|
cory1492
|
 |
« Reply #340 on: December 15, 2009, 01:11:13 PM » |
|
block 103 is 0x42C600 to 0x430800, if it's bad that can easily account for the differences - ecd usually can only correct single bit errors. 12 MHz, +-0,4% Keep in mind a 12MHz is divided to get 4MHz PLL and then divided to get 48MHz for usb and 24MHz for MCU, inaccuracies are multiplied and spread. Prefer an accurate crystal over a resonator, I guess. There are a couple people who have had to fiddle with the Vusb cap, but usually that is related to windows not finding it at all rather than dropping connections. Vusb in this instance shouldn't be needed, so as the datasheet says, "An external 220 nF (±20%) capacitor is required for stability."
|
|
|
|
|
Logged
|
|
|
|
|
Caesar555
|
 |
« Reply #341 on: December 15, 2009, 04:17:36 PM » |
|
First Post  I built the programmer and the flasher (according to this pic  ) except for the fact that I'm using an 18F2550 Pic The programmer was recognised, but the pic wasn't. Reding was possible, writing wasn't. So I built an arts2003 and that seems to work. Flashed it with pixboot.hex from PICFLASH_v3b_plus2.zip. Moved the pic to the flasher and attached it to an usb-port. It got recognised as unknown device, but it wasn't possible to install the drivers. Flashed the pic again with PicXBoot_12.hex from PicXFlash_V3.zip. After moving it to the flasher and attaching it to usb again I could install the drivers. It got recognised by PDFSUSB.exe an I flashed it with PICFLASH.hex from PICFLASH_v3b_plus2.zip. Reattached it and it was recognised as MemoryAccess. Installed drivers and everything seemed ok. But if I start nandpro with "nandpro usb: -r16 backup.bin" I get: Looking for usb interface device Could not detect a flash controller Fatal Error Should I have changed something somewhere before flashing or am I forgetting something 
|
|
|
|
|
Logged
|
|
|
|
|
dak
|
 |
« Reply #342 on: December 15, 2009, 04:35:36 PM » |
|
I am quite sure now that the issue was the resonator, because I used a crystal this time [...] and it works perfectly. Prefer an accurate crystal over a resonator, I guess.
I used a crystal from an old USB mouse, works perfectly now. Only thing i noticed is that i get a "bad flash config"/"cant find flash controller" when i let the flasher plugged in too long. I then have to plug it out for about 1min and i can read again. (@ Ceasar555 try this) Seems it stays stable long enough to dumb 256mb.
|
|
|
|
« Last Edit: December 15, 2009, 04:43:29 PM by dak »
|
Logged
|
|
|
|
MotoMax
Newbie

Posts: 7
|
 |
« Reply #343 on: December 15, 2009, 05:17:46 PM » |
|
I had this error and it was because one of wires to xbox got loose. The PIC can'y detect your Xbox. But if I start nandpro with "nandpro usb: -r16 backup.bin" I get: Looking for usb interface device Could not detect a flash controller Fatal Error Should I have changed something somewhere before flashing or am I forgetting something 
|
|
|
|
|
Logged
|
|
|
|
|
Caesar555
|
 |
« Reply #344 on: December 15, 2009, 05:39:42 PM » |
|
Hm. I'll try it. Never thougt it could be the xbox-wireing. Errormessage should say "No Flash Chip found". thought the flasher was faulty... Does the xbox powerplug need to be plugged in or should the box even be on (thougt I read it could happen that it switches on when connecting the flasher (and this doesn't matter) Thanks so far. I'll keep you informed 
|
|
|
|
|
Logged
|
|
|
|
|
maximilian0017
|
 |
« Reply #345 on: December 15, 2009, 05:56:40 PM » |
|
block 103 is 0x42C600 to 0x430800 Are those 200 extra bytes the ecc bytes? Is it possible for the PIC to have a small switch that can select between serial com and the USB flasher?, in other words can you change the vid/pid on the go or do we need to flash every time? Powersupply has to be connected, Xbox should be off when reading/writing. Oh, and for the stats , have a 18F2455 Dip version
|
|
|
|
« Last Edit: December 15, 2009, 06:05:54 PM by maximilian0017 »
|
Logged
|
|
|
|
MotoMax
Newbie

Posts: 7
|
 |
« Reply #346 on: December 15, 2009, 06:01:13 PM » |
|
If it can't see your USB flasher it will say "No device found." The Xbox power/brick needs to be plugged into the xbox but Don't turn the xbox on. The xbox should not start by itself. Rremember to unplug power for say 10secs after Writing to Nand. Hm. I'll try it. Never thougt it could be the xbox-wireing. Errormessage should say "No Flash Chip found". thought the flasher was faulty... Does the xbox powerplug need to be plugged in or should the box even be on (thougt I read it could happen that it switches on when connecting the flasher (and this doesn't matter) Thanks so far. I'll keep you informed 
|
|
|
|
|
Logged
|
|
|
|
|
docentore
|
 |
« Reply #347 on: December 16, 2009, 03:56:12 AM » |
|
Thanks so far. I'll keep you informed  Please do. I have the similar problem. My flasher is being recognised on 2 different pc's as MemoryAccess, which is good I presume. But when connected to xbox I'm getting the same error. I'm going to check wiring on xbox during the weekend, but it was working fine with lpt plug (did the rj45 mod for easier access - I might mess something up with wires) Thanks
|
|
|
|
|
Logged
|
|
|
|
|
Caesar555
|
 |
« Reply #348 on: December 16, 2009, 05:27:35 AM » |
|
Mine seems to works now. Powercord was in the box, but not in the power outlet  I read the nand 6 times on 2 pcs, but get always an error. On the bright side its always the same. R:\NandPro20a\NandPro20a>nandpro usb: -r16 nand2.bin NandPro v2.0a by Tiros Looking for usb interface device FlashConfig:01198010 Starting Block:0x000000 Ending Block:0x0003FF Press any key to begin reading nand Error: 254 reading block 22A 03FF edit: once i got error 250 instead of 254. same block  What I find odd is that degraded says (note2), the bad block is found at 0x3ff. but 0x3ff is the last block of the dump  Maybe a bad block? Can I ignore this? Is it possible to extract that block from another console I have?
