If two sequential tracks on an album are effectively one work, there are plenty of good, and even free, mixing programs that you can use to create a single merged track with the fade you want.
If the branch predictor is good enough, it really doesn't matter.
If that were the case, the G5 and the P4 wouldn't be dead ends. If that were the case, G4 wouldn't be getting more IPC than the G5. If that were the case, PIII wouldn't be getting more IPC than the P4. Pipeline does matter, in the real world.
Pipeline matters. Instruction set matters. Register file size matters.
It got toasted on SPECint too (a factor of 2x at the same clock-speed = toasted)
That's comparing gcc (which isn't SPEC-optimised) against Intel's compiler, and... is that taking dual-core results and dividing by two, or is that the single-core versus the single-core? That makes a BIG difference to the multithreading benchmarks.
The memory bus is 667MHz for the MPC8641D, not 768MHz.
OK, that's changed. They previously listed it as half clock. 667MHz on each of two busses is still nothing to sneeze at.
cache and memory latencies. These things aren't any lower on the G4 than they are on Yonah.
Cache and memory latencies are significantly aggravated by a long pipeline, which is why the G5 still gets fewer IPC than the G4 despite the G5's truly impressive design.
Yonah's pipeline is probably at least twice as long as the G4's: the P6 pipeline is 10+ stages and the latest version added at least 4 more.
Oh, yeh, Yonah isn't all that new a design. It's another spin on the venerable P6 core, with SSE3 and a better shared cache this time around.
I would like to see the source of your SPEC comments.
Apple:)
Apple has published SPEC for the MPC8461D? Where?
The benchmarks I've seen comparing the G4 (even with the e600 core) with the Core Duo are all based on the 7448 and earlier, and all of those are crippled by the slow memory bus. These results are consistent with Yonah's main performance advantage being memory speed. When you take a system with a single core and a 166 MHz memory bus (like the Powerbook) and compare it to one with dual core and a 533 MHz memory bus of course it's going to get its ass kicked... but the benchmark it really got toasted on was stream... and that's a *memory bandwidth* test.
Take an MPC8641D and put memory on BOTH of the 768 MHz memory busses and if it doesn't beat Yonah I'd be very much surprised.
Right now 3d virtual reality is about where text adventure was in the early '70s. Infocom kept the text adventure alive pretty long, but 2d and then 3d pretty much took it over, and the only place you got text role playing was in themed chat systems... MUDs.
I'm predicting that as free and cheap 3d virtual reality gets more common, the kinds of hack-n-slash stuff you see in MMORPGs now are going to become the kinds of things hobbyists put together inside virtual worlds that exist for other reasons. Right now that stuff is really crude by comparison, but it's just a matter of time before realistic graphics become as generic as verb-noun parsers and scripted objects in MUDs.
So the kinds of things that go on in 3d games now will be like LARPing in VR. SCA-type stuff. What will *companies* be doing instead? Heck if I know... interactive movies with licensed celebrity characters and paid actors?
You can pay people real money to farm stuff for you... even if they know you bought it, it's not different from showing up in a nice car and playing like a duffer.
The new dual-core G4 is fairly primitive technology by today's standards.
The G4 core is no slouch. With a short pipeline it does very well and doesn't require a sophisticated out-of-order processor to get good performance. Its biggest shortcoming is the slow bus (166 MHz, up to 200 in the 7448, though Apple doesn't provide the 7448 with a 200 MHz bus). This is similar to the problems of the P6 core (which Yonah is based on) before Intel grafted a faster FSB to it.
I would like to see the source of your SPEC comments. SO far as I know there haven't been comparable benchmarks of the MPC8641D and Yonah on similar hardware. You can't generalise from the performance of the 7448 because it's still crippled by the slow bus of the previous G4s. Do you actually have SPEC results for the 8641D on a 768 MHz bus system?
I'd just like someone to build a worthy successor to the pre-ARM Clies. I have a Sony SJ-22 and I haven't seen anything to touch it. The later Clies were gimmicky monstrosities (half of them didn't even have Jog-dials!), and Palm seems to have dropped the ball after Hawkins came back. I've tried three generations of Windows-powered devices and, well, it's still basically a desktop OS.
