Can anyone recommend a current model portable player which supports vorbis and replaygain?

My entire collection is Vorbis ogg files, and all have replaygain tags. I use Track Gain only.

I think if I am not mistaken there aren't any that do. This is precisely why you need to look into the Rockbox firmware which supports ReplayGain with Vorbis

http://www.rockbox.org/

The Sansa Fuze supports OGG Vorbis with Replaygain as of the latest firmware released this month. The options for Replaygain are "Off", "Track", and "Album"(reverts to Track if Album tags not found). I bought one last week for my upcoming vacation due to that, FLAC support, and a Rockbox port in development.

http://forums.sandisk.com/sansa/board/mess...thread.id=23276

Does the firmware in the Sansa FUSE support gapless playback of Vorbis files?

I'll second the recommendation of Rockbox. I've got a 5th gen ipod with rockbox and a customized WPS (While Playing Screen) that is fantastic. I get nearly 12 hours of vorbis out of it at normal listening volumes. It has album, track, and "track while shuffling" which will play album unless you have it in shuffle mode.

Rockbox is pretty easy to use nowadays (wasn't always the case). It's got a loader utility and once you get used to the context-based interface and the multi-layered menu system it's pretty intuitive.

The only problem with rockbox, really, is that since it's open-source developed by people who simply feel like working on it, at their own pace, usually there isn't a stable port available for anything in stores. I.E. normally a DAP is out of production by the time there is a good rockbox port for it. If you're good ordering used off of ebay, or you've got a friend who's got an unused player lieing around that they'll sell you cheap, that's no big deal, otherwise it's a deal-breaker.

Me, I'm waiting for my 5th gen to start dieing it's normal apple planned obsolescence death, so I can justify a 650mAh battery and 32gb (maybe 64gb by then!) compact flash based hard drive replacement. Till then, it's all I would have wanted out of a DAP.

Rockbox is indeed great replacement firmware.



I haven't looked at it in a while. So, they still haven't solved the battery problem with the 5^th gen iPod- I am surprised.

.

True. However, the daily builds for the more well established targets are usually very stable. Personally I am still using a much older build with my personal and third party patches on a Sansa e200. It is rock solid.

I think battery life on the 5G is about equal in both firmwares now, if thats what you're asking.

Yes. Thanks for the info.

Now that Rockbox does something the OF doesn't do (turn off the LCD controller when the backlight is off), I can beat the OF runtime with MP3 on my 5th gen.

Hard to say. They are far far better than they used to be. I remember the original 5th gen ports getting 5-6 hours. I'm not entirely sure what I should expect out of the original apple firmware on mp3 playback, as I've never actually done any rundown tests on it, but anecdotal evidence suggests that 11 to 12 hours is about normal. Anybody done rundown tests on the original firmware?

I believe that the original 5th gen had a much smaller battery, at 400/450 mAh than most other models of Ipods, some of which went as high as 1250 mAh I think. Don't ask me why. Honestly I never go more than 6 to 8 hours between being able to charge it anyway, so it's no big deal to me.

Also, not that it's really relevant to this forum, but I did a rundown test on q6 MP3 (~128kbps) and q4 MPC (~128kbps) encodes of the same album I used for my q3 Vorbis (~112kbps) rundown tests. It was a double-length Live metal performance.

Even though new versions of the firmware can utilize the second core on PortalPlayer targets (such as the 5th gen ipod) for mp3 decoding, Vorbis still beat MP3 by about 30 minutes. Interestingly enough, MPC beat Vorbis by almost 2 hours, producing just under 14 hours of runtime.

per http://www.rockbox.org/twiki/bin/view/Main/IpodRuntime , Compact Flash hard-drive replacements add several hours of runtime, as you no longer have to continually spin up and spin down a hard drive (plus you have the bonus of no moving parts to break while running/dancing/dropping the ipod, etc). I'm hoping that whenever I finally get around to cracking this thing open and replacing the hd/battery, I'll start approaching the 18 to 20 hour range... again, not because I need to. Just for bragging rights. heh.

Your link shows 24 hours by sleeping the LCD, which is slightly more then "11 to 12 hours".



That sounds weird. Did you post the results?

That 24hr runtime you mention was for the 60gb model, I have the 30gb model. The 60gb ipod has 64mb of cache (thus it spins the hard drive up less often) and has an 850 mAh battery from the factory, vs the 30gb model's 400/450 mAh (depending on who you ask) stock battery. Between the two, it adds up to a big difference.

Andree Buschmann was able to get 18 hours out of MPC on a 30gb model, but he was using a few custom patches he wrote himself (which he can do, since he is the one who does the lion's share of rockbox MPC code). I am not sure that these patches made it into the daily release codebase or not. I certainly was not able to replicate his time, although I did have my headphones set at normal listening volume whereas he had his pretty much turned off. I could see how either (or especially both) could account for the "missing" few hours.



