SYSTEM.CNF loading issue on PS2 consoles

[roberthawdon]

When trying to run the European Release of Kurushi (Intelligent Qube in NTSC regions) I get the following error:

Reading SYSTEM.CNF
Missing TCB
Missing EVENT
Missing STACK
Missing BOOT
TCB = 00000004
EVENT = 00000010
STACK = 801FFF00
BOOT = cdrom:PSX.EXE;1
Configuring kernel
Loading executable
Loading failed
Swap CD now

This is the contents of System.cnf:

BOOT = cdrom:\SCES_008.66;1
TCB = 4
EVENT = 16
STACK = 0X801FFF00

[alex-free]

The problem to PS2 System.cnf at the last sector of the disc is solved:

1. The problem is not due to the mastering of the disc. PS2 is capable of booting exe or reading system.cnf at the last sectors of the disc.

2. This problem happens to CDR only. The PS2 drive might spin up and the booting is hang at the PS1 logo screen.

3. The problem can be solved by burning the disc image to a 650MB / 74 mins good quality disc.

4. The problem is probably a PS2 drive design flaw. The same disc can run on a regular PS1 but not PS2. Strangly, the cheap CDR works fine on the PS2 if the exe and system.cnf is not at the end of the disc.

Old Sony docs actually said that anything over 72 minutes is not officially supported for burned CD-R media but that was involving the original psx. The original psx has a seek table optimized for 72 minute media I belive, and it was even recommended that the last 3 minutes contains a dummy track so that the drive isn't seeking in that area. Maybe it is more problematic on the PS2, because i.e. this is how tocperfect patching works (puts the tonyhax executable at the end of the disc).

[NotALuckyPey]

btw, sry for eng Now I have a second PS2, SCPH-30001R with modbo 760, it has a same problem with cnf and exe at end of disc: 1)With tonyhax - system.cnf not readable, exe not found 2)Without tonyhax - freezes on PS1 logo But If I try open system.cnf on disc via text editor in ULaunchELF - it actually opens and shows me all text in the configuration file. However with modchip PS2 thinks it's PS2 CD game (that's what ULE show me). That's how "modern" modchip works. I don't think this is a CD-R problem, maybe it's really a mechacon bug in PS1 mode?

[alex-free]

btw, sry for eng

Now I have a second PS2, SCPH-30001R with modbo 760, it has a same problem with cnf and exe at end of disc:

1)With tonyhax - system.cnf not readable, exe not found

2)Without tonyhax - freezes on PS1 logo

But If I try open system.cnf on disc via text editor in ULaunchELF - it actually opens and shows me all text in the configuration file. However with modchip PS2 thinks it's PS2 CD game (that's what ULE show me). That's how "modern" modchip works.

I don't think this is a CD-R problem, maybe it's really a mechacon bug in PS1 mode?

This can be both a CDR problem and a bug in PS1 mode.

I think it should be verified with 74min media that all these games boot on the older PS2 consoles.

Just think about what has all been said here:

'Faster speed/changing drive speed sometimes makes it work'

'Real PSX CDROM works'

'Has to be a game with SYSTEM.CBF at the end of the disc as the last file'

I think it really is as simple as this:

My theory is the seek table thing used in PS1 mode is less tolerable to this specific seek error (last file on disc) then a real PS1.

80min media is notoriously worse for both PS1/PS2, and the PS2 in PS1 mode seems like it can't recover gracefully after a subtle seek error when it overshoots the last file on the disc due to tighter spiral windings in 80min media. So when the drive overshoots that file due to denser packing of pits in 80min media not lining up with the expected seek distance for 72 or whatever minute CD-ROMs it doesn't know what to do because there is literally no data where it overshot to.

Anyways this should just be proven and if so this issue should be closed. I'd buy a bunch of Verbatim DataLifePlus 74min media from the 90s, burn at slowest speed, and test each title referenced.

Just my 2cents

[alex-free]

The problem to PS2 System.cnf at the last sector of the disc is solved:

1. The problem is not due to the mastering of the disc. PS2 is capable of booting exe or reading system.cnf at the last sectors of the disc.

