Rockbox Technical Forums
Support and General Use => Hardware => Topic started by: hollowman on April 24, 2010, 05:20:06 PM
-
I just installed the latest RockBox onto my X5-30gb (first time on this X5) and am getting no audio via Line Out jack (on subpack). Headphone out on main unit works fine. Also, it seems that music/audio signal is being deliberately muted as I can hear, over Line Out, the usual in-between track and power on/off switching ("whistling") noises. But as soon as music/audio starts, the output is muted. Again, only affecting Line Out.
I have another X5 (60gb) running an older RockBox version, and that works fine.
-
Could you swap installs and rule out hardware and confirm software?
-
Could you swap installs and rule out hardware and confirm software?
I don't understand what this means????
I installed v3.5.1 again. No diff.
-
Yea, it wasn't the best worded request.
From the original post it appears you are running the latest version of Rockbox on a previously unRockboxed X5, and an old Rockbox on another X5.
You are implying something is wrong with current Rockbox because your other X5 (the one with old Rockbox) works fine and the "new" X5 with "new" Rockbox does not.
I am suggesting that it is more likely you have a hardware issue - but the only way to be sure is to put "new" Rockbox on the "old" X5 and copy "old" Rockbox to the "new" X5, then repeat the test.
-
From the original post it appears you are running the latest version of Rockbox on a previously unRockboxed X5, and an old Rockbox on another X5.
YupYou are implying something is wrong with current Rockbox because your other X5 (the one with old Rockbox) works fine and the "new" X5 with "new" Rockbox does not.
...Or the way my new X5 "accepted" RBI am suggesting that it is more likely you have a hardware issue - but the only way to be sure is to put "new" Rockbox on the "old" X5 and copy "old" Rockbox to the "new" X5, then repeat the test.
The "new" X5's Line Out worked before I installed RB. Have made no other software/hardware changes to it knowlingly. I assume RB has no way of controlling Line Out as it does with Line In?
Again, and this is important: The audio signal seems to be deliberately muted. If I turn up the vol. on my Line-Out connected amp, I can hear the usual background digital spuriae noise one hears if one cranks up the vol. while unit is in Pause, Skips trk, Powers On/Off.
-
You may very well be right - but there are multiple possibilities which can not be ruled out until the test I attempted to outline is performed.
-
You may very well be right - but there are multiple possibilities which can not be ruled out until the test I attempted to outline is performed.
I am not willing to jeopardize my other X5 -- I only have two PMPs ;)
Can you suggest a way to hack RB so I can at least try to rectify the issue? Otherwise, what's the simplest way to go back to the orig. Cowon firmware?
Thx!
-
I don't see the jeopardy.
Simply copy off your current install from the "old" X5 and paste it onto the "new" one. And vise-versa. (copying to a temporary folder will make this easier)
If you don't like the results you can reverse the process.
You appear to be asking for rectification of an issue which has not been soundly demonstrated to exist. As things stand now there is not strong evidence of anything other than hardware issues. Before you mention again that this problem did not exist before the install of Rockbox on the "new" X5 - I hear you but w/o repeating the situation we very well might be chasing a coincidental hardware issue.
-
I don't see the jeopardy.
Simply copy off your current install from the "old" X5 and paste it onto the "new" one. And vise-versa. (copying to a temporary folder will make this easier)
If you don't like the results you can reverse the process.
You appear to be asking for rectification of an issue which has not been soundly demonstrated to exist. As things stand now there is not strong evidence of anything other than hardware issues. Before you mention again that this problem did not exist before the install of Rockbox on the "new" X5 - I hear you but w/o repeating the situation we very well might be chasing a coincidental hardware issue.
Regardless, you have failed to answer my question whether RB code/firmware is capable of controlling Line Out as it does with Line In? If you don't know the answer to this question, please say so. If you do, then clue me in.
IAC, my other X5 is at my residence quite a distance from current location. My current "PMP" is, alas due to the X5 snafu, my laptop.
-
I do not believe the TI TLV320AIC23B has the hardware capability to control the line-out volume.
And nothing audio-driver related has been touched on the iAudio players in almost one year. If there were a Rockbox problem I would expect it to have reared its ugly head by now. The Cowon players are not the most popular, but they have exactly the userbase I'd expect complaints from.
-
Line-out works for me on my M5 which is very similar to the X5 and especially the audio, processing and line-out hardware is the same (DAC, CPU and subpack). I tested with one of today's builds, revision 25710 to be precise.
