Re:But what about the PowerBooks!?
on
New Mac System Specs
·
· Score: 2, Informative
Pentium 4s are now around 95% RISC
No they're not. RISC is an architecture design model, not an implementation tool. A P4 is a CISC processor implemented as two or three separate processors pipelined together: there's the first stage that rewrites the CISC code as RISC code, then there's a vertical microcode processor that resembles RISC only in so far as early RISC processors were modelled on vertical microcode machines (the IBM 801 could be said to have been both), then there's the FPU and some post-instruction fixup and the hardware that manages the whole mess.
But having a RISC-like core to do the heavy lifting doesn't make it RISC, any more than having a VLIW-like horizontal microcode core makes the AMD processor VLIW. It's better to say that the P4 is a hardware emulator for the x86 instruction set with a RISC-like processor as part of the emulation. It's not RISC, though, any more than running a software Playstation emulator on a Windows box makes the MIPS RISC processor being emulated into a CISC.
none of the people I know would ever need to use it longer than that on battery
You've never been at a conference or in an airport lounge hunting for an open power point so you can get an extra half hour of work done before your talk. There's a lot of people for whom having enough slack so they can get through the morning session without a recharge is a really really big deal.
The G5 pipeline is 14 stages. That's considered long?
For a RISC processor, that's enormous. Even a 9-stage pipeline is pushing it: MIPS topped out with 8 stages and the Alpha had a 7-stage pipeline. Itanium was considered dubious because of its 10-stage one (not that that was the only thing that hurt the Itanium). The only RISCy processor I've run into with a >10 stage pipeline was the XScale... and its long pipeline REALLY hurt its clock-for-clock performance compared to the StrongARM.
That actually makes a lot of sense: you could have the single-core chip in the iBook and Mac mini, the dual-core in the Powerbook and Mac mini DV version with the HD input.
The new Freescale chips aren't out yet, and except for thermal issues they're going to take as much of a redesign as G5s would.
That's like saying that apart from the mass murders and war crimes Hitler was a great guy. The thermal issues are huge, and the latency issues are huge... the G5 is a classic Pentium-style long-pipeline model... and the point of a long pipeline is to push ther clock. But, look, the G4 is catching up with the G5 on clock!
The G4, if not for the lousy front end bus, would get more done per clock than the G5... and the new Freescale chips finally get rid of that lousy bus. No, the new G4 could well be faster, clock for clock, than a G5... even before you count in the dual-core chips.
so they can start EOLing the 32-bit stuff
64 bit is not an unmixed win. The only reason the AMD 64 bit stuff gives such a boost is that they could use the 64 bit-ness to get rid of a lot of bad instruction set design. The Power PC doesn't have that problem to resolve.
At work I use a program called OSX2X that lets me control a Mac and a PC screen next to each other from the same mouse and keyboard, using VNC server on the PC side but only sending keyboard and mouse updates and not transferring screen contents.
No, really, a Powerbook G5 with 90 minutes of battery life is a dumb idea. I want to see the dual-core G4. We all want to see the dual-core G4. Only trolls are still going on about the Powerbook G5... I hope they don't convince Apple to try it anyway...
I imagine the idea is that they would get experience with a near-vacuum environment while being close enough to get back if everything goes pear-shaped.
Equatorial is a better spot for delta-vee reasons, but the two week lunar night is a problem.
You don't want a G5 powerbook. You want a dual-G4 powerbook. the new Freescale dual-G4 chip breaks the G4 166 MHz system bus bottleneck, *and* gives you dual-core as well. It would breeze past any underclocked G5 Apple could fit in a laptop the size of a Powerbook.
Re:But what about the PowerBooks!?
on
New Mac System Specs
·
· Score: 5, Interesting
Because a G5 powerbook is "the mother of all thermal challenges" (direct quote from Apple).
You don't want a G5 powerbook. You want a dual-G4 powerbook. the new Freescale dual-G4 chip breaks the G4 166 MHz system bus bottleneck, *and* gives you dual-core as well. It would breeze past any underclocked G5 Apple could fit in a laptop the size of a Powerbook.
This is the first Physorg article that I've seen listed in/. that actually provides an offsite link for the story! Are they getting mellow, or did they just make a mistake and will go back to their usual "tarpit" methods?
There won't be surreptitious widgets stuffed into the various websites they visit in the course of a day. If a site offers a widget, it won't be automatically downloaded when you visit the site.
I find your faith in the intelligence of web developers charmingly naive.
I have used MusicMagic Mixer with iTunes and the playlists work just fine on my iPod Shuffle, so I'm really confused about how this makes your karma with the "awful Rio Music Manager" superior. Can you elaborate on that?
now just write a script to rename all those mp3s back to their original titles since the ipod renames them all
Drag and drop them into iTunes and it reads the name out of the ID3 tags, renames the files, and organises it into a nice artist-album-track directory tree for you.
