Misdreavus

36 minutes ago, needszeebs said:

Hey! I see it's been a few years, not sure if you ever ended up resolving this issue, but I found this thread because I am dealing with a similar, albeit less severe, issue. I also have a Pokemon Emerald cartridge that I purchased from Walmart myself back in like 2005 or something, so I know the cart is authentic. I picked it up last year to play through it again and started having troubles with it. Sometimes when booting the game, it does not leave the GBA title screen. Other times, it does but then the screen just goes white after. Sometimes the game appears to be working fine but then when I try to save, the game will freeze during saving (which ends up corrupting the save file). I got decently far into the game last summer and this happen to me where I lost the save data on the cartridge. I managed to get it working again and started another file which I was able to play through to completion. Last night, my Emerald cart started acting up again. I was trying to trade between my FireRed version and 3 pokemon in, the Emerald version froze while saving after the trade. It sat there for a while it wasn't saving, and then the screen just went completely black (sound was still on). After a few seconds, it brought me to the screen you posted up above saying "Save completed, game can no longer be played, returning to title screen." It did and somehow retained the save file on the Emerald, tho it also corrupted the save data on my FireRed game (also retained previous save). Thankfully, I had just backed up both games' saves on my GB Operator because when I rebooted the Emerald, I was stuck in this weird glitched trade room that I was unable to leave, so the game was essentially soft-locked. 

I thought I got it working again and played for another several hours saving multiple times and backing up the saves. However, when I booted it up this morning, those several hours were not present on the file and even the saves I backed up were corrupted and not working, so I was unable to restore the data. Something is seriously wrong with this Emerald cartridge and it's saving. I originally thought cleaning the pins would help, but it didn't really. My Emerald is still able to retain a save, but sometimes it just stops booting/ working properly and will start freezing, which eventually leads to corrupted save data (because it just won't save and then I have to turn the power off during the infinite save).

After doing some searching online, I was able to deduce that this may be a problem with the solder joints on the Flash memory chip. The PCB is quite thin and it seems the repetitive motion of inserting/ removing the cart from the GBA can cause stress and potentially fatigue the joints over time. I think it's much more likely this is what's happened to your Emerald, than the chip itself going. I've seen several videos of people fixing GB/A games that won't boot/ save properly by simply reflowing the solder joints on the mask ROM and RAM. I'll attach some links. 

Anyway, I thought I'd post this here in case anyone else is having this problem, or if you yourself have figured it out! What's interesting is that although I did play my Emerald growing up, I didn't play it nearly as much as my Ruby and FireRed version. Yet, my Emerald is the cartridge with problems and when I was researching, it seems many others have similar issues with specifically Pokemon Emerald and saving. I wonder if Emerald potentially had a factory defect that is interfering with the RAM. I haven't actually attempted to repair my Emerald yet because although I have an idea of what may be the problem, I do not have a soldering iron/ station, nor any experience soldering and the pins on the GBA carts are extremely small. I'll have to come back and update when I eventually get the confidence to attempt this, or find someone who can service GBA carts. Cheers.

 

Thanks for the detailed post. Yes, I actually did get the cart restored very shortly after I started this thread. I took it to a local retro gaming store that does repairs. They removed some of the excess solder from my attempt at a repair, and replaced the battery. As far as I know, that's all they did, and it works perfectly fine now. As to why this resolved what was seemingly an issue with the flash memory, I have no idea. But hopefully the problem won't rear its ugly head again.

14 hours ago, theSLAYER said:

I can't be certain, but I'm under the impression that you powered off completely when the text "*** saved the game" finished printing, but the textbox didn't close.

IIRC, if you were to do that on a retail Game Boy, there's a chance that the save may not have completely saved, so powering off immediately might throw you back to the previous save.

Always wait a bit, then close the app first, before powering off. This applies to all games on all consoles, not just VC Pokémon.
Impatient powering off the device could always lead to data loss. As a result, I always save twice, cause I'm paranoid.

The part that confuses me the most is the multiple saves I made throughout that time playing, and even SRing to those save points successfully, only to have the game revert to what was a few saves ago. I've really never experienced anything like this.

It would surprise me if I didn't let the text finish before powering off, considering how many times I've successfully saved a Pokémon game in my life, but I suppose it's a possibility. I was thinking the issue was more along the lines of a corrupt SD card or something like that.