Rockbox does not control the line out volume with one exception - a global (headphone) mute will also completely mute line-out, setting the volume one step above the line-out will give its full volume - a bit like on or off. I don't know why this is the case just stating what I see on my player.
If you still want to go back to the original firmware you need to install a Cowon firmware again, i.e. a firmware update by them. This Cowon support page should have all the info you need: How to upgrade iAUDIO X5 Firmware? (http://www.cowonglobal.com/zeroboard/view.php?id=B15&page=1&sn1=&divpage=1&bmenu=support&category=10&sn=off&ss=on&sc=on&select_arrange=headnum&bmenu=support&desc=asc&no=358&bmenu=support).
-
Line-out works for me on my M5 which is very similar to the X5 and especially the audio, processing and line-out hardware is the same (DAC, CPU and subpack). I tested with one of today's builds, revision 25710 to be precise.
And it works for me, too, on my other X5 60-gb.
FYI: I did swap the two X5s' HDs, and the orig. muted X5 still presents muted Line Out audio regardless of HD. Not sure what this proves?? After all, RB also flashes firmware, which stays with the unit regardless of HD used.
Rockbox does not control the line out volume with one exception - a global (headphone) mute will also completely mute line-out, setting the volume one step above the line-out will give its full volume - a bit like on or off. I don't know why this is the case just stating what I see on my player.
Yeah, I've known about this since my X5 was new. And this is what I was implying earlier. Any way you know how to manually see and control that bit?
If you still want to go back to the original firmware you need to install a Cowon firmware again, i.e. a firmware update by them. This Cowon support page should have all the info you need: How to upgrade iAUDIO X5 Firmware? (http://www.cowonglobal.com/zeroboard/view.php?id=B15&page=1&sn1=&divpage=1&bmenu=support&category=10&sn=off&ss=on&sc=on&select_arrange=headnum&bmenu=support&desc=asc&no=358&bmenu=support).
Thx. A lot on my plate right now but if/when I have progress, I'll post.
-
FYI: I did swap the two X5s' HDs, and the orig. muted X5 still presents muted Line Out audio regardless of HD. Not sure what this proves?? After all, RB also flashes firmware, which stays with the unit regardless of HD used.
Bootloader gets flashed.
Hard drive is Rockbox itself.
If you swapped HDs you swapped Rockbox.
-
I uninstalled RockBox and installed the orig. Cowon firmware. Line Out still "dead".
I went back to RB, in hopes a flag or bit would toggle. Nope!
Again, Line Out is not dead. The normal noises, such as the garbage one normally hears when one is skipping tracks, powering on/off, etc., are clearly audible in both channels. Then, as music signal starts, it's as if the Line Out audio signal is deliberately muted. Again, Phones out is unaffected.
Any ideas ?? -- I'm just about to call it quits.
Editorial remarks: The X5 is antiquated. When orig released, it had no rivals -- and not for a while. But it, for me, has always been buggy and unreliable. I've owned more than one. I put up with it, as many audiophiles have I suspect, just because it sounded good. Now, with inexpensive and/or way-better audiophile PMPs from China (e.g., Nationite flo 2, Head-Direct HiFiMan), Cowon, Apple, iRiver, etc. are clearly outmatched in the audio department. The Nationite units are cheap, too. Or you can save up for a HiFiMan.
-
I uninstalled RockBox and installed the orig. Cowon firmware. Line Out still "dead".
That is unfortunate. The evidence points pretty conclusively to hardware failure.
Again, Line Out is not dead. The normal noises, such as the garbage one normally hears when one is skipping tracks, powering on/off, etc., are clearly audible in both channels. Then, as music signal starts, it's as if the Line Out audio signal is deliberately muted. Again, Phones out is unaffected.
Any ideas ?? -- I'm just about to call it quits.
Deliberate muting as you describe is an active behavior that Rockbox does not perform, nor can perform when the original firmware is loaded. And vise versa. Consistency on this part is clearly a hardware issue.
-
Consistency on this part is clearly a hardware issue
Based on the symptoms I described, I pretty much disagree with you.
It's more likely a firmware issue...codewise, likely at a low level (e.g., "assembly", HEX, or bit level). Are you/anyone aware of a way (utility) of checking bit settings for Line Out? Perhaps via RB's Debug: View I/O mode?
-
You reinstalled the iAudio firmware, no?
One of its jobs is to setup all the hardware. "Set the bits".
If you ran with it as you reported and had the same issue you can rest assured that the cause is not firmware setting some register wrong.
Otherwise you are implying an identical bug in both firmwares.
