Jump to content

Recommended Posts

On 10/7/2020 at 7:59 AM, Mr Robot said:

I usually create save disks on my SD card and load my software from my read only TNFS server at Fujinet.Atari8bit.net. You can see server stats for it at https://atari8bit.net/fujinet/ it might help you visualise what is going on.

 

How do you get the stats of the tnfs-server for output & formatting? I'd love to do this when I put mine online.

Shell scripts run by cron produce text files and upload them to my webserver. The website reads the text files, parses them and outputs data using php.

 

I'm being too aggressive in my log clearing so some of the stats aren't reflecting reality.  I need to put a little db in place that stores the mounts so I can get the data back in a more flexible way.

 

EDIT: Thanks to bocianu there is now a usage log so I don't have to scour syslog any more!

2 hours ago, mytek said:

Hardware Version V1.3 question.

 

Is the reset connection that goes to EN still needed, and if so, what for?

 

EN is the esp32 hard reset pin. I moved it to a SMD pad that can accept a metal snap dome or just be shorted with something metal. It's not really needed since we use a button on GPIO to do a "safe" reset. The power switch is effectively a hard reset. I opted to have the smd pads "just in case".

  • Thanks 1
53 minutes ago, mozzwald said:

EN is the esp32 hard reset pin. I moved it to a SMD pad that can accept a metal snap dome or just be shorted with something metal. It's not really needed since we use a button on GPIO to do a "safe" reset. The power switch is effectively a hard reset. I opted to have the smd pads "just in case".

Perfect, thank you for making that very clear to me ? .

 

Thank you,

Michael

  • Like 1
  • 2 weeks later...
5 minutes ago, Kyle22 said:

What is the maximum capacity of SD card? I have a 64G in mine and it can't see it.

 

I think the file esp32 system lib only supports fat16 or fat32 (afaik exFat is not supported). Windows can only format fat32 up to 32G. There may be utilities for windows to format larger or you could format it with Linux ;) 

9 minutes ago, Kyle22 said:

What is the maximum capacity of SD card? I have a 64G in mine and it can't see it.

 

It really depends on the card type, I have successfully used a 64G card. It's really sort of pointless, just grab a 2G SanDisk for it, and be done with it.

 

-Thom

  • Like 1
13 hours ago, tschak909 said:

It really depends on the card type, I have successfully used a 64G card. It's really sort of pointless, just grab a 2G SanDisk for it, and be done with it.

 

-Thom

Except that, just as with CF cards before this, the smaller, slower cards are becoming scarcer, less easy to find and thus more expensive than the newer, faster and larger capacity cards. A couple weeks ago I bought a pair 32GB Sandisk cards for under $8 each without even doing any comparison shopping to save a few bucks. The 16GB versions were like $11 each. I didn’t even bother looking at anything smaller, as most of those I saw listed were oddball weird-name units that I couldn’t bring myself to trust. So my suggestion to folks is just buy some good-quality 16GB-32GB cards from Sandisk, Samsung or whomever plus a few spares and be done with it. 

  • Like 4
  • 3 weeks later...

Not sure if this is a firmware issue or a hardware issue, but I was looking at the FN webapp tonight and when checking the SIO voltage it looked quite low. But when I put a voltmeter on the IO35 pin it was at 3.3V which I would think is the top of its range being on a 3.3V based system.

 

FN_hardware.png.51ed32938d03576e9d4279020546af8e.png

 

So I naturally assumed that 3.3V should = an SIO voltage of 5V unless I'm mistaken. Any idea of why it's so far off the mark?

