android
  #1321  
Old 05-07-2011, 09:16 PM
bleargh bleargh is offline
Junior Member
 
Join Date: May 2011
Posts: 1
Default

will the "backlight on keypress" work on skip/next track too?
Reply With Quote

Advertisement [Remove Advertisement]

  #1322  
Old 05-08-2011, 03:20 AM
dfkt's Avatar
dfkt dfkt is offline
Moderator
 
Join Date: May 2006
Location: Vienna, Austria
Posts: 15,330
Default

No, it just leaves the screen turned off at volume change or pressing play/pause.
__________________
Please don't PM me with questions that can be answered in a forum thread. Don't be an idiot.
My Gear and Reviews | My RMAA Tests | IRC: #anythingbutipod on Freenode | Last.fm | Album Art Exchange | Rockbox | Replaygain
Reply With Quote

  #1323  
Old 05-08-2011, 05:17 PM
mizuhikari mizuhikari is offline
Junior Member
 
Join Date: Nov 2010
Posts: 2
Default

Hi there, I test my clip+ with the last Meier v2, channel swap build from dfkt. So far is good.
Just wonder if there is a build for clip v1 too?
Reply With Quote

  #1324  
Old 05-11-2011, 01:56 PM
keyb_gr keyb_gr is offline
Junior Member
 
Join Date: Mar 2011
Posts: 54
Default

mizuhikari: Building for multiple targets isn't all that much fun AFAICS. That's why people tend to stick to one type of player. (I have no idea how it's done on the build server. My guess is that they have one build environment per player, and each is sync'd to SVN.)

Here's a current Clip+ build of mine as of r29855, with just a couple of small things extra (FS#11304, FS#12094 "high gain", FS#12111).
rockbox-sansaclipv2-29855m-11304-12094-12111.zip
Should sound a touch better than a stock build (much like dfkt's latest one which also includes FS#11304), and those having a set with RDA5802 FM tuner chip will notice that soft-muting is disabled and sensitivity might be a teeny tiny bit better. So it's sort of an audiophile DX edition.

Nothing too fancy so far, I need to learn some more before I get into extra features.
Reply With Quote

  #1325  
Old 05-11-2011, 04:59 PM
mizuhikari mizuhikari is offline
Junior Member
 
Join Date: Nov 2010
Posts: 2
Thumb Up

keyb_gr: thank you for you reply .

yea, i know that . I just curious how that sound in clip v1. And thank's for your build, really appreciated that. I will try after i arrived at home.
Reply With Quote

  #1326  
Old 05-11-2011, 05:13 PM
skip252 skip252 is offline
Administrator
 
Join Date: Dec 2007
Location: Chicago
Posts: 5,356
Default

Quote:
Originally Posted by keyb_gr View Post
Should sound a touch better than a stock build (much like dfkt's latest one which also includes FS#11304)
You're saying that you detect an audible difference with FS#11304? saratoga said here if it did it could be committed if it actually does something. However this is the first time I've heard anyone say it actually sounds better. Do you have more information on what's different?
Reply With Quote

  #1327  
Old 05-11-2011, 05:54 PM
dfkt's Avatar
dfkt dfkt is offline
Moderator
 
Join Date: May 2006
Location: Vienna, Austria
Posts: 15,330
Default

I just measured the results with and without both parts of FS#11304. It didn't really show any difference.

I couldn't listen to it in an AB test, since I only have one Clip+, and swapping firmwares takes way too much time for my auditory memory to remember any differences.
__________________
Please don't PM me with questions that can be answered in a forum thread. Don't be an idiot.
My Gear and Reviews | My RMAA Tests | IRC: #anythingbutipod on Freenode | Last.fm | Album Art Exchange | Rockbox | Replaygain
Reply With Quote

  #1328  
Old 05-11-2011, 06:08 PM
skip252 skip252 is offline
Administrator
 
Join Date: Dec 2007
Location: Chicago
Posts: 5,356
Default

Exactly why I was wondering what differences they might be hearing. I do have more than one Clip+ and conducted volume matched A/B testing since the first build you posted with FS#11304 included. I was looking to see if it actually drove higher impedance headphones better. It hasn't for me. I've never heard a a bit of difference. Max volume is a bit quiet but very acceptable.

When I read what it did I even ran a couple of battery benches to make sure it didn't affect the battery life. With or without they all came within a few minute of each other so I saw no difference there either.
Reply With Quote

  #1329  
Old 05-12-2011, 09:17 AM
keyb_gr keyb_gr is offline
Junior Member
 
Join Date: Mar 2011
Posts: 54
Default

As I commented on FS#11304, any improvements due to higher HP amp bias current would mainly be expected driving low-impedance loads, as the AS3543 datasheet states. That's in line with what I found simulating amplifier circuits. (The mixer quality setting, by contrast, may improve distortion performance into line-level loads, assuming the effect isn't entirely masked by headphone amp distortion. A player with line-out may be needed to see the effect.)

