Archive of FSD files?

[scarybeasts]

hi Chris, thanks for indepth details of the must be protected *early* disc
(as BillC has not confirmed from pg2 )
your point 4 gave the game away, as mine is a flippy (40 on sideA, 80 sideB)
so hence does not need to check for a 40 or 80T drive

here is mine copied using UPURS, just checked & still copies easily with *COPY / BACKUP
so did remember rightly

To close this one out -- my "The Living Daylights" disc arrived. It's a 40T/80T flippy but it is protected like @billcarr's FSD. Track 2 is formatted as 18 sectors but the first 10 claim to be 256 bytes long; the sector IDs for those are in decrementing order; and the last 8 sectors are "hidden" with sector IDs 0 (and the other three header bytes zero).

Seems like another game where both protected and unprotected versions were sold.


Cheers
Chris

[vanekp]

That's a great offer -- but I'm working on a tool that should enable people to archive discs to HFE with good fidelity with only a beeb (no special hardware needed). Would you be willing to see if it is able to archive these two discs?

Few possible issues to get out of the way:
1) Currently the tool requires an 8271 equipped beeb. I intend to add 1770 support once I've purchased a 1770 board.
2) It currently only works on discs that were mastered track-at-a-time and not written since. This does seem to apply to many / most(?) commercial discs -- including my original model B Acornsoft Elite disc.
3) You need a way to get the data off the beeb. Currently it writes data to the "disc" in drive 1 and for me that's a Gotek with a DSD file mapped in.

Do you happen to have an 8271 equipped beeb to try this with?


Cheers
Chris

I did something similar on the weekend using a normal 5.25 drive and a gotek and used Enigma DiscImager to make copies of my BBC disc's to hfe's on the gotek seem to be much better working images than via Kryoflux, am waiting for the bits to come in to try with Greaseweazle.

[scarybeasts]

I did something similar on the weekend using a normal 5.25 drive and a gotek and used Enigma DiscImager to make copies of my BBC disc's to hfe's on the gotek seem to be much better working images than via Kryoflux, am waiting for the bits to come in to try with Greaseweazle.

That sounds fascinating. Would you be willing to describe the process in more detail?


Cheers
Chris

[vanekp]

I did something similar on the weekend using a normal 5.25 drive and a gotek and used Enigma DiscImager to make copies of my BBC disc's to hfe's on the gotek seem to be much better working images than via Kryoflux, am waiting for the bits to come in to try with Greaseweazle.

That sounds fascinating. Would you be willing to describe the process in more detail?


Cheers
Chris

sure was not that complicated, I 1st made a blank .hfe file on beebjit which I then copied to my USB stick to mount in the gotek.
connected my normal floppy drive as drive 0 and the gotek as drive 1 I guess it does not matter which way around you connect them.
and using the EnigmaDiscImager-109.rom with the backup command *DBACK <Sdr><Ddr>(No Trks)(0/1)
so if the source is 0 (floppy) and destination drive 1 (gotek) and 80 tracks then
*DBACK 0 1 80
and Enigma was able to make working copy's of things like doctor who and Vector2 without a problem, even vector2 could not copy doctor who (sends the heads up the spindle).

Greaseweazle I am not having much luck with it reading disks, writing a hfe to a BBC disc works fine but reading to hfe or spc is not not giving me 1:1 working copy's of the disc's, its not total junk but i get clock errors on the images more here .

[scarybeasts]

sure was not that complicated, I 1st made a blank .hfe file on beebjit which I then copied to my USB stick to mount in the gotek.
connected my normal floppy drive as drive 0 and the gotek as drive 1 I guess it does not matter which way around you connect them.
and using the EnigmaDiscImager-109.rom with the backup command *DBACK <Sdr><Ddr>(No Trks)(0/1)
so if the source is 0 (floppy) and destination drive 1 (gotek) and 80 tracks then
*DBACK 0 1 80
and Enigma was able to make working copy's of things like doctor who and Vector2 without a problem, even vector2 could not copy doctor who (sends the heads up the spindle).

Thanks, I'll give that a try!

Greaseweazle I am not having much luck with it reading disks, writing a hfe to a BBC disc works fine but reading to hfe or spc is not not giving me 1:1 working copy's of the disc's, its not total junk but i get clock errors on the images more here .

That's interesting. Given that both the KryoFlux and Greaseweazle are having trouble, perhaps it's something with the drive? The drive sounded like an older Mitsubishi. I have an older Chinon drive and had challenges with some of its electrical properties when wiring it up in.... creative ways. In the end I got it reliable by yanking out the bus termination resistor array. I think it was 150ohm and that was just too low a resistance when driving the drive with something other than the disc drive port. If you wanted to start with something less drastic, maybe try different cables -- do you have a shorter one?


