Cupl is pretty logical and simple so can't really do much wrong. There must be more to it than just swapping data, but I have no idea what. so as it stands, the feature will have to be abandoned :(PhilC wrote: 06 Aug 2019 17:00 Just a shot in the dark as I've not used CUPL yet but if it uses subroutines, is it possible you've missed a return or something?
Trudie - The Next IDE Board
-
exxos
- Site Admin

- Posts: 28354
- Joined: 16 Aug 2017 23:19
- Location: UK
Re: Trudie - The Next IDE Board
-
PhilC
- Moderator

- Posts: 7442
- Joined: 23 Mar 2018 20:22
Re: Trudie - The Next IDE Board
Nah, not abandoned. available in a future update. It's what I do with my PLC stuff. Get the basics working really well, then work on the problems.
Nothing wrong with having to work on features after you've got the board running. Maybe you need an extra pair if eyes on the problem? (Not volunteering as I haven't learned simple gal stuff yet)
Nothing wrong with having to work on features after you've got the board running. Maybe you need an extra pair if eyes on the problem? (Not volunteering as I haven't learned simple gal stuff yet)
If it ain't broke, test it to Destruction.
-
exxos
- Site Admin

- Posts: 28354
- Joined: 16 Aug 2017 23:19
- Location: UK
Re: Trudie - The Next IDE Board
A thought did pop in my head during the night.. I never checked if files would load even though then filenames were wrong..
I have a feeling that things might be right, but, the FAT tables shouldn't be byte swapped. That's OK in theory , it would explain a few things.. But how the heck would hardware work out where FAT tables are stored on the drive ?! :roll:
Edit:
So what I read so far is FAT is fixed at the start of the partition . I guess this theory also explains why HD10 is moaning about issues writing to the boot sector as that shouldn't be swapped either.
Assuming PCs can't use more than 1 partition on a CF card (maybe someone knows ?) Then it may be possible to turn off byte swapping during the first few sectors on then drive, or basically root and FAT sectors.
I think as the filenames on ST & PC both end up the same, I that it proves they are already both in the same format already, and not swapping them is the correct way. But the actual data needs to be swapped... and of course the command registers etc remain not-swapped also.
I have a feeling that things might be right, but, the FAT tables shouldn't be byte swapped. That's OK in theory , it would explain a few things.. But how the heck would hardware work out where FAT tables are stored on the drive ?! :roll:
Edit:
So what I read so far is FAT is fixed at the start of the partition . I guess this theory also explains why HD10 is moaning about issues writing to the boot sector as that shouldn't be swapped either.
Assuming PCs can't use more than 1 partition on a CF card (maybe someone knows ?) Then it may be possible to turn off byte swapping during the first few sectors on then drive, or basically root and FAT sectors.
I think as the filenames on ST & PC both end up the same, I that it proves they are already both in the same format already, and not swapping them is the correct way. But the actual data needs to be swapped... and of course the command registers etc remain not-swapped also.
-
mfro
- Posts: 124
- Joined: 13 Dec 2018 07:32
Re: Trudie - The Next IDE Board
A limitation that (Windows-) PC's only have until Win 7 for hot-swappable media. On earlier versions, there are alternatives/extensions available. Windows 10 removes this limitation as well.exxos wrote: 07 Aug 2019 09:20 Assuming PCs can't use more than 1 partition on a CF card (maybe someone knows ?)
So probably nothing you should rely on.
And remember: Beethoven wrote his first symphony in C.
-
exxos
- Site Admin

- Posts: 28354
- Joined: 16 Aug 2017 23:19
- Location: UK
Re: Trudie - The Next IDE Board
Good to know. Would probably just have to be a limitation on the ST side to 1 partition which is PC compatible. With multiple partitions, I doubt there would be a easy to work out where each partition starts.mfro wrote: 07 Aug 2019 11:12 A limitation that (Windows-) PC's only have until Win 7 for hot-swappable media. On earlier versions, there are alternatives/extensions available. Windows 10 removes this limitation as well.
So probably nothing you should rely on.
I tried to do 2 partitions on win7 and couldn't see any way to do it. Disk manager doesn't even list the options for partitions, pretty much only drive letter and format :shrug:
EDIT:
I guess if fixed partition sizes were used like 256MB each, then that could probably work as long as things remained consistent as the FAT locations would have to be hardwired into the PLD logic. Though that's a problem for another day anyway. Getting 1 working would be a huge success.
-
exxos
- Site Admin

- Posts: 28354
- Joined: 16 Aug 2017 23:19
- Location: UK
Re: Trudie - The Next IDE Board
I created a new partition on the ST (no byte swapping) just to see what data I get back. I saved "NEWDESK.INF" which appears in sector 260. So I need to work out at what point it needs to be byteswapped.
I also need to work out where the FAT starts and ends on each sector...
So far it looks like byteswapping appears to start at sector 63....
Next I will do a software byteswapped to see what changes in the data...
(this now works on ST & PC)
So sector numbers have changed... NEWDESK.INF is now "readable" on sector 197... before it was "unreadable" on sector 260.. odd...
A better side by side comparison...
So it looks like the first sectors are not swapped, but the later sectors are swapped...
I've dumped the ST and PC partitions here (first 300 sectors) as may need them later on..
I also need to work out where the FAT starts and ends on each sector...
So far it looks like byteswapping appears to start at sector 63....
Next I will do a software byteswapped to see what changes in the data...
(this now works on ST & PC)
So sector numbers have changed... NEWDESK.INF is now "readable" on sector 197... before it was "unreadable" on sector 260.. odd...
A better side by side comparison...
So it looks like the first sectors are not swapped, but the later sectors are swapped...
I've dumped the ST and PC partitions here (first 300 sectors) as may need them later on..
You do not have the required permissions to view the files attached to this post.
-
exxos
- Site Admin

- Posts: 28354
- Joined: 16 Aug 2017 23:19
- Location: UK
Re: Trudie - The Next IDE Board
Small update. I have come up with a solution to fix the issue above but I need to route all the datalines to the PLD :roll: Thankfully Icky has already done that , so we are looking at a re-spin of the board. So likely will be a few weeks before there is any more news on the project.
-
PhilC
- Moderator

- Posts: 7442
- Joined: 23 Mar 2018 20:22
Re: Trudie - The Next IDE Board
Good to hear this is still progressing well. :dualthumbup:
If it ain't broke, test it to Destruction.
-
exxos
- Site Admin

- Posts: 28354
- Joined: 16 Aug 2017 23:19
- Location: UK
Re: Trudie - The Next IDE Board
Yeah, doesn't seem to be any issues with the board over the previous IDE board. Though I have noticed a really odd "bug" which even atmel can't currently seems to figure out yet. This "bug" is actually in the previous IDE design as well, just I didn't notice it until recently :roll:
We still need to sort out the slave channel, but with my store things going on now, I'm not going to have much time to work on such things for a bit. But will try and get the slave working this week at some point.
-
exxos
- Site Admin

- Posts: 28354
- Joined: 16 Aug 2017 23:19
- Location: UK
Re: Trudie - The Next IDE Board
Just a quick note to say the project isnt dead :) now the H4 and web store is out of the way, I hope to get back to this over Xmas :)
Who is online
Users browsing this forum: ClaudeBot and 2 guests