Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
2Ghz Powerbook
- Thread starter Dronecatcher
- Start date
- Sort by reaction score
Is there a way to accurately read out the class of G4 CPU via OF or Terminal - or is opening up the machine the only way?
Bear in mind in 2005, people were buying 1.25 and 1.42Ghz Mac Minis only to discover their machines were 1.33 and 1.5Ghz - prior to Apple marketing the faster range.
It's not impossible (but is unlikely) that some of the last Powerbooks were created more equal than others.
EDIT:
http://www.theregister.co.uk/2004/09/28/freescale_g4_7448/
Bear in mind in 2005, people were buying 1.25 and 1.42Ghz Mac Minis only to discover their machines were 1.33 and 1.5Ghz - prior to Apple marketing the faster range.
It's not impossible (but is unlikely) that some of the last Powerbooks were created more equal than others.
EDIT:
http://www.theregister.co.uk/2004/09/28/freescale_g4_7448/
Last edited:
CPU information-including specific model identifiers-can be pulled from Open Firmware.
Of course, you can also "trick" open firmware-when I pulled the info on my 7448 upgrade it reported as a 7455.
All that aside, Apple has the capability to make CPUs report correctly, and I can't imagine them not doing so if indeed the Powerbook has a 7448.
Note that your linked article also mentions the 1mb of L2.
BTW, the die markings are not the ONLY way differentiate the two processors on physical examination...
What is the OF command to identify whether it's a 7447A or 7448?
When I view SystemProfiler.plist, Localizable.strings etc it just appears the machine reads out from predetermined strings rather than a specific hardware code.
What are the variants of the 7447A as the ones in my 1.33 iBook, 1.33 Powerbook and 1.67 Powerbook all appear different (significantly the Processor options in Preferences are gone on DLSD models)?
Last edited:
Hmm, I think, after booting into OF you type the following;
There will be a line like "ff888888: /PowerPC,700@0"
So you type;
I think that should show you the details of the processor, or I might be completely wrong and it won't give you any info at all - I found out about these commands a few weeks ago when I faffing around with my PowerMac trying to find out info about a problem I was having with it, so that's all I know, but I can't remember if it actually told me what model of G4 was in use.
Code:
dev / lsThere will be a line like "ff888888: /PowerPC,700@0"
So you type;
Code:
dev /PowerPC,700 .properties okI think that should show you the details of the processor, or I might be completely wrong and it won't give you any info at all - I found out about these commands a few weeks ago when I faffing around with my PowerMac trying to find out info about a problem I was having with it, so that's all I know, but I can't remember if it actually told me what model of G4 was in use.
Maybe it's the syntax but I either get "unknown word" , "can't find device" or "no active package"
http://osxbook.com/book/bonus/ancient/whatismacosx/arch_boot.html
I found the guide I used. It does work but, yea probably my syntax is borked or something. I was typing it from memory as I couldn't find the article but, yea now I have so have a read through of that.
I found the guide I used. It does work but, yea probably my syntax is borked or something. I was typing it from memory as I couldn't find the article but, yea now I have so have a read through of that.
Firstly, well done for remembering even that much from memory!
Thanks for the link and the commands do work, although there is still no designation of the CPU as 7447A or whatever.
In my 1.33 Powerbook, the CPU is version 80030101, whereas 1.33 iBook is 80030105 and so is my 17" DLSD. All machines are listed as 7447A in specs and yet the DLSD Powerbooks, to my knowledge, are the only OSX Macs that have Processor Options removed in Energy Saver Preferences - indicating frequency scaling and power management are under different control by different means.
I found this; http://pearpc.sourceforge.net/pvr.html
Apparently those numbers are PVR values. Whatever that is. But it would appear, if that table in the above link is anything to go by, that 80030101 corresponds to a revision 1.1 7447A. The x105 variant in your other iBook isn't listed but according to a post (#125) in https://www.thinkclassic.org/viewtopic.php?id=46&p=5, it's a revision 1.5 7447A, it just isn't listed on that PVR table in the first link for whatever reason.
But in short, that confirms they are indeed 7447A G4's.
Edit: Here's a (potentially) more relevant document with PowerPC PVR values:
http://www.nxp.com/files/archives/doc/support_info/PPCPVR.pdf
It also shows the 7448 PVR values as Open Firmware would evidently report them as.
