Crazy hands - Again about the ZX-NEXT HDD

 COMMENTS          
     BACK TO EARLIER      
                               
(C) 2000 Inside                          
-------------------------------------------------------
 Hello readers of "Black Crow"! By
At Max's suggestion, I decided to write an article
to his journal, I hope the information
outlined below will be useful to someone -
on. These are mainly comments on the third
his number. So, let's go...             
                               
     HDD circuit for ZX=NEXT      
     ====================      
                               
   Just some kind of diagram without explanation
information "what, where and by what software is supported
but", and the buggy commander under TyR-DOS is
clearly not what is needed. Actually, there is
There are several types of ATA/IDE controllers
(host adapters):                        
                               
 = SMUC,                      
  = ZX=NEXT,                   
  = (c) NEMO,                  
  = Sprinter.                  
                               
  There may be others, but these are the main ones.
new.                                  
                               
   The fact that SMUC is Scorpion's lotion
ka - it is known, as it is also known that he himself
Scorpion is no longer produced. And for anyone
SMUC's amateurish repetition is too challenging
fatal Sprinter IDE controller diagram
I didn’t see it and therefore I can’t say anything
huh, most likely everything is cool there as always
then, but there is no support from IS-DOS
There is no connection for this circuit, as I understand it. Now
many may be indignant >8-oh why
we need this IS-DOS. Well, first of all, what
work with HDD is complete only under full
valuable OS, I think this is understandable. Secondly
ry, there is NO other OS yet: NEOS is not in
account, because in the form in which
it's located now, it's more like
on the modernized TyR-DOS than on OPE-
RATIONAL SYSTEM.                       
                               
   The (c)NEMO circuit is simple and technologically advanced,
BUT the ZX-NEXT scheme has a number of advantages in terms of
compared to one disadvantage: two
there are more bodies (why look at two, yes-
lee). Athe advantages are:             
   It is possible to disable the control ports
ra, which reduces the likelihood of damage to in-
formations.  There is no need to organize a signal
blocking IORQ, but it’s still better to disable it
lead.  To the controller (c)NEMO signal RES/
goes directly without the possibility of blocking it -
rovki, that within the framework of the ZX-SPECTRUM ideology
(RESET button is trampling on everything)
- undesirable.  To signals CS0/ and CS1/
two addresses were spent, although according to the specifics
ATA/IDE tions are orthogonal. In ZX-NEXT
controller, port addressing is carried out
is based on the “classical” scheme - when
the device is selected by the low byte ad-
res, and already the ports of the device itself -
senior                                 
  There are IS-DOS drivers for both (ZX-NEXT
and (c)NEMO) controllers.                 
                               
   As it was in ZX-NEXT,
it’s good, but for repeating separately
This device would benefit from some
minimization of the number of IC housings.
Below is a list of changes:          
























                               
   Для  организации сигнала IORGE (блоки-
ровка IORQ) необходимо подключить элемент
"И" входами на не "висящие в воздухе" вы-
ходы дешифратора, а выходом на оставшийся
свободным  буфер ЛП8 (вход EZ). Т.к. эле-
мента "И" нет, то использован его диодный
"аналог".  После такой "минимизации" вид-
но,  что8 buildings remain. For organi-
tion of the work of the so-called Slave IDE-
devices sometimes use addressing
cable - to pin 28 on the host side
adapter is connected to ground, and the 28th
the wire going to the Slave device is re-
cut.                                  
                               
  If anyone is unable to reproduce the scheme,
mu of scattered additions, it’s better not
mock yourself and your equipment - HDD
and CD-ROMs are expensive things. I and the author of the magazine
no responsibility for crooked hands
We don’t carry ki readers. Payment is POSSIBLE (but
I, the author of this article, have nothing to do with this
I can’t firmly promise) it will be produced
dit in very small batches as you progress
slowdown of orders, if any at all
blow. The price for a "bare" printed circuit board will be
not less than $2, when combined with CMOS (according to
scheme Mg. Gluk) - no less than $4 for “naked”
board.  For example, the price for a HDD controller
(c)NEMO - $17.32. The bus will be NEMO-BUS +
separate connector for connecting to the
DMA roller, if anyone has one
available (DMA UltraSound Card, etc.). By-
Why NEMO-BUS and not ZX-BUS? Because
Hard is now produced only for it and
the most important thing is that the only one left
Gosya active IS-DOS developer Alexey
Leontyev is precisely KAU, then the tire corresponds
existing                                
                               