Palm needs to take PalmOS 4 and replace the AMX kernel with some open source alternative, and go back to their roots. Possibly using Coldfire instead of ARM will allow for mostly-native emulation...?
The mouse doesn't move vertically, it moves side to side, it's just held vertically.
I think something shaped more like a joystick might work better, but you still lose too much fine control by having to hold the mouse rather than just resting your fingers on top of it. That's a bigger problem than "gorilla arm".
"business decisions were made based on the value of those patents, and there would be deep concern from companies if that IP value vanished."
How much intellectual property value is there in any software patent that can't be recovered within, say, three to five years? What software patents are actively being used to generate significant revenue? How long are they likely to continue generating that income? Could a transition period like that make sense?
Exposé isn't cut from whole cloth, either: cool transitions between display layouts and virtual screens go all the way back to the Amiga, and virtual window managers on UNIX have used them for desktop transitions for a while now.
Similarly, thumbnail views of a larger vitual workspace are common.
Exposé combines a cool transition with a dynamically created layout, and it looks cool, but frankly I'd rather Apple work on a virtual desktop or even trying to really take advantage of the 3d capabilities of OpenGL than more bling like Expose, Genie, and Dashboard ripples.
...they didn't switch for performance, or heat, or anything like that. The cost/performance ratio of silicon is switching back and forth between architectures all the time, and right now the Power PC is doing relatively well. Apple's had it a LOT worse in the past. In fact, between the time the G5 Powermac came out and when Jobs made the announcement that he was tired of waiting for a 3 GHz G5 the G5 had actually ramped up in speed better than Intel's chips. Intel had the multiple cores, but IBM was close enough that Apple came out with the Quad G5 before the iMac Core Duo.
Jobs had been against the Power PC from the start. The switch to Power PC was made while he was away at NeXT, and the first software that came out of the Apple/NeXT merger was Rhapsody... on Power PC AND on Intel. NeXT already had multi-platform executables, and the proposed way forward in 1997 was to Yellow Box (OpenStep on Rhapsody, what became Cocoa) with existing Mac OS applications running under Blue Box (which evolved into Classic). Jobs could kill two birds with one stone... he had a path to back away from the Power PC, and he could dump the (truly awful) Mac OS API.
It was not to be.
The ISVs and customers rioted, and they went back to the drawing board, came up with Carbon as an intermediate API, and when OS X (finally) came out it supported Carbonised Mac OS apps almost natively, Cocoa, and legacy apps under emulation in Classic.
The reason they did this now, instead of in 2000 or 2003 or 2008, is because they were confident they could get away with it now. And I think the trigger happened in 2004. Late in the year they pulled the last OS-9-compatible Powermac G4 from the Apple Store... and nobody made a fuss. They'd tried it before, but every time the users were up in arms. This told them they'd finally scotched the classic serpent and could complete the conversion to NeXTstep.
Something I do truly believe in is that there is statistical evidence that suggests we are in a NATURAL warming trend. Ice data collected in the Arctic and Antarctic suggest that weather patterns are cyclical, and we are about due for a natural warming of global temperatures.
Yes, they're cyclical, but the last 4 cycles have shown long periods of cold with short peaks at or above the current temperatures. These peaks are about 10,000 years long. The last one started over 10,000 years ago.
The data is very noisy, and you can't say much with certainly, but only the most hopeful of observers could look at it and argue that the evidence suggests we're due for a warmer period... on the contrary, we're due for an ice age.
The fella is missing the biggest advantage to UNIX, and that's got nothing to do with whether you're running as root or Administrator or Fred Nurk or Minnie Bannister.
THe bissgest security advantage that UNIX has is that it doesn't have Internet Explorer, Outlook, Windows Media Player, and all the other software that uses that Typhoid Mary of the Internet, the Microsoft HTML control.
The idea that it's even vaguely acceptable to not only build the entire desktop environment about a web browser, but to make that browser... and not the applications that call it... responsible for determining what it's going to let a document do, is just so mind-blowing that seven years after I first saw what was then called Active Desktop I'm still having trouble believing that they're still doing it.
... he thought the Earth was significantly smaller than was known at the time, and if it wasn't for the sheer good luck that there was a landmass unknown to either him or his detractors in the way, and that it was so much closer to the European than the Asian side of the "world ocean", he'd have just been another leader of an ill-fated expedition.
