Jump to content
OGXbox.com

Yoinx

Members
  • Posts

    32
  • Joined

  • Last visited

Posts posted by Yoinx

  1. 1 hour ago, Prehistoricman said:

    This is Cromwell but it should be the same file

    https://github.com/XboxDev/cromwell/blob/master/drivers/pci/pci.c#L283

    Of course the memory size detection routine is in the PCI driver... 😕

    You can change line 319 to be

    if (true) {

    or you could change line 314 to be

    xbox_ram = 64;

    Technically, there's not much reason to just change either of those. If you're going to force 64MB anyway... Just clear the entire function and just have the line:

    PciWriteDword(BUS_0, DEV_0, FUNC_0, 0x84, 0x3FFFFFF); // 64 MB

    The whole function is just testing memory locations to see if there's more than 64mb to determine which hex value to write. No reason to waste the time letting it test it you're going to always write the same value.

  2. 1 hour ago, HDShadow said:

    Flashing the BIOS using a HeXen disc should be a simple matter of launching the HeXen disc (DVD-R/ImgBurn) and going to the BIOS flashing tools menu. 

    He already deleted the content of his post. So no idea what he's using.. and no it's a 3 year old necro'd thread without even the reason for the Necro.

    But if I'm not mistaken a lot of the Aladdin modchips are non-flashable knock-offs. So he may not be able to flash it anyway.

    • Like 1
  3. 10 hours ago, russlb said:

    Yeah sorry the pictures wouldn't upload; they were too big and I had to shrink them down.  The three CPU caps I replaced with 3300uf 10v which should be fine.

     

    Cap Replace 1.png

    Definitely needs a better picture. Lighting without the flash would be much better due to the glare it causes.

    But, wth is going on with those faces over by that screw hole on the right? Did you pool solder on top of them in the left bend?

  4. 7 hours ago, Craiguk said:

    Brilliant it kind of boots but goes no further than the photo 🙄

    I followed this and twisted the same wires, even tried disconnecting pin 8 still same result. 

    Do I need to ground pin 7 wire to something it mentions something about d0 but I'm just not sure... See below it mentions it.... 

    1- HDD LED +ve
    2- LAN LED +ve
    3- Common +ve
    4- Modchip power in
    5- Ground
    6- Flash Protect
    7- Chip enable/D0
    8- Bank 1 select

    To power the chip you need to connect pins 3-4

    To enable the modchip as the default boot bios you need to ground pin 7 (D0) and generally pin 8 to select Bank 1

    All the LEDs are powered by pins 3 and 5 and use an inline 220k resistor.

    eg. For my setup - boot off the chip, flash protect, and using bank 1
    I connect 3-4 and 5-6-7-8
    For Bank 2 you just disconnect pin 8

    PXL_20220811_151141900.jpg

    The diagram in the post above you has the pinout exactly opposite from what you posted. Pin1 is bank1, not HDD+.

  5. 5 hours ago, Prehistoricman said:

    You can inject voltage, aiming for a decent power consumption like 5 watts, limited to 5 volts of course, and feel the board/chips with your hands. If you can't feel anything, increase current and keep going. 

    That meter he has does temperature as well. I don't know what it's response rate to changes is, but he could take the temperature probe and touch it to various components. Time consuming, but doable.

    • Thanks 1
  6. I don't mean to be rude, or mean... But I have no idea what I just read. I'm guessing English isn't your primary language... But you may want to try to proofread/translate this a bit better.

     

    It seems like the Xbox doesn't work for your friend, but does for you. Are you sure that the power at their house is good?

    • Haha 1
  7. If you have emery cloth it might be worth trying to clean the top of contacts in the port (very lightly, you just want to expose metal not sand them down), and make sure that they're at least a little springy... Before bothering with swapping the port.

     

    Thinking about it... An emery board nail file might fit in an rj45 socket pretty well to do this.

    • Thanks 1
  8. This may be a bit of a hassle... But couldn't you just use mouser in the UK? Then build a cart based on the console5 capacitor list?