  And now = a little about NEOS.   
  ==========================   
                               
   Even though I belong to the category of those people...
deys who don't have KEMPSTON MOUSE in
2000";), but the problem of NE-
OS solved through the cache, although it was extremely barbaric
Chinese method:                            
                               
;LOAD ANY NEOS FIRMWARE           
 ORG #8000                    
        INCBIN "neos_512r"              
                               
;JUMP TO THE INSTALLER                    
        ORG #7FFD                    
        JP #C000                    
                               
;INSTALLER                               
        ORG #С001                    
        D.I.                               
        IN A,(#FB);OPEN THE CACHE    
        LD HL,#8000                 
        LD DE,0                     
        LD Sun,#4000                 
        LDIR ; LOADING THE FIREWARE INTO THE CACHE
        RST 0;PASSING THE BATON TO NEOS 
                               
;RECONDITIONING THE APPEAL TO TR-DOS         
        ORG #BD2F;IN THE WORLD #3D2F       
        JP #2FED;ANY FREE  
;IN THE FIRMWARE PLACE                        
                               
 ORG #AFED;THIS IS THE "FREE-
;BOTTOM PLACE"                             
        LD (#2A04),HL                 
        POP HL; REMOVE THE ADDRESS FROM THE STACK  
;SUBROUTINES IN TR-DOS                   
        LD (#2A00),HL;#2A00,#2A04: Junior-
;THERE ARE FREE CELLS IN THE FIRSTWARE         
        POP HL; REMOVE THE ADDRESS FROM THE STACK  
;RETURN TO NEOS,                        
        LD (#5EF8),HL;SAVE IT BY
;TO THE ADDRESS WE NEED                      
        LD HL,#5EF3                   
        PUSH HL;AND CHANGE IT TO THE REQUIRED  
;US                                     
        LD HL,(#2A00)                 
        PUSH HL,DE,SUN                   
        LD HL,#3100                   
        LD DE,#SEEC                   
        LD Sun,#000С                   
        LDIR ;LDIRIM THE SUBPROGRAM TRANSFER 
;TURN ON CACHE/TR-DOS TO THE ONE WE NEED       
;ADDRESS. I HAVE #SEEC - UNDER THE NEOS STACK 
 POR BC,DE                      
        LD HL,(#2A04)                 
        JP #SEEC;AND LET'S JUMP TO ENTU SA-
;MY SUBROUTINE.                       
                               
;AND HERE IS THE SUBPROGRAM ITSELF:                
        ORG #В100;ANY FREE IN    
;FIRMWARE PLACE.                         
;INPUT:                               
        PUSH AF                         
        INA,(#7B); TURN ON TR-DOS    
        POP AF                         
        JP#3D2F; AND GO THERE         
;OUTPUT:                              
        PUSH AF                         
        IN A,(#FB);ENABLE CACHE        
        POP AF                         
        JP #0000;AND GO NOT TO ADDRESS 0,
;AND TO THE ONE THAT IS IN THE PROCESS OF WORK         
;SAVE HERE (SEE ABOVE)            
                               
 [Text in assembler format STORM v1.3]
                               
   After this, run the recorded ta-
In this way the code block can be "evaluated"
"Vat".  Although at the application level to "Cher-
Noah Vorone-3" - firmware+commander, estimated
It is most likely impossible to destroy the SYSTEM.      
                               
 Max: oh well, it’s just a small matter - for-
 show yourself a complete set or look
 come to me for a more complete one.          
                               
   But in general the first impression is this:  
                               
- buggy disk drive driver: mine is 5`25"
Robotron, according to the driver, is not keeping up
spin up, and 3`5" Epson in turbo mode
does not initialize the first time;      
                               
- for some reason TR-DOS was selected as boot
floppy disk :-| , and MS-DOS drivers in the firmware
no at all;- message "unknoWn file system" if in
In this case, one might say so, not quite
seven honest. It would be more honest to call it "ah-
sent FS driver" because this is the message
issued only in case of MS-DOS floppy disk,
with other NEOS file systems it is beneficial
explodes;                      
                               
- interesting, but what do the authors suggest?
exclude drivers from the chain, finding
stored in ROM, and whether the operation will be invalidated
installing a driver or library by finding
at the beginning of a long chain, similar to
MOVE operation on a TR-DOS floppy disk;         
                               
- something is not visible in the start options menu
      TR=DOS 128               
      TR=DOS MORE 128          
as, for example, in GLUK RESET SERVICE. Or
hang everything with toggle switches again?        
                               
- HIMEM driver does not recognize correctly
physically available pages;            
                               