Not to mention nothing that low-level on that target has been touched in the Rockbox source for ages. If there was somehow an issue (which the evidence strongly supports the statement that there is not) it would have been seen before now.
EDIT:
If you feel the evidence does not support this, please lay out an explicit argument - I believe the case is cut and dry, though, but would like to see your reasoning.
-
If you feel the evidence does not support this, please lay out an explicit argument - I believe the case is cut and dry, though, but would like to see your reasoning.
As I've nted since the OP, I noticed the Line Out issue only after first install of RB. Perhaps static electricity discharged during this. RB Install Util. informs one of turning off unit so new firmware can install. The system may be vulnerable during this time.
-
Seriously, nobody else has this problem with current builds of Rockbox. You can't reproduce the problem on your other player. Putting the other player's build on this player, so that the hardware is initialized the same way, does not resolve the problem.
You've more or less explicitly ruled it out from being a firmware problem by trying a different, known good, firmware already.
If a random static discharge fried your player, that's obviously a hardware problem as well.
-
If a random static discharge fried your player, that's obviously a hardware problem as well.
Try this: in your X5 or any HD-based PMP, turn your vol. up to as high as you can comfortably take it. Listen while you power up/down, switch tracks. You should be able to hear digital garbage "whistling" noises, in both channels. That's what I still hear from both Line Out and 'Phones, in this PMP and others; therefore, I don't think the hardware was/is damaged as you seem to suggext.
-
You've said this a few times. I've taken this into account.
Now consider - the DAC could be damaged and not driving the line out possibly, but electrical interference can still be heard through it.
There are possibilities other than the one that you seem to focus on. And remember, you yourself verified that it wasn't the firmware by switching out firmwares.
-
Now consider - the DAC could be damaged and not driving the line out possibly, but electrical interference can still be heard through it.
Rubbish! Go back and carefully re-read thru this thread. Phones out produces HQ sound.
There are possibilities other than the one that you seem to focus on. And remember, you yourself verified that it wasn't the firmware by switching out firmwares.
I agree and I think I've already said it. I fault the X5 or other acts of Nature (incl. something I may have inadvertingly done).
That said, and WADR, several of you responders seem to be going out of the way to protect RB rather than truly problem-solve this issue. Let's try to think ... ahem ... out of the Box!
-
Now consider - the DAC could be damaged and not driving the line out possibly, but electrical interference can still be heard through it.
Rubbish! Go back and carefully re-read thru this thread. Phones out produces HQ sound.
And the fact that headphones out works proves what exactly in your mind? I was aware of this too, which is why I specifically said line out.
That said, and WADR, several of you responders seem to be going out of the way to protect RB rather than truly problem-solve this issue. Let's try to think ... ahem ... out of the Box!
No, we're trying to convince you it's not likely to be a firmware issue since as of 12 hours ago you were still harping on about it being a firmware issue, so that you can go about trying to investigate the real cause, and as of two posts ago you were claiming that it *wasn't* hardware.
We can't help you "solve" a problem until it's identified. And you keep coming back around to trying to claim it's a firmware issue even though you've ruled it out.
If it's a hardware issue, you need to fix or replace the hardware that's damaged. If you still think it's a firmware issue, provide some evidence (such as a firmware version that works alongside a firmware version that doesn't while using the same hardware, not two different players of the same type).
-
And you keep coming back around to trying to claim it's a firmware issue even though you've ruled it out.
But what exactly is firmware -- surely, low-level stuff is stuff that RB does not normally touch, but may have been accidentally affected during the orig. FW flash.If it's a hardware issue, you need to fix or replace the hardware that's damaged. If you still think it's a firmware issue, provide some evidence (such as a firmware version that works alongside a firmware version that doesn't while using the same hardware, not two different players of the same type).
Define "hardware issue", please. IMO, that's a pretty generic def. that'll get one out of an unexamining courtroom. But since you brought it up...
I've been, since the beg., going through the HW datasheets on this page...
http://www.rockbox.org/wiki/IaudioX5HardwareComponents
...notably the TI DAC. It's working normally AFAICS, but I don't know what may be causing it mute LO ... or something else between the TI chip and the LO jack (I assume there is nothing, like a buffer or switch).
-
Let's assume there is low-level stuff Rockbox doesn't touch, and it is here the problem lies.
A - Why are we talking about it on the Rockbox forum?
B - Why didn't reinstalling the Cowon firmware fix it?
I don't know why you linked to the datasheet or what that proves. The line-out and headphone-out are separate circuits on the TI DAC (as shown in the datasheet).
