Sorry for double posting but I thought this needed a first and last bump.

Apparently nobody knows or cares too much about the mindex stuff; but I do want to get this working and have a hunch that a dash launch plugin might be the easiest way.

I usually code in C# and PIC-C and have access to a full devkit with sidecar and a Jasper JTAG.

The only problem: I have no idea how to write a dash launch plugin and so far the readme and googling haven't yielded anything, could anyone link something useful?

I was wondering if anyone has any additional info/resources concerning the format the xbox uses to store user music, which is apparently just wma and a new header, plus an index file.

I've been looking into the mindex.xmi for a little bit and can construct one up to the last type 8 record (more or less) but there are a few things I'm still missing, for example:

- I don't understand what value is supposed to be in type 8 block #2 (the one that contains the total number of tracks at 0x18 - 0x1B) at 0x10-0x13 and 0x14-0x17; I first thought 0x13 was the last filename but I have one mindex that contains <last filename - 15> there (last filename: (0x)0053; total number of tracks: 65; value there: 0x44); sometimes the 0x17 value is the same as the 0x13 value, sometimes it isn't.
- I haven't found out what determines the record order within the mindex yet; probably haven't looked hard enough, or it's more or less random.

So far I've only found these two threads with info about the subject, does anyone have anything else they would share?

Alternatively: the user music player can play MP3 files from USB drives iirc, shouldn't it be rather easy to make some dedicated part of the hard drive (\music) appear as a USB drive to the user music player so we can just play MP3s?

Couldn't find any thicker wire for pic#2, couldcha? And then only 1/10th of it is soldered. Your other joints look nasty too. I'm not usually one for making fun of other people's soldering but dude... train a little. And ffs don't sell stuff like that!

Not saying that's necessarily your problem, but clean that $#!t up man, it won't hurt. And stop twisting wires, and using ones that are 5x as thick as needed. The blue cables are definitely thick enough.

Looks OK to me even though you might want to clean up your soldering and hot glue a little bit; as for the speed, that sounds pretty standard for a 2.5inch drive, but why not just attach the HDD to any desktop PC you have sitting around and compare?

I think I could use some help, I've read this thread many times but I don't see what I could have missed here.

I'm trying to RGH a buddy's slim but can't get past the CPLD programming part.

I have:
- white chinese glitch chip for slims
- LPT JTAG programmer built according to GliGli's schematics (using 15cm cables, 3 1N4148 mini-melf diodes [black stripe on the CPLD, NOT LPT side], 1 1206 220Ohm resistor and 4 0603 3.3k resistors)
- Trinity on 12625
- Intel D945GCLF2 mobo with LPT1 @ 0x378

I'm not sure whether the CPLD is programmed yet, according to some posts here and in the DX forums it is, however other people in the same forums claim it's not.

So I figured I'd rather program it first.

Now I have already installed the chip in the console, meaning 3V3 and GND are already connected (confirmed with multimeter at J1/J2 and VCC/GND of JTAG port); the console is connected to a power outlet and the readings say 3.3V.

I've already flashed XeLL built with MultiBuilder 0.5 (old, I know, upgraded to 0.93 by now - and yes, I did dump the NAND beforehand and checked my 3 dumps were identical).

The console turns on but keeps hanging at a green light (which is what I would have expected from an empty chip, no glitching within 5 minutes, tried 5 times).

As for the LPT port, in the BIOS I have 4 possible modes ("Output only", "Bi-directional", "EPP" and "ECP") which all give identical results concerning the CPLD, JTAG Cable and the error message from 360gcProg v1.4.

If I solder the chip in according to the official slim diagrams on the manufacturer's site (with 50cm CPU_RST "fix") but do NOT connect the "slim" bridge (between D1A and C5) I get "TDO stuck at 1".

It says "TDO stuck at 1" too if the programmer is not connected to the LPT port, so I guess that's the default "nothing there" error message.

If I do bridge the "slim" points I get a message saying the CPLD was not powered correctly and to check the wiring, however I've done that quite a few times now.

I've also measured the voltage between LPT port pin 1 and pin 18, it's 5.02V, so afaik it should work with the normal LPT JTAG programmer.

As I said I've also tried switching the LPT port modes in the BIOS, no matter what I set I get the above error message (not powered correctly, check wiring; I forgot the exact wording).

Then I've read Xumpy's post about bridging R2 or R3; I've tried bridging R2, bridging R3 and bridging R2 and R3 (not together, separate) but the error message always stays the same.

I think I'll try programming with Xilinx Impact next as that's the only thing I could still imagine being the problem.

Also note I've already hot-glued the chip to the xbox so I'd like to keep it in there if possible, though I might have to change the wiring to take CPU_RST and POST_OUT1 to the top.

Some pictures (click to enlarge; blue wires go to NAND programmer, yellow wires to glitch chip, colorful wires to LPT JTAG CPLD programmer):



Any ideas are greatly appreciated.

