as an aside, those 3840x2400 displays have astronomical response times that make them utterly unsuitable for video (those panels have a 50ms response lag iirc), whereas the specs on the 30" ACD point to a 16ms response time... which is VERY impressive.
" It makes a very large assumption about the way the user will set up their system. It requires that it be standing upright, or else the cooling drops in effectivness, big time."
FYI, there's a bigass heatpipe on the backside of the rev A g5 Powermac mainboards, and Apples docs on the G5 (rev A at least) state that it should be used only in an upright position:)
Okay then, an Opteron 848 is $1299 (according to pricewatch anyway)
can't find a quad Opteron mainboard price though, the Tyan K8QS looks like it's the business, but I can't find anywhere in my limited searching that's actually selling them.
Manufacturing limitations mainly, although I wouldn't be entirely surprised to see IBM starting to put edram L3 caches on their dies, as the transistor density is a lot higher than sram and the latency isn't all that far off...
"Third, the iBook was designed to use the keyboard as venting, so you'll end up with a burned out machine if you use the clampshell mode of the hack, which also would include you to trick it to go out of sleep with a USB device."
I hear that a lot, but surely that's what the vent around the hinge is for?
Ah, but if people are buying them, take take the hit when it sells, and then build _another_ one to replace it in inventory, which means the overall loss to MS is bigger;)
"Also, if Motorolla was so awful to Apple, why did Apple stick to them? Because of the CEO's decisions, that's why. A good CEO would have found a new CPU provider (or switched to IBM years ago). That's entirely Apple's fault."
It became apparent Motorola weren't going to pull out of their lull ~ mid-2001
Apple and IBM got started on the PowerPC 970.... ~mid-2001
"The only think keeping me from getting a nice, new 15" AlBook is that when I unplugged the one at the Apple Store, it said "100% charged, 1:12 remaining" and that is simply an unacceptably short battery life."
Sounds to me like they didn't calibrate the battery, because the battery life on them is a LOT better than that.
the Athlon64 non-FX is by no means a 'totally different core with different optimisations'
it's the same execution core, with the same cache, and half the memory bus width. (also, it doesn't need registered ram, so it has a slight latency advantage.)
Having listened in to a few programmers discussing that, it seems that it's pretty much only RC5 that actually manages to make full use of the Altivec unit in the 745x series chips anyway, and the benches I've seen that _haven't_ been littered with vec_dst show that the 970 scales in a fairly linear fashion over the G4 as far as altivec code codes (except it also maintains that scaling on larger datasets, since its got a fast memory subsystem, rather than a slow memory subsystem with a fast L3)
g{The G5 is NOT going to excel at Altivec optimised code, the G4 will remain the Altivec champ until IBM puts the kind of Altivec resources onto the 970 that Moto put onto the 745x series}g
That's not strictly true, the problem is that altivec that's optimal for the 745x chips is worst case for the 970, since best case on the 745x makes heavy use of the vec_dst (prefetch) instruction, which is handled serially on the 970 and hence stalls the Altivec hardware.
also the G5 is a lot less likely to run into bandwidth bottlenecks with altivec code on large datasets.
" The G5 consumes huge amounts of power (like 90W)"
Except it doesn't, at 1.8Ghz it goes through ~42W, the 2Ghz parts should be in the region of 47W each.
That particularly piece of fud can be traced back to a Register.co.uk journalist getting the power consumption figures for BOTH G5s confused (they claimed 97W each, which is about the combined figure for a dually 2Ghz rig)
Slashdot Mirror
as an aside, those 3840x2400 displays have astronomical response times that make them utterly unsuitable for video (those panels have a 50ms response lag iirc), whereas the specs on the 30" ACD point to a 16ms response time... which is VERY impressive.
You mean the C64GS ?
like so.
"Is there any such thing as a British libertarian? "
Aye, over here!
>:(
" It makes a very large assumption about the way the user will set up their system. It requires that it be standing upright, or else the cooling drops in effectivness, big time."
:)
FYI, there's a bigass heatpipe on the backside of the rev A g5 Powermac mainboards, and Apples docs on the G5 (rev A at least) state that it should be used only in an upright position
um, GLQuake needs hardware opengl support, which neither the Matrox Millennium nor Mystique had any sign of.
On the flipside, Delta do make some very quiet fans.
:)
eg, the fans in a Powermac G5
Okay then, an Opteron 848 is $1299 (according to pricewatch anyway)
can't find a quad Opteron mainboard price though, the Tyan K8QS looks like it's the business, but I can't find anywhere in my limited searching that's actually selling them.
Manufacturing limitations mainly, although I wouldn't be entirely surprised to see IBM starting to put edram L3 caches on their dies, as the transistor density is a lot higher than sram and the latency isn't all that far off...
The Amiga 3000 and 4000 had HD floppy drives, and they got 1.76MB on a HD floppy by default, with DFS they could squeeze 1.96MB onto one.
"Third, the iBook was designed to use the keyboard as venting, so you'll end up with a burned out machine if you use the clampshell mode of the hack, which also would include you to trick it to go out of sleep with a USB device."
I hear that a lot, but surely that's what the vent around the hinge is for?
actually, it's pretty much only DiVX encoding that the P4 wins big on, for other codecs, it's either a wash or the Athlon pulling ahead.
Ah, but if people are buying them, take take the hit when it sells, and then build _another_ one to replace it in inventory, which means the overall loss to MS is bigger ;)
Actually you're both wrong, kinda.
the original G4 is a G3 + Altivec + MERSI + the 604e's FPU hardware
the modern G4's bear very little resembalance to the 7400 however, having a pipeline three stages longer and a more elaborate altivec implementation.
"Also, if Motorolla was so awful to Apple, why did Apple stick to them? Because of the CEO's decisions, that's why. A good CEO would have found a new CPU provider (or switched to IBM years ago). That's entirely Apple's fault."
It became apparent Motorola weren't going to pull out of their lull ~ mid-2001
Apple and IBM got started on the PowerPC 970.... ~mid-2001
you do the maths...
" My original 5 gig got went from 11 hours to 45 mins over three years of constant use."
three years of constant use?
from a device that's been out barely over two years?
How exactly is he supposed to test the way websites render _in Safari_ using Mozilla?
"The only think keeping me from getting a nice, new 15" AlBook is that when I unplugged the one at the Apple Store, it said "100% charged, 1:12 remaining" and that is simply an unacceptably short battery life."
Sounds to me like they didn't calibrate the battery, because the battery life on them is a LOT better than that.
a 1.25Ghz G4 Powermac is $1299 ?, or you could look towards the recently refreshed ibook?
the Athlon64 non-FX is by no means a 'totally different core with different optimisations'
it's the same execution core, with the same cache, and half the memory bus width. (also, it doesn't need registered ram, so it has a slight latency advantage.)
Code can be optimised for the PPC970 execution pipeline / caching without hulking 64bit integer values and pointers around.
;)
So I'd say that they didn't hint that they weren't using the 7.0.1 G5 patch
Number of photoshop images that I've seen topping 600MB (plus application + undo buffer) = lots
number of games that use >512MB of ram = not very many
Game devs would have to be frickin' idiots to assume everyone has 4GB of ram to run their game.
Having listened in to a few programmers discussing that, it seems that it's pretty much only RC5 that actually manages to make full use of the Altivec unit in the 745x series chips anyway, and the benches I've seen that _haven't_ been littered with vec_dst show that the 970 scales in a fairly linear fashion over the G4 as far as altivec code codes (except it also maintains that scaling on larger datasets, since its got a fast memory subsystem, rather than a slow memory subsystem with a fast L3)
g{The G5 is NOT going to excel at Altivec optimised code, the G4 will remain the Altivec champ until IBM puts the kind of Altivec resources onto the 970 that Moto put onto the 745x series}g
That's not strictly true, the problem is that altivec that's optimal for the 745x chips is worst case for the 970, since best case on the 745x makes heavy use of the vec_dst (prefetch) instruction, which is handled serially on the 970 and hence stalls the Altivec hardware.
also the G5 is a lot less likely to run into bandwidth bottlenecks with altivec code on large datasets.
" The G5 consumes huge amounts of power (like 90W)"
Except it doesn't, at 1.8Ghz it goes through ~42W, the 2Ghz parts should be in the region of 47W each.
That particularly piece of fud can be traced back to a Register.co.uk journalist getting the power consumption figures for BOTH G5s confused (they claimed 97W each, which is about the combined figure for a dually 2Ghz rig)
ARM based desktop you say?
try Iyonix complete with RiscOS