- no system command to clear blocks
memory, i.e.  if in the 128th firmware from
commander nc.spr do some EXEC
nonsense, then falling out of the commander,
You can “clear” the memory only with the command
EXIT, you mean RST 0;                      
                               
- in the commander, displaying information
razhalku does not go through system
functions,but as in TR-DOS, read - without op-
rational system;                       
                               
- press and hold any button... OP!
Stopped.  We count: 4 lines of 64
symbol.  So... Sixteen characters in
name + "." + 3 characters in extension...
not counting "" and the cd command... Multiply...
let's divide... OP!  Directory nesting - 12
(read - but the king is NAKED!!!). Well this is to
to what is in IS-DOS today -
32. In WINDOWS`9x...                      
                               
    And now about IS=DOS...     
    =====================     
                               
   It has become fashionable (if I can say so,
considering the number of users ZX) ru-
go IS-DOS. Like, there is not enough memory, etc. and
etc., and especially the RST 16 viper was tortured.
But the main problem that remains there is
The volume of the device is 16M. Do you want
speed - work with system RAM dis-
com. Do you want a cold start HDD - perhaps -
hundred, and the 128th basic remains for a month
those  Do you want TR-DOS emulation - subject to availability
TR-DOS v5.12р, etc.  (where there is RAM-
disk drive) - please, and in the format
*.TRD, which in light of all sorts of compact
disks are very convenient.  Tortured RST 16 -
install IS-DOS cHik. Even RK UNZIP started up.
Which of the followingprovides (the main
goal in the present tense!) NEOS?          
                               
   What IS-DOS lacks is a driver
faith CD-ROM, so run every time with
PC users get tired of using a CD, yes
and the latter are satisfied with this fact. And so it is
it turns out that Ukraine with its Pentagons
and St. Petersburg is all in KAU-ZS-PROFI far away
are apart from each other, especially in terms of
information on ZX.  If they are talking about CACHE
generally heard, then very little. Alexey Le-
Ontiev, for example, thinks that shadow RAM
Only the St. Petersburg ZX listed above has it,
and IS-DOS cHik under CACHE is not yet available. Well
the most sore spot of IS-DOS, as they say,
moose higher - maximum volume of the device
16M - you have to split the hard drive into
several logical drives. And there are only
there can be only 8 (for ISDOS`2000). Mi-
plus two floppy drives, minus a RAM disk (I have
there are two of them - fast in the cache and system - in
HIMEM). And there is practically nothing left
th.                                      
                               
          About everything...          
          ===========          
                               
   There is nothing to write about Spectrum,
I don't agree.  And if there is nothing, then you need
write about something else, but then not in Spectrum-
publication,so as not to confuse the minds of the gullible
users. For some reason it hurts about iron...
most write in the form of “write and forget”
and only rare editions like Deja Vu, on-
example, sometimes they condescend to comment
repeated schemes.  But Hard, like
soft, needs comments. But in
The difference from soft is also in the print patterns -
nal boards, etc. For example, for a very long time
I was struggling with RU7.  When the board was lying on
semi - everything was fine (I assembled it from scratch
pentagon-512K). How it came to IBM's -
corps - problems began: not li-
byat RU`shki located close to the pi-
tania.  Pi-blocking capacitors
tania was not helped. After a long search I
made the memory work stably,
and very stable - the computer is under-
connected to a 220V network on one wire with
old refrigerator. But business: only 1
capacitor 470pF on CAS/zero line-
ki. Yes, there is a common belief that
as if you can’t put RU’shki in the panel-
ki, but I had one and a half rulers
exactly like that, because I bought RU7 on ba-
early on cheap - the seller pulled them out of
old board in which they also stood
on panels. Their legs were oxidized to
the fact that they were practically not soldered. And
nevertheless, we worked for a year without glitches and
failures. What exactly is this conversation for?
is it coming?  And I upgraded my penta
- installed 2MByte SIMMs according to the schememagazine-
la Deja Vu #7... and it began. Sims don't want to
They work stably - and that’s it. When
hanging a capacitor on CAS/, like
advised in Deja Vu #9, memory is stable
works only in turbo mode. When sub-
hanging the capacitor on RAS / vice versa -
only at 3.5 MHz. So what's better in that
circuit still signal WE/ shift and decrease
vary in duration, as was the case
author of the diagram. But I was glad that fact
the one that one and a half lines of RU'sheks were in pa-
nelkah and I removed them from the board without any
harm to the flimsy tracks of the latter, especially
those who were hiding had to be blocked.         
                               