2. This problem happens to CDR only. The PS2 drive might spin up and the booting is hang at the PS1 logo screen.

3. The problem can be solved by burning the disc image to a 650MB / 74 mins good quality disc.

4. The problem is probably a PS2 drive design flaw. The same disc can run on a regular PS1 but not PS2. Strangly, the cheap CDR works fine on the PS2 if the exe and system.cnf is not at the end of the disc.

Again, all of this makes perfect sense to me :)

[alex-free]

A solution for 80min media might be to create a patcher that puts a dummy file as the last file after SYSTEM.CNF. That way when it overshoots it's still in real data and can recover back to the actual lba of SYSTEM.CNF

[alex-free]

ps2scp.exe-test3.zip

I hacked together a patcher out of Mottzilla's demo swap patcher program that I have the source to that puts a DUMMY file that is 6 minutes long at the end of the disc AFTER SYSTEM.CNF. I'm out of CD-Rs for a minute but this might fix it if the theory is correct. If it does or doesn't let me know.

Just drag the track 01.bin file of the game onto this exe and it will patch the file with output in a cmd.exe window. Then burn the cue to a high quality 80min CD-R as slow as possible and test it in a PS2 supported by tonyhax or tonyhax international.

[roberthawdon]

Could you upload the source for this somewhere, and would it compile on Linux?

[alex-free]

Could you upload the source for this somewhere, and would it compile on Linux?

ps2scp-test3-lin64.zip linux x86_64 build above Can't release source without MottZilla's permission as this is reuing alot of his code just to test the idea out. Also it really needs to be cleaned up source wise.

If this doesn't work, I am thinking a 9 minute DUMMY file might.

Anyone that uses this, make sure you can see the DUMMY file when the CD is mounted by your computer. Checking with CDMage also doesn't hurt.

[alex-free]

I have a report this works with RE2.

Must use a multi-bin file if the game has audio tracks (like the audio tracks are separate bin files and track 01 is just a data file).

Must use a burner program that regenerates EDC (if you don't know what this is you have one. Imgburn, cdrdao with generic-mmc driver, etc all are good for this).

[roberthawdon]

Amazing work! Works fine on an 800mb CD 👍🏻

[Ledoeye]

Wish i could test this right now, but my cd burner fried last week need to buy a new one, do you guys have a good suggestion? I’m glad it’s working amazing job! Thank you @alex-free you’re a legend 👍

[alex-free]

Amazing work! Works fine on an 800mb CD 👍🏻

Do you mean 700MB? Thanks for verifying I'm gonna clean up the source and do a proper release. Please try all the games you can that have this issue.

[roberthawdon]

Amazing work! Works fine on an 800mb CD 👍🏻

Do you mean 700MB? Thanks for verifying I'm gonna clean up the source and do a proper release. Please try all the games you can that have this issue.

Haha, yes, sorry, had a brain fart and wrote something between 80 minutes and 700mb!

[roberthawdon]

Also, this was the only game I came across that had this issue.

[alex-free]

Also, this was the only game I came across that had this issue.

I heard there are a few Resident Evil 2 I know was the first test game. I want to compile a list from this thread I'll have to go through it again.

Maybe it might be easy to detect when SYSTEM.CNF is the last file via a newer build of the program as well.

[roberthawdon]

One thing I might try is applying this to games that also booted in the first place. I only had issues with one disc, which was a pre-release version of "Gex: Deep Cover Gecko" which crashed when loading some parts of the game. It might just be the buggy nature of a beta version, but it now has me thinking that maybe the PS2 overshot the end of the disc trying to load the game files and wasn't able to recover.

[alex-free]

One thing I might try is applying this to games that also booted in the first place. I only had issues with one disc, which was a pre-release version of "Gex: Deep Cover Gecko" which crashed when loading some parts of the game. It might just be the buggy nature of a beta version, but it now has me thinking that maybe the PS2 overshot the end of the disc trying to load the game files and wasn't able to recover.

Great idea! Let me know if you find anything. This patcher will work for any game just fine in theory as long as you follow the instructions

[alex-free]

New version of loader didn't help for me. Still games like resident evil fail to read cnf. Final fantasy ix finds the cnf with game exe name, but still fails to read.