On Windows and X11 I typically keep a notepad window open, and use it for all non-english text-entry. The fact that you don't need to do this with Cocoa and many Carbon apps (Carbon apps, particularly the ones based on ports from UNIX or Windows, don't always seem to hook in to OS X as well as Cocoa) is great.
But...
You should be able to copy and paste from TextEdit into the GIMP no matter what the character set. This isn't a real solution, but it's a workaround that should be acceptable for an application where text input is (or should be) a relatively rare operation.
That's correct: the warranty doesn't include shipping, and they only support shipping via a fairly pricey DHL Express service... but that's included in the $30. They sent me a return-shipment box that arrived the next business day, and the replacement unit was in my hands three business days after I dropped it off at DHL.
Let's pretend Adobe dropped everything to focus on a transition to Universal binaries right after WWDC last year.
Let's pretend that somewhere in Adobe's management structure there exists a decision maker who is capable of rational thought. Even without the Intel switch, Codewarrior has been a lame duck since 1997. That's when Apple first tried to switch to the NeXT codebase and relegate the classic Mac OS API to "Blue Box". Carbon was explicitly a transition API, at first, and Adobe seems to have decided that Apple's subsequent use of Carbon internally meant that Codewarrior was safe. Carbon no longer means OS 9, and Carbon support shouldn't be assumed to mean support for an OS 9 compatible execution environment... whether that's via Classic or just via a compatible binary format. And that's the only reason for sticking with Codewarrior.
Adobe should have had a Project Builder transition strategy underway for at least 5 years now.
Slashdot Mirror
That's fair enough.
I still think the open source path is likely to happen sooner.
I don't quite have the skills or time to undertake a task like that.
Sounds like an opportunity for an Open Source Bounty then...
I figured SGI was long gone already.
I don't see how you could possibly use this iPod inside of a protective case.
You can buy a 12-pack of PDA screen protectors that will last the lifetime of the iPod for under ten bucks.
Cut your own tracks...
If two sequential tracks on an album are effectively one work, there are plenty of good, and even free, mixing programs that you can use to create a single merged track with the fade you want.
do you think they'd consider releasing some form of Safari for Windows?
You have:
* The complete source to Webkit.
* Gtk for Windows.
What else do you need for a Windows port?
If the branch predictor is good enough, it really doesn't matter.
If that were the case, the G5 and the P4 wouldn't be dead ends. If that were the case, G4 wouldn't be getting more IPC than the G5. If that were the case, PIII wouldn't be getting more IPC than the P4. Pipeline does matter, in the real world.
Pipeline matters. Instruction set matters. Register file size matters.
It got toasted on SPECint too (a factor of 2x at the same clock-speed = toasted)
That's comparing gcc (which isn't SPEC-optimised) against Intel's compiler, and... is that taking dual-core results and dividing by two, or is that the single-core versus the single-core? That makes a BIG difference to the multithreading benchmarks.
The memory bus is 667MHz for the MPC8641D, not 768MHz.
OK, that's changed. They previously listed it as half clock. 667MHz on each of two busses is still nothing to sneeze at.
cache and memory latencies. These things aren't any lower on the G4 than they are on Yonah.
:)
Cache and memory latencies are significantly aggravated by a long pipeline, which is why the G5 still gets fewer IPC than the G4 despite the G5's truly impressive design.
Yonah's pipeline is probably at least twice as long as the G4's: the P6 pipeline is 10+ stages and the latest version added at least 4 more.
Oh, yeh, Yonah isn't all that new a design. It's another spin on the venerable P6 core, with SSE3 and a better shared cache this time around.
I would like to see the source of your SPEC comments.
Apple
Apple has published SPEC for the MPC8461D? Where?
The benchmarks I've seen comparing the G4 (even with the e600 core) with the Core Duo are all based on the 7448 and earlier, and all of those are crippled by the slow memory bus. These results are consistent with Yonah's main performance advantage being memory speed. When you take a system with a single core and a 166 MHz memory bus (like the Powerbook) and compare it to one with dual core and a 533 MHz memory bus of course it's going to get its ass kicked... but the benchmark it really got toasted on was stream... and that's a *memory bandwidth* test.
