WARNINGS TO QUADRA 630 BUYERS
Tim DeBenedictis
I need to make a few corrections here:
1) Apparently the Q630 can accept 72-pin DOS SIMMS after all; that's
what the "technical specifications" sheet buried inside the apple, er,
"Manual" says. So I guess the people who sold me thing ripped me off
$100 by telling me otherwise.
2) The manual also says that the machine has a full 1MB of video DRAM,
not 512K. That means, in theory, that I should be able to run 1024x768
monitors at 8-bit color. Well, after a little experimentation, I was
able to get my monitor to do 832x624 at 8-bit. I could not, however,
get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
possible that the machine only has 512K of video memory. The fact that I
could not get the monitor to do 1024x768 makes me think that the computer
really does only have 512K and not a full 1MB. (I know this monitor
supports 1024x768 because I've run it on a machine that did before.)
Does anyone know the facts here? My monitor is a SONY multiscan 15sf,
with a SONY adapter for all sorts of different configurations. Any help
greatly appreciated!
Tim DeBenedictis
edwa...@acf6.acf.nyu.edu
>
> I need to make a few corrections here:
>
>
> not 512K. That means, in theory, that I should be able to run 1024x768
> monitors at 8-bit color. Well, after a little experimentation, I was
> able to get my monitor to do 832x624 at 8-bit. I could not, however,
> get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
> possible that the machine only has 512K of video memory. The fact that I
> could not get the monitor to do 1024x768 makes me think that the computer
> really does only have 512K and not a full 1MB. (I know this monitor
> supports 1024x768 because I've run it on a machine that did before.)
>
> Does anyone know the facts here? My monitor is a SONY multiscan 15sf,
> with a SONY adapter for all sorts of different configurations. Any help
> greatly appreciated!
>
> Tim DeBenedictis
According to Apple Service Souce, the 630 does have 1 MB DRAM frame buffer.
This gives 16-bit on 12",13",14",15" and VGA (640*480) monitors
You get 8-bit on 16" and SVGA (800*600) monitors.
Those are the facts
Kevin
Donald J. Walli
video hardware in the 630 just plain *doesn't support* 1024x768. It's
not a common res for Apple video, and the only times I've seen it work
with most Macs is via some cable voodoo on Q700s and Q900s. I think
the PowerMacs might support that res, but on the 630, I'm not sure.
Don Walli
dwa...@mtu.edu
Tim DeBenedictis (tim...@netcom.com) wrote:
:
: [stuff deleted]
:
: 2) The manual also says that the machine has a full 1MB of video DRAM,
: not 512K. That means, in theory, that I should be able to run 1024x768
: monitors at 8-bit color. Well, after a little experimentation, I was
: able to get my monitor to do 832x624 at 8-bit. I could not, however,
: get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
: possible that the machine only has 512K of video memory. The fact that I
: could not get the monitor to do 1024x768 makes me think that the computer
: really does only have 512K and not a full 1MB. (I know this monitor
: supports 1024x768 because I've run it on a machine that did before.)
:
: Does anyone know the facts here? My monitor is a SONY multiscan 15sf,
: with a SONY adapter for all sorts of different configurations. Any help
: greatly appreciated!
:
: Tim DeBenedictis
:
--
|| Donald J. Walli dwa...@mtu.edu ||
|| Systems Administrator, Dept. of Fine Arts ||
|| Macintosh LAN administrator, CCLI, Dept. of Humanities ||
|| Michigan Technological University ||
Tim DeBenedictis
certainly a much, much, much nicer machine than my old LCII!!!
However, I have some warnings for potential buyers:
1) It only comes with 4MB of RAM. Since this is useless for any serious
work, right off the bat you'll need to buy more RAM (currently, the 4/250
configuration is the only one Apple sells.) Unlike the other Quadras,
the Q630 WILL NOT ACCEPT 72-pin DOS SIMMs, so you're forced to buy the
more expensive Mac SIMMs. I got an additional 4 MB for about $200... maybe
more than I should have paid, but what the hell...
2) The internal hard drive has closer to 240 MB than the advertized 250.
This is not a big deal. What is a big deal is the fact that it is an IDE
drive, not a SCSI drive. This means the NONE OF YOUR CURRENT HARD DRIVE
TOOLS WILL WORK on it. For instance, you can't repartition the drive with
the "Internal HD" format software or the "Apple HD SC setup" software. In
fact, all you can do is reformat it entirely. I hope to god that Norton
Utils will still work with it, and pray like hell that I don't have a
hard-drive crash...
(If anyone out there knows of Mac disk tool software that can deal with IDE
drives, PLEASE LET ME KNOW!!!!)
This could potentially cause other problems, say on networks which expect
SCSI-only chains. Fortunately this isn't a problem for me.
3) The machine stores its video memory on DRAM chips, not VRAM. While this
supposedly causes a performance hit, I haven't noticed it. The bigger
problem is that there's no way to expand the machine's video memory. The
bottom line is, if you were expecting to use a huge, 21", 1024x768
monitor in 32-bit color with the Q630, you cannot do it, noway, nohow.
The largest monitor size the machine supports is 800x600 at 8-bit color.
It does not support 632x824 at all, although it does 640x480 in 16-bit color
quite nicely.
4) The clock battery is an alkaline battery, not lithium. Macs already
have notoriously poor clocks; this is (I'm sure) a cost-cutting move on
Apple's part which only makes the performance worse. The alkaline
battery will probably have a shorter lifetime than a lithium battery
would, and may be more suceptible to leakage, rupture, etc... time will only
tell. (So to speak...)
5) There is only one expansion slot, and it's an LC 030 (not 040) PDS
slot. This means that most current expansion cards will not work at all:
no NuBus cards, no video cards, no houdini cards. The latter of these is
a disappointment to me: I'm currently hoping that sometime in the future,
the "Houdini Jr." will work with LC-slot machines, so I don't have to buy
a separate PC. I know, I know, fat chance, but I knew I was taking that
risk when I bought the machine.
On the positive side:
1) THIS MACHINE IS FASTER THAN HELL. Compared to my old
LC II, the difference is an order of magnitude. Speedometer 4.0
generally clocks it at about 30% faster than a PowerMac 7100 under emulation.
2) There's a slot for ethernet and/or an internal FAX modem. There's
also a video port, which, apparently with the right card that Apple's
about to sell, you can use to watch (and capture?) live TV on your monitor.
In fact, on the front of the case, there's a window for recieving infrared
signals from a remote controller. Pretty cool, although I doubt I'll ever
use this.
3) There's a bay for an internal CD-ROM drive. (What's the going rate
for an internal Mac CD-ROM these days? This is something I might actually
use...)
On the whole, I'm fairly pleased with the machine. It blazes through my
numerically- and graphically-intensive sky simulation programs, and
800x600 video is a lot bigger than 640x480. The case is very slim and
attractive; it takes up no more room on my desktop than my old LCII
did, and I like the option of putting the CD-ROM inside (and maybe the
modem, too.)
However, I can see that this IDE drive thing is going to be a hassle, and
in general I think that in terms of expandability this will be a very
limited machine. While Apple says it is PowerPC upgradeable, Apple also
said the Quadra 605 would be upgradeable when they first released it,
too. And the one expansion slot the thing has is almost useless in terms
of the number of cards it will work with.
All in all, what you get is not bad for $1250. Even after you add
another $200 for the 4MB of RAM, it's still a nice machine, and even
nicer when you subtract the $600 I got for my old LCII. If you want speed
and you're not much worried about huge monitor sizes or Houdini cards,
I'd reccommend the thing strongly. If future expandability concerns you,
I'd advise you to take a second look. While the machine is very good for
the money, it does have some serious limitations.
Tim DeBenedictis
tim...@netcom.com
Erik A. Speckman
Tim DeBenedictis <tim...@netcom.com> wrote:
>
>However, I have some warnings for potential buyers:
>
>1) It only comes with 4MB of RAM. Since this is useless for any serious
>work, right off the bat you'll need to buy more RAM (currently, the 4/250
>configuration is the only one Apple sells.) Unlike the other Quadras,
>the Q630 WILL NOT ACCEPT 72-pin DOS SIMMs, so you're forced to buy the
>more expensive Mac SIMMs. I got an additional 4 MB for about $200... maybe
>more than I should have paid, but what the hell...
I don't know whether or not the 630 takes DOS SIMMs but I do know that PC
SIMMs are rarely if ever any cheaper than Mac SIMMs, that extra parity
bit does cost money.
I for one get useful work done with 4 MB and a 4 MB swap but I don't do
much graphics work. (I will be buying memory when I upgrade to 7.5 though)
As for expansion slots, they are over rated IMHO. I will probably never
add a card to my C610, I certianly won't add more than one. SIMM slots
and drive bays are another matter. One SIMM slot just plain stinks,
especially on a machine with 4 MB on the motherboard. Two slots are
great to have around these days, I bet there will be a bunch of 4 MB
SIMMs for sale on the net :)
--
____________________________________________________________________________
Erik Speckman espe...@romulus.reed.edu GBDS
Todd Stout
>
> 2) The manual also says that the machine has a full 1MB of video DRAM,
> not 512K. That means, in theory, that I should be able to run 1024x768
> monitors at 8-bit color. Well, after a little experimentation, I was
> able to get my monitor to do 832x624 at 8-bit. I could not, however,
> get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
> possible that the machine only has 512K of video memory. The fact that I
> could not get the monitor to do 1024x768 makes me think that the computer
> really does only have 512K and not a full 1MB. (I know this monitor
> supports 1024x768 because I've run it on a machine that did before.)
>
Although the machine might have 1MB of RAM available for video, this
doesn't necessarily mean that the hardware supports the higher resolutions
or bit depths. (As you indicated, it's only THEORETICALLY possible.) Both
MacUser and MacWorld state that the largest monitor that the 630 supports
with its internal video is 15". Since the Apple 15" has a max resolution
of 832x624, I have a feeling that nothing above that is going to work.
(Since this is the same resolution as the Apple 16", does the 16" work on
a 630?) I admit it's strange that they'd give it the memory and then
cripple the video, but hey, this is Apple we're talking about.
Does anybody know if the 17" multiscan works on the 630 (even at 832x624)?
The only thing keeping me from buying this machine is the reported lack of
support for the Apple 17" monitor.
***************************** _-_ ******************************
Todd Stout _____ ___/_____\___ _____
tst...@mit.edu | | | .|.| MIT |.|. | | |
TSt...@aol.com |_|_| _|_|__A__|_|_ |_|_|
Mark Rogowsky
> I need to make a few corrections here:
>
> 1) Apparently the Q630 can accept 72-pin DOS SIMMS after all; that's
> what the "technical specifications" sheet buried inside the apple, er,
> "Manual" says. So I guess the people who sold me thing ripped me off
> $100 by telling me otherwise.
Yeah, so you shouldn't have posted the original misinformation and created
some minor hysteria.
>
> 2) The manual also says that the machine has a full 1MB of video DRAM,
> not 512K. That means, in theory, that I should be able to run 1024x768
> monitors at 8-bit color. Well, after a little experimentation, I was
> able to get my monitor to do 832x624 at 8-bit. I could not, however,
> get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
> possible that the machine only has 512K of video memory. The fact that I
> could not get the monitor to do 1024x768 makes me think that the computer
> really does only have 512K and not a full 1MB. (I know this monitor
> supports 1024x768 because I've run it on a machine that did before.)
The machine does not support 1024 x 768 mode, period. It has nothing to do
with VRAM, it has to do with maximum video mode support. Some Macs, like
the Quadra 605, support 1152 x 870. The trend on the low end with the PMac
6100 and the 630 is to stop support at 832 x 624. You are topped out
there, period. There is a full 1MB of video memory, but its single-ported
DRAM, not dual-ported VRAM. This hurts performance a bit, but doesn't
limit the size. That's up to the video controller circuitry.
>
> Does anyone know the facts here? My monitor is a SONY multiscan 15sf,
> with a SONY adapter for all sorts of different configurations. Any help
> greatly appreciated!
>
I'm glad to hear you can get 832 x 624, I thought the limit was 800 x 600.
Enjoy what you've got.
Mark Rogowsky
ro...@forsythe.stanford.edu <-- But not for much longer
ro...@cindy.stanford.edu <-- The QuickMail/SMTP box
John40
Sarjeant) writes:
>IMHO, this is the dumbest thing Apple has done. There is no reason why
>they should have dropped the SCSI, I think Mac users are willing to
>pay an extra hundred dollars to have SCSI in their machine. Hey - it's
>worth it. You can add just about as many SCSI drives to your SCSI
>system as you will ever need. An IDE - on the other hand - is limited
>to 2 drives in a master/slave configuration (and even that is pretty
>iffy).
The 630 has SCSI (for external connections). It is only the internal
drive that is IDE.
John
Tim DeBenedictis
: 2) The manual also says that the machine has a full 1MB of video DRAM,
: not 512K. That means, in theory, that I should be able to run 1024x768
: monitors at 8-bit color. Well, after a little experimentation, I was
: able to get my monitor to do 832x624 at 8-bit. I could not, however,
: get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
: possible that the machine only has 512K of video memory. The fact that I
: could not get the monitor to do 1024x768 makes me think that the computer
: really does only have 512K and not a full 1MB. (I know this monitor
: supports 1024x768 because I've run it on a machine that did before.)
I called Apple's tech support number, and the guy I talked to said that
the machine can in fact drive a 1024x768 monitor at 8-bit color. In fact,
it sould also be able to drive an 832x624 monitor at 16-bit color. However,
I have not been able to get the right combination of settings on the
monitor's adapter to obtain either of these configurations. I know the
monitor is capable of displaying them, since I've seen it done before with
a different Mac.
As tp the IDE drive problem, it's still there. The Apple guy confirmed that
there are no tools for partitioning IDE hard drives (wither Apple or
third party) because until now there have been no Macs with IDE drives!
Apple really needs to write a tool, like its "HD SC setup" program, that can
partition, scan, reformat, mount, unmount, etc. IDE drives.
(HINT TO SHAREWARE DEVELOPERS: WRITE ONE BEFORE APPLE DOES.)
Tim DeBenedictis
tim...@netcom.com
Wayne Folta
>than just using an internal SCSI drive, but I have to defer to Apple on
>this one; I have doubted Apple's wisdom in many things, but I assume that
>they can do their sums correctly.
MacWEEK said that IDE wasn't cheaper, per se, but that it gave Apple more
choices for suppliers, so they could make better deals. Or something like
that, subject to my interpretation.
--
Wayne Folta (fo...@cs.umd.edu)
Mark Rogowsky
Folta) wrote:
MacWeek is wrong then (although I don't recall reading that)... IDE is
cheaper, albeit marginally. The big savings will come when both the hard
drive and the CD drive are on the IDE controller. Then Apple will save
$30-50 per unit at their end, which is real money on a $1200 computer.
