Pressing CDs

[Arkhan]

So, this means (i think), that the actual problem is that HuC is making f*cked up ISO files and CDRwin fixes them.

[nodtveidt]

That might be the problem. Whatever it is though, you guys got it to work, and that's the significant part... so if you can make it work, it can be done, and that's just that.

[Arkhan]

Well, you could take the crapshoot and just use CDRwin's bin instead of huc's ISO, and see if that presses, and shows up operational.

[nodtveidt]

Gonna order some CDRs online... found a place that sells 'em, and they say that they ship to PR, but the dropdown for "State" didn't have PR on it. I was like "WTF MORONS".

[TheOldMan]

I noticed a considerable size difference as well

Hmmm... I must have been looking at the wrong file. I don't know if I still have the actual files we burned, aside from the audio.

But yeah, the larger file from cdrwin contains the ecc codes (that's the difference in size). You need those to burn the (larger) mode1 track, otherwise the sector alignment gets screwed up.

I *think* if you put it all together, you should have a masterable cd. Just don't use Nero. Or, if you do, make sure it really burns a mode 1 track (not XA/form1).
.............
It's starting to look like Nero isn't adding the ecc codes correctly -or- it isn't burning the track correctly. I can't swear thats what is happening, but it sure looks like it.
And once we figure this out and have discs, someone needs to go back through this thread and figure out we did for posterity

Now, if it will still boot and play... <crosses fingers>

[Arkhan]

So, HuC strikes again?

lol

I don't use nero.  Is it supposed to automagically assume it is supposed to add the error crap for you?

[Bonknuts]

But.. HuC isn't outputting a wrong file, it's just outputting a cooked binary. >_>

7575792 vs 6596608

The original ISO is 6,596,608 bytes
The BIN ripped with CDRWIN is 7,575,792 bytes

 There's nothing unusual about that. That's the size of a cooked binary vs a binary with mode 1 raw sector support bytes. Both those add up fine.

 You guys know what cooked binary vs raw binary is, right? All cooked binaries are converted to 2352 byte sector as they're are written to the CDR (actually, in a buffer first and it's done by the software). Storing the ECC, EDC, SYNC, HEADER, etc additional bytes is redundant. Thus why cooked format is popular. Plus, you can directly edit the bytes with a hex editor or any software much easier. Cooked or raw, it doesn't change the LBA address of the data.

 There's no such thing as a 2048 byte sector for mode 1. It's all 2352 bytes(plus 96 additional bytes for sub channel data). That is to say, all cooked and raw binary files end up being in mode 1 sector format (2352 without subchannel 96bytes) - assuming you choose mode 1. 2048byte is what's available as data. It doesn't throw anything off for the CD hardware reading the data. The LBA address and offset are the same regardless for CD data. Your LBA address is in sectors, not bytes. So if your LBA offset was 0x20 (relative to index 01) for a CD_READ, then it's gonna read 2048 bytes of data regardless that it's now the full mode 1 sector size. And those 2048bytes are going to be the same bytes as in the ISO HuC spits out (LBA_address * 0x800 = byte_address). There's another problem with an app making a RAW mode binary. If an app messes up the sync or header byte values and you burn it as is, you'll get read errors. It's more convenient to let the burning app handle this or an external app for conversion (well, IMO ).  Anyway, here's a good site showing the mode 1/2/forms break down..



 I don't doubt Nero is messing stuff up for you (making non bootable TurboCDs), but I've been using Nero for about 6 years or so without problems for making TurboCDs. I don't create mix moded discs or such in the project, I tell Nero to burn the CUE file (but I assume you guys are as well). I've been using 6.xx.x.x.x or something Nero and still use it. I have no idea about the new versions though (12 is the newest I think). I did have a problem once, but it turned out to be the drive. It wouldn't burn DAO RAW (it had to be DAO RAW/96, but I didn't include a sub channel file). I've had good results with Alcohol 120% too with burning TurboCD stuff (and even on laptop cdr drives). But if Nero is burning the data track as XA mode 2 form 2, then it shouldn't work on the PCE CD. Would probably work on the PC with some emulators as accessing the physical drive for an XA form 1 CDR data track (because it should be getting cooked data from the sector request for the app).

I don't use nero.  Is it supposed to automagically assume it is supposed to add the error crap for you?

 Every burning software that encounters a data file that's listed as mode 1/ 2048, yeah. It wouldn't burn right otherwise.

Code: [Select]

FILE "lords.iso" BINARY
  TRACK 01 MODE1/2048
    INDEX 00 00:00:00
    INDEX 01 00:47:09


 Ok, so Nero completely failed for you guys so the mastering company never got a CDR burned from Nero. Got it.

 EDIT:

  The Old Rover: If you can dump the CDR you burned in raw mode, I can show you where to look to see what format the sectors in the pregap are labeled as. That might shed some light on why eclipse is reading them as audio sectors/data (that's what the DDxxx whatever error looked like).