But you do make good points. I was able to get back to where I was, and I'll be sure to follow what you said.

4 hours ago, theSLAYER said:

I'm curious: did you close the VC app then power off your device, or did you just power it off straight away?

I can't be certain, but I may have just powered it off straight away. Can that cause this sort of loss of data? I've played through multiple VC games on a different 3DS system and I have to imagine there were plenty of times I just powered it off after saving with no consequences.

Last night, I played VC Crystal for a few hours. I vividly remember saving and SRing using the VC menu multiple times along the way (i.e. in front of the red Gyarados). Eventually I saved for the final time and powered off my 3DS. I just picked the game back up, and to my shock, I was saved in the same spot where I began the night before. All my progress was lost, despite the multiple times I saved and SR'd (but never hard reset).

What could possibly be the cause of this? Could it be a faulty SD card, or a problem with the 3DS?

19 minutes ago, theSLAYER said:

Urgh, my page refreshed and my response is gone. Gonna retype it all.

Based on the content of your quote (of what I said), I don't think you saw the edit I made to my original reply.

Basically yeah, I think it's a result of cheat codes from lazy code makers.
They found a byte that works when written to a hex address, and rather than testing to find the intended bit, they just released it as is.
 

Injected wonder cards show up in the album viewer.
Here's an example, that I literally just did to show you:

As such, it cannot be used to verify whether an event was injected or not.
Yes, cheat codes may or may not have the card (depending on what cheat codes a hypothetical person uses), but wonder card injection definitely shows the card.

(tangent: no idea why you're calling it a voucher)

You are correct, I missed the edit. But thanks, that's what I needed to know.

Also, "voucher" is just the term that the person I was speaking with was using. I wasn't sure what else to call it.

5 minutes ago, theSLAYER said:

Injecting wonder cards typically don't raise event flags. Only redeeming the wonder card in-game raises the correct event flags, as this is something the game does normally anyway.

Some lazy cheat codes raise a bunch of event flags; they raise other event flags, not just the intended one.
Non-lazy cheat codes don't have this issue.

So yeah, injecting the wonder card or using cheat codes from "non-lazy code makers" are fine.

Thanks for the reply. Yeah, I've heard that some of those cheat codes can cause things like all gold Frontier symbols, etc.

What I was particularly wondering about was whether this voucher would be visible in the case of an injected Old Sea Map, or whether you only get that if it was genuinely obtained from the event with the wireless adapter. The person I'm speaking with believes it's the latter, and thus its presence or lack thereof could be used to indicate the origin of the WC (2005 event vs. injected).

I'm speaking with someone who has a source for a copy of Pokémon Emerald with a supposedly genuine (obtained from one of the 2005 events) Old Sea Map. They said that as far as they know, if the WonderCard is injected using a tool such as the one offered here, the "voucher" that appears when you connect the wireless adapter and go to the Mystery Gift screen will not be there. Is this the case, or will the voucher be there even in the case of the injection? I would think (and was told on Discord) that it would be present, but seeing as how I'm getting conflicting answers, I figured I'd start a discussion so that someone who's 100% certain can help me.

Thanks in advance.

15 hours ago, Hide said:

You should check the threads in the sector that holds the LE26FV10N1TS chip, that is the flash memory. Put very roughly (but it is not exactly as I say), the threads in the back are inbound communications and the threads in the front are the lines from the cartridge to the GBA system (Do you notice how the threads in the front go from the cartridge to GBA connection pins?). After you check the "inbound" links, check the "outbound" links.

AFAIK, Pokémon games employ these chips to save data:

Macronix: MX29L010TC-15A1, 1m (128kb) flash memory chip.
Sanyo: LE26FV10N1TS, 1m (128kb) flash memory chip.

I don’t know if you can swap one chip-model with the other (it shouldn’t be an issue unless they work differently); however, it is possible to remove an original chip from one cartridge and put it into another.

If it helps, here's a website about "home-brew" projects with Gameboy electronics (including GBA). This link features an -original- schematic of a GBA cartridge; it doesn't have the same layout as Pokémon Emerald, but it could help you to map the relevant connections of your flash chip.

This person took an original Macronix chip from a Pokémon game to make "his own" cartridge.