One of the ADC's is used to sample the SIO voltage, and it's calculated in this routine:

int SystemManager::get_sio_voltage()
{
    // Configure ADC1_CH7
    adc1_config_width(ADC_WIDTH_12Bit);
    adc1_config_channel_atten(ADC1_CHANNEL_7, ADC_ATTEN_11db);

    // Calculate ADC characteristics
    esp_adc_cal_characteristics_t adc_chars;
    esp_adc_cal_characterize(ADC_UNIT_1, ADC_ATTEN_DB_11, ADC_WIDTH_BIT_12, 1100, &adc_chars);

    int samples = 10;
    uint32_t avgV = 0;
    uint32_t vcc = 0;

    for (int i = 0; i < samples; i++)
    {
        esp_adc_cal_get_voltage(ADC_CHANNEL_7, &adc_chars, &vcc);
        avgV += vcc;
        //delayMicroseconds(5);
    }

    avgV /= samples;

    if (avgV < 501)
        return 0;
    else
        return (avgV * 5900 / 3900); // SIOvoltage = Vadc*(R1+R2)/R2 (R1=2000, R2=3900)
}

-Thom

4.78V is exactly what I show on my Fujinet page as well, always has been, and I have my Fujinet powered with an external 5V USB PSU. There was a discussion about low voltage readings off of the Fujinet web config site in another thread in this sub-forum. I remember @DrVenkman and/or @_The Doctor__ talking about it being too low a reading, for me and for someone else who started the conversation who's reading was even lower, IIRC 4.75V. And it was suggested I test my line voltages, so I did.

 

I've tested the +5V off of the SIO, controller port and the PBI I installed; though I was unaware that the Incognito supplied +5V to pin 47/48, so I connected +5V from the mother board during installation. But they all read a rock steady 4.89V from OEM "universal" 31VA PSU. Slightly lower than I'd like, but I've never had an issue so far. I've half a dozen 31VA universal PSU's for my Atari drives and 800/1200Xl, so I may see if I can find one with a slightly higher voltage output and swap them.

 

I haven't tried putting a voltmeter to my Fujinet yet,

 

EDIT: Actually, I just checked my Fujinet config page and it's shows a voltage of 4.757 now! Which, IIRC, is exactly the voltage another user was getting on his. My real voltage from the SIO is still showing 4.89V.

Edited by Gunstar
  • Like 1
2 hours ago, mytek said:

So I naturally assumed that 3.3V should = an SIO voltage of 5V unless I'm mistaken. Any idea of why it's so far off the mark?

It could be an issue with the code @tschak909 posted. I whipped that up very early on. I have confirmed readings with my multimeter and compared them to the voltage displayed and it's very close, ie, SIO voltage measured at the FujiNet receptacle pin matched the displayed voltage.

Well to be honest, I did end up tweaking my resistor divider because my actual SIO voltage is the same as Gunstar's at 4.89V and I was seeing 4.78V. So I tweaked the values to bring it up to 3.3V on the voltage sense line, but still got 4.78V on the web app. That's when I really started getting suspicious.

 

So based on what I was seeing, it suggests that the ADC is not getting read, and that the 4.78 is a fixed responseEDIT: I'm not saying this is actually what is happening in the code, but more a description of what it appears to look like based on changing the input voltage and yet getting the same reading.

7 hours ago, mytek said:

So based on what I was seeing, it suggests that the ADC is not getting read, and that the 4.78 is a fixed responseEDIT: I'm not saying this is actually what is happening in the code, but more a description of what it appears to look like based on changing the input voltage and yet getting the same reading.

I opened an issue on Github so I don't forget about it. The only point of it originally was to check if the Atari was on or not. It's no longer used for anything other than the display. If it's not easily fixed or fixable at all, I'll remove it.

Just now, mozzwald said:

I opened an issue on Github so I don't forget about it. The only point of it originally was to check if the Atari was on or not. It's no longer used for anything other than the display. If it's not easily fixed or fixable at all, I'll remove it.

Fair enough ;-)

You might lose a data point for troubleshooting with this removed... if it was working it would let you know when an Atari and fujinet combo had power issues possibly

Sorry to go this page. I unsuccessfully tried to access from my e-mail to the actual page. 

The issue is with the latest FujiNet update. After update it can access to the website

showing folders like usually, but I am not able to open any folders except my SD card inserted

to FujiNet. Trying to open any other the yellow light (right-top) is not flashing just light solid.

