Keep in mind that both the about box in the first image and the CPU monitor in the second image are moved around the screen while the videos are playing.
I could be wrong - but if apple gave me a quad g4 to play with - under the provision that I keep it under my hat... well, lets say that you wouldn't find pictures of it on the net!
This skepticism seems a little hysterical to me ... hasn't it been pretty much accepted that Apple has been testing quad-processor configurations for a while now? It doesn't mean they'll ship one, now or ever.
That said, I would've thought a quad box could launch CPU Monitor in under 5 bounces.
<strong>With all the info we have now I think it's an IBM 4-way 604e server.
People got the old Mac OS to run on these with a lot of hacking, so I wouldn't be surprised if OS X would run as well. Then the movies aren't palying yet the bars are about 40%.
Whatever it is, it's not a 4xG4 and even less a G5.
(btw the G5 does not use multiple cores per die, what you mean is the Power4)
G-news</strong><hr></blockquote>
I thought the G5 was to be truer to the Book E spec. and so is to be multi-core per die. Am I mistaken.
<strong>This skepticism seems a little hysterical to me ... hasn't it been pretty much accepted that Apple has been testing quad-processor configurations for a while now? It doesn't mean they'll ship one, now or ever.
That said, I would've thought a quad box could launch CPU Monitor in under 5 bounces. </strong><hr></blockquote>
This man speak much sense.
The movie (check it out) makes all the stuff about putting multiple CPU meters next to each other a nonsense - it shows the "standard" CPU monitor window with a title bar and spaces between each meter - this would be very hard to fake. In fact, the movie looks very genuine. It appears the creator is running a newer version of the CPU monitor too, which might account for some other issues (like it floats above the dock, not below it.)
We know the box (at one time) contained at least one, and maybe up to six 604 processors. I'm curious to know if this is a hacked Darwin/Mac OS X running on a multi-processor IBM box (doesn't explain "new" CPU monitor), or a real Apple, maybe running 10.2.
There's also the puzzle of how this stuff appears on a UK-based web site - I'd guess philpot's web site - and where it came from (the keyboard menu suggests to me it was also in the UK (I'm in the UK, but have a US flag on my menu bar because I usually need to type #, not £, but sometime I need to type £ more than #).)
Why can't we have a movie with the Apple System Profiler showing us what we really want to know?
If it is a multi 604-based box, the person is trying to pull a fast one - pretending it uses a USB keyboard, and _not_ coming clean about the processor. But if you had Mac OS X running on such a box, you wouldn't post teasers, it'd be such a great hack.
If it really is a quad-G[45], then I can understand the lack of clear documenting evidence, because until that appears, it is still "deniable".
This actually makes me think it _is_ a real quad-G[45], depite the fact it takes 5 bounces to launch the CPU monitor...
It can hold up to 4 604@332Mhz. There are constant rumors that the f50 is able to run mac os 9 with a few hacks. If this is the case mac os X should be a piece of cake. But even if this is all true, don't hold your breath. These puppies go for $10000 used and a dual G4@1Ghz is probably twice as fast ......
It seems to me that we are witnessing a brilliant fake.
Anyone who can program the Mac could "double up" the bars from a "reversed engineered" CPU Monitor, right?
That's a real question--i'm not sure, but it seems reasonable.
A little variance in how high the bars can get, in relation to the real processor(s) on the machine (surely sitting off "camera") and there we have it.
Note how it's the same build but my 7455 is identified as a 7450 while the one in the pic is not. Notice how his kernel is Darwin 5.2 and mine is Darwin 5.2.2. Notice how only one cruncher was started? Notice how the keyrate is displayed differently?
The reason why the processor bars go so high is because Snapz Pro is grabbing such a large portion of the screen at a relatively high frame rate.
Try it.
</strong><hr></blockquote>
Well that was in reply to my message, so I took up your challenge. I downloaded Snapz Pro (moki, if you're reading this, can't you drop the 9 megs of foreign language manuals from the default download), and snapped a largish chunk of my screen as I launched CPU Monitor.
It launched in 1.5 bounces on my dual G4/533. OK, so maybe it was in memory, thus launched faster, so then I launched a few other similarily-sized programs that I don't normally use. None of them took overly long to launch - one or two bounces. At no point did the computer feel sluggish.
I was capturing up to 1000x800 pixels, and aiming for 20 frames per second. (In actual fact it got no-where near that, but Snapz Pro _was_ taking 100% of one of my CPUs.)