EDIT: Holy $#!t, turns out I had more luck than brains today: the CHIP IS INDEED PRE-PROGRAMMED! Bought from DX around 2-3 months ago.

I thought - why not give it another try, as I don't seem to be too lucky with that LPT JTAG Programmer (desoldered that and the bridges at R2 and R3, kept the "slim" bridge). And f*** me, after 20 seconds my display was tinted in blue.



If someone happens to find the problem with my programmer (?) I'd still appreciate that for the future but at least my buddy's gonna be happy now

P.S.: If anyone wonders why I said it wouldn't glitch before - well, that's what I had thought; now that I think about it it comes to mind that my buddy was over here yesterday and I soldered all of the stuff then. We were drinking and watching old music vids, so I switched to the PC via HDMI switch; and when I switched back to check, obviously the console must have rebooted (as it usually does when losing and then gaining HDMI connectivity again); that's why I thought it didn't work. Shouldn't drink heavily while trying to glitch, even though the soldering does seem ok in retrospect so at least there's that

While I'm at it... my buddy says thanks to gligli, Tiros, cjack, Redline99, arnezami, tmbinc (who is sorely missed in the 360 scene, see you when Durango hits...) and all the others who made this possible.

EDIT2: In the meantime I had a chance to build a 14719 RGH image with MultiBuilder 0.93; I flashed it to the console, read it back, found no differences, put the xbox back together and voila - always (at least so far) boots within 10-30 seconds.
I'm tempted to call bull$#!t on the "can't put in metal cage, screws up RGH" and "you need do try 50 different caps" stories... the thin wires I used worked just fine and I was able to screw everything back together witout any problems; no cap change either, only used the 50cm CPU_RST wire (didn't feel like playing around with that so maybe that's unnecessary as well).

EDIT3: After 15 boots times have varied between 10 seconds and 2 minutes 10 seconds, 10 of those times it was below 30 seconds.

Why not just use any old AT(X) or small new PSU for <= 5€/$ instead? I mean - do what you gotta do, but that sounds like an accessory that's useful to exactly 7 people in the world.

Yeah so I'm aware lots of people have done this before me but I recently found a reason to resurrect my own "360pc" project (need parallel port, main PC has no more PCI slots).

Initially it was going to contain a 15" display in the upper case half (where the 80mm fan is now) but unfortunately that cracked right in the middle when I moved last year.

So anyway, here it is, a triumph of free spirited german engineering; main ingredient is hot glue; I didn't feel like wasting too much time on this (and it shows, but I don't care).

All cuts were done with a cutter knife that I heated up with a lighter; built during the night so no Dremel.

I might paint it some day as it does look pretty $#!tty but definitely not as long as it's -10°C outside.

Components that I had lying around:
- Ugly old Xbox case
- Mobo & CPU Intel D945GCLF2 Mini-ITX
- PSU Delta Electronics DPS-125DB A Rev. 00 (125W) (bought new for 4€, unfortunately sold out now... you could also try the 85W version, the SEASONIC SS-200SFD or the HIPRO HP-K1517F5 for <10€/$)
- RAM 2GB Kingston
- HDD WD 120GB from an old Xbox
- Ancient 80mm fan (PSU is built sandwich style in its original casing, would overheat without additional fan in Xbox case)
- USB header connector for 2x front USB
- 20 to 24 pin ATX power adaptor (turns out I could leave that out, board works fine with only 20 pins too)
- 5.25" molex power to 4 pin P4 power adaptor
- $#!tty smart phone to take $#!tty pics with

Power button and front USB work as expected, connect button is now reset button; laptop DVD burner lying around but missing adaptor, would easily fit.

Might use MMC slots to make external SATA and power connectors.

Runs fine with XPSP3, no display/HIDs connected, controlled via VNC (after re-installing XP).



















Regards

Would a retail game triggered reboot (on 9199, if it was reproducable) do any good? Judging from arnezami's rebooter thread it might but I'm not sure. A simple no will suffice, thanks.

Guess not.

Alright, I took a little break from trying to get this to work and I think I found the problem now... I'm just not sure how to solve it yet.

Yes, I can flash all my other 360s with this flasher and the same cables.

Well the box doesn't boot anymore and when flashing the NAND the SB doesn't get any hotter than the rest of the PCB (according to my IR thermometer 26,6°C).
The NAND looks ok as far as I can tell, took those with a cheap microscope:



Tried it but didn't seem to change anything.


I looked at the trace on the front of the board and it seems ok; on the bottom of the board I messed up when first trying to install the Cygnos v2 and removed some of the green laquer protecting the PCB, thus shorting it with RB. I have removed the short a few weeks ago though (just checked again, no connection).


That's what happened back then. But there's no short there anymore, however I think I've found the culprit now...


This happened back when the 'official' ground points were the big golden pads the DVD drive sits on. And there is a short, although it doesn't seem constant (the reading jumps around between 0.01 and 0.68kOhm, but it sometimes jumps to '1' for half a second too).
Any ideas how I can get rid of that short? I've tried with different kinds of desoldering braid, a desoldering pump and up to ~300°C on the iron without any luck beyond what is already removed.


