Enigma Tape Magazine

[TMD2003]

Achtung!

I've now had a bit of time to work on Enigma 1. I've trimmed Louise's original WAV file to the correct length so the tape starts within about half a second, and ends where it's supposed to. It's had all the hideous noise flattened so that only the loading signals and a bit of tape hiss are present. And, where the Players CD Player ("Program: -cd-") had a false start, I've cut that out - but other than that, there are no alterations made to the lengths of the gaps between blocks, which is how it's supposed to be. And it all loads perfectly in Fuse, using F8 to toggle the tape on and off.

What I have not been able to do is make it into a working TZX - it will definitely require someone who knows how to handle the turbo loading blocks with the high-pitched pilot tones, because the basic version of MakeTZX is completely flummoxed.

So here's the cleaned-up WAV that should give alternative programs the best chance of success:

Enigma Tape Magazine 1 - WAV/RAR file

[TMD2003]

Achtung nochmal!

I've now done the same processing on Enigma issue 2. Note that I HAVE REMOVED THE LINK ABOVE and re-uploaded Enigma 1 with a LibreOffice spreadsheet, with timestamps for each block in the processed WAV. This should help whoever is going to do the final TZXing with where all the STOP THE TAPE prompts should be inserted that emulators pay attention to. I don't always see on screen what the header name is - either it's been POKEd away or it's black-on-black - but where I do know the header name, I've included that.

In issue 2:

• The editorial instructs those who want back issues (of which there was only one) to specify if they want it on tape or +D disc. That means that there are disc copies of Enigma out there that should also be considered MIA... what are the chances of those being found? What's Hamish Rust doing these days? It seems he was the one with the +D who made the +D utilities so I assume it'd have been him making the discs in the first place...
• It also confirms the complete list of what was on Issue 1, and that petrol was £1.96 per gallon in 1989 (or 43.1p per litre). Sounds about reasonable for what I remember, given that the Gulf War at the end of 1990 put the price above 50p per litre for the first time. The Bank of England's inflation calculator says that's the equivalent of £1.085 now. In my neck of the woods it's around £1.40 and it was a lot worse than that this time last year... draw your own conclusions.
• And there's a review of Renegade III. Brace yourselves...

I would now say there is further evidence that the extra junk on the end of the raw rip of Enigma #1 was an early work-in-progress copy of issue 2. The Denizens demo that didn't quite load (one of the headers wasn't recognised) was presented entirely with the regular loader, whereas here on issue 2 (where it loads and runs properly) it uses the turbo loader. Also, all the headers were in capitals on that tape, but are in lower case on issue 2. I'll see if I can get the "issue 1" version working and see if it's markedly different.

ISSUE 1 e1final.rar: https://mega.nz/file/5GghwL4B#PCh9gUewj ... HTcFPuTW6U
ISSUE 2 e2final.rar: https://mega.nz/file/VWIRyZZa#jYyzla8Vt ... Ic_AXY6fx4

Everything that I can possibly get to work (i.e. it doesn't require extra knowledge of the workings of a +D, or a disc with erased files on it...) is working, using Fuse and loading at regular speed, so the WAVs are ready for TZXing.

As long as there's someone who knows how to handle those turbo blocks, that is!

I'll have a shot at issue 3 this afternoon.

[TMD2003]

And in issue 3:

• "Good news! Enigma is now available on +3 disc!" Oh, joy. That's another probably-long-lost format we're hunting for!
• There's another tape magazine called F-S-S from round about the same time. This isn't even mentioned in the archive at all... unless it was short for something that is archived.

@LouEnPointe's "README" files that are stashed with the original WAVs are her observations while ripping the tapes to WAV that I might want to note. This one, apparently, was highly troublesome, and was full of noise, so I've given it the most comprehensive de-noise-ification that I could manage. The W...I...D...E gaps between the blocks might not have been how they were originally - some of them are over 30 seconds - it could be that Louise has sampled and re-sampled the tape in bits and stitched them all together. The original WOS preservation team from Martijn's day were very anal about preserving tapes exactly as they were found originally, i.e. thou shalt not use the TZX Beautifier on pain of being cast out to use an unexpanded VIC-20 for all eternity, so if the 30-second gaps were original, they stay, but if not, should I find out how long they really were and trim the gaps to length? Or do we not care about that? A TZX with STOP THE TAPE prompts in it, surely, will not preserve the gaps anyway, as that's considered to be a gap of 0 ms.

There are also two Bytes: blocks - "D1" and "D2" respectively - that contain text (I can tell by the loading sounds, that's how much of a Spectrum nerd I am), and are unassembled machine code designed to be used with the assembler in Issue 1. Using a real +2B, I've assembled these files and I can confirm that they do make border effects, as specified in the dot-matrix-printed info sheet.

