Page 2 of 2

Re: Miracle Systems FLP size determining algorithm known?

Posted: Sun Oct 20, 2019 12:59 pm
by stephen_usher
Well, something is definitely wrong with the detection somehow. I installed the logging feature of FF and got this when I tried to format on the Sandy Q-Board:

Code: Select all

Write 8[09]/9... 6039 us
IMG Bad Sector Offset: 4294961184 -> 9
Write 0[01]/9... 3512 us
Write 1[02]/9... 5982 us
Write 2[03]/9... 7468 us
Write 3[04]/9... 7479 us
Write 4[05]/9... 7630 us
Write 5[06]/9... 7363 us
Write 6[07]/9... 6424 us
Write 7[08]/9... 7478 us
Write 8[09]/9... 8282 us
Well, I won't get any further until the developer returns from his week away next week. (Half-term, maybe?)

Re: Miracle Systems FLP size determining algorithm known?

Posted: Mon Oct 21, 2019 6:37 pm
by stephen_usher
Further testing of ED disk images shows that they're very unreliable. It seems that the Gotek can't keep up with the data rate, at least during formatting. (QL ED disks have a higher data density/rate than the PC ED format, possibly pushing the technology to edge.)

I've been asked by the Flashfloppy maintainer to test file access using a pre-prepared image. Unfortunately I don't have an ED drive or disk. Does anyone know of a QL emulator which supports ED floppy images so that I can generate an image?

P.S. Having looked, there seems to be no emulators which support the ED images, so I'm going to have to abandon the attempt at getting the Gotek+Flashfloppy supporting them.

Re: Miracle Systems FLP size determining algorithm known?

Posted: Tue Oct 22, 2019 2:08 am
by NormanDunbar
You can create a blank ED image with qltools - https://github.com/NormanDunbar/qltools/releases

HTH

Cheers,
Norm.

Re: Miracle Systems FLP size determining algorithm known?

Posted: Tue Oct 22, 2019 8:23 am
by stephen_usher
NormanDunbar wrote:You can create a blank ED image with qltools - https://github.com/NormanDunbar/qltools/releases

HTH

Cheers,
Norm.
Excellent! I'll take a look.

Re: Miracle Systems FLP size determining algorithm known?

Posted: Tue Oct 22, 2019 9:36 am
by Derek_Stewart
stephen_usher wrote:Further testing of ED disk images shows that they're very unreliable. It seems that the Gotek can't keep up with the data rate, at least during formatting. (QL ED disks have a higher data density/rate than the PC ED format, possibly pushing the technology to edge.)

I've been asked by the Flashfloppy maintainer to test file access using a pre-prepared image. Unfortunately I don't have an ED drive or disk. Does anyone know of a QL emulator which supports ED floppy images so that I can generate an image?

P.S. Having looked, there seems to be no emulators which support the ED images, so I'm going to have to abandon the attempt at getting the Gotek+Flashfloppy supporting them.
Hi,

Does the Gotek support ED Disk formats, my Gotek drives only go up to HD format.

Can you tell me the model of the you Gotek drive that can read/write to ED Disk formats.

Re: Miracle Systems FLP size determining algorithm known?

Posted: Tue Oct 22, 2019 6:24 pm
by stephen_usher
It can potentially. The Flashfloppy firmware merely needs a disk description in the IMG.CFG file. Apparently it can handle the PC ED format, but that has a smaller sector size, giving more time to process the data due the gaps.

Re: Miracle Systems FLP size determining algorithm known?

Posted: Tue Oct 22, 2019 7:10 pm
by stephen_usher
Right, I formatted an ED image using qltools and copied it onto the USB drive, created an IMG.CFG file with containing the QL disk geometries and loaded it into the Gotek.

I can write and read files from the images without issue, so it looks like the formatting process just tips over the edge of what Flashfloppy is capable of in terms of data rate.

Re: Miracle Systems FLP size determining algorithm known?

Posted: Thu Oct 31, 2019 11:34 am
by NormanDunbar
NormanDunbar wrote:You can create a blank ED image with qltools - https://github.com/NormanDunbar/qltools/releases
Ok, I updated the SinclairQL repository for QLTools this morning and merged in all my latest changes. This repository is now at 2.15.5 levels. I've created a new release under the following URL:

https://github.com/SinclairQL/qltools/releases/latest

It's probably best to clone and release from this repo, rather than from mine.


Cheers,
Norm.