Just for $#!ts and giggles, here are some of the errors in NandPro:



Thanks for your help guys.

Thanks for the replies but I'm already using a USB flasher.

As far as the shorting goes - i did it the right way (unplug PSU, wait a few mins, short pins, plug in PSU, wait a few secs, un-short the pins, try to write NAND) and over 9000 times now. Sorry, I just noticed I formulated it wrong in my first post.
I've stopped trying now since you confirmed what I had read in other threads, just thought it couldn't hurt.

Blackaddr's thread seems to be about 'buffered LPT' related problems, not USB. And I don't even get to block 0x200 anymore.
Wrong thread, you meant the one in the hacking forums... I don't have any transistors lying around though and I'm not sure I want to spend any more money on this box.
I could try removing the JTAG wiring of course and see if that makes a difference, I'll do that next. I think I did that a few months ago too though... we'll see.
So it shouldn't make a difference in my situation (at least yet) anyway, right?


I've redone the JTAG wiring a few times, using new diodes and wires every time. Redid the NAND wiring too.
I'm 100% confident it's not my wiring, I've done this often enough and the flasher with the same cables works on all my other consoles (only have Jaspers and Xenons but yeah).

Can I dump that 'idea' about the PRE pin? I'm probably misinterpreting the datasheet but the way I see it you can block a range of, uh... blocks from reading/writing?
If there's a short somewhere that causes 'random' values to be sent as the limits of the blocked blocks (ugh), wouldn't that explain why most times I can't write any blocks, one time until 0x5C, one time until 0x1FC and a few times until somewhere below 0x20?
Sounds pretty far fetched when re-reading though...

OT: If this box won't come back to life I have another Xenon I could give to my buddy, even with Freeboot0032 already on it, but it also has RRoD 0102.
I have a heatgun, preheater and cheapo chinese ebay reballing kit I never used before and read the tuts/threads on the subject on XboxExperts, some of those guys used the same cheap AOYUE station I have.
Are those tuts for reflowing (actually I'd like to try reballing but the process would be roughly the same I suppose) any good/accurate/are there better ones? The link to the MannCorp PDF there is down and the other tuts I found were mostly about how to load profiles on IR stations...

Excuse me if any of this sounded edgy or anything, this old box is starting to become a giant waste of time though.

Yeah hi, I'm new here - well, not really.
I've been trying to get my oldest Xenon to work but have all sorts of trouble with the NAND.

This was basically my test box for everything, had XeLL on it, then a Cygnos, then XBR... now I've desoldered the Cygnos and am trying to bring the box back to life.
To the best of my recollection the box wouldn't boot at all anymore (w/o error code), the PSU showed a normal orange light though.

When reading/writing/erasing the NAND I usually get error 0 or error 8021 on every block (USB flasher).
Despite getting errors on every block my first dump of this round of trying to fix it contained valid data, including the MS copyright string.

Also, sometimes I can read/write/erase parts of the NAND... one time i got to block 5C, one time to 1FC and a few times something in between, then errors on (more or less) every block.
IIRC the last time I tried to fix this box a few months ago I could (sometimes) read/write/erase the NAND, but only the first 200 blocks (like Blackaddr recently mentioned in another thread here, but I don't even get that far anymore).

I also tried shorting pins 41-44 on the NAND for 3 secs after a few seconds before and a few seconds after plugging in the PSU (after a few mins w/o power). I tried it dozens of times by now, shorting for 1-15 secs. Tried the alternative pins too (29-32 according to the datasheet?). I think shorting those pins doesn't even apply to my case since the NAND always gets detected by NandPro, and the flashconfig is also correct (01198010) but I figured I didn't have much to lose.

Then there was this thing with the Cygnos toolbox a few months ago, that didn't want to write the onboard NAND without problems anymore either afair; but it had some sort of option to disregard the 'bad block' messages it seemed to get from the NAND on every block... can this be done by hand/without the Cygnos too? Or is that the same as just leaving NandPro (2.0e) running if it errors on every block?

I checked and re-did all connections multiple times, I can't see any shorts between the NAND pins either (and never soldered directly to it anyway so didn't bother to whip out the multimeter) and cleaned all relevant areas with a toothbrush and alcohol many times.
The flasher works just fine with other 360s.

The datasheet mentions a 'Power On Read Enable' pin to 'enable and disable the Lock mechanism and Power On Auto Read', that couldn't have anything to do with this could it? Could this explain why some people said they pulled a screwdriver across all pins on the right side of their NAND to get it working again?
It' the only pin on the right side of the NAND that's connected (on the top of the 360 board) aside from the I/O pins btw, unless the second VCC/VSS are connected below the NAND.

Next I was thinking of replacing the NAND but figured I should ask here first before I rip the NAND off the Cygnos v2/out of a 8955 box w/o dvd key I was given.

This box with XBR/Freeboot would make an awesome birthday present for a buddy. Any pointers appreciated, thx.