PPCenter :: devblog

PPcenter. Arts and craft for my Sega Saturn. Since 1847 !

Pseudo Saturn Kai 6.074 released !

Written by cafealpha 2 comments
Pseudo Saturn Kai 6.074 released ... more than three months ago  Sorry for the late news, but that's still some quite acceptable delay in comparison to frequency I post articles in this blog

The release was initially announced on SegaXtreme forums, because it is my favorite forums for Saturn things (more than 11 years I registered to it !
Big thanks to the forum moderation/maintenance/support team for their steady good work ). Also the news was announced on SegaXtreme first ... because writing an article there takes less time than writing an article on my blog
So since I'm a lazy person, let's slightly arrange this article from what I wrote on SegaXtreme forums !


What's new in this version ? Well, there's basically, not a lot of major features (sorry !), but fix of a critical bug affecting Gamer's Cartridge, and addition of some "minor, but that I wanted to add someday" features.

______________
Features for Gamer's Cartridge users

Indirect Dump
: Dump cartridge (boot cartridge, or official memory cartridge, or any other cartridge) by compressing its contents to internal backup memory, and then extracting it to SD card when Gamer's Cartridge is inserted.
Because of small size of internal backup memory, it may be required to repeat the "insert memory cartridge → Indirect Dump → Turn off Saturn → Insert Gamer's Cartridge → Indirect Dump → Turn off Saturn" cycle several times, but at the end cartridge is dumped
Requires some motivation and free time on user side (It took me 45 minutes in order to dump my good old very first official memory cartridge), but this way of dumping doesn't requires to unsolder cartridge memory chips, which is the most important for me !
Note : if you are the lucky owner of Saturn FDD, then floppy disc is used instead of internal backup memory, and consequently dump is significantly faster and simpler.


Left : Screen display after compressing dump data
to internal backup memory
Right : Screen display after writing dump data from
internal backup memory to SD card
Theses pictures were taken on Indirect Dump's first successful test

One extra application of this feature was to dump ROMs from legacy cartridges such as X-Terminator, Satellite, ST Key and "8MEG memory cartridge". Curious people wanting to test theses cartridges theses ROMs can found them available for flashing (at your own risks) in Pseudo Saturn Kai Save Data Manager.
Special thanks to A Murder Of Crows, AtariBorn and SaturnGuru who dumped their cartridges ... and reported many bugs from early versions of Indirect Dump feature 

 
Indirect Dump development : the cut scenes
Left : dump data corruption (see below for details)
Right : Dump file CRC check error. Fortunately, only dump
computation routine was buggy, and dump feature itself was working


Save Data corruption bugfix : This corruption problem concerns saves on Saturn internal memory, with save data size making save sector allocation table finishing exactly at the end of a sector. Very few games are concerned, and the only example I could find so far is Albert Odyssey.

I found this problem when testing Indirect Dump : data compressed then written to internal save memory wasn't then correctly read and decompressed to SD card.
I initially thought this problem to be caused by Indirect Dump itself, because at the time I had this problem, I just finished to mess with data alignment error when preparing indirect dump's compressed data, so thought that this was the next bug to fix regarding Indirect Dump
Quick investigation showed that contents of read data wasn't as expected, with a small portion of data being same as data at unrelated offset ... so maybe a problem when compressing and adding header etc to Indirect Dump save data ? It took me then a around one hour of testing and code review until I was sure that culprit was actually not Indirect Dump itself, because re-read of save data didn't matched with written save data !
Then, testing Indirect Dump with same input data and virtual memory cartridge disabled worked as expected, which 99% confirmed that problem was located on virtual memory cartridge side !

So yes, "virtual memory cartridge" actually manages save data handling for all backup devices (internal save memory, memory cartridge, and even Saturn FDD too) : this is possible because Gamer's Cartridge hooks Saturn backup library at relatively high software level.
In comparison with 1:1 hardware emulation of save memory cartridge, this has the advantage of being free of capacity limitation inherent to memory cartridge's chips, and also allows to store save data individually on SD card (two files per save, vs big file containing all saves for 1:1 cartridge emulation). On the other side, all the save data handling software has to be re-written, so care has to be taken when re-implementing it

So, this is quite an uncomfortable bug I have to face : save data are corrupted when being written to internal save memory, and Gamer's Cartridge was released around 6 months before, with consequently cartridges potentially corrupting save data in the wild.
Fortunately, while being hard to find, the bug wasn't hard to fix  Basically, this was a timing error when updating current save data block ID, as described below :


Internal Save Data corruption related source code
Left : before fix
Right : after fix