I am thinking the PS-EXE of FF IX is the last file on disc (instead of SYSTEM.CNF). Patcher should make it work on 80min media as well :)

[the7thchild]

Is the patcher working? Anyone can confirm the 74mins CDR discovery?

從我的iPhone傳送

Alex @.***>於2023年6月16日 09:42寫道：



New version of loader didn't help for me. Still games like resident evil fail to read cnf. Final fantasy ix finds the cnf with game exe name, but still fails to read.

I am thinking the PS-EXE of FF IX is the last file on disc (instead of SYSTEM.CNF). Patcher should make it work on 80min media as well :)

— Reply to this email directly, view it on GitHubhttps://github.com/socram8888/tonyhax/issues/24#issuecomment-1593950967, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A45OHGMLPKGX4KTJ2WTIAY3XLO2YRANCNFSM4Z2EMC2Q. You are receiving this because you commented.Message ID: @.***>

[alex-free]

I cleaned up the source, renamed it to 'PS2 6 Minute Patcher', and made binaries for Linux x86, Linux x86_64, Windows x86, and Windows x86_64. Same usage, but it puts a dummy file named 'PS26MP' as the last file on disc (to ensure it doesn't conflict with an already existing DUMMY file, if that ever would be an issue).

ps26mp-1.0.1-linux_x86_64_static.zip ps26mp-1.0.1-linux_x86_static.zip ps26mp-1.0.1-windows_x86.zip ps26mp-1.0.1-windows_x86_64.zip

[alex-free]

I went out and bought the worlds worst CD-Rs going to test this out myself with the all the games mentioned on this thread

[alex-free]

Bug confirmed with SCPH-10000 Japanese launch PS2

[alex-free]

Confirmed this totally works with Kurushi!

[the7thchild]

Hi Alex,

May be the game size is small, Kurushi worked for me even without the patch on 80mins CDR, although the system.cnf is located at the last LBA.

從我的iPhone傳送

Alex @.***>於2023年6月19日 00:24寫道：



Confirmed this totally works with the Kurushi!

— Reply to this email directly, view it on GitHubhttps://github.com/socram8888/tonyhax/issues/24#issuecomment-1596196939, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A45OHGJPB6XOFAXYXM4GQCDXL4TTFANCNFSM4Z2EMC2Q. You are receiving this because you commented.Message ID: @.***>

[roberthawdon]

Kurushi was the reason I opened this issue. It might be a case that your system handles 80min CDRs better than mine. Until patching the disc image, I had never been able to boot Kurushi on my PS2.

Hi Alex, May be the game size is small, Kurushi worked for me even without the patch on 80mins CDR, although the system.cnf is located at the last LBA. 從我的iPhone傳送 Alex @.>於2023年6月19日 00:24寫道：  Confirmed this totally works with the Kurushi! — Reply to this email directly, view it on GitHub<#24 (comment)>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A45OHGJPB6XOFAXYXM4GQCDXL4TTFANCNFSM4Z2EMC2Q. You are receiving this because you commented.Message ID: @.>

[alex-free]

Hi Alex,

May be the game size is small, Kurushi worked for me even without the patch on 80mins CDR, although the system.cnf is located at the last LBA.

從我的iPhone傳送

Alex @.***>於2023年6月19日 00:24寫道：



Confirmed this totally works with the Kurushi!

—

Reply to this email directly, view it on GitHubhttps://github.com/socram8888/tonyhax/issues/24#issuecomment-1596196939, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A45OHGJPB6XOFAXYXM4GQCDXL4TTFANCNFSM4Z2EMC2Q.

You are receiving this because you commented.Message ID: @.***>

I was able to reproduce this exact issue with the maxell CD-Rs (the worlds worst CD-Rs). Without patching it can't read it. With patching it can.

I'm not sure if high quality 80 min CD-Rs would work without patching because I'm still waiting for my Verbatim DataLifePlus 80min CD-Rs to arrive.

This bug is confirmed not to be present in anything SCPH-50000 or newer. Only SCPH-10000-SCPH-39004 have the seek issue. Is your console one of the affected systems?

[the7thchild]

Very nice to hear that. It seems that the patch will help the PS2 to handle a lot of scenarios. I will test resident evil 2 tonight and see if that’s going to work!

