[comp.sys.ibm.pc] Why 4.77MHz on clones?

wcs@ho95e.UUCP (01/27/87)

A lot of PC clones offer two speeds, 4.77 and 7.x or 8.0 MHz, and claim
the 4.77 is for compatibility.  What is the compatibility problem?
Add-on hardware?  Stupid copy-protection?  Do I need to care?
		Thanks;
-- 
# Bill Stewart, AT&T Bell Labs 2G-202, Holmdel NJ 1-201-949-0705 ihnp4!ho95c!wcs

muth@amdahl.UUCP (01/28/87)

In article <1292@ho95e.ATT.COM>, wcs@ho95e.ATT.COM (#Bill.Stewart) writes:
> A lot of PC clones offer two speeds, 4.77 and 7.x or 8.0 MHz, and claim
> the 4.77 is for compatibility.  What is the compatibility problem?
> Add-on hardware?  Stupid copy-protection?  Do I need to care?

Some software (mainly games) uses timing loops that are based on 4.77 MHz.
If you run these programs on machines with faster clocks (like a clone or
an AT) you end up with some real interesting results.

There may also be add-on hardware that counts on 4.77 MHz. I know of one
LAN adapter that works on AT's running at 6MHz, but not at 8MHz.
-- 
John A. Muth           ...!{ihnp4,hplabs,sun,nsc}!amdahl!muth

darrylo@hpsrlc.UUCP (01/28/87)

     If you play games, you generally need the 4.77MHz speed to make
real-time games run at the correct speed (otherwise, it runs *much* too
fast).  I've also heard that some copy protection schemes need a 4.77MHz
clock.

     -- Darryl Okahata
     hplabs!hpcea!hpsrla!darrylo

mattw@phred.UUCP (01/29/87)

Some of the high speed xt clones 7.16, 8.00 Mhz will not format or
'diskcopy' at the high speed. This is due to many factors including
the ram speed and DMA channel circuit topology. Some of the clones
change the speed of the clock to the DMA circuit during high speed
clock operation ie: if the speed of the CLK88 line is 8Mhz the DMA
clock is lowered to 4Mhz. This will allow the clone to format diskettes
at the high speed. Another problem with some of the 'Turbo' clones is
that the high speed design does not preserve the 14.x Mhz clock for
the proper operation of the CGA video system (monochrome will continue
to work fine) and the interrupt for the real time clock, tone, and
keyboard scan circuits. The result of this is that the CGA card will
not work and the real time clock will not keep time at the high speed.
There is a plethora of xt 'TURBO' clones on the market. Some are
correctly implemented others are not. I would not buy another without
throughly analysing the high speed operation. Good luck on your 
purchase. <8^)

tes@whuts.UUCP (01/30/87)

> In article <1292@ho95e.ATT.COM>, wcs@ho95e.ATT.COM (#Bill.Stewart) writes:
> > A lot of PC clones offer two speeds, 4.77 and 7.x or 8.0 MHz, and claim
> > the 4.77 is for compatibility.  What is the compatibility problem?
> > Add-on hardware?  Stupid copy-protection?  Do I need to care?
The PCDOS copy, backup, restore, commands actually fail at speeds
other than 4.77.  I did not believe this when I installed my Fast88
at 7.4.  (Excellent product, excellent documentation, very cheap
*real* improvement w/o need for slot).  The Documentation said
that the switch on the Fast88 was to allow these programs to run
at 4.77.  Non-Microsoft/IBM copy programs do not have these
problems, proving stupid S/W, not H/W as the problem.
-- 
    -----                   Terry Sterkel
  -====----            AT&T Bell Laboratories
  ---------    {harvard|allegra|ulysses|ihnp4}!whuts!tes
    -----         [opinions are obviously only my own]

pozar@hoptoad.UUCP (01/31/87)

> In article <1292@ho95e.ATT.COM>, wcs@ho95e.ATT.COM (#Bill.Stewart) writes:
> > A lot of PC clones offer two speeds, 4.77 and 7.x or 8.0 MHz, and claim
> > the 4.77 is for compatibility.  What is the compatibility problem?
> > Add-on hardware?  Stupid copy-protection?  Do I need to care?

Side_note: Ever wonder why 4.77 was picked?   The 8088 bus was designed 
around the Multi-bus specs.  As I remember 4.77 or some mutilple was use
on the Multi-bus for timeing.

-- 
        Tim Pozar
UUCP    pozar@hoptoad.UUCP
Fido    125/406
USNail  KLOK-FM
	77 Maiden Lane
	San Francisco CA 94108
terrorist cryptography DES drugs cipher secret decode NSA CIA NRO IRS
coke crack pot LSD russian missile atom nuclear assassinate libyan RSA

madd@bucsb.bu.edu.UUCP (02/03/87)

In article <1292@ho95e.ATT.COM> wcs@ho95e.ATT.COM (#Bill.Stewart) writes:
>A lot of PC clones offer two speeds, 4.77 and 7.x or 8.0 MHz, and claim
>the 4.77 is for compatibility.  What is the compatibility problem?
>Add-on hardware?  Stupid copy-protection?  Do I need to care?

There are a lot of things that you might want 4.77 for.  What happens
if your favorit space game, which at 4.77 mhz is almost impossible,
suddenly has double the horsepower?  Makes life tough.  Also, some
cards can't handle the ~8 mhz modes (though this is very uncommon in
my experience).  Finally, many diskette drives/controllers have timing
problems with 8 mhz modes.  This is very apparent with the THE PC+
machine that a friend of mine has.  Works perfect except that it will
occasionally lock up from timing problems with the diskette drive (if
it's used) when running @ 8 mhz.