Cheers
Chris

[vanekp]

Greaseweazle I am not having much luck with it reading disks, writing a hfe to a BBC disc works fine but reading to hfe or spc is not not giving me 1:1 working copy's of the disc's, its not total junk but i get clock errors on the images more here .

That's interesting. Given that both the KryoFlux and Greaseweazle are having trouble, perhaps it's something with the drive? The drive sounded like an older Mitsubishi. I have an older Chinon drive and had challenges with some of its electrical properties when wiring it up in.... creative ways. In the end I got it reliable by yanking out the bus termination resistor array. I think it was 150ohm and that was just too low a resistance when driving the drive with something other than the disc drive port. If you wanted to start with something less drastic, maybe try different cables -- do you have a shorter one?


Cheers
Chris

It could be the drive though when I use it on the BBC it seems to work fine, reading and writing. Mine does not have the terminator in and was wondering if that was a problem but you say you have had problems with it in, also tried putting 1k resistors in there, no difference.
My Greaseweazle is the F1: Based on the STM32F103 MCU and the "Blue Pill" development board, and is directly connected to the new drive cable I made, I did wonder if it needs the 1.5K pull up resistors as I have no interface board just directly wired, and did wonder about the cable length as I put a 1m cable on it, its not mentioned anywhere that it needs to be as short as possible but I guess I could try that.
Otherwise it could be the drive itself as you mentioned. Its a Mitsubishi M4853-112M, I have no other BBC drive to try to eliminate that as the problem.

[scarybeasts]

Greaseweazle I am not having much luck with it reading disks, writing a hfe to a BBC disc works fine but reading to hfe or spc is not not giving me 1:1 working copy's of the disc's, its not total junk but i get clock errors on the images more here .

That's interesting. Given that both the KryoFlux and Greaseweazle are having trouble, perhaps it's something with the drive? The drive sounded like an older Mitsubishi. I have an older Chinon drive and had challenges with some of its electrical properties when wiring it up in.... creative ways. In the end I got it reliable by yanking out the bus termination resistor array. I think it was 150ohm and that was just too low a resistance when driving the drive with something other than the disc drive port. If you wanted to start with something less drastic, maybe try different cables -- do you have a shorter one?


Cheers
Chris

It could be the drive though when I use it on the BBC it seems to work fine, reading and writing. Mine does not have the terminator in and was wondering if that was a problem but you say you have had problems with it in,

To be fair, I've also had plenty of problems with the resistor array out

A 1m cable with no terminator array might be pushing it. The thing most likely to fail is the high frequency signals such as the read pin and the write pin.
See: https://retrocmp.de/fdd/general/resistor.htm


Cheers
Chris

[vanekp]

Shortened the cable to 10cm still the same result, terminated both sided with 1K resistors with much the same results.
Does not leave a lot over but the drive itself, even though it works fine on the BBC but not on both Kryoflux and Greaseweazle

[scarybeasts]

Shortened the cable to 10cm still the same result, terminated both sided with 1K resistors with much the same results.
Does not leave a lot over but the drive itself, even though it works fine on the BBC but not on both Kryoflux and Greaseweazle

In case you had a scope, it'd be interesting to capture the read pin output at the greaseweazle end of things:
- Are the high/low voltages clearly distinguished within the TTL acceptable ranges?
- Are the edges of the read pulses nice and square?
- Are the read pulse widths within spec?

I also have an in-development BBC disc program. It can tell you the rotation speed of the drive in case you're curious


Cheers
Chris

[vanekp]

I have a rather old 35Mhz scope this is at 1v per dev 5us

not sure how its suppose to look but its at-least 5v pp

and yes it may be an idea to check the rotation speed of the drive, so your program will come in handy.

[scarybeasts]

I have a rather old 35Mhz scope this is at 1v per dev 5us

Wow! Looks like it belongs in a museum

While browsing around, I did find these greaseweazles for sale: https://www.sellmyretro.com/offer/detai ... b-_b-43881

One of the lines in the description of changes (from the original board) caught my eye:
"Added stronger buffers for driving 5.25" drives with higher current pull-up resistors."

So it does sound like there are some 5.25" drives out there with more challenging electronic setups...


Cheers
Chris

[danielj]

Yup, it's a more robust design. Some people were having issues with older 5.25" drives (touch wood I've not had any). It's mentioned in the Facebook group. It also has proper esd protection on the USB port and can be powered with 12VDC.

I am halfway through putting a talk/presentation together on this stuff for tomorrow's abug. God help us all...

[scarybeasts]

and yes it may be an idea to check the rotation speed of the drive, so your program will come in handy.

Try the attached SSD, a very early prototype of some disc software I'm working on.
Use INIT0 or INIT1 depending if you want to check drive 0 or 1. It should auto-detect 8271 vs. 1770 and then hit the chip directly, reporting the chip type and the measured drive speed. You can then run TIME a few times to see if the speed is consistent (there is some minor jitter with the 8271 because querying the chip has a high latency).

