tschak909 Posted April 11, 2020 Author Share Posted April 11, 2020 ok, methinks there needs to be a slight circuit fix. -Thom 1 Quote Link to comment Share on other sites More sharing options...
tschak909 Posted April 11, 2020 Author Share Posted April 11, 2020 I have spent a considerable amount of time building the plumbing required to handle different TCP/IP protocols using very abstracted interfaces that the Atari can efficiently handle. Here you see the same test programs being used to send via both the TCP and UDP protocols. 3 Quote Link to comment Share on other sites More sharing options...
Fierodoug5 Posted April 11, 2020 Share Posted April 11, 2020 8 hours ago, jamm said: Interesting. Tested with FN (I don't know why I keep wanting to write "FJ") plugged in to USB for power. Tested voltage at SIO-side of R8 (AUDIN): Not plugged in to SIO: 0v Plugged in to 800XL SIO while running audio self-test: multimeter reads ~1.5v, but slowly "drains" down and stays steady at 0.08v Plugged in to 800 SIO sitting at SDX prompt: 3.75v Thanx for confirming. Don't know why I didn't think to check for voltage on the aud in pin. Quote Link to comment Share on other sites More sharing options...
jamm Posted April 11, 2020 Share Posted April 11, 2020 2 hours ago, Fierodoug5 said: Thanx for confirming. Don't know why I didn't think to check for voltage on the aud in pin. Can you test and make sure you're seeing the same thing? Very unlikely we'd have two different electrical issues with the same end result, but I'd rather be positive... Quote Link to comment Share on other sites More sharing options...
Fierodoug5 Posted April 11, 2020 Share Posted April 11, 2020 Yup, will check it when I get home from work tonight Quote Link to comment Share on other sites More sharing options...
goldy/gmg aka lopez453 Posted April 11, 2020 Share Posted April 11, 2020 This is how 800XL audio circuits look like. To make it cleaner I just cut out the relevant part of schematics, covered unimportant things and added text to wires. When you try to measure voltage, I think the 4,7 uF capacitor is being charged and that is the reason for falling voltage. When it's charged the measured voltage is near zero it "blocks" direct current and let's the waweforms thru. I will try to dig out the 800 schematics and verify 3 Quote Link to comment Share on other sites More sharing options...
Fierodoug5 Posted April 11, 2020 Share Posted April 11, 2020 Out of curiosity, what are we using sio audio in for on the Fujinet? Quote Link to comment Share on other sites More sharing options...
tschak909 Posted April 11, 2020 Author Share Posted April 11, 2020 11 minutes ago, Fierodoug5 said: Out of curiosity, what are we using sio audio in for on the Fujinet? Many awesome things will come from that pin. It's literally connected to an 8-bit DAC that can be controlled by the ESP firmware. One of the first examples of this is literally a port of S.A.M. (done by Sebastian Macke) that @jeffpiep has ported to the ESP32, so there will be a "voice" device... without the blanking that happens under Atari hosted S.A.M. Other things? * I want to be able to stream cassettes with audio tracks. * MP3 or WAV playback * Seriously, it's an 8-bit DAC that we can use for sound output! -Thom 5 Quote Link to comment Share on other sites More sharing options...
goldy/gmg aka lopez453 Posted April 11, 2020 Share Posted April 11, 2020 (edited) 800 does not have a DC isolating capacitor between sio connector and the audio circuit input... So in your Fujinet device put 100nf capacitor in the Audio path (between resistor and the sio connector audio pin) and it should be better. Edited April 11, 2020 by goldy/gmg aka lopez453 4 Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted April 11, 2020 Share Posted April 11, 2020 (edited) If there is a problem with matching audio up (impedance, voltage, current) perhaps optocoupler isolation, isolation caps, transformer, you know the usual suspects Edited April 11, 2020 by _The Doctor__ Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted April 11, 2020 Share Posted April 11, 2020 Okay someone else posted at the same time but the result will be similar Quote Link to comment Share on other sites More sharing options...
Fierodoug5 Posted April 11, 2020 Share Posted April 11, 2020 (edited) 15 minutes ago, goldy/gmg aka lopez453 said: So in your Fujinet device put 100nf capacitor in the Audio path (between resistor and the sio connector audio pin) I should (hopefully) have a capacitor in stock I can try tonight. Edited April 11, 2020 by Fierodoug5 Quote Link to comment Share on other sites More sharing options...
Kyle22 Posted April 11, 2020 Share Posted April 11, 2020 22 minutes ago, goldy/gmg aka lopez453 said: 800 does not have a DC isolating capacitor between sio connector and the audio circuit input... So in your Fujinet device put 100nf capacitor in the Audio path (between resistor and the sio connector audio pin) and it should be better. I think you meant to say 4.7 uF instead of 100 nf. Quote Link to comment Share on other sites More sharing options...
goldy/gmg aka lopez453 Posted April 11, 2020 Share Posted April 11, 2020 (edited) 4 hours ago, Kyle22 said: I think you meant to say 4.7 uF instead of 100 nf. basically it does not matter so much, but at 800 the pokey is connected to the audio circuit thru 100nf, so I tried to match the sio audio path with pokey audio path. in xl series the sio sound is mixed in later stage (after the pokey is amplified by the Op amp) together with parallel bus via 4,7uF capacitor and Sio via 4,7uF capacitor. it needs more testing when there will be actual music sound coming from FN dacs. lower capacity could filter out the bass sounds, higher could cause short pulse sound during powering on and still interfere with pokey volume in 800. Edited April 11, 2020 by goldy/gmg aka lopez453 1 Quote Link to comment Share on other sites More sharing options...
