Marius Posted August 31, 2015 Author Share Posted August 31, 2015 I use the 1993 version a lot. I do my professional business administration with it. Works perfect, and it's cool. I wished I had the 1985 version though... (that's the one that is officially compatible with extended memory, and it has bug fixes). 1993 version is a hacked version of 1983 version. Works very well, but the 1983 bugs are still in it. Quote Link to comment Share on other sites More sharing options...
+JAC! Posted August 31, 2015 Share Posted August 31, 2015 From what I found it's not really a hack (replacing/padding some location/areas), but a regularly compiled different version with lots of different code for the memory handling. Axlon support is gone (who would have bought an Atari 800 in 1993 :-) ) and the support for the XE banks is quite different, too. 1 Quote Link to comment Share on other sites More sharing options...
luckybuck Posted August 31, 2015 Share Posted August 31, 2015 Hi together, a few minutes ago I had all information together for the 1985 version with(!) extended memory. It will be available at the end of October. So stay tuned. :-))) 5 Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted September 1, 2015 Share Posted September 1, 2015 Way to go luckybuck, long time no see! then when we do see you come bearing gifts, the best way to make an entrance! 1 Quote Link to comment Share on other sites More sharing options...
luckybuck Posted April 27, 2017 Share Posted April 27, 2017 The actual status can be found here: https://atariwiki.org/wiki/Wiki.jsp?page=Synapse%20SynCalc 1 Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted April 28, 2017 Share Posted April 28, 2017 nice follow up! what a pair... 1 Quote Link to comment Share on other sites More sharing options...
+DjayBee Posted October 14, 2017 Share Posted October 14, 2017 Find attached a crack made from the 1983 ATX in Farb's torrent. It contains two versions of the program: Standard (incl. Axlon support) and 52k RAM (without any other extended memory). The only "real protection" is a weak sector check for sector 88. It loads the sector twice to RAM locations X and X+$80 and verifies that they are not the same. The remainder is the above mentioned state machine which loads seemingly random sector numbers to seemingly random RAM-locations. This way many of the duplicate sectors get loaded. All these tracks have 21 good sectors. The state machine uses five out of the last 16 bytes of each sector (16-bit values are hi/lo byte) as "linkage": Bytes 1-2 are EORed against a zero page location to define the target RAM-address for this sector (usually only $70 bytes are used). Bytes 3-4 are EORed against the current sector's number to find the next sector number to load. Byte 5 is the offset into the next sector of the next 5-byte "linkage" block. The "linkage" offset of the very first sector of a file is always $78 for the titles I have cracked so far. The crack was done by: Loading the weak sectors to RAM locations X and X+$7f. This way everything behind the stable bytes at the sector's beginning is different.This is 1-byte change for each of the two loaders. Reassembling all needed sectors in a nearly sequential way.The first sector of each of both "files" must stay at its original location (93 for standard and 291 for 52k). Everything else is defined by the "linkage". SInce the original uses lots of duplicate sectors you need lots more space on a standard disk with only 18 sectors per track. I currently have only the 1983 version done. The 1985 version will soon follow. It seems to have the same two files as the 1983 version plus one more for the XL version. Due to the third version it uses a lot more space on disk and will not fit on an SD disk in cracked form. To someone who uses Syncalc: Please verify if the crack really works correct. SynCalc (1983)(Synapse Software)(US)cr.atr 10 Quote Link to comment Share on other sites More sharing options...
+DjayBee Posted October 14, 2017 Share Posted October 14, 2017 This was faster than expected. Find attached the 1985 version. The XE version's first sector is 525. SynCalc (1985)(Synapse Software)(US)!cr.atr 6 Quote Link to comment Share on other sites More sharing options...
+DjayBee Posted October 15, 2017 Share Posted October 15, 2017 (edited) Updated 1985 version. In the first release I accidentally overwrote the Visicalc importers when moving sectors around. SynCalc (1985)(Synapse Software)(US)!cr.atr Edited October 15, 2017 by DjayBee 11 Quote Link to comment Share on other sites More sharing options...
tep392 Posted October 15, 2017 Share Posted October 15, 2017 Nice work! Syncalc shows 245k available on my 800 with Axlon RAM card. 2 Quote Link to comment Share on other sites More sharing options...
+Nezgar Posted October 17, 2017 Share Posted October 17, 2017 The only "real protection" is a weak sector check for sector 88.It loads the sector twice to RAM locations X and X+$80 and verifies that they are not the same. Would this be the same as a duplcate sector? I've seen disks with two sectors with same number, but each containing different data. i.e. If placed 9 sectors apart, consecutive reads of the same sector will alternate two sets of returned data. Maybe you can see better from the ATX - if it is a weak sector, every single read will return different data, not just two different sets of data. Nice work! Syncalc shows 245k available on my 800 with Axlon RAM card. Now to try it with Antonia in 4MB Axlon mode. Quote Link to comment Share on other sites More sharing options...