Mark
Mark Rogowsky
> Tim DeBenedictis (tim...@netcom.com) wrote:
> : 2) The internal hard drive has closer to 240 MB than the advertized 250.
> : This is not a big deal. What is a big deal is the fact that it is an IDE
> : drive, not a SCSI drive. This means the NONE OF YOUR CURRENT HARD DRIVE
> : TOOLS WILL WORK on it. For instance, you can't repartition the drive with
> : the "Internal HD" format software or the "Apple HD SC setup" software. In
> : fact, all you can do is reformat it entirely. I hope to god that Norton
> : Utils will still work with it, and pray like hell that I don't have a
> : hard-drive crash...
>
> This was design to cut costs. IDE hard drives are typically cheaper
> than SCSI couterparts, although I've found the price difference to be
> minimal.
It is minimal. $10-30 at the OEM level. Add in $20-40 once CD-ROM drives
can be put on the IDE chain and you're talking $30-70 (I said $30-50
elsewhere, but you get the idea). That's several percentage points added
to the gross margin on machines that cheap.
>
> IMHO, this is the dumbest thing Apple has done. There is no reason why
> they should have dropped the SCSI, I think Mac users are willing to
> pay an extra hundred dollars to have SCSI in their machine. Hey - it's
> worth it. You can add just about as many SCSI drives to your SCSI
> system as you will ever need. An IDE - on the other hand - is limited
> to 2 drives in a master/slave configuration (and even that is pretty
> iffy).
Guess what, you have absolutely no clue what you're talking about. First
of all, they didn't drop SCSI at all. There is both an external SCSI port
and an internal SCSI connector (for the CD-ROM drive). Second of all, SCSI
will likely be the exclusive interface on the high end Macs for some time.
Since the 8100 tower (and the next-generation supra-8100 tower) supports
internal arrays, it would need a multi-connector SCSI bus. I don't know
whether those large machines will have any IDE inside (perhaps for the
stock drive?).
Third of all, you are of the same mistaken school that pervades this
group: that money is not a factor. I picked up the newspaper today and
looked at an ad for a 486SX2/66 computer from Compaq for $1100. It
includes a keyboard, some cheap software, and a decent sized hard disk.
Apple needs to compete with machines like that. Witness the Performa 636,
which costs about the same and -- for marketing reasons only -- is touted
by Apple as a 66MHz machine (when all of us who have any grasp of this at
all understand that it's a 33MHz machine). $100 more for any feature is
not going to cut it. Period.
>
> The only thing I can see the IDE (aside from cost reduction) for is
> providing a springboard for a second-generation Houdini board that
> performs on par with any PC clone. In other words, a 100% clone clone
> in your Mac. This clone could run OS/2, Linux, BSD386, MS_DOS, Win NT,
> etc.
Actually, IDE is entirely irrelevant for this purpose. OS/2, MS-DOS,
WindowsNT all support SCSI without any problem. I presume Linux and BSD386
do, too, but I've never used either. Regardless, you don't need IDE for a
super-Houdini.
>
> An interesting prospect, and it will be neat to see what Apple does
> with all of this. I have a sneaky suspicion that the Q630 is going to
> get dropped within 6 months.l
I have a sneaky suspicion that when the next generation of PowerMacs comes
out (and right now MacWeek's Henry Norr seems to be suggesting May, which
I presume is based on some fairly good inside sources) that the Quadra 630
will sport a fancy new logic board with a PowerPC chip on it. The form
factor will likely live for the next half decade or so -- it is the
replacement for both the LC and 6100 form factors.
>
> But this is no reason not to buy one. It's inexpensive and it will
> sport a host of multimedia features (TV, VCR, audio/video heaven). For
> the consumer, it could be neat. For the computer user, I have some
> doubts.
I am glad you see it as acceptable to purchase. The only people who should
shy away at all are those who need large monitor support and/or want a
PowerPC now. The 6100 is cheaper than the 630 + the PowerPC upgrade board.
Mark Rogowsky
> Tim DeBenedictis (tim...@netcom.com) wrote:
>
> : 2) The manual also says that the machine has a full 1MB of video DRAM,
> : not 512K. That means, in theory, that I should be able to run 1024x768
> : monitors at 8-bit color. Well, after a little experimentation, I was
> : able to get my monitor to do 832x624 at 8-bit. I could not, however,
> : get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
> : possible that the machine only has 512K of video memory. The fact that I
> : could not get the monitor to do 1024x768 makes me think that the computer
> : really does only have 512K and not a full 1MB. (I know this monitor
> : supports 1024x768 because I've run it on a machine that did before.)
>
> I called Apple's tech support number, and the guy I talked to said that
> the machine can in fact drive a 1024x768 monitor at 8-bit color. In fact,
> it sould also be able to drive an 832x624 monitor at 16-bit color. However,
> I have not been able to get the right combination of settings on the
> monitor's adapter to obtain either of these configurations. I know the
> monitor is capable of displaying them, since I've seen it done before with
> a different Mac.
The Apple tech is wrong. The 630 does not support 1024 x 768.
>
> As tp the IDE drive problem, it's still there. The Apple guy confirmed that
> there are no tools for partitioning IDE hard drives (wither Apple or
> third party) because until now there have been no Macs with IDE drives!
> Apple really needs to write a tool, like its "HD SC setup" program, that can
> partition, scan, reformat, mount, unmount, etc. IDE drives.
>
Look for some third-party products within 90 days to do just this.
> (HINT TO SHAREWARE DEVELOPERS: WRITE ONE BEFORE APPLE DOES.)
Hint to shareware developers: stay away from this. Writing driver software
and the like is not for the feint of heart. You will get flamed to hell by
anyone who uses your shareware driver and sees the hard disk go kerblooey
do to some unforseen incompatibility.
Matt Burch
>2) The manual also says that the machine has a full 1MB of video DRAM,
>not 512K. That means, in theory, that I should be able to run 1024x768
>monitors at 8-bit color. Well, after a little experimentation, I was
>able to get my monitor to do 832x624 at 8-bit. I could not, however,
>get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
>possible that the machine only has 512K of video memory. The fact that I
>could not get the monitor to do 1024x768 makes me think that the computer
>really does only have 512K and not a full 1MB. (I know this monitor
>supports 1024x768 because I've run it on a machine that did before.)
FACT: The 630's video controller uses dual frame buffers to speed
video capture perfromance.
THEORY: The fact that it has dual frame buffers effectively cuts in
half the "apparent" video memory available, hence the observed
"512K limit" on screen res.
Makes sense to me...
btw, WHERE DID YOU GET YOUR Q630?!? I've been calling every place I
can think of, but it's always "they're not in yet, sir..."
--
-----------------------------------------------------------------------
Matt Burch | There are very few problems that can't be solved
mbu...@ksu.ksu.edu | with the suitable application of photon torpedoes.
-----------------------------------------------------------------------
Eric Sarjeant
: 2) The internal hard drive has closer to 240 MB than the advertized 250.
: This is not a big deal. What is a big deal is the fact that it is an IDE
: drive, not a SCSI drive. This means the NONE OF YOUR CURRENT HARD DRIVE
: TOOLS WILL WORK on it. For instance, you can't repartition the drive with
: the "Internal HD" format software or the "Apple HD SC setup" software. In
: fact, all you can do is reformat it entirely. I hope to god that Norton
: Utils will still work with it, and pray like hell that I don't have a
: hard-drive crash...
This was design to cut costs. IDE hard drives are typically cheaper
than SCSI couterparts, although I've found the price difference to be
minimal.
IMHO, this is the dumbest thing Apple has done. There is no reason why
they should have dropped the SCSI, I think Mac users are willing to
pay an extra hundred dollars to have SCSI in their machine. Hey - it's
worth it. You can add just about as many SCSI drives to your SCSI
system as you will ever need. An IDE - on the other hand - is limited
to 2 drives in a master/slave configuration (and even that is pretty
iffy).
The only thing I can see the IDE (aside from cost reduction) for is
providing a springboard for a second-generation Houdini board that
performs on par with any PC clone. In other words, a 100% clone clone
in your Mac. This clone could run OS/2, Linux, BSD386, MS_DOS, Win NT,
etc.
An interesting prospect, and it will be neat to see what Apple does
with all of this. I have a sneaky suspicion that the Q630 is going to
get dropped within 6 months.l
But this is no reason not to buy one. It's inexpensive and it will
sport a host of multimedia features (TV, VCR, audio/video heaven). For
the consumer, it could be neat. For the computer user, I have some
doubts.
--
============================================================================
sarj...@netcom.com | I'm crying. Sitting on a cornflake | ls -l -a
Eric W. Sarjeant | waiting for the van to come. (Beatles) | Arnold_S*
Bob Knowlden
From: Eric Sarjeant, sarj...@netcom.com
Date: Sat, 6 Aug 1994 17:09:25 GMT
In article <sarjeantC...@netcom.com> Eric Sarjeant,
-a
>Eric W. Sarjeant | waiting for the van to come. (Beatles) |
Arnold_S*
I think there's a false impression here, if not an outright error:
The Q630 series HAS an *external* SCSI port; I'm a bit surprised that
using an internal IDE drive plus a new controller costs less in total
than just using an internal SCSI drive, but I have to defer to Apple on
this one; I have doubted Apple's wisdom in many things, but I assume that
they can do their sums correctly.
Bob Knowlden
(Not speaking for...)
MIT/Lincoln Lab
MacFAQ
> I called Apple's tech support number, and the guy I talked to said that
> the machine can in fact drive a 1024x768 monitor at 8-bit color. In
> fact, it sould also be able to drive an 832x624 monitor at 16-bit color.
> However, I have not been able to get the right combination of settings
> on the monitor's adapter to obtain either of these configurations. I
> know the monitor is capable of displaying them, since I've seen it done
> before with a different Mac.
DataTek sells a switchable VGA adapter that works wonders with my Quadra
605 and a Sony 17" Multiscan. I can easily run it in 19" (1024 X 768)
mode. Because it's a switchable adapter, you can use it with a variety of
Macs, monitors, and resolutions. The adapter runs fifteen dollars or so. I
don't have DataTek's phone number at home, or I'd post it. Email me if you
need it.
Les Jones
Author of the AOL and ZTerm FAQs
available at sumex in info-mac/comm/info
Eric Sarjeant
: In article <sarjeantC...@netcom.com>, sarj...@netcom.com (Eric
: Sarjeant) wrote:
: > Tim DeBenedictis (tim...@netcom.com) wrote:
: > : 2) The internal hard drive has closer to 240 MB than the advertized 250.
: > : This is not a big deal. What is a big deal is the fact that it is an IDE
: > : drive, not a SCSI drive. This means the NONE OF YOUR CURRENT HARD DRIVE
: > : TOOLS WILL WORK on it. For instance, you can't repartition the drive with
: > : the "Internal HD" format software or the "Apple HD SC setup" software. In
: > : fact, all you can do is reformat it entirely. I hope to god that Norton
: > : Utils will still work with it, and pray like hell that I don't have a
: > : hard-drive crash...
: >
: >
: > IMHO, this is the dumbest thing Apple has done. There is no reason why
: > they should have dropped the SCSI, I think Mac users are willing to
: > pay an extra hundred dollars to have SCSI in their machine. Hey - it's
: > worth it. You can add just about as many SCSI drives to your SCSI
: > system as you will ever need. An IDE - on the other hand - is limited
: > to 2 drives in a master/slave configuration (and even that is pretty
: > iffy).
: Guess what, you have absolutely no clue what you're talking about. First
: of all, they didn't drop SCSI at all. There is both an external SCSI port
: and an internal SCSI connector (for the CD-ROM drive). Second of all, SCSI
: will likely be the exclusive interface on the high end Macs for some time.
: Since the 8100 tower (and the next-generation supra-8100 tower) supports
: internal arrays, it would need a multi-connector SCSI bus. I don't know
: whether those large machines will have any IDE inside (perhaps for the
: stock drive?).
Then Apple is even dumber than I thought. Why add the unecessary
confusion by including IDE in a system that is clearly equipped to
handle SCSI?
As has already been said, all of your existing SCSI tools will not
work. And why does SCSI have to be an "exclusive" club? You must be
kidding me, SCSI has been part of the Mac almost from the getgo. It's
not exclusive by any means, you can get 1 gig SCSI for $700 now.
That's clearly on par with IDE.
: Third of all, you are of the same mistaken school that pervades this
: group: that money is not a factor. I picked up the newspaper today and
: looked at an ad for a 486SX2/66 computer from Compaq for $1100. It
: includes a keyboard, some cheap software, and a decent sized hard disk.
: Apple needs to compete with machines like that. Witness the Performa 636,
: which costs about the same and -- for marketing reasons only -- is touted
: by Apple as a 66MHz machine (when all of us who have any grasp of this at
: all understand that it's a 33MHz machine). $100 more for any feature is
: not going to cut it. Period.
Oh, money is definitely a factor. I would like to buy a PowerMac 8100 for
$10, but it's not going to happen by replacing the internal SCSI with
an IDE. That's the bottom line.
Apple should be working on volume, that's part of the reason SCSI hard
drive prices have been falling (and that is the reason IDE has
always been so cheap).
The Mac needs to prove itself a Mac. IMHO, adding IDE is Apple's
attempt to prove how dumb they can be and is not a positive reflection
on the Mac in general. If you have been watching the PC arena, the IDE
interface is on the way out. There is already a new IDE level 2
standard to support larger hard drive size, and further "enhancements"
loom on the horizon. Apple is joining the wrong bandwagon at the wrong
time.
Maybe 5 years ago I could have seen an IDE in a Mac, but with the
price difference so minimal (and Apple so willing to include SCSI in
the system anyway), I don't understand the reasoning.
Probably the 630 will sport a PPC chip very shortly, but I suspect
that Apple will rethink the IDE hard drive. If you've already got the
SCSI controller, you may as well use a SCSI drive. It may seem like a
moot point, or maybe even a quibble, but it's silly to throw another
standard into the Mac. The Mac has already been well standardized on
the SCSI, and Apple needs to apply some leverage in the PC marketplace
with the SCSI feature.
SCSI is a standard. Sure IDE may be faster because it doesn't have to
deal with a host adaptor, but you could always go to a Fast SCSI. The
performance benefits of IDE (ATA) are so marginal that you really
cannot percieve it. The real benefit of SCSI in your Mac is that (up
until now) your Mac has been SCSI and nothing else.
: I am glad you see it as acceptable to purchase. The only people who should
: shy away at all are those who need large monitor support and/or want a
: PowerPC now. The 6100 is cheaper than the 630 + the PowerPC upgrade board.
Thank you!
There's nothing wrong with the Q630 in general, it's just Apple's
silly selection of an IDE hard drive instead of a SCSI. Since SCSI is
running the CD-ROM internally, you may even be able to get around it.
