flashjazzcat Posted January 28, 2018 Share Posted January 28, 2018 (edited) I just rigged up the Sparkfun board. The board came with no headers but luckily, after rifling through my boxes of bits, I found a female header strip just long enough to break in two and fit to the board. So - with the FTDI adapter in place and no other peripherals connected, I tried setting it up with RespeQt. I got absolutely nowhere (with neither the U1MB HSIO driver nor the SDX driver) until I set the handshaking protocol to "None", which was somewhat unexpected and seems far from ideal. At this point, I was able to get a disk directory of virtual drive B: at the SDX prompt. SDX SIO driver works, as does the U1MB PBI driver. Then I connected the XF551 as drive 1 and tried again. RespeQt virtual drive B: was now unresponsive. XF works fine, but no access via the FTDI adapter. LED flashes as the SIO sends out command frames, but I get a NAK error. Interestingly, virtual drive 2 still works after I power on the XF551 up until the point I actually query the floppy disk. Once a transfer has occurred on the XF551, the RespeQt drive is toast again. If I then power off the XF, the RespeQt volume becomes accessible again. If I then disable the U1MB PBI HSIO driver and try again, I get identical results (and this would be via the SDX SIO drivers). So I think it's safe to conclude this isn't a software issue. I'm not sure what handshake the Sparkfun board is meant to use, but perhaps this is at the root of the issue somehow. EDIT: I see this FTDI adapter is supposed to use a DSR handshake, so I guess step one is figuring out why that doesn't work for me. Heh... Forgot to put the handshake jumper on the 1088XEL board. Did that, and everything works perfectly, including XF and FTDI in tandem, using PBI drivers or the SDX drivers. So: jumpering RI, CTS or DSR on the six pin header right in front of the Sparkfun board would be my recommended troubleshooting step for all reported issues. Edited January 28, 2018 by flashjazzcat 1 Quote Link to comment Share on other sites More sharing options...
+Firedawg Posted January 28, 2018 Author Share Posted January 28, 2018 @FJC - I'm not experiencing any issues with real and virtual drives working together or separately while using RespeQT. I have 4 1050's daisy chained up connected to the 1088XEL with FTDI adapter connected to the PC (Win10) which is given a Serial Port Assignment of COM3. With U1MB bios settings for PBI BIOS enabled and SIO driver: HSIO, SIO devices: D1-D4. As long as I do not use a virtual drive D1 thru D4 I'm good, once I do I receive an error as earlier mentioned. For RespeQT settings I have Port name: COM3, Handshake method: CTS (I have used DSR with same results), High speed mode baud rate: 57600 (3x), C/E response delay: 300. I do not have a working XF551 at the moment to compare apples to apples with DrV's setup. @ FJC or DrV - If there is any configuration that you would like to test let me know. Quote Link to comment Share on other sites More sharing options...
flashjazzcat Posted January 28, 2018 Share Posted January 28, 2018 Since you say you're not experiencing any issues using real and virtual drives together, but then report an issue when using a virtual drive on D1-D4, I assume you mean there's a problem when you've disconnected all the 1050s and then mounted a virtual disk on one of the first four drive numbers. If that's the case, then I can't reproduce the problem here, since I have a virtual drive on D2: with the exact same settings as yours (aside from the handshake method). If you're not first disconnecting all the 1050s - or at least disconnecting the 1050 with the same drive ID as the virtual drive in RespeQt - then that's your issue, but I'm sure you're already aware you don't want a physical drive and a virtual drive on the same ID. Quote Link to comment Share on other sites More sharing options...