Take an MPC8641D and put memory on BOTH of the 768 MHz memory busses and if it doesn't beat Yonah I'd be very much surprised.
Right now 3d virtual reality is about where text adventure was in the early '70s. Infocom kept the text adventure alive pretty long, but 2d and then 3d pretty much took it over, and the only place you got text role playing was in themed chat systems... MUDs.
I'm predicting that as free and cheap 3d virtual reality gets more common, the kinds of hack-n-slash stuff you see in MMORPGs now are going to become the kinds of things hobbyists put together inside virtual worlds that exist for other reasons. Right now that stuff is really crude by comparison, but it's just a matter of time before realistic graphics become as generic as verb-noun parsers and scripted objects in MUDs.
So the kinds of things that go on in 3d games now will be like LARPing in VR. SCA-type stuff. What will *companies* be doing instead? Heck if I know... interactive movies with licensed celebrity characters and paid actors?
You can pay people real money to farm stuff for you... even if they know you bought it, it's not different from showing up in a nice car and playing like a duffer.
The new dual-core G4 is fairly primitive technology by today's standards.
The G4 core is no slouch. With a short pipeline it does very well and doesn't require a sophisticated out-of-order processor to get good performance. Its biggest shortcoming is the slow bus (166 MHz, up to 200 in the 7448, though Apple doesn't provide the 7448 with a 200 MHz bus). This is similar to the problems of the P6 core (which Yonah is based on) before Intel grafted a faster FSB to it.
I would like to see the source of your SPEC comments. SO far as I know there haven't been comparable benchmarks of the MPC8641D and Yonah on similar hardware. You can't generalise from the performance of the 7448 because it's still crippled by the slow bus of the previous G4s. Do you actually have SPEC results for the 8641D on a 768 MHz bus system?
IBM had no good competitor to Yonah and Conroe.
Freescale did.
I'd just like someone to build a worthy successor to the pre-ARM Clies. I have a Sony SJ-22 and I haven't seen anything to touch it. The later Clies were gimmicky monstrosities (half of them didn't even have Jog-dials!), and Palm seems to have dropped the ball after Hawkins came back. I've tried three generations of Windows-powered devices and, well, it's still basically a desktop OS.
Palm needs to take PalmOS 4 and replace the AMX kernel with some open source alternative, and go back to their roots. Possibly using Coldfire instead of ARM will allow for mostly-native emulation...?
The mouse doesn't move vertically, it moves side to side, it's just held vertically.
I think something shaped more like a joystick might work better, but you still lose too much fine control by having to hold the mouse rather than just resting your fingers on top of it. That's a bigger problem than "gorilla arm".
"business decisions were made based on the value of those patents, and there would be deep concern from companies if that IP value vanished."
How much intellectual property value is there in any software patent that can't be recovered within, say, three to five years? What software patents are actively being used to generate significant revenue? How long are they likely to continue generating that income? Could a transition period like that make sense?
Exposé isn't cut from whole cloth, either: cool transitions between display layouts and virtual screens go all the way back to the Amiga, and virtual window managers on UNIX have used them for desktop transitions for a while now.
Similarly, thumbnail views of a larger vitual workspace are common.
Exposé combines a cool transition with a dynamically created layout, and it looks cool, but frankly I'd rather Apple work on a virtual desktop or even trying to really take advantage of the 3d capabilities of OpenGL than more bling like Expose, Genie, and Dashboard ripples.
...they didn't switch for performance, or heat, or anything like that. The cost/performance ratio of silicon is switching back and forth between architectures all the time, and right now the Power PC is doing relatively well. Apple's had it a LOT worse in the past. In fact, between the time the G5 Powermac came out and when Jobs made the announcement that he was tired of waiting for a 3 GHz G5 the G5 had actually ramped up in speed better than Intel's chips. Intel had the multiple cores, but IBM was close enough that Apple came out with the Quad G5 before the iMac Core Duo.
Jobs had been against the Power PC from the start. The switch to Power PC was made while he was away at NeXT, and the first software that came out of the Apple/NeXT merger was Rhapsody... on Power PC AND on Intel. NeXT already had multi-platform executables, and the proposed way forward in 1997 was to Yellow Box (OpenStep on Rhapsody, what became Cocoa) with existing Mac OS applications running under Blue Box (which evolved into Classic). Jobs could kill two birds with one stone... he had a path to back away from the Power PC, and he could dump the (truly awful) Mac OS API.
