I disagree and I'll tell you why. They big iron AMD stuff being good isn't hard to fathom. AMD bought the IP from DEC for the Alpha chip. That sucker was a beast back in the day. Thus it's not a stretch to see why they did very well in the Server sector.
The only notable DEC IP I am aware of in AMD products was the EV6 bus used on the K7, but that's gone on the K8. A lot of DEC engineers did jump to AMD, and I think that's the influence you were thinking of.
I disagree and I'll tell you why. They big iron AMD stuff being good isn't hard to fathom. AMD bought the IP from DEC for the Alpha chip. That sucker was a beast back in the day. Thus it's not a stretch to see why they did very well in the Server sector.
That was 7 years ago. You'd have thought Intel would have got the idea that a fast serial interconnect was a good thing by now. And they have but it's still over a year off.
Quote:
Originally posted by hmurchison
The Pentium M was still competitive even at 90nm. AMD seems to be getting there. I saw a roadmap with Athlon at 35 watt TDP so both companies are minding the wattage which is good. However Intel IMO has more experience at the low watt/high performance architecture.
AMD Turions run at 25 to 35W. They don't do a dual core yet though which is due in a matter of months at speeds of up to 2.66Ghz supposedly. This is still at 90nm. AMD seem to have much lower TDP requirements at 90nm than Intel.
Quote:
Originally posted by hmurchison
Woodcrest will be tough to beat even for the next Opteron (85 series ??) coming out. Think about the effects on a data center when you have servers that have Woodcrest chips running %30 cooler. It adds up quickly.
The 185, 285 and 885 were out the day before Intel's developer forum. Heat and power consumption is one concern in a datacentre but I usually buy them based on performance under load and AMD Opterons trounce Intel currently, especially running 64bit Linux. The new chips will be great if you want really fast CPUs but in general datacentre use are overkill. Most people buy behind the curve. I tend to use Opteron 244s which run at 1.8Ghz. For a web server which is rarely CPU bound, they beat the Xeons hands down for shifting web pages and running PHP/Perl. And the CPUs cost about $120 each. :-)
I'll sell you an Intel Xeon server if you want one but IME the Opterons are quicker, cheaper, faster and more reliable.
Quote:
Originally posted by hmurchison
Hey I like AMD just like anyone but Intel isn't a company you just count out. They'll hit 45nm first and if CSI hits the ground running (their Hypertransport competitor) then things begin to get interesting. I don't expect ondie memory controllers to come anytime soon. FB-DIMMs reduce latency I don't see it as much of problem
Well, they've cancelled CSI once already and look to be a year late now. It's generally not a problem though till you get to servers or Quad-core. If Intel get to 45nm when AMD have just got to 65nm then they'll continue to do well. AMD have reportedly also had problems beating their K7/K8 design with some false starts and renames. So, anyone's guess really though you can probably bet on Intel winning the process war.
For a PowerMac, I still think it's going to be a closely run thing AMD v Intel. At least there's one good thing to come out of the Intel switch and that's AMD are there to keep Intel hungry. Unlike Freescale and IBM.
AMD Turions run at 25 to 35W. They don't do a dual core yet though which is due in a matter of months at speeds of up to 2.66Ghz supposedly. This is still at 90nm. AMD seem to have much lower TDP requirements at 90nm than Intel.
TDP is different to average power draw. On battery Core Duos draw very little power, the same isn't true for Turions, which is why AMD notebooks still don't compare to centrino.
I wish Apple would just buy Adobe. Then Apple could take a page out of the MS playbook and make windows users wait for their version and make it real buggy.
I'm not entirely up to date on the matter, so I may be wrong but..
Apple ~ $25 billion
Adobe ~ $28 billion
Isn't that about right?
In any case, it isn't currently possible for Apple to purchase Adobe, unless Steve starts dipping into his own pockets..
EDIT:
Wow. In the last year things have really changed !!:
In any case, it isn't currently possible for Apple to purchase Adobe, unless Steve starts dipping into his own pockets..
Is it better, as in Pixar vs Disney, to "steal" the creative talent given the layoffs at Adobe/Macromedia?
I would think that morale isn't at an all time high.
If you buy a company folks can leave. If you hire the folks, presumably they want to be with you. You may not get the specific IP or branding but I think Apply has decent enough branding as is.
do you aquire other smaller software firms and build your own "adobe killer" strategic aquisitions??
Corel has big range of Applications that Apple has made familiarity with in it's own software. Plus Corel Has Painter, CorelDraw Graphics Suite, and now Paint shop Pro. Plus many others Like Word Perfect Office to start.
Corel has big range of Applications that Apple has made familiarity with in it's own software. Plus Corel Has Painter, CorelDraw Graphics Suite, and now Paint shop Pro. Plus many others Like Word Perfect Office to start.
You may be on to something there. I've always liked Paint Shop Pro, which at least for me, an amateur user, has always been more intuitive to use than Photoshop. That said, Corel doesn't really offer too many software products for Mac OS X, if I recall, including Paint Shop Pro, so significant investment would be required by Apple to make anything important come out an acquisition.
