
for each message. In both examples the freerunner was laying on the
table with the touchscreen on the top. Indeed there is something
strange there:
16ByteStream Method (data_stream.read(buffer,16,null)):
59 fe d1 4a 5b 55 1 0 0 0 0 0 0 0 0 0
uSec: 22
59 fe d1 4a 1c 7a 1 0 3 0 0 0 48 0 0 0
uSec: 8751
59 fe d1 4a 92 7a 1 0 3 0 2 0 2 4 0 0
uSec: 23
59 fe d1 4a b2 7a 1 0 0 0 0 0 0 0 0 0
uSec: 21
59 fe d1 4a 71 9f 1 0 3 0 1 0 90 0 0 0
uSec: 9125
59 fe d1 4a ce 9f 1 0 3 0 2 0 de 3 0 0
uSec: 23
59 fe d1 4a ef 9f 1 0 0 0 0 0 0 0 0 0
Snippet method:
X Message arrived
uSec: 462
Y Message arrived
uSec: 8501
Z Message arrived
uSec: 363
SYN: 0 0 0
uSec: 16
X: -72, Y: 54, Z: 306
X Message arrived
uSec: 464
Y Message arrived
uSec: 402
Z Message arrived
uSec: 370
SYN: 0 0 0
uSec: 16
X: -54, Y: 72, Z: 306
X Message arrived
uSec: 466
Y Message arrived
uSec: 27613
Z Message arrived
uSec: 402
SYN: 0 0 0
uSec: 16
X: -72, Y: 54, Z: 306
X Message arrived
uSec: 2038
Y Message arrived
uSec: 384
Z Message arrived
uSec: 361
SYN: 0 0 0
uSec: 15
Regarding the smoothness of changes, when i decode "manually" the 16
byte stream the changes are rather smooth (e.g. for Z=1000 in the
current position, then 980, then 956 then 880, etcc.). When i decode
it by the code snippet the changes are also "smooth" BUT the ranges
for X, Y and Z are totally asymmetric and not in the same range. E.g.
when I put the freerunner on the X axis I get the value -32 (with -X i
get -90) while when i put it on the Z axis +306 (-Z gives -162 as
result).
About the sampling, I am not sure what u mean. I use:
data_stream.skip(8,null); //for time
eventtype=data_stream.read_uint16(null); //is decoded correctly as 0, 2 or 3
code=data_stream.read_uint16(null); //is decoded correctly as 0,1,2
x_value=data_stream.read_int32(null); // strange behaviour IMHO
which should correspond to the decoding described in:
http://wiki.openmoko.org/wiki/Accelerometer_data_retrieval
_______________________________________________
devel mailing list
devel@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/devel