Even the date of these early issues is missing on the archive. I can confirm that the first four issues are all from 1989, issues 5-8 were from 1990, and logically, issue 9 should have been 1990 (issue 8 was Aug-Sep 1990) but... maybe Christmas or making a more professional-looking tape got in the way? Issue 9 is listed on the archive as 1991.

Gap quandaries aside, everything works, hence a third WAV file can be given a clean bill of health. Furthermore, I've collected some screenshots from the first three tapes that could go on the archive straight away. Super! Smashing! Great! Bendy Bullies all round! Of course, whoever can make these into fully functioning TZXs will win Bully's Special Prize, which is definitely not a speedboat. No, really, it isn't.

ISSUE 3 e3final.rar: https://mega.nz/file/sa4UQKqA#N30Bt3yiZ ... LmOFPlf_PM

[TMD2003]

Issue 4 consists entirely of the magazine, with no extra demos or tiny utilities that (a) require a +D, (b) Hamish Rust wrote in five minutes between mouthfuls of Irn-Bru, (c) both.

On the loading screen, the programmers have forgotten how to spell "independent", having had it correct on issue 3. In that same issue, they'd printed an indignant letter chastising them for their poor spelling - which hasn't improved much. Maybee their ordishoning four a riteing job with the Grauniad.

It also contains a larger than usual number of reviews, some of which are for Ocean's Hit Squad re-releases. Of particular interest was Batman - the first game, the 3D isometric game similar to Head Over Heels; the review says it was written in 1985, before the launch of the 128K Spectrum, but has some excellent 128K sound effects. Really? I'm going to have to check that. And I don't expect the game to load in 128K anyway, as the tape was encoded with Speedlock 1, which was 48K only - and that's for both the original release and the Hit Squad version. I'm going to have to investigate that further and see if Enigma's reviewer was stretching the truth. I expect to "don" a nylon wig, paint my face with some spare orange paint I have in the garage, point at them and shout "YOU ARE FAKE NEWS!" But... are they?

In very definitely not fake news, I can confirm that the WAV of this tape - which I've given a similarly thorough spruce up as issue 3 - is fully working, every block loads, and is available for anyone who will attempt to TZX it. I have heard that .ODS files, such as the one containing the timestamps in this package, can be opened even on Bill "Trust me, I'm a multi-billionaire, what could possibly go wrong?" Gates' spreadsheet program, I think it's called Extra Large or something like that.

ISSUE 4 e4final.rar: https://mega.nz/file/5WYR1T4Z#XJ80D110S ... ZmT_eRL350

There are a couple of options I might try to TZX what I've got - the first is to compress the WAV into a CSW file, though I might need DOSBox for that, or at the very least, the old laptop with 32-bit Windows 7 on it. If that works, I may be able to make a PZX file, but that doesn't run on too many emulators, and I see no way of converting it to TZX (there is no pzx2tzx.exe in the PZX Tools I have), only the other way round.

I might also try feeding the WAV directly into MakeTZX or Taper and see what they can do.

Next up is issue 5, which is on the database... in retrospect I may have been wrong about it containing SAM Coupé data, though the impression I get from the first four issues was that the Enigma team were desperate for its release; it had just seen the light of day by the time of issue 5, hence the opening message when it starts loading, but I didn't look into it properly, which I will this time. I'll compare the TZX on the archive to Louise's tape rip and see if they match.

[TMD2003]

Curse, drat, dagnabbit, semprini, Belgium, Commodore and lots of other naughty words.

Taper's sampler was a dead loss - it has to create an internal VOC file, and it can't possibly due to running on DOSBox. It crashed with no hint of how to avoid the FATAL ERROR.

It's a good job I downloaded the entire WOS mid-2017 archive when I did, because it means I still have access to the CSW tools. However, the files it produces are not guaranteed to work. The program runs OK on the old laptop, which is good - but making a CSW of e4final.wav causes a crash as soon as it tries to load the turbo-loading blocks. By the looks of things, MakeTZX converts a WAV file to CSW internally before converting to TZX, and it's completely bamboozled by these turbo-loading blocks with their 150%pitched pilot tone. I've tried the Direct setting with MakeTZX and all the turbo blocks are being registered with a speed of around 70% of normal, when it should be more like twice that.

Glass is the only file I know of that has anything similar. Once the custom loader starts, it alternates between blocks of high speed (around 150%) and regular speed, but all the blocks have high speed (150%) pilot tones. Someone, somewhere, managed to make a TZX of this game, as far back as 1998, in what I assume to be the early days of TZX, given the dreadful problems I had round about that time trying to get the format to work with X128 (which it didn't, even though the documentation said it did).