Since I really didn't want to fix half of the bug (and having to fix the remaining half on other release), next step is to add a test bench tool for internal save data memory access !
"Test bench" is a program that automatically tests a lot of patterns for a given function or module. Ideally, all major modules should have their test bench, but in the real world, making test bench programs takes time, so programs are usually manually tested with 2 or 3 patterns, and then developer prays for all other patterns being bug-free 
... And to tell you the truth, this test bench was the largest one I wrote so far  It's not impossible a bug to be still hidden in an untested pattern, but anyway I'm quite confident about the reliability of internal save data support now 

Here is how test bench screen looks like on my dev environment :

SatLink and Yabause couple showing test bench screen.
SatLink and Yabause : my best friends when testing Saturn programs


Screen copy above was taken after fixing internal save data support, and I forgot to take a similar screen copy when bug wasn't yet fixed ... latest Save Data Manager CDROM's + Gamer's Cartridge with firmwave 6.037 should display error in test bench screen, so it would be funny to try it ... maybe someday 
BTW, Test Bench is still available in Save Data Manager CD-ROM (it took me a bit of time to add test bench screen, so didn't wanted to remove it after fixing internal save data corruption bug !), but this test bench screen requires a trick in order to be accessed : did you found it ?

That's all about internal save data corruption : I fixed it, and checked a lot of patterns about internal save data access, so similar problem shouldn't happen in the future !
Additionally, I didn't received any feedback from users about this bug, so let's hope it didn't corrupted game saves elsewhere than on my dev laptop



Autodump : Automatically dump saves from internal backup memory to SD card, so that they can be restored in the case they are lost because of depleted CR2032 battery, or save data manipulation error, etc.
With this feature, you won't have to say anymore "I lost all my saves because of that $%QTY@!#% CR2032 battery" 
Since this feature uses SD card, it is available on Gamer's Cartridge only. And no, I won't write save data into Action Replay's cartridge flash ROM : flash ROM chips are not designed in order to be written frequently, hence using flash ROM as read/write storage area may prematurely brick your Action Replay cartridge.
If you're not convinced about flash ROM limited write cycles, I can make some "cartridge self destruction" feature that would continuously write and erase random data to flash ROM, but don't claim that your cartridge no longer boots after that


An alternate solution in order to keep internal save data change the chip holding internal save memory to a FRAM one that doesn't requires battery to retain saves data, as detailed in db-electronics.ca.
(there are maybe other alternate solutions, so please let me know in the comments if I missed one)

The advantage of this FRAM mod is that saves are not lost when CR2032 battery is depleted. There are however several disadvantages :
 - This is an hardware mod, which requires to remove a chip, then solder another one, hence quite difficult to do.
 - CR2032 battery is still required in order to keep clock and language informations.

For being the developer of Pseudo Saturn Kai and Gamer's Cartridge, my opinion is a bit biased, but autodump is a better solution, because autodump is not an invasive mod (plug cartridge vs change chip on Saturn motherboard), and the FRAM solution doesn't allows to dump save data to modern media device. Anyway, I don't force you to like my projects, so if you don't like autodump, then don't use it


Virtual Floppy Disk Drive. Simulate Saturn Floppy Disc Drive (FDD) behavior, with save data stored on SD card.
This feature Worked fine with Dezaemon 2, which can now handle 10 saves at once (5 saves from cartridge, and 5 other saves on FDD ), but I didn't tested for other games using FDD, so any feedback is welcome.
This is a feature available on Gamer's Cartridge only too. Thanks to this feature, floppy disk access is available without having to plug any floppy disk drive to Saturn 



______________
Features for Action Replay and Gamer's Cartridge users

Proper support of FDD in Save Data Manager : Save Data Manager now supports both two partitions of Saturn floppy discs for all common operations such as copy, move, delete, import, dump, etc.
Special thanks to Dezaemon DB for big assistance in beta testing


Backup devices informations from Save Data Manager :
First device is internal backup memory : the easiest to handle
Second device is backup memory cartridge,
unconnected (zero partitions detected) in this picture
Third device is FDD : two partitions, wrong unit ID returned,
and always set at third position in devices information
array, even when cartridge is not available.
Basically, FDD is a real trouble-maker when trying to handle backup devices



Save Data Manager development cut scenes : trying to format FDD
(format for both two partitions is displayed in a single picture)
Don't worry about the horrible display of error
messages : it's (finally) improved in version 6.074


Soft Reset Patch
: Change the "A+B+C+Start key combo" soft reset from "exit to multiplayer screen" to system reset. This idea is not from me, but from neuroacid (rmenu developer).


More saves to import from Save Data Manager
. The most noticeable addition is saves from Urawaza Dataro cartridge, containing quality saves for Japanese games. Special thanks to Madroms for dumping the cartridge, and providing me its dump file 


Other
: Fixed many bugs, and added other bugs ... the usual routine 


(Not a software feature, but) future users of Gamer's Cartridge may appreciate the couple of new labels, designed by beebaraka-sensei (twitter) for artwork itself and Dezaemon DB (twitter) for the final editing of the labels. Many thanks to them for the very good work !


