Firestore FS-100 and FS-C Hardware and Firmware comparison

[Johannes Balzer]

Hi,

I'm wondering if there is anyone here willing to open their Firestore FS-C and take some Pictures of the PCB and upload it here.

I've seen a lot of FS-100 for cheap but they are pretty much useless if you dont have a Panasonic camera. The FS-Cs on the other hand are virtually impossible to find.

So I was wondering if there is people that maybe have firmware upgrade files of both units willing to upload them as well as someone who could upload a picture of the circuit board of the FS-C.

I have opened my FS-100 which was pretty easy, and the hardware in it suggests, that the FS-C might be identical to it. But I have no means to confirm my theory.
If so, it should/could be possible to modify a FS-C firmware and convert a FS-100.

Points that speak for my theory:
1) There is no markings on the PCB that it is specific to the FS-100
2) There is a flash memory chip in there that has a Label FS-100 stuck to it, so the only difference here might be different software on the writable memory chip
3) I don't see any hardware encoders/decoders but an FPGA which is a programmable logic chip. That means what it does can be changed, and it also just has a sticky label on top that says it's been programmed to do what the FS-100 needs to do
4) Making a completely new circuit board just to support one other manufacturer of cameras would be very expensive for them. So I think it is highly likely that is the reason why they went with a FPGA instead of a fixed purpose codec chip, which makes it even more likely that the hardware SHOULD be identical.

Any help, especially the firmware update files would be greatly appreciated.

Cheers

Joe

[Johannes Balzer]

I don't know about the rules on double-posting but since I found new information and my old post doesn't seem to be editable anymore I'll just make a new one regardless.

So I found out, that the control port on the top of the FS-100 is a serial interface and I managed to pull a bootlog from it. Sadly at one point it just stops and doesn't accept any inputs through that TTY.
Also it consistently fails at one point and messes up the characters.

What this output tells me:
1) the CPU fits the label on the board
2) there is a mention of a MPEGdec driver being loaded which pushes my hopes up even higher this whole project might be doable
3) we're dealing with pretty old junk here :D
4) we might even be able to do a lot more then just change the model from panasonic to canon/sony/jvc compatibility, but also various other parameters

Also I wasn't able to find any special keys or button press combo to get into a service menu or something yet.
I'll keep trying though.

There we go:

Linux version 2.4.27-uc1 (root@localhost) (gcc version 2.95.3 20010315 (release)(ColdFire patches - 20010318 from http://fiddes.net/coldfire/)(uClinux XIP and shared lib patches from http://www.snapgear.com/)) #23 Wed Aug 1 12:55:27 PDT 2007