Plus, it had no idea where the end of the file was, and never produced a finished TZX that would have been borked anyway.

What can possibly be going wrong? It's at times like this that I am reminded that I am not a true Spectrum geek with the highly specialised arcane knowledge required for a task like this, just some regular Joe who writes CSSCGC entries, mainly in BASIC, and has absolutely no understanding whatsoever of the fine minutiae of how emulated tape files work, let alone the C/C++ code that made the utilities in the first place.

All I can do, therefore, is make compressed WAVs that work. I can do no more than that. It's over to the proper experts for the rest.

What MakeTZX has revealed, at least, is the filenames of the blocks that didn't appear on screen, so I've updated the timestamp files. And so, I need to update all the downloads...

ISSUE 1 e1final.rar: https://mega.nz/file/dPYSFBCB#BmFHuHuZd ... L_7-a9oQ50
ISSUE 2 e2final.rar: https://mega.nz/file/4P41nTYK#7balPgHAn ... slua__Ztvs
ISSUE 3 e3final.rar: https://mega.nz/file/BPQjCY5Y#DiLmWA7Ew ... OVwvKeEG8E
ISSUE 4 e4final.rar: https://mega.nz/file/ZPBm3RxS#BgeqId8oq ... J1RZ9WJupU

[^m00h^]

Hi TMD2003,

can you upload ETM5 to mega.nz please.

[TMD2003]

That's today's project. It'll take longer as I'm comparing with the version Dr. Titus has already made, just to confirm it's all intact. Updates will come soon enough, and I should be able to get #6 done as well if there are no unforeseen hurdles.

[^m00h^]

That's great! Also ETM6 from Dr.Titus does not have one block with an article, the first article aicr, and Dr. Titus took it from ETM5.

[TMD2003]

Issue 5 has been something of a slog, mainly because I was right initially - there are SAM Coupé blocks on this tape.

It looks like Dr. Titus' version of this tape is mostly fine, although it lacks the STOP THE TAPE prompts that will cause emulators to stop the tape (well, duh) where necessary. These can be put in using Taper easily enough. Also, there are three files - one is the tape magazine itself, the other two are the supplementary programs at the end of the tape.

Everything works on the WAV I've processed, on the Spectrum at least - except for the Bytes: block at the end, which is more uncompiled code intended for use with the assembler in issue 1. When loading raw data (as this is), this assembler requires capitals in the header and whoever saved the data to tape (Hamish?) was half-asleep and used lower-case characters, so it won't load and will have to be manually loaded in and resaved with a different capital header to test it. As you'd expect for my diligence and general devotion to duty when it comes to Spectrum projects, this is what I did (and needed to dig out a test tape from the bowels of the Sinclair archives). The code loaded, depacked, and... didn't do anything upon execution. Bah!

So what we need to do now, is make this tape work with a SAM Coupé as well as a Spectrum.

I've consulted a range of SAM emulators to see if I can crack what's going on. I have access to SimCoupe, ASCD, ZEsarUX and Speccy - these are the only four I know of, and they all have variable percentages of the full SAM experience.

From the SAM Coupé section in this issue, because Jon and Hamish were really keen on the don't-call-it-a-Super-Spectrum:
"Remember, it's best to use a fairly high volume on your tape deck to get them to load. Black and white stripey (sic) borders while loading usually mean that it's not working."
This is about to become relevant.

The first block on the tape is the BASIC loader, which displays "Enigma" rather than "Program:" because the title is CHR$ 22+CHR$ 0+CHR$0+"Enigma"+CHR$ 6. Clever. That's probably how Alcatraz did it. The program runs, and then tells us it will skip over the next three files "because they're Coupé ones". We then get:
Turbo loading data block (83 bytes)
Turbo loading data block (2389 bytes)
Bytes: coupe_c CODE 42842,75
Turbo loading data block (77 bytes)

These all have the same higher pitched pilot tone and loading signal (about 133% of normal speed, not 150% as I've said before) as I've been used to on these Enigma tapes so far. The Spectrum does not even start to recognise these blocks, not even the pilot tone - it just flashes the border red and cyan. This is even though "Bytes: coupe_c" has a regular-length Spectrum header, rather than the longer header of the SAM Coupé.

And here is where we run into trouble, because a SAM Coupé header should be 80 bytes long - I've even consulted the SAM Technical Manual to confirm it. The 83-byte block is its header, and its length has been incorrectly determined during the TZX conversion. Dr. Titus may not have realised that this was SAM Coupé code and didn't know it was wrong.