I would guess that Snapz Pro is single-threaded, so it can only use one processor, leaving the other(s) free for real processing. I also presume that it records into memory, but at any rate disk access didn't seem too sluggish for me. So, if this quad-processor box is has 4 G5s, DDR memory and a ATA/133 controller, why would CPU monitor take 5 bounces???
Thus, I'm even less convinced that the computer in the movie is a quad-G4 or G5.
[quote]
<strong>
(he he, some people don't want a G5)
</strong><hr></blockquote>
I really do want to believe this thing is a quad-G5, but without real evidence - like from the Apple System Profiler - I remain skeptical.
Personally I think it is a fake - this is my reasoning:
1. Box can easily be IBM server as discussed above - even if PowerPC does not need to have ported MacOS X as could be just displaying a screen shot jpeg on full screen - the cpu monitor can easily be photoshopped onto the jpeg.
2. The moving picture needn't be running on the box in question but could be running on a bonafide shipping mac. It only then requires basic programming to display a fake "new" CPU monitor.
3. If SnapzPro is single-threaded as tested out above, why would the CPU load be distributed relatively evenly across the 4 processors - surely one would be full capturing the video and the others would be relatively low just playing back a couple of small quicktime movies?
If it's really running mac os X then loading the cpu monitor would be slow, simply because this is an old hardware.</strong><hr></blockquote>
Because a box is just a box - you have no idea what is actually inside it.
Plus, it's not the same box - where is the "big red switch" seen in the spy pics in the picture from your URL?
I'd guess the box actually is the IBM, before Creo has done their stuff to it. IBM has discontinued this box, but it is possible that Apple had some kicking around which they might have used to house a test box.
I currently believe that if it is real, and it would be an extraordinarily eloborate hoax if it isn't, then it is likely to be a hacked-up Darwin & Mac OS X running on quad 604s, but if so, wouldn't this be a cool enough hack for someone to want to boast about that, rather than pretend that it is something it is not?
Another question... If Snapz Pro takes about 100% CPU time on my dual G4/533, why doesn't it take anything like that on this box? Maybe it really is a hoax...
That's my last statement on the matter until Philpot finds more time to continue this elaborate deception, or sneeky peak into Apple's future (delete as applicable).
Comments
we should start a 'who has more CPUs' thread
<a href="HTTP://homepage.mac.com/ssmurphy/image1.jpg"; target="_blank">HTTP://homepage.mac.com/ssmurphy/image1.jpg</a>;
and
<a href="http://homepage.mac.com/ssmurphy/image2.jpg"; target="_blank">http://homepage.mac.com/ssmurphy/image2.jpg</a>;
Keep in mind that both the about box in the first image and the CPU monitor in the second image are moved around the screen while the videos are playing.
So what do you think now?
Later Steve
to fix the URL's
[ 02-05-2002: Message edited by: ssmurphy ]
[ 02-05-2002: Message edited by: ssmurphy ]
[ 02-05-2002: Message edited by: ssmurphy ]</p>
This had me going - right up until I opened my utilities folder and made a duplicate copy of the cpu monitor app:
<a href="http://www.elroyonline.net/misc/fake_quad.gif"; target="_blank">http://www.elroyonline.net/misc/fake_quad.gif</a>;
I could be wrong - but if apple gave me a quad g4 to play with - under the provision that I keep it under my hat... well, lets say that you wouldn't find pictures of it on the net!
That said, I would've thought a quad box could launch CPU Monitor in under 5 bounces.
<strong>With all the info we have now I think it's an IBM 4-way 604e server.
People got the old Mac OS to run on these with a lot of hacking, so I wouldn't be surprised if OS X would run as well. Then the movies aren't palying yet the bars are about 40%.
Whatever it is, it's not a 4xG4 and even less a G5.
(btw the G5 does not use multiple cores per die, what you mean is the Power4)
G-news</strong><hr></blockquote>
I thought the G5 was to be truer to the Book E spec. and so is to be multi-core per die. Am I mistaken.
(NOT)
[ 02-05-2002: Message edited by: P.hilB.ot ]
[ 02-05-2002: Message edited by: philbot ]</p>
[ 02-05-2002: Message edited by: Alpha Mac ]</p>
<strong>This skepticism seems a little hysterical to me ... hasn't it been pretty much accepted that Apple has been testing quad-processor configurations for a while now? It doesn't mean they'll ship one, now or ever.
