VOGONS

Common searches


Temple of Apshai Trilogy

Topic actions

First post, by Sardonyx

User metadata
Rank Newbie
Rank
Newbie

As some probably know, the Temple of Apshai Trilogy was just released on Steam yesterday. They're distributing the DOS version with DOSBox.

After playing it for a bit, I noticed it plays very oddly for Apshai. For one, the monsters do almost no damage, to even lightly armored characters. I've never gotten any experience, even from completely clearing a level. And the amount of silver it gives out is crazy. A couple trips in to the room with a copper ingot will easily net thousands of silvers. There's no challenge at all - and that's bizarre for Apshai. It's usually quite difficult starting out.

It's been a while since I played so I tried the versions of Trilogy for C64 and Atari, as well as the original Temple of Apshai for C64 just to make sure it's not bad memory. Even with the same character stats, those versions are as difficult as I expect.

Overall I noticed in the version on Steam:

  • The monsters do very little damage
  • The game never gives any experience (always 0)
  • The game gives way more silver than items are valued at
  • Monsters don't move when resting (taking 0 steps)

I also tried downloading a few different versions for DOS I found online and they play identically to the one on Steam.

So...has the DOS version always been broken? Is this a case of a bad/trained/hacked copy being put online and everyone's passing around that copy, including the current rights holder? Is this some type of copy protection that's taking all the challenge out of the game?

Just wondering if anyone has any more info on this, or ideally if someone has an original DOS version on disk they could test or compare to what's on Steam/online.

Reply 1 of 24, by Jorpho

User metadata
Rank l33t++
Rank
l33t++

An intriguing question. There are certainly other old DOS games on Steam that are just lazy repackagings of warez. (I'm thinking of Drakkhen.)

Sardonyx wrote on 2020-10-20, 21:27:

I also tried downloading a few different versions for DOS I found online and they play identically to the one on Steam.

Were you using them all in DOSBox? An obvious step would be to try them on authentic hardware, or in PCem, or even in a more recent version of DOSBox.

You can easily compare different versions by checking the checksums of the files.

I see there's at least one video on Youtube; how does that compare?
https://www.youtube.com/watch?v=aoTYCTUKfK0

Reply 2 of 24, by Sardonyx

User metadata
Rank Newbie
Rank
Newbie
Jorpho wrote on 2020-10-20, 22:01:

An obvious step would be to try them on authentic hardware, or in PCem, or even in a more recent version of DOSBox.

I ran them in DOSBox. Unfortunately I don't have any vintage PC hardware on hand. I'm not familiar with PCem, but I'll take a look.

Jorpho wrote on 2020-10-20, 22:01:

You can easily compare different versions by checking the checksums of the files.

That's a good idea - I think they'll all be the same, but I will verify.

Jorpho wrote on 2020-10-20, 22:01:

I see there's at least one video on Youtube; how does that compare?
https://www.youtube.com/watch?v=aoTYCTUKfK0

Looks about the same as what I'm seeing. He setup a mediocre character with no armor or shield and is only occasionally taking 1% damage at a time in combat. He should be getting destroyed.

This is what I would expect from someone playing with a random (non-tweaked) character: https://www.youtube.com/watch?v=jmuEtgzjVyY

Reply 3 of 24, by Sardonyx

User metadata
Rank Newbie
Rank
Newbie

I generated MD5 sums and compared them across the Steam release and other archives I could find. They're all the same. Some will occasionally include a save game, character file, or batch file to start it, but all of the game's executable and data files are identical.

If someone has Disk version they could compare against that would be amazing. Here's the MD5 Sums from the Steam release:

