daniel.haxx.se

One day in late 2001 I had a talk with my brother Björn and our mutual friend Linus about how their portable MP3 player was very cool but the software/firmware on it was rather limited and lame in many ways. (Correct, I didn’t own any portable music player at that time). As usual, we brought up the idea about being able to hack it yourself and oh how good couldn’t you make it then? Not very long afterwards, we had a mailing list setup for discussing how to reverse engineer and improve the Archos Player firmware. Personally, I had no device yet but it sounded like good fun so I subscribed and participated from the start. After a few months, I got myself an Archos Recorder to be able to get down on the metal too, and the Recorder was also slightly different and thus brought some more challenges to the team.

Archos Recorder

The Recorder was a step forwards since it provided better sound and a (gasp!) graphical LCD. Some of the first work I did in the Rockbox project was to work on the code for the LCD, to bring text to it using fonts and to provide line drawing routines etc. Keeping the entire screen in a separate “frame buffer” that is updated to the screen with a lcd_update() call was an early design decision that has stuck ever since.

We took the Rockbox a long way supporting more and more of the early SH-based Archos targets, including the V2s, the FM and the Ondio series. But eventually of course the models started to get hard to get and out of production. It was time to start looking into moving to other targets, and other targets would more or less force us into a world with software audio codecs!

I got my original Recorder stolen, but I had it replaced and put in a 80GB disk and it was much rejoicing.

iriver h140

We did scan the market for targets that used somewhat standard components for which we could get specs and docs and we found the iriver h1×0 series and the work began. As usual we got plenty of help from everywhere and it didn’t take too long to show the nay-sayers that we could indeed transition Rockbox into the future with software codecs. We also took it to the h3×0 series with its color screen not that long after and my golly, didn’t an entirely new world of opportunities open? 40GB of disk was just about enough to hold most of my music collection.

iAudio X5

What’s the fun of Rockbox if I couldn’t follow along? I got the iAudio X5 early on in the porting effort and joined in and got my first color-screen target. Of course I didn’t like how the mere 20GB disk narrowed what music I was bringing with me, but hey some sacrifices had to be made for the greater good of advancing Rockbox!

Rockbox was now booming and flourishing, coming to new targets all over and getting more and more developers involved.

Sansa E260

A guy from SanDisk contacted us asking about a Rockbox port to their Sansa E200 series, and even though they sent me a bunch of targets etc they never provided any docs or actual help on the effort of porting Rockbox to these babies. Not even the figuring out the firmware format, as that was instead made by our own secret super-hero MrH.

Meizu M6

Time flies and soon enough (like in the late 2007) none of the targets Rockbox ran fine on were no longer being manufactured and started to get hard to get in shops all over the world. The eternal race to get Rockbox ported to a currently manufactured model of course just got more important.

It was almost two years since I got the Sansa e200 series from SanDisk and it was time to join in the efforts of bringing Rockbox to some Meizus. The amount of interested people and the existence of a (leaked) data sheet for the main SoC helped me settle for buying this.

Cowon D2

As I’ve already explained, I bought a D2 too at the same time I got the Meizu so that I could do comparison for people and just play around some extra.

Of course, my timing is dubious as I got my two new targets exactly at the same period in my life when I went back to work (almost) full-time from having been on paternity leave for six months. Together with my extra “admin duties” such as the euro devcon and gsoc 2008 happening, I really haven’t had much time to actually dive into low-level fiddling with the ports yet. Hopefully I soon get adjusted and get some time to really help out.

Marc Guay has been doing great progress on the SanDisk Sansa C100 port of Rockbox lately, and today he showed-off some recent proofs of the state of affairs with this early and crude screen shot:

I learned about the new consortium called RF4CE (Radio Frequency for Consumer Electronics) that seems to consist of a range of the consumer electronics giants (but oddly I cannot find any actual web site explaining any details on this gang) and they’re obviously working on a new remote control standard based on the IEEE 802.15.4 standard so hopefully we’ll soon see remote controls go to the 2.4GHz band and fight the growing number of devices there, to make remote controls more functional on wider ranges than the somewhat limited infrared thingies current devices use.

Oh, and of course that’ll force us to buy new “all-in-one” remote controls soonish…

Freescale press-release, Sony press-release

My 19″ (Acer) LCD screen at work died the other day. It started flickering a lot and when I tried to switch it off/on to see if the flickering would go away, it wouldn’t go back on again.

I got a lowly 1280×1024 screen to borrow until my new screen would arrive and I got to experience that using the panning screen thing of Kubuntu (my desktop would remain 1600×1200 since I preferred not to reboot or even shutdown my X) could easily get the mouse coordinates confused and then I had to hurry up to the upper left corner to “reset” the coordinates and then I could again click on the correct items on the screen…

Very well, I got a 24″ 240BW Philips. This being a widescreen thing (they’re really taking over the world, those widescreen versions) it prefers 1920×1200 resolution. My computer had one of them intel graphics chips (82Q963/Q965) on the board and dang, I got no good resolution no matter how I poked the xorg.conf file. After trying out 915resolution, I learned that the graphics card couldn’t in fact show the necessary resolution so I had to insert a card that can. I got a dreaded Nvidia and now I’m using the evil proprietary drivers to get DVI at this resolution…

But I must say that with my 19″ at 1600×1200 and my 20″ at home at the same resolution, bumping up the screen size yet another notch to 24″ makes me think 1920×1200 is kind of on the low side for such a huge screen. I did a quick check on my favourite tech stuff compare site (prisjakt.nu) but it seems there’s no screens with higher resolutions at this size?