That said, I would've thought a quad box could launch CPU Monitor in under 5 bounces.
This man speak much sense.
The movie (check it out) makes all the stuff about putting multiple CPU meters next to each other a nonsense - it shows the "standard" CPU monitor window with a title bar and spaces between each meter - this would be very hard to fake. In fact, the movie looks very genuine. It appears the creator is running a newer version of the CPU monitor too, which might account for some other issues (like it floats above the dock, not below it.)
We know the box (at one time) contained at least one, and maybe up to six 604 processors. I'm curious to know if this is a hacked Darwin/Mac OS X running on a multi-processor IBM box (doesn't explain "new" CPU monitor), or a real Apple, maybe running 10.2.
There's also the puzzle of how this stuff appears on a UK-based web site - I'd guess philpot's web site - and where it came from (the keyboard menu suggests to me it was also in the UK (I'm in the UK, but have a US flag on my menu bar because I usually need to type #, not £, but sometime I need to type £ more than #).)
Why can't we have a movie with the Apple System Profiler showing us what we really want to know?
If it is a multi 604-based box, the person is trying to pull a fast one - pretending it uses a USB keyboard, and _not_ coming clean about the processor. But if you had Mac OS X running on such a box, you wouldn't post teasers, it'd be such a great hack.
If it really is a quad-G[45], then I can understand the lack of clear documenting evidence, because until that appears, it is still "deniable".
This actually makes me think it _is_ a real quad-G[45], depite the fact it takes 5 bounces to launch the CPU monitor...
Hmmmm.......
<a href="http://www.creo.com/products/software_solutions/data/Server5000/index.asp"; target="_blank">http://www.creo.com/products/software_solutions/data/Server5000/index.asp</a>;
It can hold up to 4 604@332Mhz. There are constant rumors that the f50 is able to run mac os 9 with a few hacks. If this is the case mac os X should be a piece of cake. But even if this is all true, don't hold your breath. These puppies go for $10000 used and a dual G4@1Ghz is probably twice as fast ......
Anyone who can program the Mac could "double up" the bars from a "reversed engineered" CPU Monitor, right?
That's a real question--i'm not sure, but it seems reasonable.
A little variance in how high the bars can get, in relation to the real processor(s) on the machine (surely sitting off "camera") and there we have it.
That will spit this back at ya, heres my G4
[code]
[localhost:/usr/bin] patrick% hostinfo
Mach kernel version:
Darwin Kernel Version 5.2:
Fri Dec 7 21:39:35 PST 2001; root:xnu/xnu-201.14.obj~1/RELEASE_PPC
Kernel configured for up to 2 processors.
1 processor is physically available.
Processor type: ppc7400 (PowerPC 7400)
Processor active: 0
Primary memory available: 1024.00 megabytes.
Default processor set: 52 tasks, 163 threads, 1 processors
Load average: 0.73, Mach factor: 0.71
[localhost:/usr/bin] patrick%
</pre><hr></blockquote>
<a href="http://www.geocities.com/locke300/Picture.jpg"; target="_blank">http://www.geocities.com/locke300/Picture.jpg</a>;
I don't know if it is real or not, but I thought I would post the link here and see what you guys thing.
Purhaps?
[ 02-05-2002: Message edited by: Falcon ]</p>
[ 02-05-2002: Message edited by: Jeremiah Rich ]</p>
Duh.
The reason why the processor bars go so high is because Snapz Pro is grabbing such a large portion of the screen at a relatively high frame rate.
Try it.
(he he, some people don't want a G5)
<strong>I hate to add more fuel to the fire, but I was pointed to this:
<a href="http://www.geocities.com/locke300/Picture.jpg"; target="_blank">http://www.geocities.com/locke300/Picture.jpg</a>;
I don't know if it is real or not, but I thought I would post the link here and see what you guys thing.</strong><hr></blockquote>
Here's what my RC5 output looks like:
[code][24] ceugene@duet:~> unix/dnet/dnetc
distributed.net client for Mac OS X Copyright 1997-2001, distributed.net
RC5 PowerPC and AltiVec assembly by Dan Oetting
Enhancements for 604e CPUs by Roberto Ragusa
Please visit <a href="http://www.distributed.net/"; target="_blank">http://www.distributed.net/</a>; for up-to-date contest information.
dnetc v2.8016-470-CTR-01102321 for Mac OS X (Darwin 5.2.2).
Please provide the *entire* version descriptor when submitting bug reports.
The distributed.net bug report pages are at <a href="http://www.distributed.net/bugs/"; target="_blank">http://www.distributed.net/bugs/</a>;
Using email address (distributed.net ID) '[email protected]'
[Feb 06 12:26:27 UTC] Automatic processor detection found 2 processors.
[Feb 06 12:26:27 UTC] Loading crunchers with work...
[Feb 06 12:26:27 UTC] Automatic processor type detection found
a PowerPC 7450 (G4) processor.
[Feb 06 12:26:27 UTC] RC5: using core #5 (crunch-vec-7450).
[Feb 06 12:26:27 UTC] RC5: Loaded FCDF6A09:80000000:8*2^28 (58.00% done)
[Feb 06 12:26:27 UTC] RC5: Loaded FCDF6A5E:A0000000:6*2^28 (14.70% done)
[Feb 06 12:26:27 UTC] RC5: 3 packets (17.00 stats units) remain in
buff-in.rc5
[Feb 06 12:26:27 UTC] RC5: 5 packets (34.00 stats units) are in buff-out.rc5
[Feb 06 12:26:27 UTC] 2 crunchers ('a' and 'b') have been started.
.....10%.....20%.....30%.....40%.....50%.....60%.. ...70%.....80%.....90%....100
[Feb 06 12:27:54 UTC] RC5: Completed FCDF6A09:80000000 (8.00 stats units)
0.00:01:26.71 - [10,397,907 keys/s]
[Feb 06 12:27:54 UTC] RC5: Loaded FCDF6A66:B0000000:5*2^28
[Feb 06 12:27:54 UTC] RC5: Summary: 1 packet (8.00 stats units)
0.00:01:26.71 - [10.39 Mkeys/s]
[Feb 06 12:27:54 UTC] RC5: 2 packets (12.00 stats units) remain in
buff-in.rc5
Projected ideal time to completion: 0.00:02:30.00
[Feb 06 12:27:54 UTC] RC5: 6 packets (42.00 stats units) are in buff-out.rc5
.....10%.....20%.....30%.....40%.....50%.....60%.. ...70%.....80%.....90%....100
[Feb 06 12:28:39 UTC] RC5: Completed FCDF6A5E:A0000000 (6.00 stats units)
0.00:02:12.03 - [10,394,827 keys/s]
[Feb 06 12:28:39 UTC] RC5: Loaded FCDF6AB0:C0000000:4*2^28
[Feb 06 12:28:39 UTC] RC5: Summary: 2 packets (14.00 stats units)
0.00:02:12.39 - [17.17 Mkeys/s]
[Feb 06 12:28:39 UTC] RC5: 1 packet (8.00 stats units) remains in
buff-in.rc5
Projected ideal time to completion: 0.00:01:40.00
[Feb 06 12:28:39 UTC] RC5: 7 packets (48.00 stats units) are in buff-out.rc5
.....10%.....20%.....30%.....40%.....50%.....60%.. ...70%.....80%.....90%....100
[Feb 06 12:30:04 UTC] RC5: Completed FCDF6A66:B0000000 (5.00 stats units)
0.00:02:09.14 - [10,392,390 keys/s]
[Feb 06 12:30:04 UTC] RC5: Loaded FCDF6AB1:60000000:8*2^28
[Feb 06 12:30:04 UTC] RC5: Summary: 3 packets (19.00 stats units)
0.00:03:36.58 - [16.69 Mkeys/s]
[Feb 06 12:30:04 UTC] RC5: 0 packets remain in buff-in.rc5
[Feb 06 12:30:04 UTC] RC5: 8 packets (53.00 stats units) are in buff-out.rc5
[Feb 06 12:30:04 UTC] Connected to us.v27.distributed.net:2064...
[Feb 06 12:30:04 UTC] The keyserver says: "Happy cracking. Hail Eris!"
[Feb 06 12:30:04 UTC] Retrieved project state data from server. (cached)
[Feb 06 12:30:05 UTC] RC5: Retrieved stats unit 49 of 49 (100.00%)
[Feb 06 12:30:07 UTC] RC5: Sent 8 packets (53.00 stats units) to server.
[Feb 06 12:30:07 UTC] Connection closed.