Either way, I tried this tape in SimCoupe, which has a useful tape browser but annoyingly can't display it and run the emulator at the same time - it must be closed first. The initial Spectrum block is unrecognised and the loading lines are black and white. The 83-byte header is recognised as "Basic: Enigma", but I saw the briefest flash of black and white lines just before the main data loaded. I told you that would be relevant. There is then a "Stop the tape" prompt, and a rudimentary menu where we can save to disc or not. So I select "don't save", the tape starts again, and... black and white lines. Blocks will "load", but nothing ever executes.

SimCoupe also accepts CSW input, so I took the WAV file I've cleaned up here, and attempted to convert it to CSW. It still defines the SAM header as 83 bytes, and completely fails to load in any way. It just keeps on trying to load the SAM-only blocks over and over again without ever getting there.

ZEsarUX looks like it's having a bit more success in that it finds and displays the headers up to "Bytes: piccy", but said "piccy" is never displayed as it is on the Spectrum, and some shash appears in the bottom two lines. What this emulator doesn't have is the loading stripes, more like a border that alternates between the two colours. At least those two colours are blue and yellow rather than black and white, but it still fails at "block4", by which time the screen is completely covered with shash and changed to a rather MSX-esque palette. ALSO: ordinarily, I find the popup screens annoying whenever the emulator changes screen resolution, but this time it's useful, as it does at least say it's moved into Spectrum-compatible mode when "Bytes: piccy" has loaded. This does confirm that Jon and Hamish were careful to make sure that change happened, and the tape image isn't loading correctly, rather than it being an error on their part.

Speccy (and I'm using the latest version, 5.9) gives the same result as ZEsarUX. All the shash is in the same place on the screen, in the same colours.

ASCD can only flashload tapes, so I can't keep track of what's working and what's not. All I know is, it doesn't like the turbo loading blocks (because it gives an error in a Windows box) and goes on to produce the same result as the other two emulators.

I can at least conclude from this that any of the emulators that isn't SimCoupe (usually the default choice) would be the one to use to test whether or not any TZX output is SAM-compatible. I'm opt for Speccy 5.9 as it's the newest out there.

I wanted to test what the output from DEVICE t35 would sound like - but none of the SAM emulators will make a new TZX file. The best I can do on this front is with ASCD, where the command SAVE "file" will create a new .TAP file in the last accessed directory, always called TAPE.TAP (and presumably overwritten every time - beware!) Unfortunately, these are exactly the same format as Spectrum .TAP files and don't record anything other than standard-speed data. Hence, whether I use DEVICE t or DEVICE t35, the data on the .TAP file is identical. However, I suspect it was unlikely that it was originally produced using DEVICE t35; it was probably from the same Spectrum used to make these magazines in the first place, with the header built up from scratch and saved using the same tape-speed-changing routine they'd used for the first four issues.

The burning question of the day is: can the SAM code be fixed, can it be accurately TZXed, and will any emulator run it properly anyway?

STOP PRESS: I just had an idea. Is it worth being a complete and total autist, in that examining the waveform of the "83"-byte block should confirm whether it is actually 83 bytes or a badly-encoded 80, and also it should reveal the number of bytes in the next block. The SAM Technical Manual spells out the structure of the header with the detail that should allow this to be possible...

All this may very well be out of Dr. Titus' hands. SAM Coupé emulation never really left the Stone Age, relative to the Spectrum. We just have to work with what we've got. And also, these are Spectrum tapes that the SAM only emulated in its own special way, but given that Jon and Hamish were so enthusiastic about its release, you'd think they'd have tested this tape extensively to ensure that it worked perfectly on a SAM just as much as it would on a 128K Spectrum.

As requested by @^m00h^, here's the file I've just been working on.

ISSUE 5 e5final.rar: https://mega.nz/file/8KgSDKjD#KDxSJ8An6 ... JdDE9gr6RI

[flatduckrecords]

Playing the WAV into my SAM seems to work, here’s how “piccy” is meant to look:



My guess for the 83-byte header is the TZX tool adding a length (2 bytes) and a parity/checksum byte? Of the header proper the second 40 bytes can be ignored anyway since they are just a comment field, all the critical stuff is in the first 40. I think SAM also sends a byte before the block (0x01 for “header” and 0xFF for “data”) but I’m not sure if they are saved to tape (they are sent over the network, which shares the same header format).

I’ll report back if the BASIC loads and runs for me, and I’ll share the results of the “save to disk” option.

@TMD2003 would you like me to make a recording of a t35 program / data block? I could do a couple of speeds if you want to compare them.

Update: The save-to-disk thing churned through the tape saving each data block to disk, but I couldn't get the magazine to actually run. I'll try again tomorrow!

[TMD2003]

Playing the WAV into my SAM seems to work, here’s how “piccy” is meant to look:

That's certainly more than I've ever seen it do. Looks like SAM emulation falls short of what we need, here. Have you tried sending Dr. Titus' TZX through TZX2WAV to see how it compares? It should match.