    I don't know the specs off hand for what you'll be looking for, but this search for instance *should* be what you can choose from for C1B1. They should all be fairly standard  radial caps, they provide the capacitance and voltages on that list. However, the lead spacing may be an issue as I don't know what it is for all of them. Technically you could just measure them all on your existing board as you build your cart. 

     

    You'd also be able to pick your preferred brands this way. I'd recommend sticking with something like Panasonic. However, I have no idea if you'd save much money going this route.

     

    Edit:

    Found this more specific list on badcaps.net which should give you the exact specs of what you need.

    • Like 1
  9. On 5/18/2022 at 8:24 AM, Onaga said:

    255.255.255.0 ?

    Sorry, slow reply... I got tied up with stuff.

    That's a subnet mask, not a DNS server. From everything else I've seen posted after this comment, the only thing that I can say that I've seen act odd is depending on which dashboard was set to default, my xbox would act weird for IP addressing. Some dashes would get their addresses right, some wouldn't. Then if you launch another dash from that one, it's network would be all screwy and not correct itself. Maybe try re-installing a dash that you *know* should work right and reconfigure it?

    I know on my network, between the OGXbox and the Xbox 360... the have an oddity because I run my network on the 192.168.0.0/16 or 255.255.0.0 subnet (depending on your preference). However, sometimes the units will get an IP address but use the typical 255.255.255.0 subnet, which obviously causes issues. 

  10. Ended up installing it like I thought it should go in this 1.6, booted up fine with the evox bios. So everything's good. It's constantly on, (blue light) as long as there's standby power. But I think it was like that in the last one too. *shrug*

     

    Thanks for the help!

  11. Sorry I didn't reply to this very quickly. I ended up bricking the 1.6 I was working on. I tsop'd the 1.2 that I had, but now I have another 1.6 that someone at work gave me so I'm back to this.

    I ended up looking at my wire install connector again... It looks like my connection may have had one of the wires ripped out of it as I have a pin in the connector for pin 15 (left of the gray wire in the below image where get if facing the camera) with no wire. However looking at the installation image that I can find (that's never in English) doesn't show a wire there. I also don't have a ring terminal, neither does the install image. I can add one, but where? Wire 2? I would assume that is the pin header could ground to pin 2 the wire install would as well.

    Here are pictures of the chip and the wire install. I didn't pull the 5v wire that fits to the red connector, or the 3 pin one for d0, lan/HDD LEDs from the old box yet. And the selector board is connected but not pictured.

    My original attempt on my bricked box was to solder the wires directly over to lay as if they matched the header locations/orientation. But any number of other things could have been going on with that unit. I'm assuming that these should just straight over to the lpc in the right order, unless I'm missing something. I'll have to go the lpc rebuild on this new box, but that's not a huge deal.

    I'm thinking that in your note under the lpc pinout you may be confusing the x2 and 2.6? As you mention d0 in this writing harness, but that's on the white connector on this chip, not the lpc wiring harness.

    20220518_215950.jpg

    20220518_220032.jpg

    20220518_220146.jpg

  12. I don't know if it will make much difference, but if you painted the shield, you should definitely clean the paint off of the standoffs where the screws go through the motherboard. The metal contacts around the holes serve as a common connection on circuit boards like this. Things *may* work fine if one has a bad connection, but that's never guaranteed. If the screw threads were in contact with metal the entire way, it wouldn't matter much. However, I believe that thread into plastic so the pressure from the screw head holding the board pads against the shield are what's giving you a connection.

     

    I'd start off just sanding the paint off the flat tops.

     

    • Like 1
  13. The x2.6 had an accessory programmer that could be used from PC, but it requires a parallel port.

    I'm actually curious about different methods of reflashing the x2.6 chip myself. I saw piprom which looked promising. But I'm not sure if it is intended to work on a modchip vice the onboard prom in the xbox

  14. 43 minutes ago, Rocky5 said:

    script builds myprograms.xml out of the gamelist.xml for that system and loads the menu. 