+Firedawg Posted January 28, 2018 Author Share Posted January 28, 2018 Since you say you're not experiencing any issues using real and virtual drives together, but then report an issue when using a virtual drive on D1-D4, I assume you mean there's a problem when you've disconnected all the 1050s and then mounted a virtual disk on one of the first four drive numbers. If that's the case, then I can't reproduce the problem here, since I have a virtual drive on D2: with the exact same settings as yours (aside from the handshake method). If you're not first disconnecting all the 1050s - or at least disconnecting the 1050 with the same drive ID as the virtual drive in RespeQt - then that's your issue, but I'm sure you're already aware you don't want a physical drive and a virtual drive on the same ID. No problems with RespeQT D1-D4 when physical drives are disconnected as stated before. Totally understand not using the same ID for drives real or virtual. My confusions is understanding DrV's issue associated with not able to use real and virtual drives together. At the moment I have no issues with my FTDI other than maybe interjecting more confusion to the mix, so I'll step back for now. Thanks Jon. 1 Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted January 28, 2018 Share Posted January 28, 2018 (edited) Perhaps a testing protocol should be issued to all to test this, but it's looking like I posted and suspected a while back... Group, please plug in your sparkfun and run respeqt set a bootable disk in drive 1 slot of respeqt, and have a real physical drive turned on as drive 2 with a bootable disk loaded. try to access both. post results next set a bootable disk in drive 2 slot of respeqt, and have a real physical drive turned on as drive 1 with a bootable disk loaded. try to access both. post results please include a close up picture of your sparkfun board so we can see your setup and chip versions Edited January 28, 2018 by _The Doctor__ Quote Link to comment Share on other sites More sharing options...
flashjazzcat Posted January 28, 2018 Share Posted January 28, 2018 I thought the handshake jumper would be the charm, but it appears I was the only one daft enough to leave it off (Herb has his connected properly). So, two of us have no issues with mixing real and virtual drives providing ID conflicts are avoided. No clue what Herb's issue is yet. 1 Quote Link to comment Share on other sites More sharing options...
+Firedawg Posted January 28, 2018 Author Share Posted January 28, 2018 Group, please plug in your sparkfun and run respeqtset a bootable disk in drive 1 slot of aspeqt, and have a real physical drive turned on as drive 2 with a bootable disk loaded. try to access both. Results: Successful access to drive 1 RespeQt. Successful access to drive 2 real physical drive. Set a bootable disk in drive 2 slot of aspeqt, and have a real physical drive turned on as drive 1 with a bootable disk loaded. try to access both. Results: Successful access to drive 2 RespeQt. Successful access to drive 1 real physical drive. Note - This works when I'm using one drive turn on at a time, as I usually have 4 on. I know Herb is using just one drive so this testing is more appropriate to his setup. 1 Quote Link to comment Share on other sites More sharing options...
+Firedawg Posted January 28, 2018 Author Share Posted January 28, 2018 I thought the handshake jumper would be the charm, but it appears I was the only one daft enough to leave it off (Herb has his connected properly). So, two of us have no issues with mixing real and virtual drives providing ID conflicts are avoided. No clue what Herb's issue is yet. It's hard communicate solely in text as we inadvertently leave out much needed info at times. I know it must be frustrating on your end trying to diagnose issues out of sight and little info or worst bad info. Quote Link to comment Share on other sites More sharing options...
flashjazzcat Posted January 28, 2018 Share Posted January 28, 2018 Yes indeed, and I'm concluding that your original issue was indeed caused by having a virtual drive on the same ID as one of the floppies (which is an absolute universal no-no, regardless of the devices in question). 1 Quote Link to comment Share on other sites More sharing options...
flashjazzcat Posted January 28, 2018 Share Posted January 28, 2018 (edited) In other news, I finally hooked up Rapidus: Appears to work, but I may make a video of more elaborate testing. It's going to be difficult to test the XEL loader since the MBPI is not only completely blocked by the Rapidus PCB, but I'd need to rig up a pass-thru connector for the three Rapidus signals. As it is, Rapidus is sitting on two stacked 40-pin sockets in order to clear U1MB (owing to my replacement three-way header) and the connectors going into the MPBI. I'm actually inclined to remove the Rapidus-specific features of the default BIOS plugin and replace the Rapidus menu item with a generic "Device M1" label, not because of any disdain for Rapidus, but because a) I don't envisage many people fitting Rapidus to a 1088XEL, and b) strictly speaking, the M1 signal placement and labelling on the MPBI connector implies that it's considered to be a general purpose IO signal. Of course, restoring the Rapidus-specific functionality would be as simple as flashing the 1KB Rapidus plug-in to the U1MB with UFLASH. EDIT: Early signs are that the SIDE loader (which I flashed to enable use of the SIDE2, since the MPBI is blocked) is encountering heavy read errors in 65C816 mode, so I doubt I'll labour over this avenue of investigation for too long. Edited January 28, 2018 by flashjazzcat 6 Quote Link to comment Share on other sites More sharing options...