It was not to be.
The ISVs and customers rioted, and they went back to the drawing board, came up with Carbon as an intermediate API, and when OS X (finally) came out it supported Carbonised Mac OS apps almost natively, Cocoa, and legacy apps under emulation in Classic.
The reason they did this now, instead of in 2000 or 2003 or 2008, is because they were confident they could get away with it now. And I think the trigger happened in 2004. Late in the year they pulled the last OS-9-compatible Powermac G4 from the Apple Store... and nobody made a fuss. They'd tried it before, but every time the users were up in arms. This told them they'd finally scotched the classic serpent and could complete the conversion to NeXTstep.
The quoted text is by Michael Crichton, not Gregory Benford. Can you explain what relevance Crichton's comments have to Benford's scheme?
Something I do truly believe in is that there is statistical evidence that suggests we are in a NATURAL warming trend. Ice data collected in the Arctic and Antarctic suggest that weather patterns are cyclical, and we are about due for a natural warming of global temperatures.
Yes, they're cyclical, but the last 4 cycles have shown long periods of cold with short peaks at or above the current temperatures. These peaks are about 10,000 years long. The last one started over 10,000 years ago.
The data is very noisy, and you can't say much with certainly, but only the most hopeful of observers could look at it and argue that the evidence suggests we're due for a warmer period... on the contrary, we're due for an ice age.
The fella is missing the biggest advantage to UNIX, and that's got nothing to do with whether you're running as root or Administrator or Fred Nurk or Minnie Bannister.
THe bissgest security advantage that UNIX has is that it doesn't have Internet Explorer, Outlook, Windows Media Player, and all the other software that uses that Typhoid Mary of the Internet, the Microsoft HTML control.
The idea that it's even vaguely acceptable to not only build the entire desktop environment about a web browser, but to make that browser... and not the applications that call it... responsible for determining what it's going to let a document do, is just so mind-blowing that seven years after I first saw what was then called Active Desktop I'm still having trouble believing that they're still doing it.
What are these people thinking?
... he thought the Earth was significantly smaller than was known at the time, and if it wasn't for the sheer good luck that there was a landmass unknown to either him or his detractors in the way, and that it was so much closer to the European than the Asian side of the "world ocean", he'd have just been another leader of an ill-fated expedition.
On Windows and X11 I typically keep a notepad window open, and use it for all non-english text-entry. The fact that you don't need to do this with Cocoa and many Carbon apps (Carbon apps, particularly the ones based on ports from UNIX or Windows, don't always seem to hook in to OS X as well as Cocoa) is great.
But...
You should be able to copy and paste from TextEdit into the GIMP no matter what the character set. This isn't a real solution, but it's a workaround that should be acceptable for an application where text input is (or should be) a relatively rare operation.
That's correct: the warranty doesn't include shipping, and they only support shipping via a fairly pricey DHL Express service... but that's included in the $30. They sent me a return-shipment box that arrived the next business day, and the replacement unit was in my hands three business days after I dropped it off at DHL.
Adobe is relatively cold to the Mac Community for some reason or an other.
Oh, to be a fly on the wall where Apple and Adobe went at it over Yellow Box and Display Postscript. I bet there were harsh words on both sides.
Let's pretend Adobe dropped everything to focus on a transition to Universal binaries right after WWDC last year.
Let's pretend that somewhere in Adobe's management structure there exists a decision maker who is capable of rational thought. Even without the Intel switch, Codewarrior has been a lame duck since 1997. That's when Apple first tried to switch to the NeXT codebase and relegate the classic Mac OS API to "Blue Box". Carbon was explicitly a transition API, at first, and Adobe seems to have decided that Apple's subsequent use of Carbon internally meant that Codewarrior was safe. Carbon no longer means OS 9, and Carbon support shouldn't be assumed to mean support for an OS 9 compatible execution environment... whether that's via Classic or just via a compatible binary format. And that's the only reason for sticking with Codewarrior.
Adobe should have had a Project Builder transition strategy underway for at least 5 years now.