+mizapf Posted June 15 Author Share Posted June 15 This looks like a shift of 1 (AA = 10101010 -> 01010101 = 55). This may have no effect* actually, but it is interesting to see why there is a shift at all. *The reason is that the 1 bits indicate a flux change, and this may come one position sooner or later. You can check this with my Java tool "AnalyseHFE" Usage: java AnalyseHFE <hfeImage> <track> <head> {"raw" | "cell" | "ocell" | "byte" | "sect" } ["noaddr"] [<outfile>] AnalyseHFE.zip 3 Quote Link to comment Share on other sites More sharing options...
rgjt Posted June 15 Share Posted June 15 I can't seem to get rid of this error when opening tiimagetool v3.02. Escape characters for unprintable characters should not be in the TI character set. Please review Preferences. Under Preferences, the Character conversion rules is all blank. I'm using openjdk version 17.0.11 64-bit under Debian Linux wormwood. There are no entries in the tiimagetool.log, all blank. With the version 2.4.7, I have no such error. 1 1 Quote Link to comment Share on other sites More sharing options...
+mizapf Posted June 15 Author Share Posted June 15 This is not the field that the error message refers to. Have a look at the "Output" tab. There is a field about unprintable characters. The former default was a period (.) which turned out to be a very bad idea, as it caused an escape sequence for each appearing period (in particular in BASIC programs, creating output like "DSK1..FILE"). I decided to better nag you with a warning than to let this default cause further problems. To fix that, pick a character from outside the TI character set. For example, you can use § or » or th etc. If you add a % character, this will append the hex code of the unprintable character behind that special character (like »81) 2 Quote Link to comment Share on other sites More sharing options...
F.G. Kaal Posted June 16 Share Posted June 16 (edited) 20 hours ago, F.G. Kaal said: Seems that some unwanted inversions is missing or active. Or the bits are all shifted some positions to the left or right Oh, you also mentioned this 🙂. It doesn't matter if en extra shift doesn't matter (what a strange sentence), it just is impossible to compare HFE images created with different tools. Going to play with the Analyze tool now. Edited June 16 by F.G. Kaal 2 Quote Link to comment Share on other sites More sharing options...
+mizapf Posted June 16 Author Share Posted June 16 Since I'm currently about to pick up my work on restoring the HFE write support in MAME, the discussion is indeed very welcome. 2 Quote Link to comment Share on other sites More sharing options...
F.G. Kaal Posted June 16 Share Posted June 16 2 hours ago, mizapf said: Since I'm currently about to pick up my work on restoring the HFE write support in MAME, the discussion is indeed very welcome. My wish (for months now) is to add simple HFE support (Creating empty HFE image and converting V9T9 DSK to HFE) to Ti99Dir. I was examining your java code (from a previous version of TIIMT) and translated that to C code to play with. I learned a lot from this exercise about interleave and skew and all the sync bytes before and after the data, I thought it was just a simple conversion, how wrong can I be. Then I noticed that the code for HFE support was changed in TIIMT v3.0.2. Also translated that to C code for Ti99Dir and then I spotted big differences in the HFE file data. So, I did a quick test this afternoon. I created an empty TEST.DSK and only changed the first 2 bytes of the first 18 sectors like so (track number, sector number); 00 02, 00 03 ... 00 08 and 01 00, 01 01, 01 02, ... 01 08 Test 1: Convert TEST.DSK to HFE with the HxCFloppyEmulator.exe, copied the file to a CF-card and placed this in the Lotharek floppy simulator on my Geneve and used DU2k to walk through the sectors. Everything okay here obviously. Test2: Convert TEST.DSK to HFE with Ti99Dir with the translated code from an older version of TIIMT. There where not much differences in the data if I compare the image created by HxCFloppyEmulator.exe and Ti99Dir. For as far I can tell the differences are beyond the track data (after >61B0) and before the next track data. Checked this file on my Geneve. Could read track 0, sector 0 but could never step to the next sector or read another sector. Oh Bummer. Test3: Convert TEST.DSK to HFE using TIIMT v3.0.2. Checked this file on my Geneve. Can read all tracks and sectors. I am a bit surprised with this result because of all the differences in the file data compared with the HxCFloppyEmulator.exe version but an extra shift in the data doesn't seem to matter after all, it just gives me a feeling of loose sand 🤔but anyway congrats to you ... good job Michael. I have still have to do a test 4, that is trying or just comparing the translated Java code of TIIMT v3.0.2 with Ti99Dir. If I did the translation correctly the files must be the same. But I also found out this afternoon that there is a command line interface that can be used with the hxfce.exe part of the HxCFloppyEmulator.exe. I am playing with the thought to just call hxfce.exe with the correct command line options. 3 Quote Link to comment Share on other sites More sharing options...