My guess for the 83-byte header is the TZX tool adding a length (2 bytes) and a parity/checksum byte? Of the header proper the second 40 bytes can be ignored anyway since they are just a comment field, all the critical stuff is in the first 40. I think SAM also sends a byte before the block (0x01 for “header” and 0xFF for “data”) but I’m not sure if they are saved to tape (they are sent over the network, which shares the same header format).
I’ll report back if the BASIC loads and runs for me, and I’ll share the results of the “save to disk” option.

See my new thread in the Really Anally Technical Section, just to see if I can get the answer by brute force.

@TMD2003 would you like me to make a recording of a t35 program / data block? I could do a couple of speeds if you want to compare them.

For future reference, that would be incredibly useful. At least that way I'll have some kind of audible comparison.

Tell you what - do it with Mines of Magrathea - it's just a .TAP with one BASIC loading block, about 21K of it. I'd be interested to know if any of the SAM emulators will be able to load it if it's saved with DEVICE t35; I'll probably have to use ZEsarUX as it's the only one I've seen that'll accept WAV input (unless the CSW conversion works).

You realise you've just volunteered to be the official SAM Coupé tester for these tapes, right? There's nothing like proper hardware for a rigorous test, and that's something I can't do, not least because I don't have 500 quid to throw at a potentially dodgy eBay seller when one finally turns up...

[TMD2003]

That's great! Also ETM6 from Dr.Titus does not have one block with an article, the first article aicr, and Dr. Titus took it from ETM5.

I have now investigated Louise's rip of issue 6 to see what the truth is.

Dr. Titus' TZX efforts once again consist of three files - the magazine, and two TZXs of the supplementary programs - which, after a bit of investigation, are two parts of the same program that requires a +D and a disc in the drive to work. The program is Steve Nutting's +D Filer, which is listed as MIA, but it may be that it was never released "in the wild" and Enigma was the only place it was available. I've seen other programs listed as MIA that are available only via magazines, even if it was something that everyone knew about like Crash or YS, rather that one of these obscure tapezines with a small circulation. Either way, I made a blank +D image, reloaded the Filer program and it does at least execute, copying itself onto the disc ready for use.

What I have further discovered, as I've been using MakeTZX solely to discover the filenames of the headers that don't display on screen, is that MakeTZX does not recognise any normal speed blocks after it's picked up on a turbo block. This would explain why Dr. Titus cut the Filer program into two TZXs - the original WAV had to be cut that way, it's just that he didn't put it all back together again. However, the magazine also needed to be cut the same way so that "Bytes: loader_c" could also be found.

So, onto the magazine section with the missing block: Dr. Titus' version is titled "ETM-5-6.TZX", which raised my suspicions further that all is not as it should be with it. And sure enough, there is a discrepancy: "block1" is 21,518 bytes and it should be 19,346 bytes. The other blocks all match the temporary TZX I made of the WAV. On loading "block1", it is indeed a copy of the section of the magazine from issue 5, where the editor tries to write with a "dreadfull (sic) hangover" rather than celebrating Enigma's first birthday.

Dr. Titus left a note in the TZX file that says "Note: Engine from ETM-5", which implies that the entire section from the initial "Program:" to "block1" had to be replaced. If that's true, then other than "block1", nobody would ever have noticed. The loading screen from the WAV is the same, the "save to disc" menu is the same, the text in the headers is the same, and there's the same background music along the course of the magazine.

The only other difference I can find is, rather than re-displaying the loading screen when a new block is loaded, Dr. Titus' version just displays a message "ENIGMA ©91 E.S.D." - which further indicates that this might have been from some kind of re-release (i.e. was this one included as a Sinclair User bonus covertape?) The dot-matrix-printed information sheet gives the release date for this issue as April-May 1990, not 1991.

And something I should have mentioned last time: Scroll down too far from the first introduction page (with the explanation of the controls) on both issues 5 and 6, and it will crash. This is in "block1"; Dr. Titus' version of issue 5, and the WAVs of issues 5 and 6, all do this.

Either way, now that the proper version of issue 6 has been recovered, completely intact and all working (as ever, I've checked the cleaned-up WAV with Fuse), it's worth redoing this one.

ISSUE 6 e6final.rar: https://mega.nz/file/MeQ3VSCI#_R3bYBZp5 ... 7a6EfunFYg

@flatduckrecords - does this one work on your SAM Coupé?

[flatduckrecords]

@flatduckrecords - does this one work on your SAM Coupé?