I have no means of testing whether the changed settings really have any audible effect, with only one player and all. I'll have it tested by someone who regularly conducts measurements using Triple.fi 10 Pros as a load though.

(All I can tell is that my Clip+ sounds glorious on my trusty HD590s, with distortion performance limited by the recordings in almost all cases.)

Last edited by keyb_gr; 05-12-2011 at 10:12 AM.
Reply With Quote

  #1330  
Old 05-12-2011, 10:04 AM
bertrik bertrik is offline
Rockbox Developer
 
Join Date: Jul 2009
Posts: 34
Default

Quote:
Originally Posted by skip252 View Post
When I read what it did I even ran a couple of battery benches to make sure it didn't affect the battery life. With or without they all came within a few minute of each other so I saw no difference there either.
Actually this is quite an interesting fact, as possibly reduced battery run-time was an important argument to *not* enable high-quality mode (AFAIK). If having high-quality enabled or not does not make a noticeable difference in battery run-time, we might just as well enable high-quality mode by default.
Reply With Quote

  #1331  
Old 05-12-2011, 10:15 AM
dfkt's Avatar
dfkt dfkt is offline
Moderator
 
Join Date: May 2006
Location: Vienna, Austria
Posts: 15,330
Default

Any more battery benchmarks with/without FS#11304 needed? I could do some as well.

I could also record the Clip+'s output (with various loads), to see if the patch makes any audible difference with various phones.
__________________
Please don't PM me with questions that can be answered in a forum thread. Don't be an idiot.
My Gear and Reviews | My RMAA Tests | IRC: #anythingbutipod on Freenode | Last.fm | Album Art Exchange | Rockbox | Replaygain
Reply With Quote

  #1332  
Old 05-12-2011, 10:49 AM
skip252 skip252 is offline
Administrator
 
Join Date: Dec 2007
Location: Chicago
Posts: 5,356
Default

I wouldn't want anything committed using my memory from that far back. I have a relatively recent battery benchmark using r29511-110303. I'll grab a build with #11304 enabled, set the same conditions and do another soon as get the player fully charged. It should be finished sometime tomorrow.
Reply With Quote

  #1333  
Old 05-12-2011, 01:48 PM
keyb_gr keyb_gr is offline
Junior Member
 
Join Date: Mar 2011
Posts: 54
Default

Here are the measurements (muchas gracias to industrious Bad_Robot, who did them twice to ensure repeatability):

Standard build
My build, with FS#11304

As you can see, it's not a ginormous OMG!!!1 type difference, but it's there - the distortion spectrum is somewhat more well-behaved with FS#11304 in. It's about 2 dB less for the 5th harmonic, up to about 6 dB for the 21st. And 1 kHz is near the impedance maximum for tf10Pros.

If someone wants to check this in (provided the runtime tests turn out OK), you might as well take the 11304-enabled version of the diffs from FS#12111.

Last edited by keyb_gr; 05-12-2011 at 02:02 PM.
Reply With Quote

  #1334  
Old 05-14-2011, 02:48 PM
skip252 skip252 is offline
Administrator
 
Join Date: Dec 2007
Location: Chicago
Posts: 5,356
Default

There's a drop in runtime in this battery bench with FS#11304 vs. the last one, 15:13:41 vs 15:50:28. I used the build dfkt posted here so the circumstance aren't exactly the same but very close imo. Everyting else was the same as this
Quote:
This battery benchmark was done today using a playlist made from 100 files split evenly between the external card and the internal memory. 50 -q5 Vorbis and 50 -V4 .mp3. The equalizer was on. The volume was adjusted to between -25 and -19 depending on how noisy things were around me . The backlight was set to 10 seconds. I listened with my usual set of Senn HD 428s connected. Those are of 32 ohm headphones.
so it's more of a real life test.