    Yeah, I was trying to make sense of how that was happening in the scripts. I hate to question how someone does things, especially when I admittedly didn't know much about xbmc... And I'm sure it would break things when the themes changed... but wouldn't it be faster to just prebuild each system's entire myprograms.xml when scanning the roms then substitute the entire prebuilt file in depending on what was chosen rather than combining/rebuilding it each time? IE, store the whole thing next to the gamelist.xml and just soft link it or something when called.

    I've looked at a few of the scripts already trying to figure out how it was building them and there is a lot of string manipulation (and in XML) which unfortunately is not super efficient in Python... and that's probably just exacerbated on the Xbox. Even combining the files by looping through them and reading each line for thousands of lines should be expected to take a few seconds at least on Python, especially on the Xbox hardware.

     

    amazing work on this and hopefully the questions don't make me seem unappreciative.

     

     

  15. 4 hours ago, Donnie-Burger said:

     

    I don't have any other drives to test with, but I had no graphical glitching in xbpartitioner that I could tell. I was ftp-ing in xbmc at around 4-5MB/s which seems fairly typical. And like I said, everything plays perfectly fine including Xbox games from the HDD. The only thing that's been slow is the menus. Unfortunately, I also don't have any ide interfaces for my computers anymore... And not to interested in purchasing a usb-ide adapter (this Xbox is just for my daughter to play on, and she has a 360 with pretty much the same stuff on it).

    I can live with things as they are, I just wanted to make sure I wasn't missing something dumb/obvious that would cause this as far as settings went.

    • Like 1
  16. 1 hour ago, Donnie-Burger said:

    put up:

    hard drive

    addapter

    cable

     

    models

    The harddrive is a Maxto 4A250J0, which i'm just realizing is an old 5400rpm drive... so I can't expect  much out of it. It's not locked. There's no adapter, and it's just the OEM ribbon cable. (Partition Layout attached)

     

    I ended up deleting the XBMC-Emustation folder and copying it back over via FTP and just putting 1 NES rom into the roms folder on 'E' and letting it scan. The menu does come up a lot faster like that, around 3-4 seconds. The gameslist.xml was only 2kb though, so that's not so surprising. It might just be an issue with a slow hard-drive. Though, I thought that the XBOX only gets like 10MB/s through its IDE controller anyway so I wouldn't have expected it to be *that* slow off of a slow drive. 

    I went ahead and rescanned them all back in and the gamelist.xml is around 47kb for that system (NES) and it's back to ~25 seconds for the game list menu to open in the basic view. I know that the list has to be processed and all that, but I can't imagine that the hard-drive speed would have that huge of an effect for a 47kb list to be read off of it. Is it trying to validate that everything exists as it processes the list or something (this would make sense for hdd seek speeds and such to slow it down)? Is there a way to disable that so that it just shows the list without validation? Obviously, if something doesn't exist anymore since the games were scanned this could cause a crash... But I'd rather that being a problem and the list opening quickly.

    I have fast game parsing on per your previous suggestion.

    partition_layout.jpg

    • Like 1
  17. 15 hours ago, Rocky5 said:

    Run Emustation from E and have the emulators, media and roms folder on a larger partition. 
     

    don’t run it from F and defo not G as the seek times are horrendous the more you fill those partitions. 

    That's probably where I'm going wrong. I put the roms on f and put the media on g. Emustation is on E

     

    I'm only using a 250gb hdd, with the f/g partitions around 115gb each. If they shouldn't go on e/f/g where *should* I put them for better results?

    Or should I have made an extra partition with like 10-20gb to store them?

    • Like 1
  18. 3 hours ago, Bowlsnapper said:

    Do you have a shitload (hundreds) of OG Xbox games on the HDD? That causes loading times exactly of that sort of length. Caused by the Dashboard scanning for games every single time. The 2TB Origins install behaves that way. Ever single time the dash boots. Total dealbreaker for me, man.

    I don't, maybe like 30 Xbox games. But it's behaved like this even when I only had the dashboard and game gear roms loaded. So, I don't think it's an issue of hdd usage.

     

Board Life Status


Board startup date: April 23, 2017 12:45:48
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.