Best regards Charles

從 Windows 的郵件https://go.microsoft.com/fwlink/?LinkId=550986傳送

寄件者: Robert @.> 傳送時間: 2023年6月19日 上午 07:27 收件者: @.> 副本: @.>; @.> 主旨: Re: [socram8888/tonyhax] SYSTEM.CNF loading issue on PS2 consoles (#24)

Kurushi was the reason I opened this ticket. It might be a case that your system handles 80min CDRs better than mine. Until patching the disc image, I had never been able to boot Kurushi on my PS2.

Hi Alex, May be the game size is small, Kurushi worked for me even without the patch on 80mins CDR, although the system.cnf is located at the last LBA. 從我的iPhone傳送 Alex @.>於2023年6月19日 00:24寫道：  Confirmed this totally works with the Kurushi! — Reply to this email directly, view it on GitHub<#24 (comment)https://github.com/socram8888/tonyhax/issues/24#issuecomment-1596196939>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A45OHGJPB6XOFAXYXM4GQCDXL4TTFANCNFSM4Z2EMC2Q. You are receiving this because you commented.Message ID: @.>

— Reply to this email directly, view it on GitHubhttps://github.com/socram8888/tonyhax/issues/24#issuecomment-1596305129, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A45OHGMQ7H7BSBWUMHWQAODXL6FF3ANCNFSM4Z2EMC2Q. You are receiving this because you commented.Message ID: @.***>

[alex-free]

Before anyone tries this with a LibCrypt game (Dino Crisis Europe), don't waste the CD-R, yet...

I am working on some kind of generic Libcrypt1/LibCrypyt2 patcher... it will be combined with a future version of this patcher since it's all my code :)

[NotALuckyPey]

I checked patcher with 2 games on SCPH-39008: 1)RE Directors Cut Dualshock - now works fine 2)Pitfall 3D: for now tonyhax can read cnf and exe, but freezes on switching video mode. (maybe a bad laser, console has warranty sticker and never been opened) I will try to recheck later on SCPH-30001R (Problem with disc spindle)

[the7thchild]

The PS26MP patcher works like a charm. Resident Evil 1 Director's Cut now works on the PS2 with 80mins super cheap Verbatim CDR! Thanks Alex and everyone!

[alex-free]

thoughts @socram8888 ?

[socram8888]

@alex-free I'm wondering if it'd be possible to workaround this in software, by asking the drive to first seek to the last_lba*74/80, so the controller sees the last LBA is there to adjust its seeking, then seek normally.

[alex-free]

I'm still working on the documentation for this, but this should work for Dino Crisis Europe:

1) Patch with PS26MP. 2) Patch with this patcher:

aprip-1.0.3-windows_x86.zip aprip-1.0.3-windows_x86_64.zip aprip-1.0.3-linux_x86_static.zip aprip-1.0.3-linux_x86_64_static.zip

the syntax is this:

./aprip -b magic word here track 01 bin here

So for Dino Crisis Europe, my command was:

./aprip -b 6C3A '/home/alex/Downloads/Dino Crisis (Europe)/Dino Crisis (Europe) (Track 1).bin'

You can find the magic words at this site for most games: https://www.psdevwiki.com/ps3/PS1_Custom_Patches

Same deal, burn with some software that can regenerate EDC. Still a WIP but indeed a solution.

The aprip linked above should work for every LibCrypt2 game (most titles) allowing them to be burned by i.e. IMGBURN instead of CloneCD in that weird format.

I can definitely do LibCrypt1 support too, and LibCrypt3/4 probably can be made to work eventually as well. But I feel like 80% of games are LC2 if they have LC.

[alex-free]

I checked patcher with 2 games on SCPH-39008: 1)RE Directors Cut Dualshock - now works fine 2)Pitfall 3D: for now tonyhax can read cnf and exe, but freezes on switching video mode. (maybe a bad laser, console has warranty sticker and never been opened) I will try to recheck later on SCPH-30001R (Problem with disc spindle)

Can you link the redump page to the PitFall 3D game in question?

[NotALuckyPey]

Can you link the redump page to the PitFall 3D game in question?

http://redump.org/disc/3190/

[alex-free]

Can you link the redump page to the PitFall 3D game in question?

http://redump.org/disc/3190/

I didn't have any issue with PitFall 3D, without even using the patcher. Same story with Dino Crisis Europe, I didn't need to use the PS26MP patcher for either to work.