rgjt Posted June 17 Share Posted June 17 (edited) On 6/15/2024 at 1:30 PM, rgjt said: I can't seem to get rid of this error when opening tiimagetool v3.02. Escape characters for unprintable characters should not be in the TI character set. Please review Preferences. Under Preferences, the Character conversion rules is all blank. I'm using openjdk version 17.0.11 64-bit under Debian Linux wormwood. There are no entries in the tiimagetool.log, all blank. With the version 2.4.7, I have no such error. Deleted the period in Replace unprintable characters by under the Output Tab and that solved the error message. Edited June 17 by rgjt Additional comment 1 Quote Link to comment Share on other sites More sharing options...
rgjt Posted June 17 Share Posted June 17 On 6/15/2024 at 1:34 PM, mizapf said: This is not the field that the error message refers to. Have a look at the "Output" tab. There is a field about unprintable characters. The former default was a period (.) which turned out to be a very bad idea, as it caused an escape sequence for each appearing period (in particular in BASIC programs, creating output like "DSK1..FILE"). I decided to better nag you with a warning than to let this default cause further problems. To fix that, pick a character from outside the TI character set. For example, you can use § or » or th etc. If you add a % character, this will append the hex code of the unprintable character behind that special character (like »81) Got it resolved now. Thanks. 1 Quote Link to comment Share on other sites More sharing options...
+mizapf Posted June 17 Author Share Posted June 17 Actually, it took me some time to find out why BASIC listings had all these periods doubled (like OPEN #1:"RS232..BA=9600..DA=8"), until I figured out that this was an unplanned effect of the special character treatment. Quote Link to comment Share on other sites More sharing options...
+Schmitzi Posted June 18 Share Posted June 18 12 hours ago, rgjt said: Deleted the period in Replace unprintable characters by under the Output Tab and that solved the error message. There was no period in this field, but after 100 tries I recognized that there was a space there, so removing this space solved the error for me at startup thx 2 Quote Link to comment Share on other sites More sharing options...
rgjt Posted June 18 Share Posted June 18 4 hours ago, Schmitzi said: There was no period in this field, but after 100 tries I recognized that there was a space there, so removing this space solved the error for me at startup thx That error was a head scracther for me until mizapf cleared things up. Anyway, the latest version of tiimagetool is a fantastic piece of sofware and the goto tool for any TI enthusiast especially as I work almost solely under a Linux environment. 2 Quote Link to comment Share on other sites More sharing options...
F.G. Kaal Posted June 18 Share Posted June 18 On 6/16/2024 at 8:29 PM, F.G. Kaal said: I have still have to do a test 4, that is trying or just comparing the translated Java code of TIIMT v3.0.2 with Ti99Dir. Found a bug in my .DSK read function (was always reading sector 0) but this didn't fix TEST2 but TEST4 was a success. Using the to C converted Java functions from Micheal's TIIMT 3.0.2 resulted in a HFE image that was almost the same as the image created by TIIMT 3.0.2. The only difference was in the areas with unwritten data in the HFE image. In the HFE image created by TIIMT 3.0.2 these where all 0x00 and in my C version these where all 0xCD which just means uninitialized data in the debug Visual studio environment. Just need to clear the whole track array after doing a malloc. Micheal's work is looking really good but being a lazy programmer I am going for the command line options of the hxfce.exe program. With this method I can add more .DSK formats and have less testing to do . 3 Quote Link to comment Share on other sites More sharing options...
