Software emulation 1818VG93
ZXNet echo conference «real.speccy»
dedicated to remaking TR-DOS with the creation software emulation of VGshki.
I propose to discuss. Tell me who and what you think about this, and I’ll read...
Maybe I’ll answer... :o)))
--
Timonin Maxim
TEXT:
=======================================
BACKGROUND
It all started with the fact that I became the owner of a Spectrum, but not the usual ZX-128
with a disk drive and AY, and ATM-turbo 2+ with a bunch of additional bells and whistles, such as advanced graphics, megabytes of memory, a more flexible management structure of the address space, a PC keyboard controller and, very importantly, a built-in IDE HDD controller. And, what is also important, I had software for these bells and whistles in a significant amount, mainly for the CP/M operating system. But time passed, and from the software I mainly added only what was written for TR-DOS. The stack of floppy disks in the box in front of the computer grew (I
actively used the Internet in search of new products for Specky), and no matter how hard I tried
to arrange and classify them, every major gathering with my beloved
ZX led to the formation of a shapeless pile of 5.25" and3.5" floppy disks with games, systems and demos on the desktop, in which one could drown, to put it figuratively. This was all the more offensive since more than forty floppy disks of software for CP/M had already been transferred to the hard drive and conveniently located there. But
such a pleasant prospect awaited, at best, only a couple of dozen more floppy disks with iS-DOS, and even then only if I get the appropriate driver somewhere. The severity of the problem will not decrease at all. It was necessary to do something, especially since information began to arrive that on Scorpio, with the help of the SMUC card and the “shadow”, this problem has already been largely solved. But what is good for Scorpio is not always good for others. Spectrums, and even Zones with MOA tried their best in this... :o)
I had no
desire to turn my Spec into Scorpio-compatible (namely, this method
connecting SMUC to other ZX was offered by MOA), and in the case of ATM it was almost technically impossible,
Scorpio and ATM-2+ have too different architectures (in which The ZX architecture is only
one of the possible ones).that, for the time being, dreams remained dreams. The situation began to change after
I began to establish contacts with other active spektrumists and, above all, with ATM players, which was greatly facilitated by the creation of the ATM support site.
BIRTH AND DEVELOPMENT OF THE PROJECT
By establishing contacts with other spektrumists, I looked for like-minded people,
the first of whom were ART and T!M0N. In correspondence with them, the contours of a project designed to solve this problem began to emerge. Later
ART, due to changes in life (graduation from college, etc.), temporarily disappeared from view, and T!M0N and I, having agreed on the main positions, began to slowly
work on the project. The original plan was to repeat the SMUC scheme, but without using a shadow engine - that is, the screw had to be divided into TRD images, sorted into subdirectories. The main alteration, of course, had to undergo TR-DOS, which had to be made to work with the hard drive, at least through point #3D13. At the same time, it was planned to create a new operating system, which, partly located in the same TR-DOS ROM (BIOS), and partly loaded from the hard drive, would manage the change of TRD images in virtual drives. Work on researching TR-DOS
and ways to change the firmware began when a new member
of the group joined us - UKMS[z]. From that moment on, thanks to him, the principlesprojects have changed dramatically. It was he who convinced us of the possibility of full emulation of the 1818VG93 controller on the Spectrum, at least when working with memory. Moreover, he already had preparations for such an emulator, since he had also been thinking for a long time about how to load the HDD on his ATM with work. Establishing
contact through this site with other ATM users gave him
confidence that his developments were needed by someone else besides himself. From that moment on, the development of the project took a straight path. In addition, a
discussion began on the principles of a new multitasking graphical operating system
for the ZX, running from a hard drive and using new firmware to
run TRD software. But if the future of the OS project is still uncertain, then the new
TR-DOS firmware is already, rather, not a project, but a reality:
PRINCIPLES AND PROSPECTS OF THE PROJECT
Today I can say with confidence that the new TR-DOS firmware already
exists (it is currently being tested and catching the latest
bugs), and I want to outline what new it will bring to the world Spectrum.