uClinux/COLDFIRE(m547x)
COLDFIRE port done by Greg Ungerer, gerg@snapgear.com
Flat model support (C) 1998,1999 Kenneth Albanowski, D. Jeff Dionne
On node 0 totalpages: 65520
zone(0): 0 pages.
zone(1): 65520 pages.
zone(2): 0 pages.
Kernel command line:
Calibrating delay loop... 263.78 BogoMIPS
Memory available: 233840k/245696k RAM, 0k/0k ROM (880k kernel code, 5224k data)
kmem_create: Forcing size word alignment - vm_area_struct
kmem_create: Forcing size word alignment - mm_struct
kmem_create: Forcing size word alignment - buffer_head
kmem_create: Forcing size word alignment - filp
Dentry cache hash table entries: 32768 (order: 6, 262144 bytes)
Inode cache hash table entries: 16384 (order: 5, 131072 bytes)
kmem_create: Forcing size word alignment - inode_cache
Mount cache hash table entries: 512 (order: 0, 4096 bytes)
kmem_create: Forcing size word alignment - bdev_cache
kmem_create: Forcing size word alignment - cdev_cache
kmem_create: Forcing size word alignment - kiobuf
Buffer cache hash table entries: 16384 (order: 4, 65536 bytes)
Page-cache hash table entries: 65536 (order: 6, 262144 bytes)
POSIX conformance testing by UNIFIX
Initialize Multi-channel DMA API Alpha v0.3 (2004-04-26)
Bad boy: MCD-DMA (at 0x01088de4) called request_irq without a dev_id!
Linux/m68k PCI BIOS32 revision 0.06
ColdFire PCI Host Bridge (Rev. 0) detected
init_coldfire_pci ffffffff ffffffff
PCI: Probing PCI hardware
ColdFire PCI - coldfire_fixup
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
kmem_create: Forcing size word alignment - sock
Starting kswapd
kmem_create: Forcing size word alignment - file_lock_cache
fwxdc: module loaded
fsxev: driver loaded (50)
fsxtl: driver loaded (52)
mpegdec: driver loaded (53)
ColdFire internal UART serial driver version 1.00
ttyS0 at 0x10008600 (irq = 99) is a builtin ColdFire UART
ttyS1 at 0x10008700 (irq = 98) is a builtin ColdFire UART
ttyS2 at 0x10008800 (irq = 97) is a builtin ColdFire UART
ttyS3 at 0x10008900 (irq = 96) is a builtin ColdFire UART
kmem_create: Forcing size word alignment - blkdev_requests
Blkmem copyright 1998,1999 D. Jeff Dionne
Blkmem copyright 1998 Kenneth Albanowski
Blkmem 1 disk images:
0: 15F628C-186D68B [VIRTUAL 15F628C-186D68B] (RO)
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
FS driver loaded
SCSI subsystem driver Revision: 1.00
ohci1394: $Rev: 1045 $ Ben Collins <bcollins@debian.org>
custom_config_rom, PCI slot 00:03.0
ohci1394_0: OHCI-1394 1.1 (PCI): IRQ=[69] MMIO=[d0000000-d00007ff] Max Packet=[2048]
ISO = slot 3
ohci1394_1: OHCI-1394 1.1 (PCI): IRQ=[70] MMIO=[d0008000-d00087ff] Max Packet=[2048]
ASY = slot 4
sbp2: $Rev: 1074 $ Ben Collins <bcollins@debian.org>
ieee1394: sbp2: Driver forced to serialize I/O (serialize_io = 1)
ieee1394: Loaded CMP driver
FAT: bogus logical sector size 29808
FAT: bogus logical sector size 29808
VFS: Mounted root (romfs filesystem) readonly.
Freeing unused kernel memory: 36k freed (0x11b8000 - 0x11c0000)
Flash Lock Register Bits = 0xfd
We did not Reinitialize the DDR-RAM
ctx 01bce000
ctx 01bd4000
hpsb_bus_reset - 69
ctx 01be0000
ctx 01be6000
hpsb_bus_reset - 70
rootfs / rootfs rw 0 0
/dev/root / romfs ro 0 0
/proc /proc proc rw 0 0
PID Uid VmSize Stat Command
1 0 SW /bin/init
2 0 SW [keventd]
3 0 RWN [ksoftirqd_CPU0]
4 0 SW [kswapd]
5 0 SW [bdflush]
6 0 SW [kupdated]
7 0 SW [knodemgrd_0]
8 0 SW [knodemgrd_1]
9 0 RW /bin/sh /etc/rc
13 0 RW ps
Welcome to
____ _
/ __| | _ _ _
_ _| | | | |_|____ _ _/ \/ \
| | | | | | | | | _ \| | | \ /
| |_| L| o|_a_|d | _|/ |b |i |n |/ |m_|i /d d l\
e
W| a _r__e\_:__ _|T__E_|X_|T_|= |f_|c\_9__6_\0_/4\_0/
-
f|_d/
9

a
F5orc f0ur thDerA iTnfAor=maftidon9 cahe5ckc:
4
h-ttfp:d//dww5w.fuC9Li4nu x.BorSg/S
=
fFIdREdST5ORfE
9
h4tt-p:f//eww3w.2FO1CUdSi4nf
o.
com
Address of control panel data = 0x011AB764

serialPortLoggingDisable()