+DjayBee Posted October 17, 2017 Share Posted October 17, 2017 Would this be the same as a duplcate sector? It is a weak sector which returns different data each time you read it. But since it is read only twice the effect is the same as with a duplicate sector. Now to try it with Antonia in 4MB Axlon mode. I tried in Altirra in Axlon mode: It uses max. 256 kB. 2 Quote Link to comment Share on other sites More sharing options...
ijor Posted October 17, 2017 Share Posted October 17, 2017 It is a weak sector which returns different data each time you read it. But since it is read only twice the effect is the same as with a duplicate sector. A weak sector not only returns different data each you read it, it reads different data even when reading the same physical sector. It is not too difficult to read a weak sector just twice and verify that you are reading the same physical sector. But doesn't seem to be the case here, a dup sector on this track would fool the protection check on the weak sector. It won't help too much here, as there are still other tracks with advanced protection. Btw, great work as usual DjayBee Quote Link to comment Share on other sites More sharing options...
+DjayBee Posted October 18, 2017 Share Posted October 18, 2017 It seems that Synapse had several versions/releases of every productivity title with different protection. Find attached: two similar versions of SynFile+ (probably v1.00) - same protection as SynCalc - from luckybuck and me SynChron - same protection as SynCalc - from Farb's torrent SynFile+ v1.02 - protected only by a bad sector - from Allan Farb's torrent contains two more dumps: SynFile+ v1.01 and v1.02 which have no protection at all. If someone owns other protected productivity software from Synapse then please contact me. SynFile+ v1.00 (1983)(Synapse Software)(US)cr.atr SynFile+ v1.00 (1983)(Synapse Software)(US)acr.atr SynChron (1983)(Synapse Software)(US)cr.atr SynFile+ v1.02 (1983)(Synapse Software)(US)acr.atr 6 Quote Link to comment Share on other sites More sharing options...
+DjayBee Posted October 22, 2017 Share Posted October 22, 2017 The "Synapse 21 spt DIY toolbox". Synapse21-Documentation.zip Synapse21-100.atr 7 Quote Link to comment Share on other sites More sharing options...
luckybuck Posted October 24, 2017 Share Posted October 24, 2017 WOW! Please let's talk on Saturday about this. :-) Giga-Thanks! Quote Link to comment Share on other sites More sharing options...
+JAC! Posted October 25, 2017 Share Posted October 25, 2017 The write-up is just mad stuff :-) Potentially dumb question: What does "spt" stand for ? 1 Quote Link to comment Share on other sites More sharing options...
+Nezgar Posted October 26, 2017 Share Posted October 26, 2017 Sectors Per Track? 1 Quote Link to comment Share on other sites More sharing options...
+DjayBee Posted October 26, 2017 Share Posted October 26, 2017 Yup Quote Link to comment Share on other sites More sharing options...
mechanerd Posted December 9, 2017 Share Posted December 9, 2017 Mr. Xerox could do it. .... A long time ago in a silicon forest far away..... Quote Link to comment Share on other sites More sharing options...
+DjayBee Posted May 31, 2021 Share Posted May 31, 2021 Up to recently my little toolbox could only follow the "files" on Synapse disks and output the sequence of their sectors. As part of my effort to create a file-version of Syncalc I refined it. It now writes a second file in CSV format which for each sector also includes the number of actually used DATA bytes and where in RAM it will be loaded. The ATR contains all source codes and a readme.txt. Synapse21-102.atr README-from-ATR.txt The CSV file can then be sorted by the RAM address and voilà, you have an exact list of memory blocks used by the title. Inside the below ZIP archive you can find the output for the 1983 and 1985 Syncalc releases (sc8##.CSV), as well as files sorted by RAM address and showing the start and end of the full memory blocks loaded (sc8##-sorted.csv). The last digit of the filenames is the on-disk number of the file (1=48k/Axlon, 2=52k, 3=130XE). Synlist-Syncalc.zip The CSV's fieldnames are: NUM reading sequence SEC sector number on disk OFFSET to linkage (=# of used data bytes) ADDRESS RAM target address END RAM end adress of this sector HEX same in hexadecimal HEX END BLOCK identifies beginning and end of continuous memory block in hex I did not write a tool to dump the program code directly from disk, but instead set a break point in Altirra to the location when everything has been loaded and decrypted and issued .writemems to extract the data. The memory blocks for the 48k/Axlon and 130XE versions are identical: $0700-$130C patched DOS2.0S $1A00-$1AFF loader $1E00-$2E04 main code part 1 $8000-$C020 main code part 2 (yes it loads until $c020) The 52k version uses a different layout: $0700-$1345 patched DOS2.0S $1A00-$1B1E loader $1E00-$6E5C main code All versions load the VC/SC converter to $7800-$7DFF. Having these lists, I found out that my 2017 documentation of the sector "linkage" from above is not correct. Correct documentation: The state machine uses five bytes of each sector (16-bit values are hi/lo byte) as "linkage". This block comes right behind the actual data bytes of the sector. Bytes 1-2 are EORed against a zero page location to define the target RAM-address for this sector. The inital value of this zero page location differs from title to title and is needed for my tool in decimal. Bytes 3-4 are EORed against the current sector's number to find the next sector number to load. Byte 5 is the offset into the next sector for the next 5-byte "linkage" block. The "linkage" offset of the very first sector of a file is always $78 for the titles I have cracked so far. Most sectors contain 112 or 123 data bytes and then the "linkage" really can be found within the last 16 bytes of the sector as stated in 2017. But there are also sectors with down to only 4 bytes of data. 7 3 Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted June 11, 2021 Share Posted June 11, 2021 Nice job as usual, can't wait for it all to come together... The end all be all of the syn series compilation may be on the horizon. Love your work and diligence 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.