No, it seems to have the same problem as issue #5 - of the "three SAM files" mentioned in the Spectrum loading screen there seems to be only two on the tape (I think the missing one is CHARSET). The BASIC loader doesn't load the files by name though — it just does LOAD "" CODE — so it's pulling in the two "SAM" data files that are there, as well as the (Spectrum compatible) "piccy" file - but of course interpreting it as executable CODE rather than a SCREEN$.

Reading through the code there IS reference to "coupe_c" (which does exist), followed by "charset" and "code" which don't appear on the tape. There is a code file on tape called "loader_c" that's not referred to by name in the program, so I tried loading that instead of the file named "code" (and skipping the charset); but that didn't work. I've got some more time this evening so I'll try again.

Here's a t35 of MoM:
mom_t35.wav
I think it's too quiet to load back in, but I thought it was good enough to hear what it's supposed to sound like.

It might just be a bizarrely improbable concidence, but were you aware the MGT version Mines of Magrathea occupies 42 disk sectors? That's dedication to the bit, Jim. 🫡

[TMD2003]

No, it seems to have the same problem as issue #5 - of the "three SAM files" mentioned in the Spectrum loading screen there seems to be only two on the tape (I think the missing one is CHARSET). The BASIC loader doesn't load the files by name though — it just does LOAD "" CODE — so it's pulling in the two "SAM" data files that are there, as well as the (Spectrum compatible) "piccy" file - but of course interpreting it as executable CODE rather than a SCREEN$.

Hmmm. I'd assumed that the three blocks the Spectrum was supposed to skip were:
(1) The 83 (80?) byte SAM header that says Basic: Enigma
(2) The body of the SAM Basic
(3) Bytes: coupe_c, regular ZX header and data

Apparently not.

Reading through the code there IS reference to "coupe_c" (which does exist), followed by "charset" and "code" which don't appear on the tape. There is a code file on tape called "loader_c" that's not referred to by name in the program, so I tried loading that instead of the file named "code" (and skipping the charset); but that didn't work. I've got some more time this evening so I'll try again.

"Bytes: loader_c" is a routine that sets up the Spectrum to receive the high-speed pilot tones and data. Something similar occurs in the supplementary programs - in the issue 1 assembler, for instance.

Enigma 5's information sheet says: "SAM Coupé owners type DEVICE T then hit F7 and follow on screen prompts".
Enigma 6's information sheet says: "SAM Coupé owners load the Spectrum emulator tape and follow on screen prompts".
So even though the two issues have the same initial loaders, it may have been that they forgot to include "charset" and "code" on Enigma 5, realised their error after the first few had been sent out, and then abandoned the SAM-only part of the loader and told SAM owners to treat it as a Spectrum-only program.

Here's a t35 of MoM:
I think it's too quiet to load back in, but I thought it was good enough to hear what it's supposed to sound like.

Interesting. Sounds like "t35" refers to 350% of regular loading speed rather than 135%. I wouldn't bank on anything saved with DEVICE t35 reloading with any reliability; if you're using metal tapes and a Nakamichi Dragon, then it might work, but an old-school shoebox-style tape recorder and a cheap "Type zero" tape? No chance. Powerload was "only" 200% and that had its problems.

At least this confirms the SAM blocks on these Enigma tapes weren't saved that way.

It might just be a bizarrely improbable concidence, but were you aware the MGT version Mines of Magrathea occupies 42 disk sectors? That's dedication to the bit, Jim.

Yesterday I made a kebab out of minced lamb I'd had in the freezer since November, swatted a moth that had been eating my carpet, and wore my new hiking boots, which I'd made sure were leather so they didn't leak. The day before, I'd accidentally dropped a snail that was hanging onto the underside of the plastic storage shed in my backyard. And, while digging the triffid away that was growing out of control out the front, my spade sliced through a worm. I'd intended to plant some lavender in there but I over-watered it last summer and it died. It turns out they were all incarnations of the same entity, who is now screaming "COINCIDENCE? WHAT HAVE YOU GOT AGAINST ME, YOU MULTIPLE-ME-MURDERER?" in a dark cave somewhere, probably just outside Hemel Hempstead. God was unavailable for comment, having just disappeared in a puff of logic.

Now, this, on the other hand...

[TMD2003]

Right! This will be my last contribution for today. I've got a cake with a lot of candles to blow out ("not as many as the rest of us, grumble snarl froth" - 95% of this forum), and I'm going to spend the evening doing that, and throwing large quantities of grilled flesh and fermented barley juice down my throat courtesy of a newly-opened local Turkish takeaway and some local-ish breweries. Try and stop me, I dare you.

Anyway - Louise's rip of Enigma #7 is worth preserving, because it's a variant on the one that is already present on the database. Maybe, again, it was a case of the WOS preservation people from T'Olden Days being ultra anal, with copies of (for instance) the original Ocean release of Batman, and the Hit Squad re-release, both being preserved and hosted. Sure, the Hit Squad release was in a smaller case, but as far as I can see, the tapes were identical.