Comments
Originally posted by hmurchison
I disagree and I'll tell you why. They big iron AMD stuff being good isn't hard to fathom. AMD bought the IP from DEC for the Alpha chip. That sucker was a beast back in the day. Thus it's not a stretch to see why they did very well in the Server sector.
The only notable DEC IP I am aware of in AMD products was the EV6 bus used on the K7, but that's gone on the K8. A lot of DEC engineers did jump to AMD, and I think that's the influence you were thinking of.
Originally posted by hmurchison
I disagree and I'll tell you why. They big iron AMD stuff being good isn't hard to fathom. AMD bought the IP from DEC for the Alpha chip. That sucker was a beast back in the day. Thus it's not a stretch to see why they did very well in the Server sector.
That was 7 years ago. You'd have thought Intel would have got the idea that a fast serial interconnect was a good thing by now. And they have but it's still over a year off.
Originally posted by hmurchison
The Pentium M was still competitive even at 90nm. AMD seems to be getting there. I saw a roadmap with Athlon at 35 watt TDP so both companies are minding the wattage which is good. However Intel IMO has more experience at the low watt/high performance architecture.
AMD Turions run at 25 to 35W. They don't do a dual core yet though which is due in a matter of months at speeds of up to 2.66Ghz supposedly. This is still at 90nm. AMD seem to have much lower TDP requirements at 90nm than Intel.
Originally posted by hmurchison
Woodcrest will be tough to beat even for the next Opteron (85 series ??) coming out. Think about the effects on a data center when you have servers that have Woodcrest chips running %30 cooler. It adds up quickly.
The 185, 285 and 885 were out the day before Intel's developer forum. Heat and power consumption is one concern in a datacentre but I usually buy them based on performance under load and AMD Opterons trounce Intel currently, especially running 64bit Linux. The new chips will be great if you want really fast CPUs but in general datacentre use are overkill. Most people buy behind the curve. I tend to use Opteron 244s which run at 1.8Ghz. For a web server which is rarely CPU bound, they beat the Xeons hands down for shifting web pages and running PHP/Perl. And the CPUs cost about $120 each. :-)
I'll sell you an Intel Xeon server if you want one but IME the Opterons are quicker, cheaper, faster and more reliable.
Originally posted by hmurchison
Hey I like AMD just like anyone but Intel isn't a company you just count out. They'll hit 45nm first and if CSI hits the ground running (their Hypertransport competitor) then things begin to get interesting. I don't expect ondie memory controllers to come anytime soon. FB-DIMMs reduce latency I don't see it as much of problem
Well, they've cancelled CSI once already and look to be a year late now. It's generally not a problem though till you get to servers or Quad-core. If Intel get to 45nm when AMD have just got to 65nm then they'll continue to do well. AMD have reportedly also had problems beating their K7/K8 design with some false starts and renames. So, anyone's guess really though you can probably bet on Intel winning the process war.
For a PowerMac, I still think it's going to be a closely run thing AMD v Intel. At least there's one good thing to come out of the Intel switch and that's AMD are there to keep Intel hungry. Unlike Freescale and IBM.
Originally posted by aegisdesign
AMD Turions run at 25 to 35W. They don't do a dual core yet though which is due in a matter of months at speeds of up to 2.66Ghz supposedly. This is still at 90nm. AMD seem to have much lower TDP requirements at 90nm than Intel.
TDP is different to average power draw. On battery Core Duos draw very little power, the same isn't true for Turions, which is why AMD notebooks still don't compare to centrino.
Originally posted by backtomac
I wish Apple would just buy Adobe. Then Apple could take a page out of the MS playbook and make windows users wait for their version and make it real buggy.
I'm not entirely up to date on the matter, so I may be wrong but..
Apple ~ $25 billion
Adobe ~ $28 billion
Isn't that about right?
In any case, it isn't currently possible for Apple to purchase Adobe, unless Steve starts dipping into his own pockets..
EDIT:
Wow. In the last year things have really changed !!:
Apple = $55,336,146,880
Adobe = $22,584,241,740
- Xidius
Originally posted by >_>
In any case, it isn't currently possible for Apple to purchase Adobe, unless Steve starts dipping into his own pockets..
Is it better, as in Pixar vs Disney, to "steal" the creative talent given the layoffs at Adobe/Macromedia?
I would think that morale isn't at an all time high.
If you buy a company folks can leave. If you hire the folks, presumably they want to be with you. You may not get the specific IP or branding but I think Apply has decent enough branding as is.
Vinea
Originally posted by NOFEER
do you aquire other smaller software firms and build your own "adobe killer" strategic aquisitions??
Corel has big range of Applications that Apple has made familiarity with in it's own software. Plus Corel Has Painter, CorelDraw Graphics Suite, and now Paint shop Pro. Plus many others Like Word Perfect Office to start.
Originally posted by onlooker
Corel has big range of Applications that Apple has made familiarity with in it's own software. Plus Corel Has Painter, CorelDraw Graphics Suite, and now Paint shop Pro. Plus many others Like Word Perfect Office to start.
You may be on to something there. I've always liked Paint Shop Pro, which at least for me, an amateur user, has always been more intuitive to use than Photoshop. That said, Corel doesn't really offer too many software products for Mac OS X, if I recall, including Paint Shop Pro, so significant investment would be required by Apple to make anything important come out an acquisition.