--
============================================================================
Ron Ross
ro...@forsythe.stanford.edu (Mark Rogowsky) wrote:
> The form
> factor will likely live for the next half decade or so -- it is the
> replacement for both the LC and 6100 form factors.
I'm surprised to hear you say this, since no Apple form factor I can think
of has survived for as long as five years, with the exception of the 128K
-> Classic II box.
Myself, I expect the "centris" slim line box to go next year and possibly
the "vx" box as well (purely speculation). What's your take on this?
--
Ron Ross ron...@panix.com CIS: 73060,373
Erik A. Speckman
Ron Ross <ron...@panix.com> wrote:
>In article <rogo-06089...@cindy.stanford.edu>,
>ro...@forsythe.stanford.edu (Mark Rogowsky) wrote:
>
>> The form
>> factor will likely live for the next half decade or so -- it is the
>> replacement for both the LC and 6100 form factors.
>I'm surprised to hear you say this, since no Apple form factor I can think
>of has survived for as long as five years, with the exception of the 128K
>-> Classic II box.
Margins aren't what they used to be and new case designs cost money.
Still, 5 years seems a bit extreme to me as well.
>Myself, I expect the "centris" slim line box to go next year and possibly
>the "vx" box as well (purely speculation). What's your take on this?
I still wonder if we will se some x100 "plus" machines that use the same
cases as the existing PowerMacs an only differ by being 10-20% faster than
the existing models.
Excepting that possibility, I was under the impression that none of the
PowerMac form factors would survive the next major release of PowerPC
machines. I though the 8XXX case was going to make way for a slightly
larger mini tower and I thought the middle desktop would be redesigned as
well.
I don't really care. I din't expect to be able to upgrade my C610 to the
second generation anyway.
Mark Rogowsky
> : Guess what, you have absolutely no clue what you're talking about. First
> : of all, they didn't drop SCSI at all. There is both an external SCSI port
> : and an internal SCSI connector (for the CD-ROM drive). Second of all, SCSI
> : will likely be the exclusive interface on the high end Macs for some time.
> : Since the 8100 tower (and the next-generation supra-8100 tower) supports
> : internal arrays, it would need a multi-connector SCSI bus. I don't know
> : whether those large machines will have any IDE inside (perhaps for the
> : stock drive?).
>
> Then Apple is even dumber than I thought. Why add the unecessary
> confusion by including IDE in a system that is clearly equipped to
> handle SCSI?
To save money on the drives -- and the CD readers in the future. Period. It
does not increase confusion for the 99+% of Mac owners who never change out
their internal hard drives.
>
> As has already been said, all of your existing SCSI tools will not
> work. And why does SCSI have to be an "exclusive" club? You must be
> kidding me, SCSI has been part of the Mac almost from the getgo. It's
> not exclusive by any means, you can get 1 gig SCSI for $700 now.
> That's clearly on par with IDE.
I have no idea how you chose to quote "exclusive" in the manner that you
did. My point, quoted above, is that on high-end Macs, there will be no IDE
(in all probability) because of the need for complex internal SCSI buses.
And, again, to note your price quote, the issue is (a) pricing on the low
end, for the included internal drive and (b) pricing at the OEM level, what
Apple pays wholesale. I'm not necessarily that excited about having IDE,
but that $20 per machine more that Apple can make adds up over 750,000
Macs... It's almost pure margin, i.e. pure profit, and in part means the
difference between a new millenium with Apple and a new millenium without
it. Worth the tradeoff, it says here.
>
> : Third of all, you are of the same mistaken school that pervades this
> : group: that money is not a factor. I picked up the newspaper today and
> : looked at an ad for a 486SX2/66 computer from Compaq for $1100. It
> : includes a keyboard, some cheap software, and a decent sized hard disk.
> : Apple needs to compete with machines like that. Witness the Performa 636,
> : which costs about the same and -- for marketing reasons only -- is touted
> : by Apple as a 66MHz machine (when all of us who have any grasp of this at
> : all understand that it's a 33MHz machine). $100 more for any feature is
> : not going to cut it. Period.
>
> Oh, money is definitely a factor. I would like to buy a PowerMac 8100 for
> $10, but it's not going to happen by replacing the internal SCSI with
> an IDE. That's the bottom line.
Again, you are without a clue. Your analogy is so far from relevant I won't
even discuss it in detail. Suffice it to say that your original misguided
point was about an additional $100 wholesale cost for a SIMM slots. I
explained that $100 is way too much. Apple is looking to save $10 and $20
at a time in manufacturing costs since every $10 is one more point on the
gross-margin chart -- it may be the difference between profitability of the
line and non-profitability.
>
> Apple should be working on volume, that's part of the reason SCSI hard
> drive prices have been falling (and that is the reason IDE has
> always been so cheap).
Apple can't make SCSI as cheap as IDE now any more than it could in the
past. It is going for volume with low-cost, high-performance,
feature-packed machines like the Q630. It gets there with tradeoffs: IDE
sted SCSI, DRAM sted VRAM, single SIMM slot, LC slot sted NuBus/PCI/040
PDS, etc. etc.
>
> The Mac needs to prove itself a Mac. IMHO, adding IDE is Apple's
> attempt to prove how dumb they can be and is not a positive reflection
> on the Mac in general. If you have been watching the PC arena, the IDE
> interface is on the way out. There is already a new IDE level 2
> standard to support larger hard drive size, and further "enhancements"
> loom on the horizon. Apple is joining the wrong bandwagon at the wrong
> time.
>
Apple will support the EIDE standard when it becomes relevant to do so.
EIDE is a superset of IDE and means IDE is stronger than ever, not on its
way out. Read any of the PC trade press, and you'll see that it all agrees
that IDE's renaissance means SCSI will be more common, but not as
ubiquitous as IDE. Apple is joining the bandwagon to save money and take
advantage of economies of scale in IDE drive manufacturing.
> Maybe 5 years ago I could have seen an IDE in a Mac, but with the
> price difference so minimal (and Apple so willing to include SCSI in
> the system anyway), I don't understand the reasoning.
You don't understand so I'll try it again using round, simple numbers.
[This is oversimplified, world, so don't flame me, OK...] Let's say the
Quadra 630 sells for $1000 to distributors. Let's say it cost Apple $800 to
build it using an internal SCSI drive, 2 SIMM slots, VRAM, and a NuBus
slot. The gross margin is then $200, 25% of $800. The net margin is far
smaller, since it gets chewed up by other expenses like marketing and what
not (I forget how margin calcs work, but you get the idea). Now, let's say
they use IDE, 1 SIMM slot, DRAM and an LC slot. The cost is now $700. But
they can sell it for the same $1k. Now, the gross margin is $300, or 43%.
Even if they now use the cost savings to keep prices low and sell it for
$900. The gross margin is $200, or 28% -- a 3-point jump, which is
significant.
>
> Probably the 630 will sport a PPC chip very shortly, but I suspect
> that Apple will rethink the IDE hard drive. If you've already got the
> SCSI controller, you may as well use a SCSI drive. It may seem like a
> moot point, or maybe even a quibble, but it's silly to throw another
> standard into the Mac. The Mac has already been well standardized on
> the SCSI, and Apple needs to apply some leverage in the PC marketplace
> with the SCSI feature.
You are completely wrong, which no longer shocks me. When the PPC chip is
on board in a 630 form factor, IDE will be spreading throughout the Mac
line like that mystery virus in "The Stand." It will be used in almost
every Powerbook (perhaps all of them), and perhaps other desktop machines.
It will use for the CD-ROM in the 630 box as well as the hard disk. The
"SCSI feature" will still be available externally on all Macs (save the
dockless Duos) and internally on all high-end (and probably medium-end)
desktops.
>
> SCSI is a standard. Sure IDE may be faster because it doesn't have to
> deal with a host adaptor, but you could always go to a Fast SCSI. The
> performance benefits of IDE (ATA) are so marginal that you really
> cannot percieve it. The real benefit of SCSI in your Mac is that (up
> until now) your Mac has been SCSI and nothing else.
EIDE is faster than Fast SCSI-2, although Wide and Fast SCSI-2 is faster
than EIDE. This is all irrelevant in the way IDE is being used in the 630,
where it outperforms regular SCSI. EIDE drives are cheaper than Fast SCSI-2
drives, even if its just by a little. And as for your "real benefit," it's
gone so come to grips with it. Future Macs may well support IDE, SCSI and
FireWire -- a trio of hard-disk connectivity options. Perhaps they'll even
add Fiber Channel, for a foursome!
> There's nothing wrong with the Q630 in general, it's just Apple's
> silly selection of an IDE hard drive instead of a SCSI. Since SCSI is
> running the CD-ROM internally, you may even be able to get around it.
You can get around it. A future 630-box Mac will use IDE for the CD-ROM. At
that point, there may or may not be an internal SCSI connector, I can't say
for sure.
Mark Rogowsky
ro...@forsythe.stanford.edu <-- Currently reflects to the QuickMail box
Mark Rogowsky
ron...@panix.com (Ron Ross) wrote:
> In article <rogo-06089...@cindy.stanford.edu>,
> ro...@forsythe.stanford.edu (Mark Rogowsky) wrote:
>
> > The form
> > factor will likely live for the next half decade or so -- it is the
> > replacement for both the LC and 6100 form factors.
>
> I'm surprised to hear you say this, since no Apple form factor I can think
> of has survived for as long as five years, with the exception of the 128K
> -> Classic II box.
You may be right. Five years could be a bit much. The LC seems to be on the
way out in 1994-early 1995. When did it come? I thought that was 1990 when
the first "low-cost" Macs came out. That would be 4+ years for LC...
>
> Myself, I expect the "centris" slim line box to go next year and possibly
> the "vx" box as well (purely speculation). What's your take on this?
The Centris slim is absolutely terminal with the 630 intro. The info in
MacWeek implied, and I'll agree at this point, that the vx/7100 box would
survive, perhaps with minor modifications, into the next generation. As a
3-slot desktop machine seems in the offing for the business standard, this
all makes sense.
Mark Rogowsky
ro...@forsythe.stanford.edu <-- Currently reflects to the QuickMail box
Mark Rogowsky
wrote:
You can switch the adapter all day and it will never do 1024 x 768 on a
Quadra 630. Since that is what this thread is about, I thought I'd rain on
this notion before it gains any credence. On the Q605 and many other Macs,
17" multiscans can often reach the 1024 x 768 resolution.
Mark Rogowsky
Speckman) wrote:
> In article <ronross-0708...@ronross.dialup.access.net>,
> Ron Ross <ron...@panix.com> wrote:
> >I'm surprised to hear you say this, since no Apple form factor I can think
> >of has survived for as long as five years, with the exception of the 128K
> >-> Classic II box.
>
> Margins aren't what they used to be and new case designs cost money.
> Still, 5 years seems a bit extreme to me as well.
See my other post. I'll go with 4 years.
>
> >Myself, I expect the "centris" slim line box to go next year and possibly
> >the "vx" box as well (purely speculation). What's your take on this?
>
> I still wonder if we will se some x100 "plus" machines that use the same
> cases as the existing PowerMacs an only differ by being 10-20% faster than
> the existing models.
I don't believe there will ever be one in the 6100 box. I could be wrong.
If the Big Five Intro slated for next year really is pushed back to May,
there may be some interim steps, but even that seems unlikely.
>
> Excepting that possibility, I was under the impression that none of the
> PowerMac form factors would survive the next major release of PowerPC
> machines. I though the 8XXX case was going to make way for a slightly
> larger mini tower and I thought the middle desktop would be redesigned as
> well.
Here I disagree, based primarily on MacWeek. The Tsunami/TNT machines are
slated for the larger-than-8100 box. The Nitro machine is alleged to use
the 8100 box. The Catalyst machine is alleged to use a 7100-like box --
perhaps virtually identical. The Alchemy machine, i.e. the multi-form
factor machine, is almost certainly slated for the 630 and the LC 500
series.
>
> I don't really care. I din't expect to be able to upgrade my C610 to the
> second generation anyway.
Good for you! Betting on upgrades before the machines are even out is poor
gambling. Buying a IIvx, like I did, after the 7100 intro and upgrading it
proved pretty shrewd :-).
Mark Rogowsky
ro...@forsythe.stanford.edu <-- Currently reflects to the QuickMail box
TedHowe
ron...@panix.com (Ron Ross) writes:
>> The form
>> factor will likely live for the next half decade or so -- it is the
>> replacement for both the LC and 6100 form factors.
>
>I'm surprised to hear you say this, since no Apple form factor I can
think
>of has survived for as long as five years, with the exception of the 128K
>-> Classic II box.
>Myself, I expect the "centris" slim line box to go next year and possibly
>the "vx" box as well (purely speculation). What's your take on this?
Actually, there is one other form factor that lasted at least five years:
The IIcx/IIci/Quadra 700 form factor was at least 5 years - the IIci
lasted nearly 4 years itself if I'm not mistaken.
Ted Howe
ted...@aol.com
Ron Ross
> I don't really care. I din't expect to be able to upgrade my C610 to the
> second generation anyway.
Apropos of this, I spoke to Apple's PMac product mgr at MacWorld and
suggested that it was more important than ever to keep the 6100 upgrades
available through the next round of PMacs. He wasn't too encouraging
about the prospect :(
Ron Ross
> The IIcx/IIci/Quadra 700 form factor was at least 5 years - the IIci
> lasted nearly 4 years itself if I'm not mistaken.
I make it '89-'93.
L.H....@student.lut.ac.uk
>In article <timmydCu...@netcom.com>, tim...@netcom.com (Tim
>DeBenedictis) wrote:
>
>>
>> 2) The manual also says that the machine has a full 1MB of video DRAM,
>> not 512K. That means, in theory, that I should be able to run 1024x768
>> monitors at 8-bit color. Well, after a little experimentation, I was
>> able to get my monitor to do 832x624 at 8-bit. I could not, however,
>> get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
>> possible that the machine only has 512K of video memory. The fact that I
>> could not get the monitor to do 1024x768 makes me think that the computer
>> really does only have 512K and not a full 1MB. (I know this monitor
>> supports 1024x768 because I've run it on a machine that did before.)
>>
>
>Although the machine might have 1MB of RAM available for video, this
>doesn't necessarily mean that the hardware supports the higher resolutions
>or bit depths. (As you indicated, it's only THEORETICALLY possible.) Both
>MacUser and MacWorld state that the largest monitor that the 630 supports
>with its internal video is 15". Since the Apple 15" has a max resolution
>of 832x624, I have a feeling that nothing above that is going to work.
>(Since this is the same resolution as the Apple 16", does the 16" work on
>a 630?) I admit it's strange that they'd give it the memory and then
>cripple the video, but hey, this is Apple we're talking about.
>
L.
Ask yourself what a PowerMac DRAM video can do.
--
_____________________________________________________________________________
L.H....@student.lut.ac.uk Email me for a copy of the Mac screensaver FAQ
L.H....@student.lut.ac.uk
>Tim, WRT your monitor troubles: it seems entirely plausible that the
>video hardware in the 630 just plain *doesn't support* 1024x768. It's
>not a common res for Apple video, and the only times I've seen it work
>with most Macs is via some cable voodoo on Q700s and Q900s. I think
>the PowerMacs might support that res, but on the 630, I'm not sure.
>
>Don Walli
>dwa...@mtu.edu
>
>Tim DeBenedictis (tim...@netcom.com) wrote:
>:
>: [stuff deleted]
>:
>: 2) The manual also says that the machine has a full 1MB of video DRAM,
>: not 512K. That means, in theory, that I should be able to run 1024x768
>: monitors at 8-bit color. Well, after a little experimentation, I was
>: able to get my monitor to do 832x624 at 8-bit. I could not, however,
>: get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
>: possible that the machine only has 512K of video memory. The fact that I
>: could not get the monitor to do 1024x768 makes me think that the computer
>: really does only have 512K and not a full 1MB. (I know this monitor
>: supports 1024x768 because I've run it on a machine that did before.)
I was under the impression that the 630 used DRAM-based, not VRAM-based
video, and that that imposes the same limitations on video suuport that the
PowerMac DRAM ports have.
L.
Colin Stuckless
>>IMHO, this is the dumbest thing Apple has done. There is no reason why
>>they should have dropped the SCSI, I think Mac users are willing to
>I think there's a false impression here, if not an outright error:
>The Q630 series HAS an *external* SCSI port; I'm a bit surprised that
>using an internal IDE drive plus a new controller costs less in total
>than just using an internal SCSI drive, but I have to defer to Apple on
>this one; I have doubted Apple's wisdom in many things, but I assume that
>they can do their sums correctly.
This is correct... the SCSI controller is still in there, hence the
external SCSI port. I'd be willing to bet you could even remove the
internal IDE drive and put a SCSI in there if you wanted to. It'd
just be a simple cabling issue.
As far as adding a new controller (IDE) to cut costs, I've been
told that adding IDE to a motherboard adds a few dollars at most to the
cost at the manufacturing level. And while the price difference between
IDE and SCSI drives aren't real huge, I would imagine that with the
volumes that a company like Apple would be be dealing with it
could make a differnce... especially in the notebook market, where
I believe there's a larger gap between SCSI/IDE pricing (2.5" drives).
--
___________________________________
Colin Stuckless col...@cs.mun.ca
Memorial Univeristy of Newfoundland
Michael Smith
>The Q630 series HAS an *external* SCSI port; I'm a bit surprised that
>using an internal IDE drive plus a new controller costs less in total
>than just using an internal SCSI drive, but I have to defer to Apple on
>this one; I have doubted Apple's wisdom in many things, but I assume that
>they can do their sums correctly.
Bear in mind that an IDE "controller" is usually a 1-of-8 decoder and a
16-bit bidirectional bus transciever - about 30c worth of parts, and
easily implemented in some spare corner of a gate array.
I won't soapbox about IDE, but for this application, surprisingly, it's
actually a reasonable design decision.
>Bob Knowlden
--
# mike smith : mi...@apanix.apana.org.au - Silicon grease monkey #
# "The question 'why are the fundamental laws of nature mathematical' #
# then invites the trivial response 'because we define as fundamental #
# those laws which are mathematical'". Paul Davies, _The_Mind_of_God_. #
Brian V. Hughes
wrote:
>Ask yourself what a PowerMac DRAM video can do.
While this is analogous, they are not the same situation. The 630
series does reserve 512k DRAM for video display, the Power Macs, however,
reserve 768k DRAM. That is why they can drive a 640x480 monitor in
Thousands of colors where the 630 can only do 256 colors.
-Hades
Brian A. Cole
]
] In article <Cu49r...@lut.ac.uk>, L.H....@student.lut.ac.uk (Lloyd Wood)
This isn't the whole story. The DRAM-based video on the 6100/7100/8100
steals 768K of the system RAM. (In this regard, they are like the IIci
and the IIsi).
However, the 63x machines have a separate bank of DRAM chips soldered to
the motherboard that is used exclusively for the video frame buffer.
(In this regard, it is like most PC-clone video cards).
Remember, VRAM is _not_ faster than DRAM as far as read and write speeds.
The reason VRAM is "faster" than DRAM for video is because it is dual-
ported, and thus can be written (say by an app) and read (to update the
screen) at the same time. However, Apple has implemented some special
video caching circuitry on the 63x machines so that alleviates most of
the throughput problems associated with single-ported DRAM (this is what
the trade rags say, anyway). If Apple did a good job, it should be only
minutely slower than the standard VRAM-based video subsystem they could
have designed for the 63x machines.
The video subsystems in the 660av and the 840av are VRAM-based, but I
think you would have to agree that the 63x's video subsystem is better.
On the 660av/840av, the VRAM is divided into two smaller (of course)
chunks when you have the video-in displayed in an on-screen window, thus
limiting the bit-depth available to you. The 63x can perform this feat
without dividing the VRAM -- an amazing feat -- I'd like to see how they
did that.
brian
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~ _______ _____ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Brian A. Cole |_ _ | | ___| http://www.cs.wisc.edu/~tuc/
Head Consultant | | | |__| |___ The speed of time is
t...@cs.wisc.edu |_| |__________| 3600 seconds per hour.
Jude Giampaolo
We just got a 630 at work today and it runs just fine in thousands of
colors mode at 640x480.
--
Jude Giampaolo Case Western Reserve University
jc...@po.cwru.edu Cleveland, Ohio
giam...@snowhite.eeap.cwru.edu Electrical Engineering
Name withheld upon request
have vram, either 1 or 2 megs depending on whether or not they are a av.
Since I brought up the P.M, what is the software that is needed for
giving voice commands to a non av? My school said they werent even sure
if the *AV's* come with softwill that enables it to receive voice
commands (but it does come with text to speech software). How does a
person with a non av PM get both text to speech and speech to text software?
Brian A. Cole
]
] The powermacs have vram instead of dram? Mt schiool price list says they
The non-AV 6100 has a DRAM-based video subsystem. The other PowerMacs
have two seperate video subsystems, one DRAM-based, the other VRAM based.
If you happe two have two monitors, you can plug them both in to a PowerMac
right out of the box (the non-AV 6100 excluded). Well, alomost right out
of the box -- you might need an adapter to use the DRAM-based video
subsystem's funky AudioVision connector.
Bill Rosenberg
wrote:
> In article <timmydCu...@netcom.com>, tim...@netcom.com (Tim
DeBenedictis) writes:
> >
> > I need to make a few corrections here:
> >
> >
> > 2) The manual also says that the machine has a full 1MB of video DRAM,
> > not 512K. That means, in theory, that I should be able to run 1024x768
> > monitors at 8-bit color. Well, after a little experimentation, I was
> > able to get my monitor to do 832x624 at 8-bit. I could not, however,
> > get 1024x768 at all. Now, 832x624=519168 bytes or 507K, so it's clearly
> > possible that the machine only has 512K of video memory. The fact that I
> > could not get the monitor to do 1024x768 makes me think that the computer
> > really does only have 512K and not a full 1MB. (I know this monitor
> > supports 1024x768 because I've run it on a machine that did before.)
> >
> > Does anyone know the facts here? My monitor is a SONY multiscan 15sf,
> > with a SONY adapter for all sorts of different configurations. Any help
> > greatly appreciated!
> >
> > Tim DeBenedictis
>
> According to Apple Service Souce, the 630 does have 1 MB DRAM frame buffer.
> This gives 16-bit on 12",13",14",15" and VGA (640*480) monitors
> You get 8-bit on 16" and SVGA (800*600) monitors.
>
> Those are the facts
>
> Kevin
With a 1 MB frame buffer, shouldn't you be able to get 16-bit color at
800x600 resolution? The way I understand it, the required RAM would be:
800*600*16 = 7680000 bits = 960000 bytes
Am I missing something obvious here?
Bill
Stephen Russell
> I was under the impression that the 630 used DRAM-based, not VRAM-based
> video, and that that imposes the same limitations on video suuport that the
> PowerMac DRAM ports have.
>
It's not part of the standard DRAM used for application/data memory.
Stephen Russell
(Jude Giampaolo) wrote:
> In article <hades-08089...@fpcgb-kip9.dhmc.dartmouth.edu>,
> Brian V. Hughes <ha...@dartmouth.edu> wrote:
> >In article <Cu49r...@lut.ac.uk>, L.H....@student.lut.ac.uk (Lloyd Wood)
> >wrote:
> >
> >>Ask yourself what a PowerMac DRAM video can do.
> >
> > While this is analogous, they are not the same situation. The 630
> >series does reserve 512k DRAM for video display, the Power Macs, however,
> >reserve 768k DRAM. That is why they can drive a 640x480 monitor in
> >Thousands of colors where the 630 can only do 256 colors.
>
> We just got a 630 at work today and it runs just fine in thousands of
> colors mode at 640x480.
It is 1MB of DRAM, not 512k. But it is dedicated DRAM, on the logic board,
not part of the regular system memory.
Brian A. Cole
]
] I, for one, am not disappointed that Apple has started using IDE in
] addition to SCSI. It gives you more choices (you can replace your
] internal IDE or get an external SCSI) and lets you save money in the
] cases where IDE is cheaper (I believe that although the difference
] between SCSI and IDE is not as significant as it once was, it is still
] enough to make a difference to most buyers). Don't react to IDE as if
] it's something horrible just because THEY (those clone makers) use it.
You're right that "it's bad because PC-clones use it" makes no sense.
The way I see it, there are two valid reasons no to like IDE in Macs:
1) The technology is inferior.
Yeah, you can make these arguments, but the don't amount to much.
The transfer/access times are about the same as SCSI, and 95% of
mac users won't even notice the difference. Let's just say that
IDE is not as elegant as SCSI and leave it at that. Yes, elegence
is semi-important to some Mac buyers, but no big deal on this front.
2) It's a support nightmare.
This is much more important, and something that business purchasers
will have to think about hard before making buying decisions. Even for
home buyers, this is going to be a pain sometimes. Consider just the
kind of questions we get here on c.m.s.h, like "can I put my old HD I
just replaced in an enclosure and use it as an external drive?" -- the
answer would have been "yes" but now it's "maybe. you can unless you
have a q630, p630, p636, or pb150".
IMHO, there was no reason for Apple to use IDE in the 63x line. The price
difference is almost nil for 3.5" drives. I do, however, think they were
fully justified in using IDE drives on the pb150 (2.5" drives have a much
bigger price delta between SCSI and IDE). Maybe Apple was thinking some-
thing along the lines of "We're going to FireWire in a year or two anyway,
so IDE will be good practice supporting multiple HD protocols" -- it's
possible.
brian
PS. I've been trying to get a straight answer for this question for a while,
so I'll ask it again:
Is it possible to yank out the IDE HD from a q630 and replace it
with a SCSI drive? The q630 must have an internal SCSI ribbon
for the CD-ROM drive, but I'm unsure if:
1) replacing the two-connector ribbon with a standard three-
connector ribbon will work
2) termination issues can be resolved
3) the q630's motherboard can handle not having an IDE drive
connected to it
4) The replaced IDE drive (Apple ROMs?) will work in my father's
PC-clone.
Please let me know if you have answers to any of these questions.
(It's less likely to be possible, but I'd also like to know it it's
possible to replace a pb150's IDE HD with a SCSI hd).
Ishir Bhan
addition to SCSI. It gives you more choices (you can replace your
internal IDE or get an external SCSI) and lets you save money in the
cases where IDE is cheaper (I believe that although the difference
between SCSI and IDE is not as significant as it once was, it is still
enough to make a difference to most buyers). Don't react to IDE as if
it's something horrible just because THEY (those clone makers) use it.
Ishir Bhan finger ib...@husc.harvard.edu
ib...@digitas.org
Valentin Richter
Bill_Ro...@qmail4.sp.trw.com (Bill Rosenberg) wrote:
> With a 1 MB frame buffer, shouldn't you be able to get 16-bit color at
> 800x600 resolution? The way I understand it, the required RAM would be:
>
> 800*600*16 = 7680000 bits = 960000 bytes
>
> Am I missing something obvious here?
No, you don't. My Performa 475 has 1MB of VRAM and it does 16bit color on
my Apple16" (832 * 624 pixels) just fine. It let me even connect to a 21"
monitor (1152 * 870 pixels) with 8bit color. So why Apple didn't take the
monitor capabilities of the 475 series for the 630 and made everything
just faster and more expandable is plain mystery to me.
Well, not plain, I think manufacturing costs had to do with it. But I
think their decision was penny wise and pound foolish. It makes the 630 an
easier buy because it costs a couple of dollars less but what about the
damage an unsatisfied customer can do?
Valentin
-----
Valentin Richter
ric...@informatik.uni-muenchen.de
L.H....@student.lut.ac.uk
Seems like the easiest way to avoid the CPU/video lockout that is noticeable
on the IIci and IIsi, at the expense of some wasted RAM.
A. D. Anbar
t...@calico.cs.wisc.edu (Brian A. Cole) writes:
> 2) It's a support nightmare.
> This is much more important, and something that business purchasers
> will have to think about hard before making buying decisions.
Yes, but if price keeps the machines out of consideration for business
users in the first place, then no-one's going to get to asking about
support costs. Anything that makes Macs more price-competative without
sacrificing performance and important features should be welcomed. And,
for someone who's informed (i.e. a professional consultant), this
shouldn't be too major a support headache. Support costs for Macs should
still be far below those of PCs.
Try and make this concrete: if a client asked you whether they should go
out and purchase 10 Q630s or 10 486's, would the IDE drive in the Q630 be
anywhere near as important a consideration in your recommendation as the
cost of the machines? Would it be in the picture at all? I think we're
talking 2nd or 3rd order concerns, here.
> Even for
> home buyers, this is going to be a pain sometimes.
"Sometimes", at worst. More like "rarely".
> Consider just the
> kind of questions we get here on c.m.s.h, like "can I put my old HD I
> just replaced in an enclosure and use it as an external drive?" -- the
> answer would have been "yes" but now it's "maybe. you can unless you
> have a q630, p630, p636, or pb150".
Yes, but those of us on c.m.s.h are hardly a typical cross-section of Mac
users! The vast majority never expand beyond the hard drive in the
original machine. If they do, most will just buy an external drive
anyway. Since Apple's not dropping external SCSI, Macs still have an
edge over PC's in ease of expandability here.
Ariel
---------------------------------------------------------------------
A. D. Anbar anb...@caltech.edu
Mail Stop 170-25 Tel: (818) 395-6982
Division of Geological and Planetary Sciences Fax: (818) 796-9823
California Institute of Technology
Pasadena, CA 91125
---------------------------------------------------------------------
A graduate education teaches you how to REASON, not how to THINK!
---------------------------------------------------------------------
Dan Magorian
Aug 1994 16:48:38 GMT
All you avid MacWeek junkies missed an important point in this IDE thread.
Don't you all think people at Apple hated the idea of IDE, due to it's
technical inferiority (ever tried to attach 2 from different mfgrs in a PC?
Esp Connor, they're incompatible with everyone else's. The whole thing is a
wretched PC hardware kludge) and the extra work of emulating the old ST506
command set in PC Roms?
Of course they did, and it's doubtful that the mgr cost difference would have
forced the change. What did is the negotiations for the PowerPC Reference
Platform. Basically, IBM demanded that it (PREP) have IDE, and Apple demanded
that it have ADB (Desktop bus). So it has both.
Not sure whether the 630 is the first PREP mac, or just on the way there.
Dan
Nathan W. Fullerton
>The video subsystems in the 660av and the 840av are VRAM-based, but I
>think you would have to agree that the 63x's video subsystem is better.
I agree that the 630 uses a more clever method, but not a better one.
See below.
>On the 660av/840av, the VRAM is divided into two smaller (of course)
>chunks when you have the video-in displayed in an on-screen window, thus
>limiting the bit-depth available to you. The 63x can perform this feat
>without dividing the VRAM -- an amazing feat -- I'd like to see how they
>did that.
They did it by limiting the bit depth at higher resolutions and limiting
the video capture resolution to 320x240.
((640x480x2) + (320x240x3))<1MB, ((832x624x1) + (320x240x3))<1MB,
((832x624x2) + (320x240x3))>1MB
On the 630, since there's not enough video RAM (yes, I know it's DRAM)
for the video-in buffer and 832x624@thousands, you simply don't get the
option to use thousands. On the 840 & 660, you can turn off the
reserved video-in buffer to use the the full VRAM for graphics.
On the 630, if you increase the video window size above 320x240, the
pixels in the video capture window are doubled. On the 660 & 840 (even
with 1MB of VRAM) you can go up to ~480x360 before the lines double.
The increase in quality is noticeable.
I was really psyched when I read the magazine article at first, but on
further analysis it appears that they traded 832x624x16bit capability
for 640x480x16bit + video capture. I'd rather suffer the inconvenience
on a depth switch than not be able to run 16" mode in thousands. Of
course, the 630's system is easier for a new home user to understand.
The basic 630 info here (320x240 video-in resolution limit, graphics
mode restrictions) came straight from the Apple PR on
ftp.support.apple.com.
Peace,
Nathan
Multimedia Lab, Robotics Institute, Carnegie Mellon University, Pittsburgh, PA
(412)268-6972, http://videomedia.pc.cs.cmu.edu/
Erik A. Speckman
Bill Rosenberg <Bill_Ro...@qmail4.sp.trw.com> wrote:
>In article <1994Aug5...@acf6.acf.nyu.edu>, edwa...@acf6.acf.nyu.edu
>With a 1 MB frame buffer, shouldn't you be able to get 16-bit color at
>800x600 resolution? The way I understand it, the required RAM would be:
>
>800*600*16 = 7680000 bits = 960000 bytes
>
>Am I missing something obvious here?
Maybe, people have already mentioned other possible reasons (ie.
compromises for video capture), but another possibility is memory
organization and bandwidth.
The 800*600*16 bit screen is refreshed at 60-75 HZ. That is aproximately
65 Mbytes/second just to read out the image to the monitor.
I think this represents a significant amount of the bandwidth available
from a single 4 byte wide bank of 80ns DRAM. And then, of course, you
have to write to video memory to update the screen.
Apple could have used two baks but that would have increaced component
costs and taken more motherbaoard real estate, which also cost money.
I am pretty sure the machines that use VRAM and support at least 16 bits
on monitors >=800x600 have the memory organized as at least two banks.
--
____________________________________________________________________________
Erik Speckman espe...@romulus.reed.edu GBDS
Russell Paden
Multiscan HG monitor (CPD-1304S) that's worked faithfully with my IIsi
for nearly two years now. When I turn on the Quadra 630, the top half of
the screen is all wavy (you can't tell what anything is). The trash can
is clear at the bottom, but the the menu bar at top is totally
distorted. When I hook up the monitor to the IIsi it works fine. So,
after reading this thread, I'm still not clear whether I've got to get a
new monitor, or if there's something wrong with my new Quadra.
Any help will be appreciated.
Thanks,
--------------------------------------------------------------------------
| Russ Paden | "Life is like a dog sled team. If you're not the |
| ru...@indirect.com | lead, the scenery never changes." --Anonymous |
--------------------------------------------------------------------------
L.H....@student.lut.ac.uk
>Apple needs to compete with machines like that. Witness the Performa 636,
>which costs about the same and -- for marketing reasons only -- is touted
>by Apple as a 66MHz machine (when all of us who have any grasp of this at
>all understand that it's a 33MHz machine).
And Byte's review of the BlackBird excludes it from that group of people
who actually know anything about computers.
"The heart of the 540c is a Motorola 68LC040 clocked at 66MHz. Because the
rest of the system is clocked at 33MHz, overall system performance should
be around that of a 33MHz system, except when portions of an application's
code reside in the processor's caches." - Tom Thompson, p144, Byte Aug 94.
A '66MHz' 040's caches operate at the same speed as a 33MHz 040, funnily
enough. This mislabelling should be put to rest.
L.
$100 more for any feature is
>not going to cut it. Period.
>>
>> The only thing I can see the IDE (aside from cost reduction) for is
>> providing a springboard for a second-generation Houdini board that
>> performs on par with any PC clone. In other words, a 100% clone clone
>> in your Mac. This clone could run OS/2, Linux, BSD386, MS_DOS, Win NT,
>> etc.
>
>Actually, IDE is entirely irrelevant for this purpose. OS/2, MS-DOS,
>WindowsNT all support SCSI without any problem. I presume Linux and BSD386
>do, too, but I've never used either. Regardless, you don't need IDE for a
>super-Houdini.
>>
>> An interesting prospect, and it will be neat to see what Apple does
>> with all of this. I have a sneaky suspicion that the Q630 is going to
>> get dropped within 6 months.l
>
>I have a sneaky suspicion that when the next generation of PowerMacs comes
>out (and right now MacWeek's Henry Norr seems to be suggesting May, which
>I presume is based on some fairly good inside sources) that the Quadra 630
>will sport a fancy new logic board with a PowerPC chip on it. The form
>factor will likely live for the next half decade or so -- it is the
>replacement for both the LC and 6100 form factors.
>>
>> But this is no reason not to buy one. It's inexpensive and it will
>> sport a host of multimedia features (TV, VCR, audio/video heaven). For
>> the consumer, it could be neat. For the computer user, I have some
>> doubts.
>
>I am glad you see it as acceptable to purchase. The only people who should
>shy away at all are those who need large monitor support and/or want a
>PowerPC now. The 6100 is cheaper than the 630 + the PowerPC upgrade board.
>
>Mark
>ro...@cindy.stanford.edu
>
>Mark Rogowsky
>ro...@forsythe.stanford.edu <-- But not for much longer
>ro...@cindy.stanford.edu <-- The QuickMail/SMTP box
Mark Rogowsky
Cole) wrote:
> You're right that "it's bad because PC-clones use it" makes no sense.
> The way I see it, there are two valid reasons no to like IDE in Macs:
>
> 1) The technology is inferior.
> Yeah, you can make these arguments, but the don't amount to much.
> The transfer/access times are about the same as SCSI, and 95% of
> mac users won't even notice the difference. Let's just say that
> IDE is not as elegant as SCSI and leave it at that. Yes, elegence
> is semi-important to some Mac buyers, but no big deal on this front.
You call your reasons "valid" but then you knock off Reason 1 right hear.
If it's "semi-important" it's not an issue. Anyone who choose not to buy a
Mac because it has IDE is an idiot. Especially 'cause the alternative is a
PC, which will doubtless have IDE. Oh, and if you don't choose to replace
the internal drive ever, the fact you have IDE may never even occur to you
as at all relevant.
>
> 2) It's a support nightmare.
> This is much more important, and something that business purchasers
> will have to think about hard before making buying decisions. Even for
> home buyers, this is going to be a pain sometimes. Consider just the
> kind of questions we get here on c.m.s.h, like "can I put my old HD I
> just replaced in an enclosure and use it as an external drive?" -- the
> answer would have been "yes" but now it's "maybe. you can unless you
> have a q630, p630, p636, or pb150".
Fact: most medium and large size businesses do not even consider the Mac
anymore as a standard desktop machine even though the price/performance for
standard desktops now equals or exceeds that of Intel-based PCs. Fact:
cheaper Macs and better marketing are the only way to change that -- hence
the possibility of IBM being authorized to make Mac clones for the Fortune
1000 market (or whatever it is); IBM markets to big firms better.
Businesses are no more of less likely to buy Macs than before.
Oh, and regarding the "can I reuse my drive" question, I offer two
solutions: 1) Buy the new drive as an external and keep the old one
internal -- this is common 2) sell the old drive and move on. It's not a
relevant selling point either and will add to "confusion" for a tiny subset
of users performing an action one time. Big deal.
>
> IMHO, there was no reason for Apple to use IDE in the 63x line. The price
> difference is almost nil for 3.5" drives.
Your humble opinion is wrong. If they saved even $10, that increases the
margin one point. If it sets the stage for using an IDE CD-ROM and an IDE
HD later, with a savings of $50, it makes a huge difference in margins. The
machines may be profitable instead of unprofitable.
> I do, however, think they were
> fully justified in using IDE drives on the pb150 (2.5" drives have a much
> bigger price delta between SCSI and IDE). Maybe Apple was thinking some-
> thing along the lines of "We're going to FireWire in a year or two anyway,
> so IDE will be good practice supporting multiple HD protocols" -- it's
> possible.
>
Not a factor, either. Supporting IDE has nothing to with supporting
FireWire and "multiprotocol" drive support of IDE/SCSI is vastly difference
from SCSI/FireWire. Apple would not go through with the IDE drives as an
"exercise" to the external market. If that was their goal, it'd be internal
use only.
> brian
>
> PS. I've been trying to get a straight answer for this question for a while,
> so I'll ask it again:
> Is it possible to yank out the IDE HD from a q630 and replace it
> with a SCSI drive? The q630 must have an internal SCSI ribbon
> for the CD-ROM drive, but I'm unsure if:
> 1) replacing the two-connector ribbon with a standard three-
> connector ribbon will work
It should work no problem.
> 2) termination issues can be resolved
So long as you either (a) attach the hard drive between the board and the
CD and leave the drive unterminated or (b) attach the CD between the board
and a terminated drive, while un-terminating the CD, you should be fine.
> 3) the q630's motherboard can handle not having an IDE drive
> connected to it
Don't know, but I bet yes.
> 4) The replaced IDE drive (Apple ROMs?) will work in my father's
> PC-clone.
Don't know, but I bet yes.
> Please let me know if you have answers to any of these questions.
> (It's less likely to be possible, but I'd also like to know it it's
> possible to replace a pb150's IDE HD with a SCSI hd).
This is almost certainly impossible. There would not likely be enough room
inside the PB for a SCSI connector, which would necessarily be different
from the IDE connector.
Mark Rogowsky
ro...@forsythe.stanford.edu <-- Currently reflects to the QuickMail box
Hartmut Kocher US/ESA 60/1L/2? #5629
> In article <rogo-06089...@cindy.stanford.edu> ro...@forsythe.stanford.edu (Mark Rogowsky) writes:
> >Apple needs to compete with machines like that. Witness the Performa 636,
> >which costs about the same and -- for marketing reasons only -- is touted
> >by Apple as a 66MHz machine (when all of us who have any grasp of this at
> >all understand that it's a 33MHz machine).
>
> And Byte's review of the BlackBird excludes it from that group of people
> who actually know anything about computers.
>
> "The heart of the 540c is a Motorola 68LC040 clocked at 66MHz. Because the
> rest of the system is clocked at 33MHz, overall system performance should
> be around that of a 33MHz system, except when portions of an application's
> code reside in the processor's caches." - Tom Thompson, p144, Byte Aug 94.
>
> A '66MHz' 040's caches operate at the same speed as a 33MHz 040, funnily
> enough. This mislabelling should be put to rest.
>
From what I heard, the 68040 doubles clock speed internally. Therefore,
they decided to state the internal clock instead of the bus clock.
This is consistent with the PC world where they also state internal
frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
externally with 33MHZ).
--
+==============================|==============================+
| Hartmut Kocher | |
| Technical Consultant | All opinions expressed here |
| Rational GmbH | are my own. |
| Rosenstrasse 7 | |
| 82049 Pullach im Isartal | I know you guessed it, |
| Germany | but it keeps my lawyer happy.|
| Email: h...@rational.com | |
+==============================|==============================+
A. D. Anbar
Dan Magorian <mago...@ni.umd.edu> writes:
> Don't you all think people at Apple hated the idea of IDE, due to it's
> technical inferiority (ever tried to attach 2 from different mfgrs in a PC?
> Esp Connor, they're incompatible with everyone else's. The whole thing is a
> wretched PC hardware kludge) and the extra work of emulating the old ST506
> command set in PC Roms?
>
> Of course they did, and it's doubtful that the mgr cost difference would have
> forced the change. What did is the negotiations for the PowerPC Reference
> Platform. Basically, IBM demanded that it (PREP) have IDE, and Apple demanded
> that it have ADB (Desktop bus). So it has both.
I'm no expert, but...
I don't see why including IDE in the PREP spec forces Apple to include an
IDE drive in a Mac. As long as the Mac OS/ROMS are designed so they
_can_ handle IDE, so that the MacOS _can_ run on an IDE-equipped PREP
machine (i.e. some future IBM product), everyone should be happy. If
Apple then chooses to market machines which support only a subset of the
complete spec., that should be acceptable to everyone, too.
Your scenario would only make sense if PREP was going to support only
IDE, but not SCSI. Then Apple would be forced to use IDE to remain
PREP-compliant. I presume this is not the case- it would be silly if
PREP did not include support for both standards.
The decision to actually implement IDE must have some other motive (e.g.
cost).
David H Brown
>In article <hades-08089...@fpcgb-kip9.dhmc.dartmouth.edu> ha...@dartmouth.edu (Brian V. Hughes) writes:
>>[...] That is why they can drive a 640x480 monitor in
>>Thousands of colors where the 630 can only do 256 colors.
>
I doubt that Mr. Hughes is sure about this... I just pulled down the Monitors
control panel on my 630 that I picked up Sunday and (using the oldish Apple
13" Trinitron monitor [640x480]) set it to thousands of colors. No problem.
Rendered models from Ray Dream Designer look much nicer :-)
--
bro...@stolaf.edu (Dave Brown) -- "And still the universe extends to a place
that never ends... which is maybe just inside a little jar" (YW).
Dan Magorian
Date: 10 Aug 1994 04:40:37 GMT
In article <329lo5$8...@gap.cco.caltech.edu> A. D. Anbar, anb...@caltech.edu
writes:
Since they were already having for political reasons to go to the expense of
supporting
IDE drives, they might as well take advantage of the small cost savings by
using an
internal IDE drive. Putting a lot of work into software and hardware
support for IDE
then not using it would have made no sense at all from a marketing
perspective.
Of course, users would have liked it better.
Dan
Bao Chau Ha
>> >Apple needs to compete with machines like that. Witness the Performa 636,
>> >which costs about the same and -- for marketing reasons only -- is touted
>> >by Apple as a 66MHz machine (when all of us who have any grasp of this at
>> >all understand that it's a 33MHz machine).
>>
>> And Byte's review of the BlackBird excludes it from that group of people
>> who actually know anything about computers.
>>
>> "The heart of the 540c is a Motorola 68LC040 clocked at 66MHz. Because the
>> rest of the system is clocked at 33MHz, overall system performance should
>> be around that of a 33MHz system, except when portions of an application's
>> code reside in the processor's caches." - Tom Thompson, p144, Byte Aug 94.
>>
>> A '66MHz' 040's caches operate at the same speed as a 33MHz 040, funnily
>> enough. This mislabelling should be put to rest.
>>
>
>From what I heard, the 68040 doubles clock speed internally. Therefore,
>they decided to state the internal clock instead of the bus clock.
>This is consistent with the PC world where they also state internal
>frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
>externally with 33MHZ).
>
at 66 Mhz. There is no 040 equivalent of the 486DX2-66. You may not
know that the 486DX2-66 is almost twice as fast as a 486DX33,
eventhough they are clocked at the same speed. I would love to be
proven wrong that the 68LC040 in the 540C is actually running twice
the speed of the 68040 in a Quadra 950, although both are clocked
at 66 Mhz.
Bao
Brian V. Hughes
bro...@lars.acc.stolaf.edu (David H Brown) wrote:
>In article <CuA6o...@lut.ac.uk> L.H....@student.lut.ac.uk (Lloyd Wood)
writes:
>>In article <hades-08089...@fpcgb-kip9.dhmc.dartmouth.edu>
ha...@dartmouth.edu (Brian V. Hughes) writes:
>>>[...] That is why they can drive a 640x480 monitor in
>>>Thousands of colors where the 630 can only do 256 colors.
>>
>>Are you sure about this?
>
>I doubt that Mr. Hughes is sure about this... I just pulled down the Monitors
>control panel on my 630 that I picked up Sunday and (using the oldish Apple
>13" Trinitron monitor [640x480]) set it to thousands of colors. No problem.
>Rendered models from Ray Dream Designer look much nicer :-)
As has already been pointed out, the 630 does come with 1MB of
dedicated DRAM for video displays, not 512k of system DRAM as I had
stated. I'm not sure where I got the 512k figure from, but this is what
happens when you don't check your data tables to verify your information.
-Hades
mis...@csc.canterbury.ac.nz
> they decided to state the internal clock instead of the bus clock.
> This is consistent with the PC world where they also state internal
> frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
> externally with 33MHZ).
doubled. IT ISNT. The 040 has two clock pins , bclk and pclk, pclk is doubled.
pclk is used for some floating point operations ( perhaps all ), while the
integer unit runs off bclk. I guess this is the reason for the 040's excellent
floating point speed.
feel free to correct me if Im wrong.
Jon.
Erik A. Speckman
Hartmut Kocher US/ESA 60/1L/2? #5629 <koc...@lts.sel.alcatel.de> wrote:
>>
>> "The heart of the 540c is a Motorola 68LC040 clocked at 66MHz. Because the
>> rest of the system is clocked at 33MHz, overall system performance should
>> be around that of a 33MHz system, except when portions of an application's
>> code reside in the processor's caches." - Tom Thompson, p144, Byte Aug 94.
>>
>> A '66MHz' 040's caches operate at the same speed as a 33MHz 040, funnily
>> enough. This mislabelling should be put to rest.
>>
>
>From what I heard, the 68040 doubles clock speed internally. Therefore,
>they decided to state the internal clock instead of the bus clock.
>This is consistent with the PC world where they also state internal
>frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
>externally with 33MHZ).
Let us just say that Lloyd knows more than you do.
The 040 takes a 1x and a 2x clock. The important thing is that the
pipeline is advanced with the 1x clock. A 486 DX or DX/2 takes only a 1x
clock. The DX/2 has internal circutry (a PLL?) which is used to
synthesize the double speed clock.
Erik A. Speckman
Actually, I thought the DX and DX/2 are both clocked at 33 MHz. The DX/2
synthesizes the 2x clock internally.
Other than that, I agree with you. Ther is no 040 equivalent to the
DX/2, or if there is, it isn't used in any Macs.
The LC040 in the 540C is running at the same speed as the 040 in the
Q650, Q800 and Q950.
LEE BRIAN
Huh???? I had that the DX2's have a PLL that doubles the input clock.
i.e., 33MHz clock comes in and is doubled to 66. even if what I heard
is wrong and a 66 clock DOES drive the DX33, it is definitely divided
by 2 on-chip. I also agree that a 68040 should be rated as the internal
clock speed, as I also heard that it doubles the clock input. It is
most likely that all on-chip cache accesses proceed at 66MHz.
later,
bjl
(sorry, my grammar is messed up today...)
--
Brian Jonathan Lee (aka "hojo") | "Beef satay?!?! Not beef satay again!!!!"
b...@ecf.toronto.edu | "XMen! XMen! Rescue Kitty from the caves!"
b...@eecg.toronto.edu | "Evil thy name is NETREK!"
Ludis Langens
>In article <habaoch.94...@lab30.eng.auburn.edu>,
>Bao Chau Ha <hab...@eng.auburn.edu> wrote:
>>That is wrong. Both the 486DX33 and 486DX2-66 are clocked externally
>>at 66 Mhz. There is no 040 equivalent of the 486DX2-66. You may not
>>know that the 486DX2-66 is almost twice as fast as a 486DX33,
>>eventhough they are clocked at the same speed. I would love to be
>>proven wrong that the 68LC040 in the 540C is actually running twice
>>the speed of the 68040 in a Quadra 950, although both are clocked
>>at 66 Mhz.
>
>Huh???? I had that the DX2's have a PLL that doubles the input clock.
>i.e., 33MHz clock comes in and is doubled to 66. even if what I heard
>is wrong and a 66 clock DOES drive the DX33, it is definitely divided
>by 2 on-chip. I also agree that a 68040 should be rated as the internal
>clock speed, as I also heard that it doubles the clock input. It is
>most likely that all on-chip cache accesses proceed at 66MHz.
Sigh, this was debated to a useful conclusion several months ago. The
68040 (and 68LC040) (and all previous 68040's) are clocked with two
clocks: a bus clock and a second clock that is exactly double the first.
Thus, all '040 machines from Apple to date could be given one of the
following speeds: 20/40 (Centris 610), 25/50 (Quadra 700), 33/66 (lots
of machines), or 40/80 (Quadra 840AV).
Some parts of the CPU run from the slower clock. Other parts run from
the higher speed clock. However, integer instructions boundaries are
tied inexorably to the (slower) bus clock. This is true of ALL '040's
shipped to date. This means both the 950 and 540C have 33MHz CPU's!
Ludis Langens
lu...@netcom.com
Jim Clark
that there are several possibilities out there, but am
not aware of the companies providing such accelerators
and their cost, whether user group/educational discounts
are available etc.
Any help would be appreciated...
jim clark
ut martin
math/cs
John Roldan
: In article <1994Aug10....@lts.sel.alcatel.de>,
: koc...@lts.sel.alcatel.de (Hartmut Kocher US/ESA 60/1L/2? #5629) wrote:
: > > A '66MHz' 040's caches operate at the same speed as a 33MHz 040, funnily
: > > enough. This mislabelling should be put to rest.
. . .
: > frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
: This group has decided, correctly, that the marketing claim is patently
: false. Read Eric Speckman's SPEC posting for proof that not only isn't the
: 040/33 a clock-doubled chip, the spurious claim that said chip matches the
: performance of the 486/66 is also patently false.
: Mark Rogowsky
I didn't notice SPEC figures in the faq . . care to direct me to this
posting or any other document showing 486 and 68030,68040 SPEC figures?
I always wondered where they sit on the SPECtrum. Yes, I know actual
mileage may vary . . . Is there a SPEC newsgroup ?
Mark Rogowsky
<mago...@ni.umd.edu> wrote:
This post is so completely devoid of fact and loaded with misinformation.
[I open with that harsh assessment so that followers of this group know to
take me seriously here.]
First of all, Apple has been working on IDE drive support in the Macintosh
for some time, perhaps as much as 3-4 years or longer, certainly 18-24
months.
The reason they have been doing this is plain and simple: cost, cost and
cost. When Apple began doing the engineering, the first PowerPC Macintosh
was a vision, not a product, and prep (no caps back then) was short for
prepare [as in "Prep him for surgery!"]
When it became apparent that the EIDE spec would allow for CD-ROMs to be
placed on an IDE chain, the decision was sealed. The cost differential --
at the all-important manufacturing level -- between building a Mac with an
IDE drive and IDE CD-ROM vs. building one with a SCSI HD and a SCSI CD-ROM
is potentially $50 per machine. Go back in the thread and read my argument
on gross margins to understand how tremendous $50 is at that stage. True,
the 630 does not use an IDE CD-ROM at this time, but beginning to build
IDE machines that are even $10 cheaper is the first step toward more
general use of IDE.
Second of all, the presence of a SCSI port and in the future a FireWire
port means that all the configuration headaches associated with IDE are
not going to be relevant on the Macintosh. You can simply move ahead with
plug and play SCSI devices as you always have.
Third of all, and here's where the post quoted above demonstrates its
utter irrelevance: the Quadra 630 is based on the 68040 chip, a CISC chip
designed by Motorola and used in millions of Macintoshes. The PReP
specification (PowerPC Reference Platform) defines hardware built around
the PowerPC chip, designed primarily by IBM in conjunction with Motorola
(and Apple). That chip is used in hundreds of thousands of Macs at this
time. Given that the Quadra 630 is not a PowerPC-based computer, it cannot
be a PReP Mac in any manner!
The PReP specification, while addressing IDE, absolutely does not require
IDE. Consider the case of DOS/Windows. 99.9% of all Windows users boot
from an IDE hard drive, but they need not. They could pull the IDE drive
and boot from a SCSI drive (I hope I'm not confused here, I know in OS/2
you can do this). Just because PReP contains IDE doesn't mean a
PReP-compliant machine would need to have an IDE drive. It merely means it
would have to support IDE drives -- which Apple could do through plug-in
cards.
The hullabaloo about Apple not adopting PReP because of its use of IDE, a
parallel port, etc. is ludicrous. Apple would not need to add any hardware
to its machines to fulfill some arbitrary hardware requirement. The issues
involve things including, but not limited to, (a) Apple's unwilligness to
cede control of its hardware design to some IBM-led consortium which
might, in the future, go in directions Apple does not like (b) Apple's
unwillingness to make significant changes to its OS to allow said OS to
bootstrap a PReP machine to life [those opening seconds of a PReP machine
are very different than they are on a Macintosh] (c) IBM's unwillingess to
make sufficient changes in the spec [Note: 0 PReP machines have shipped
while 300,000 PowerMacs have -- who should be compromising?] to mollify
Apple and meet Apple's concerns.
In short, the Quadra 630 has about as much to do with all of this as
Forrest Gump does with a history text book. Both may be good things and if
you believe the fictions about one, it could relate to the other.
End of line.
Mark Rogowsky
koc...@lts.sel.alcatel.de (Hartmut Kocher US/ESA 60/1L/2? #5629) wrote:
> > enough. This mislabelling should be put to rest.
> >
>
> From what I heard, the 68040 doubles clock speed internally. Therefore,
> they decided to state the internal clock instead of the bus clock.
> This is consistent with the PC world where they also state internal
> frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
> externally with 33MHZ).
As Bill Clinton would say:
"No. No. No. No. No. No. No. No. No. No."
This group has decided, correctly, that the marketing claim is patently
false. Read Eric Speckman's SPEC posting for proof that not only isn't the
040/33 a clock-doubled chip, the spurious claim that said chip matches the
performance of the 486/66 is also patently false.
Mark Rogowsky
L.H....@student.lut.ac.uk
>In article <CuA72...@lut.ac.uk>, L.H....@student.lut.ac.uk writes:
>> In article <rogo-06089...@cindy.stanford.edu> ro...@forsythe.stanford.edu (Mark Rogowsky) writes:
>> >Apple needs to compete with machines like that. Witness the Performa 636,
>> >which costs about the same and -- for marketing reasons only -- is touted
>> >by Apple as a 66MHz machine (when all of us who have any grasp of this at
>> >all understand that it's a 33MHz machine).
>>
>> And Byte's review of the BlackBird excludes it from that group of people
>> who actually know anything about computers.
>>
>> "The heart of the 540c is a Motorola 68LC040 clocked at 66MHz. Because the
>> rest of the system is clocked at 33MHz, overall system performance should
>> be around that of a 33MHz system, except when portions of an application's
>> code reside in the processor's caches." - Tom Thompson, p144, Byte Aug 94.
>>
>> A '66MHz' 040's caches operate at the same speed as a 33MHz 040, funnily
>> enough. This mislabelling should be put to rest.
>>
>
>From what I heard, the 68040 doubles clock speed internally. Therefore,
>they decided to state the internal clock instead of the bus clock.
>This is consistent with the PC world where they also state internal
>frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
>externally with 33MHZ).
The 68040 does *NOT* double clock speed internally.
Next.
L.
A. D. Anbar
Dan Magorian <mago...@ni.umd.edu> writes:
> >The decision to actually implement IDE must have some other motive (e.g.
> >cost).
>
> Since they were already having for political reasons to go to the expense of
> supporting
> IDE drives, they might as well take advantage of the small cost savings by
> using an
> internal IDE drive. Putting a lot of work into software and hardware
> support for IDE
> then not using it would have made no sense at all from a marketing
> perspective.
As rogo nicely pointed out, the R&D to implement IDE on the Mac was
underway long before PREP was a factor. And, even if you _were_ right
about politics being the original motivator, Apple is hardly unfamiliar
with spending money on technologies which it chooses not to market for
profit (e.g. various Newton variants; the "Star Trek" emulator; Houdini;
I'm sure there are others).
> Of course, users would have liked it better.
>
This contradicts what you just said- from a marketing perspective, it
makes no sense to piss-off your users. If this really made a difference
to most users, then Apple would just blow-off IDE. The fact is that IDE
delivers all the needed functionality to the vast majority of Mac users,
will cause only minor headaches for a small minority of tech-support
folks, and saves Apple $$$ downstream. (The only functional loss due to
IDE drives seems to be the PB 150's inability to function in SCSI disk
mode, which I assume is due to the fact that it does not have a SCSI disk
inside. This is a real loss to a minority of PB users). Back-room
politics really doesn't enter into it.