DRM is the only answer to protecting Sony's own copyrights, as they have the rights to a lot of music distribution already. What is the alternative?
No DRM. I mean the format the music Industry WANTS us to use, the one they bemoan the loss of sales in, has no DRM whatsoever, *and* is higher quality than any of the DRM-protected formats. I mean if the Industry really thoughjt DRM was anything but a lever to control the market with they'd quit shipping CDs today!
What is the big deal here?
The big deal is this: if you send someone a secret message, and you send them the key to read the message, then they can read the message. Right? Well, that's what DRM does, it sends me the music, it sends me the key to play the music. Do that, and there is NO WAY ON GOD'S GREEN EARTH to stop me from listening to the music. And if I can listen to it, I can re-record it, and at that point... hey... the DRM is gone gone gone.
DRM can't keep people from copying music. All it can to is to cause the first person who copies it to lose a miniscule amount of detail that they're never going to hear when they play it on an MP3 player as they walk down a busy street.
And in exchange for that completely irrelevant cost, they force everyone who buys their music to go through some variant of "Mix, Burn, Rip" to ensure that they can still listen to their music when MusicMatch^WNapster goes out of business.
the company that is the most relevant in the personal audio world these days doesn't make it easy for you to transfer your music back off of their pocket music device, either
Stop thinking that you are going to get away with stolen music for ever its illegal.
I pay for my music, thank you very much. And I convert it to MP3 or unencrypted AAC format so I can keep my own music even if Real or Napster or Apple quits supporting their key management servers.
But this time you're way out of line technically as well as philosophically. Sony's "obfuscation" of MP3 files has nothing to do with DRM. You HAVE the unencrypted files before you install them!
Slashdot Mirror
Pentium 4s are now around 95% RISC
No they're not. RISC is an architecture design model, not an implementation tool. A P4 is a CISC processor implemented as two or three separate processors pipelined together: there's the first stage that rewrites the CISC code as RISC code, then there's a vertical microcode processor that resembles RISC only in so far as early RISC processors were modelled on vertical microcode machines (the IBM 801 could be said to have been both), then there's the FPU and some post-instruction fixup and the hardware that manages the whole mess.
But having a RISC-like core to do the heavy lifting doesn't make it RISC, any more than having a VLIW-like horizontal microcode core makes the AMD processor VLIW. It's better to say that the P4 is a hardware emulator for the x86 instruction set with a RISC-like processor as part of the emulation. It's not RISC, though, any more than running a software Playstation emulator on a Windows box makes the MIPS RISC processor being emulated into a CISC.
none of the people I know would ever need to use it longer than that on battery
You've never been at a conference or in an airport lounge hunting for an open power point so you can get an extra half hour of work done before your talk. There's a lot of people for whom having enough slack so they can get through the morning session without a recharge is a really really big deal.
The G5 pipeline is 14 stages. That's considered long?
For a RISC processor, that's enormous. Even a 9-stage pipeline is pushing it: MIPS topped out with 8 stages and the Alpha had a 7-stage pipeline. Itanium was considered dubious because of its 10-stage one (not that that was the only thing that hurt the Itanium). The only RISCy processor I've run into with a >10 stage pipeline was the XScale... and its long pipeline REALLY hurt its clock-for-clock performance compared to the StrongARM.
That actually makes a lot of sense: you could have the single-core chip in the iBook and Mac mini, the dual-core in the Powerbook and Mac mini DV version with the HD input.
The new Freescale chips aren't out yet, and except for thermal issues they're going to take as much of a redesign as G5s would.
That's like saying that apart from the mass murders and war crimes Hitler was a great guy. The thermal issues are huge, and the latency issues are huge... the G5 is a classic Pentium-style long-pipeline model... and the point of a long pipeline is to push ther clock. But, look, the G4 is catching up with the G5 on clock!
The G4, if not for the lousy front end bus, would get more done per clock than the G5... and the new Freescale chips finally get rid of that lousy bus. No, the new G4 could well be faster, clock for clock, than a G5... even before you count in the dual-core chips.
so they can start EOLing the 32-bit stuff
64 bit is not an unmixed win. The only reason the AMD 64 bit stuff gives such a boost is that they could use the 64 bit-ness to get rid of a lot of bad instruction set design. The Power PC doesn't have that problem to resolve.
At work I use a program called OSX2X that lets me control a Mac and a PC screen next to each other from the same mouse and keyboard, using VNC server on the PC side but only sending keyboard and mouse updates and not transferring screen contents.
When's the Mac Mini due for an update
Maybe by Thanksgiving? It's only been out since January... Apple doesn't do refreshes that fast.