Dezaemon-themed labels ... could you find out the odd label ?

The design on theses labels is themed on Dezaemon 2, and as you can see, the Dezaemon character on the game cover grew in the right way 
I personally do appreciate the white themed one





Before being asked "where are the cheat codes ?", please let me reply :)
Cheat codes support is still under development. Menu in order to select game and its cheat codes is 90% done, but the bit of code in order to apply codes is still TBD. I didn't touched this feature for around half a year mostly because of lack of free time, motivation, and being busy in developing other features. I plan to finish this for "next-next release" however.
Update three months after writing the explanation above : cheat codes on Pseudo Saturn Kai are working on few games !! 
There are many things remaining until reaching an "OK to release" state, but at least the cheat codes core feature seems to work, which is quite a relief for me :)

By the way, there's a trick cheat codes selection menu in Pseudo Saturn Kai 6.074 : did you found it ? It won't do anything else than showing games list and cheats for each games, so it's not worth spending more than 10 minutes in looking for the appropriates key(s) combo and timing however.


Yes, this release have (at least) two Easter eggs available Coincidence with release timing during Easter period ? I think not


Apart from cheat codes, there are many other things scheduled for future releases : I initially planned to release everything at once, but had to make this intermediary release in order to fix the save data corruption bug on Gamer's Cartridge, so stay tuned


PS : Special thanks to Stac for beta testing  All the features above wouldn't exist without his help !!!

Read more Pseudo Saturn Kai 6.074 released !

Bit shift in Saturn minimalistic C program

Written by cafealpha 4 comments
While making a Saturn program running directly from ROM, constraint is usually to have program as small as possible. Why ? Because I would like to make Pseudo Saturn Kai continue supporting Action Replay, which have limited ROM size

So, in that situation, I use homemade, minimalistic stdlib, containing usually only memcpy and memset functions, and everything goes well.
But for some reason, current program was begging for __ashlsi3 and __ashrsi3 functions.
... What's these things ? __ashlsi3 is an assembly function doing left shifts of data, and as you probably guessed, __ashrsi3 does the same, but with right shifts

No problem, let's ask my friend google, and copy & paste theses functions without minding about their contents Theses functions are standard functions provided by C library, so sources can be found here and there.


An example of what __ashrsi3 function looks like

A day after that, here comes the time to do basic testing of this ROM code
Let's first start with simple things such as screen display ...


Text display ... not as expected ?!

Hrm, it's probably a problem in my printchar routine ?

printchar source code : this is the core thing
behind text display in my programs


But that's a very long time I didn't touched this routine, which works well on Pseudo Saturn Kai, and works even better in vdp1ex from Charles MacDonald sample programs, because all my printchar routines come from there
But just in case of, let's try to tune one " <<3 " into " *8 " in the code above, and see if that changes something. Theses two operations are equivalent, so this should give the same result, but ... Saturn screen becomes all black

Okay, there's definitely something weird regarding that printchar, but I just re-used it as-is from other project, so what's going wrong with my it ?!
[...]
And then (finally) a light-bulb above my head appears Maybe there's something wrong regarding the __ashlsi3 and __ashrsi3 routines ?
And, yes there was something wrong regarding theses routines ... I wrote a paragraph about them on the first half of this article, so it would be surprising the problem came from somewhere else

In more accurate words, routines are OK, which is normal since they are written by people smarter than me But the calling convention of __ashrsi3 was not as my gcc was expecting ?
Well, that's just a guess from the display results, because it seems to display the same pixel 8 times on each row of a given character, which may happen when ignoring shift parameter in __ashrsi3 function.

Anyway, rather than finding in details what's wrong, let's fix the problem First, with assembly by hand, in desperately trying to change input parameter handling ... I don't remember exactly what I wrote, but that was something like pushing a register to stack, moving parameter to it, finally restoring register, putting nop instructions everywhere, etc ... nothing difficult, but over my extent in assembly programming

The result is ... a mixed success :

Trying to modify __ashrsi3 routine by hand ... well, it seems that
some shift values are not correctly handled


Okay, so let's re-ask google about that __ashrsi3 routine ... and a different implementation arrives in search results. Let's try it


__ashrsi3

And the execution result :

Execution result with second version of __ashrsi3 function

Yeah, it works this time !!! And don't ask me why, because I didn't took time in understanding what's different in that second version of __ashrsi3 function
And, rather than understanding why it works ... let's close that assembly pandora box before some other mystic bugs pop from it

I will probably see about this in the future, but that was absolutely out of the scope of today's programming, and in software world, too much digression leads to freeze of projects, which I would like to avoid in order to go forward to next pending project 