Mr Robot Posted January 28, 2018 Share Posted January 28, 2018 I finally got chance to start my first 1088XEL today in between the making of Sunday lunch. Mrs. Robot loves all the cute stripy things! 7 Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted January 28, 2018 Share Posted January 28, 2018 Group, please plug in your sparkfun and run respeqtset a bootable disk in drive 1 slot of aspeqt, and have a real physical drive turned on as drive 2 with a bootable disk loaded. try to access both. Results: Successful access to drive 1 RespeQt. Successful access to drive 2 real physical drive. Set a bootable disk in drive 2 slot of aspeqt, and have a real physical drive turned on as drive 1 with a bootable disk loaded. try to access both. Results: Successful access to drive 2 RespeQt. Successful access to drive 1 real physical drive. Note - This works when I'm using one drive turn on at a time, as I usually have 4 on. I know Herb is using just one drive so this testing is more appropriate to his setup. That's basically what I was doing last night and tonight. So long as I have the USB cable connected between the Sparkfun board and the XEL, my physical drive becomes inaccessible. Unplug the Sparkfun board and the physical drive works great. I tried this with my 1050 set to both D1: and having .ATRs set in the D2: and D3: slots in RespeQt and D1: empty, and with my 1050 set to D2: with an .ATR mounted as D1: in RespeQt and D2: empty. My handshake settings on the Sparkfun board are set to DSR as is RespeQt and have been all along. Should I try another handshake setting just for fun? I'll probably repeat all of the above again tonight after I have a relaxing Sunday afternoon beer, because hey, I built an XEL-CF-II today and it works great. Thanks for yet another amazing widget Michael, and Jon for writing the code that makes it all go. 3 Quote Link to comment Share on other sites More sharing options...
flashjazzcat Posted January 28, 2018 Share Posted January 28, 2018 Thanks for the feedback and generous remarks. I'll go ahead and release the firmware in the next couple of days, since I'm satisfied we're not dealing with software issues here (or at least nothing particular to the HSIO driver, which you might corroborate by disabling it in the BIOS and performing the same tests with the default SDX SIO driver, if you have not already done so). I can't say that changing the handshake will make a difference, but since it can be tested in under a minute, why not? Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted January 28, 2018 Share Posted January 28, 2018 Thanks for the feedback and generous remarks. I'll go ahead and release the firmware in the next couple of days, since I'm satisfied we're not dealing with software issues here (or at least nothing particular to the HSIO driver, which you might corroborate by disabling it in the BIOS and performing the same tests with the default SDX SIO driver, if you have not already done so). I can't say that changing the handshake will make a difference, but since it can be tested in under a minute, why not? Will do! I'm still sipping my beer (a delightfully rich IPA from Stone Brewing called "Vengeful Spirit IPA". At 7.3% ABV, I'm sure more than two or three would be vengeful indeed. ). I'll try to be systematic, testing every handshake setting, with and without the PBI HSIO enabled. I'll be using RespeQt connected to the Sparkfun board, and to eliminate one more variable, I'll be using my second 1050 drive which is *NOT* Happy-enhanced. It's just a bone-stock drive. Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted January 28, 2018 Share Posted January 28, 2018 please show us your spark fun board using picture close enough to read the chips and see the circuit board lay out... and if everyone will follow the protocol so we can get more data that would be awesome 1 Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted January 28, 2018 Share Posted January 28, 2018 In other news, I finally hooked up Rapidus: Jon, did you socket your Sparkfun board? it looks taller than others, Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted January 28, 2018 Share Posted January 28, 2018 please show us your spark fun board using picture close enough to read the chips and see the circuit board lay out... and if everyone will follow the protocol so we can get more data that would be awesome Sure, but it's the same board spec'd on the project BOM, the BOB-12731 , with an FT232R chip. As you can see, the jumper is set to DSR. Tests soon ... pint glass is almost empty. Quote Link to comment Share on other sites More sharing options...