# MD5 checksums generated by MD5summer (http://www.md5summer.org)
# Generated 10/20/2020 7:16:30 PM

c92f2665d3d56bbdf5dd1944120cb26b *LEV1.BAD
3bd1647a75bd75f536bd0db1913f1b0a *LEV10.BAD
21f03d670f8ed949767b744cfdf85c3d *LEV11.BAD
9522b9b3db225aac5abad0edafa1bd6d *LEV12.BAD
53c20beaa07064e8bf71a0b90a7b1da8 *LEV2.BAD
a54b22988e603d8176909028e320b38d *LEV3.BAD
278c8b0537cffc8da2e2abeb5109222e *LEV4.BAD
489bac35bad310c279b43f52f7ab1177 *LEV5.BAD
55075621c29d8cd6c24f88c6f4e93703 *LEV6.BAD
678cf48329799050328ae6dd3e8aabd3 *LEV7.BAD
72b5cb9034ebacbfeda031d9ab843c24 *LEV8.BAD
77ac6062e30994e56d7429b64920d83e *LEV9.BAD
90cf5ccb5abaf7329fbcb7363cb1876c *LEVEL1.LVL
c9887d600cb96dae10e1109e5ce514d7 *LEVEL10.LVL
a672d3c86defb340b194c0b7170288b1 *LEVEL11.LVL
f045432fbe5af94a02b31d876c1960fe *LEVEL12.LVL
b2b14942e85f17cf427982961b6fff19 *LEVEL2.LVL
b76c396531391696f04fe27575c9e756 *LEVEL3.LVL
72d0c4a6870b6bc119b516d4fa339552 *LEVEL4.LVL
b61271a4ed421d0a93c6f74483ac2d93 *LEVEL5.LVL
73bf6785af9ec96e9612e621e48563cd *LEVEL6.LVL
8c2d9302758ffe9558d7511b3ad53fdc *LEVEL7.LVL
ec484b39d8d431058a33f450921f77d9 *LEVEL8.LVL
5f6ef98bfb7e78167eb76083c8303475 *LEVEL9.LVL
9ad3ec887eb4d018db6d1759f76a33a3 *STARS.ANI
a575ba280aba5707effb10c5aa8d08b5 *SWORD.PLR
ed8ac3652880712b0cbfa658552c6808 *TITLE.SCR
cc07e66fbde869caae60c31038138693 *TRILOGY.000
9020683f8a2e24a2daf7f483e8289234 *TRILOGY.001
ca7bee9b116137bc6881c23db355b04c *TRILOGY.002
188430de49ff1500eb767660c50a8b81 *TRILOGY.003
4af67ddf68283de286614d07448c8492 *TRILOGY.004
a124ea49c3b49897c7af2a2a22ed3862 *TRILOGY.005
95ed9092d799117f90ebf7dfe8cdc64c *TRILOGY.006
0ddfd9caa1693b795a2fbc6e83d2e93d *TRILOGY.007
53ee2d3d92eb685be52dd3da7e1942c9 *TRILOGY.008
f364406d98eb07dd1625baaec9fb5c14 *TRILOGY.009
77575b3b9ff3f3544a9c8dc20d59a265 *TRILOGY.COM

Reply 4 of 24, by Sardonyx

User metadata
Rank Newbie
Rank
Newbie

One more follow up. I tracked down everything for PCem. I set up an IBM PC with DOS 3.30 and the made a 360K floppy disk image from the Steam version of the game. It plays exactly the same in PCem. I went through a couple rooms, killed some Mosquitos and Swamp Rats, picked up the Copper Ingot, a piece of trash, and left the dungeon with 96% health. That little adventure netted me 450 silver and 0 experience. The Copper Ingot is worth 20 silver and the Trash is worth 0. I'm not sure of the Experience values, but every kill and treasure is worth some. So something's definitely not right.

Reply 5 of 24, by Jorpho

User metadata
Rank l33t++
Rank
l33t++

I agree it would be swell if someone with the original disks would be able to verify the checksums.

But in the meantime, one other vague possibility comes to mind: what if you run the game in PCem, but set the (emulated) system clock to a date in 1986? Maybe the programmers used the system date for a random number seed or something, and never anticipated that the game would be run decades later.

Reply 6 of 24, by Sardonyx

User metadata
Rank Newbie
Rank
Newbie
Jorpho wrote on 2020-10-21, 03:01:

But in the meantime, one other vague possibility comes to mind: what if you run the game in PCem, but set the (emulated) system clock to a date in 1986? Maybe the programmers used the system date for a random number seed or something, and never anticipated that the game would be run decades later.

It was a good thought. I hadn't thought of trying the date. Even in PCem I just automatically set the date to today.

So I gave it a shot, setting the date to 05-11-1986. It didn't change anything. The randomly generated character I used this time was particularly crappy. Just out of curiosity I tried it out in some of the higher levels and the enemies can do a few % damage to me. They should have one hit killed me.

Reply 7 of 24, by NewRisingSun

User metadata
Rank Oldbie
Rank
Oldbie

TDC has a Kryoflux image of a 720 KiB disk of the game. The files are identical to yours, with the exception of additional "AUTOEXEC.BAT" and "GO.BAT" files.