Everything checks out fine on the back of the board; the multimeter appears to confirm that everything that is supposed to be connected is indeed connected.  What exactly does this mean considering my game doesn't register at all after the attempted repair years ago (the GBA acts as if nothing is in it)?

Are you saying to do the same thing on the front?  I don't believe I'll be able to trace every connection on the front since many of them are covered by the printed text or other things such as the SRAM or ROM.

5 hours ago, Hide said:

By the way, something "funny" about your cartridge is that its chips do not have ID data. Most chips have a printed number and/or the brand (this is how you start checking what they do). Your cart should look like these (pictures attached):

Actually, if what you're referring to is the printed text on the SRAM and ROM, mine does have that; it was just not visible with the lighting on the pic in the first post.  This is what my cart looks like after the attempted repairs were done on it.  I just removed the battery prongs for the time being.

You may notice a small bit of excess solder on the pins of the ROM; this has been brought to my attention before, and I removed it as best I could.  There may still be a small bit left (see pins 3-6, top to bottom, on left side), but I was unable to get that off, and I'm going to see if someone can help me with it.

EDIT: As for the continuity tests, are there any particular origin/destination points I should check?  There are obviously a lot of them.

EDIT2: To go along with my question in the first edit, here is some work I did with the photo of the back to make it easier to trace the points.

18 hours ago, Hide said:

Sometimes the boards corrode with time, in other occasions falls break the boards and create microscopic gaps; these things disrupt communications within the board. If this was the problem, all you have to do is check for continuity.

A blueprint is useful for this matter, but you could trace the threads of the chip without one. Just check the board from the back, not the front.

This is a common problem in circuit boards; regarding micro-soldering, the common practice is not transplanting the chip, but bridging the faulty terminals to their destinations using wires (at least for the first attempt to repair the board).

You can check this generic video on continuity.

Thanks for your reply. If I'm understanding you correctly, are you saying to trace the fine lines on the back and see if a point on the left and its corresponding point on the right register on a multimeter?

7 minutes ago, theSLAYER said:

Not that I can refer you to a specific video, but I've seen videos of repairmen dealing with circuit boards, and what they tend to do, is to place the prongs of the multimeter around the chip, to see if any voltage passes (to know if the chip is still functional), or to place them at the beginning of the chip and beginning of the where the current should originates from, to see if the voltage travels right up to the chip (no breakages in between circuits etc).

However, without a specific blueprint for the voltages that should be across the board, we can't just do that haphazardly.
Or course, you could do a simple compare and contrast with a working board, however I'm not certain you want to put the working board at risk
(from you know, not knowing what you're doing )

From what I understood, it's not the save that you value, but the cart itself.
Provided it's only the chip that's dead, and not the circuits on the board (or some other factor),
someone with the appropriate tools can try to transplant a working chip onto it,
but it's hard to save it'll work, and the cost and time taken will probably be high.

I do have other non-Pokémon games that I no longer play and would be willing to, more or less ,"ruin" in the name of testing the board, but I opened them up and they are structurally different inside (no battery, etc.).

While I certainly did value the save file I had on this cart at the time, I know it's long gone, so at this point, the cart is all I have left, and it would be nice to have another Emerald.  Your point about the cost and time of a potential repair is entirely fair though.

Also, after doing some research, I found a potentially relevant thread on Reddit, and if the numbers in the first reply are accurate, it makes no sense that the flash memory would fail after 4 years.  The second reply also mentions that I would need to flash the new memory first before swapping it in.

5 hours ago, theSLAYER said:

And here I thought it could all be explained by the presumption that you had a fake cart.
The explanation honestly fits, if it were a fake cart.

Not that I believe that GameStop sells fake carts, but I don't know what's their policy, especially if it's related to second hands goods.
Or maybe they had some kind of mix up. Who knows at this point.

I used to own a fake cart that battery ran dry (not that I knew it was fake until after the fact),
and this was exactly what happened. Once the battery ran dry, I could no longer save the game.

I don't know how it would be possible. I also wonder what you should do.
Replace the entire chip? Check the traces on the circuit board? Run a multimeter before and after the chip to check if it's dead?

It's possible the memory chip itself is dead, or unable to draw power from your battery.
But here's the thing: as far as I know, on a real cart, the battery isn't for running the memory. It only runs the RTC.