|
|
|
|
« Last Edit: December 16, 2009, 05:32:18 AM by Caesar555 »
|
Logged
|
|
|
|
|
iLLNESS
|
 |
« Reply #349 on: December 16, 2009, 05:37:57 AM » |
|
Mine seems to works now. Powercord was in the box, but not in the power outlet  I read the nand 6 times on 2 pcs, but get always an error. On the bright side its always the same. R:\NandPro20a\NandPro20a>nandpro usb: -r16 nand2.bin NandPro v2.0a by Tiros Looking for usb interface device FlashConfig:01198010 Starting Block:0x000000 Ending Block:0x0003FF Press any key to begin reading nand Error: 254 reading block 22A 03FF edit: once i got error 250 instead of 254. same block  What I find odd is that degraded says (note2), the bad block is found at 0x3ff. but 0x3ff is the last block of the dump  Maybe a bad block? Can I ignore this? Is it possible to extract that block from another console I have? thats how the nand works. the bad block is at 22a, but it was remapped to 3ff. a 16mb nand has 16.5mb of space. the first 16mb are the regular data, and the last 512kb are for remapping bad blocks/etc.. atleast, as far as i know. if you get another bad block im guessing its remapped to 3fe.. that or the 22a block is mapped to 3fe and the new bad block is remapped to 3ff.
|
|
|
|
|
Logged
|
|
|
|
|
cory1492
|
 |
« Reply #350 on: December 16, 2009, 07:55:11 AM » |
|
iLLNESS the 0.5M 'extra' is spare data area (each block has 512bytes of that 0.5M associated to it) and is needed to stow error correction hashes and things like LBA numbers, any blocks left for wear leveling or bad block relocation are in the normal 16M area. With 360, it seems they left some empty ones at the tail end of NAND for some cases such as that one (data outside lflash needed remapping.) block 103 is 0x42C600 to 0x430800 Are those 200 extra bytes the ecc bytes? Is it possible for the PIC to have a small switch that can select between serial com and the USB flasher?, in other words can you change the vid/pid on the go or do we need to flash every time? A block is 0x4200 bytes in a dump, basically a page is 512bytes user data with 16bytes spare data following, 32 pages per block on small block nand. Can use dual device setup, it's documented in the 18F sample code from microchip, the samples are usually CDC (virtual com port over usb) and MASS storage. I'm pretty certain at this point the issue comes down to unstable resonator. Even with an unstable resonator one could dump/write in chunks instead of all in one go. But when connected to xbox I'm getting the same error. Tends to work best: unplug everything from xbox and usb, connect pic to xbox, plug in xbox, plug in pic to pc.
|
|
|
|
|
Logged
|
|
|
|
|
dak
|
 |
« Reply #351 on: December 16, 2009, 08:24:25 AM » |
|
I changed the 220n cap on vsub to a 473n. I am reading my 256MB now the third time in a row without a problem. I think it was both. Using a crystal resonator plus the bigger cap. Thanks to those who helped.