Well... these are not. I could tell as soon as "Program: ETM7" appeared on screen. It's "Program: Enigma 7" on the existing version.

Again, there are SAM Coupé blocks at the start. The existing version has a header of 81 bytes; I temporarily TZXed this WAV and the equivalent file is 83 bytes, which is at least consistent with issues 5 and 6. Most significantly, the "Bytes: charset" is present in the SAM blocks, whereas it isn't in the existing version. "Bytes: code" still isn't there, although maybe that's what "Bytes: coupe_c" should be doing.

The loading screen that has been uploaded here isn't the actual loading screen that reappears between loading the magazine blocks, it's the credits screen. Issues 5 and 6 used the credits screen as the loading screen, whereas issues 1-4 all had a proper loading screen and issue 7 has gone back to that format. This can be rectified, presumably.

A few more blocks in the magazine have header text in capitals. Bytes: COUPE_C and all the numbered BLOCK* (1-9) blocks are capitalised in the existing version, and they're lower case on this tape rip.

The existing version is "side 1" for the magazine and "side 2" for everything else. Louise's rip is one long file.

This rip had the +D Hacker program next, again with different capitalisation, then the 48K version of Jekyll & Hyde, then a SAM Coupé program, again with a header that MakeTZX judges to be 83 bytes.

The existing version has 48K Jekyll & Hyde first on "side 2", with a few more capitalisation changes, then a 128K version with four more code blocks involved. Then it's +D Hacker, also with different capitalisation, and finally the SAM Coupé program, with a header length of 81 bytes.

So this one does need TZXing after all, and all it needs is someone who can do the job properly, and if the SAM header is the right length, so much the better. Maybe MakeTZX can't do that. Maybe a perfect SAM Coupé tape will always come out as 83 bytes this way. If there are experts on that front, make yourselves known.

ISSUE 7 e7final.rar: https://mega.nz/file/ML5mSIQB#spXqf8yqd ... wEXz48a0CI

Issue 8 is only available as a TAP so that'll play havoc with the turbo loaders. That is definitely going to need a wash and brush up. There's not a lot left for me to do, then - even though in the issue 7 it proudly declares "Enigma is going monthly!" That aged like milk left out in the Sonoran desert in the middle of July.

[flatduckrecords]

Right! This will be my last contribution for today. I've got a cake with a lot of candles to blow out ("not as many as the rest of us, grumble snarl froth" - 95% of this forum), and I'm going to spend the evening doing that, and throwing large quantities of grilled flesh and fermented barley juice down my throat courtesy of a newly-opened local Turkish takeaway and some local-ish breweries. Try and stop me, I dare you.

Felicitations on this auspicious occastion! I've two-score of the blighters to extinguish huff-and-puff-like next week. Enjoy your evening.

Again, there are SAM Coupé blocks at the start. The existing version has a header of 81 bytes; I temporarily TZXed this WAV and the equivalent file is 83 bytes, which is at least consistent with issues 5 and 6. Most significantly, the "Bytes: charset" is present in the SAM blocks, whereas it isn't in the existing version. "Bytes: code" still isn't there, although maybe that's what "Bytes: coupe_c" should be doing.

Well I've run into my first Coupé Tape Loading Errors with this edition, but overall it's fairing better:


ETM7, block 3 (I think)

Using the BASIC loader from ETM6 and the CODE and files from ETM7 it seems to be working better.

[TMD2003]

Felicitations on this auspicious occastion! I've two-score of the blighters to extinguish huff-and-puff-like next week. Enjoy your evening.

I did, in the presence of The Cruiser and his 2005 interpretation of The War Of The Worlds - also featuring Eowyn from LOTR as his ex-wife, an 11-year-old Dakota Fanning (with already five years' acting experience) as his daughter, Tim Robbins as a nutter with at least one thing in common with an Artilleryman, Steven Spielberg shouting "ACTION!", Morgan "You're Reading This In My Voice" Freeman narrating, and John "Star Wars" Williams writing the soundtrack. The WRONG soundtrack. Because you can keep your Imperial March, when we hear about intelligent life watching us from afar, waiting to make their move, I am expecting to hear... THIS! As is everyone else since 1978, and it's not right without it. It's not right!

Were you, by any chance, a very late addition to the Sinclair world? Anyone born in the 1980s who hangs around here is usually Russian (e.g. Denis Grachev, born 1983, as this forum has told me).

Well I've run into my first Coupé Tape Loading Errors with this edition, but overall it's fairing better:
Using the BASIC loader from ETM6 and the CODE and files from ETM7 it seems to be working better.