As a example, I started to adapt yeti3d engine to Saturn in 2010 (7 years ago !), and this ended in ... developing a SD card based memory cartridge for Saturn !
The point above is is not a joke : menus used in my yeti3d adaptation are the origin of menus used in Pseudo Saturn Kai, which shows the continuity (co-consanguinity ? ) in my projects. And after getting yeti3d working a bit, I really would like to load levels from something else than CD-ROM, which was from PC (via adequate link cable) on a first time, and which will (should ?) then evolve to SD card.
That would be cool to make a Saturn game not requiring to burn CD-ROM somedays, but before that I need to finish neighboring side quests


Evolution of my Saturn projects

To conclude this article, let's say it was terribly fun to see some bits of the internals behind C language
By the past, I remember I did something similar with my yeti3d adaptation, but remained at "C language level" : optimization done at that time was (IIRC, ) to avoid 4 bits shifts, because theses are not available in a single SH-2 CPU instruction. Avoiding 4 bits shifts was simply done by merging two fixed point operation in a single one (or something like that : I did this 7 years ago !), and this actually gave some improvements in 3D scene rendering Theses were the good times ... I sometimes think "I'm Getting Too Old For This Shit", but that's only to motivate to finish my old projects

But before that, Pseudo Saturn Kai, and then Kicad and Quartus are waiting for me


BTW, in the case you wondered about what was today's programming session, you probably guessed it was about exception handler addition to Pseudo Saturn Kai. This not a new feature, since it was available in Pseudo Saturn 0.83x ... in fact, I grabbed some sources from Pseudo Saturn 0.83x in order to implement this exception handler (thank you CyberWarriorX !) This exception handler will be available as a small add-don to cheat codes feature in Pseudo Saturn Kai next major release.

Read more Bit shift in Saturn minimalistic C program

Cartridge labels available !

Written by cafealpha no comments
Finally available :]

Last June, I initially wanted to release Saturn Gamer's Cartridge with custom label, but some unexpected problems regarding label quality (not resistant to water) delayed its release, consequently, Gamer's Cartridges manufactured until now had only shell, but no label.

"Stronger" label retained for use by Gamer's Cartridge is designed for car tuning purpose, so waterproof, and even alcohol proof ! I suppose this will satisfy the requirements of typical Gamer's Cartridge user
(※) Just a (common sense) note : while labels are waterproof, cartridges themselves aren't  So no, you still can't play Saturn while taking a bath

data/images/20160914_labels.tb.jpg
Cartridges with their labels.
Bottom-right one (without LEDs) is USB dev cart.


Custom labels were the last main milestone regarding Gamer's Cartridge design : I won't design custom boxes for my cartridges.
Why ? Because I want Gamer's Cartridge users to ... use their cartridges, not collect (shelf) them

You are Saturn Gamer ? Want to share your Saturn saves with your emulator or friends ? Want to enjoy Dezaemon 2 game but don't want to purchase a modchip + memory cartridge ? Then Saturn Gamer's Cartridge is for you


Special thanks to my friend Darius who selected paper, printed labels sheet, and cut each labels
Darius provided pictures he took when cutting labels, so please enjoy the labels making-of


data/images/20160818_labels_scan.jpg
Label sheet scanned just after print-out.
Red/blue lines are used as help when cutting labels.



data/images/20160819_labels_tools.jpg
Tools used : ruler and cutter.
And also Darius' God Hand and eyes (not pictured)


data/images/20160819_labels_row_cut.jpg
In order to optimize time , labels are first cut row per row.

data/images/20160819_labels_individual_cut.jpg
After row cutting, labels are cut individually

Read more Cartridge labels available !

Could finally write a file to my SD card :]

Written by cafealpha 3 comments
I could finally write a file to my SD card ... from my Saturn
Well, I just tested by dumping my Saturn's BIOS, and didn't tried further, because I don't want to find and fix other bugs today

data/images/20150716_bios_dump_ok.tb.png
Top : my Saturn BIOS, dumped from my Freewing device 5 years ago
Bottom : BIOS file in my SD card dumped today
Hex Editor says that both files are identical

This fixes a nearly 3 years old bug
At that time, I connected a SD card to my second pad port on my Saturn and could read files successfully. However, file write sometimes failed ... I thought about many possible causes of failures (not enough 3.3V power supply ? Bad luck ? etc etc), but it was a software bug : writing sectors to SD card actually takes longer than reading sectors, and I didn't waited enough for my SD card to finish write.
Changing poll timeout count from 500 clocks to 50000 clocks seems to fix the problem. (Q&D fix, but if it works fine, I won't change it )

So now, as SD card read/write access works on both yabause and real hardware, it is time to to use it in order to develop new features for my Saturn

data/images/20150716_memcart_rev2f.tb.jpg
My SD card cartridge for Saturn (Q&D prototype revision )

Edit : wow, current cartridge firmware build ID (incremented on each make/build) is 4200

Read more Could finally write a file to my SD card :]

Rss feed of the tag