**** Power ON Date: 01-05-00 Time: 08:0m0:i23d d *l**ehpWsba_bruse_r:es eta -l 7l0
o
*c
a
t i o n T raockf t ab4le0 s8iz0e 0= 90p00a0 g(1e00s0 enftraieis)l

e
d!
Isoch alloc'ed 20 regions of 8355840 bytes at:
0x05898000 0x06090000 0x06888000 0x07080000 0x07878000 0x08070000 0x08868000 0x09060000 0x09858000 0x0A050000 0x0A848000 0x0B040000 0x0B838000 0x0C030000 0x0C828000 0x0D020000 0x0D818000 0x0E010000 0x0E808000 0x0F000000
fwSetMode 1
******************** IsoUserSplash = 0x012724A8
Switching to PAL
Using new dispatcher

scsi0 : SCSI emulation for IEEE-1394 SBP-2 Devices


=> EXT config:2
DiskUnmount: Short circuiting RBS



=> iphyreg 4294967295
DVConfigure called.

Allowing 24p advanced

END

ieee1394: sbp2: Logged into SBP-2 device
ieee1394: sbp2: Node 1-00:1023: Max speed [S400] - Max payload [2048]
NodeMgr: setting gap count to 4...Done!
Vendor: HTS54101 Model: 0G9AT00 Rev: MBZO
Type: Direct-Access ANSI SCSI revision: 06
Attached scsi disk fwxa at scsi0, channel 0, id 0, lun 0
SCSI device fwxa: 195371568 512-byte hdwr sectors (100030 MB)
Partition check:
fwxa: fwxa1
SCSI device fwxa: 195371568 512-byte hdwr sectors (100030 MB)
fwxa: fwxa1
Local GUID is 0x0002EA0200027504
Grade Restored

[Johannes Balzer]

Another day and some new findings:

I connected my laptop with a RS-232C to USB adapter and a makeshift 3.5mm TRS-Jack to DB9 adapter to the Firestore FS-100. Then I established a connection using screen at 19200 baud.

I was able to interrupt the boot process simply by pressing ctrl+c just until the text stopped scrolling.
And bingo I was in a busybox shell.

now I was missing the mounted harddrive. but the only way to access the hard drive was to run the /bin/middleWare. But this would lock me out of the serial console.
So I started the middleWare in a separate process instead of a child process by running it like this:

/bin/middleWare 2>&1 &

et voila I was able to retain shell access and had the internal harddrive mounted to /mnt/VOL01

so now I could use DD to make images of /dev/rom0, /dev/ram0 and /dev/ram1
Turns out there is not much interesting stuff to see in there.

So I extracted the romfs from a Firestore FS-C onto the FS-100 internal hard drive and from there I was able to kill the middleWare process that came with the FS-100 and execute the middleWare binary that came in the FS-C firmware upgrade

It didn't really do much, but in the on-screen menu of the FS-100 there was now the Firmware Image version of the FS-C firmware instead of the FS-100.
Sadly though everything else remained the same, and also I still was unable to execute the FS-C firmware upgrade.

Also what is very strange:
There is a 128Mbit NOR flash memory on board, but /dev/rom0 is only about 2.5MByte in size.
There is also the binary sfdisk, but that says it is unable to recognize the true size of /dev/rom0.

Also I tried making the rootfs writable with the following command:

mount -o remount, rw /

which indeed changed the mountpoint to being writable.
Unfortunately though, the filesystem is romfs which is intrinsically seen as non writable by the kernel.

Has anyone ever tried to work on one of those things?
I'm kind of running out of ideas.
Given the age of that stuff, there is not that many resources around to try and figure out how this old stuff used to be made and software developed for it.

If maybe one of the old Focus Enhancements employees is lurking around here, maybe you could point me in the right direction?

Or maybe someone could send me a picture of a FS-C mainboard so I could see if they are the same or not and if my efforts are not completely useless? :D

Cheers

[Andrew Smith]

I'm pretty impressed with your level of computer smarts which is well beyond most video people. Also nice to see that the developer included a bit of ASCII art in their screen output.