In the most recent Lugradio podcast Episode 19 Season 5 at roughly 1h30 into the show, we got to hear a user’s write-in explain to the hosts about the benefits of using Rockbox on your ipods. Although the hosts aren’t very impressed… They also later on mention that they did get “a lot of mail about Rockbox” so obviously it is getting quite known out there.

Back at FSCONS ‘07, I asked the guy doing the Openmoko presentation about whether they are going GTK or Qt, as his talk mentioned both and he didn’t really spell it out on what horse they were putting their money on. He then thought it was a really funny question and went on to explain how the Openmoko is like a small computer that can run anything you want. A bit like he was educating us that embedded devices do have CPUs that can run actual software. As if they wouldn’t have a main branch and a main development selecting one of these particular toolkits…

Many moon laps later, I discussed Openmoko with a friend over a few stouts at Snaxx 18, and he explained how he’d got one of the dev boards a long time ago and had kept up and tried a lot of versions of it and that it basically never worked to even make simple phone calls. He gave me the impression that perhaps the project wasn’t really that well run if it after this long still don’t have even the most basic functionality present and running stable.

Therefor I was happy to listen to TWIT 143 about the Dash Express and them telling me about it being based on the Openmoko platform. It felt like solid proof they are moving in the right direction then, so at least parts of the project must be functioning!

Then, to round it up it was with a big grin I read about the recent news that they are abandoning GTK and are now going to use E17 and Qt instead. Not a trace of any “this is like a small computer it can run anything” talk now (although I do understand that the statement was just something from this person and not any public endorsement from the project or so). This very same Ars Technica article says the first Openmoko based phone called Freerunner is going into “mass production stage next month” (that would be June 2008).

Personally, I can still see how making Rockbox run as an application on an Openmoko device would be a very cool thing.

I got the latest Linux Journal issue (#170, June 2008) the other day and while reading through it I fell over the article about a guy who found a GPL license agreement among the papers for his brand new Sony TV.

… and there he also mentioned that they use curl! Fun! Apparently in their “Bravia Internet Video Link” product. Whatever that is… “Stream, browse and watch internet video and much more on your compatible BRAVIA® HDTV with the BRAVIA® Internet Video Link. It’s like extra channels for your new HDTV.”

Robert Menes uploaded some photos of an early Rockbox port running on his SanDisk Sansa m240 (v1), and this much thanks to Dave Chapman’s early work. This port still lacks NAND driver and various important things so you still need tcctool to get code onto the target, but at least there’s evidence of progress now!

I have one of them mp3 capable mobile phones and I have a 4GB NAND flash inserted in it that is packed with music I like. Yet I never end up using it as a music player.

I see people everywhere use their phones for music and I repeatedly read and hear the soon coming death of the portable music player being predicted not far away by opinion-expressing know-it-allers.

My phone plays mp3 files just fine, but there are several reasons why I don’t use it for that. The primary one being that it gets a lousy battery run-time if I do that, and if I’d run down the battery all the way when listening to music then how would I be able to use the phone for regular voice? With a separate (Rockbox) device I can listen to music until the last drop of power goes out without hampering my communication abilities.

In my particular case, my phone’s lack of a proper standard USB port and it’s lack of anything but “full speed” (and yes full speed is less than high speed and is a lot slower than it sounds) when connecting it using the custom cable to my Linux box are two more reasons. Not to mention that it has this “database-only” approach to the music which I really don’t like – but yeah, I can learn to live with it.

Besides, it’ll be a while longer until I can hack my phone to run Rockbox and thus work the way I want it. Let’s hope Android or OpenMoko or similar efforts actually make it possible one day.

As I suspected and guessed in my blog post yesterday, Jason Kridner of Texas Instruments responded to the mailing list and confirmed that the “open platform” currently doesn’t even have a free-to-use assembler for the DSP in the DaVinci (which thus has less free tools available than the DM320 series!) and the gcc port seem to be mostly an idea so far:

I’m not aware of any solid plans on a gcc port yet, but I can confirm that TI plans to offer C64x+ C compiler and assembler tools similar to the way we provide the C54x tools for the current OSD. The restrictions and registration might not be exactly the same, but my view is that the important thing is to get something out there that any hobbyist can use for free. It doesn’t make a lot of sense for someone doing coding for use in their own living room to need to pay $3000+ for a full set of development tools when all they need is a C compiler they can run on their Linux box.

I acknowledge that Neuros really seem to make efforts to make things truly open and free, but TI’s ways are often far from straight-forward and obvious. Jason refers to his presentation from Lugradio live, but I don’t see how that clarifies anything on the openness front.