No, I was going to do so, but found out you had to have a wiki account with them to do so, and I wasn't about to go bother them in IRC just to post my rundown tests. I'm sure they get people doing that all the time.

I'll throw em up here if you want to take a look though.

Vorbis q6
LAME MP3 q4
MPC q4

As you'll notice, my memory was a bit off in my previous post. First, I used q4 on LAME, not q6. I actually got well in excess of 14 hours with MPC. Also, Vorbis only beat LAME by about 18 minutes (although it did beat it).

Also, in case you're wondering, the reason I used different bitrates is simply as a matter of real-world testing. It's the same reason I tested with the phones at -15db instead of -56db, I wanted to know what the Ipod could do under the circumstances in which I'd actually use it. I can't ABX Vorbis at q3, so that's what I encode for my Ipod at. I have never had good results going below 128kbps on mp3s, so that's what I encode with for portables on the rare occasion I encode to mp3. MPC... well simply it isn't optimized for crap under 160kbps, and I think q4 was actually the lowest option available to me when I was encoding them, so 128kbps it is, heh.

After the test, If I didn't already have 22gb of music encoded into Vorbis q3, I would have thought seriously about using q4 MPC instead. I may still take a look back into it after I see how SV8 performs with Rockbox... although I do still have a sentimental attachment to Vorbis for some reason.

FWIW #1,Your vorbis battery bench has an odd start offset in comparison to your MP3 one, and ends at a significantly lower battery voltage.
FWIW #2, the 60/80GB players have only a 600mah battery. 20H vs 14H claimed stock runtime when using Apple testing procedures. I always got ~10% more than that when benching the OF.
FWIW #3, -56db is not artificially low for many of us IEM users. I started benching at a higher volume simply so I could quickly hear if the player was dead or not.

The start offset is weird, I know. It seems to start counting time from when you start the plugin, but it only starts recording log entries from the time you start playing music. At least, that is the way I read the logs. for my rundown times posted at the start of the files, I simply subtracted the start offset from the last entry and used that as my total playtime. I started playing at the exact time I pulled the charging plug, so that at least should be accurate.

Regarding the difference in ending voltages, I have no explanation. I noticed that in all 3 tests the ipod shut down well before the player reached "0%". I simply assumed it was an error with rockbox's time estimations not matching my specific battery exactly. As to why the voltage levels at shutdown differed between one test to the next, I have no explanation... unless the measurements were wrong. The battery certainly was dead after the mp3 test, as well as the other two. I was interested in the time at which the machine would shut itself off, which is what happened.

I was unaware that the 60gb was 600 mAh, not 850 mAh. As I do not have one, I was going by information off of a website which was apparently misinformed.

Do you really run your IEMs at -56db? I just checked and found -58db is the absolute minimum on my Ipod. I understand that the IEMs are providing 15 or 20db of sound isolation, and they require very little in the way of power to drive them in the first place, but -56db still sounds really low. What do you use?

Andre specified that he was using the stock earbuds in his tests though, so I don't think it applies here. I think he was specifically trying to limit the amount of battery spend on driving the earbuds, which makes me wonder why he bothered to attach them in the first place. Additionally he noted that he had line out disabled, which makes both of the previous two points rather redundant anyway.

I drive mine at ~ -50 when at home.

To get a closer approximation of the runtime you appear to be in the camp which really needs to monitor the actual shut-down time. I'm assuming your battery is older and the voltage is dipping (temporarily) quite a bit when under heavy load (as in spinning up the hard drive). What you're possibly seeing as the end of the battery bench file is not the actual shut down time, but the last record written to disk before the disk spinup which triggered shutdown.

EDIT: That being said I'm not seeing disk-spinup dips elsewhere in the log I looked at more carefully.
Oh, also feel free to totally ignore the "Time Left" column. It is a calculated value and has no impact on anything.

EDIT: Oh, also, at least in the case of Rockbox (and when I tested (give me a while to look more carefully through the change log to see if there is a good reason my tests are now invalid based on functions since them)) 64MB RAM vs 32MB RAM = very little runtime change.

.
Not yet. Whether the developers get to it before Rockbox does I couldn't say.

Thanks for the info.



What do you mean here. Rockbox does support gapless playback of Vorbis files.

Edit: I guess you mean before Rockbox is ported to the FUSE.

The Rockbox port for the Fuze line is not ready yet for most people. You have to dual-boot to load songs over the USB in the official firmware because the USB drivers don't exist at this moment (and it looks like this feature may take a while to arrive due to a lack of specs on the IO chip).

http://forums.rockbox.org/index.php?topic=14064.0

Elsewhere, one of the Sansa mods said the developers are looking into adding gapless support which may or may not make it into the June(ish?) firmware update.

I understand that- I am still doing that with my older firmware for my e280. I am one of the early adopters of the e200 port and used to provide an unofficial build. However, I haven't looked at the current Rockbox in a while, since I already have all the features that I want and it is very stable. I am also too busy/lazy to update my personal patches.



Thanks for the info.