.....10%.....20%.....30%.....40%.....50%.....60%.. ...70%.....80%.....90%....100
[Feb 06 12:30:23 UTC] RC5: Completed FCDF6AB0:C0000000 (4.00 stats units)
0.00:01:43.36 - [10,387,766 keys/s]
[Feb 06 12:30:23 UTC] RC5: Loaded FCDD65F3:70000000:8*2^28
[Feb 06 12:30:23 UTC] RC5: Summary: 4 packets (23.00 stats units)
0.00:03:55.94 - [19.87 Mkeys/s]</pre><hr></blockquote>
Note how it's the same build but my 7455 is identified as a 7450 while the one in the pic is not. Notice how his kernel is Darwin 5.2 and mine is Darwin 5.2.2. Notice how only one cruncher was started? Notice how the keyrate is displayed differently?
<strong>
The reason why the processor bars go so high is because Snapz Pro is grabbing such a large portion of the screen at a relatively high frame rate.
Try it.
</strong><hr></blockquote>
Well that was in reply to my message, so I took up your challenge. I downloaded Snapz Pro (moki, if you're reading this, can't you drop the 9 megs of foreign language manuals from the default download), and snapped a largish chunk of my screen as I launched CPU Monitor.
It launched in 1.5 bounces on my dual G4/533. OK, so maybe it was in memory, thus launched faster, so then I launched a few other similarily-sized programs that I don't normally use. None of them took overly long to launch - one or two bounces. At no point did the computer feel sluggish.
I was capturing up to 1000x800 pixels, and aiming for 20 frames per second. (In actual fact it got no-where near that, but Snapz Pro _was_ taking 100% of one of my CPUs.)
I would guess that Snapz Pro is single-threaded, so it can only use one processor, leaving the other(s) free for real processing. I also presume that it records into memory, but at any rate disk access didn't seem too sluggish for me. So, if this quad-processor box is has 4 G5s, DDR memory and a ATA/133 controller, why would CPU monitor take 5 bounces???
Thus, I'm even less convinced that the computer in the movie is a quad-G4 or G5.
[quote]
<strong>
(he he, some people don't want a G5)
</strong><hr></blockquote>
I really do want to believe this thing is a quad-G5, but without real evidence - like from the Apple System Profiler - I remain skeptical.
Cheers...
1. Box can easily be IBM server as discussed above - even if PowerPC does not need to have ported MacOS X as could be just displaying a screen shot jpeg on full screen - the cpu monitor can easily be photoshopped onto the jpeg.
2. The moving picture needn't be running on the box in question but could be running on a bonafide shipping mac. It only then requires basic programming to display a fake "new" CPU monitor.
3. If SnapzPro is single-threaded as tested out above, why would the CPU load be distributed relatively evenly across the 4 processors - surely one would be full capturing the video and the others would be relatively low just playing back a couple of small quicktime movies?
No sorry, I need more convincing than that.
<a href="http://www.qualprint.com/technical/techimage/brisque.gif"; target="_blank">http://www.qualprint.com/technical/techimage/brisque.gif</a>;
(credit: pic found by Wacky in spymac forum)
If it's really running mac os X then loading the cpu monitor would be slow, simply because this is an old hardware.
<strong>Why do do still argue when it's clear that this is no G4 neither G5? It's a 604 machine from creo called brisque:
<a href="http://www.qualprint.com/technical/techimage/brisque.gif"; target="_blank">http://www.qualprint.com/technical/techimage/brisque.gif</a>;
(credit: pic found by Wacky in spymac forum)
If it's really running mac os X then loading the cpu monitor would be slow, simply because this is an old hardware.</strong><hr></blockquote>
Because a box is just a box - you have no idea what is actually inside it.
Plus, it's not the same box - where is the "big red switch" seen in the spy pics in the picture from your URL?
I'd guess the box actually is the IBM, before Creo has done their stuff to it. IBM has discontinued this box, but it is possible that Apple had some kicking around which they might have used to house a test box.
I currently believe that if it is real, and it would be an extraordinarily eloborate hoax if it isn't, then it is likely to be a hacked-up Darwin & Mac OS X running on quad 604s, but if so, wouldn't this be a cool enough hack for someone to want to boast about that, rather than pretend that it is something it is not?
Another question... If Snapz Pro takes about 100% CPU time on my dual G4/533, why doesn't it take anything like that on this box? Maybe it really is a hoax...
That's my last statement on the matter until Philpot finds more time to continue this elaborate deception, or sneeky peak into Apple's future (delete as applicable).