There may be some other reasons, but these are popular ones.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
                   - Jim Frost * The Madd Hacker -
UUCP:  ..!harvard!bu-cs!bucsb!madd | ARPANET: madd@bucsb.bu.edu
CSNET: madd%bucsb@bu-cs            | BITNET:  cscc71c@bostonu
-------------------------------+---+------------------------------------
"Oh beer, oh beer." -- Me      |      [=(BEER) <- Bud the Beer (cheers!)

jkg@gitpyr.UUCP (02/03/87)

In article <1696@hoptoad.uucp> pozar@hoptoad.UUCP (Tim Pozar) writes:
>Side_note: Ever wonder why 4.77 was picked?   The 8088 bus was designed 
>around the Multi-bus specs.  As I remember 4.77 or some mutilple was use
>on the Multi-bus for timeing.

I always thought it was because the crystal input to the 8284 (8088 clock
driver) is 3X the processor speed - IBM PCs and compatibles use a 14.31818
Mhz crystal to provide basic timing. It just so happens that 14.31818/4 (a
couple of D flip-flops !) is 3.58 - very close to the colorburst crystal 
frequency. I can just see the conversation at the design meeting now...

IBM Design Engineer 1 - "Oh boy! - we saved ourselves one whole chip!"

IBM Design Engineer 2 - "Yeh - and now we can drive a color TV directly 
			for a monitor!"

In Unison -		"Won't the bean counters be proud of us!"


						Jim Greenlee
-- 
The Shadow...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!jkg

Jryy, abj lbh'ir tbar naq qbar vg! Whfg unq gb xrrc svqqyvat jvgu vg hagvy lbh
oebxr vg, qvqa'g lbh?!

brewster@watdcsu.UUCP (02/03/87)

> > A lot of PC clones offer two speeds, 4.77 and 7.x or 8.0 MHz, and claim
> > the 4.77 is for compatibility.  What is the compatibility problem?
> > Add-on hardware?  Stupid copy-protection?  Do I need to care?

> Side_note: Ever wonder why 4.77 was picked?   The 8088 bus was designed 
>around the Multi-bus specs.  As I remember 4.77 or some mutilple was use
>on the Multi-bus for timeing.
>Tim Pozar

    Compatibility problem : early PC software used software timing loops,
    which will fail when you change the speed that code executes at.
    
    Why 4.77 MHz ?  See IBM Tech Ref 1-4

    The microprocessor operates at 4.77 MHz.  This frequency is derived
    from a 14.31818 MHz crystal, the frequency of which is divided by 3
    for the microprocessor clock, and divided by 4 to obtain the 3.58
    MHz color-burst signal required for color televisions (also used
    by colour televisions in Canada).

    Result : 4.77 MHz too save cost of 1 crystal.

    Copied without permission but in the spirit of page v of the preface :
    "the information in this manual is for reference".

    The original PC was put out at a time when running the processor
    at 7.15 MHz, if even possible, would have been far too risky a
    move for a conservative company like IBM.  I believe that it was
    only last year that you could start to buy a 6 MHz AT from IBM,
    where clone companies talk of selling 16 MHz AT's, and can 
    probably deliver reliable 10 MHz product.
                                                   
						   Try not  to become  a  man
UUCP  : {decvax|ihnp4}!watmath!watdcsu!brewster    of success but rather  try
Else  : Dave Brewer, (519) 886-6657                to  become a  man of value.
                                                         Albert Einstein

goldy@hound.UUCP (02/04/87)

In article <1696@hoptoad.uucp>, pozar@hoptoad.UUCP writes:
> > In article <1292@ho95e.ATT.COM>, wcs@ho95e.ATT.COM (#Bill.Stewart) writes:
> > > A lot of PC clones offer two speeds, 4.77 and 7.x or 8.0 MHz, and claim
> > > the 4.77 is for compatibility.  What is the compatibility problem?
> > > Add-on hardware?  Stupid copy-protection?  Do I need to care?
> 
> Side_note: Ever wonder why 4.77 was picked?   The 8088 bus was designed 
> around the Multi-bus specs.  As I remember 4.77 or some mutilple was use
> on the Multi-bus for timeing.
> 

Ever try playing some of the games like DIGGER or PACMAN or STARGATE
at double or faster speeds???

That's why you need to be able to slow the machines down!

Mark H. Goldberg
..ihnp4!hound!goldy

byrnes@ge-dab.UUCP (02/11/87)

Line eater food....

Yes, you need to care about the 4.77 speed.

Many pc add ons won't work at 8mhz.

If you're a gammer, try playing Atari's Defender
at 8mhz! Many other real time display programs work
too fast at 8 mhz. 

So, yes you do need to worry, 4.77 is the "standard"
8mhz is the "Try it and hop I don't loose anything
important mode".

Good Luck

                                             Arthur J. Byrnes 
UUCP:   ...!mcnc!ge-rtp!ge-dab!byrnes        General Electric
GEnet:  advax::byrnes                        1800 Volusia Ave, Rm 4412
Voice:  +1 904 258 2507                      Daytona Beach, FL 32015

Disclaimer; These views are those only of the author, Arthur.