Kyle22 Posted April 12, 2020 Share Posted April 12, 2020 The 100nf 1nf is the bypass to ground if you look carefully. The 47uf 4.7 is the input cap. Look at the 800xl schematic. Sorry about the initial typos. I have my weak glasses on. :) 1 Quote Link to comment Share on other sites More sharing options...
Kyle22 Posted April 12, 2020 Share Posted April 12, 2020 @tschak909, is it possible to scrape https: sites using this? This would be an amazing thing for my BBS. Quote Link to comment Share on other sites More sharing options...
Fierodoug5 Posted April 12, 2020 Share Posted April 12, 2020 ok, put a 4.7uf cap between sio pin "AUD IN" and the resistor. Now have perfect sound on the 800. Voltage on the pin starts about a volt or so and slowly drops to zero, bout the same as the pin was doing on the 800xl without the cap in place. 4 1 Quote Link to comment Share on other sites More sharing options...
Fierodoug5 Posted April 12, 2020 Share Posted April 12, 2020 Still works fine on the 800xl as well 5 Quote Link to comment Share on other sites More sharing options...
tschak909 Posted April 12, 2020 Author Share Posted April 12, 2020 1 hour ago, Kyle22 said: @tschak909, is it possible to scrape https: sites using this? This would be an amazing thing for my BBS. http and https will be supported protocols, yes. I am at the start of implementing that right now. Idea is to use AUX1 to specify a particular HTTP verb, so. 4 = GET 12 = POST 13 = PUT This alongside AUX2 for CR/LF<->EOL translation, which is shared across all protocols: 1 = CR 2 = LF 3 = CR/LF 2 Quote Link to comment Share on other sites More sharing options...
Kyle22 Posted April 12, 2020 Share Posted April 12, 2020 24 minutes ago, tschak909 said: http and https will be supported protocols, yes. I am at the start of implementing that right now. Idea is to use AUX1 to specify a particular HTTP verb, so. 4 = GET 12 = POST 13 = PUT This alongside AUX2 for CR/LF<->EOL translation, which is shared across all protocols: 1 = CR 2 = LF 3 = CR/LF @JohnPolka Check this out. Will it work? Quote Link to comment Share on other sites More sharing options...
tschak909 Posted April 12, 2020 Author Share Posted April 12, 2020 in the end, you're literally getting a CIO interface, so for an HTTP GET: 5 DIM A$(256) 10 OPEN #1,4,3,"N:HTTP://WWW.FOO.COM/" 20 TRAP 60 30 INPUT #1;A$ 40 PRINT A$ 50 GOTO 30 60 CLOSE #1 2 Quote Link to comment Share on other sites More sharing options...
tschak909 Posted April 12, 2020 Author Share Posted April 12, 2020 For those debugging audio, maybe this will help, schematic on page 7-1: http://www.atarimania.com/documents/Atari_410_Program_Cassette_Recorder_Field_Service_Manual.pdf 2 Quote Link to comment Share on other sites More sharing options...
JohnPolka Posted April 12, 2020 Share Posted April 12, 2020 1 hour ago, tschak909 said: in the end, you're literally getting a CIO interface, so for an HTTP GET: 5 DIM A$(256) 10 OPEN #1,4,3,"N:HTTP://WWW.FOO.COM/" 20 TRAP 60 30 INPUT #1;A$ 40 PRINT A$ 50 GOTO 30 60 CLOSE #1 To clarify, Will this change to line 10 also work? 10 OPEN #1, 4, 3, "N:HTTPS://WWW.FOO.COM/" 1 Quote Link to comment Share on other sites More sharing options...
tschak909 Posted April 12, 2020 Author Share Posted April 12, 2020 Just now, JohnPolka said: To clarify, Will this change to line 10 also work? 10 OPEN #1, 4, 3, "N:HTTPS://WWW.FOO.COM/" Yes. -Thom 2 Quote Link to comment Share on other sites More sharing options...
goldy/gmg aka lopez453 Posted April 12, 2020 Share Posted April 12, 2020 8 hours ago, Kyle22 said: The 100nf 1nf is the bypass to ground if you look carefully. The 47uf 4.7 is the input cap. Look at the 800xl schematic. Sorry about the initial typos. I have my weak glasses on. yes thats what I said for 800xl and also described plain 800 has 100nf in the pokey path and no dc filtering capacitor in the sio path and the reason why to consider 100nf too. important is that now 4,7uF was sucessfully tested to be working for both. 1 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
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.