Mod parent redundant. :)
No, really, a Powerbook G5 with 90 minutes of battery life is a dumb idea. I want to see the dual-core G4. We all want to see the dual-core G4. Only trolls are still going on about the Powerbook G5... I hope they don't convince Apple to try it anyway...
G4s are only PPC CPUs, but G5s are PPC-64 CPUs...
Unless he's got more than 4G of RAM in that powerbook, all PPC-64 will do is slow him down.
I've been using Alpha almost since it was released, and unless you really need the address space you're better off with 32-bit pointers.
I imagine the idea is that they would get experience with a near-vacuum environment while being close enough to get back if everything goes pear-shaped.
Equatorial is a better spot for delta-vee reasons, but the two week lunar night is a problem.
I do think Apple will _call_ the 8641-based laptops 'G5's though
How about "G5 Mobile"?
While I doubt the code-name is an example of this
Why?
You don't want a G5 powerbook. You want a dual-G4 powerbook. the new Freescale dual-G4 chip breaks the G4 166 MHz system bus bottleneck, *and* gives you dual-core as well. It would breeze past any underclocked G5 Apple could fit in a laptop the size of a Powerbook.
Because a G5 powerbook is "the mother of all thermal challenges" (direct quote from Apple).
You don't want a G5 powerbook. You want a dual-G4 powerbook. the new Freescale dual-G4 chip breaks the G4 166 MHz system bus bottleneck, *and* gives you dual-core as well. It would breeze past any underclocked G5 Apple could fit in a laptop the size of a Powerbook.
This is the first Physorg article that I've seen listed in /. that actually provides an offsite link for the story! Are they getting mellow, or did they just make a mistake and will go back to their usual "tarpit" methods?
There won't be surreptitious widgets stuffed into the various websites they visit in the course of a day. If a site offers a widget, it won't be automatically downloaded when you visit the site.
I find your faith in the intelligence of web developers charmingly naive.
Yeah, you have to consider formats. mp4/aac bring nothing to the table I don't get through ogg and mp3
ogg and mp3 bring nothing to the table I don't get through AAC.
except for DRM
That's not part of MP4/AAC, that's an Apple proprietary layer on top of it. The same thing could be done with ogg or mp3.
Karma sounds better
Bet you it doesn't, the iPod Shuffle is damn good, and has better bass response than any other player I've tried... including the disk-based iPods.
and has superior onboard tools for handling playlists
The point of this exersize was NOT having to fiddle with playlists while you're listening.
I don't have to base my mixing decision on file formats.
Ah, it supports MP4/AAC as well?
NO device, it seems, supports all the possible formats. You always have to include formats in your calculations.
I have used MusicMagic Mixer with iTunes and the playlists work just fine on my iPod Shuffle, so I'm really confused about how this makes your karma with the "awful Rio Music Manager" superior. Can you elaborate on that?
There seems to be a widely held misconception that MP3 is an open format.
In practice it is, the MP3 genie is ALSO out of the Fraunhofer bottle.
Apple allows you to create a DRMed file, they don't try and DRM your own files behind your back.
now just write a script to rename all those mp3s back to their original titles since the ipod renames them all
Drag and drop them into iTunes and it reads the name out of the ID3 tags, renames the files, and organises it into a nice artist-album-track directory tree for you.
DRM is the only answer to protecting Sony's own copyrights, as they have the rights to a lot of music distribution already. What is the alternative?
No DRM. I mean the format the music Industry WANTS us to use, the one they bemoan the loss of sales in, has no DRM whatsoever, *and* is higher quality than any of the DRM-protected formats. I mean if the Industry really thoughjt DRM was anything but a lever to control the market with they'd quit shipping CDs today!
What is the big deal here?
The big deal is this: if you send someone a secret message, and you send them the key to read the message, then they can read the message. Right? Well, that's what DRM does, it sends me the music, it sends me the key to play the music. Do that, and there is NO WAY ON GOD'S GREEN EARTH to stop me from listening to the music. And if I can listen to it, I can re-record it, and at that point... hey... the DRM is gone gone gone.
DRM can't keep people from copying music. All it can to is to cause the first person who copies it to lose a miniscule amount of detail that they're never going to hear when they play it on an MP3 player as they walk down a busy street.
And in exchange for that completely irrelevant cost, they force everyone who buys their music to go through some variant of "Mix, Burn, Rip" to ensure that they can still listen to their music when MusicMatch^WNapster goes out of business.
Stop thinking that you are going to get away with stolen music for ever its illegal.
I pay for my music, thank you very much. And I convert it to MP3 or unencrypted AAC format so I can keep my own music even if Real or Napster or Apple quits supporting their key management servers.
But this time you're way out of line technically as well as philosophically. Sony's "obfuscation" of MP3 files has nothing to do with DRM. You HAVE the unencrypted files before you install them!