Maybe someone who knows what they're doing can come along and advise you, but my impression would be you should just buy a new cart.

 

Thanks for your reply.  I actually did get a new cart; however, this cart has sentimental value to me despite the save file being gone, so I'd like to get it up and running if possible.  The new cart I got eventually had its battery run dry, and it still saved normally afterwards, so I doubt the battery was responsible.

Here is a picture I took when I first opened the cart (prior to the attempted replacement of the battery).  I believe the flash memory is the black rectangle located directly above the word "FLASH," right?  The traces around it are extremely small, so I'm not even sure if someone who is proficient at microsoldering could successfully do a "transplant."

Do you mind elaborating on exactly what you meant regarding the multimeter?  That definitely sounds like something I could do.

I have an Emerald cart that I got from GameStop on the release date in 2005, so I'm as sure as I can be of its legitimacy.  Back in 2009, I started getting a series of error messages when saving the game. Ultimately, my save file was deleted on its own.  Attached are photos I took of the bizarre messages I got.  I've shared these on other Pokémon sites and no one seems to have seen them before.  Excuse the poor quality / blurriness; these were taken with a digital camera over a decade ago, and some were captured from video that I took.

Not knowing at the time that the battery was not the home of the save file, I tried to replace the battery, only to get excess solder on my game and damage it.  (I'm looking at repairing that, and it's outside the scope of this post anyway.)  If I get my game running again, am I likely to have the same save corruption happen, or was what happened probably a unique case involving something going on with my particular save file?  I had never used any external devices with it, though, so that would be a surprise to me.

If it is indeed something that will probably occur again, is it at all possible to replace flash memory in the GBA cart?

7 hours ago, theSLAYER said:

The gamefaqs one. As for the one you posted, I didn't look at it, but the fact that it listed getting some events from Digiex suggest the user just inserted everything in. I'll probably need to take a look at it. (Don't get me wrong; I love Digiex, I just feel as though if the user grabbed and combined events into a save, he could do the same with the aurora ticket)

 

That's a good point about the Gamefaqs one actually. While gameshark has been around (hence the backup), I don't think event injection was around, in terms of event wondercards. People probably just used the gameshark code to enable the event.

If you get around to taking a look at it, I'd appreciate it if you'd let me know what you think.

2 minutes ago, theSLAYER said:

I know you earlier said that you can't load the .sps file (the shiny celebi one), but I just want you to know that VBA supports it.
It's under "Import Gameshark Snapshot"

I have doubts whether this was legit, or obtained with Morpheo's injector.

When you say you have doubts about it being legit, which save file are you referring to - the one on GameFAQs or the one I linked to in the 3rd post?  The one on GameFAQs is from 2004, and being so close to the event dates (and likely before we had discovered how to to inject Wonder Cards) is a decent reason to believe it's legit IMO.

21 minutes ago, ICanSnake said:

It is possible to convert it without emulation, there's a site that can do it. I have the .sav of it myself, but unfortunately it's currently on my broken laptop and it's taking a long time to get fixed.

I found a site that can do it for DS games, but not GBA; that feature hasn't been added yet.  Do you know what type of GBA Gameshark I would need to restore the save?  I see two different types (1, 2).

22 minutes ago, ICanSnake said:

I believe this GameFAQs save upload by ShinyCelebi should be what you're looking for.

Thanks.  Unfortunately I need it in the form of a .SAV file, and I don't work with emulation (as far as I know, using an emulator is the only way to convert .SPS into .SAV).  Is there anything else I can do?

1 hour ago, theSLAYER said:

I don't think those save exists. We've had a hard time trying getting even a legitimate one for Gen 3 event research purposes.

You could try checking here, to see if anyone uploaded their save.

Thanks.  I did run across this.  In the description, the author claims to have gotten the AuroraTicket from the PCNY.  Any thoughts on its legitimacy?

I'm looking for a save file for FR/LG/E with an Aurora Ticket (with the corresponding Deoxys not yet caught) that was obtained from one of the official events back in 2004/2005.  I know the NA version of the ticket has leaked and anyone can get it (and that one such will be indistinguishable from an "authentic" one), but I'm looking for one that was obtained directly from the actual event, if possible.