Good news is the LibCrypt patching works in aprip, Dino Crisis works burned via normal software. Official release of that here: https://alex-free.github.io/aprip

I think it may be hard to figure out a list of affected games that need PS26MP since it seems to vary console by console. So far only Kurushi has been able to reproduce the issue for me (unpatched, can't read SYSTEM.CNF)

[alex-free]

Interesting, I get this issue on a SCPH-1000 PS1 and a SCPH-10000 PS2 for Medievil (any version) but not on my SCPH-5501. I'm out of the worlds worst CD-Rs and can't test anything more yet

[alex-free]

It seems my SCPH-1000 won't read any 80 minute Maxell Music CD-Rs, so I remain firm that this is a PS2 specific issue. I have reproduced the issue in all the releases of MediEvil on my SCPH-10000 PS2. Due to an issue with the old PS26MP v1.0.1 MediEvil did not work after being patched. I have addressed this in v1.0.2, and renamed the tool 'PSX80MP'.

I am still working on an official release of the patcher, but I will attach them to this post: psx80mp-1.0.2-linux_x86_64_static.zip psx80mp-1.0.2-linux_x86_static.zip psx80mp-1.0.2-windows_x86.zip psx80mp-1.0.2-windows_x86_64.zip

Please note that the PAL releases of MediEvil (Europe, France, Spain, Italy, Germany) contain LibCrypt v1 protection. I have recently added LibCrypt v1 patching support to my aprip patcher: https://github.com/alex-free/aprip . You can run the PSX80MP patcher and then the aprip patcher and then burn with i.e. IMGBURN or cdrdao with the default generic-mmc driver and it will all work!

[alex-free]

@alex-free I'm wondering if it'd be possible to workaround this in software, by asking the drive to first seek to the last_lba*74/80, so the controller sees the last LBA is there to adjust its seeking, then seek normally.

A way to do this is to read the directory record sector(s) (starts at LBA 22) for the second to last file on disc and then seek to that. Then seek to the last file from there?

The thing is, there are games which have the SYSTEM.CNF as the second to last file, and then the PS-EXE as the very last. And we have seen on this thread that the PS-EXE in that case still can't be read so I'm really not sure if it will work. I am thinking it is still really probable it will overshoot due to tighter spiral winding in 80 minute media no matter what.

[alex-free]

https://alex-free.github.io/psx80mp

I have thoroughly tested every single game mention on this thread:

PitFall 3D Resident Evil Director's Cut Resident Evil 2 Kurushi MediEvil MegaMan X4

And every single game besides pitfall 3D reproduces this issue with my launch Japanese PS2 SCPH-10000 via my fork of Tonyhax. The last version of the patcher linked is confirmed to fix every affected game I've tried. Hurrah 😃

[alex-free]

Just found something relevant from Sony docs:

Also what no $ said here: https://www.psxdev.net/forum/viewtopic.php?t=3977&start=40

[gng4-github]

wasted many CD-Rs for the game Rescue Shot (Europe) to make it work with PS2 3XXXX until found this patcher, it works! Thank you very much.

however, after using the patch, PS1 console did not recognize the game, also not working with GS/AR CD after switching CD (both PS1/PS2)

hope there is a fix for this...

**this game and a few others: Ghoul Panic (Europe), Mighty Hits Special (Europe), must play with CD in order for light gun/mouse to work (other 2 no need the patch)

[alex-free]

wasted many CD-Rs for the game Rescue Shot (Europe) to make it work with PS2 3XXXX until found this patcher, it works! Thank you very much.

however, after using the patch, PS1 console did not recognize the game, also not working with GS/AR CD after switching CD (both PS1/PS2)

hope there is a fix for this...

**this game and a few others: Ghoul Panic (Europe), Mighty Hits Special (Europe), must play with CD in order for light gun/mouse to work (other 2 no need the patch)

Interesting, it should work fine on both if it was working on the PSX before. I'll look into it

[socram8888]

@alex-free I'm wondering if it'd be possible to workaround this in software, by asking the drive to first seek to the last_lba*74/80, so the controller sees the last LBA is there to adjust its seeking, then seek normally.