Jude Giampaolo
Allan Levy <allan...@his.com> wrote:
>
>No, the 630 does not have a parallel port, and Apple is not cooperating with
>the IBM folks on the PREP standard last I heard - good try though.
I wish they would......
--
Jude Giampaolo Case Western Reserve University
jc...@po.cwru.edu Cleveland, Ohio
giam...@snowhite.eeap.cwru.edu Electrical Engineering
Christian Stieber
> >From what I heard, the 68040 doubles clock speed internally. Therefore,
> >they decided to state the internal clock instead of the bus clock.
> >This is consistent with the PC world where they also state internal
> >frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
> >externally with 33MHZ).
> >
> But at least in the PC world they have the grace to call the chip a DX2
> to give a clue that it only clocks the bus at half the speed. Similarly
> for the new DX4 chips.
... but the DX4 should be called DX3. Seems they decided that DX4 looks
better.
Christian
--
// Christian Stieber sti...@informatik.tu-muenchen.de
\// Certified Amiga Developer PGP public key available
--------------------------------------------------------------------------
"Life starts at '030, fun starts at '040, impotence starts at '86"
keyboard not connected -- press F1 to continue
Larry Hamilton
Store (Dallas) does not have their's yet. If so, what flavor does the CD
come in, IDE or SCSI? I priced CDs this weekend (for a PC) and the IDE
drives were $150 while a SCSI drive was closer to $300.
On a PC if you try to put two hard drives from different manufacturers on
the same IDE controller you are going to have a least a few problems
getting it to work. Would this also be the case on the Mac with a hard
drive and a CD ROM from different vendors?
Larry Hamilton
l...@unicomp.net
David Hough
>
>they decided to state the internal clock instead of the bus clock.
>This is consistent with the PC world where they also state internal
>frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
>externally with 33MHZ).
>
to give a clue that it only clocks the bus at half the speed. Similarly
for the new DX4 chips.
Dave
--
*****************************************************************************
* G4WRW @ GB7WRW.#41.GBR.EU AX25 * *
* da...@llondel.demon.co.uk Internet * Stop the World! I want to get off! *
* g4...@g4wrw.ampr.org Amprnet * *
*****************************************************************************
Allan Levy
Dan Magorian writes in a message on 10 Aug 94 to All in COMP.SYS.MAC.HARDWARE:
DM> Don't you all think people at Apple hated the idea of IDE,
DM> due to it's technical inferiority (ever tried to attach 2 from
DM> different mfgrs in a PC? Esp Connor, they're incompatible with
DM> everyone else's. The whole thing is a wretched PC hardware
DM> kludge) and the extra work of emulating the old ST506 command
DM> set in PC Roms?
DM> Of course they did, and it's doubtful that the mgr cost difference
DM> would have forced the change. What did is the negotiations
DM> for the PowerPC Reference Platform. Basically, IBM demanded
DM> that it (PREP) have IDE, and Apple demanded that it have ADB
DM> (Desktop bus). So it has both.
DM> Not sure whether the 630 is the first PREP mac, or just on
DM> the way there.
No, the 630 does not have a parallel port, and Apple is not cooperating with
the IBM folks on the PREP standard last I heard - good try though.
Sincerely,
Allan E. Levy
L.H....@student.lut.ac.uk
>i.e., 33MHz clock comes in and is doubled to 66. even if what I heard
>is wrong and a 66 clock DOES drive the DX33, it is definitely divided
>by 2 on-chip. I also agree that a 68040 should be rated as the internal
>clock speed, as I also heard that it doubles the clock input.
Please state the 'source' where you heard this fallacy, and then I can
quote the relevant parts of the Motorola datasheet at you.
>It is
>most likely that all on-chip cache accesses proceed at 66MHz.
Nope.
L.H....@student.lut.ac.uk
Yup.
> I would love to be
>proven wrong that the 68LC040 in the 540C is actually running twice
>the speed of the 68040 in a Quadra 950, although both are clocked
>at 66 Mhz.
The 68LC040 is the 68040 without an FPU. Whether it's described as
33MHz (roughly the same as a 33MHz 486SX) or 66MHz (optimistic
marketing hype), it will be slower than the 33MHz 68040, because it
loacks that FPU.
It's not the frequency you get that counts; it's how you use it that
matters.
L.
L.H....@student.lut.ac.uk
>In article <1994Aug10....@lts.sel.alcatel.de> koc...@lts.sel.alcatel.de (Hartmut Kocher US/ESA 60/1L/2? #5629) writes:
>>
>>From what I heard, the 68040 doubles clock speed internally. Therefore,
>>they decided to state the internal clock instead of the bus clock.
>>This is consistent with the PC world where they also state internal
>>frequencies instead of the bus clock (e.g., a 66MHz 486 is clocked
>>externally with 33MHZ).
>>
>But at least in the PC world they have the grace to call the chip a DX2
>to give a clue that it only clocks the bus at half the speed. Similarly
>for the new DX4 chips.
>
1/2.5. That's marketing for you.
Calling a 68040 a 'DX2' is false, as it does not clock-double the external
frequency, and integer operations take place at the bus speed, not double
the bus speed.
David Smith
>In article <ronross-0708...@ronross.dialup.access.net>,
>ron...@panix.com (Ron Ross) writes:
>
>Actually, there is one other form factor that lasted at least five years:
>
>The IIcx/IIci/Quadra 700 form factor was at least 5 years - the IIci
>lasted nearly 4 years itself if I'm not mistaken.
>
>Ted Howe
>ted...@aol.com
The Q700 didn't use the same box as the IIci/cx, the Q700's is wider adn
designed to be set sideways.
For number of machines that have had the same box it would likely be the
LC box. How many machines have had that box?
--
---- David A. Smith | The opinions above are mine and do not reflect
- smi...@cs.csee.usf.edu | those of the University of South Florida.
- das...@suntan.eng.usf.edu |
| PGP/RIPEM key avalible via finger.
edwa...@acf6.acf.nyu.edu
> In article <32461p$r...@search01.news.aol.com> ted...@aol.com (TedHowe) writes:
>>In article <ronross-0708...@ronross.dialup.access.net>,
>>ron...@panix.com (Ron Ross) writes:
>>
>>Actually, there is one other form factor that lasted at least five years:
>>
>>The IIcx/IIci/Quadra 700 form factor was at least 5 years - the IIci
>>lasted nearly 4 years itself if I'm not mistaken.
>>
>>Ted Howe
>>ted...@aol.com
>
> The Q700 didn't use the same box as the IIci/cx, the Q700's is wider adn
> designed to be set sideways.
>
> For number of machines that have had the same box it would likely be the
> LC box. How many machines have had that box?
>
The ci and the 700 have the same form factor in the sense that Apple sold a ci
to Q700 upgrade kit, and that they shared some of the same parts.
Both the ci and th Q700 can be placed vertical or horizontal, though most
opt for the ci to be placed horizontal, and the Q700 to be placed vertical.
There was an LC, LCII, LCIII, LC475, and like a butt-load of Performas that
used an LC box:
400,405,410,430,450,460,466,467,475,476
Hope this helps (?)
-Kevin
LEE BRIAN
>In article <CuBzG...@ecf.toronto.edu> b...@ecf.toronto.edu (LEE BRIAN) writes:
>>Huh???? I had that the DX2's have a PLL that doubles the input clock.
>>i.e., 33MHz clock comes in and is doubled to 66. even if what I heard
>>is wrong and a 66 clock DOES drive the DX33, it is definitely divided
>>by 2 on-chip. I also agree that a 68040 should be rated as the internal
>>clock speed, as I also heard that it doubles the clock input.
>
>Please state the 'source' where you heard this fallacy, and then I can
>quote the relevant parts of the Motorola datasheet at you.
well, it was most likely the comp.sys.* groups. if the data sheet
says otherwise, I heard wrong then.
later,
bjl
BTW, are ANY 68k family members internally clock doubled?
David Edwards
: That is wrong. Both the 486DX33 and 486DX2-66 are clocked externally
: at 66 Mhz.
Huh? A '486DX33 clocked externally at 50MHz will cook itself in it's own
juices. (I have determined this through experimentation. I would advise NOT
trying this at home, especially with a loved one's computer...)
(The below info is derived from a very early model 486DX50 motherboard.
How early? It has one of the "buggy" first 50MHz 80486 chips... Obviously,
things may have changed for later-model motherboards.)
A '486DX50 (NOT a DX2 50...) uses an external 50MHz crystal. Logically,
then, it can be assumed that it is clocked at 50MHz (The one I'm currently
playing with is... I just poked in there with a frequency counter while
the chip was out of it's socket...). A DX33 is the same chip in a
different speed grade; therefore, it is reasonable to assume that the chip
is clocked externally at 33MHz. (The 66MHz clock modules frequently seen
on DX33 motherboards frequently are fed into a D-type flip-flop that
cleans up the waveform wonderfully. This flip-flop is frequently hidden
somewhere inside a vast sea of gates on one of the support chips.)
: There is no 040 equivalent of the 486DX2-66. You may not
: know that the 486DX2-66 is almost twice as fast as a 486DX33,
Faster, yes. Significantly faster (50% or so...), yes. Twice as fast?
No. They average about 10% faster than a DX50 (About on par with a DX55?)
in my experience, not twice as fast as a 486DX33.
--
____________________________________________________________________
David Edwards || I will not do it, in my car, I will not speak ||
U of MO, Rolla || for UMR. I would not, could not in a plane, I ||
d...@umr.edu || would not, could not on a train. I will not ||
|| speak for UMR, I will not do it, Sam you are. ||
====================================================
DRAEGER M B
: Bear in mind that an IDE "controller" is usually a 1-of-8 decoder and a
: 16-bit bidirectional bus transciever - about 30c worth of parts, and
: easily implemented in some spare corner of a gate array.
: I won't soapbox about IDE, but for this application, surprisingly, it's
: actually a reasonable design decision.
I think this was a rather smart move to introduce this standard into the
Mac world. To the end-user it does not really matter where his data
comes from - as soon as there are at least some formatters out there -
and besides the fact that IDE drives tend to be a little bit cheaper (as
pointed out, this adds up) some poster pointed out that especially
notebook HD's (smallest form factor) tend to get out with IDE first -
the market is bigger - and then with SCSI. Could make a time difference,
too. As I said, as soon as some utils are out there, I will stop to be
concerned,as IDE-Mac need not impose the same limits as IDE-DOS (only
two HD's in a computer, HD's in most general sense): There still is
SCSI, and there maybe another spare corner on a gate array, too. I'm
waiting for some IDE-Mac CD's to show up ;-)!
Matthias
Ludis Langens
>BTW, are ANY 68k family members internally clock doubled?
Most members of the 683xx family of microcontrollers have a PLL that creates
a 16MHz (or 25MHz) clock from a 32768Hz (i.e. 32.768KHz) reference. The
exact frequency generated is software programable. The reason for using
a 2^15Hz crystal is that is very cheap (every digital watch has one) and
it does not consume very much power.
Ludis Langens
lu...@netcom.com
edwardsk_acf...@nisc.fidonet.org
bjl_ecf_t...@nisc.fidonet.org
>In article <CuBzG...@ecf.toronto.edu> b...@ecf.toronto.edu (LEE BRIAN)
writes:
>>Huh???? I had that the DX2's have a PLL that doubles the input clock.
>>i.e., 33MHz clock comes in and is doubled to 66. even if what I heard
>>is wrong and a 66 clock DOES drive the DX33, it is definitely divided
>>by 2 on-chip. I also agree that a 68040 should be rated as the internal
>>clock speed, as I also heard that it doubles the clock input.
>
>Please state the 'source' where you heard this fallacy, and then I can
>quote the relevant parts of the Motorola datasheet at you.
well, it was most likely the comp.sys.* groups. if the data sheet
says otherwise, I heard wrong then.
later,
bjl
BTW, are ANY 68k family members internally clock doubled?
--
dpe_saucer...@nisc.fidonet.org
phiss27_alinga_...@nisc.fidonet.org
David Lerner
>Is the Quadra 630 available with a CD ROM yet? My local Computer City
>Store (Dallas) does not have their's yet. If so, what flavor does the CD
>come in, IDE or SCSI? I priced CDs this weekend (for a PC) and the IDE
>drives were $150 while a SCSI drive was closer to $300.
The Quadra 630 is only shipping in one configuration - 4/250, without CD.
You can add an Apple CD300Plus internal drive for about $319.
Several of the 63X Performa and LC models are shipping with a CD-ROM, but
those models have the 68LC040 processor, missing the FPU that you might
possibly want one day.
Because of the design of the 630s and the way that drives plug directly
in with special adapters (instead of using cables), I can only assure you
that Apple CD300Plus SCSI CD-ROMs will fit. Other brands might fit, but
you will need to get an Apple bracket and connector adapters, and if the
connectors on your drive are not aligned identically to the ones on
Apples drive (and I don't think that there is any standard for this) the
non Apple drive will not fit.
You cannot use an IDE CD-ROM in any current Mac model.
--
David Lerner/Tekserve Corporation/Macintosh service and repair/212 929-3645
163 West 23rd St, New York, NY 10011 USA/fax 212 463-9280/dle...@panix.com
Larry Hamilton
me that IDE is a little cheaper than a SCSI drive with little or no
performance difference so what's the big deal? It would also seem to me
that Q630 users should be able to put in and IDE CD ROM into the Q630
(note that Apple is not currently doing this) for about half the price
of what a SCSI CD ROM costs (you can have up to two devices on and IDE bus).
Now a real complaint on the Q630 is the monitor size limitation. Apple
almost would have to do something like limit the size to 15" on purpose.
Maybe it would hurt the PPC sales or maybe they are content with their
market share who knows - Apple will never tell.
What will really help the Mac (Macintosh users not necessary Apple) is
clones. If Apple really does license it's priority hardware we should be
able to purchase any configuration of a Mac we want similar to the way you
buy PCs today. If you don't want IDE don't buy it, I priced a PC last
week with SCSI drives instead of IDE. Even had my choice of buses PCI or
ESIA.
Comments?
L.H....@student.lut.ac.uk
>In article <CuE67...@lut.ac.uk>, <L.H....@student.lut.ac.uk> wrote:
>>In article <CuBzG...@ecf.toronto.edu> b...@ecf.toronto.edu (LEE BRIAN)
>writes:
>>>Huh???? I had that the DX2's have a PLL that doubles the input clock.
>>>i.e., 33MHz clock comes in and is doubled to 66. even if what I heard
>>>is wrong and a 66 clock DOES drive the DX33, it is definitely divided
>>>by 2 on-chip. I also agree that a 68040 should be rated as the internal
>>>clock speed, as I also heard that it doubles the clock input.