Thanks.

2 hours ago, Kaphotics said:

you likely edited the backup save file, not the active one

(the sav stores two files in the event one is corrupt).

I'm familiar with the two save banks; however, I wasn't aware that I had a choice in which one I edited.  How do I change between them in PkHeX?

Awhile back, I edited my game's playtime in PkHeX to force the checksum to be a value that yields a shiny Jirachi from the Colosseum Bonus Disc.  I'm trying to do that again; however, I'm not sure what exactly I did the first time, and I'm having a bit of difficulty.  Here's what I'm doing:

1. I open the .sav in HxD Hex Editor and locate the checksum.

2. I open the .sav in PkHeX and edit the playtime.

3. I go to File -> Export SAV -> Main and save my file.

4. I locate the file and manually add .sav to the file name.

5. I open the new file in HxD, browse to the same location I found the checksum at in step 1, and view it.

The checksum doesn't change though.  Why would this be?  Am I doing something wrong?  I know if I do it properly it should change since I've done this before (with an earlier version of PkHeX).

9 hours ago, Sabresite said:

I would take an area of all 00's, and change one value to 01 and see how PKHex recalculates checksum.  Then change it back to 0, and change the next offset to 1 and see how that changes the checksum.  Then change the value back to 0, and change another 00 to 0x2 or 0xF, and see how that changes the checksum when PKHex recalculates it.  You will find it is arithmetic and you will be able to easily find a pattern of changes to get the checksum you want.

Good luck!

I took the very first value in HxD (block 00 in offset 0) and changed it to 01, but after opening in PKHeX and saving, my checksum did not change.  Why would this be?

And thanks!

EDIT: I did, however, get the checksum to 00 F5 by changing my TID to 44405 (and keeping my SID the same at 40171).  How can I tell if this is a valid TID/SID combo given the vast amount of seeds that could be obtained with a live battery?  The only way I can think of is to search each minute of each hour of each calendar day, which is impossible.

EDIT2: I got the checksum to 00 F5 by changing my game time to 5 hr 26 min 30 sec.

EDIT3: I GOT IT!  Thanks everyone for all your help!

1 hour ago, ajxpk said:

Well... first I think you should get an understanding about what an checksum is.
A checksum is the sum from calculated data and it is for checking purpose. It's in fact to there to protect the data from being manipulated and to prevent errors. Whenever data is saved and stored the checksum gets recalculated. Now if you change data from an block the old checksum that was based on the state prior the editing becomes invalid. The same goes vice versa, meaning if you just edit the the checksum the checksum becomes invalid. Once you understand this it should also become logical for you that it simply isn't possible to change the checksum without changing anything inside the block. This has nothing to do with PKHeX. PKHex just uses the same mechanic that the game uses for getting the correct results... which just makes sense, right? So whatever you do to your checksum, PKHeX is correcting it. (Thankfully!)

So... what you are going to do is changing data from the block and then you use PKHeX to get your new checksum.
You could of course do all of it using PKHeX. I personally recommend to use the Playtime, just stick to the minutes and hours...
Otherwise you can mess with the IDs, but keep in mind that in RS not every TID/SID combination might be legal. So better use FR/LG or Emerald in case you want to reuse the save file for something else... (It doesn't affects the Jirachi, I just say this because I think TID/SID combinations aren't considered in legality checking yet from what I know... so I don't want to encourage to mess with it.)

Thanks for the explanation!  So how do I know what to change the playtime to?  Do I have to just tweak it, check the checksum in HxD, and repeat until I get my desired checksum?

Also:

1) I don't know about all, but I would think there would be a lofty combination of TID/SID in R/S if there is a live battery at the point of generation, as that gives R/S more seeds to work with.

2) The Jirachi bonus disc is only compatible with R/S.

58 minutes ago, Sabresite said:

It is done automagically when you save.

Okay.  So you're saying to change that block first, rather than directly changing my checksum to 00F5?  How do I know what to change that block to (either manually in a hex editor, or by changing something in PKHeX)?

3 minutes ago, Sabresite said:

You have to change data in that block so when the checksum is regenerated by PKHex, it is 0xF500.

Pardon my ignorance, but would you mind explaining how PKHeX regenerates the checksum?  I don't see an option for it.