   A little higher I mentioned that I do not bring
There are drawings of printed circuit boards. But there's nothing here
no surprise - no corresponding
current editor. And in general, when I
people who are far from ZX ask what I can-
what to do with it, I don’t even know what to do
answer.  The first thing I want to say right away is
Well, I'm typing texts. But that's not really
honestly, since typing texts for ZX
without the ability to rewrite them on a PC - this is
cannot be called the word "DO", in the sense
"WORK".                              
                               
N.V.  The fact that it is not possible to transfer texts to the PC
perhaps - not entirely true - the same isdos
will allow you to write to almost without problems
msdosdisk any text file. Criterion
"to do" cannot be defined only by this,
otherwise it turns out that if the PC does not read
floppy disks formatted on zx, then on
her work cannot be called “work”... ;) 
                               
   Here is another topic that can be covered -
sorry, because after reading numerous
magazines, it seems that
there is no problem as such - everything is fine -
but is written and read between IBM and ZX, but
in fact this is not the case. The problem of inconsistency
controller capacity has been affected
only in ZXPOWER #2 magazine. And that's just
at the level of the fact of existence, and not as with
this fight with 100% efficiency.
PC emulators write all sorts of emu-
lators, write all sorts of formatters and
copiers, but from the Spectrum it’s impossible
was to rewrite it on IBM, so, at least with
50% confidence in successful completion
operation, and still - it’s impossible! And
then copiers for the new FAT I something
I haven’t met that. No, well, an IBM floppy disk
copied, but looks MS-DOS directory
somehow strange.                          
                               
   And that Spectrum is supposedly becoming obsolete,
so it’s not the fault of megahertz and megabyte -
This computer just ended up in our CIS
(forgive me) asshole hands: everyone's friend
don't care about the other.  And it's not even my faulthackers and pirates 100%, just semi-
This is a vicious circle:         
                               
- Hard manufacturers don’t care about the stan-
darts Sinclair Research ltd, AMSTRAD, and
most importantly - at each other: for example,
there is LPRINT-3, but we don’t care, we are
let's try and come up with something like that on AY
or even more confusing; there are #7FFD, #DFFD, and
we don't care - we'll figure it out               
                               
             76543210          
   #1FFD .X.X....          
   #7FFD X....XXX          
                               
 and it doesn’t matter that there are intersections with the port
extensions PENTAGON-512, the main thing is technology
logical!  But I will outdo everyone on this issue -
zero makers of one cool car -
we spat on NEMO-BUS and ZX-BUS together
taken - we will have an ISA!               
                               
- soft producers don’t care about production
Hard's drivers, most importantly - on CONJUNCTURE
soft market:                              
                               
    - we will not support HIMEM: it
    no one has (although "sociological
    surveys" conducted by some
 by our magazines back in 1997
    they called that there is at least 256k
    for the majority). But for some reason Mazda-
    According to some, Windows is ra-
    they work great on 128 cars,and on machines with HIMEM;                
                               
    - we write programs only for ourselves,
    and don’t care about what others need,
    they won't pay us any money anyway
    accidentally indicate prices at the end
    always more than $1 for a 640k floppy disk, and
    PC CD 640M costs $2. As a result, the necessary
    There is NO soft.                          
                               
   A great thing (in terms of statistics) was done
launched PLAYGEAR GROUP by releasing a CD - immediately
you can see how much soft-TRASH was released for
Spectrum.                               
                               
   Something is missing in this link -...
USER. And he remained like a CLASS in the Spectrum -
sya?   What should he use?  Soft-trash on
non-standard cars?  There are only
Hard-&soft-makers (thank you), hackers,
swappers, etc. The situation is reminiscent of rock
concerts (not pop-rock, but rock), where
at concerts in most cases the music
there are many more cants than listeners. But
"Rock and roll is dead, but I'm not yet" they sang
in the decade before last, and Spectrum
dead, but I'm not yet" they started singing like that
in 1993 (see ZX-REVIEW published by INFORCOM). Yes
that in the light of such analogies we still have
a chance to change something! But it needs to change already
            NOW!            
                               
 When preparing an articlewere used:
                               
soft: IS-DOS TECH EDITOR               
        ART-STUDIO 128                   
                               
literature:                              
  Guk M. "PC Hardware. Encyclopedia"
lopedia".                                
                               
address: 61143                   
       Ukraine                 
       Kharkov              
       st. Bolshaya Koltsevaya   
       building 14, apt. 61          
       Lazarev Andrey         
                               
FIDO: 2:461/48.3 (for Inside)