No sound, no action. ...

 

FujiNet Version: 0.5.ed9d13bd
Version Date/Time: 2021-03-15 00:55:47
Build Date/Time: Mon Mar 15 03:32:59 UTC
2021

 

Peter

2 hours ago, KPeter said:

Sorry to go this page. I unsuccessfully tried to access from my e-mail to the actual page. 

The issue is with the latest FujiNet update. After update it can access to the website

showing folders like usually, but I am not able to open any folders except my SD card inserted

to FujiNet. Trying to open any other the yellow light (right-top) is not flashing just light solid.

No sound, no action. ...

 


FujiNet Version: 0.5.ed9d13bd
Version Date/Time: 2021-03-15 00:55:47
Build Date/Time: Mon Mar 15 03:32:59 UTC
2021

 

Peter

This thread is for hardware discussion. Please start a new thread or use the Testing and Bug Reporting thread and provide serial debug output while trying to access a TNFS server. You can use the FujiNet Flasher to capture this information.

Any thoughts on using one of the spare I/O pins (IO12 or IO15) for an SD card detection pin?

 

Since the SD card detect switch is wired with one side to GND, the other side could go to an I/O pin with a pull-up resistor, and then if a new card were inserted, the firmware could be coded to detect that occurrence and issue a Soft Reset to initialize and mount the new card automatically.

 

SD_interface.thumb.png.021e54dececec1df9c3b9829399b1337.png

  • Like 5
  • Thanks 1
35 minutes ago, mytek said:

Any thoughts on using one of the spare I/O pins (IO12 or IO15) for an SD card detection pin?

 

Since the SD card detect switch is wired with one side to GND, the other side could go to an I/O pin with a pull-up resistor, and then if a new card were inserted, the firmware could be coded to detect that occurrence and issue a Soft Reset to initialize and mount the new card automatically.

 

SD_interface.thumb.png.021e54dececec1df9c3b9829399b1337.png

Nice idea!  I love the way that the CF based HDD solutions using fjc's BIOS will auto-detect card swaps and just work seamlessly.

  • Like 2
37 minutes ago, mytek said:

Any thoughts on using one of the spare I/O pins (IO12 or IO15) for an SD card detection pin?

 

Since the SD card detect switch is wired with one side to GND, the other side could go to an I/O pin with a pull-up resistor, and then if a new card were inserted, the firmware could be coded to detect that occurrence and issue a Soft Reset to initialize and mount the new card automatically.

 

SD_interface.thumb.png.021e54dececec1df9c3b9829399b1337.png

I also like this, is this feasible

@mozzwald ?

 

-Thom

2 hours ago, mytek said:

Any thoughts on using one of the spare I/O pins (IO12 or IO15) for an SD card detection pin?

 

Since the SD card detect switch is wired with one side to GND, the other side could go to an I/O pin with a pull-up resistor, and then if a new card were inserted, the firmware could be coded to detect that occurrence and issue a Soft Reset to initialize and mount the new card automatically.

 

1 hour ago, tschak909 said:

I also like this, is this feasible

I think this can be done. I'm probably gunna go with IO12 due to the current board layout and IO12 is close to the SD slot. The firmware will need to check for the pull-up at boot and enable this feature if available. Will try to bodge up a board when I have time.

  • Like 4
3 hours ago, mozzwald said:

I think this can be done. I'm probably gunna go with IO12 due to the current board layout and IO12 is close to the SD slot. The firmware will need to check for the pull-up at boot and enable this feature if available. Will try to bodge up a board when I have time.

I'll do the same on my end when you roll out an update to test.

 

Thanks for being receptive to the idea :)

9 minutes ago, mytek said:

I'll do the same on my end when you roll out an update to test.

 

Thanks for being receptive to the idea :)

Of course, besides being a good idea, this project belongs to the community. We want it to improve. :)

 

-Thom

 

  • Like 2

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...