I have a copy of keyb_gr's build and will give that a try when time permits. That should be fairly soon.
Attached Files
File Type: txt battery_bench_‎May ‎13, ‎2011.txt (61.4 KB, 17 views)
Reply With Quote

  #1335  
Old 05-16-2011, 03:43 AM
Timar Timar is offline
Member
 
Join Date: Apr 2010
Posts: 107
Default

Quote:
Originally Posted by skip252 View Post
There's a drop in runtime in this battery bench with FS#11304 vs. the last one, 15:13:41 vs 15:50:28. I used the build dfkt posted here so the circumstance aren't exactly the same but very close imo. Everyting else was the same as this so it's more of a real life test.
Hmm, dunno, but "real life test" sounds like "way too many unknows variables" to me. The difference in runtime could have a lot of causes: you could have played more ogg files the one time, more files from external memory, you could have turned on the backlight more often etc. pp. I think if we really want to know if FS#11304 makes a difference we have to compare it under "laboratory conditions"...
Reply With Quote

  #1336  
Old 05-19-2011, 09:37 AM
dfkt's Avatar
dfkt dfkt is offline
Moderator
 
Join Date: May 2006
Location: Vienna, Austria
Posts: 15,330
Default

Here's a new build, with two new patches, besides the usual ones:
  • FS#11891 manages to achieve gapless playback with "badly" encoded MP3s (ie. most anything that isn't LAME).
  • FS#12111 by keyb_gr (see above) includes FS#11304 and additionally disables output mixer AGC (automatic gain control).

r29898M-110519-sansaclipplus (FS#9305, FS#10849, FS#11577 Meier v2, FS#11891, FS#12111+11304, channel swap) - dfkt.zip
__________________
Please don't PM me with questions that can be answered in a forum thread. Don't be an idiot.
My Gear and Reviews | My RMAA Tests | IRC: #anythingbutipod on Freenode | Last.fm | Album Art Exchange | Rockbox | Replaygain
Reply With Quote

  #1337  
Old 05-19-2011, 10:33 AM
dfkt's Avatar
dfkt dfkt is offline
Moderator
 
Join Date: May 2006
Location: Vienna, Austria
Posts: 15,330
Default

Here's an RMAA test, default Rockbox compared to FS#12111+11304: http://rmaa.elektrokrishna.com/Temp/...11%2B11304.htm

There are some very tiny differences between the builds, as keyb_gr already mentioned above. The build with FS#12111+11304 included seems to be a tiny bit louder than the default build (less than 1dB). So I did a second set of measurements, where I increased the input volume on my sound card by ~1dB for the stock build, to be more or less equally loud as the patched build (or even a little louder): http://rmaa.elektrokrishna.com/Temp/...Matched%29.htm - THD and IMD results are still the same, ie. a little better with the patched build.
__________________
Please don't PM me with questions that can be answered in a forum thread. Don't be an idiot.
My Gear and Reviews | My RMAA Tests | IRC: #anythingbutipod on Freenode | Last.fm | Album Art Exchange | Rockbox | Replaygain
Reply With Quote

  #1338  
Old 05-19-2011, 10:09 PM
lestatar's Avatar
lestatar lestatar is offline
Ultra Senior Member
 
Join Date: Dec 2009
Location: Hong Kong now, but NYC always
Posts: 4,657
Default

Quote:
Originally Posted by dfkt View Post
Here's a new build, with two new patches, besides the usual ones:[LIST][*]FS#11891 manages to achieve gapless playback with "badly" encoded MP3s (ie. most anything that isn't LAME)....SNIPPED...
@dfkt, just saw this. How the hell did you manage this? Sweet!
__________________
DAPs:2xRBFuze8+16GB|SonyE345|ZenV+,Micro,Xtra40GBx2|RBG igabeatFX|RCAOpal
IEMs:PanaHJE900|Nuforce 700x|HippoVB|iMetal590,i490|SM PL-21|CX300|EP630,Aurvana
Guitars:IbanezS540,JS1200|Ovation
PoolCues: a bunch
A Glossary for Newbies
Reply With Quote

  #1339  
Old 05-20-2011, 06:33 AM
dfkt's Avatar
dfkt dfkt is offline
Moderator
 
Join Date: May 2006
Location: Vienna, Austria
Posts: 15,330
Default

You have to ask dryrock how he did it. I just added his fine patch to the build.

It is a somewhat controversial patch for some Rockbox developers, when you read the comments: http://www.rockbox.org/tracker/task/11891

I couldn't find/hear anything detrimental with that patch yet. It does its job on FhG/Xing/etc stuff and leaves proper LAME MP3s with gapless headers alone.
__________________
Please don't PM me with questions that can be answered in a forum thread. Don't be an idiot.
My Gear and Reviews | My RMAA Tests | IRC: #anythingbutipod on Freenode | Last.fm | Album Art Exchange | Rockbox | Replaygain
Reply With Quote

  #1340  
Old 05-22-2011, 01:48 PM
keyb_gr keyb_gr is offline
Junior Member
 
Join Date: Mar 2011
Posts: 54
Default

Quote:
Originally Posted by dfkt View Post
The build with FS#12111+11304 included seems to be a tiny bit louder than the default build (less than 1dB).
Interesting. That could be the effect of the disabled output mixer AGC in FS#12111.
Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump



All times are GMT -5. The time now is 07:05 PM.