In beebjit, the speed is a perfect 3125 (3125x 64us for 1 rev) as you might expect in an emulator.

On my model B, the Gotek gives 3121. This is a tiny bit fast, probably just the Gotek has a more accurate crystal than the old beeb?
My Mitsubishi / TEAC drives are 3115 / 3116, so also a tiny bit fast. My older Chinon is 3133, so a tiny bit slow.


Cheers
Chris

[vanekp]

ran it a few times and it comes back with 3171 or 3167 each time and 3123 on my gotek

[scarybeasts]

ran it a few times and it comes back with 3171 or 3167 each time and 3123 on my gotek

Interesting. That's about 1.5% slow. The standard rotation speed rating is often +- 1.5% so within bounds. Some older drives have a variable resistor on the bottom surface that can be twiddled to change speed a little. Pictures of the M4853 do show a variable resistor on the bottom but I have no idea what it does


Cheers
Chris

[vanekp]

Yes some drives you can adjust the motor speed, never looked under my drive to see if it has one.
its hear (left side)

Adjusted it and it comes back with 3123 or 3127 (8271 Disc controller chip)

[vanekp]

I gave Greaseweazle another go this evening after having adjusted the speed yesterday, this is what BoneCruncher looked like before viewtopic.php?f=32&t=19066&start=30#p270265
Now a copy looks like this (one bad spot but that's in a non data area)

The Disc is getting tired even started screeching at me quickly popped the disc out and gave the heads a clean, so think the disk is not in very good condition anymore, but the speed seems to make a lot of difference with Greaseweazle, I guess it was just out of tolerance.

[mad]

There are in excess of 1172 images now, but there aren't many "new" forms of protection on the later titles, just variations on a theme.

Thanks for taking an interest, I look forward to seeing the results. If you PM me, I can send you some / all as necessary.

Hi Bill,
I am unable to PM you requesting links to FSD files (PM feature is unavailable on my account).
Are the files availble to anyone who requests them, or are they just for the BBC elite?.

Thanks.

[scarybeasts]

Hi,

Are you actually after the FSD files, or will HFE conversions do? Arguably, HFE is more useful (e.g. you can put the on a Gotek).


Cheers
Chris

[mad]

Hi Chris,

I already have a collection of a few HFE files, maybe created by you?.

It's FSD files i am looking for, if you can help point me in the right direction.

thanks!

[vanekp]

Hi Chris,

I already have a collection of a few HFE files, maybe created by you?.

It's FSD files i am looking for, if you can help point me in the right direction.

thanks!

Chris has them sitting here, he did post the link somewhere on the forum for it.

https://drive.google.com/drive/folders/ ... zDoG3Bdyky

[scarybeasts]

It's FSD files i am looking for, if you can help point me in the right direction.

I don't have the FSDs online. In case it helps, here's a large number of them converted to HFE:
https://drive.google.com/drive/folders/ ... zDoG3Bdyky


Cheers
Chris

[mad]

Hi,

Thanks both for the HFE links.

While very much appreciated, I already have those.
It is the 'complete' original .FSD archive i would like to have.

[Pernod]

It is the 'complete' original .FSD archive i would like to have.

Why?

[billcarr2005]

Why?

Perhaps to easily read the data from any image using an "off the shelf" hex editor, since it's stored as easy to read bytes

I don't have a ready to use archive of FSD files, since very often I created them to quickly understand the layout of the disks and then create SSD from them.
I believe scarybeasts did a lot of working knocking them into shape when developing beebjit.
Of course, there's no easy way to write FSD back to original disks either
But if you want any specific titles, I'd be happy to email them to you, time permitting!

[mad]

Why?

Perhaps to easily read the data from any image using an "off the shelf" hex editor, since it's stored as easy to read bytes

exactly!

I don't have a ready to use archive of FSD files, since very often I created them to quickly understand the layout of the disks and then create SSD from them.

That would be even better in a few title cases and save me the hassel.

however i am still unable to send PM or contact you.
maybe you could PM me a contact email?

[scarybeasts]

Why?

Perhaps to easily read the data from any image using an "off the shelf" hex editor, since it's stored as easy to read bytes

exactly!

Ah yes, that's an important use case. It's not 100% the same, but I recently added an option to beebjit, -opt disc:dump-sector-data.
It can be used to go from less observable file formats (e.g. HFE) to a file of raw bytes. I intend to script across all my HFEs and grep for interesting strings -- fragments of source code or hidden messages perhaps.

I'm currently using the option to quickly look for text inside SCP files coming off my Greaseweazle.


Cheers
Chris

[mad]

Hi Chris,

your work on the HFE conversions and emulator tools are really appreciated.
I guess it will be the standard goto format for many old machine original floppy disks in time.

the more conversion tools available, the better.
keep up the good work!.