As might have been implied by what I wrote, I wasn't expecting it to work because it's still missing one block, unless the SAM BASIC has changed in this version and is actually expecting to LOAD "coupe_c" CODE rather than LOAD "code" CODE. (That will be very confusing for anyone who hasn't been following the developments on this thread.)

Meanwhile, I will crack on with issue 8 - and issue 9, because there's every chance that it'll be a variant on what's already archived.

[flatduckrecords]

Because you can keep your Imperial March, when we hear about intelligent life watching us from afar, waiting to make their move, I am expecting to hear... THIS! As is everyone else since 1978, and it's not right without it. It's not right!

Were you, by any chance, a very late addition to the Sinclair world? Anyone born in the 1980s who hangs around here is usually Russian (e.g. Denis Grachev, born 1983, as this forum has told me).

That is, as we Millennials^* are wont to say, a legit banger. Which is to say, yup! I mentioned elsewhere I think but I got my first Spectrum in ‘89 or ‘90. Growing up this would have been more my jam, but I soon discovered music with guitars in so it’s okay.

As might have been implied by what I wrote, I wasn't expecting it to work because it's still missing one block, unless the SAM BASIC has changed in this version and is actually expecting to LOAD "coupe_c" CODE rather than LOAD "code" CODE. (That will be very confusing for anyone who hasn't been following the developments on this thread.)

No, unfortunately the BASIC block itself seems to be corrupted. Most of it loads but there’s some garbled bytes towards the end. I’ve not compared 5/6/7 side by side but it seems pretty much the same at first glance.

I disassembled coupe_c (it’s only about 100 bytes) and its job seems to be setting up SAM’s memory and screen registers. It’s a bit confusing in isolation (a couple of JumP instructions didn’t make sense to me). Tape 7 does include both “coupe_c” and “code”. I think “code” must be the music+driver, or maybe the magazine engine itself, I’ve not looked closer yet.

Unfortunately loading content blocks from earlier issues just seems to crash, but issue 7 blocks seem to be mostly working. There’s no music on SAM (it has a different sound chip of course) but e.g. the starfield screen effect works the same way as Spectrum.

Oh and the missing charset file - turns out it’s just a copy of the Spectrum ROM font - so overall it’s pretty indistinguishable. Probably easier just to load the Spectrum version in the SAM Spectrum emulator (as per the instructions you mentioned) anyway!

On the rising edge of the Millennial-window so to speak, but a millennial nonetheless.

[TMD2003]

I've also got some Enigma tape magazines if extra samples will help out?
I've got issues 1, 3, 5, 7, 8 ans 9 plus a disassembler cassette.

ACHTUNG! A rip of your issues 8 and 9 will be required.

I just read Louise's notes before trying to clean up her version of issue 8 and the dropout in one of the magazine blocks is, unfortunately, unrepairable - there's a clear gap in the signal. It's followed by a section where the amplitude veers wildly off-centre but the data is still there, which might not have been terminal.

The version of issue 8 on the archive, that's a TAP file, only has four blocks of the magazine in it anyway. It was ripped some time in 1996, and doesn't have a "normal speed" version of the ruined block 7.

Meanwhile, Louise further notes on issue 9 that it has a very short block 3 which can't possibly contain everything that the info sheet says it does. The copy on the archive is also incomplete; it has only four magazine blocks and nothing else, and was ripped in the early TZX days of 1998. Was the version of this magazine hosted by Sinclair User just a shareware-style sampler that was designed to encourage SU readers to subscribe to Enigma?

EDIT: something is certainly amiss with this issue. The very short block 3 is stated to be SAM-only - there's one page in block 2 that does this, but block 3 consists solely of this message (and crashes on any attempt to scroll right through it). @Gilby - let's see what your Issue 9 looks like, and if it has the same flaws.

[LouEnPointe]

Apologies at the consternation with some of the SAM loading errors. I had no way of checking those. The drop out in issue 8 is unrecoverable by azimuth tweaking (I tried for about 30 mins) but I was so conscious of the tapes maybe going wrong given their age and I wanted to get as much of them as possible given their missing status. It took a long time to retrieve some of these issues given the noise on the tapes themselves.
They're mostly there, there's a possible alternative source to fill in the blanks (yay), and you all have something to chew over now.
Thanks for your work TMD2003 - I appreciate you picking up the wavs and running with them (my technical knowledge stops after a point).
Thanks,
Louise
PS Yes there were stops and start in the wavs. I had to tweak azimuth as I was going along and I had to leave gaps so I could go back and redo the block if needed. In some cases the gap between blocks was under 5 seconds which made things so awkward given the tape noise as well.

[flatduckrecords]

Thanks for all your hard work @LouEnPointe!