PJRC.COM Offline Archive, February 07, 2004 Visit this page on the live site |
| ||
Shopping Cart Checkout Shipping Cost Download Website |
Home | MP3 Player | 8051 Tools | All Projects | PJRC Store | Site Map |
You are here: MP3 Player Freq. Asked Questions | Search PJRC |
|
FAT32 formatting is required. With Microsoft's utilities, you generally need at least 500 megs to use FAT32. It is theoretically possible to go down to 256 megs using special formatting utilities (the minimum cluster size supported is 4k, and FAT32 requires at least 65525 clusters).
The 0.6.10 firmware scans all the directories at startup to build a list of all the files. The speed is approx 120 files/second. I recently made some improvements that get this to 140-150 files/second, and my hope is to eventually make it must faster. If you load up an 80 gig drive with 16000 five megabyte MP3 files, you'll wait approx 2 minutes for the initial scan. So today's directory scan speed put a bit of a practical limit on the number of files you can reasonably have on the drive... depending on how patient you are.
The board will also work without any SIMM at all, but it will run the old 0.1.3 firmware which has only minimal features.
The circuit board was designed for the original Micron SIMMs. In recent years, SIMM formats have changed. Most of these will work with the latest firmware revisions, but at a reduced capacity. There is no simple way to identify the SIMM format, but you can read this lengthy explaination of the technical details.
168 pin DIMMs will not fit into the 72 pin SIMM socket. There are no plans to change the design to use a 168 pin DIMM socket. The DIMM socket is longer, and would result in a longer circuit board. The large number of pins would also require using a fine-pitch surface mount chip, instead of the hobbist-friendly 84 pin PLCC Xilinx FPGA (which has just barely enough pins to work with a SIMM). If you don't have a SIMMs, we usually have some used SIMMs in stock. We test these SIMMs with the player to make sure they are compatible.
Usually we are able to ship orders made with a credit card on the next business day after we receive the order. Other payment options add some delay, and if you are mailing us a check or money order, we need to wait to receive it.
We're hoping he will provide more details for others who want to attempt this. Following PJRC's privacy policy, we haven't posted his email address (we're hoping to do this soon if he says it's ok)... but if you are serious about adding USB, we can forward your email to him.
We are working with Techknowman to offer his backlight LCD at the PJRC store, already assembled and ready to plug in.
Previously, LCD conversion kits were provided by Dean Woodyatt, but saddly he ran out of EL sheets and stopped making them. His site still has useful information and a nice step-by-step instruction page.
Also, a regular topic in the User Discussion Group is ideas and plans for cases.
Some work has been done in the firmware to create an interactive "tree list" type file selector. This is a work in progress (mostly due to hard work from Tom and Matthew). If you're interested in hacking on the firmware, this feature needs you!
If you use the board without the display, there are only buttons to move to the next and previous file. Directory navigation requires the LCD.
Some time ago group of students undertook a project to add ATAPI support to this player. They had some partial success. Here are seven pages they wrote about their project.
ATAPI CDROM drives use the same signals as IDE disk drives, but the communication protocol is different. They also use a different filesystem, ISO9660 instead of FAT32. All that is needed to support ATAPI CDROM drives is code in the firmware to use the ATAPI protocol and the ISO9660 filesystem code. When this firmware is available, it may be loaded onto existing boards to make them read CDROMs as well as hard drives.
The most common problem is inadaquete power provided to the board. The 12 volt input requires a DC voltage between 9 to 16 volts. If a laptop drive is used, the current rating should be at least 800 to 1000 mA. The low voltage input is intended for NiCD or NiMH batteries. Again, a laptop drive will consume considerable power while it's motor is spinning up to full speed.
When the board does fail, the Xilinx XCS10XL chip is usually the first chip to "blow", and it usually fails in a short-circuit manner that tends to protect the other chips on the board. Failures are usually caused by a higher voltage momentarily shorted to the 3 volt power lines on the board, or to the IDE connector pins.
For really long wires (over 6-12 inches), it's a good idea to add a small resistor in series with each line, soldered close to the board. While not absolutely necessary, this will help protect the board is the event that one of the wires shorts to, say, a 12 volt power line. 1K resistors would be a good choice.
TODO: LCD pushbuttons exist now (the answer above was originally written before the LCD/pushbuttons existed)... add info about extending the cable length. In a nutshell, you can easily extend the 4 conductor cable to mant feet (a few meters), or perhaps longer with special wires. The red and black lines should use 18, 16 or other heavy gauge wire to avoid losing voltage needed by the LCD circuitry.
Now it's true that this answer has dodged the question, but we really are avoiding making promises about when particular features will be ready. You could read through the recent news page and development history page to get an idea of the general pace that this project has.