Unless somebody unearths a 360 KiB disk of the game with different files (such as this one or this one), I would say that the finding is just that the PC trilogy version (written in Turbo Pascal) was made a lot easier for some reason. For what it's worth, the original 1982 PC version (written in BASICA) is as difficult as expected.

Reply 8 of 24, by Sardonyx

User metadata
Rank Newbie
Rank
Newbie
NewRisingSun wrote on 2020-10-21, 06:41:

TDC has a Kryoflux image of a 720 KiB disk of the game. The files are identical to yours, with the exception of additional "AUTOEXEC.BAT" and "GO.BAT" files.

Unless somebody unearths a 360 KiB disk of the game with different files (such as this one or this one), I would say that the finding is just that the PC trilogy version (written in Turbo Pascal) was made a lot easier for some reason. For what it's worth, the original 1982 PC version (written in BASICA) is as difficult as expected.

Thanks for verifying that. I did see an AUTOEXEC.BAT and GO.BAT in some of the archives. They're not present in the Steam release though.

It's so strange that the PC version of Trilogy would differ from all the others. The releases for the Atari ST and Amiga are from the same time and play the same as all the other releases.

I would characterize this version more as "broken" or "bugged" rather than just easier. The experience and progression mechanic is completely non-functional.

I'm tempted to get one of the 360K disks, but I'm not sure if I have a 5.25" drive for PC any more. I think I still have a 386 in storage, but most of the vintage/retro hardware I have is for 8-bit era systems (Atari/Commodore).

One other oddity - I've spent some time scouring the internet looking for a photo of the 3.5" DOS disk. The only one I can find looks like the label was printed on a dot matrix printer and is text only. It doesn't have any of the standard logos, the Epyx name or anything normally associated with the game. Which is bizarre as the other versions have professionally printed labels. The 5.25" disk seems to be fairly common, I can find lots of photos of it. I can also find photos of the Atari ST and Amiga disks from the same time and they have professional quality labels on them with the expected logos.

It could just be a buggy release, but I'm still curious about the 720K floppy that was archived. I'm not very familiar with TDC - does it contain photos of the actual copy of the game the disk image was made from?

Reply 9 of 24, by NewRisingSun

User metadata
Rank Oldbie
Rank
Oldbie

The only one I can find looks like the label was printed on a dot matrix printer and is text only.

Sounds like a "Slash" release.

I'm not very familiar with TDC - does it contain photos of the actual copy of the game the disk image was made from?

Not that I know of. But the disk was originally uploaded to TDC by "keropi", who frequents vogons.org as well; maybe he could photograph his copy if he still has it.

Reply 11 of 24, by Jorpho

User metadata
Rank l33t++
Rank
l33t++

Well, it wouldn't be the first time a game got a seemingly-functional PC release that was actually badly broken. (Metal Gear is a spectacularly inexplicable mess.)

But it's a shame that someone went through the trouble of getting whatever rights are necessary to redistribute this game, only for anyone who tries it to get an unrepresentative experience.

NewRisingSun wrote on 2020-10-21, 06:41:

Unless somebody unearths a 360 KiB disk of the game with different files (such as this one or this one), I would say that the finding is just that the PC trilogy version (written in Turbo Pascal) was made a lot easier for some reason.

Is it written somewhere that it was done with Turbo Pascal?

Someone managed to painstakingly reconstruct the original Turbo Pascal source of ZZT through an exhaustive effort of reverse engineering.
https://blog.asie.pl/2020/08/reconstructing-zzt/

Someone feeling unusually quixotic could conceivably do the same thing here and perhaps determine if there was some trivial bug at fault.

Reply 13 of 24, by Sardonyx

User metadata
Rank Newbie
Rank
Newbie
keropi wrote on 2020-10-21, 15:18:
here you go, this is my disk/box: […]
Show full quote

here you go, this is my disk/box:

M3iqXhL.png

😀

Thanks for sharing these! That's interesting. From what I can find, Keypunch published a release under license from Epyx in 1989. It looks like a lower cost re-release. They're known for their budget releases, so maybe something happened with this version.

I wonder if this is the same thing that's on an Epyx release from '86 or if Keypunch some how ended up releasing a bad version. 🤔

It's certainly possible there never was a good version for DOS, but it's also possible what was released earlier by Epyx was different.