You keep using the word "mute" with no supporting evidence. I'm not trying to nit-pick words, but I believe that poor word choice can lead to poor logic and poor conclusions.
The fact you are able to pick up electrical interference when plugged into the line-out in no way signifies a working, yet muted, line-out circuit. It could be a symptom of that, but it could also be a symptom of a non-working line-out.
My beef is it appears to me a conclusion was reached and the some of the evidence is being held up in support of that conclusion, while ignoring the other (and I feel overwhelming) evidence to the contrary.
Fact: You installed Rockbox.
Fact: Line Out does not work now (I assume the line out was proved working immediately prior to the install.) At best this is correlation.
Fact: You can hear what sounds like electrical interference on the line out now.
Theory: Rockbox broke line-out
Evidence For: Correlation.
Evidence against: Reinstall of Cowon firmware does not restore line out.
Evidence against: Nobody else amongst the hundreds if not thousands of people with the same hardware has reported the problem.
Evidence against: Rockbox audio code for the iAudio players has not changed in 10 months IIRC.
Evidence against: The really low-level setup is done by the Cowon bootloader IIUC, which we use to launch our bootloader.
Now consider - the DAC could be damaged and not driving the line out possibly, but electrical interference can still be heard through it.
Rubbish! Go back and carefully re-read thru this thread. Phones out produces HQ sound.
As I say, this proves nothing. Separate circuits.
EDIT: Not as separate as I thought. See page 9 of the datasheet. I see no way to mute the line-out inside the DAC itself without muting the headphone out
I agree and I think I've already said it. I fault the X5 or other acts of Nature (incl. something I may have inadvertingly done).
That said, and WADR, several of you responders seem to be going out of the way to protect RB rather than truly problem-solve this issue. Let's try to think ... ahem ... out of the Box!
I resent what I see as a misportrayal of the situation.
My challenging assumptions of yours is hardly defensive behavior on my part.
If you want problem solving - let's start with sound reasoning.
-
If you want problem solving - let's start with sound reasoning.
Study the block diag in the TI DAC's datasheet (http://focus.ti.com/lit/ds/symlink/tlv320aic23b.pdf). Look at where the Line Out ckt is compared to HP Out and HP's vol cntrl. Then, realize (as noted elsewhere in this thread) that RB does disable LO when vol is minimized ("0"). I need to check the status of this bit. Can you tell me how to access it in RB or via any other utility?
BTW: do you know of a source for X5 schematics? I cannot obtain them as I usually can with other hifi gear I own.
BTW: Yes, I am absolutely certain that, at the most 15 min prior to 1st install of RB, LO worked. Correlation? ...technically yes, but -- as much as I like RB -- I can't give it the full benefit of doubt here. Maybe something in this X5 was getting ready to "go" anyway, and RB was merely the last straw. Maybe!
-
I need to check the status of this bit. Can you tell me how to access it in RB or via any other utility?
Easiest way would be to write a plugin that printed. However, since the value of registers is reset whenever the power is disconnected, the value you read in rockbox may or may not be the same value as in the OF. Getting the value in the OF would be very, very hard.
BTW: do you know of a source for X5 schematics? I cannot obtain them as I usually can with other hifi gear I own.
I don't know any MP3 players with schematics. These aren't typically published.
BTW: Yes, I am absolutely certain that, at the most 15 min prior to 1st install of RB, LO worked. Correlation? ...technically yes, but -- as much as I like RB -- I can't give it the full benefit of doubt here. Maybe something in this X5 was getting ready to "go" anyway, and RB was merely the last straw. Maybe!
If it happens in the OF then its obviously a hardware issue. Rockbox can't change how the OF works. It sounds like you've burned out the line out amp.
-
If you want problem solving - let's start with sound reasoning.
Study the block diag in the TI DAC's datasheet (http://focus.ti.com/lit/ds/symlink/tlv320aic23b.pdf). Look at where the Line Out ckt is compared to HP Out and HP's vol cntrl.
If you notice in the block diagram you quote, line-out separates from headphone out before the variable amplifier. You'll note there is no controllable hardware after said split on the line-out. In other words nothing to mute the line-out without muting the headphones.
You'll also note that functional blocks which can mute are explicitly labeled as capable of muting, and none of those are in the audio chain in question.
Then, realize (as noted elsewhere in this thread) that RB does disable LO when vol is minimized ("0"). I need to check the status of this bit. Can you tell me how to access it in RB or via any other utility?
No bit to check.
The headphone and line-out are both disabled at minimal volume, that happens upstream of the split.
I feel insulted by what I see as a repeated implication that I am ignoring parts of your messages.