mizapf

No, CF7+ is still on my todo list for MESS.

Simple copy operations between DSK and hard disk images are quite easy to do with TIImageTool (I was too lazy to start the emulation first every time, hence I wrote that tool.) Open both images, mark the files with the mouse (+ CTRL or SHIFT), CTRL-C, go to the other image, CTRL-V.

You know, according to MESS policy, if there's such a cartridge, and MESS does not support it, I'll have to fix that (some day).

Hmm ... the biggest cartridge supported with 379i banking was the 128 K version. I did not notice we went up to 1 meg in the meantime. The 128 K version allowed for setting the bank with the addresses 6000 - 601E (even), i.e. 16 banks with 8 K each. Are there real 1 M cartridges, or do we talk about emulation artifacts?

It was most likely me who created that RPK file. I certainly explained it some times, but I'll give another short explanation here. The RPK file is a ZIP file which contains the contents of GROMs and ROMs of a cartridge. In most cases this is a file for the complete GROM set, up to 40 KiB long for 5 GROMs, and one or two files for the ROM area 6000-7FFF. Also in the ZIP there is an XML file called layout.xml which declares which ROM files go to which memory area. I'd recommend to unzip the RPK file and to read the layout.xml to get an idea how to build an RPK file. As for the format change, previously, there were the separate bin files which had to be mounted in different slots. This was the original way of working with cartridges. The RPK format is far superior to that way since you do not have to remember which ROM files to include, and into which slot they belong. Moreover, it opened the way to use more than one cartridge at a time, and also other enhancements like GRAMKracker became possible. The RPK format was introduced by me in MESS five years ago, with release 0.131. Until release 0.146 of May 2012, that is, two years ago, both systems (RPK and the legacy bin format) were supported next to each other, at which point I pulled out the support for bin. A migration time of 3 years is definitely long enough. The removal of the bin support made the code much better maintainable and stable. There is another cartridge format which is compatible to the RPKs to some extent, the ZIP files. They contain the cartridge contents without the layout.xml, which is now distributed with MESS as a single XML file for all cartridges. It happened with TIImageTool, not with MESS. TIFILES are single files. DSK files are complete disk images. You cannot "convert" between those files; you have to save the TIFILES on the DSK image. This is possible by importing in TIImageTool: Open the DSK file, import the TIFILES files. The TOSEC cartridge bins are not useful anymore except for archival. I moved them into the "tosec" directory. The DSK files are OK - I picked 10 of them randomly and opened them in TIImageTool without problems. In the /emulators/cartridges directory there are no more bin files, only RPKs and ZIPs (both kinds not intended to be unpacked).

No floppy: You did not forget to plug in a disk controller, did you? (-peb:slot8 hfdc) During runtime you can check that in the OSD menu, "Slot devices". When you figured out what arguments you need for the desired configuration, you should write that in a batch file, and start that one next time. No need for QMC2. === The BATCH file is a TIFILES file which can be imported as desired (Edit/Import files...). I remember you had the problem that it is inside a RXB directory, so how come that? What did you do to get that result? ... Just checked your video clip (Posted Wed May 28, 2014 2:21 PM), you said: "All I did was I tried to convert it into an RPK". I don't understand - you wanted to convert a disk image into a cartridge? RPK files are cartridge contents (RomPacKages), so they have not really anything to do with DSK image files. I'd be curious to know what you actually did. Also, in your disks subfolder there is this BATCH.dsk file. Could you upload that as well please? The Harddisk.hd in the ZIP archive you uploaded is OK, I thought of the one that caused your "Negative seek" error message (HARDDRIVE1.hd in your MESS folder).

Rich, it is sufficient if you keep your directory of disks and cartridges, remove the exe files and ini files, and simply install a fresh release over it. Don't kill your complete installation every time, you're doing more work than necessary. MESS does not change anything in the system outside of its folder, no entries in the registry etc. Concerning the 32K mem, you should either choose whether you want to use the internal 32K (console mod) or the external 32K card. Also here, do not use both. You would not enhance your real console with 32K inside and then continue using the external 32K card either. The internal 32K is active by default. RPK files do not have anything to do with disk controllers. RPK files can only be inserted into cartridge slots. Or did you want to write "which dsk file"? If so, I can say that as long as you use only one disk controller, the flop1 image is always used for the first drive, the flop2 for the second and so on, so this matches DSK1, DSK2, ... When we have more than one controller, each controller adds its drives to a list, and there will be more flop arguments. I'm not sure but I would expect that the card in the lower-numbered slot gets the first set of drives (like flop1...flop4), the next card gets the rest (like flop5...flop8). This has nothing to do with how the controllers internally map that to DSK1...DSKx. === Apart from that, Rich, we still have some loose ends here, and before you are digging up more issues, let's fix the previous ones first. I'd be interested to learn why TIImageTool reports the "Negative seek" when opening your hard disk image. Can you send a copy of the image to me, or can you upload it somewhere? I expect the image has not been properly formatted. If this is the case, I can enhance the error output in TIImageTool to give more meaningful output. Also, I'd like to know what you did to create that RXB batch disk. And then, concerning the lockup, you should try to start it without GRAMKracker first and then see whether it still locks up. Select "single" as the cartridge slot option.

Did you get the file from http://qmc2.arcadehits.net/wordpress/download/ and follow the instructions to add the repository? I don't have Ubuntu, so I cannot test, but you should say what you mean by "does not seem to work". Error messages like missing libraries, error on start, does not show up at all etc.?

The problem goes a bit further. The problem is that the MAME core loads the images via the "-flop1", "-flop2", "-flop3", "-flop4" command line arguments. At that point, the TI emulation in MESS is not even initialized. I guess (but I'd have to spend some deeper thoughts on that) it is not possible to find out what controller plans to go for DSK5-8 and what controller takes the DSK1-4. It could well be that "flopX" and "DSKX" only correspond for a single controller configuration.

My suggestion to only use one controller is due the fact that I did not test what happens with two controllers run at the same time. Of course, you can get them working on different CRU addresses, but you have to remember that. A worse problem is what floppy drive is connected to what controller. Currently (with the legacy floppy implementation) we have Floppy Disk 1 to 4; these are associated to the floppy controllers, but I cannot say for sure which controller gets which floppy drive. See the problem? With the redesigned floppy support we will get floppy drives as slot devices under the respective controller, so this will allow for a better control. @Rich: In case of trouble, it always makes sense to strip down the configuration to the smallest set that is needed to run the software. Also, we don't have a SCSI emulation yet, you need not look for it. So if you want to use a hard disk, you should use HFDC in slot 8 and no additional BwG or TI FDC.

Rich: @1:37: You are obviously using the natural keyboard. This has the (minor) glitch that keystrokes in the partial mode also slip through to the emulation. This does not happen with the emulated keyboard. For that reason, when you press TAB in the partial mode in order to show the menu, this is translated to FCTN-7, and the emulation reacts. This is a known issue, sorry. @4:07: I cannot reproduce this problem on my system. I always get a directory list for DSK2 as well as for HARD1. We should make sure you are using a current MESS release. Can you find out, possibly using the output in the log window of QMC2? @4:24: Please only use one disk controller, either HFDC or TI FDC - not both! GROM port should be "single", and you should probably remove either the SAMSmem or the Horizon. I'm wondering why this config looks like this. It does not look that way in QMC2. What about simply starting with a minimum config like single gromport, evpc, and HFDC in slot8? You may complain that I don't explain why it's not working, but how could I, considering the fact that I never ran into the trouble you're showing in the video clip? I can try to collect the experience from users and then see what needs to be explained... @6:09: If the QUIT key does not work, first check it works on the master title screen (maybe you have to check the key mapping). I don't know whether the Box demo allows interrupts, so QUIT may not be working anyway. You can always exit by resetting the emulation: Go to partial mode (ScrLock), press F3. @6:55: Why are you using GRAMKracker? Are you sure that it is properly configured? Check the configuration on the right.

Hello Rich, @2:45: chdman is a command line tool. If you double-click it, it opens a new window, prints out some usage help, and terminates, which also makes the window disappear. You must use chdman in a command line. But, as Robert already said, there's the feature in TIImageTool to make life a lot easier here. Don't bother with chdman. The same is true for imgtool (see why I called my tool "TIImageTool"). @3:30: How did you create the harddrive image? The reported "IOException: Negative seek offset" is a consequence of an unwanted result while calculating the track to read. I am not entirely sure where and why it occurs. Maybe you can upload that HARDDRIVE1.hd so I can try it myself. I'm quite sure that there is either no file system on it, or the file system is invalid. Sorry if my tool does not catch all possible cases of broken images. @4:00: I don't quite see the point here. However, I don't know what is actually in the BATCH.dsk file and how you created it. To TIImageTool it looks as if there is a directory, which itself contains the text file. The directory listing you are showing in the background, when you used your RXB/REA, agrees with that. So both are telling there is a directory on the disk. Maybe they are right? @4:34: Rich, please be aware that TIImageTool was primarily written by me to help me with my own working with images. Everything it contains is something I felt necessary to make image usage easier. Of course, I am happy to see if other people do enjoy its features as well; for that reason I am offering it for free download. I'm certainly willing to help where needed, but I can only invest a limited amount of my free time, since there are many other things waiting for completion. If you complain that the manual is written by someone who already knows the tool, you are pretty close. The manual was never intended as a tutorial, only as a reference. It was neither intended to guide you out of trouble. And it was never intended to be a "good manual". Just complete. @7:00: The funnelweb80.dsk does not have anything to do with MESS. You may have picked it up somewhere, but it certainly is not part of the MESS distribution. That is, I don't know who it created, and how. It looks quite large to me, having 2812 free sectors and 2948 used ones, which makes 1.4 MiB. I am not sure this format is properly supported on the emulated HFDC or BwG (must be 2 sides, 80 tracks, with 36 sectors each). @7:40: Concerning the F keys, it depends whether you are using the "emulated" or the "natural" keyboard. For the emulated keyboard, the case is clear: The TI console does not have F keys either. The TI console has a FCTN key (mapped to Alt on the PC keyboard), and you press it together with the number keys. For natural keyboard there are indeed mappings of FCTN-x to Fx. I just tried, and in Extended Basic I get the last line by pressing F8. (I don't like the natural mode because of its latency.)

You mean like GUI experience or so? Or from a programmer's point of view? I wouldn't say that there are no standards. You can run programs compiled for various kinds of Linux distributions, as long as you find the required libraries. And these libraries establish some standard, albeit on a very fine-granular base. One argument I often encounter is: Who's there to help me out of trouble? Many people seem to decide on their computing environment almost completely in terms of nearest help. Certainly, there are more Windows users, thus the best choice would be Windows. Just in case something breaks. Also, the advantage of having a choice between different distributions, all with their particular focus, is not really honored by people. In contrast, they say there'll be no chance for Linux as long as there is such a diversity. As for information technology, in my view the market concept has already failed completely, and I don't think it can really be reanimated. With the consequence of having a choice, there just a lack of interest.

Tursi, as you master both worlds, why do you prefer Windows for your private projects? I am just curious, not trying to argue pro or con. I admit I never tried to get deeper into Windows than for using applications.

I once considered a specific measure for boot performance: How long does it take from turning on the PC until you get a display of a selected web page? This includes the time for booting, for logging on, and for launching an application. I know that Windows is pretty quick to show the login screen, but it takes some time after logging in before the system becomes responsive because a lot of services still have to be started.

Hello Rich, first of all, although a reboot seemed to have solved some problem, this is less likely. A reboot would have allowed Windows to load some updated driver, but we're strictly on the application layer. So there is most probably something else that changed; maybe when you last exited QMC2, settings were saved and on re-entry, things worked as expected. Probably. Some comments: When you want to return up one level in the OSD (menu), just press ESC; this is quicker. You need not step down all the way to the bottom of the menu. at 2:07: You are missing the HFDC controller on the right side. This means you won't have floppy support. On the other hand, you were able to launch MESS with floppy controller somehow. But I noticed you started the emulation under the "Played" tab. I suppose QMC2 took the configuration you used last time. Suggestion: You should make sure that all devices on the right are properly selected, then use the Save button on the right edge of the window to create a defined configuration. It will show below the "Default configuration". The problem with starting the "played" instances is that it takes the previous configuration, not the current one. So suppose you found an error this time, or you change the inserted disk, and then you start the emulation under "played", QMC2 will likely ignore all your current changes. This may look confusing. at 4:40: You get a selection "cart1" ... "cart4". This surprises me, as you should only get one "cart" slot. The four slots were in earlier releases of the TI emulation in MESS, or you have to specify "-gromport multi", which I did not notice. On the other hand, QMC2 seems to be started with 0.153 (timecode 2:01). Do you have multiple MESS releases on your system? Maybe you should move away the older ones because there may be issues with the mess.ini file between different MESS releases. at 5:14: You just left the OSD, but you did not switch back to full keyboard mode! This is a very popular thing which I found myself doing several times as well. Please always make sure you return to full keyboard mode before continuing in the emulation. One thing that will happen in the partial mode is that the P key is mapped to "Pause". Whenever you now type P, the screen is dimmed, and the emulator seems to hang. In fact, it is just paused.

OK, we have a cartridge type "paged379i". You simply concatenate all ROM files to a single 32K file (starting with bank 0), name it for instance "allroms.bin". Then create a file "layout.xml" with this content: <?xml version="1.0" encoding="utf-8"?> <romset version="1.0"> <resources> <rom id="romimage" file="allroms.bin"/> </resources> <configuration> <pcb type="paged379i"> <socket id="rom_socket" uses="romimage"/> </pcb> </configuration> </romset> and create a ZIP file with this layout.xml and the allroms.bin. Finally change the zip suffix of the file name to rpk.

How do you switch the four 8K ROMs? Maybe this is already available as a cartridge type in MESS.

I remember we gave them names: Swastikas Hoppers Stamps Car doors Yum-Yums Things with the weird propulsion Dust clouds Hawks Propellers Twisters Twin fans Death spades Springs Elephants Brushes Toppers Forks Death spindles Whatevers Morning stars I think it is pretty long ago since I got past the toppers

Rich, can you do a little video clip on that, as you recently did? This is easier for me to get the issue.

I did not think of MESSUI because it is not in the main development process anymore; one of the developers picked it up when it was in a completely dysfunctional state. If it works for you, use it. And ... after a quick glance over it, it seems to me that it still is a Windows-only thing. SDLMESS users (builds for Linux and Mac) never had the MESSUI.

The name is ninerpedia. :-) I noticed the information there is outdated, as usual ... I moved to link from the MESS to the MAME download page.

Can you point me to the location of that link? Where is it on AtariAge? Rich, you are welcome to collect some text that you deem be suitable for starters. Maybe I'm too deep in it to see where the tripwires are. I actually tried several times to set up a starter's manual, but it always ends up in a much too long text, repelling all possibly interested people. It's as if you wanted to write a manual to a toolbox and explain all applications for a hammer, the pincers, the screwdrivers etc. Also, QMC2 is not really a user interface that makes it much easier; it simply arranges all the complexity in a proper window. For that reason I don't use it, I have my own start scripts in Linux, but I admit that people who want to stay away from command line operations should get some more help here. We have some plans for enhancing the OSD menu (ScrLock/Tab) to become a web service. For us this means we can build a TI-specific UI for MESS which could make handling much easier.

In turn, we have five different addressing modes and 16 registers. And a MPY and DIV instruction (ask C64 people ).

Exploring my feelings a bit deeper I admit I also have some sympathies for Hunt The Wumpus. :-) I hate other games. Wumpus does not even *try* to be sophisticated. It is effective in its own way. In contrast, there are games like Moonsweeper. It tries to be good but fails completely. The graphics may be a bit over average, but a gameplay of roughly a few minutes (having seen everything in that time) is really unacceptable. Follow-up is Buck Rogers.