Apparently those numbers are PVR values. Whatever that is. But it would appear, if that table in the above link is anything to go by, that 80030101 corresponds to a revision 1.1 7447A. The x105 variant in your other iBook isn't listed but according to a post (#125) in https://www.thinkclassic.org/viewtopic.php?id=46&p=5, it's a revision 1.5 7447A, it just isn't listed on that PVR table in the first link for whatever reason.
But in short, that confirms they are indeed 7447A G4's.
Edit: Here's a (potentially) more relevant document with PowerPC PVR values:
http://www.nxp.com/files/archives/doc/support_info/PPCPVR.pdf
It also shows the 7448 PVR values as Open Firmware would evidently report them as.
Last edited:
FWIW, I just went to the trouble of disassembling my 17" DLSD and photographing the processor. Here it is. I'm intentionally showing the RAM slots to show that it indeed has DDR2 RAM, but the die markings are also quite legible.
Thanks for digging these up.
I'm pretty sure the DLSDs do have a last variant of 7447A (to explain the frequency scaling) but that still isn't accounted for or described far as I've seen.
Given that the guys you linked to at thinkclassic were busy editing ROMs to fool a boot of OS9, I wouldn't be surprised if it was found that the last production run of DLSDs outstripped supply of 7447As, so 7448s with an edited ROM were pushed out instead.
[doublepost=1466272846][/doublepost]
So, it's a 7447B (despite the board saying 7447A) - now we are getting somewhere!
BTW, my processor was made the same week in 2005(week 35) as the other posters claimed "7448" as per the photos he has now removed.
BTW, in the interest of full disclosure in case anyone tries to claim the image was edited-I take all photographs in RAW mode on my DSLR and use Lightroom to convert the files to JPEGs. All that was done to this particular file was crop it, although some online tools will show it as edited due to the fact that it was handled by Lightroom. LR leaves a "signature" in the EXIF data. I am happy to provide the full, complete .CR2(camera RAW file) on request.
I don't think anyone in this forum would doubt the veracity of your photo
But for the record - it's easy to edit EXIF data or dupe it by other means...So, I'm currently Googling 7447B for clues to it's unique frequency scaling unless anyone's got there first?
It might be worth mentioning since I don't think anyone else picked up on it but Everymac says that because of mrsaggytits or whatever his name is - he actually states in that topic on the 68k mac forum you linked to that he contacted Everymac and badgered them to update their info based on what he claims to be the truth of the matter.
I wasn't using that as a point of reference, nor was I using the Low End Mac index that puts a 7457B into the DLSD
I realised that after I found his posts.
This might be of some use. It's the only thing I've been able to find that makes any reference to the 7447B. There's also a now long-dead PDF named "MC7447ARQI.pdf" that seemed to answer people's questions back in the day when asking about the "upcoming 7447B chips".I don't think anyone in this forum would doubt the veracity of your photo
So, I'm currently Googling 7447B for clues to it's unique frequency scaling unless anyone's got there first?
Btw as to the unique scaling you might have better luck searching for e600 given they renamed the G4 to that after apple stopped using it.
There's this one too on the same site.
Edit: fixed Xenforo's god awful markup bollocks.
Lol fair do's.I realised that after I found his posts.
There's this one too on the same site.
Edit: fixed Xenforo's god awful markup bollocks.
Thanks for links, alas nothing in there explicitly detailing a new speed/power management in the 7447B (at least not in plain english terms).
To logically account for the freq scaling in the DLSD I'll have to assume this:
As (according to Open Firmware) the 1.33 iBook has the same CPU, it doesn't require intelligent, automatic speed/power management because of lower speed ceiling - whereas 1.67Ghz in the DLSD does.
To logically account for the freq scaling in the DLSD I'll have to assume this:
As (according to Open Firmware) the 1.33 iBook has the same CPU, it doesn't require intelligent, automatic speed/power management because of lower speed ceiling - whereas 1.67Ghz in the DLSD does.
I think this might be the most relevant information I can find that answers your questions.
Edit: if it doesn't take you to the right part, it's under "Overview ... > System Software > Power Management".