+mizapf Posted June 18 Author Share Posted June 18 Thanks, I'd just like to do a SWPB on my vowels, it's really just Michael. As far as I know, in Irish it is actually Mícheál. 1 Quote Link to comment Share on other sites More sharing options...
+dhe Posted June 18 Share Posted June 18 It might be good to update the release on github. 1 Quote Link to comment Share on other sites More sharing options...
+mizapf Posted June 18 Author Share Posted June 18 You are right, I did not bother about the release packages. The source code is updated to 3.0.3, though. I'll put the 3.0.3 release on Github soon. 1 Quote Link to comment Share on other sites More sharing options...
+dhe Posted June 18 Share Posted June 18 I release it's not that easy also... With Java 8, or with something recent! Quote Link to comment Share on other sites More sharing options...
+mizapf Posted June 20 Author Share Posted June 20 I just published the new release 3.0.3. 3 Quote Link to comment Share on other sites More sharing options...
chris36 Posted June 21 Share Posted June 21 On 6/18/2024 at 1:18 PM, F.G. Kaal said: Just need to clear the whole track array after doing a malloc. calloc() Quote Link to comment Share on other sites More sharing options...
F.G. Kaal Posted June 21 Share Posted June 21 On 6/18/2024 at 8:03 PM, mizapf said: Thanks, I'd just like to do a SWPB on my vowels, it's really just Michael. As far as I know, in Irish it is actually Mícheál. Haha .. sorry but just checked to be sure. The directory on my harddisk with your name got the correct order of all the characters. Somehow my vingers got lost on the keyboard 2 Quote Link to comment Share on other sites More sharing options...
+Schmitzi Posted June 22 Share Posted June 22 Hi, I downloaded the 3.03 master.zip from Github but there is no "tiimagetool.jar"-file inside as mentioned in the install instructions ? 🤔 Quote Link to comment Share on other sites More sharing options...
Nick99 Posted June 22 Share Posted June 22 1 hour ago, Schmitzi said: Hi, I downloaded the 3.03 master.zip from Github but there is no "tiimagetool.jar"-file inside as mentioned in the install instructions ? 🤔 I just downloaded it from Github and it has the .jar-file in it. 2 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted June 22 Share Posted June 22 5 hours ago, Schmitzi said: Hi, I downloaded the 3.03 master.zip from Github but there is no "tiimagetool.jar"-file inside as mentioned in the install instructions ? 🤔 @SchmitziGo to https://github.com/mizapf/tiimagetool/releases/tag/v3.0.3 and download it. the file is there. 1 Quote Link to comment Share on other sites More sharing options...
+MarkB Posted June 22 Share Posted June 22 On 6/18/2024 at 7:03 PM, mizapf said: Thanks, I'd just like to do a SWPB on my vowels, it's really just Michael. As far as I know, in Irish it is actually Mícheál. Yep it is 🙂 Quote Link to comment Share on other sites More sharing options...
+mizapf Posted June 22 Author Share Posted June 22 3 hours ago, khanivore said: Yep it is 🙂 And that's because if there's an I left from the CH, you can't have an A right from it, right? Quote Link to comment Share on other sites More sharing options...
+MarkB Posted June 22 Share Posted June 22 1 hour ago, mizapf said: And that's because if there's an I left from the CH, you can't have an A right from it, right? Ooh, you're testing my memory there. Yes now I remember the rule from school "caol le caol, leathan le leathan". You can't have a slender vowel (i or e) on one side of a consonant group and a broad vowel (a, o or u) at the other side so yes it has it be e first after the ch. 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.