flashjazzcat Posted January 28, 2018 Share Posted January 28, 2018 Jon, did you socket your Sparkfun board? it looks taller than others, I did: it's on female headers I found in the cupboard. I guess this will mess up IO panel alignment, but I don't much like soldering stuff like this into the board. 1 Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted January 28, 2018 Share Posted January 28, 2018 I did: it's on female headers I found in the cupboard. I guess this will mess up IO panel alignment, but I don't much like soldering stuff like this into the board. I am thinking the same, better to have it socketed just in case Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted January 28, 2018 Share Posted January 28, 2018 Okay, I tested with all three Sparkfun jumper settings (plus the jumper removed and NONE set in RespeQt), with and without PBI HSIO - same results as before. With a virtual disk mounted as D1: in RespeQt and my 1050 set as D2:, the physical drive does not respond to commands at all. However, and here's where it gets mildly interesting from a troubleshooting standpoint, if I set my 1050 to D1: and mount a virtual disk in D2: in RespeQt, the disk in D1: spins up on boot and sometimes (rarely) when I try a basic DIR command from the SDX command line. Doesn't matter what protocol, though it seemed like it took a bit longer with the Sparkfun board and RespeQt set to CTS than with the other options. It still times out with a 138 error however. As soon as I unplug the USB cable from my PC, I can access the 1050, whether set to D1: for the first set of tests, or D2: in the second. Quote Link to comment Share on other sites More sharing options...
flashjazzcat Posted January 28, 2018 Share Posted January 28, 2018 I wonder if something on the FTDI board's USB connector is shorting one of the SIO signals (perhaps intermittently). Puzzling. 1 Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted January 29, 2018 Share Posted January 29, 2018 Problem solved. And the solution has been right under my nose since Thursday, the first day I booted the machine. It's been under your noses, too, in the photos I've already posted. It's so simple, yet I haven't seen anyone mention this before, either in Michael's design/beta build thread, nor in this one yet. Have you figured it out yet? Since (with luck) at least a few dozen more people will be building out boards in the next few months, this photo is worth sharing. Hopefully all I've gone through - lots of wasted effort as it turns out - will have been worth it if someone remembers someone's problems in the future and how simple the solution was for me: As I was reflowing solder on TOP of the Sparkfun board, just for the hell of it as a last-ditch hail mary, I saw this little switch. I pulled up the datasheet but it doesn't mention the function of the switch at all. On a lark I powered down the 1088XEL, moved the switch from the left position (where it was when it arrived and has been from the beginning) to the right position. I then rebooted the 1088XEL. My 1050, still configured as D1:, spun up and SIO noises came from the 1088XEL as SpartaDOS did its initial disk polling. I did a double-take to confirm but yes, I had disk images mounted as D2: - D4: in RespeQt. I did D1:DIR and immediately the disk spun up again and in a couple seconds the directory appeared. I then swapped over to D2: and repeated the exercise. The directory loaded up. I repeated for D3: and D4:, then back to D1: and yep, it still worked. Then I went to D5: and D6: (two of the partitions on the card in my XEL-CF-II) and they all responded perfectly as well. To ensure this wasn't a fluke, I powered everything down cold, booted back up and it all still works. So, tl;dr: if you can't use real drives and virtual drives through the Sparkfun board, make sure that #$@$#@ switch is in the RIGHT position. I deserve another beer after all this. 7 Quote Link to comment Share on other sites More sharing options...
+Firedawg Posted January 29, 2018 Author Share Posted January 29, 2018 I'm having one with you!!! 1 Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted January 29, 2018 Share Posted January 29, 2018 this is why I wanted close up shots on everybody's board, to compare them, which you have now provided a very nice close up shot of ! and the prominent select before our eyes. 1 Quote Link to comment Share on other sites More sharing options...
chevymad Posted January 29, 2018 Share Posted January 29, 2018 (edited) Finished my board, and it even seems to work! lol Had some issues programming the pics with a pickit3. Wish I'd gotten the joy2pic. Still need to update the u1mb. I plugged a cart in and I don't get anything at all though? And I'm not sure how to access the xegs slot of the u1mb either? Edit, I should say I either get a black screen with a cursor upper left, or self test. Edited January 29, 2018 by chevymad 7 Quote Link to comment Share on other sites More sharing options...
Like us on Facebook
Follow us on Twitter
AtariAge on Instagram
Watch us on YouTube
Chat on Discord
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.