A way to do this is to read the directory record sector(s) (starts at LBA 22) for the second to last file on disc and then seek to that. Then seek to the last file from there?

No. Say, the SYSTEM.CNF is the last block in a 320000 sector image. Instead of asking the disc to seek for that sector, you'd first seek for 320000*71/80, then seek for the real data block.

What I essentially wonder is if you can "train" the firmware on the PS2's CD reader to learn that the data is more tightly packed than it expects, since it should be able to handle slight manufacturing differences between CDs.

[alex-free]

@alex-free I'm wondering if it'd be possible to workaround this in software, by asking the drive to first seek to the last_lba*74/80, so the controller sees the last LBA is there to adjust its seeking, then seek normally.

A way to do this is to read the directory record sector(s) (starts at LBA 22) for the second to last file on disc and then seek to that. Then seek to the last file from there?

No. Say, the SYSTEM.CNF is the last block in a 320000 sector image. Instead of asking the disc to seek for that sector, you'd first seek for 320000*71/80, then seek for the real data block.

What I essentially wonder is if you can "train" the firmware on the PS2's CD reader to learn that the data is more tightly packed than it expects, since it should be able to handle slight manufacturing differences between CDs.

As it is fatal to seek past the end of the disc, you'd still need to read the directory record to find out the true last LBA (some of these effected games don't even have that many sectors). That would add quite a bit of overhead but testing it out would be most likely worth it.

[alex-free]

Something else I wonder is what happens when the game goes to read the last file on the disc with, someone above mentioned the patcher might help with that case as well. From what I can tell though in-game usually the retry on seek error code is much more generous then the boot sequence/tonyhax loader.

This whole issue is because (mostly) SCEE devs didn't listen to Sony docs and Sony didn't enforce it's own rules knowing the hardware because it worked fine with 71 minute media (and 74 minute CD-Rs even as we have seen). It is almost like the most effective non-intentional anti-piracy measure ever if you think about it.

Lastly, about your idea with training the drive. If the drive can seek to the SYSTEM.CNF (which is the second to last file in some games) and then can't seek to the last file (the PS-EXE derived from the SYSTEM.CNF) then I'm not sure if that would work because I would think it goes from a very high LBA to another very high LBA towards the end of the disc and still fails.

I will of course however test anything you make, although my only working PS2s are my Japanese SCPH-10000 launch consoles (my USA one stopped reading CDs) so I'll have to adapt your code for International.

[alex-free]

I bought a 'PlayStation Master Disc 71PS', the ones used for testing on the debugging station/sending out to Sony to make the CD-ROMs, and I can confirm that they are 71 minutes long.

ATIP info from disk: Indicated writing power: 4 Is not unrestricted Is not erasable ATIP start of lead in: -11849 (97:24/01) ATIP start of lead out: 319725 (71:05/00) Disk type: Long strategy type (Cyanine, AZO or similar) Manuf. index: 25 Manufacturer: Taiyo Yuden Company Limited

[gng4-github]

...... ... after using the patch, PS1 console did not recognize the game, also not working with GS/AR CD after switching CD (both PS1/PS2) ......

not sure if this is relevant, after using the patch, when verifying the burned disc, reported error on sector 22, and the verifying process stalled at 9X%

[alex-free]

...... ... after using the patch, PS1 console did not recognize the game, also not working with GS/AR CD after switching CD (both PS1/PS2) ......

not sure if this is relevant, after using the patch, when verifying the burned disc, reported error on sector 22, and the verifying process stalled at 9X%

That is expected if your talking about IMGBurn because PSX80MP doesn't update EDC/EEC data, it lets IMGBurn do that during the burn.

[gng4-github]

found from attic an unused Mistsubishi CD-R, checked atip: it looks like the same as Verbatim DataLifePlus, will that prove they are the same?

also like to know about "Indicated writing power," what does it mean (couldnt find any related info)?

have some Ricoh TYPE 74 HR100 Premium CD-R (74min), slim PS2 can read, but not phat 3000X, 50/50 chance 550X, atip shows writing power 4, not like cheap CD-R as 5 (readily readable by PS/PS2), possible to tweak?