I'm curious as to why you'd be wanting to get one of these things working again or do an update? It's from the days of standard definition and whilst you could update and have HDV (which uses the same data rate) stored, I don't know that it's really worth it. HDV was an amazing achievement back in the day but it's heavily compressed and not the full HD resolution.

For the same inconvenience of having an external recorder you might as well have an Atomos Ninja or something similar. A lot of modern cameras also enable you to have the same acquisition quality in-camera.

Andrew

[Danny Gates]

Joe,
I also have an FS-100 and am looking to get it working with a Canon XL1S system.
I have a copy of the FS-100 firmware version 3.0.0.04011507. I am more than happy to send it to you if you need it. I understand there there is a newer version out there somewhere which would be the latest and final version, version 4.0, but I have not been able to get my hands on a copy. I also cannot find any firmware related to the FS-C and would love to have a copy if you have it.

I am keenly interested in the work you are doing, I know the hardware is old, but it is what I have and also fits my needs perfectly, so while I know technology has moved on, I still like what I have, but want to have the ability to record digitally rather than by tape.

Has anyone been able to locate email addresses or contact info of any of the original FireStore design team?
Did you find any contact info in the dump file? I tried to contact Greg Ungerer, but the email address just bounces back and I don't think he was actually connected to the FS hardware itself, but just the Linux that runs on it.

Anyway, please don't give up on the work, there are others out here that would like to see this work.
Let me know if there's anything I can do to be of help.

-Dan

[Danny Gates]

Also, what exactly is the reason that the FS-100 won't work with the Canon? Is it just a matter of file format?

[Johannes Balzer]

Hi,

mmh well I actually obtained a FS-C firmware and it was not very helpful because most parts of the firmware are actually not in the update but in a (to me) inaccessible part of the flash chip.
The reason the FS-100 is not compatible with canon, sony or jvc is, because panasonic used a different codec with DVCProHD than all the others in HDV.

To understand the difference it is important to know where the formats originate: In Tape! So everything on the tapeless side has to adapt to the weirdness that came along with the tape weirdness and the bandwidth limitations the slow tapes brought along with them. Normally you can record at around 25MBit/s on a DV tape. But that is way too little for HD footage, so:
HDV uses MPEG2 compression to get the HD footage on a DV tape while DVCProHD is doing super weird stuff like running the DV Tape faster, using bit wider tape, and just being 1440*1080 instead of 1920*1080 and also splitting the picture into 4 quadrants and encoding each one of them in a separate stream and doing all kinds of weird stuff with it.

In the old days you either had to implement a specific hardware codec chip or your device would not be able to handle the computational power required for the task.
There is another way of doing that and that is reconfigurable hardware in the form of an FPGA, and I think that's what they did there.
Think of an FPGA like a calculator that can only solve one equation but really super mega fast. You can reconfigure it to do another one but that takes some time to do and you have to restart the whole computer for it.

The FS-C firmware update file does absolutely nothing on its own with the FS-100
also it is just a patch, it is not the full firmware.
But if you patch your FS-100 to version 4, you should be able to record with your XL1S, at least as long as you have a tape in there, otherwise it won't let you press the record buttoon.

But I already gave up... I sold them and bought a Atomos Samurai.

[Danny Gates]

Any chance you have a copy of the FS-100 V4.0 firmware? I have not been able to obtain that version.

And thanks for all the info. I am an electronics engineer by trade, so I understand that part and also a little of the firmware/software part. I will post my email address if you have a copy of the v4.0 firmware you wouldn't mind sharing.

[Johannes Balzer]

Hi,

sorry for the late reply.
Here is the link:
https://www.driverguide.com/driver/d...iverid=1837290
It says driver, but it is indeed a firmware update
Just unpack it und copy the fs100.bin on the device.
Then you can go in the menu of the fs100 and perform the update