|
|
|
|
|
Logged
|
|
|
|
|
Caesar555
|
 |
« Reply #352 on: December 16, 2009, 08:26:02 AM » |
|
ok thanks. so nothing to fear? dumped another console without errors. will use that one for testing. (hopefully) one last thing: when comparing my dumps in 360 flash dump tool my console with the bad block has no pairing information and the 3 LDV-values are 0  the box without dumperrors has information there. normal or bad? edit: when trying to write I got Looking for usb interface device Incompatible ARM hardware for this version of NandPro! replugging the flasher in the usb-port did the trick flashed without errors. now I get E71 after powering it on  but thats for another topic edit2: E71 was because I powered the console before doing the jtag-hack. noticed that (e79). did the jtag-hack and powered it on again. after reading the nand again there differences to the one I flashed. Flashed it again and now it works 
|
|
|
|
« Last Edit: December 16, 2009, 12:21:08 PM by Caesar555 »
|
Logged
|
|
|
|
|
gepal
|
 |
« Reply #353 on: December 17, 2009, 05:37:17 AM » |
|
I enjoyed reading this thread, good work. I've made ART2003-LVP programmer and SPI flasher found in this post http://www.xboxhacker.net/index.php?topic=12306.msg84524#msg84524When I plug in USB flasher to PC, XP shows me "Unknown device" instead of "Memory Access" In device properties I see that vendor id and product id of "Unknown device" is all zeroes, so I suppose this is the problem. I think that I've successfully flashed my 2455 PIC. I'm using files found in first post. Flashing steps: detect device, erase, write, verify, and read. All steps without errors. One thing I've noticed when I compare PIC dump after flashing with original V3 boot loader PicXBoot_12.hex is that they differs. Is it normal ? Do I need to manually set vendor and product id's ? Edit: flashed pic with PicXBoot.hex dumped pic to file dump.hex compared dump.hex and PicXBoot.hex - differs flashed PIC with dump.hex dumped PIC to dump2.hex compared dump.hex and dump2.hex - both files identical !!!
|
|
|
|
« Last Edit: December 17, 2009, 06:34:57 AM by gepal »
|
Logged
|
no
|
|
|
|
Caesar555
|
 |
« Reply #354 on: December 17, 2009, 07:05:17 AM » |
|
This could be because of the settings in the configblock. http://products.foxdelta.com/programmer/art2003/ART2003-LVP.pdf 3d page I noticed that they changed when reading the pic. but don't know why... I had an wire error in my flasher causing it to delete the pic everytime I tried to use it.... But as it got not erased any more I had a vid and pid. but couldn't install the driver. But I used the V3b_plus2-pack from cory. After that I used the files linked in the first post, which u use and it worked like a charm. I'd say if you verify the pic after flashing and there is no error it is flashed correctly. Maybe you also have an error in your built flasher. maybe you should try corys v3b_plus2-files and have more luck with them. Since my flasher seems to work here are some pics. Wires: red=5V, black=ground, green=wire. The red wire on the bottom only connects pin 13 to the resistor (under pin 14). It got too hot while soldering, so the isolation came a bit off...  
|
|
|
|
|
Logged
|
|
|
|
|
gepal
|
 |
« Reply #355 on: December 17, 2009, 07:32:00 AM » |
|
This could be because of the settings in the configblock. http://products.foxdelta.com/programmer/art2003/ART2003-LVP.pdf 3d page I noticed that they changed when reading the pic. but don't know why... I had an wire error in my flasher causing it to delete the pic everytime I tried to use it.... But as it got not erased any more I had a vid and pid. but couldn't install the driver. But I used the V3b_plus2-pack from cory. After that I used the files linked in the first post, which u use and it worked like a charm. I'd say if you verify the pic after flashing and there is no error it is flashed correctly. Maybe you also have an error in your built flasher. maybe you should try corys v3b_plus2-files and have more luck with them. Since my flasher seems to work here are some pics. Wires: red=5V, black=ground, green=wire. The red wire on the bottom only connects pin 13 to the resistor (under pin 14). It got too hot while soldering, so the isolation came a bit off...   Maybe, but AFAIK PicXBoot.hex contains config block which is preloaded when you open file in winpic800, so I don't think this is the problem, because after write/dump/compare procedure configblock stays exactly the same.
|
|
|
|
|
Logged
|
no
|
|
|
|
|
|
gepal
|
 |
« Reply #357 on: December 17, 2009, 08:02:58 AM » |
|
Yes, I did it without success. Still, I got VID and PID all zeroes. Any suggestions ? edit: can somebody confirm that dump from PIC which was flashed with PicXBoot.hex and the PicXBoot.hex itself is the same ? or what differences I can expect (if any) ?
|
|
|
|
« Last Edit: December 17, 2009, 08:11:29 AM by gepal »
|
Logged
|
no
|
|
|
|
nicmon
|
 |
« Reply #358 on: December 17, 2009, 08:40:55 AM » |
|
Hi, is there any other variant of PIC18F2455 that we could use instead? e.g. PIC18F2550
|
|
|
|
|
Logged
|
|
|
|
|
Arakon
|
 |
« Reply #359 on: December 17, 2009, 09:09:15 AM » |
|
dumps are usually binary, hex files are basically text.. it's normal that they don't match up. use the verify function to be sure.
|
|
|
|
|
Logged
|
I do NOT give support by email, PM, ICQ or whatever. Anyone annoying me that way will have his balls removed. With a rusty butterknife. Slowly. And I'll enjoy doing it.
|
|
|
|