Either way it's an interesting bit of history.

Reply 14 of 24, by NewRisingSun

User metadata
Rank Oldbie
Rank
Oldbie

I have asked on TDC for anybody to contribute an image of the 360 KiB version. Let's see what happens.

The 360 KiB version is listed as having some kind of key disk copy protection scheme in a list of copy-protected games by Moebius, of The Good Old Days. Since the commonly-available version does not require a key disk, it obviously must have been modified in some way.

Reply 15 of 24, by Namrok

User metadata
Rank Oldbie
Rank
Oldbie

Turbo Pascal you say...

This might be old hat here, but once upon a time I made an attempt to play the DOS version of Wizardry 1 in dosbox. The RNG seemed totally FUBARed. I was going out of my mind, all my characters were constantly losing stats every time they leveled up. I never once successfully raised a dead character. My experience wasn't mapping to any of the blogs or discussion I saw about Wizardry on other platforms. I went looking through what these chances were supposed to be. Even found this old document called the "Wizisystem" I think that exhaustively detailed all the formulas Wizardry uses. At the end of my quest, I found a github repository of someone that had decompiled the pascal code Wizardry was written in for the Apple II, and double checked the formulas myself.

Then I noticed the rand function for Pascal on the Apple II was different from the rand function for Pascal in MSDOS. If memory serves, one returns 0-255, and the other returns 0.0-1.0. Or something like that. I'm more or less convinced whoever ported it to MSDOS fucked this up, as certain RNG outcomes seemed clamped to one end actually playing it.

Makes me wonder if something similar happened here.

Win95/DOS 7.1 - P233 MMX (@2.5 x 100 FSB), Diamond Viper V330 AGP, SB16 CT2800
Win98 - K6-2+ 500, GF2 MX, SB AWE 64 CT4500, SBLive CT4780
Win98 - Pentium III 1000, GF2 GTS, SBLive CT4760
WinXP - Athlon 64 3200+, GF 7800 GS, Audigy 2 ZS

Reply 16 of 24, by NewRisingSun

User metadata
Rank Oldbie
Rank
Oldbie

The 5.25" disk's content is identical to the 3.5" disk's, except for the fact that TRILOGY.COM is wrapped with the Formaster Copylock keydisk checking routine. So, no long-lost version with correct gameplay. 🙁

Reply 17 of 24, by Sardonyx

User metadata
Rank Newbie
Rank
Newbie
NewRisingSun wrote on 2020-10-22, 15:45:

The 5.25" disk's content is identical to the 3.5" disk's, except for the fact that TRILOGY.COM is wrapped with the Formaster Copylock keydisk checking routine. So, no long-lost version with correct gameplay. 🙁

Thanks for helping track it down. That's unfortunate 🙁

Is there any chance the game executable is looking for something on the disk (weak bits, half tracks, etc.) and running differently if it doesn't find them? I know that's a long shot and perhaps the disk imaging method already accounts for those types of protection schemes.

I'd love to know more about how this was released in this state. It's a terrible representation of the game and unfortunately what a lot of people will experience now that it's on Steam.

Reply 18 of 24, by NewRisingSun

User metadata
Rank Oldbie
Rank
Oldbie
Sadonyx wrote:

Is there any chance the game executable is looking for something on the disk (weak bits, half tracks, etc.) and running differently if it doesn't find them?

No. When I say that the keydisk protection scheme "wraps" the program file, I mean that the keydisk checking routine replaces the protected program's original starting point. It looks for the specially-formatted track, and if it finds it, executes the protected program normally by branching to the protected program's original entry point, and if it does not find it, just bails out to DOS. So by the time you see the title screen, the protection is already gone and done.

And the game behaves identically even with a protected disk image in an emulator that supports protected disk images. 😀

Last edited by NewRisingSun on 2020-10-22, 21:30. Edited 1 time in total.

Reply 19 of 24, by Jorpho

User metadata
Rank l33t++
Rank
l33t++
Sardonyx wrote on 2020-10-22, 21:14:

I'd love to know more about how this was released in this state.

It would make for an interesting story, no doubt – but like I said, reverse-engineering the code is far from a trivial matter.

(I'm a little surprised no one's done a deep-dive into that DOS port of Wizardry. That series has no shortage of fans.)