Page 4 of 7
Re: RISKY MSX Rev2
Posted: 28 May 2025 18:11
by ahmsx
terriblefire wrote: 28 May 2025 14:55
Yes I have empirically shown than LM1117s do not need tantalum caps. They work fine without it. Datasheets, despite what people think, are often wrong and frequently works of fiction
Probably the tantalum recommendation on the datasheet has to do more with the time the LM1117 was brought to market. And further revisions of the document probably didn't bother analyzing current day alternatives. But the datasheet hints at the ESR (aside the capacitance itself) as an important factor for stability, so I'm sure valid alternatives (even using a capacitor+resistor) could be used instead of tantalum if needed.
But in any case, the specific use of the LM1117 done here and in most circuits, where an already regulated +5V is fed to its input, probably doesn't require those conservative "works-in-all-scenarios" values as the regulation is way more easy than if the LM1117 was used in more harsh conditions with randomly fluctuating input voltages.
Re: RISKY MSX Rev2
Posted: 28 May 2025 18:36
by ahmsx
Had to flash the firmware (using WCHISPTool_CMD and a USB-A cable) twice again despite the flashing tool reporting a successfull flash, but looks good. The USB pen drive that freezed the previous version when using LFNs works fine with 3.0 beta.
Flashed manually via the menu about five roms flawlessly.
One minor suggestion: I think it is useful to report the firmware version on the "Insert USB" screen too, just in case things go bad before the USB can be mounted.
Good work, thanks.
Re: RISKY MSX Rev2
Posted: 28 May 2025 18:59
by terriblefire
I've copied this exact same circuit from the TF520, 530, 534. 328, 330, 536, TFMSX. These are all unregulated.
The data sheet here seems to be the limit of what was tested. Nothing more
Re: RISKY MSX Rev2
Posted: 28 May 2025 22:50
by arkadiusz.makarenko
I am almost there.
ExFAT doesn't include navigation structure like "." And ".."
So I decided to add another button to the mix. Now after hitting Backspace you will go one dir up. Previously you had to navigate to page 1 to go up. As pages load fairly slowly sometimes it is pain to go though like 20 pages. (I think I will add buttons to navigate faster by like 5 pages at a time)
Another issue is that I can't sort content. There is not enough ram on left to do it. So it has to appear as it is stored in filesystem.
Re: RISKY MSX Rev2
Posted: 28 May 2025 22:55
by arkadiusz.makarenko
ahmsx wrote: 28 May 2025 18:36
Had to flash the firmware (using WCHISPTool_CMD and a USB-A cable) twice again despite the flashing tool reporting a successfull flash, but looks good. The USB pen drive that freezed the previous version when using LFNs works fine with 3.0 beta.
Flashed manually via the menu about five roms flawlessly.
One minor suggestion: I think it is useful to report the firmware version on the "Insert USB" screen too, just in case things go bad before the USB can be mounted.
Good work, thanks.
I can't help with programming. WCH tools are sort of dated, and Linux ones are quite bad.
Saying this open source ones are not much better for this chip, and don't offer memory configuration anyway.
I have added firmware versions at the start. On top of it I did improve how auto flashing at the boot works. There was annoying No file found message. Now it check for file presence before attempting to flash it.
PS.
I have done release. I think it is good enough now
https://github.com/arkadiuszmakarenko/R ... tag/v3.1.0
Re: RISKY MSX Rev2
Posted: 29 May 2025 22:56
by ahmsx
Works mostly fine here.
I tested some USB pen drives which I haven't tested before with other firmware versions and on some of them (old ones, FAT16 or FAT32) the flashing menu stops at "USB Enumeration complete.".
When that happens, if you press some keys and then unplug the drive and plug another one that works, it loads the file list of the new drive correctly, but all key presses that you made during the "USB Enumeration complete." screen are then suddenly taken into account in the file menu (the keys are buffered but not consumed until the file menu shows).
That causes unwanted results because the keys pressed were not intended for the file menu.
Maybe keys should be consumed while in the "Insert USB" phase, or the key buffer should be flushed when entering the file menu to avoid this situation.
Good progress with the file menu.
Thanks.
Re: RISKY MSX Rev2
Posted: 29 May 2025 23:00
by ahmsx
Oh, and minor nitpick: the version shown when booting the firmware is 3.0.1 but the .bin file name and release refers to 3.1.0.
Re: RISKY MSX Rev2
Posted: 30 May 2025 17:01
by arkadiusz.makarenko
ahmsx wrote: 29 May 2025 22:56
Works mostly fine here.
I tested some USB pen drives which I haven't tested before with other firmware versions and on some of them (old ones, FAT16 or FAT32) the flashing menu stops at "USB Enumeration complete.".
When that happens, if you press some keys and then unplug the drive and plug another one that works, it loads the file list of the new drive correctly, but all key presses that you made during the "USB Enumeration complete." screen are then suddenly taken into account in the file menu (the keys are buffered but not consumed until the file menu shows).
That causes unwanted results because the keys pressed were not intended for the file menu.
Maybe keys should be consumed while in the "Insert USB" phase, or the key buffer should be flushed when entering the file menu to avoid this situation.
Good progress with the file menu.
Thanks.
Issue is mounting the drive. I did see similar issue at hot swapping the drive. I guess looking over mount may not be as good idea as I thought.
Ps
You don't need to press buttons, if it detected usb disconnect then it will clear data, and enumerate it again.
Re: RISKY MSX Rev2
Posted: 30 May 2025 18:00
by ahmsx
arkadiusz.makarenko wrote: 30 May 2025 17:01
Issue is mounting the drive. I did see similar issue at hot swapping the drive. I guess looking over mount may not be as good idea as I thought.
Ps
You don't need to press buttons, if it detected usb disconnect then it will clear data, and enumerate it again.
Yes, its the same issue.
You know, impatient users like me start pressing keys when computers seem frozen with a static screen... ESC, Enter, Up, Down, Left, Right, Space, Shift, ...
Maybe printing a message like "Unplug and plug a FAT-formatted USB drive again" when the mount fails and the firmware does not proceed to the file menu would help in this case.
But flushing the keys buffer too will always avoid the file menu moving around like crazy consuming all the impatient users' previous random key presses.
Thanks.
Re: RISKY MSX Rev2
Posted: 30 May 2025 19:20
by arkadiusz.makarenko
ahmsx wrote: 30 May 2025 18:00
arkadiusz.makarenko wrote: 30 May 2025 17:01
Issue is mounting the drive. I did see similar issue at hot swapping the drive. I guess looking over mount may not be as good idea as I thought.
Ps
You don't need to press buttons, if it detected usb disconnect then it will clear data, and enumerate it again.
Yes, its the same issue.
You know, impatient users like me start pressing keys when computers seem frozen with a static screen... ESC, Enter, Up, Down, Left, Right, Space, Shift, ...
Maybe printing a message like "Unplug and plug a FAT-formatted USB drive again" when the mount fails and the firmware does not proceed to the file menu would help in this case.
But flushing the keys buffer too will always avoid the file menu moving around like crazy consuming all the impatient users' previous random key presses.
Thanks.
@ahmsx Try this
https://github.com/arkadiuszmakarenko/R ... tag/v3.1.1
I have tidy up enumeration process, and removed re-enumeration (which is causing issues on slow drives).
It did fix my issue related to hot-swapping USB, so maybe it will fix it for you as well.