1) The ability to “break away” from the real flop when working with TR-DOS. Now
640Kb disk is just a logical formation that can
exist on any media in the form of a TRD file. Although using one
POKE operator it will be possible to enable the previous TR-DOS operating mode forany
device - ABCD. Now ANY real operating system will be able
to use almost all the wealth of TR-DOS software - be it iS-DOS, be it CP/M, be it Vasya_Pupkin-DOS.
2) As a consequence of this, it will be possible to store more than 90% of the software under TR-DOS on the hard drive and any other media (for example, CD-ROM or FDD in 720Kb
MS-DOS format). The only limitation is that the virtual images use the standard TR-DOS format. Although
theoretically it is possible to support FDI images with a non-standard format, but for
this the upper memory of even a megabyte Spectrum may not be enough. Apart from
this limitation, nothing will prevent you from using TRD software from the hard drive, since
the VG93 software emulator already now, with bugs that have not yet been fully caught,
allows you to run programs from a RAM disk with the most non-standard
loaders, including music ones.
3) Completely new possibilities will open up in using the MAGIC button, as
soon The NMI processing procedure in
TR-DOS will be radically redesigned. In addition to the standard function, it will be possible to use this
button to call up a resident stored in the upper memory next to the RAM disk
TR-DOS. As a resident, it will be possible to record any program there, but
first of all, it can and should be used to organize the service
work of a virtualTRD with HDD and other devices, for example - for
loading a new TRD from the hard drive (for example, when starting a game on two
images), recording changes in the virtual TRD to a screw, returning from a
TRD program to the operating system (which, I repeat, can be anything!)
or vice versa, continuing, after the operations performed, the execution of the
TRD program. Now you won’t need to press
RESET to return from many programs. It will be enough to exit via MAGIC to the resident menu and select the
"exit" option.
4) Finally, the upper memory will no longer be used only for
copying, but will become a necessary and integral part of the Spectrum,
used for a virtual RAM disk and the resident. And most importantly,
it will make sense to connect new, most non-standard external devices,
because with the “virtualization” of TR-DOS, the importance of other axes will sharply increase,
for which the problem of connecting new ultra-modern drives will
consist only in writing the appropriate driver, and the virtual
RAM-TRD will absolutely not care where the software was loaded into memory. And this
means that ZX-SPECTRUM will cross another significant step in its
glorious development.
Today the project is 90% ready. The VG93 emulator has been completely written, which just needs to be optimized a little and a couple of bugs caught. Still to be done
new MAGIC handler andintegrate the new TR-DOS into the general firmware, which is different for different clones. It is also necessary to adapt it to different
upper memory standards, since so far it only works with
ATM-turbo2+ memory, being written on this clone.
Minimum requirements for Specky for the new firmware are 1024Kb RAM.
Optimal - 1024Kb RAM and IDE HDD, otherwise why is this new firmware
needed at all?
B In the future, it is possible that the firmware will be modified for clones with 512 and 256 kilobytes of memory. A solid TRD, of course, will not fit into such memory,
but it will be possible to place smaller SCL files, which is already good.
The main advantage of the project, if it is successfully completed, will be that
it will open up prospects for the creation and implementation of new OS projects,
since previously the obstacle to this was the reluctance of users
to switch to such OSes, while losing the opportunity to use
the huge baggage of TR-DOS. As a result, most of the time other
operating systems, even when installed on the ZX, stood idle, while
the old TR-DOS worked tirelessly, creaking disk drives, ignoring the
connected from a high bell tower, but practically virgin
empty hard drive.
From the moment the project is completed, I hope everything will go differently, namely
as it should would...
Project participants-
Maksagor
UKMS[z]
T!M0N
Gratitude is also expressed to Warlord, NUTS, ART and all those others
who expressed interest and support (moral and intellectual) for our
undertaking.
Timonin Maxim aka Maxagor.
===========================
THE END