Recording Enhancements Pack

[NPswimdude500]

I haven't read all of this thread, so this might have been suggested already, but since there's space, would it be possible to include the name of the recorded file in the recording screen?


^since there's room, how 'bout a line that says something like "Name: rec_0001.wav" ?

[petur]

Not shure if this is related to the enhancement pack or not.

There is a problem with automatic file splitting during recording if it should split at max. file size (max. 2GB FAT file size - 1MB).

I found the bug. It's not related to the enhancement pack at all. The bug is in 'pcm_recrd.c' in function 'pcmrec_start()':

Line 644 is wrong:

Code: [Select]

    num_rec_bytes = 0;

it should be:

Code: [Select]

    num_rec_bytes = pre_chunks * CHUNK_SIZE;

to take the prerecorded chunks into account.

Right. I wrote a long answer on this and reviewed the code while writing. As a result I got logged out before posting and my answer is lost. Damn.

To keep the long answer short: the bug is that num_rec_bytes is used in the recording screen (imho the wrong place) to determine file splitting.

I'll keep it like that for now as it involves too much rewriting... your bugfix should do.

I'll also increase the margin while I'm at it

edit: I think I messed this one up as well when trying to increase peak detection sensitivity...

edit2: margin is fine I think, so I'll leave it like that

[vinylivo]

Thanks for comitting fix, other users will be even more thankful than me because I can always use a patch otherwise.

I don't know what you mean by 'the bug is that num_rec_bytes is used in the recording screen (imho the wrong place)...' because in my thinking this variable should reflect the number of recorded bytes. And the prerecorded bytes are a part of it in the event of starting an actual recording.
I agree there could be other ways of determining file splitting, perhaps by a get_file_size function or something similar.

[petur]

I don't know what you mean by 'the bug is that num_rec_bytes is used in the recording screen (imho the wrong place)...' because in my thinking this variable should reflect the number of recorded bytes.

Well, there is also a variable that holds the number of bytes written to file, and somehow I feel that variable (and location) is more suited for file splitting. Now the gui thread decides this...

[Mmmm]

By jove...you know what? I think I've done it!!!..........

Although, I am very excited, and when I'm excited I say things that I regret later because I then find that I haven't done it at all!! 

I also talk a lot of rubbish!
 

I must now eat, then I'll look again to check whether I really have done it and do a spot of tidying up and re-naming of strange variables. Then I'll put a patch up that you can all test and tell me of any problems you find.

And Petur, I'll start the long process of compiling for all those targets 

[groovetube66]

Hi all.  Just got my ihp-120 and have been following these threads about recording for a while.  Got a couple of questions:

• With Rockbox + recording enhancments patch is there no longer a setting for ext. mic?  I saw mic and line in.  I tried a test recording using line in and Audio Technica Pro24 stereo electret mic both with and without the 1.5v battery.  The sound was good but levels were low.  Turning up the gain so that digital gain was in effect caused noticeable hiss... or was I hearing the mic self-noise?
• Is there a build that contains the Recording Enhancements with the multiple AGC settings and peak metering?
• Is there a playback setting so you can see the same peak level meters you see while recording?

This Rockbox and iRiver is cool stuff.  If someone ever writes the code for multitrack recording with sync it will be even more amazing.

[NPswimdude500]

Is there a build that contains the Recording Enhancements with the multiple AGC settings and peak metering?

There most certainly is!
http://homepage.sunrise.ch/mysunrise/remartinelli/html/rockbox.html

pilot000 from Misticriver compiles an "optimized" version of the bleeding edge builds. It's more of an experimental build, but it includes that patch as well as a few other neat todbits....
(discussion on MR: http://www.misticriver.net/showthread.php?t=36943)

[Mmmm]

Ok then.... It appears that I really have done it! I wasn't dreaming.

So here is the (hopefully) CVS commitable version of the remote support patch.

It would be fantastic if some people could test it and tell me if they notice anything odd or give any other comments. (whataboutbob how about it? )

The peakmeter is enabled in the rWPS now too (just type %pm in your chosen position)

I haven't made any pre-compiled builds yet for you non-patchers to try and I won't replace the patches in post#1 yet either... not until I'm convinced nothing is amiss with it! 
also no changeable fonts version yet (although I couldn't resist doubling the peakmeter size  )

I've still got to do loads of testing to make sure it compiles and runs on all the other targets too before it can be committed...

Enjoy... 

[attachment deleted by admin, too old]

[bennis]

Wow! I have installed the new build with the agc functions on it. I will be testing today during a concert:) so far looks good! On inputs I can;t however choose for mic input ... only line and buildin mic...

I like the histogrammy thing...gives a smooth idea about the levels .. even without recording it works:) GREAT tool

[petur]

mic = internal mic (mono)
line-in = line-in connector

However, the line-in connector supplies power for an external mic, in fact the iriver firmware 'line-in' and 'ext-mic' are nothing more than the line-in with a different gain range.

I don't know what voltage the h1xx puts on the line-in connector but my h340 gives a bit more than 3V (so better than external 1,5V power!)

[paulheu]

Am I right to assume the new peakmeter patch will require a new agc patch as well? it seems at first glance some stuff changed enough to break the current agc patch..

[Mmmm]

New version...now compiles under every target (except the ones that won't compile anyway (12 & 20).

Oh and I haven't tested 5 or 6 because I couldn't get the versions of binutils and gcc to make the crosscompiler to test them with...

But apart from that it is just about perfect I think... Anyone found any problems yet?

paulheu: Yes I think it will break the present AGC patch..but that's life eh? 

[attachment deleted by admin, too old]

[paulheu]

paulheu: Yes I think it will break the present AGC patch..but that's life eh? 

It is.. but since I prefer to have agc available I will have to hold off on a test run.. 

[paulheu]

I've been fiddling with the agc patch and actually have it working with the new peakmeter/remote patch. Oddly though I only get two letters and some (gain) values in the AGC mode display on the screen.. Things seem to work fine it's just the  display that's odd..

I've attached the patch I've done. As this is my first 'actually change code' patch I hope I'm not breaking anything .. I'm only trying to learn something here.. so shoot me if I am being silly..

EDIT: removed the patch.. It's no longer valid..

[Mmmm]

Paulheu...just quickly looking at your patch... I think that for the following part:

Code: [Select]

+                    lcd_puts_style(0, 7, buf, STYLE_INVERT);
+                else
+                    lcd_puts_style(0, 6, buf, STYLE_INVERT);
+			}
+            else
+            {
+                if(global_settings.rec_source != SOURCE_MIC)
+                    lcd_puts(0, 7, buf);
+                else
+                    lcd_puts(0, 6, buf);
             }

instead of using 6 & 7, try using 7 & 8...

(looks as if the gain settings and AGC mode are printing on the same line!)
Could be wrong though....