>>
>>Please state the 'source' where you heard this fallacy, and then I can
>>quote the relevant parts of the Motorola datasheet at you.
>
>well, it was most likely the comp.sys.* groups. if the data sheet
>says otherwise, I heard wrong then.
>
>later,
>bjl
>
>BTW, are ANY 68k family members internally clock doubled?
>
Alan Coopersmith
|I didn't notice SPEC figures in the faq . . care to direct me to this
|posting or any other document showing 486 and 68030,68040 SPEC figures?
There's a list of SPECmarks for a large variety of systems available
via anonymous ftp from ftp.cdf.toronto.edu in the file /pub/spectable.
|I always wondered where they sit on the SPECtrum. Yes, I know actual
|mileage may vary . . . Is there a SPEC newsgroup ?
Not SPEC-specific, but comp.benchmarks covers the topic.
--
------------------------------------------------------------------------
Alan Coopersmith Internet: al...@OCF.Berkeley.EDU
Open Computing Facility or: al...@CSUA.Berkeley.EDU
University of California, Berkeley Bitnet: alanc@ucbocf
Brian V. Hughes
Hamilton) wrote:
>A lot has been discussed here about the IDE drive in the Q630. Seems to
>me that IDE is a little cheaper than a SCSI drive with little or no
>performance difference so what's the big deal?
Actually, the money that Apple saves is on the logic board because
internal IDE is about $15 cheaper than internal SCSI. I don't think there
is any actual savings on the hard drive itself. In general 3.5" IDE and
SCSI drives are basically the same price. They do, however, save a bundle
on the 2.5" IDE drives that are used for the Powerbook 150. Expect to see
IDE drives in all future Powerbooks, because of this.
>It would also seem to me that Q630 users should be able to put in and IDE
CD >ROM into the Q630 (note that Apple is not currently doing this) for
about half >the price of what a SCSI CD ROM costs (you can have up to two
devices on and >IDE bus).
I don't think this is currently possible, but it will be very soon.
>Now a real complaint on the Q630 is the monitor size limitation. Apple
>almost would have to do something like limit the size to 15" on purpose.
>Maybe it would hurt the PPC sales or maybe they are content with their
>market share who knows - Apple will never tell.
Just a quick technical point here, there is no monitor size
limitation. Any size monitor that can display 640x480, 800x600, or 832x624
will work with the internal video of the 630 series. It would be nice if
larger resolutions were supported, but the 630 series is an entry level,
budget priced system.
-Hades
John Gall
: >I also agree that a 68040 should be rated as the internal
: >clock speed, as I also heard that it doubles the clock input.
: Please state the 'source' where you heard this fallacy, and then I can
: quote the relevant parts of the Motorola datasheet at you.
I also have read that the 040's internal speed is double the bus speed.
BUT, I have a source for my info. I kept calling Apple's Fax Info and got
datasheets on just about everything they offered. One of those is about
the 68040 Microprocessor and it DOES state that the internal speed IS
double the bus speed.
John Gall
sia...@ramp.com
Thomas Deweese
>In article <fa0_940...@nisc.fidonet.org> bjl_ecf_t...@nisc.fidonet.
>>BTW, are ANY 68k family members internally clock doubled?
>In the Intel sense (instructions cycle internally at twice bus speed), no.
Actually it was my understanding that all the more recent
680x0 chips could be 'internally clock doubled' they just weren't
forced to be clock doubled. So where as a 486DX2/66 can not take
advantage of a 66Mhz mother board. A 40Mhz 68040 could run on a 40Mhz
mother board or on a 20Mhz mother board... It is simply a matter of
grounding some pins or something.
Now this is all a little fuzzy since Apple hasn't take advantage
of this ability for a while (and they got blasted for it last time
they did, which is why I think this whole discussion is so funny).
--
Thomas DeWeese
dew...@image.kodak.com
"The only difference between theory and practice is
that in theory there isn't any." -- unknown
Mark Rogowsky
> In article <32ofl9$o...@kodak.rdcs.Kodak.COM> dew...@image.kodak.com (Thomas Deweese) writes:
> >In article <CuIzp...@lut.ac.uk>, <L.H....@student.lut.ac.uk> wrote:
> >>In article <fa0_940...@nisc.fidonet.org> bjl_ecf_t...@nisc.fidonet.
> >>>BTW, are ANY 68k family members internally clock doubled?
> >
> >>In the Intel sense (instructions cycle internally at twice bus speed), no.
> >
> > Actually it was my understanding that all the more recent
> >680x0 chips could be 'internally clock doubled' they just weren't
> >forced to be clock doubled. So where as a 486DX2/66 can not take
> >advantage of a 66Mhz mother board. A 40Mhz 68040 could run on a 40Mhz
> >mother board or on a 20Mhz mother board... It is simply a matter of
> >grounding some pins or something.
>
> Where do these idiots come from?
I believe it is the Land of Dairy Queen, located in the middle part of the
United States. We apologize to the rest of the civilized world...
Mark Rogowsky
ro...@forsythe.stanford.edu <-- Currently reflects to the QuickMail box
ro...@cindy.stanford.edu <-- The QuickMail/SMTP box
Evan Torrie
>In article <32k9f7$j...@glock.ramp.com> sia...@glock.ramp.com (John Gall) writes:
>>I also have read that the 040's internal speed is double the bus speed.
>Gee, we have the choice here of:
>1. The Motorola 68040 datasheets. Written by engineers for engineers and
>2. Apple's Fax Info. Written by PR and marketing people who are barely
>Bear in mind that this entire thread has been railing against Apple for its
>description of the 040 capabilities. That a number of Apple engineers read
>these groups, and no Apple engineer has posted on this topic. You think
>they'd be qualified to, wouldn't you?
Here's a post from Apple engineer Dale Adams, which appeared about three
months ago when this came up the first time with the stupid Apple marketeer
who pushed the change to "dual-clocked" nomenclature:
From comp.sys.mac.hardware
From: dale_...@quickmail.apple.com (Dale Adams)
Date: Fri, 20 May 1994 13:32:51 GMT
Newsgroups: comp.sys.mac.hardware
Subject: Re: 66/33Mhz PowerBooks what marketing hype is this?
In article <Cq08t...@ryn.mro.dec.com>, ba...@rock.enet.dec.com wrote:
>
>
> There seems to be an awful lot of confusion about these new powerbooks and
> their clock doubling. Nobody seems to know what they are talking about.
> These 68LC040 processors that are being used are not the same chips as in the
> Quadras. Motorola took the design and shrunk it, and added power management,
> so that they would only use 3-4 Watts or so. I'm not sure whether an FPU
> is on chip, but hasn't been used, or what. (That is basically what Motorola
> did in the desktop 68LC040. They turned off the existing FPU and sold it for
> less money. It's actually the same as the regular 68040 w/FPU.)
> As far as clocking goes, the processors are exactly the same. For instance,
> the old Quadra 800 at 33Mhz, actually runs internally at 66Mhz. This is a
> fact. I know, because I worked at Apple as a student sales representative.
> It is just not well documented.
>
> I do agree that all this is just a marketing thing. But, what you can
> conclude is that a 33/66Mhz 68LC040 in the PowerBook is just as fast as the
> 68040 in the Quadra 800, except when floating point operations are involved.
> You can't compare the 486DX and DX2, since Motorola has no equivalent. The
> 68040 has always been "clock-doubled".
>
> I hope this clears up many of the mis-conceptions.
No, unfortunately you just created a slew of new ones.
The facts:
1) The 68LC040 is not a standard 68040 with a disabled FPU. The 68LC040 is
a modification of the 68040 without the FPU logic. This results in a
smaller, and therefore cheaper and lower-power chip.
2) The 68LC040 does not have power management hardware. The current chips
are lower power than the original 68LC040s simply because they are now
produced in a smaller geometry process.
3) A 33 MHz '040 (LC or otherwise) does not run internally at 66 MHz, at
least not in the sense that the internal pipelines are clocked at that
speed. It is most certainly not "clock-doubled" in the same sense that an
x86DX2 processor is. The '040s do receive a clock that's 2X the processor
(& bus) clock speed. However, this is used to generate a set of internal
quadrature clocks that are used to gain some timing advantages in various
parts of chip. Instructions are not executed at a 66 MHz rate.
In the future, please donUt post this type of mis-information. All it does
is perpetuate old myths that many of us have tried very hard to squelch.
- Dale Adams
--
------------------------------------------------------------------------------
<A HREF="http://liber.stanford.edu/~torrie/>Evan Torrie</A>.
Stanford University, Class of 199? tor...@cs.stanford.edu (finger for PGP)
"Freedom is the freedom to say that two plus two make four. If that is
granted, all else follows." _1984_, George Orwell.
John Vickers
>>I also have read that the 040's internal speed is double the bus speed.
>>BUT, I have a source for my info. I kept calling Apple's Fax Info and got
>>datasheets on just about everything they offered. One of those is about
>>the 68040 Microprocessor and it DOES state that the internal speed IS
>>double the bus speed.
Ho-Hum.
I'd just like to point out, for the benefit of any 286/386/386SX
owners/sellers out there who are feeling a tad inadequate in
the clock speed department that all [*] these chips run off
a double-frequency clock input. For example, an Intel 386DX33
requires a 66.7MHz clock; so we can all now call it a "386-66"
or even a "386-67" :-).
*: AMD 386's can be run off a single speed clock, which is rather
disappointing for these puposes :-(.
--
John Vickers
Owner of well-thumbed 68040UM/AD.
Cambridge, UK jo...@rhizik.demon.co.uk
Chacon, Octavious A.
>
>>>I also have read that the 040's internal speed is double the bus speed.
>>>BUT, I have a source for my info. I kept calling Apple's Fax Info and got
>>>datasheets on just about everything they offered. One of those is about
>>>the 68040 Microprocessor and it DOES state that the internal speed IS
>>>double the bus speed.
O.k. let me get it straight, i buy a clock chipping kit for my 660av, it
comes with a 16.5mhz qrystal, the bus now runs at double the speed (33mhz) and
the cpu runs at (66mhz), this is what I understand from all the discussion.
Logic tells me differently though, I would imagine that the bus operates at
16.5 mhz and the cpu runs at 33mhz i.e. clock doubled.:-)
-OctAVious
L.H....@student.lut.ac.uk
>L.H....@student.lut.ac.uk writes:
>
>>In article <32k9f7$j...@glock.ramp.com> sia...@glock.ramp.com (John Gall) writes:
>>>I also have read that the 040's internal speed is double the bus speed.
>
>>Gee, we have the choice here of:
>>1. The Motorola 68040 datasheets. Written by engineers for engineers and
>>2. Apple's Fax Info. Written by PR and marketing people who are barely
>
>>Bear in mind that this entire thread has been railing against Apple for its
>>description of the 040 capabilities. That a number of Apple engineers read
>>these groups, and no Apple engineer has posted on this topic. You think
>>they'd be qualified to, wouldn't you?
>
> Here's a post from Apple engineer Dale Adams, which appeared about three
>months ago when this came up the first time with the stupid Apple marketeer
>who pushed the change to "dual-clocked" nomenclature:
[post deleted]
Well, I'm glad that I was wrong (I missed Dale's post the first time
around), that there's an Apple engineer willing to give technical
information that Apple marketing would rather not have given out, and that
I can now simply email this post at anyone who says that an 040 is
'clock-doubled' in the Intel sense.
Dan Ortiz
6100/60s? I know my Q630 seems like a rocketship when it comes to scanning and
saving.
Dale Adams
(Thomas Deweese) wrote:
> In article <CuIzp...@lut.ac.uk>, <L.H....@student.lut.ac.uk> wrote:
> >In article <fa0_940...@nisc.fidonet.org> bjl_ecf_t...@nisc.fidonet.
> >>BTW, are ANY 68k family members internally clock doubled?
>
> >In the Intel sense (instructions cycle internally at twice bus speed), no.
>
> Actually it was my understanding that all the more recent
> 680x0 chips could be 'internally clock doubled' they just weren't
> forced to be clock doubled. So where as a 486DX2/66 can not take
> advantage of a 66Mhz mother board. A 40Mhz 68040 could run on a 40Mhz
> mother board or on a 20Mhz mother board... It is simply a matter of
> grounding some pins or something.
> Now this is all a little fuzzy since Apple hasn't take advantage
> of this ability for a while (and they got blasted for it last time
> they did, which is why I think this whole discussion is so funny).
No, no, no, no, no! Where does this stuff come from? (Granted, Apple's
latest marketing is more than a bit to blame.)
The '040 is _not_ clock doubled in the same sense as the DX2 Intel
processors. Yes, it does receive a clock that's 2X the speed of the bus
clock, but the internal instruction pipeline is not clocked at that speed
(which is the typical meaning of a clock doubled processor). The 2X clock
is used to produce a set of internal quadrature clocks to improve timing
relationships between internal stages, but is not used to clock the
instruction pipe at a 2X rate.
Also, there is not a pin on the processor that tells it to run at its bus
at half speed. (Thinking of the '060, perhaps?) Any system that
implements that sort of thing does so with logic external to the processor.
No '040 systems from Apple do this (although the old, now defunct IIvx
did, but that machine had an '030 processor).
Sorry for the late post, but my newserver is about a week behind.
- Dale Adams
Zoltan Kocsi
inputs. BCLK is the bus clock, PCLK is the system logic clock.
The speed of the part is the speed of the bus clock, that is a 25MHz
chip requires a 25MHz BCLK and a 50MHz PCLK on its inputs.
The internal PLL does NOT double any clock. It merely synchronises the
internal PCLK (which is just a delayed version of the PCLK supplied on
the pin) to the rising edge of BCLK, that is, it's just an adjustment
for the unavoidable phase shift between the BCLK and PCLK inputs.
Even this feature is limited: the manual says that the skew between BCLK
and PCLK rising edges must be less than +/- 9ns for a 25MHz part.
(Well, the min. pulse with on PCLK is 9.5ns ...)
The Intel stuff is said to run with double speed internally. The Moto
manual, on the other hand, says that the PCLK/BCLK business allows
the bus to run half the speed of the CPU internals. Since both give you
the bus clock freq, it seems to me to be the same thing.
However, there's a big difference:
In instruction execution timing (as far as I know) Intel gives you the
number of internal clocks while Motorola gives you the number of bus clocks.
That is, if both chips have an instruction which executes in a single
clock (best case), a 25MHZ 68040 executes them with 25MIPS while
a 25/50MHz 486 with 50MIPS. Not as if MIPS would mean anything.
Therefore, I see some reason of calling a clock doubled Intel chip
a 50...100MHz part, but in the case of the 68040 it's just sheer marketing
rubbish. Well, assuming that I'm right in the above. :-)
Zoltan
--
Zoltan Kocsi <zol...@research.canon.oz.au>