[Arkhan]

...You've also never sent a burned CD-R in to get pressed, so for all you know, you could have the same "working" discs as Rover.  I've used Nero in the past to create discs to play on my CD-R.  I wouldn't trust them to get pressed.  Rover's shit he burns at home in Nero plays on hardware fine, but f*cks up at the pressing house.  It may work on a the TGCD (where perhaps it is a bit lazy with reading through the ECC)...

but a pressing setup is a bit (a lot) more picky.  "has to be perfect" picky.

I'm saying, a cooked binary may be throwing the whole process off, where a raw bin/cue extracted properly from CDRWin with the ECC injected properly before-writing doesn't. 

So, Nero ends up writing a disc that accidentally runs on hardware, but is not correct.

[nodtveidt]

Tom, Nero burns the ISO as MODE2, even if the cuesheet says to burn as MODE1. CDRWIN confirms this.

[Arkhan]

hah, so its burning as MODE2 (Form 1?), which has an 8 byte header at the start of the data section...

whereas Mode 1 has these 8 bytes as unused at the END of the data section.

probably causing an overlap....

Confirming that Nero is incorrectly burning the cooked ISO that HuC produces.  Rather than stuffing in the proper ECC where needed like CDRwin does, it just decides HEY this is close enough! HAH!

... and ruins everything.    Like I said, it will accidentally work if you play it, but won't confirm to the presser trying to create a Mode 1 disc.  There is crap where it shouldn't be.

Sweet.

[nodtveidt]

I verified it by using Nero to burn the ISO by itself to CDR, then used CDRWIN to rip the track from the CDR. Indeed, the resulting .cue stated MODE2/2352. Mounting the ISO with Alcohol 120% and then ripping that with CDRWIN resulted in MODE1/2352... success.

[TheOldMan]

Mounting the ISO with Alcohol 120% and then ripping that with CDRWIN resulted in MODE1/2352... success.

Don't get too happy yet
Can you burn a bin/cue with alcohol 120? (I've never used it). If so, you need to build the cue and burn a test disc to make sure all the audio, etc works. And that it boots and runs.

If that works, you -should- have a good-to-go master. Here's hopeing

<crosses other set of fingers...and some toes>

[Bonknuts]

Awesome work Nodt! Wow, so your Duo is actually reading a mode 2 form 1 data format. Mode 2 form 2 (which is just normal mode 2) was defined in Yellow book, but XA form 1 wasn't defined until 1991. Just speculation but maybe there's 'firmware' in the CD units that are more up to date. And some CD units that can only read mode 2 form 2, are reading the same data sectored but without applying ECC. Either way, mode 2 is for the birds...

 Just curious, have you tried burning the ripping cue/bin from the virtual drive by CDRWIN, with Nero and rechecking it to see if Nero follow the raw format to the T?

 This doesn't just help out for pressing, but for people burning CDRs too (on those finicky systems).

[nodtveidt]

Don't get too happy yet
Can you burn a bin/cue with alcohol 120? (I've never used it). If so, you need to build the cue and burn a test disc to make sure all the audio, etc works. And that it boots and runs.

If that works, you -should- have a good-to-go master. Here's hopeing

<crosses other set of fingers...and some toes>

Gonna order some CDRs and test it... the ones I have right now are inadequate.

Awesome work Nodt! Wow, so your Duo is actually reading a mode 2 form 1 data format. Mode 2 form 2 (which is just normal mode 2) was defined in Yellow book, but XA form 1 wasn't defined until 1991. Just speculation but maybe there's 'firmware' in the CD units that are more up to date. And some CD units that can only read mode 2 form 2, are reading the same data sectored but without applying ECC. Either way, mode 2 is for the birds...

 Just curious, have you tried burning the ripping cue/bin from the virtual drive by CDRWIN, with Nero and rechecking it to see if Nero follow the raw format to the T?

 This doesn't just help out for pressing, but for people burning CDRs too (on those finicky systems).

Yes, the Duo was reading the MODE2 discs just fine, as long as I was using high-quality media and not cheap shit. I can verify this as well by ripping any one of the burned CDRs I've made over time. However, I wouldn't have known to even do this if not for Arkhan and TheOldMan, so the credit really goes to them.

[Arkhan]

Awesome work Nodt! Wow, so your Duo is actually reading a mode 2 form 1 data format. Mode 2 form 2 (which is just normal mode 2) was defined in Yellow book, but XA form 1 wasn't defined until 1991

I think basically everyones Duo/TGCD can read these.  I used to burn downloaded games with Nero back in the day before I got CDRwin to operate.  I've also had good luck with Alcohol 120% (It burns Bin/Cue, OldMan).

My guess is that since the CD Rom hardware is so old, its a bit lazy with things, and eventually does just find the data it wants, even if its a bit off.

My TGCD, white PCE CD, Duo-R, and the spare TGCD I sold, all read discs burned lazily with nero back when I was ~12 or 13. 

It's accidental workage.

I'd also bet that if the pressing machine would actually allow the stamper to be made, the discs produced, while "incorrect" would still actually work. 

Nodt, I guess this means we actually earned our Special Thanks spots in the credits? lol.  (unless demoing it at a con was good enough? hah)