Thanks for that - very interesting. Pity Apple didn't reference it to a particular CPU instead of just the vague "PowerPC G4" - and if the 7447B is used in other machines, I'd have thought that feature would be an advantage no matter the rated top speed.
EDIT: Sorry, the Apple Developer documentation does link to the CPU used - 7447A...doh.
You can see how speculation can arise...
Last edited:
Hopefully this matter has been settled, but here is some additional information on the physical differences between the 7447x series processors and the 7448.
First of all, I think it's prudent to first post a photograph of a 7448 processor. This particular processor is in a Newertech branded 2ghz upgrade.
I'll draw your attention to a couple of key points on this processor, one of which is manufacture date(2008) and also the particular shape of the 8 in 7448. In addition, look at the other markings on the processor, which identify the rated clock frequency and presumably other properties of the processor.
Note the location and size of the 7448 marking at the bottom of the package. It spans aproximately two of the surface mount components around the perimeter.
Another key consideration is both the aspect ratio and size of the die, as the 7448 was noted for being both smaller and more square than the 7447x.
Note that the 7448 is almost exactly 8mm wide, and slightly over 7mm tall.
Now, let's compare this to the 7447B. On cursory examination, one thing which stands out is that the 7447A marking on the package is almost entirely under one of the surface mount components.
Second, let's examine the size of the 7447B in my 17" DLSD.
Note that the width of the 7447B is virtually identical to the 7448 at 8mm, but the height is just a shade under 10mm. In other words, the 7447B is decidedly and noticeably rectangular.
Now, let's examine the photograph of the claimed "7448" in a DLSD that was posted briefly a few days ago.
I will point out a couple of anomalies with the claimed 7448 processor.
First of all, take a careful look at the die markings. Although the processor APPEARS to say 7448 on the die, note that the form of the "8" looks much more like the "B" on my 7447B as opposed to the "8" on my 7448. Otherwise, the die marking are virtually identical to the processor in my 17" DLSD, right down to the manufacture date. The die markings on the known 7448 are quite different.
Second, I'll again draw your attention to the position and relative size of the package marking relative to other surface mount components. It is more consistent with the arrangement seen on 7447x series processors.
Finally, again pay attention to the aspect ratio-the claimed 7448 has an aspect ratio in line with a 7447x processor, and not with a 7448.
I certainly invite @asaggynoodle to refute any points which I've made if he can do so factually.
First of all, I think it's prudent to first post a photograph of a 7448 processor. This particular processor is in a Newertech branded 2ghz upgrade.
I'll draw your attention to a couple of key points on this processor, one of which is manufacture date(2008) and also the particular shape of the 8 in 7448. In addition, look at the other markings on the processor, which identify the rated clock frequency and presumably other properties of the processor.
Note the location and size of the 7448 marking at the bottom of the package. It spans aproximately two of the surface mount components around the perimeter.
Another key consideration is both the aspect ratio and size of the die, as the 7448 was noted for being both smaller and more square than the 7447x.
Note that the 7448 is almost exactly 8mm wide, and slightly over 7mm tall.
Now, let's compare this to the 7447B. On cursory examination, one thing which stands out is that the 7447A marking on the package is almost entirely under one of the surface mount components.
Second, let's examine the size of the 7447B in my 17" DLSD.
Note that the width of the 7447B is virtually identical to the 7448 at 8mm, but the height is just a shade under 10mm. In other words, the 7447B is decidedly and noticeably rectangular.
Now, let's examine the photograph of the claimed "7448" in a DLSD that was posted briefly a few days ago.
I will point out a couple of anomalies with the claimed 7448 processor.
First of all, take a careful look at the die markings. Although the processor APPEARS to say 7448 on the die, note that the form of the "8" looks much more like the "B" on my 7447B as opposed to the "8" on my 7448. Otherwise, the die marking are virtually identical to the processor in my 17" DLSD, right down to the manufacture date. The die markings on the known 7448 are quite different.
Second, I'll again draw your attention to the position and relative size of the package marking relative to other surface mount components. It is more consistent with the arrangement seen on 7447x series processors.
Finally, again pay attention to the aspect ratio-the claimed 7448 has an aspect ratio in line with a 7447x processor, and not with a 7448.
I certainly invite @asaggynoodle to refute any points which I've made if he can do so factually.
Last edited: