atarimac

Mark, I retested, this time looking at the drive sector count instead, using a watch with a second hand. To sector 271, it was 22 seconds. Now, I am running on a quad-core MacPro, which is going to be significantly faster than your PowerBook G4, (or the WIN95 laptop). USB is a software intensive bus (which is why Firewire 400 is faster than USB2 for disk drives, even though USB2's raw bus speed is greater 480 vs 400). A quick calculation, if it's 271 sectors, at 128 bytes/sector + say 5 bytes overhead for the read command, you are looking at 36043 chars, divided by 1920 chars/sec (19200 baud is the normal SIO rate, without some sort of DOS speedup, and 10 bits/char), the theoretical minimum is 18.77 seconds. There is going to be some delay in the command signal processing on each sector, and I think I was actually measuring from turning the power switch on, so there is a couple of second delay there to. So it looks like the MacPro and the FTDI are as fast as you can get for standard SIO. Mark

Mark, On my MacPro, with the FTDI dongle, I loaded the game to the Load Runner splash screen in about 22 seconds. (I wasn't watching the sector display, I can do that again later) I did note something strange though. I hadn't used SIO2OSX much since I got my new MacPro, and when I set up the MacPro I had rearranged my USB setup and Hubs. With the Dongle plugged into my Belkin USB 2.0 Hub, I got much slower performance (around 55 seconds), and it had many stalls during the load, and sometimes wouldn't load correctly. (I'm not sure why the Hub would cause this, however, the dongle is a USB 1.0 device, so with a 2.0 Hub, it will use something called Split Transactions, which might affect it. I can try other hubs to see if that has an effect. I'd be curious if you were using a Hub or had the Prolific dongle plugged directly in. I'm also amazed the Prolific even worked at all. In my testing with a couple of Prolific dongles early in the development of SIO2OSX, I couldn't get them to function at all. But perhaps Prolific changed their drivers and now they partially work. Mark

All, Not sure if everyone noticed (posts were in emulation forum), but version 3.9.1 of Atari800MacX has been released that fixes this (available from www.atarimac.com). I also checked the changes into the atari800 CVS, so that version of the emulator for other platforms should be fixed as well (I tested the Linux version). Thanks, Mark

Thanks for the kudos guys. If you have feature requests, let me know, as I'm starting the 4.0 version. Not sure how soon it will be released,as I have quite a bit of work to do on it.....

All, I found an issue with State Saving in version 3.9.0 of Atari800MacX while I was working on integrating the newest Atari800 base emulator code for my 4.0 release. I've issued a simple bug fix version of 3.9.1. You can download it from www.atarimac.com. Thanks, Mark

Here's one another SIO2OSX user says he bought and it works, and is FTDI. It's German, so not sure what shipping, etc are from UK, but here is the link he gave me: http://www.er-tronik.de/shop2/product_info...210d62d6c892343 I know that it is a pain trying to find the right adapter, and I'm sure that's why Steve at Atarimax.com created his own USB dongle. The issue with most of the Serial dongles, is that they are not real time on any of the handshake/control lines. They work fine for "3-wire" serial applications where all you need is transmit, receive, and ground. However, to emulate SIO, you have to use one of the modem control lines, and the timing has to be accurate. The reason FTDI works is that it returns the modem status lines in every USB frame (or once per 1ms). The other dongles are making their driver poll for the lines, by sending a separate USB message and waiting for the reply, and the latency is more than the SIO emulation can handle. With anybody who has a MacPro (not many I know), the other option which I'm pretty sure would work is using a PCIE 16550 Serial card. I'm pretty sure OSX has a 16550 standard. But that only helps MacPro users, and the cards aren't real cheap either.

Ooops....Thanks Al ......a link was helpful. (And folks, Al should know, since he hosts the site...Thanks again to him).

Fletch, Number 5 to be exact And I'll have to eat my words, as I went and counted my Kagi email receipts. Seems I've sold 27 of them so far. So the good news is you were an early adopter, the bad news is that it's not as exclusive of a club as I once thought. Everyone, And as I said, it's really a matter of how much time it will take, which I could determine if I had the interface I needed to code to. I hate to say that's a factor, but my day job takes more time than when I started this, and believe it or not, SIO2OSX proceeds won't pay the mortgage payment Thanks, Mark

Spookt, I won't say no offhand, but besides the cost of the adapter, it's going to be a question of how big a job it is. The first step there would be for Steve to release the information on it, before I could determine that. I asked him for it when he first came out with the USB version (a couple of years ago now I think), and he had indicated he was going to release it, but I've never seen anything on it. Mark

All, The SIO2USB adapter doesn't work with SIO2OSX. It would require a specific Mac OSX driver, or some other specific code (user space) for that adapter, and he hasn't publicly published any details on the interface. Cost wise there isn't a huge difference. A USB Serial dongle is about $20, and you can find a serial cable for $10 or under, so getting a Serial SIO2PC device is about the same cost as the SIO2USB> Even if I did have the data, I wouldn't have a huge incentive to add the support. First, I'd have to buy one of the adapters, and the market for SIO2OSX is quite limited (I've only sold 15 or so of them in the time it's been out) so at $25 per copy of SIO2OSX, I wouldn't get much return on my time. Thanks, Mark

Is this emu essentially a mac version of a800win+ (sorry, never owned a mac), if so, perhaps you can team up with unixcoffee in writing/programming the next version of a800win+ as well and incorporate some of the exclusive features of the mac version.....just a thought, that is all Carmel, The emulator is a Mac version of Atari800. a800win+ was originally developed from the Atari800 code base, but from an old version, which is what Unixcoffee is trying to bring up to date. Some of the features of Atari800MacX wouldn't be easy to bring over, as they depend on features of the Mac OS that aren't in Windows (printer emulation being the best example). I'm afraid I wouldn't be much help on a800win+, since my knowledge of Windows programming is very thin (not that I couldn't learn, but I don't have the time...) Mark

kurtm, You can use Action, Mac/65, etc. with the Atari800MacX emulator on the Mac, as I added that capability a few versions ago. It will only work with the standard cart image, not the Max Flash image, as the emulator doesn't have a way to know that a SpartaDosX Max flash image is a SpartDosX, and not some other Max Flash cartridge. I've talked to one of the other Atari800 team members about adding this to the core emulator (Atari800), and that may happen in the near future. Of course Atari800WinPlus would have to be added by the guys currently looking at rebuilding it (as my knowledge of Windows programming is very thin.) Mark

All, Tonight I released an updated version of the Atari800MacX emulator. It's mainly a bug fix release, I hope to be adding some new features in the next release. The following are the changes: Bugs Fixed: Fixed issue with 8Mbit Flash Cartridge images not working. Fixed issue with XEP80 Emulation not working with the new SpartaDos cartrdiges Fixed issue when Joystick emulation was turned off with control menu, joystick trigger was always reported as pressed. Fixed issue with Right keyboard Meta keys not being recognized on newer Macs. This replaces the bad fix which was issued with version 3.8.0.

I would like a set also. Thanks, Mark.

I got mine as well today, and wanted to add to the kudos. It's a great job, very professionally done, and something I will be using for years to come. Thanks again.... Mark

All, I've found the problem. It had to do with the way the emulator was simulating the XEP80 input. The old version had no timing in it, and was simply depending on the number of times the port was read. However, the SpartaDOS driver sometimes reads twice in the same bit period, which was causing it to fail. There were several other logic errors in the XEP80 emulator, which I fixed as well. If you would like a Beta version to help me test it, I will be putting one together in the next couple of days. Message me and I will let you know where to get it. Right now it works with the original driver and the SpartaDOS driver from the new cart. It seems to have issues with the original SpartaX cart image I have, but I'd prefer it work with the new version. Also, when we determine it is solved for sure, I'll make sure the Atari800 CVS gets updated as well. Thanks, mark

I think they look fine, as good or better than many of the other manual scans I've seen. Thanks for the hard work!

All, I'm the author of Atari800MacX, and you are correct, it looks like the Sparta Cart XEP80 driver isn't working with the emulator, most likely the emulator's fault. I had tested most of the demo disks and XEP80 programs I could find (including AtariWriter80). However, I never tried Sparta Cart with it, most likely because I never had one back in the day (and still don't ). To answer the question about Atari800, as far as I know, they are using the same code base that I am. I think one of the guys from the core team added it to the standard SDL version. There seems to be two problems with the emulation. You are seeing the "not found on joystick port 2" if you ever do a coldstart. It works the first time you start the emulator, insert a boot disk and the cartridge, but then a coldstart causes the above message. I tried making a coldstart reset the XEP80 emulation state variables, but that doesn't do it, so I'm not sure what exactly is causing this. And when I say "works" above, I'm being generous . I get the same garbage that you guys see on the emulated XEP80 screen. If you do a warmstart however, you will get a little added garbage, so it's not totally dead. If I had to guess, it looks like some sort of problem with the cursor/location control commands, perhaps something that Sparta is doing that none of the other programs I tested with did. I will look at it more as I get time, however, my day job has me pretty swamped right now. Stay tuned....... Thanks, Mark

From earlier posts and threads (and searching on my part), no there doesn't seem to be a copy of this online. It would be worth preserving, as it seems to be one of the very worthwhile languages for the Atari that is not well documented.

If you are talking about using Atari800MacX to talk to telnet BBS's, you can do that. You should use your favorite Atari Terminal program (like BobTerm), and make sure R: emulation is turned on in the emulator. Then to "Dial" out in Bobterm, when you go to terminal mode, type: ATDI hostname port i.e. ADTI thebbs.com 23 If you are talking about using a real serial port as a 850 simulator, Atari800MacX doesn't have that capability right now....

The printer emulation's are quite complex, I suspect I spent a least a month on each one of them. I won't say I will not ever do another one, but at the moment I don't have the free time for it. The 1029 was never released here in the states, so I would have to have a manual or technical documentation to be able to emulate it, as well as sample printouts. Mark

tjb, Thanks for the compliments, and I'm glad to hear it's running well on Leopard. I have Leopard, but haven't installed it yet. Mark No big deal, I'll just pick up a USB to serial adaptor. I just finished installing Atari800MacX on my Mac Mini running Leopard and it works great! I'm new to the Mac so I'm just getting started with emulation on the Mac. I really like the media status window window, it's much easier than going through the menus. Very nice application. tjb

No, it does not. I have talked with Steve about it, and he has talked about releasing sample code that would allow others to use the interface without a kernel driver, but at this point, it's not possible for me to support it. Mark

All, I've now released version 3.8.1, which has pulled the keyboard "fix" back out. It was causing more problems, and needs to be debugged further. I've also updated the online documentation, and removed the references to my local drive fromt he HTML. Thanks, Mark

Finally, one user has noted there is an issue with the built in help, some of the images on the Preferences panels pages have references to my local drive. I will release a new 3.8.1 version in the next couple of days fixing that, in the meantime, use the PDF manual if you have issues.